THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S The Sourcebook on the Foundations of Social Protection Delivery Systems synthesizes real-world experiences and lessons learned of social protection delivery systems from around the globe. It DELIVERY SYSTEMS a broad view of social protection, covering various intended populations such as poor or low-inc families, unemployed workers, persons with disabilities, and individuals facing social risks. It discu many types of interventions that governments provide to individuals, families, or households, in Leveraging Digital Technology for categorical programs, poverty-targeted programs, labor benefits and services, disability benefits services, and social services. Adaptive seeks to Protection Social The Sourcebook in address concrete “how-to” Pakistan questions, including: How do countries deliver social protection benefits and services? How do they do so effectively and efficiently? How do they ensure dynamic inclusion, especially for the most vulnerable and needy? Melis Guven, Zaineb Majoka, and Gul Najam Jamy How do they promote better coordination and integration—not only among social protec programs but also among programs in other parts of government? How can they meet the needs of their intended populations and provide a better client experience? The delivery systems framework elaborates on the key elements of that operating environment. framework is anchored in core implementation phases along the delivery chain. Key actors, inclu people and institutions, interact all along that delivery chain. Those interactions are facilitated b communications, information systems, and technology. This framework can apply to the delivery one or many programs and to the delivery of adaptive social protection. The Sourcebook structures itself around eight key principles that can frame the delivery systems min 1. There is no single blueprint for delivery systems, but there are commonalities, and those common elements constitute the core of the delivery systems framework. 2. Quality of implementation matters, and weaknesses in any of the core elements will nega affect the entire system, reducing the impacts of the program(s) they support. 3. Delivery systems evolve over time, in a nonlinear fashion, and their starting points matter 4. Efforts should be made to “keep it simple” and to “do simple well,” from the start. 5. The “first mile”—people’s direct interface with administrative functions—is often the weak link in the delivery chain; improving it may take systemic change but will greatly improve efficiencies and mitigate the risk of failures on the frontlines. 6. Social protection programs do not operate in a vacuum, and thus their delivery systems s not be developed in silos; synergies across institutions and information systems are poss and can improve program outcomes. 7. Social protection delivery systems can contribute more broadly to government’s ability to other sectors, such as health insurance subsidies, scholarships, social energy tariffs, housi benefits, and legal services. 8. The dual challenges of inclusion and coordination are pervasive and perennial and encou the continuous improvement of delivery systems, through a dynamic, integrated, and hum centered approach. © 2024 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Disclaimer This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Attribution Please cite the work as follows: Guven, Melis, Zaineb Majoka, and Gul Najam Jamy. 2024. The Evolution of Benazir Income Support Programme’s Delivery Systems: Leveraging Digital Technology for Adaptive Social Protection in Pakistan. Washington, DC: World Bank Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover and Layout Design: A.DelaRoche Designs Photography credits: World Bank Photo Collection (Flickr Attribution- NonCommercial-license), Shutterstock (Shutterstock standard image license), and Wikicommons (licensed under the Creative Commons Attribution 4.0 International license) THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS Leveraging Digital Technology for Adaptive Social Protection in Pakistan Melis Guven, Zaineb Majoka, and Gul Najam Jamy THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS Leveraging Digital Technology for Adaptive Social Protection in Pakistan CONTENTS Acknowledgements 1 Acronyms 2 Executive summary 3 1. Introduction 7 2. The socioeconomic context of Pakistan 11 3. Social protection in Pakistan: from the past to the present 14 3.1 The launch and expansion of the Benazir Income Support Programme 16 3.2 The launch and expansion of the National Socio-Economic Registry 23 3.3 Provinces roll out social protection programs aimed at adaptive social protection 33 The Sourcebook on the Foundations of Social Protection Delivery Systems synthesizes real-world experiences 4. The Social Protection and Delivery for learned lessons Chain BISP of social protection delivery systems from around 36 the globe. It t 4.1 ASSESSa broad view of social protection, covering various intended populations such as 37 poor or low-inc 4.1.1 Outreach and communications 37 families, unemployed workers, persons with disabilities, and individuals facing social risks. It discu 4.1.2 Intake and registration and assessment of needs and conditions 43 many types of interventions that governments provide to individuals, families, or households, inc 4.2 ENROLL categorical programs, poverty-targeted programs, labor benefits and services, 45 disability benefits 4.3 PROVIDEservices, and social services. 52 The Sourcebook seeks to address concrete “how-to” questions, including: 4.4 MANAGE 59 How do countries deliver social protection 4.4.1 Case management and grievance redress mechanism benefits and services? 59 4.4.2 Monitoring How do conditionalities for the Benazir they do so effectively and Taleemi Wazaif and Nashonuma CCTS efficiently? 64 4.4.3 Monitoring of payments 67 How do they ensure dynamic inclusion, especially for the most vulnerable and needy? 4.4.4 Exit decisions 68 How do they promote better coordination and integration—not only among social protec programs but also among programs in other parts of government? 5. Monitoring and evaluation 70 How can they meet the needs of their intended populations and provide a better client experience? 6. Conclusions and recommendations 73 The delivery systems framework elaborates on the key elements of that operating environment. References framework is anchored in core implementation phases along the delivery chain. 78 Key actors, inclu people and institutions, interact all along that delivery chain. Those interactions are facilitated by communications, information systems, and technology. This framework can apply to the delivery one or many programs and to the delivery of adaptive social protection. The Sourcebook structures itself around eight key principles that can frame the delivery systems min FIGURES Figure 1: Social protection delivery chain 9 Figure 2: Persistent human capital index gap between the richest and poorest group over decades 13 Figure 3: The BISP’s budget allocation and number of beneficiaries 17 Figure 4: BISP’s annual budget allocation 17 Figure 5A: Poverty rates by district 18 Figure 5B: BISP beneficiaries by district 18 Figure 6: Percentage of total population covered by social assistance 20 Figure 7: Change in real value of benefit level 21 Figure 8: Change in adequacy of BISP transfer 22 Figure 9: NSER-NADRA Workflow 24 Figure 10: Process cycle of dynamic registry 28 Figure 11: NSER Data-Sharing process with other institutions 30 Figure 12A-12B: Population affected by floods 32 Figure 13: The outreach phase of the social protection delivery chain 38 Figure 14: Modes of communication about the WeT sub-program 40 Figure 15: Intake and registration and assessment of needs and conditions 44 Figure 16: The enrollment stage of the social protection delivery chain 45 Figure 17: BTW admission process flow 47 Figure 18: Flow of BTW android application 49 Figure 19: Process flow for enrollment of PLW & children 50 Figure 20: Process flow for the nutrition CCT 51 Figure 21: Provision of payments 52 Figure 22: Evolution of BISP payment system 53 Figure 23: BISP payment workflow 55 Figure 24: Kafaalat case management (GRM) 60 Figure 25: BISP payments complaints management system 61 Figure 26: Assessment of delivery systems 74 Figure 27: Proposed expanded data integration between NADRA and the NSER 75 Figure 28: Depiction of a knowledge discovery platform 76 TABLES Table 1: Human capital indicators by region 12 Table 2: Separation criteria for Benazir Taleemi Wazaif 69 BOXES Box 1: The impact of the Benazir Income Support Programme on poverty and household welfare 21 Box 2: Poverty score card based on poverty means test 23 Box 3: An exemplary partnership: the collaboration between BISP and NADRA 24 Box 4: Moving toward a dynamic social registry: lessons from other countries 26 Box 5: Using the NSER in emergency response situations 31 Box 6: Provincial social protection programs 33 Box 7: Local governments participate in social protection delivery systems 34 Box 8: Enhancing women’s empowerment through identity cards 38 Box 9: BISP beneficiary committees 41 Box 10: Raising awareness of disaster preparedness using social protection schemes 42 Box 11: Using the BISP payment systems in shock response 57 Box 12: Kenya’s Inua Jamii payment system 58 Box 13: The Grievance Redress mechanism in Turkiye 63 Box 14: Jordan’s National Aid Fund 64 Box 15: Roles in the verification compliance with CCT co-responsibilities: country examples 65 Box 16: Leveraging savings and financial inclusion to support beneficiaries beyond cash transfers 69 Box 17: Management tools to promote quality in delivery system: lessons from Brazil 72 ACKNOWLEDGEMENTS “The Evolution of Benazir Income Support Programme’s Delivery Systems: leveraging digital technology for Adaptive Social Protection in Pakistan” was prepared by a team led by Melis Guven and Zaineb Majoka. The core team from the World Bank Group’s Social Protection and Jobs Global Practice consisted of Gul Najam Jamy, Polly Jones, and Murium Hadi. Syed Muzaffar Ali’s technical background paper on information systems was instrumental in shaping the report’s relevant sections. William Cook provided valuable feedback on payment systems. Surat Nsour provided references on outreach and communications. The team also received support on visuals from Fahad Mirza and Atsuhiro Oguri, and on graphic design from Laura Restrepo and Andres de la Roche. Deepika Davidar edited the final report. We are grateful for the guidance and feedback received from Cem Mete, Stefano Paternostro and Amjad Zafar Khan. Phillippe Leite, Sohail Abbasi, and Amna Khan provided valuable inputs on social protection initiatives implemented by provincial governments. We also appreciate the guidance and strategic direction received from Najy Benhassine (Country Director for Pakistan) and Nicole Klingen (Human Development Regional Director, South Asia Region). We acknowledge the helpful comments received from peer reviewers: Junko Onishi, Yoonyoung Cho, and Kenichi Chavez (at the concept note stage) and Iftikhar Malik, Sarah Coll-Black, and Ubah Thomas Ubah (at the decision review stage). Waleed Anwar and Syed Farrukh Ansar helped to organize the concept note and decision review meetings. The Sourcebook on the Foundations of Social Protection Delivery Systems synthesizes real-world experiences and lessons learned of social protection delivery systems from around the globe. It We would like to extend our special appreciation to Shazia Marri, the Chairperson of the BISP and Federal a broad view of social protection, covering various intended populations such as poor or low-inc Minister for the Ministry of Poverty Alleviation and Social Safety (2022-2023), Secretary Yusuf Khan, Additional Secretary Tahir Noor,families, and NSER unemployed Wing Directorworkers, persons General Naveed with Akbar disabilities, for their guidanceand individuals and facing support. We social risks. It discu are also many types of interventions that governments provide to individuals, families, Ali thankful for our informative technical discussions with a number of BISP staff including Ahmer Athar, Hamid or households, in Khan, Khurram Shehzad, Bakhsh Mehar, Hazoorprograms, categorical Saqib Mumtaz, Nouman poverty-targeted programs, Ali, labor Farhan Abbasi, and benefits andNadia Fakhar. services, disability benefits Our discussions with Arshad Liaquat Chaudhry, Director General of the BISP Punjab Central Zone, during a field services, and social services. visit to Lahore were extremely helpful in expanding our understanding of how the BISP’s delivery systems function on the ground. The Sourcebook seeks to address concrete “how-to” questions, including: We would also like to thankHow do countries Dr. Mukhtar deliver Ahmad and social Sumaira protection Sagheer benefits for their and valuable services? time and insights. How do they do so effectively and efficiently? Howdiscussions The report also benefitted from with Sir do they ensure Baber and Muqueet Michael inclusion, dynamic Shahzad. especially for the most vulnerable and needy? How do they promote better coordination and integration—not only among social protec programs but also among programs in other parts of government? How can they meet the needs of their intended populations and provide a better client experience? The delivery systems framework elaborates on the key elements of that operating environment. framework is anchored in core implementation phases along the delivery chain. Key actors, inclu people and institutions, interact all along that delivery chain. Those interactions are facilitated b communications, information systems, and technology. This framework can apply to the delivery one or many programs and to the delivery of adaptive social protection. The Sourcebook structures itself around eight key principles that can frame the delivery systems min 1. There is no single blueprint for delivery systems, but there are commonalities, and those common elements constitute the core of the delivery systems framework. 2. Quality of implementation matters, and weaknesses in any of the core elements will nega affect the entire system, reducing the impacts of the program(s) they support. 1 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 3. Delivery systems evolve over time, in a nonlinear fashion, and their starting points matter ACRONYMS 4Ps Pantawid Pamilyang Pilipino Program NDEP National Data Exchange Platform APBS Aadhaar Payment Bridge System NDMA National Disaster Management Authority API Application Programming Interface NGO Non-Governmental Organization ASP Adaptive Social Protection NICOP National Identity Card for Overseas BBC BISP Beneficiary Committee Pakistani BDC BISP Debit Card NOC No objection certificate BI Business intelligence NPGP National Poverty Graduation Program BISP Benazir Income Support Programme NSER National Socio-economic Registry BTW Benazir Taleemi Wazaif (Education CCT) NSPS National Social Protection Strategy BVS Biometric Verification System OPM Oxford Policy Management CAPI Computer Assisted Personal Interviewing PAPI Pencil and Paper based personal CHM Complaints and Handling Mechanism interviewing CT Cash transfer PASS (Ministry of) Poverty Alleviation and Social CCT Conditional Cash Transfer Safety CMS Case Management System PBM Pakistan Bait-ul-Mal CNIC Computerized National Identity Card PCMS Payments Case Management System DG Director General PDMA Provincial Disaster Management Authority EPI Expanded Programme on Immunization PIN Personal Identification Number ETL Extract, transform and load PKLI Pakistan Kidney and Liver Institute and EWS Early Warning System Research Center FCDO Foreign Commonwealth and PKR Pakistan Rupee Development Office POS Point-of-sale FSP Food Support Program PLW Pregnant and Lactating Woman GDP Gross Domestic Product PMT Proxy Means Test GEWEL Girl’s Education and Women’s PPAF Pakistan Poverty Alleviation Fund Empowerment and Livelihood PPP Purchasing Power Parity G2P Government-to-person PSA Public service announcement GPS Global Positioning System PSP Payment Service Provider GRM Grievance redress mechanism PSPA Punjab Social Protection Authority HCI Human Capital Index QR Quick Response HIES Household Integrated Economic Survey RAAST Roshan Digital Account Asaan (easy) and ID4D Identification for Development Swift Transactions IPF Implementing partner firm SAM Severe Acute Malnutrition ISAS Integrated Social Assistance System SM Social mobilization KP Khyber Pakhtunkhwa SNF Specialized Nutritious Food KPI Key Performance Indicator SPJ Social Protection and Jobs LMA Limited Mandate Account TAT Turnaround time MAM Moderate Acute Malnutrition UCMS Unified Case Management System MIS Management information system UCT Unconditional Cash Transfer MPI Multi-dimensional Poverty Index UNITAR United Nations Institute for Training and NADRA National Database and Registration Research Authority USP Universal Social Protection NAF National Aid Fund WeT Waseela-e-Taleem NARC National Agricultural Research Center THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 2 The Sourcebook structures itself around eight key principles that can frame the d 1. There is no single blueprint for delivery systems, but there are commo common elements constitute the core of the delivery systems framew 2. Quality of implementation matters, and weaknesses in any of the core affect the entire system, reducing the impacts of the program(s) they s 3. Delivery systems evolve over time, in a nonlinear fashion, and their sta 4. Efforts should be made to “keep it simple” and to “do simple well,” from 5. The “first mile”—people’s direct interface with administrative functions link in the delivery chain; improving it may take systemic change but w efficiencies and mitigate the risk of failures on the frontlines. 6. Social protection programs do not operate in a vacuum, and thus their not be developed in silos; synergies across institutions and informatio and can improve program outcomes. 7. Social protection delivery systems can contribute more broadly to gov other sectors, such as health insurance subsidies, scholarships, social e benefits, and legal services. EXECUTIVE SUMMARY 8. The dual challenges of inclusion and coordination are pervasive and p the continuous improvement of delivery systems, through a dynamic, centered approach. Pakistan, the world’s fifth-most populous country, fuel costs, the government launched the Benazir is a lower-middle-income state with a per capita Income Support Programme (BISP) in 2008. BISP began GDP of US$1,505 (2021) (“GDP per Capita-Pakistan” as an unconditional cash transfer (UCT) program n.d.). While poverty has decreased significantly over (now known as Kafaalat) and has since become the the past two decades, one-quarter of the population country’s flagship social safety net to support the still lives below the poverty line, and these people poorest families in the country. In 2010, BISP was are disproportionately located in rural areas. The established as an autonomous institution, with COVID-19 pandemic worsened the situation for poor fiscal autonomy and support from all political parties, and vulnerable households, and the floods in 2022 which made its expansion possible. Its coverage has further pushed millions of people below the poverty significantly expanded over time, and two conditional line. Human capital accumulation is sluggish in cash transfers (CCTs) have been added with the aim of Pakistan, with the country’s low education and health encouraging families to invest in the human capital of indicators leading to low labor productivity. Stark their children. BISP now covers between 26 percent to inequalities in human capital accumulation between 30 percent of the population in each province. wealth quintiles and gender disparities exacerbate poverty and low human development indicators. The program has been successful in mitigating Pakistan is also among the countries that are most the negative impact of poverty on households as vulnerable to climate change-related disasters which well as protecting them against shocks. BISP has pose a continuous threat to lives and livelihoods. had significant positive impacts on consumption expenditure, food consumption, child nutrition Over the past 15 years, Pakistan has made significant security, asset retention, the mobility of women, strides in establishing and strengthening its social investments in health and education, and savings. protection system. Prior to the 2008 global financial To date, BISP has provided USD 1 billion in regular crisis, social safety net initiatives in Pakistan were and reliable support (World Bank n.d.). Between 2011 ineffective because of their non-objective targeting and 2019, the percent of BISP beneficiaries below and limited coverage. In response to rising food and the poverty line fell from 90 percent to 72 percent; 3 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs however, its impact on poverty diminished by 2019, With data and technology, BISP has constantly not be developed in silos; and can improve program primarily because of eroding value of benefit level. improved its delivery systems to be more accessible 7. Social protection delivery s other sectors, such as heal Investments in building systems over the years were and responsive to the needs of its target population. benefits, and legal services 8. The dual challenges of incl realized when the government was able to rapidly BISP relies on the National Socio-Economic Registry the continuous improveme centered approach. expand BISP, both vertically and horizontally, to (NSER) to identify and enroll eligible families. The NSER respond to the COVID-19 pandemic and the floods was updated in 2021 with majority of data collected in 2022. It is essential to continue investing in such through a door-to-door survey.1 BISP uses a biometric programs to reduce poverty and inequality in the verification system (BVS) when making payments. country. Beneficiaries receive their payments after being biometrically verified by their thumbprint through The BISP benefit level has been increased to ATMs or at point-of-sale (POS) agents. BVS is linked improve its adequacy, but it still accounts for only with the database of the National Database and 10 percent of average household consumption. The Registration Authority (NADRA), which enables real- value of BISP’s cash transfer was not increased until time verification at every step. 2017, causing a steady reduction in the adequacy of benefits. Although the benefit level has been The strength of the NSER is largely due to its increased on an ad hoc basis in the last few years, it operating ecosystem. Its collaboration with NADRA is still inadequate, especially given the recent drastic has been crucial to its progress. Both databases increase in fuel and food prices and the highest were major technological and capacity investments ever recorded inflation rate in March 2023. A recent to benefit the poor and vulnerable. This ecosystem mechanism to review the transfer value resulted in also supported BISP’s efforts to enhance its payments a 25 percent increase in the UCT amount, effective system by transitioning from cash to digital payments January 2023. The current value accounts for almost using BVS to ensure that identity is verified at every 10 percent of the average consumption; however, for stage. As the ecosystem in the country continues to BISP to have any impact on poverty, the benefit level evolve, BISP can enhance the way in which it assesses must at least be 15 percent of the average household the welfare status of households by complementing consumption. the NSER data with information from administrative databases of other institutions such as vehicle The Government of Pakistan introduced two administration, tax administration, and land records CCTs programs—Waseela-e-Taleem (WeT) and administration especially in the provinces. Nashonuma—to enhance investments in human capital. Launched in 2012, WeT aimed to increase Prior investments in technology enhanced the primary school enrollment rates and human capital ability of BISP’s delivery systems to respond to accumulation among poor children, especially shocks. BISP was globally recognized for its rapid girls. WeT has now been renamed the Benazir response to the pandemic using cash transfers. Taleemi Wazaif (BTW) program and has been Further, in the aftermath of the 2022 floods, BISP expanded to cover primary, secondary, and higher provided one-time emergency transfers to 1.76 million secondary education for children aged between 4 families. To identify beneficiaries, BISP used both NSER and 22 years who come from families that receive and big data analysis. By triangulating NSER data with the BISP UCT. Impact evaluations conducted in other sources—for example, using GPS data from the 2016 show significant positive impacts in terms NSER in combination with flood exposure data from of increasing school enrollment and reducing the United Nations Institute for Training and Research grade repetition, with a higher impact on girls (UNITAR—BISP) was able to precisely target those than on boys. The program has promoted school areas where families were most vulnerable to floods. enrollment, attendance, and graduation, with female students receiving stronger incentives than males. Efforts to further improve BISP’s delivery systems Subsequently in 2020, the government introduced are ongoing, with a focus on making them dynamic Nashonuma, a health and nutrition CCT that targets and even more responsive to shocks. To make pregnant and lactating mothers and children aged the NSER more dynamic, BISP has established between 0 and 24 months old from families that registration desks at the tehsil (subdistrict) level receive the BISP UCT. where households can register to be covered by the program and existing beneficiary households 1 The door-to-door survey was complemented with temporary desk-based data collection kiosks in 2020 and 2021. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 4 can update their information. Through a structured There are challenges to making the NSER more process which involves interim periodic steps, the dynamic and increasing its role in other social entire NSER database will be completely updated programs. To make the NSER more dynamic, BISP every four years, thereby eliminating the need for must actively seek out potential beneficiaries who complicated and costly door-to-door surveys in may have been excluded. Also, program managers the future. However, the NSER will conduct regular could use human-centered design techniques to surveys in disaster-prone areas identified by the understand how people interact with the system and National Disaster Management Authority (NDMA) to identify ways to improve the client experience. and the Provincial Disaster Management Agencies Streamlining and automating how BISP exchanges (PDMAs). In case of a disaster, the registry desks will data with partner organizations could make the serve as a platform for processing new registrations, process more efficient and enable the NSER to be recording losses, and re-evaluating the proxy means a dynamic registry. To ensure that both the NSER test (PMT) used to establish eligibility for emergency and NADRA stay current, policy makers could extend cash assistance offered by social safety nets or the integration of the NSER with NADRA to facilitate other public/private institutions. This approach was automated updates of beneficiaries’ changed successfully implemented during the 2022 floods. attributes, such as marital status or disability. Because However, the success of this approach depends the NSER is a valuable national asset, policy makers on the proximity of these data collection centers could develop a comprehensive national vision for to people, especially those living in remote areas. the registry so that it can increasingly deliver benefits Here, provincial local governance structures can be for programs other than BISP. leveraged to ensure that alternative measures can be put in place in areas where mobility and distance can Further, using business intelligence (BI) software limit access to these data collection sites. with data visualization and data analytics features could unleash the potential of the data held by Despite making progress in outreach and NADRA and the NSER and make them available communication, BISP has faced capacity challenges through a knowledge discovery platform. BI software that have hindered the development of an effective can be designed to collect and process large amounts communication strategy. Effective communication of unstructured data and prepare it for analysis with is critical for social protection systems and could be the aim of improving decision-making and increasing used at every stage of delivery. Several draft strategies operational efficiency. Provincial social protection produced by BISP have recognized communication as programs and disaster relief management authorities the institution’s weakest link. The communications could be given access to the knowledge discovery unit is understaffed and, as a result, community platform to drill down into specific geographic areas mobilization efforts such as BISP beneficiary and apply filters to socioeconomic attributes to committees (community-level groups of female analyze their impact. After thorough analysis, finalized beneficiaries) have not been sustainable because of a criteria could be used to extract data to be verified lack of capacity, purpose, and protocols. As BISP moves by a cloud-based management information system toward on-demand data updates and registration (MIS). BISP is developing a prototype for a cloud-based through the dynamic registry, outreach will become MIS for users of NSER data, which is expected to be even more critical, particularly to marginalized and piloted in collaboration with the provinces to assess remote populations. While mobile registration its effectiveness before a potential scale up. vehicles are currently being used to actively search for potential beneficiaries, more planning and resources BISP’s shift to a new payment system is expected to are required to scale up this effort. BISP could leverage improve beneficiaries’ experience and to support BTW’s compliance monitors to conduct outreach financial inclusion but needs to include a more and community activities on its behalf. It could comprehensive grievance redress mechanism. also explore involving local notables to increase With BISP’s transition to a new payment system, awareness of BISP’s programs. Communication efforts the number of participating financial institutions could also focus on informing beneficiaries of the is expected to increase. Consequently, procedures program’s grievance mechanism as there is currently related to grievance redress, funds reconciliation, and limited awareness of this among beneficiaries. other administrative activities may become more complicated, necessitating careful management 5 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s by BISP. Introducing a unified grievance redress the organization to: (a) identify weaknesses in its other sectors, such as heal mechanism for all of its sub-programs will significantly program implementation and to make the necessary benefits, and legal services 8. The dual challenges of incl improve its grievance redress management. At improvements. For example, if an evaluation revealed the continuous improveme centered approach. present, each cash transfer program has its own MIS that payment disbursements were being delayed in for receiving complaints, which is integrated with a certain area, BISP could take corrective action to the complaints systems in each participating bank. speed up the delivery of the payments; (b) increase The new mechanism is anticipated to streamline the transparency, foster citizen engagement, and ensure entire grievance redress process. Recognizing that that the program is publicly accountable; and (c) build an increase in the number of participating financial trust in the program among its beneficiaries and institutions may increase challenges with how stakeholders. Process evaluations and spot checks grievances are handled would help in the design of are best practices in program management, as they BISP’s unified case management system (UCMS). help to ensure that programs are being implemented as intended and that program resources are being As Pakistan’s social protection system expands, used efficiently and effectively. coordination between provincial and federal agencies will prevent fragmentation and duplication BISP’s commitment to monitoring and evaluation and ensure more effective management of BISP local (M&E) includes impact evaluations. These offices. Provincial governments are in the process evaluations are conducted by external firms, under of establishing social protection programs and the guidance and supervision of BISP’s M&E Wing and delivery systems. Seeking additional opportunities are focused on assessing the program’s effectiveness to collaborate with provinces at multiple stages of in achieving its intended outcomes. Four rounds of the delivery chain will help BISP ensure the efficient such evaluations have been conducted since BISP’s delivery of social protection across the country. inception, and an impact evaluation firm is presently Effective collaboration will be particularly vital in being hired to conduct additional rounds. As with the context of the BWT (education) and Nashonuma the process evaluations, the impact evaluations are (health) CCTs because the provinces are responsible conducted using a collaborative approach, with BISP for providing services that are critical to the more and its key stakeholders identifying the indicators to dynamic NSER becoming a reality. To address the be measured. The evaluations measure the program’s need for more on-the-ground presence to implement impact on beneficiaries’ well-being in both the short the CCTs and ensure a more dynamic NSER, BISP and long run as well as outputs and outcomes has increased its field staff (which has presented against defined benchmarks using the theory of a management challenge for BISP headquarters in change. These impact evaluations have enabled BISP Islamabad). to gather valuable insights into its effectiveness in supporting vulnerable populations. This information A series of process evaluations and spot checks has helped BISP identify areas for improvement has helped BISP improve its operations. Since and make necessary adjustments to the program. its inception in 2008, BISP has monitored the In addition, like the process evaluations and spot implementation of its sub-programs through regular checks, the impact evaluations have helped to build external operational reviews, impact assessments, confidence and trust in BISP among beneficiaries and beneficiary feedback surveys and various other third- stakeholders by providing transparent and objective party evaluations2. These evaluations have helped assessments of the program’s performance. 2 These included: (i) the NSER survey 2010-11 with regular process evaluation and spot checks conducted throughout the survey; (ii) the NSER update 2016-21 with regular process evaluation and spot checks conducted throughout the survey; (iii) an operational review of the education CCT and continuous beneficia- ry feedback since inception in 2012; (iv) an ongoing operational review of the health CCT; (iv) four rounds of impact evaluations of the BISP UCT using the same panel of households between 2010 and 2019; and (v) several years of external reviews and beneficiary surveys funded by the UK Foreign, Commonwealth, and Development Office (FCDO). All of the above produced quarterly and annual reports and even weekly reports when needed. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 6 common elements constitute the core of the delivery systems framewor 2. Quality of implementation matters, and weaknesses in any of the core el affect the entire system, reducing the impacts of the program(s) they sup 3. Delivery systems evolve over time, in a nonlinear fashion, and their starti 4. Efforts should be made to “keep it simple” and to “do simple well,” from th 5. The “first mile”—people’s direct interface with administrative functions—is link in the delivery chain; improving it may take systemic change but will efficiencies and mitigate the risk of failures on the frontlines. 6. Social protection programs do not operate in a vacuum, and thus their de not be developed in silos; synergies across institutions and information s and can improve program outcomes. 7. Social protection delivery systems can contribute more broadly to govern other sectors, such as health insurance subsidies, scholarships, social ene benefits, and legal services. 8. The dual challenges of inclusion and coordination are pervasive and pere the continuous improvement of delivery systems, through a dynamic, int centered approach. INTRODUCTION KEY MESSAGES The need for effective and timely social protection to protect the well-being of households before, during, and after the shock has been underscored by the COVID-19 pandemic and subsequent global crises. Countries with dynamic social protection delivery systems can quickly adapt and respond to the changing needs of population, thus enabling Adaptive Social Protection (ASP). Transitioning to ASP requires effective delivery systems which are also key to achieving universal social protection. The Government of Pakistan launched the Benazir Income Support Programme (BISP) in 2008 as an autonomous institution to deliver the government’s pro-poor social policy initiatives. Over the past 15 years, the government has made significant investments in its social protection system and has built strong legal and institutional arrangements, leveraged technology, and developed dynamic delivery systems that have collectively improved the effectiveness and efficiency of its programs. 7 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th Compounding global crises have highlighted Effective delivery systems will be essential to 6. Social protection programs not be developed in silos; the importance of providing effective social achieve universal social protection and ensure and can improve program 7. Social protection delivery s protection. Social protection can be defined as shock-responsive social protection systems. other sectors, such as heal benefits, and legal services policies and programs that help individuals and Achieving universal social protection (USP), which 8. The dual challenges of incl the continuous improveme societies to manage risk and volatility, protect them provides access to social protection for all whenever centered approach. from poverty and inequality, and help them access and however they need it, will be critical for economic opportunities. Social protection aims to effectively reducing poverty and boosting shared achieve this by increasing people’s resilience, equity, prosperity. Delivery systems constitute the operating and opportunity, the three goals of social protection, environment for implementing social protection through a range of instruments in the domains of benefits and services. In recent years, delivery systems social insurance, labor and economic inclusion, have improved significantly because of the rapid and social assistance and care. The unprecedented development and deployment of: (i) identification crisis caused by the COVID-19 pandemic, which systems, (ii) social registries, and (iii) payment systems. was immediately followed by the global shocks of All three are complementary and used beyond social the war in Ukraine and climate emergencies, has protection. Each has proven critical for improving reinforced the need for responsive social protection the quality of delivery of social protection programs. initiatives. It has also highlighted new and rapidly In many countries, social protection programs are evolving possibilities for designing and delivering not only the biggest users of these systems but have social protection that are more suited to responding also contributed to their widespread adoption (World to crises (World Bank 2022c). Bank 2022c). How countries make progress toward achieving USP will depend on political and economic The COVID-19 shutdowns across the world created factors along with the available fiscal space but, an urgent imperative to respond to the needs of in all cases, building effective delivery systems is a the poor and vulnerable, and many countries had prerequisite. For example, whether delivery systems to rely on existing social protection systems to do are static or dynamic has implications for how this. Across the world, countries introduced more than inclusive and responsive social protection programs 200 social protection measures (Gentilini et al. 2022), can be. The pandemic highlighted the importance of most of which leveraged existing systems but some dynamism to enable adaptive social protection (ASP) of which used technology to rapidly launch, scale up, systems, which will become increasingly necessary and deliver new programs. Since then, governments as countries try to support households in the face have responded to other socioeconomic shocks (such of threats from climate change and other crises. A as food price hikes, climate shocks, and fuel shortages) timely and adequate response is critical to protecting by providing cash transfers to millions of people, human capital before, during, and after a shock hits. including informal sector workers who are not the traditional beneficiaries of safety nets. According to The Government of Pakistan has made significant the World Bank tracker of social protection and jobs investments in its social protection system over the responses to COVID-19, 223 economies introduced past 15 years. The Benazir Income Support Programme social protection measures that compensated around (BISP) was launched in 2008 as an autonomous 1.3 billion people for income lost because of lockdowns institution to deliver the government’s pro-poor social and the economic downturn (Gentilini et al. 2022). policy initiatives. A key motivation for its creation was the recognition that publicly funded social protection It became evident that countries that had invested programs were limited in terms of adequacy, in their social protection systems were able to targeting efficiency, and their ability to respond respond to the pandemic crisis more quickly and to shocks. Complementary legal and institutional effectively. Examples include Jordan, Pakistan, and arrangements, openness to leveraging technology, and Türkiye, countries that had made prior investments in the development of dynamic delivery systems have their delivery systems. Since the frequency with which led to significant improvements in the effectiveness crises occur is accelerating, investing in strengthening and efficiency of its programs. BISP launched its first and building systems is more important than ever unconditional cash transfer (UCT) program, which was to increase countries’ preparedness to face various also called BISP at that time but is now referred to as shocks. While countries are still recovering from Kafaalat, to mitigate the adverse impact of food, fuel, COVID-19, the impact of the global inflation crisis and financial crises on the poor. Over the years, the highlights the continued importance of efficient number of beneficiaries has increased from 1.7 million social protection delivery systems to protect families to 9 million families. BISP has also introduced vulnerable groups. two conditional cash transfers (CCTs)— Waseela-e- THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 8 Taleem for education and Nashonuma for health—that programs, how they can be leveraged to promote prioritize human capital accumulation. better coordination and integration, and how social protection programs can provide a better client BISP has expanded its coverage and improved its experience for their intended populations. delivery systems, making it more accessible and responsive to the needs of its target population. This report documents the progress that Pakistan Delivery systems are a prerequisite for providing has made so far in improving its systems for social protection benefits and services and include delivering social protection to its people. The identification systems, social registries, payment government has increasingly relied on data systems, and management information systems. and technology to increase the efficiency and The government has increasingly used data and effectiveness of the program. BISP UCT (Kafaalat), technology to improve the efficiency and effectiveness the country’s largest social assistance program in of delivery systems, mainly through leveraging the terms of both budget allocation and number of Computerized National Identity Cards (CNICs) system beneficiaries, has been responsible for the most and linking it with the NSER and payment systems. innovative developments in the delivery of benefits. The NSER currently covers over 35 million families Its delivery systems have evolved significantly over and has data on household demographics and other time expanding in scope from simply delivering the socioeconomic indicators that are used to determine UCT to becoming a system that other programs can eligibility for benefits. The government has also started leverage to identify beneficiaries and deliver benefits. using BISP’s delivery systems to respond to economic It has flexibility to be scaled up, both horizontally and climate shocks, which has improved the pace of and vertically, in times of shock. This did not happen response. overnight: the government has consistently invested time and resources over the past decade and a half How countries deliver social protection has to improve how it functions. By documenting that implications for outcomes. Analyzing different social journey, using the Social Protection Delivery Chain protection delivery systems around the world can Framework developed by the World Bank in the provide answers to crucial questions such as how “Sourcebook on the Foundations of Social Protection various elements of delivery systems come together Delivery Systems,” (Lindert et al. 2020), this report to ensure that programs function as intended, can be a resource for domestic and international how they provide beneficiaries with easy access to stakeholders (Figure 1). FIGURE 1 SOCIAL PROTECTION DELIVERY CHAIN ASSESS ENROLL PROVIDE MANAGE Eligibility Determination Provision of Beneficiaries Exit Decisions, Assessment and of Benefits Notification Benefits Compliance, Notifications Intake & of Needs and Enrollment and Service and and/or Updating and and Case Outreach Registration Conditions Decisions Package Onboarding Services Grievances Outcomes RECURRING 1 2 3 4 5 6 7 CYCLE 8 9 PERIODIC RE-ASSESSMENT Source: Lindert et al. 2020 Note: The stages that are common to most programs include outreach, intake and registration, assessment of needs and conditions, eligibility and enrollment, payment of benefits and provision of services, and monitoring and management, including how beneficiaries exit the program. People and institutions interact all along the delivery chain. Those interactions are facilitated by several factors including communications, information systems, and technology. 9 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services The report contains an end-to-end review of the social assistance delivery chain 8. The dual challenges of incl the continuous improveme centered approach. including outreach, intake and registration, assessment of needs and conditions, eligibility and enrollment, payment of benefits, provision of services, and program monitoring and management, including how beneficiaries exit the program, since the inception of BISP to June 2023. THE REPORT HAS SIX SECTIONS INCLUDING THE INTRODUCTION. • SECTION 2 describes the socioeconomic context. • SECTION 3 outlines the history of social protection in Pakistan focusing on BISP (coverage, spending, results, institutional arrangements and role in adaptive social protection). • SECTION 4 assesses the BISP’s delivery chain. • SECTION 5 discusses its monitoring and evaluation practices. • SECTION 6 concludes and provides recommendations to the Government of Pakistan on how to further improve the delivery of benefits to those most in need. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 10 common elements constitute the core of the delivery systems framewor 2. Quality of implementation matters, and weaknesses in any of the core el affect the entire system, reducing the impacts of the program(s) they sup 3. Delivery systems evolve over time, in a nonlinear fashion, and their starti 4. Efforts should be made to “keep it simple” and to “do simple well,” from th 5. The “first mile”—people’s direct interface with administrative functions—is link in the delivery chain; improving it may take systemic change but will efficiencies and mitigate the risk of failures on the frontlines. 6. Social protection programs do not operate in a vacuum, and thus their de not be developed in silos; synergies across institutions and information s and can improve program outcomes. 7. Social protection delivery systems can contribute more broadly to govern other sectors, such as health insurance subsidies, scholarships, social ene benefits, and legal services. THE SOCIOECONOMIC 8. The dual challenges of inclusion and coordination are pervasive and pere the continuous improvement of delivery systems, through a dynamic, int CONTEXT OF PAKISTANcentered approach. KEY MESSAGES Pakistan had made significant gains in poverty reduction; however, recent shocks, including the COVID-19 pandemic and the floods of 2022, have significantly reversed those gains. Pakistan is behind its regional counterparts in human capital accumulation and has huge gender disparities. This, combined with an increasing incidence of climate shocks, poses a continuous threat to the lives and livelihoods of the poor and vulnerable. Over the past two decades, poverty rates declined A narrow export base coupled with weak production in Pakistan; however, the impact of the COVID-19 has prevented the country from capitalizing on rich pandemic reversed that trend and inconsistent resources. Nevertheless, between 2001 and 2018, over economic growth, recent natural disasters and 47 million people rose out of poverty (World Bank surging inflation pose obstacles to future progress. n.d.). However, the pandemic pushed two million Pakistan, the world’s fifth-most populous country, is a people below the poverty line (World Bank 2021a) lower middle-income country with a per capita GDP of and the floods in 2022 pushed an estimated 8.4 US$1,505 in 2021. While the promise of rapid sustained million to 9.1 million people into poverty (World Bank economic growth has always existed, Pakistan has 2022b). The current surge in food and transport prices exhibited a cyclical boom-and-bust pattern. brought the annual rate of consumer price inflation in March 2023 to 35.37 percent, the highest ever in 11 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme Pakistan (Asif 2023), making even basic necessities Asia average of 48 percent and comparable to Sub- centered approach. unaffordable for millions of people. This situation will Saharan Africa with an average of 40 percent (Table be exacerbated by the country’s ongoing balance of 1) The country has the second highest number of payments crisis. out-of-school children in the world (UNICEF 2019). A consequence of such low human capital is that Pakistan lags its regional counterparts on human Pakistan’s labor productivity lags that of its peers capital accumulation. In 2021, a child born in Pakistan in South Asia (Pakistan Bureau of Statistics 2019). was forecast to be only 41 percent as productive National education and health indicators, including when she grew up as she could have been if she had school enrollment, adult literacy, and child mortality, had a complete education and enjoyed full health are worse than in other countries with similar levels (World Bank 2020a). This rate is lower than the South of economic development. TABLE 1 HUMAN CAPITAL INDICATORS BY REGION Middle Europe & Latin East & Sub- INDICATOR East Asia Central America & North North South Saharan & Pacific Asia Caribbean Africa America Asia Africa Pakistan HCI Component 1: Survival Probability of survival to age 5 0.98 0.99 0.98 0.98 0.99 0.96 0.93 0.93 HCI Component 2: Education Expected Years of School 11.9 13.1 12.1 11.6 13.3 10.8 8.2 9.4 Harmonized test scores 432 479 405 407 523 374 374 339 HCI Component 2: Health Survival rate, from age 15 to 60 0.86 0.9 0.86 0.91 0.92 0.84 0.74 0.85 Fraction of children under 5 0.76 0.9 0.85 0.82 - 0.69 0.69 0.62 not stunted Human Capital Index (HCI) 0.59 0.69 0.56 0.57 0.75 0.48 0.4 0.41 Source: Ersado et al. 2023 There are stark inequalities in human capital in the poorest households (Agha and Sohail 2016). accumulation with long-term negative consequences Among those in the richest two quintiles, Pakistan’s for poor households. Figure 2 on the following page level on the HCI is around the low-middle-income shows that that the gap in human capital outcomes country average of 0.48. However, the HCI level among between households in the richest two quintiles and households in the poorest two quintiles is well below households in the poorest two quintiles has grown in the Sub-Saharan African average. With low levels of the past 30 years. For example, research shows that productivity and lifelong earnings, poor households Pakistan has raised its overall Human Capital Index can remain stuck in intergenerational poverty. (HCI) level by investing more in richer rather than THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 12 FIGURE 2 PERSISTENT HUMAN CAPITAL INDEX GAP BETWEEN THE RICHEST AND POOREST GROUP OVER DECADES High Income Countries: 0.70 (Year 2020) 0.50 0.49 Low Middle Income Countries: 0.48 (Year 2020) Richest: 0.44 0.46 0.42 Sub-Saharan Regions: 0.40 (Year 2020) 0.37 Richer: 0.32 0.33 0.31 Poorer: 0.28 Poorest: 0.26 0.28 1990 Source: Ersado et al. 2023 2012 2017 Note: Calculations are based on Demographic and Health Surveys (1990-2017) YEAR Gender disparities reinforce poverty and low human Besides the lingering impacts of COVID-19 and development indicators. Pakistan ranked third-from- the intensifying impacts of higher prices, climatic last on the World Economic Forum’s 2020 Global shocks pose a continuous threat to lives and Gender Gap Index. Lower levels of female education livelihoods. Pakistan is among the top 10 countries and literacy, poor access to health services, and social most vulnerable to climate change including floods, and cultural practices that limit female labor force droughts, and heatwaves which endanger the lives participation (FLFP) leave female-headed households and livelihoods of millions of people each year. These particularly at risk. Pakistan’s FLFP rate of 21 percent is crises necessitate continued investments in expanding less than half the average of 46 percent for the rest social protection systems and reducing inefficiencies of the world (“Labor Force Participation Rate Female- and fragmentation. Strong social protection delivery Pakistan” n.d.). The COVID-19 pandemic compounded systems such as IDs, social registries, payment these disparities by limiting the fiscal space available systems, and management information systems (MIS) to close the infrastructure and social sector deficits can enhance the effectiveness and efficiency of social that inherently affect women and girls more than men. protection programs in responding to shocks both ex post and ex ante. 13 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan not be developed in silos; synergies across institutions and informat and can improve program outcomes. 7. Social protection delivery systems can contribute more broadly to go other sectors, such as health insurance subsidies, scholarships, socia benefits, and legal services. 8. The dual challenges of inclusion and coordination are pervasive and the continuous improvement of delivery systems, through a dynami centered approach. SOCIAL PROTECTION IN PAKISTAN: FROM THE PAST TO THE PRESENT KEY MESSAGES Pakistan’s early social safety nets schemes (Pakistan Bait-ul-Mal and Zakat) were weakly targeted, had limited coverage, and were mired in bureaucratic and administrative inefficiencies. In response to the global financial crisis in 2008, the government launched the Benazir Income Support Programme (BISP) and, in 2010, the government established BISP as an autonomous national institution responsible for delivering its pro-poor social policy. Financial autonomy, administrative flexibility, and transparent accountability mechanisms have enabled BISP to make large investments in its programs as well as its delivery systems. Coverage of BISP Kafaalat has expanded from 1.76 million families to 9.2 million families while budget allocation to BISP increased from PKR 34 billion in 2008 to PKR 360 billion in 2022. BISP’s delivery systems rely on the National Socio-Economic Registry (NSER), Computerized National Identity Cards, and Biometric Verification System-based payments. Launched in 2010, the NSER has become the backbone of Pakistan’s social protection delivery systems and is currently used by more than 30 programs and agencies. The NSER collects demographic and socioeconomic information on each household which is used to calculate a poverty score based on a Proxy Means Test (PMT). The poverty score serves as an objective criterion to identify beneficiaries. The NSER is moving toward an on-demand registration system, which will ensure a complete update every four years. BISP recently established a steering committee for the NSER with representation from all provinces, which will strengthen data-sharing among entities while also improving data quality and use. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 14 Historically, social protection programs in Pakistan Early social safety net programs in Pakistan had focused on social security in the formal sector. limited reach and effectiveness. Still in operation, Social security schemes were first introduced in Zakat & Ushr, established in 1980, and Pakistan Bait- the 1950s and included the Government Servants’ U-Mal (PBM), established in 1991, are the earliest social Pensions-cum-Gratuity Scheme (1954), the Public protection programs in the country. Zakat is financed Sector Benevolent Fund and Group Insurance Scheme by voluntary deductions from the commercial bank (1969), the Employees’ Social Security Scheme (1970), accounts of citizens and corporations. Local Zakat Workers’ Welfare Funds (1971), and the Employees’ committees, established under the umbrella of Old Age Benefits Institution (1976). These programs the Ministry of Religious Affairs, identify the most are limited to employees in the formal sector, even needy and marginalized Muslims through informal though the informal sector represents 81 percent mechanisms such as consultations with community of total employment in Pakistan, including most leaders. In 2007, it supported around 1.6 million workers from the bottom quintiles. As a result, families (World Bank 2007). The Food Support Program majority of workers, especially those who are poor (FSP) run by PBM used to be the main government- and vulnerable, are excluded from such protection. run cash transfer program. Implemented by the Ministry of Social Welfare and Special Education, With the Development Policy Framework (DPF) it provides a fixed amount to poor families to 2005–2010, social protection for the poor became support their basic needs. District-level committees a policy objective for the government. The DPF comprised of governmental and non-governmental included provision of cash and non-cash support, representatives identify and select PBM participants protection from shocks, and investment in human using informal and community-based targeting capital as key tools to support the poor and methods. The welfare status of the beneficiaries is vulnerable. In 2007, the government adopted the not independently verified. National Social Protection Strategy (NSPS) to develop the foundations of an integrated and comprehensive The financial crisis of 2008 revealed the limitations social protection system that would: (i) cover of Zakat & Ushr and PBM. Both programs were weakly everyone in need, especially the poorest and the targeted, lacked transparency and accountability, had most vulnerable; (ii) promote household investments limited coverage, and were mired in bureaucratic and in human and physical assets, including health, administrative efficiencies. Both programs delivered nutrition, and education; and (iii) protect them from cash manually, which was slow, inefficient, and lacked shocks by increasing their resilience and breaking transparency. Since most benefits were one-off the intergenerational cycle of poverty (“Government transfers, they did not provide recipients with regular, of Pakistan Planning Commission Annual Plan 2007- consistent support. 08” 2007). The strategy also reviewed existing social protection initiatives concluding that they were largely ad hoc with multiple duplicate and overlapping programs (Government of Pakistan, 2007). The strategy prioritized households with female heads as well as those with many children and recognized the needs of marginalized populations. 15 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services 3.1 THE LAUNCH AND EXPANSION OF THE BENAZIR INCOME 8. The dual challenges of incl the continuous improveme centered approach. SUPPORT PROGRAMME In 2008, following a sharp rise in fuel and food BISP4 has always been led by a chairperson, usually prices, the government launched BISP as its a federal or state minister, who provides strategic flagship social safety net program to respond to direction. As a minister, BISP’s Chairperson is part of the shrinking purchasing power of poor households the federal Cabinet, which keeps social protection with a UCT of PKR 1,000 per month.3 BISP showed firmly on the agenda of the highest policymaking that the government was committed to designing forum in the country and in national discourse. The a safety nets program based on international best main functions of the Board are: (i) approving the practices and leveraging technology for increased eligibility criteria for receiving financial assistance; (ii) efficiency, transparency, and accountability. The approving BISP policies, regulations, and budget; and program fell short of its initial plan to reach 3 million (iii) ensuring transparency by monitoring the program. of the poorest families in Pakistan but was able to The Chairperson heads the Board and is expected reach 1.74 million families. to provide strategic guidance and ensure the implementation of the program in accordance with In 2010, the government established BISP as an the Board’s decisions. The BISP Secretary heads the autonomous institution to serve as a vehicle management team, which includes several Director to deliver its pro-poor social policy through Generals (DGs) at the national headquarters as well as redistribution and targeted support to the poor. in the provinces and is the principal accounting officer The BISP Act 2010 established BISP as an institution of the program. Both the Chairperson and Secretary responsible to “enhance financial capacity of the are appointed by the federal government with the poor; formulate and implement comprehensive approval of the Prime Minister. policies and targeted programs for the uplift of underprivileged and vulnerable people; and reduce Over time, the budget and coverage of BISP Kafaalat poverty and promote equitable redistribution of expanded reflecting a policy shift away from wealth” (Ministry of Law and Justice, n.d.). The Act reactive measures and toward providing long-term created an enabling environment through a unique safety nets. Providing small, frequent, and reliable governance structure that provided BISP with: the cash transfers to poor households not only mitigates freedom to hire from the private sector wherever their current poverty but also leads them to make specialized skills were needed; a budget line item in more investments in their human capital (such as the national budget that provided it with financial health and education), which can help them to reduce autonomy and sustainability; an independent poverty in the future (Loeser et al. 2021). In line with Management Board with 50 percent members the NSPS, the government significantly expanded the from the non-government sector to guide the budget and coverage of the BISP (Figure 3). By 2022, the strategic direction; and transparent accountability budget allocation had increased to PKR 360 billion or arrangements where the internal audit function 0.58 percent of GDP. BISP coverage increased from 1.74 reports directly to the Board. In 2019, the government million families in 2008 to 9 million families in 2023. established the Poverty Alleviation and Social Safety (PASS) Division and brought all federal and provincial social protection programs, including the BISP, under its authority. 3 At the time of its launch, the UCT was referred to as BISP but is now called Kafaalat 4 From here on, ‘BISP’ will refer to the institution. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 16 FIGURE 3 THE BISP’S BUDGET ALLOCATION AND NUMBER OF BENEFICIARIES a. BISP allocation as percent of GDP b. Number of BISP beneficiary families, millions 9 0.52% 7.9 0.47% 5.23 5.37 5.07 5.21 4.92 5.02 5.06 4.65 4.33 4.36 0.35% 0.35% 0.35% 0.31% 0.30% 0.31% 0.32% 3.1 2.58 0.26% 0.27% 0.25% 0.27% 0.26% 1.76 2009 2012 2014 2016 2018 2020 2022 2008 2010 2012 2014 2016 2018 2020 2022 YEAR YEAR Note: 2022 GDP projected numbers from IMF Pakistan Staff Report Source: BISP Administrative Data Note: 2022 GDP projected numbers from IMF Pakistan Staff Report FIGURE 4 BISP’S ANNUAL BUDGET ALLOCATION (PKR BILLIONS) BUDGET ALLOCATION (PKR), BILLIONS 246,06 235,125 194,325 102 115 106,7 116 97 70 70 75 50 50 34 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Source: BISP Administrative Data 17 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program Currently, the beneficiaries of the BISP’s unconditional with higher levels of poverty have a higher proportion 7. Social protection delivery s other sectors, such as heal transfer are spread across all provinces and regions of BISP beneficiaries. This indicates that potential benefits, and legal services 8. The dual challenges of incl of Pakistan, but there are considerable geographical beneficiaries in remote areas of the country are being the continuous improveme centered approach. variations in coverage, which indicate potential missed by the program. An ongoing assessment of exclusion errors. A closer look at the distribution of the NSER focuses on finding systematic patterns in beneficiaries (Figure 5B) and the prevalence of poverty exclusion of population from the registry, which will across districts (Figure 5A) reveals that coverage in help reduce these coverage gaps. Balochistan; however, other provinces with districts FIGURE 5A FIGURE 5B POVERTY RATES BY DISTRICT BISP BENEFICIARIES BY DISTRICT Source: Authors’ calculations based on HIES 2018-19 and NSER data In 2012, BISP introduced WeT, a CCT program BTW is designed to give households an incentive designed to encourage beneficiaries of BISP Kafaalat to ensure school enrollment, attendance, and to invest in their own human capital. Initially, WeT graduation of their children, especially girls. Since covered primary education for children, from BISP more girls (48 percent) are out of school than boys (35 Kafaalat beneficiary families, between ages 4-12 percent), the BTW incentive for sending girls to school years. It now covers primary, secondary, and higher is higher than that for sending boys. For boys who education for children and young people between are enrolled and have 70 percent school attendance the ages of 4-22 years in 159 of Pakistan’s 170 districts. each quarter, the program pays the household PKR WeT was renamed the Benazir Taleemi Wazaif (BTW) 1,500, while for girls who achieve the same goals, the program after it expanded beyond the primary level. program pays PKR 2,000 (“Benazir Taleemi Wazaif” An impact evaluation of WeT (2016) found that it had n.d.). Families are provided with a one-time bonus raised school enrollment and attendance with a of PKR 3,000 when girls complete primary education 10-percentage point increase in enrolment rates for and transition to secondary school. The number of both girls and boys (Oxford Policy Management 2016). enrolled children since the inception of the program THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 18 is 10 million, including 8.4 million primary school growth is monitored while the mother participates students. In 2023, there were 7.8 million actively in awareness sessions covering topics like hygiene, enrolled children (World Bank 2021c and BISP breast-feeding, dietary diversity, immunizations, and Administrative Data). However, nearly 21.5 million antenatal and postnatal care. By making cash transfers children (between the ages of 4 and 18) remain out- conditional upon participation in these awareness of-school children, contributing to low human capital campaigns, it is expected that the women’s increased accumulation, particularly among the poor, girls, and knowledge of nutrition will influence the quality and the rural population. quantity of the foods that their families consume, leading to lower stunting rates among their children In 2020, the government introduced Nashonuma, a (Field and Maffioli 2021). health and nutrition CCT, in 14 districts. The stunting rate in Pakistan is 37.6 percent among children under The proven positive impact that the BISP Kafaalat 5 years of age; this is higher than the stunting rate and CCTs have had on improving household welfare of neighboring countries, Bangladesh (28 percent), and other indicators of well-being provides a solid and India (35.5 percent) (World Bank, “Prevalence of foundation for further expansion of these programs, Stunting”). Since the first 1,000 days are the most but coverage remains limited. Although there is important developmental phase in a person’s a common belief that when people receive cash life, Nashonuma targets pregnant and lactating support, they stop working and become dependent mothers and children aged 0 to 24 months among on welfare, Ambler and de Brauw (2019) found that beneficiary families receiving the BISP UCT. Every receiving the BISP transfers had no aggregate impact enrolled beneficiary mother and/or child can receive on household labor supply. In fact, they found an a maximum of 11 quarterly payments. The quarterly increase in labor force participation among men in benefit amount is PKR 2,000 for pregnant and beneficiary households. Other studies have shown lactating mothers and mothers of male children in the that while the transfers have had a limited impact target age group and PKR 2,500 for mothers of female on reducing poverty, they have had a significant children in that age group (“Benazir Nashonuma” n.d.). impact on reducing inequality and increasing food Beneficiaries also receive specialized nutritious food. At security and women’s empowerment (Box 1). While each visit to one of the program’s facilitation centers the government has increased the coverage of the located in tehsil-level health facilities, the child’s BISP Kafaalat and CCTs, it stands at 19.8 percent of 19 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s the population,5 much lower than other countries only able to complete it in 2021. More recently, other sectors, such as heal in the region such as Bangladesh and Nepal where BISP has established 647 facilitation centers across benefits, and legal services 8. The dual challenges of incl social assistance covers around 40 percent of the Pakistan where people can register and update their the continuous improveme centered approach. population (Figure 6). information in NSER. Both the NSER and the BVS- based payments system are linked with the NADRA BISP’s delivery systems have improved over time, database to ensure that individuals are verified at making it more accessible and responsive to the every step, from registration and enrollment to the needs of its target population. Delivery systems delivery of the payments. are the prerequisite for providing social protection benefits and services. To identify and enroll eligible families, BISP relies heavily on the NSER. After remaining static for five years, the NSER started its update in 2016 but due to multiple delays, was 5 Calculated using the total number of BISP beneficiaries (9 million families, with the average family size in the NSER being 5.2) and Pakistan’s total population (235 million people) FIGURE 6 PERCENTAGE OF TOTAL POPULATION COVERED BY SOCIAL ASSISTANCE 40.37% 40.14% 29.15% 28.13% 19.84% Bangladesh Indonesia Nepal Pakistan Sri Lanka Source: ASPIRE, World Bank Note: Numbers for all countries except Pakistan are from World Bank’s ASPIRE database using data from the latest year available. Coverage in Pakistan is calculated using BISP’s current number of beneficiaries (9 million) THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 20 BOX 1 THE IMPACT OF THE BENAZIR INCOME SUPPORT PROGRAMME ON POVERTY AND HOUSEHOLD WELFARE A 2016 impact evaluation of BISP concluded that it had had a significant impact on increasing consumption expenditure, food consumption, child nutrition security, asset retention, mobility of women, investments in health and education, and savings among its beneficiary households. The evaluation also found that it did not have a significant impact on poverty rate but reduced the poverty gap by 3 percentage points (Cheema et al, 2016). These results are consistent with a more recent study from 2020 that used panel data from 2013, 2014, 2016, and 2019. This study found that BISP beneficiaries experienced a reduction in poverty only during the first three years of intervention (2011-2014) when around 25 percent of households graduated from “ultra-poor” to “poor” and “vulnerable” categories. After that, their poverty rates remained more or less consistent. The analysis also looked at the multidimensional poverty index (MPI) and found that the proportion of MPI poor decreased in every round. Of the sample, 91 percent were MPI poor in 2011 but by 2019, this had decreased to 73 percent. However, when using a regression discontinuity model on 2019 data, no significant impact was found on headcount poverty or MPI. A possible reason for this limited impact on poverty is the declining real value of the benefit over time. At the time of BISP’s launch in 2008, the monthly transfer value was PKR 1,000 (PKR 3,000 per quarter), which, despite inflation and other changes in the economy, remained unchanged until 2018. In that year, the BISP FIGURE 7 CHANGE IN REAL VALUE OF BENEFIT LEVEL 2333 Nominal Real 2000 1667 1000 997 2008 2010 2012 2014 2016 2018 2020 2022 Source: Authors’ calculations based on BISP administrative data 21 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program CHANGE IN ADEQUACY OF BISP TRANSFER 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme centered approach. benefit level was increased by 66 percent, bringing it up to PKR 5,000 per quarter (PKR 1,667 per month). At the same time, the adequacy of the benefit (defined as its share of the average monthly consumption of households in the lowest quintile) had also steadily gone down from 9.8 percent in 2008 to 4.2 percent in 2017, though it was increased by 6.6 percent by the increase in 2018. However, Nayab and Farooq (2020) concluded that this raise was not sufficient and for the BISP UCT to have any impact on poverty reduction, the benefit value should cover at least 15 percent of the total consumption of these households. FIGURE 8 CHANGE IN ADEQUACY OF BISP TRANSFER UCT nominal value 9.85% UCT as share of average consumption among 20% poorest 8.64% 7.69% 6.64% 6.54% 5.85% 5.29% 6.64% 1667 4.98% 4.7% 4.43% 4.2% 1000 2008 2010 2012 2014 2016 2018 Source: Authors’ calculations based on HIES 2011, 2014, and 2019. Note: Adequacy is defined as the total transfer amount received by all beneficiaries in a quintile as a share of the total welfare of beneficiaries in that quintile. The government has subsequently increased the benefit level, and it currently stands at PKR 8,750 per quarter. This brings the real value to the same level as it was in 2008. The increases have been ad hoc given that the BISP has no indexation policy in place. With the drastic increase in fuel and food prices since 2022 and the highest ever recorded inflation in March 2023 (35.37 percent), there is a need to adjust the benefit level so that households can maintain their levels of consumption. However, it should be also noted that almost 40 percent of Kafaalat beneficiaries receive either BTW or Nashonuma transfer, where a family on average receives around PKR 5,000, with significant variation depending on the number of eligible children. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 22 3.2 THE LAUNCH AND EXPANSION OF THE NATIONAL SOCIO-ECONOMIC REGISTRY Launched in 2010, the NSER started as a targeting (Box 2), which is then used as a proxy for a household’s tool for BISP Kafaalat but is now the backbone of socioeconomic status. Over time, with investments in Pakistan’s social protection delivery system with human resource and technology and innovation, the multiple other programs and institutions, including NSER’s utility has increased significantly and, currently, disaster response initiatives, leveraging this data for more than 30 programs and institutions rely on this targeting. It is a comprehensive database designed data for identification and targeting. It has also been to capture and consolidate information on the socio- used to respond to disasters such as COVID-19 and economic status of individuals and households and the floods of 2022. covers around 87 percent of the country’s population. Using these indicators, it calculates a poverty score BOX 2 POVERTY SCORE CARD BASED ON POVERTY MEANS TEST The compilation of a poverty score card relies on the use of a proxy means test (PMT), which measures a small set of variables that capture those individual and household characteristics that are comparatively easy to verify. Collectively, these variables can proxy for household welfare. Although a PMT is not as accurate as a full consumption aggregate in measuring poverty, it requires considerably less information, which makes it easy to collect and verify. It is also less susceptible to short-term fluctuations as it relies on asset and household information, such as household size and house or land ownership, which tend to be fairly static. In Pakistan, household PMT scores were calculated using variables that had been used in the questionnaire for the 2007-2008 Household Integrated Economic Survey (HIES) collected by the Pakistan Bureau of Statistics. A questionnaire was designed in Urdu for the door-to-door survey in 2010, which collected household data on 23 selected variables in the following broad categories: (i) household and individual characteristics; (ii) ownership of durable goods and housing characteristics; and (iii) ownership of productive assets, especially land holding, livestock, and farm equipment. The PMT formula was then applied to new data on these variables to devise a poverty score for each household. BISP first collaborated with NADRA in 2010-2011 Besides offering information on their various to collect household data through a door-to- demographic and social indicators, individuals door survey that was called the Poverty Scorecard must have a CNIC to register with the NSER (see Targeting Survey. BISPs partnership with NADRA was Annex 1 for database design). The first time these critical for the intake and registration process during data were collected, they were to be used only for 2010-2011. The Poverty Scorecard Targeting Survey the BISP Kafaalat, so each household was registered was launched in mid-2010 in all districts of the country with the ever-married female as the contact person. and took slightly over a year to complete. Data were To register, the ever-married female had to have a collected using pencil and paper-based personal CNIC, and she was encouraged to enter at least one interviewing (PAPI). The collected data were passed additional CNIC from her family, in most cases, her on to NADRA, which was responsible for data entry husband’s. Each ever-married female represents one and verification, and calculation of poverty scores. family within a household, so a household can have Approximately 27 million households were surveyed, more than one registered family. Each entry is linked out of which 25.5 million went on to be registered in to a cell phone number used by BISP to communicate the database, constituting around 87 percent of the with the household. Recently, BISP has required population (Box 3). Each household was assigned a beneficiaries to give their bank account information poverty score based on the PMT formula. for payment purposes. 23 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program BOX 3 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme AN EXEMPLARY PARTNERSHIP: centered approach. THE COLLABORATION BETWEEN BISP AND NADRA The collaboration between BISP and NADRA has been crucial for the effective implementation of the program since 2008. In the initial stages, NADRA’s database helped BISP to verify the identities of potential beneficiaries. This helped to ensure that the program was effectively targeted toward the intended beneficiaries and reduced the potential for fraud or mismanagement of funds. NADRA also provided crucial technical assistance by supporting the development of the program’s MIS, which was instrumental in managing and monitoring data related to BISP beneficiaries, payments, and other program activities. This collaboration enabled the BISP to effectively manage program data, resulting in increased transparency and accountability in its operations. The two organizations closely collaborate to ensure that NSER is dynamic. They recently launched a desk-based data collection approach to register households that were not included in the 2018-19 NSER survey and updated household roster information, and corrected information discrepancies in households such as missing CNICs. The collaboration with NADRA has made BISP’s database more accurate and complete, ensuring that eligible households are not left out of the program. The collaboration between BISP and NADRA has been mutually beneficial. NADRA had struggled to increase the number of registered women because of social and cultural barriers and a lack of awareness of the benefits of registering with the database. (CONTINUED ON NEXT PAGE) FIGURE 9 NSER-NADRA WORKFLOW Missing Registration Discrepancy in NSER data Roster Update Citizen Tehsil Desk Office NSER DB Identity confirmation Gender/DOB Status Confirmation Family composition Data in English Marital status Disability Online NADRA APIs THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 24 BOX 3 (Continued) However, BISP required beneficiaries who did not have CNICs to obtain them within a specified period to continue receiving payments. This incentivized millions of women to acquire CNICs, which, in turn, helped NADRA to expand its database. Overall, the partnership between BISP and NADRA is an excellent example of how collaboration between government agencies can have a significant positive impact on society. It has proved to be beneficial for both organizations as it has helped BISP to be transparent and accountable in its operations and enabled NADRA to expand its database to include millions of women who previously did not have a legal identity. Having a legal identity is crucial for women’s empowerment and inclusion in society as it allows them to access various services and opportunities. Even though the survey suffered from a few initial technology which made the process more efficient, design and implementation challenges, spot BISP faced significant challenges in completing the checks found it to be robust. Challenges included: survey. Computer-assisted personal interviewing (i) a disproportionately higher representation of the (CAPI) was used to collect data which significantly rural poor; (ii) outdated “weights”6 since the PMT reduced the data processing time. However, there formula was based on HIES 2007-08, and conditions were multiple setbacks with the data collection had changed; (iii) variations in the distribution of process. Besides the COVID-19 pandemic, a key poverty across regions compared to HIES poverty challenge was that to collect data in districts with estimates; and (iv) the need for more quality control security concerns, BISP was required to obtain a No- questions to reduce the incidence of misreporting Objection-Certificate for their private contractors. This and fraud. To ensure that the data collected by each process took a long time. BISP devised an alternative firm was high quality, BISP hired a firm to conduct spot strategy: data were collected in those districts by checks to verify the data on a sample of over 67,000 government schoolteachers who were trained, households. The spot checks confirmed that the data monitored, and supervised by BISP staff. Census collection methods used by the contracted firms were sweep surveys are time-consuming particularly in robust and estimated the accuracy of the poverty highly populated countries like Pakistan with hard- scorecard survey to be 95 percent.7 to-reach regions. It took until 2021 to complete the survey and about six months to clean the data before Since proxies for poverty and their relative weights they were available to be used. The updated NSER, change over time, the PMT formula was revised in covering 35 million households or 87 percent of the 2015 (using variables from HIES 2013-14) to improve population, was formally launched in January 2022. targeting. The proxy indicators and weights were updated based on the correlation between the In parallel, a desk-based data collection approach indicators and household consumption. This step was launched in collaboration with NADRA to was aimed at increasing the coverage of poor in urban register households that had been missed in the areas, as the poverty rates based on PMT scores had door-to-door sweep of the NSER update. Centers underestimated poverty in urban areas and over- catered for: (i) registration of households missed in represented those in rural areas. Another indicator the door-to-door survey, (ii) update of household was added to capture location effects based on roster information; and (iii) update of information the taxonomy of the National Agricultural Research of households marked with discrepant information, Center (NARC), which classified districts into seven such as missing CNICs. These centers operated for agroclimatic zones based on their geographical five months (120 working days) after which one or proximity and similarities in climate. two centers per tehsil remained functional for a year to continue the update process. Most of them have Concerned that the NSER data were outdated, BISP now become permanent. Around 11 percent of total initiated a second national door-to-door survey families in the NSER registered via desk-based data in 2016. While the second survey benefited from collection centers. 6 These weights quantify the impact of each variable on household welfare 7 International Development Strategies, n.d. 25 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs The 2016-2021 survey was the last nationwide door- composition. The shift toward having a dynamic NSER not be developed in silos; and can improve program to-door survey as the BISP is gradually moving to an is expected to: (i) reduce inclusion and exclusion 7. Social protection delivery s on-demand registration system which will ensure errors more efficiently, by enabling households to other sectors, such as heal benefits, and legal services dynamic updates of the NSER and will significantly regularly update their information and (ii) save time 8. The dual challenges of incl the continuous improveme reduce the cost of data collection. Currently, there and financial resources.8 However, success depends centered approach. are 647 dynamic registration centers – at least one on accessibility of these centers as people living in in each tehsil – where people can register in the remote areas as well as the poorest of the poor have NSER if they were left out during the door-to-door significant mobility constraints. As the BISP phases in survey. However, already registered households will the dynamic update of the NSER, it can learn valuable also be allowed to update their information if their lessons from countries like Brazil, Türkiye, and Chile data was last updated more than three years ago or that have already established more dynamic social if there is any significant change in their household registries (Box 4). 8 The cost per survey in 2011 was around US$2.7 whereas during the update, it was around US$1.7. BOX 4 MOVING TOWARD A DYNAMIC SOCIAL REGISTRY: LESSONS FROM OTHER COUNTRIES Social registries like the NSER can be powerful tools for improving the targeting, coverage, integration, and operations of social programs. However, these improvements depend on the quality of the data, which is a function not only of the validation and verification procedures when data are first collected but also of how current the data are. Static data does not allow for good management of beneficiaries over time, whereas a dynamic registry can offer up-to-date information to enable timely decision-making, ensure that programs remain relevant to households’ current situations, and avoid faulty targeting, among other problems. Brazil, Türkiye, and Chile use two types of information to update the data in their social registries: self-reporting by registered families and administrative databases. In all three countries, registered households are obligated to report specific changes in their circumstances. Some of the self-reported changes require accompanying documentation, and officials make random home visits to verify the information provided by the households. In Brazil and Chile, households must confirm their information in the social registry once every two years even if there are no changes to be reported, otherwise they may not be eligible to apply for a new program or may stop receiving their current payments. NSER has also started using multiple administrative databases (such as travel history, bank statements, and public employment records) to verify and validate the information in the registry; this can be further enhanced by leveraging other databases with large coverage like land records and other employment databases. NSER requires households to bring supporting documents if there is any change in family size or composition, but these changes must first be updated with NADRA. The downside is that people often forego a day’s work to visit these offices and the financial implication can prevent them from registering in the first place. NSER has opened a few registration centers where NADRA services are also available onsite, but it has not been scaled up. Moreover, there are currently no random home visits, which if incorporated, can improve data accuracy. A key requirement for making a social registry more dynamic is the existence of a wide network of service points to which people can come to register or change their information. NSER currently has 164 centers, but given mobility constraints, more are needed. (CONTINUED ON NEXT PAGE) THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 26 BOX 4 (Continued) Some services are available online but unlike Brazil, Türkiye, and Chile where access points are supplemented with online service windows, Pakistan will have to use other innovative solutions to improve access as literacy rates are low and access to the internet is limited in Pakistan. Experience suggests the important role of institutional arrangements, procedures, and incentives (for both households and administrative staff) in supporting the implementation of a dynamic social registry. These include the authorizing framework for the social registry; inter-governmental agreements for information sharing as well as other areas; well-defined technical protocols; and monitoring, oversight and controls. To make the NSER a more dynamic registry, BISP has information if any update in the NADRA database has established registration desks at the tehsil level. already been made, including supplying missing CNICs Census-type surveys can omit eligible households or replacing invalid ones; (iv) update information for multiple reasons, and households’ demographic on their changed circumstances after experiencing and socioeconomic characteristics also change over a shock; (v) update data on their socioeconomic time, thus affecting their eligibility for benefits. BISP status every two years; and (vi) recertify themselves desks were set up in tehsils to address these issues. In as existing beneficiaries every three years. There will some cases (for example, after the floods in 2022), BISP be an update of the entire registry every four years set up additional temporary desks to widen access by executing a PMT calculation within six months of for people affected by shocks. Households can use collecting updated data from types (v) and (vi) in three these registration desks to: (i) register if they were and a half years (Figure 10). So far, 2 million families omitted during the door-to-door survey; (ii) update have either registered or updated their information information if their household characteristics qualify through the desk-based data collection centers. them to file an appeal (iii) update beneficiary roster 27 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services FIGURE 10 PROCESS CYCLE OF DYNAMIC REGISTRY 8. The dual challenges of incl the continuous improveme centered approach. Start Type Type Type 1 2 3 Door to Door Update of Missed Implementation of Roster Update Completed Information Appeal Cut Off Missed HH Coverage Appeal Cut Off PMT 32-35 Birth Registration Missing CNIC of Death Registration Potential Beneficiaries Vulnerability inclusion filters are: Marital Status Update Discrepancy Gender Update Resolution Update of Presence of one of more widow Disability Update PMT Cut CNICs of all adult members member(s) Off: 32 Presence of one or Linked with NADRA Operational measures database to ensure B-Form prior more members with to adding children 65 years of age Size of household 3 Re-calculation of PMT Volunteer opt-out or less Completed Immediate-on going Subject to increase Regular in the number of Kafaalat beneficiaries Type Type Type Type 4 5 6 7 Shock Responsive Cyclic Socio-Economic Re-certification of Update Update Update Existing Beneficiaries the Registry For all kinds of Non-Beneficiaries will be called for After 3 years of After collecting disasters: re-survey after two years to update survey, updated data from socio-economic status, in different beneficiaries will both type 5 & 6 in 3 NDMA/PDMAs will PMT brackets (e.g.) be asked to go notify the areas and a half years, a 35.01-40: First 3 months through the fresh PMT NSER will resurvey 40.01-45: Next 3 months mandatory notified areas calculation will be 45.01-50: Next 3 months recertification/ executed within the using desk based 50.01 & Above: Next 3 months reassessment data collection next 6 months and Once initiated, all HHS under these Beneficiaries will all the existing NSER centers categories will be processed in be given 6 data will be updated Re-calculation of phases and process will be completed months for the after every 4 years. PMT within a one-year time frame resurvey Immediate in Every 2 years Every 3 years Completed Every response to shock 4th Year Source: BISP NSER Wing THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 28 These continuous updates mean that the NSER BISP, as the keeper of NSER, is trying to make database will be fully updated every four years this data available to other public and private through desk-based data collection, eliminating sector organizations through semi-automatic the need for massive door-to-door surveys in the systems. Over 30 organizations including provincial future. A door-to-door survey will be carried out social safety net programs, federal and provincial on a small sample to cross-check the accuracy of government ministries, NGOs, health institutes, data collected at desks and to discourage CNIC. national poverty alleviation programs, disaster In addition, the NSER will conduct surveys in areas relief management authorities, and research specified by the National Disaster Management and policy institutes have accessed data from Authority (NDMA) and the Provincial Disaster the NSER on a need-to-know basis. Given the Management Agencies (PDMAs) in case of a disaster. breadth and sensitivity of the NSER’s data, only Also, in the event of a disaster, the registry desks in the specific information needed by the user for the tehsils will be used to process new registrations, planning or implementation purposes is shared record losses, and reassess the PMT for emergency after sensitive information is removed and the data cash assistance provided by social safety nets or are anonymized. Data sharing can be done using other public or private institutions. This system was an API that ensures the automated two-way flow used during the 2022 floods. of information by secure means, which in some cases can be bi-directional, for example, the Punjab In the absence of recent household survey or Social Protection Authority (PSPA) (Figure 11). Here census data, it is hard to conduct external validity bi-directional means that while the recipient entity checks on the accuracy of NSER data. The current can use NSER data, it can also contribute to the estimates of coverage are based on 2017 census database. PSPA uses the National Data Exchange population numbers which, given Pakistan’s annual Platform (NDEP) to retrieve and upload data to and population growth rate of 1.8 percent, may overstate from the BISP. This is an automated process with no NSER coverage. Moreover, while the latest HIES human intervention required and hence requires Survey was conducted in 2018-19, Pakistan has since technological capacity. experienced the COVID-19 pandemic, the 2022 floods, and record-breaking high inflation. Multiple studies The use of APIs for data sharing is a crucial step have posited a substantial increase in poverty. in the automation of the process, but it is just the However, in the absence of recent household beginning. Multiple APIs could be incorporated to survey data, it is impossible to make any accurate standardize asset information and geo-location. calculations of poverty rates, Available estimates The NDEP could act as a consolidated platform for of coverage and of exclusion and inclusion errors registering and retrieving data on the beneficiaries of are based on simulations rather than on concrete various social protection programs. However, the use evidence. To address this problem, the World Bank is of APIs has its limitations because they rely on the currently fielding a survey to collect household data CNIC as input for data retrieval. Partner organizations that can be used to assess the quality of NSER data. will have to conduct their own registration process first to be able to collect other information from Previously, the NSER database was a collection of the NSER against the provided CNICs. This could multiple datasets, but it is now one consolidated make information-sharing a tedious and ineffective database. Each phase of data collection used to exercise since the organization that collects the be stored separately in the NSER database, which, CNIC can also gather the rest of the information with though convenient during project implementation, minimal extra effort. Additionally, the NSER cannot created multiple hurdles to effectively using the full provide a poverty score for a specific household, as dataset. Data are now entered and retrieved from BISP has decided to provide only the PMT ranges. This the consolidated NSER database, which allows for could restrict the efficacy of partner organizations. better data quality checks and data security. The consolidated database enables BISP to provide reliable and consistent reports to both internal and external stakeholders. 29 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s other sectors, such as heal FIGURE 11 NSER DATA-SHARING PROCESS WITH OTHER INSTITUTIONS benefits, and legal services 8. The dual challenges of incl the continuous improveme centered approach. Get PMT CNIC PMT Range Post Prog. Enroll Beneficiary CNIC, Name & Address Benefit code, Value & Duration Get Prog. Enroll Beneficiary CNIC, Name & Address Benefit code, Value & Duration National Data Exchange Platform (NDEP) Cognitive APIs NSER Partner Organizations Source: Ali 2023 While the NSER can provide partner organizations possible to export data from one database and import with a wealth of valuable data, it could benefit from it into another, regardless of the underlying technology. receiving additional information from partner Similarly, web services at extract transform and load organizations. The NSER does not synchronize any (ETL) and presentation layers9 are compatible with information received from partner organizations. By most database management systems. However, data- setting up mechanisms for two-way data exchange, sharing between platforms will not work if databases the NSER could become a more dynamic and useful have incompatible designs and have to follow the resource. By incorporating updated information program-specific requirements and operational from other federal and provincial agencies such decisions of their respective organizations. For as death and birth records, marriage and divorce instance, a database may have separate columns for documents, disability status, and land records, the “first name” and “last name,” while another database NSER would become a more dynamic registry. This may have the full name in a single column. Even this would make the NSER a more comprehensive and complexity can be overcome if the organization carries up-to-date repository of information and enhance out a well-planned data transformation exercise. its value as a resource for partner organizations and other stakeholders. The absence of harmonized data security protocols and the lack of technical capacity are also obstacles Given all of the current advances in technology, to optimizing data sharing between organizations. interoperability between platforms is no longer Feeder agencies such as NADRA and NSER that contain a significant challenge. Most reputable databases, sensitive information require that the recipient agencies including open-source databases, use industry- mitigate any data security risks before any data is standard data encoding mechanisms, making it shared. The receiving agencies can do this by adopting 9 The presentation layer ensures the information that the application layer of one system sends out is readable by the application layer of another system. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 30 international security standards and taking proactive each time data is shared. A streamlined data-sharing measures to ensure data privacy and confidentiality. policy and automated processes would improve the Many organizations also lack adequately trained IT staff situation. This is especially important for disaster and often require external organizations to go through risk management where response time is short and long unautomated processes to access their data, absence of data sharing and data security protocols, creating a bottleneck in the data-sharing process. Since limited technical capacity, and lack of organizational the process is manual, it requires human intervention links can exacerbate the situation (Box 5). BOX 5 USING THE NSER IN EMERGENCY RESPONSE SITUATIONS When Pakistan went through COVID-19-related lockdowns, the NSER update was still incomplete. The government recognized that mounting an effective emergency response would require finding a way to include all vulnerable populations in the registry. NSER data were mainly being collected via door-to-door surveys, but desk-based data collection—in which where people visited temporary data collection kiosks to share their data —had been piloted in a few districts. Since the pilot results did not show any significant quality difference between the desk-based and door-to-door data, the government quickly scaled up its desk-based data collection effort. Only those households that had not previously been included in the NSER were allowed to register while households already registered were not allowed to update their data. The wealth of information in the NSER makes it ideal not only for planning social protection initiatives but also for responding to shocks. Since COVID-19, it has been used to respond to multiple shocks as programs can be rapidly expanded either vertically or horizontally. Given that it contains the CNIC numbers and cell phone numbers of all individuals in the database, it can also be used to collect additional data from these people for decision-making and planning purposes. The extent of NSER’s coverage enabled the government’s COVID-19 response to identify 11.9 million vulnerable families and provide them with a one-time emergency cash transfer of PKR 12,000. In addition, the existing 4.3 million BISP Kafaalat beneficiaries also received emergency transfers of the same value. The updated NSER meant that it was possible to expand the program horizontally as well as vertically extremely rapidly, with the emergency transfers being announced only eight days after the March 2020 lockdowns began. It is worth noting that these emergency transfers were the largest in Pakistan’s history and were made possible only because of nationwide cash transfer delivery system developed under BISP. The government leveraged NSER again in 2022 after the country was hit by floods that directly affected around 33 million people. This unprecedented level of devastation required an efficient response. NSER enabled the government to identify all the families that were in flood-affected districts by triangulating information with other databases. For example, by using GPS data from NSER and flood exposure data from the United Nations Institute for Training and Research (UNITAR), it was possible to identify areas where families most vulnerable to floods were concentrated. The maps (Figure 12) show that majority of those affected by floods were in the province of Sindh where, in most areas, more than 80 percent of the population falls in the bottom 60 percent of welfare distribution. The data provided powerful insights into where the scarce resources should be channeled. As a result, one-time emergency payments of PKR 12,000 each were made to 1.76 million families. 31 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program POPULATION AFFECTED BY FLOODS 7. Social protection delivery s other sectors, such as heal benefits, and legal services FIGURE 12A FIGURE 12B 8. The dual challenges of incl the continuous improveme centered approach. Source: Estimates based on NSER and UNITAR data Note: Bottom 60 percent of the welfare distribution is based on Proxy Means Test score that is calculated using data on demographic and socioeconomic indicators. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 32 3.3 PROVINCES ROLL OUT SOCIAL PROTECTION PROGRAMS AIMED AT ADAPTIVE SOCIAL PROTECTION Pakistan’s social protection system is further progress in building provincial institutional capacity, developing as the provinces introduce and scale up with both Punjab and Sindh having set up social their programs. The Eighteenth Amendment to the protection authorities with independent boards. On Constitution of Pakistan in 2010 gave the provinces the programming side, Punjab has rolled out multiple self-governing, legislative, and financial autonomy. programs, while Sindh is preparing to roll out a CCT All social sector programs and policies fell under (Box 6). Currently, provincial programs have low provincial jurisdiction. There has been considerable coverage, but they are expanding. BOX 6 PROVINCIAL SOCIAL PROTECTION PROGRAMS The Punjab Social Protection Authority (PSPA) provides its beneficiaries with a range of benefits, including a universal health and nutrition CCT program in 11 districts of South Punjab aimed at improving health and nutrition outcomes among the target population. Other cash transfer programs include various subsidies targeted to families in the two lowest income quintiles in Punjab. The ration subsidy program aims to provide subsidies to about a million families on the purchase of selected items. The cash assistance to the transgender community and senior citizens’ social insurance programs also come under the jurisdiction of the PSPA. The PSPA is also planning to introduce a labor market program that will provide employability and poverty graduation services to migrant workers, women, and young people. The Sindh Social Protection Authority has also launched initiatives aimed at improving health, reducing malnutrition, improving women’s well-being, and fostering youth employment. The precursor of the Sindh Social Protection Authority had been running food security programs that focused on reducing acute malnutrition in children and mothers. The other initiatives of Sindh’s social protection system include efforts to improve women’s well-being and internship programs for young people focused on increasing their employability and promoting their graduation from poverty. Coordination and collaboration between provincial At the policy and programmatic level, a partnership and federal agencies will avoid fragmentation and between federal and provincial agencies with duplication. The expansion of the social protection clear roles and responsibilities and based on their system presents opportunities as well as challenges. respective technical and operational advantages Better coordination could help to expand coverage, can further enhance the efficiency and effectiveness avoid leakages, and prevent double dipping. of the country’s social protection system. While Provincial governments depend on the NSER data BISP has been expanding its coverage continuously, to plan and roll out their programs, but they were it will have to coordinate closely with the provincial not able to access this data until recently. Now, the governments before expanding existing programs federal government has a data-sharing agreement or introducing new ones. Provinces can leverage with the provinces to give them access to NSER data the advantage of local presence by increasing as well as access to information on whether specific the coverage of both the population and specific households receive BISP transfers. Further, rather vulnerabilities and also by strengthening links with than investing scarce resources to create parallel NSER data systems. Moreover, provincial programs systems and structures, provincial governments are integrated with local health and education could benefit from BISP’s well-functioning and well- governance structures which can promote established delivery systems. improvements in addition to incentivizing service 33 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; utilization. Establishing institutional arrangements these 11 districts, the KPK government will provide and can improve program 7. Social protection delivery s to support social protection delivery systems can be the CCT benefit to those not eligible for BISP (i.e., their other sectors, such as heal benefits, and legal services complex, and there is no one-size-fits-all solution. PMT score is above the BISP cut-off) and those below 8. The dual challenges of incl the continuous improveme The partnership between BISP and the Khyber the eligibility cut-off will be referred to BISP. To ensure centered approach. Pakhtunkhwa (KPK) provincial government is still seamless coordination with BISP, each CFC has BISP in the early stages, but it can provide a blueprint staff on-site who will register potential beneficiaries. for other provinces. The provincial health CCT is a The BISP management could benefit from the universal program implemented in 11 districts, where experience of other countries on the respective roles health and other services (such as birth registration and responsibilities of federal and local governments and CNICs) are provided through Citizen Facilitation in social protection (Box 7). Centers (CFCs) which serve as one-stop shops. In BOX 7 LOCAL GOVERNMENTS PARTICIPATE IN SOCIAL PROTECTION DELIVERY SYSTEMS There is no blueprint for the best institutional arrangements to support social protection delivery systems. Roles and responsibilities vary from country to country. In addition, these arrangements are dynamic and tend to evolve over time. In all countries, including Pakistan, the overall context including the level of decentralization, the capacity of local governments, and political dynamics influence what institutional options are available to deliver social protection benefits and services. The BISP delivery chain is fully centralized with its deconcentrated local offices reporting directly to the central program agency, like the arrangements that applied for the Prospera program in Mexico. At the same time in Pakistan, there are important areas of the delivery chain where the central agency collaborates with the provinces. For example, when the PMT used for targeting the BISP was updated in 2021, the provinces formally participated in the technical discussions and influenced the process. BISP obtains information from provincial education departments to assess the supply needs of public schools participating in the BTW education CCT. BISP shares data from NSER with provinces to use to target their own programs. In countries that are more decentralized than either Mexico or Pakistan, local governments often participate in specific activities along the delivery chain, generally using information systems, rules, and guidelines developed at the national level within the framework of bilateral agreements between the national and local levels. In these countries, the national level has fewer client-facing responsibilities, as in these examples. Chile: Most social Colombia: For the CCT, municipal Brazil: Municipalities are protection interventions authorities appoint a liaison and responsible for outreach, are designed and provide supporting infrastructure. enrolling participants, and financed at the They are responsible for outreach, following up on cases of non- national level and enrolling participants, handling compliance for the CCT and are implemented by grievances, transmitting data on for registering people in and municipal authorities. beneficiaries’ compliance with updating the social registry. Registrations and updates conditions, and supporting groups State governments provide for the social registry are of mother volunteers who help to technical assistance and other handled by municipal communicate with participants. support to the municipalities. staff, who are trained and Municipal authorities register The national government certified by the Ministry of households with the social registry covers some administrative Social Development and supervised by a specialized division in costs for municipalities through the Family. the Ministry of Planning. performance-based financing. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 34 BISP relies heavily on its field offices for the The first National Social Protection Conference implementation of its program activities, which could in May 2023 was attended by stakeholders at the provide an entry point for better coordination and federal and provincial levels. The government could integration with provincial programs. BISP established consider establishing a committee for more general a network of local field offices to increase access to the consultation and collaboration in developing a social program for its beneficiaries and the public. There are protection strategy for the country. This will help in 16 zonal offices (including a central zone in the capital identifying roles and responsibilities for each entity. of each province and region), 160 district offices, and over 400 tehsil offices. This structure facilitates the BISP and its delivery systems are critical to ensure implementation of the program at the local level. While the adaptiveness of Pakistan’s social protection the zonal offices supervise the network of district system. UCTs can be an appropriate tool for shock and tehsil offices, the tehsil offices are responsible response for two reasons: (i) if the UCT is linked to for recording and addressing any complaints from robust and adaptive systems, its coverage or benefit BISP beneficiaries through a complaint management level can be rapidly scaled up or down in response to system, monitoring payments to beneficiaries, carrying changing needs, thus avoiding the need to spend time out field visits,10 and educating beneficiaries about the and resources on developing a response from scratch program. However, these structures can be integrated and (ii) there is evidence that cash transfer recipients with the provincial administrative structures which are are likely to recover more quickly from shocks and at the Union Council level. This can further improve hence are more resilient to shocks (Loeser et al. 2021; the quality of services by enhancing the monitoring Ivaschenko et al. 2019). BISP was instrumental in function. Pakistan’s response to the COVID-19 and vertically and horizontally expanded with the existing beneficiaries Monitoring how well the various steps along the receiving a top-up transfer and an additional 10.5 delivery chain are implemented is critical to any million vulnerable families being given emergency program’s effectiveness. The performance of staff transfers. It took eight days to make the first transfers. working directly with potential or actual beneficiaries Similarly, after the 2022 floods, BISP provided one-time is critical in this regard. In some countries (such emergency transfers to 2.78 million families. as Brazil and Chile), these individuals work for local governments. In Pakistan, they work in BISP’s Better coordination with both disaster management decentralized local offices and report to the central authorities as well as provincial social protection agency. In either situation, it is important to monitor agencies is required to improve the effectiveness the activities of these staff, and to ensure that they are and timeliness of shock response. In 2012, the performing their jobs well. government developed a national strategy for managing catastrophic events. The Federal Disaster To improve coordination with the provinces and Response Action Plan (FDRAP) outlines contingency to ensure a consultative decision-making process, plans, minimum resources and approval processes BISP recently established a steering committee for that can ensure a swift response to shocks. It the NSER and organized the first National Social recommends cash transfers as the preferred Protection Conference. The steering committee has shock response modality and specifies roles and representation from all provinces and regions as well responsibilities of partner agencies, including national as technical experts. The objective is to ensure that and provincial disaster management authorities, BISP, BISP delivery systems generally, and specifically the NADRA, the Ministry of Finance, and commercial banks. NSER, can be leveraged by social protection programs This model was deployed to respond to floods in Sindh in the country. It aims to create a sense of ownership in 2012 and later parts of it were adapted to respond among the federating units so that the NSER is not to temporarily displaced persons (TDPs) of the conflict seen as a BISP specific registry but as a public good affected Federally Administered Tribal Area (FATA) in that can be used for planning and implementation. 2015 (Smith and Bowen 2020). BISP has taken a lead The committee will provide guidance and technical role in planning and implementation but there is a advice on key activities such as revision of the PMT need to mainstream the FDRAP and to build on it by formula, data collection process, etc., where provinces including provincial governments in planning, resource can play a significant role given their local presence. mobilization, and delivery of shock responses. 10 Field visits are conducted to: (i) check POS agents during disbursement of cash transfers, to ensure no fraud or lack of service takes place; (ii) to visit CCT lo- cations (schools and health facilities) to meet beneficiaries and observe ongoing operations; and (iii) to create awareness about any new activity or initiative. 35 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 6. Social protection programs do not operate in a vacuum, and thus th not be developed in silos; synergies across institutions and informat and can improve program outcomes. 7. Social protection delivery systems can contribute more broadly to go other sectors, such as health insurance subsidies, scholarships, socia benefits, and legal services. 8. The dual challenges of inclusion and coordination are pervasive and the continuous improvement of delivery systems, through a dynami centered approach. THE SOCIAL PROTECTION DELIVERY CHAIN FOR BISP KEY MESSAGES The four main phases of the delivery system are: assess, enroll, provide, and manage. Each phase encompasses a series of activities that promote, communicate, identify, locate, register, pay out, monitor and streamline the delivery of the cash transfer to its beneficiaries. As BISP’s role in disaster response increases, intended beneficiaries, particularly those in remote areas, need to be made aware of their eligibility, the registration processes, and available benefits. In 2010, BISP conducted a nationwide door-to-door survey to compile a database, now known as National Socio-Economic Registry (NSER), which assigned each household a PMT score, which was used as a proxy for measuring poverty. Eligible families receive a letter which contains information on frequency and number of payments, delivery mechanism of payments, expected date of first payment, address of BISP office, duration of eligibility, and details of how to update their data. BISP currently makes all its payments digitally to ensure safe, secure, swift, and convenient delivery and tracks payments flow through its MIS wing, which enables it to monitor frequency of withdrawals as well as dormant accounts. BISP is working on integrating its grievance redressal mechanism with the development of a UCMS, which will allow complaints data to be stored in one single place. Institutional coordination, both horizontal and vertical, is critically important for monitoring beneficiaries’ compliance with conditionalities. BISP has a team of 900 field officers who regularly carry out monitoring and compliance verification. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 36 As with most benefits and services, BISP cash enrolled and onboarded. Once the eligibility decision transfers follow a well-organized delivery chain with has been made, the registrant becomes a beneficiary. four main phases: PROVIDE—the payment of benefits is a key stage of ASSESS—through outreach and communication, the delivery chain. This is a recurring action and one the government lets potential beneficiaries know of the main points of contact between a program and about proposed social protection initiatives; its beneficiaries. For cash transfers, there is a move to conducts intake and registration to ensure that the digitizing payments to ensure their safe, secure, swift, intended population is registered efficiently, and and convenient delivery. their information is recorded accurately at points of contact such as local offices, mobile teams, social MANAGE—a case management system and workers; and phones. The needs and conditions efficient grievance redressal mechanism of BISP gives of registered applicants are assessed using pre- beneficiaries a voice in the process. Monitoring will determined criteria that can be based on different reveal if beneficiaries are receiving timely payments factors such as socioeconomic welfare, risk profiles, or in full while an exit strategy presents an effective labor profiles to determine their eligibility for specific way to exit recipients who are no longer eligible for programs or a mix of benefits and services. the program. ENROLL—the target population (individuals, families, In the following sub-sections, each phase and what or households) enter this stage as registrants and it entails will be presented and explored. become beneficiaries if they are deemed eligible and are 4.1 ASSESS This sub-section presents the first phase in the software applications to support both front-office delivery chain, which consists of outreach, intake and back-office functions, database management and registration, and assessment of needs and and interoperability. Social registries don’t operate in conditions. Social registries play a critical role in isolation and are usually part of broader information supporting the assess phase of the delivery chain. systems supporting social programs, including Social registries are information systems that allow beneficiary registries and administration systems, for flows of information on registrants (individuals, payments administration, and case management households) and their socioeconomic conditions that systems. Since NSER, Pakistan’s social registry, has inform decisions on enrolment, policy coordination, already been discussed in section 3.2 above, this and monitoring. As information systems, their basic section will not delve further into the details of NSER. architecture includes data intake and exchange, 4.1.1 Outreach and Communications Outreach relies on effective communication to raise to be assessed. The key goal is to ensure that the awareness among potential beneficiaries about the target population has the relevant information and existence of social protection programs. The delivery understanding of the planned interventions and is of most social protection programs starts with willing and able to access the program. It is important outreach. This stage typically involves communication that vulnerable groups, such as women and girls, and interactions to build awareness of the program people with disabilities, senior citizens, indigenous among the population, inform the target population populations, people in remote areas, people more about the program, and encourage them to come exposed to climate shocks, and gender and sexual forward and provide information for their eligibility minorities receive information about benefits. 37 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th FIGURE 13 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s THE OUTREACH PHASE OF THE SOCIAL PROTECTION DELIVERY CHAIN other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme centered approach. RECURRING 1 2 3 4 5 6 7 CYCLE 8 9 1 Outreach When BISP’s unconditional transfer subprogram her. The TV commercials, newspaper, and radio ads, (now called Kafaalat) was launched, the choice banners, and brochures explained the aims of the to name it after the late Prime Minister Benazir program as well as its intended beneficiaries. They Bhutto made it recognizable to the population also referred to the religious duty of helping the poor and ensured the political buy-in of both the ruling and needy, and focused on family welfare. As an and opposition parties. One of the most popular indirect way to promote female empowerment in leaders of Pakistan, Benazir Bhutto, was assassinated a highly conservative society, the UCT was provided in 2007 during her election campaign. In 2008, her to ever-married women whether they were married, party won the elections. BISP was launched the divorced, separated, or widowed, lived alone and same year and all political parties supported the had no family (defined as a spouse and children). The decision to name the program after Benazir Bhutto. emphasis on marital status presented the program as To announce the program and attract the attention family oriented. Having a CNIC was also a requirement of the public, the government invested heavily in for registration (Box 8). print and media campaigns that featured pictures of BOX 8 ENHANCING WOMEN’S EMPOWERMENT THROUGH IDENTITY CARDS According to the results of the 2018 ID4D-Findex survey, one in every two women in low-income countries lacked an identification document or its equivalent, preventing them from participating in public life and claiming their rights as full citizens and restricting their autonomy both inside and outside the household. In Pakistan, an identity card allows citizens to vote, own property and other assets, seek membership of political parties, and gain access to government welfare schemes and services, including hospitals and schools. In this way, identification facilitates an individual’s participation in economic, social, and political activities. The Government of Pakistan deliberately made the CNIC a pre-requisite for women who wanted to register for BISP. Within four years of the program’s launch, there was a 72 percent increase in the number of CNICs issued, with a 94 percent increase among women. By 2012, 40 million Pakistani women possessed CNICs. The impact of this on women’s empowerment was unprecedented. The BISP impact evaluations found that women with CNICs reported having a stronger sense of identity than they ever had before. They were also more aware of their rights and were eager to vote and share their opinions. Source: World Bank 2016 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 38 Initially, responsibility for outreach and identifying as beneficiary families were required to send their beneficiaries was delegated to members of the children to school on a regular basis in a country with Parliament. The 442 members of both houses of many out-of-school children. It was only through Parliament were responsible for raising awareness of communication that issues that keep children out the UCT program and for outreach in their respective of school, such as child labor and social norms, constituencies (USAID 2012). Each member was also could be addressed. Having developed a draft asked to identify 8,000 potential beneficiaries who communications strategy for WeT, BISP and rolled were considered the most deserving and needy it out in five pilot districts. The strategy used street in their constituency. This approach was adopted theater, television, radio, and announcements in because at that time there was no credible database mosques to communicate key messages to BISP with high coverage rates. The assumption was beneficiaries to encourage them to enroll their that parliamentary representatives were close to children in school and thus become eligible for WeT. their constituencies and best placed to identify An evaluation assessed the reach and effectiveness the poorest households. They were expected to of the strategy and found that most respondents encourage those households to step forward and received information through local notables volunteer the information needed to register them (Figure 14). Focus group discussions revealed that for the program. However, in a country of 180 million no women reported receiving messages through people, members of Parliament did not have close TV, radio, mosques, texts, or robo-calls. Based on contact with everyone in their constituencies. the responses from beneficiaries, the most effective means of communication was announcements from When the poverty scorecard was introduced to local notables, with 40 percent of beneficiaries having determine eligibility for the Kafaalat UCT program received information on the WeT this way. This was in 2009, BISP had no comprehensive formal followed by street theater (13 percent) and mosque communications strategy. Given that BISP was the or church announcements (7 percent). Among largest national safety net program in the country, those who received information from mosque or a well-thought-out communications approach was church announcements, 71 percent understood the needed not only to inform the broader population message completely (Mott MacDonald 2014). These about the program and build its credibility but also results indicate that the communications strategy to inform potential beneficiaries about its objectives in the five districts was not necessarily effective and rationale, its intended target groups, benefits, and point to the need to develop tailored modes and payment and grievance redress mechanisms. of communication that can be easily accessed and This would have helped to create transparency and understood by potential beneficiaries. accountability. However, communications took place via the first door-to-door survey that collected data Although the 2016 draft communications strategy for the NSER. Communications teams deployed a acknowledged that communications had been few days before data collection started to hold the BISP’s weakest link, no strategy was officially community meetings, meet community leaders, adopted, and communications and outreach make announcements through mosques, and continued to appear in operations manuals under distribute flyers, posters, and brochures to inform “social mobilization.” BISP has drafted several communities about survey dates and processes. communications strategies since 2012. Some have Third party spot checks on a sample of locations been comprehensive, covering a range of areas from assessed the performance of the communications consensus-building activities to promoting dialogue firm. However, given the size of the country and the on and increasing public awareness of social safety remoteness of locations, the survey could not cover nets. The strategies recommended using information the entire country. material that highlighted stories of success not only in improving the socioeconomic status of poor There was a stronger focus on communications and families but also empowering them to provide outreach when the WeT program was introduced a better future for their children. The materials in 2012, but again, no formal communications would also show how program activities had the strategy was ever approved. Communications potential to reduce poverty at the grassroots level, and outreach were critical for the success of WeT focusing on program implementation, beneficiary 39 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme centered approach. FIGURE 14 MODES OF COMMUNICATION ABOUT THE WET SUB-PROGRAM 38.80% 12.90% 7.30% 0.60% 0.30% 0% Local Notables Street Theatre Mosque or Church Radio TV Robo Call Source: Mott MacDonald Limited (2014) participation and responsibilities, the payment continuously for long periods of time; (ii) a lack of cycle, and grievance and accountability mechanisms clarity regarding the role and purpose of the BBCs; (Government of Pakistan, 2016.). (iii) the absence of adequate capacity-building for the BBCs; (iv) the absence of any protocols defining BISP set up BISP beneficiary committees (BBCs) to the BBCs’ structure, functioning, or processes aid outreach and communication, but this proved (including communication protocols); and (v) a lack to be unsustainable. BBCs were local groups of of capacity and resources within the BISP to manage beneficiary women, set up by third parties, and and monitor the committees (Box 9 on the following led by a volunteer, an uncompensated “mother page). leader”. By 2018, more than 100,000 committees had been formed throughout Pakistan. However, their efficacy was questionable as an increasing number of them became dormant or were abandoned by BISP. Possible reasons for this were: (i) the lack of remuneration for the mother leaders, which discouraged them from leading the groups THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 40 BOX 9 BISP BENEFICIARY COMMITTEES A BISP beneficiary committee (BBC) was a group of 5 to 40 beneficiaries, led by a woman or mother leader who was elected or nominated by the group. The composition of these groups was based on women’s availability and geographical context. The BBC model was adopted under BTW and was implemented through implementing partner firms (IPFs). The IPFs were tasked with establishing, organizing, mobilizing, sensitizing, educating, and guiding these beneficiary committees. The main task of each BBC was to increase the enrollment of primary school aged children in BISP beneficiary households through regular follow up with beneficiaries and by raising awareness among BISP beneficiary families regarding the importance of education. The secondary objectives of BBCs included raising awareness among poor women about BISP and its processes, increasing enrollment of primary school aged children, ensuring retention of already enrolled children, and reducing the drop-out rate through behavior change campaigns and regular follow up with the mothers of children enrolled in BTW. The BBC meetings also provided a platform for beneficiaries to raise concerns with the program and seek resolution. BBCs were formed at the community level as a local information sharing and facilitation forum for BTW beneficiaries. BISP field offices, with the support of dedicated BTW staff, were mandated to oversee the social mobilization activities carried out by the IPF in their respective geographical areas. IPFs were responsible for conducting frequent meetings of the BBCs and facilitating mobilization of beneficiary mothers to complete the registration, compliance monitoring, and case management. IPFs also acted as the intermediary communication and coordination channel between BISP and its beneficiaries. The BBC model was piloted in a few districts in 2015-16 and gradually expanded to the remaining BTW districts. As of August 2019, there were 98,724 BBCs present at different administrative levels in 50 BTW districts across Pakistan. However, the BBCs have been defunct since 2018 due to the closure of IPFs contracts. IN 2019, the BISP Board decided to exclude the social mobilization activity under BTW. The decision was based on the Oxford Policy Management’s (OPM) evaluation report which found a mixed response from beneficiaries in terms of performance, effectiveness, efficiency, and usefulness of BBCs and mother leaders. Although the BBCs were an effective platform for social mobilization, they faced serious challenges with implementation and management. Analysis showed that the intended objectives of the social mobilization activity executed via the BBC model were not achieved and had high-cost implications. Therefore, BISP explored and employed other cost-effective avenues for social mobilization. In 2020, BISP hired about 900 compliance monitoring BISP has integrated communications and outreach officers for the BTW program and also gave them into its various functions but has not developed its some outreach and communication functions. The capacity to implement comprehensive and frequent main job of these officers is to visit schools to verify communications and outreach campaigns. BISP’s attendance by children from beneficiary families; outreach and communication efforts have been ad however, they are also responsible for communicating hoc and often take a backseat to other aspects of with the community about BTW-related activities. program implementation. The BISP headquarters has However, it is not clear how much time they have hired only one to two people for communications available for communications and outreach after whereas a program of this size requires more human performing their compliance monitoring tasks. resources and technical support. Currently, BISP’s 41 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; communications and outreach is mainly conducted training for beneficiaries of government-led social and can improve program 7. Social protection delivery s through press releases issued regularly to the print protection programs (Box 10). Sessions include other sectors, such as heal benefits, and legal services media and through BISP’s senior management on targeted messaging on disaster preparedness, 8. The dual challenges of incl the continuous improveme social media channels. resilience, and adaptation to those households that centered approach. are most vulnerable to shocks and most likely to As many countries are increasingly working toward engage in negative coping mechanisms when shocks making their social protection systems more occur. However, the formal integration of shock adaptive, some are holding beneficiary education preparedness and tailored disaster management sessions to provide relevant information to the training into social protection beneficiary education most vulnerable households. Several countries appears to be limited (World Bank 2022d). have integrated disaster preparedness and resilience BOX 10 RAISING AWARENESS OF DISASTER PREPAREDNESS USING SOCIAL PROTECTION SCHEMES In the Philippines, the country’s flagship CCT Pantawid Pamilyang Pilipino Program (4Ps) requires beneficiary households to attend monthly family development sessions (FDS). One of the topics covered at these sessions is ‘Towards a Safe and Prepared Community’ and it includes basic concepts such as hazard, disaster, capacity, vulnerability and risk, community and family preparedness, and the need to prepare an emergency kit and a disaster checklist. Trained city/municipal facilitators run the sessions which include group exercises and discussions. The modules are available in Tagalog, Ilocano, and Bisaya. Information materials are produced by the facilitators and serve as social marketing materials. Evaluations have shown that the sessions have had a notable impact on disaster awareness and protection, with over 60 percent of beneficiaries claiming that their preparedness for disasters is very high. In Mexico, the former CCT program Prospera (presently discontinued but previously known as Oportunidades and Progresa before that) collaborated with Mexico’s Civil Defense to provide disaster preparedness training for social protection beneficiaries. The cash transfers provided under this program were conditional on a set of co-responsibilities including participation in health and self- care workshops. Prospera contributed to disaster risk management through the deployment of field workers and beneficiaries to communicate and educate the community about possible hazards and risk mitigation strategies. An assessment, based on discussions with operational personnel of the states with recurring disaster events, suggested there was room for improvement, including a need to strengthen training on the following aspects: prevention and protection actions to be taken in the event of disasters, modifications that may occur in the operation of the program during the contingency (replenishment of documents, advance of support, exemption from co-responsibilities) and how to contact staff operational during an emergency. Jamaica, unlike the Philippines and Mexico, aims to increase disaster preparedness and protect livelihoods by disseminating information tailored to poor households. In 2017, the Ministry of Labor and Social Security and Office of Disaster Preparedness and Emergency Management produced printed and audiovisual information to inform poor and vulnerable households about disaster preparedness and recovery. These materials included two public service announcement (PSA) videos as well as posters, booklets, and brochures. All of the materials stressed the need to secure important identification documents and to contact key officials during emergencies. The materials were also disability sensitive, using sign language interpretation in the PSA videos and giving specific attention to the needs of visually impaired persons in the brochure and factsheets. Source: World Bank (2022d) THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 42 Communications and outreach are becoming even high transportation costs (Cheema et al. 2020). more important as BISP evolves. With the move to make the NSER more dynamic, beneficiaries will be Other than frequent outreach to the public, BISP will able to go to local BISP offices to register or update need to ensure that people in remote areas, persons their information instead of BISP having to conduct a with disabilities, women with limited mobility, and door-to-door survey to gather information. This will other marginalized groups are not left out of and are require more outreach and communication to inform able to access the new registry centers. There has beneficiaries about the programs, their eligibility been some promising use of “active search” methods conditions and how and when they can update their in some countries, whereby program administrators information. BISP has adjusted its benefit levels and has or local officials proactively reach out to people living expanded its program both horizontally and vertically in remote areas or vulnerable populations who may in response to shocks such as the COVID-19 pandemic. otherwise be overlooked. Brazil has a dynamic registry When this happens, the expansion of eligibility needs (Cadastro Unico) that has operated on an on-demand to be communicated effectively to ensure high basis since 2007, and an active outreach strategy uptake rates from the target population. Anecdotal (busca ativa) was developed under its Brasil Sem evidence and some evaluations suggest the efficiency Miséria (Brazil Without Poverty) initiative to find and of communication through word-of-mouth, but such register all extremely poor families not in the social communication is often limited to timing of payments. registry (WWP n.d.). In Pakistan, BISP made a similar An evaluation of BISP programs in 2020 found that arrangement by using mobile registration vehicles to poor communication to its beneficiaries regarding reach hard-to-reach groups, especially in remote areas when payments would be released led them to make with difficult terrain as in Balochistan. More planning multiple trips to their point of sale or ATM, incurring and resources will be needed to scale this up. 4.1.2 Intake and Registration and Assessment of Needs and Conditions The objective of intake and registration is that the client’s interaction with the point of contact can be intended population is registered efficiently, and either person-centered or service oriented. their information is recorded accurately. Intake is the process of initiating contact with clients and A key feature of intake and registration is whether gathering their information, while registration consists it allows anyone to apply for social programs at any of recording and verifying their information. In some time. This is known as dynamic inclusion which means cases, the information reported by registrants is that access to intake and registration is open and complemented by data from other administrative continuous, usually using an on-demand application systems. Following outreach efforts, the people window as the point of contact. The accessibility of the (individuals, families, or households) who want to point of contact enables individuals to register to have apply for assistance engage with the relevant agency their potential eligibility for social programs assessed and provide their information and documentation. when in need or for current beneficiaries to update The information on the applicant is then verified and their information when their circumstances change. used to assess their needs and conditions. These points of contact are particularly important when developing “adaptive social protection systems Intake and registration require structures and (Leite et al. 2017).” processes that provide registrants and applicants with a point of contact. These points of contact The needs and conditions of registered applicants can be: (i) a local office, service window, or kiosk; (ii) are assessed using pre-determined criteria that can a mobile team; (iii) social workers, frontline staff, or be based on different factors such as socioeconomic enumerators; or (iv) phones and other digital service welfare, risk profiles, or labor profiles. The data windows. They can be managed by central agencies collected on the individuals, families, or households (through staff or contractors) or by local governments. during the intake and registration phase are used to It is important whether the client interface is assess their needs and to determine their eligibility temporary or permanent. It matters who initiates the for specific programs or a mix of benefits and services. “contact”: is it the applicant (demand-based) or the The choice of instruments and techniques used to administrator (en masse data collection)? Finally, the assess applicants’ needs and conditions depends 43 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d FIGURE 15 link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs INTAKE AND REGISTRATION & ASSESSMENT OF NEEDS AND CONDITIONS not be developed in silos; and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme 1 2 3 4 5 6 7 RECURRING 8 9 centered approach. CYCLE 2 3 Assessment of Intake & Needs and Registration Conditions on the characteristics of those being targeted. the targeting efficiency was rather mixed; an analysis One method of classifying applicants is based on based on household survey found that 16 percent of demographic characteristics, such as gender or age (in BISP households possessed more than three acres of the case of demographic categorical programs) while land, even though this was an exclusionary criterion another relies on caseworker assessments (common (Nayab and Farooq 2020). in employment and social service assessments). A third method uses automated tools to aggregate key In Phase 2 (2010-2023), a new intake and registration assessment indicators (typically used for aggregating questionnaire was designed around variables to be socioeconomic welfare measures). Finally, statistical used to calculate a proxy means test (PMT). Before tools, such as predictive analytics and data integration the full roll out, the government decided to conduct a and analytics, can be used to generate profiles pilot test of the efficacy of using poverty scorecards to of individuals or families especially in situations identify the most vulnerable. A rapid evaluation of the when limited data are available (these methods are pilot test in 16 districts found that, while most of the sometimes used for labor profiling or for predicting households selected by Members of Parliament were social risks). Often, agencies use a combination poor, targeting based on poverty score cards yielded of instruments and techniques. Since intake and even poorer households. There was only a 15 percent registration and the assessment of needs and overlap of potential beneficiaries between the two conditions are typically carried out together (Figure methods among households that had a PMT score 15), this sub-section discusses the BISP’s experience of less than 15. The overlap increased as the poverty with both stages. score cut-off increased. BISP wanted to prioritize targeting the very poorest households, for which the Within the first five years of BISP, there were major poverty score card was found to be a more efficient changes in how intake and registration and the method (Nikitin et al. n.d.). assessment of needs and conditions were conducted. There were two phases and each used a different Once the survey was completed, each household methodology to assess needs and conditions. was assigned a poverty score, which determined its eligibility for BISP’s unconditional cash transfer, During Phase 1 (2008-2009) of BISP, intake and now known as Kafaalat. The range for the poverty registration depended on Members of Parliament scores was between 0 and 100 with the poverty which had its drawbacks. Each Member of Parliament, line corresponding to 28 on the PMT measure. BISP based on their knowledge of local communities, established 16.17 as the cut-off score, meaning that had to identify the poorest households in their all households below this score would be eligible to constituency and submit a form for each identified receive the transfer. The cut-off score was determined household. The level of participation and enthusiasm based on BISP’s available financial resources and, among parliamentarians for the exercise was mixed hence, did not imply that the households with scores as there were concerns about a lack of transparency. higher than 16.17 were not poor. Within its limited The parliamentarian system had two main drawbacks. budget, BISP prioritized targeting the poorest and the First, it caused the program to have credibility most vulnerable. About 7.7 million households were problems with the public because it had the potential found to be eligible for BISP’s unconditional transfer to promote clientelism (Pakistan Today 2011). Second, based on the cut-off score. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 44 4.2 ENROLL The enrollment stage in the delivery chain of their eligibility and enrollment status; and (iv) involves determining which registrants qualify onboarding beneficiaries by introducing them to for which benefits and services. The objective is program(s) and gathering additional information, if that eligibility is effectively determined according necessary (Figure 16). The inputs to this stage are the to pre-determined criteria, that the right benefit results of the assessment of the applicant’s needs and service packages are provided to meet the and conditions, the program’s eligibility criteria, and needs of the target population, and that eligible the available budget. At this stage of the delivery registrants are enrolled and onboarded efficiently chain, the target population can be individuals, with minimal leakage to ineligible populations. This families, or households. They enter this stage as stage of the delivery chain involves several phases: registrants and become beneficiaries if they are (i) determining eligibility and making enrollment deemed eligible and are enrolled and onboarded. decisions; (ii) establishing each beneficiary’s package Once the eligibility decision has been made, the of benefits and services; (iii) notifying all registrants registrant becomes a beneficiary. FIGURE 16 THE ENROLLMENT STAGE OF THE SOCIAL PROTECTION DELIVERY CHAIN RECURRING 1 2 3 4 5 6 7 CYCLE 8 9 Eligibility and Notification and 4 5 6 Enrollment Decisions Onboarding Determination of Benefits and Service Package 45 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program When the BISP’s Kafaalat UCT program started Data on any households who could not be contacted 7. Social protection delivery s in 2008, the eligibility of potential beneficiaries are handed over to the NSER Wing of BISP. other sectors, such as heal benefits, and legal services was determined first by Members of Parliament 8. The dual challenges of incl the continuous improveme and then by applying exclusion filters. After each When beneficiaries are informed about their centered approach. Member of Parliament selected 8,000 eligible eligibility, they are advised on how the payments families in their districts, the data were verified by will be made. Each ever-married woman within NADRA using multiple verification and selection the household will receive her own cash benefits filters. For families to be eligible, their monthly independently and each will have to use their own income had to be less than Rs. 6,000, and the CNIC to withdraw payments. The letter sent to female applicant had to have a CNIC. Furthermore, beneficiaries includes the following information: several specific categories of vulnerable groups were (i) the frequency (quarterly) and amount of benefit identified, including widowed or divorced women and the fact that the government may change it in with no adult male members in the family, families the future; (ii) how the payments will be delivered; with physically or mentally challenged members, (iii) the expected date of the first payment; (iv) the and families with members suffering from a chronic address of the closest BISP tehsil office where the disease, all of whom still had to meet the income recipient women or household can obtain additional cutoff and possess a CNIC. Moreover, several information or file a complaint; (v) the duration of exclusion filters were used on the basis of which eligibility (three years), after which the household’s families could be deemed ineligible.11 eligibility will be re-evaluated; and (vi) details of how and where the household will need to submit its In 2010, the BISP started using the poverty data again (if they fail to update their information, scorecard to determine eligibility for Kafaalat. the household will be declared ineligible). The government decides the eligibility cut-off point based on expected number of families,12 availability Benazir Taleemi Wazaif (Education CCT) of funds, and the value of benefit (BISP 2019). BISP continues to use filters to exclude households Before the roll-out of the BWT program (then based on their apparent relative wealth, although known as WeT) in 2012, BISP gathered data on the exclusion filters list no longer includes ‘receiving public and private schools to determine their income support from other government programs’. available enrollment capacity, infrastructure, With the data from the intake and registration phase accessibility, and location. This information helped of the Kafaalat program, a beneficiary list is created BISP to understand if schools in a specific location and subsequently an enrollment plan is prepared in would be able to absorb the increased demand collaboration with the payment agency. after the BTW rollout. BISP obtained this information mainly from provincial education departments for Beneficiaries are then contacted and informed public schools since they collected such data. Now of their eligibility for benefits. The main way of compliance monitors for the BTW conduct supply communicating eligibility to households is through capacity assessments when needed for all public a letter. Any undelivered letters are returned to the and private schools identified as having enrolled BISP tehsil office. If these letters cannot be delivered children of BISP beneficiary families (See 4.4.2). after the second attempt, these lists are displayed at the tehsil office. The NSER survey recorded both landline and cell phone numbers for households so that those with functioning numbers could be contacted by phone calls or secure texts or both. 11 This required NADRA to triangulate the applicants’ CNICs with other databases to identify if any family member: (i) was in government or semi-government employment including authorities, departments, and armed forces of Pakistan; (ii) received a pension exceeding PKR 6,000 from any government or semi-gov- ernmental agency; (iii) received any post-retirement benefits from any government department or agency; (iv) owned more than three acres of agricultural land or more than 80 square yards (three marla) of a residential house or plot; (iv) received income support from any other government programs such as Pakistan Bait-ul-Mal (PBM) Food Support Scheme; (v) had a machine-readable passport; (vi) had a National Identity Card for Overseas Pakistanis (NICOP) holder; or (vii) was an account holder at a bank other than the National Bank of Pakistan, Habib Bank Limited, United Bank Limited, Muslim Commercial Bank, Allied Bank Limited, Bank of Punjab, Bolan Bank, Khyber Bank, First Women Bank, Zarai Taraqiati Bank, Khushhali Bank and all microfinance banks (Khan and Qutub 2010). 12 BISP programs are targeting families instead of households where a family is defined by the presence of an “ever married” woman, irrespective of whether she is living with her husband or whether she has children. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 46 FIGURE 17 BTW ADMISSION PROCESS FLOW SCHOOL / COLLEGE VERIFICATION School or College Child admission verification from School Headmaster Child Eligibility Criteria If child is eligible 1. Child has not completed Primary Education 2. Child Age (4-22Y) from BENEFICIARY active BISP beneficiaries BISP Beneficiary BISP Beneficiary will visit the Tehsil office for Child admission 1. Inactive Beneficiary 1. Meets Eligibility 2. Not 4-22Y Child 2. Child B-Form No 3. Graduated Grade 12 3. Child Physical Presence Registration Yes Tehsil Office No School Verification and Compliance No Certificate Monitors School Verification Enrollment Certificate Yes Payment Generation Payment Agency Amount transferred to beneficiary account Fund Cash Transfer Received to Beneficiary Process End in LMA 1 Account (LMA 2) 47 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme centered approach. BISP has improved the eligibility determination process resulted in an upsurge in applications and and enrollment process for the BTW using digital enrollments in the BWT in 2021-2022. An Android- technology. In the past, the program faced several based online application made it possible to apply challenges, including high administrative costs, using a mobile device such as a cell phone (Figure 18). and a high error and fraud rate due to the paper- Not only did this make the application process easier, based approach. However, digitalization, initiated it reduced the number of visits a beneficiary had to in 2020, has resulted in significant improvements make to the tehsil office. For example, through the in program intake, registration, and compliance online app, potential beneficiaries can check: (i) their monitoring (the latter is discussed in the monitoring children’s eligibility by entering their CNIC number section). BISP developed Android-based applications instead of visiting the tehsil office and (ii) whether that are now used by compliance monitors in field the NSER has a complete roster of their children. If offices to determine eligibility, enroll beneficiaries, the roster is incomplete, beneficiaries will first have and monitor compliance. Since BTW targets only to update their family data on the NADRA database. the children of active beneficiaries of Kafaalat, BISP verifies the status of Kafaalat beneficiaries in real After the enrollment stage is completed, an time from the NSER to avoid inclusion errors. To educational information session is organized to enroll in the BTW, beneficiaries must bring a copy give the women of beneficiary families an overview of their children’s B-forms13 and a school admission of their co-responsibilities and to underscore the slip with information on the child’s class and school importance of education. Sessions last between name verified by the teacher. The verification 10 to 15 minutes and are held for groups of 15 to 20 process includes checking the data on the B-forms women. The focus of the sessions is the importance against the NADRA database in real-time so that no of education and the need to ensure that their “ghost” child is enrolled. Information on the enrolled children maintain regular school attendance so students, such as pictures, admission slips, and other that the family can continue to receive the cash documentation, is archived to create a roster of benefit. Beneficiaries are also informed about the eligible BTW families based on their existing Kafaalat potential consequences of non-compliance and household information. If the Kafaalat beneficiary the value of the cash transfers. The sessions also family meets all requirements, the enrollment explain the program’s grievance redress process process is completed during their first visit to the and encourage them to take an active role in their tehsil office (Figure 17). children’s education and future success. Digitalization has also facilitated beneficiaries’ interaction with the program’s administrative systems. The launch of the BISP’s online application 13 A birth certificate in Pakistan is known as a Child Registration Certificate (CRC) or B-form. The document is used to register minors under the age of 18 years. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 48 FIGURE 18 FLOW OF BTW ANDROID APPLICATION Beneficiary Search Child Registration Child Enrollment 1. Active UCT Members 1. 1. Add new child option through 1. School selection from list 2. Open mother's details B-Form number verified from of schools and enroll a 3. Define Province, NADRA child after running District, Tehsil, UC, & 2. Enter child details e.g. B-Form business rules Village of Mother Number, DoB, Age and run business 2. Upload school Admission rules form with EMIS Code 3. Upload B-Form or Link to already 3. If school is not listed, uploaded B-Form 4. Take picture of details of child enroll- Child ment will be recorded 4. Add Father's Name and CNIC and school will be sent Number for MSCA 5. Generate Unique Child ID a. After completion of MSCA, child will be auto enrolled on beneficiary visit date The Nashonuma program is only open to active Facilitation centers have been set up in tehsil health Kafaalat beneficiaries who are pregnant or lactating facilities. Eligible candidates use their CNICs to mothers with at least one child between the age cross-check their status with the NSER database. If of 0 and 23 months. The program is only available they are active Kafaalat beneficiaries, they need to to residents of districts with a high prevalence of present these documents at the facilitation center: stunting based on the latest National Nutrition Survey (i) a pregnancy test result or a confirmation report data and where no similar program is offered by the and an EPI vaccination card if they are pregnant (or provincial government. To be eligible, households a tetanus vaccination card if in the mother is in her must have at least one pregnant or lactating woman second trimester); (ii) a B-form for the child and an member or a child aged between 0 and 23 months. EPI card for proof of the mother’s vaccination if she is The program has a maximum age limit of 18 months lactating; and (iii) a B-form and an EPI card for proof of for a child to be eligible at entry and a maximum the child’s vaccination if he or she is between the ages of two children who can be enrolled per family. By of 6 and 23 months old. For a child to be eligible for setting these criteria, the Nashonuma program aims the Nashonuma program, their birth must be recorded to provide targeted support to those who are most in in NADRA’s family database, which requires a birth need to improve the health and nutrition of mothers certificate to verify that the child is part of the applicant and young children in the selected program areas. family. By ensuring that all necessary documents are brought to the doctor’s office at the facilitation center, All beneficiaries of the Kafaalat program who might the Nashonuma program can accurately track and qualify for the Nashonuma CCT are required to visit a monitor the progress of eligible beneficiaries and nearby Nashonuma facilitation center to have their provide them with the necessary support to improve eligibility determined and to enroll in the program. their health and well-being (Figure 19). 49 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program FIGURE 19 PROCESS FLOW FOR ENROLMENT OF PLW 7. Social protection delivery s other sectors, such as heal AND CHILDREN TO NASHONUMA benefits, and legal services 8. The dual challenges of incl the continuous improveme centered approach. Potential Beneficiary Development of compliance Verification from profile of PLW & children CNIC verification NSER data 0-23 months of age Enter the current Automatic generation of next month/trimester of the steps indicating required pregnancy/lactation actions/visits, proposed timelines period/ age of child for the visits and amount Source: Nashonuma Project Document, December 2020 Nashonuma beneficiary families are divided into center to verify their required immunization, attend three groups: (i) those who are not malnourished a nutrition awareness session, receive specialized (ii) those with moderate acute malnutrition (MAM) nutritious food (SNF), and record their child’s growth. and (iii) those with severe acute malnutrition (SAM). From the second visit onwards, they have to bring All groups receive immunizations at the tehsil health empty packs of SNF to prove that they consumed the facility where the Nashonuma facilitation center is contents, and the packs are refilled for them. Once located. If a family is in the MAM or SAM category, all these activities are completed, they are referred they are referred to a health facility for treatment to the BVS counter to have their identity verified and are helped to apply for benefits related to through their thumb print before receiving the cash antenatal and post-natal care and safe delivery transfer, which is available immediately after their under the Sehat Insaf Program. All beneficiaries first visit to the facilitation center (Figure 20). must visit the compliance counter at the facilitation THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 50 FIGURE 20 PROCESS FLOW FOR NASHONUMA HEALTH FACILITY LHWs/ MOBILIZATION Eligible PLWs and child BISP MLS (Beneficiary directed for Processing) (0-23 M) from eligible BISP Beneficiaries Active BISP 1. No Active Beneficiaries Beneficiaries 2. No PLV's and Child (0-23 M) No Registration Registration Counter Yes Not Verified Normal Category MAM/SAM Yes MAMISAM Doctor's GROWTH VERIFICATION IMMUNIZATION MONITORING TREATMENT Verification PLWs and Child (0-23 M) Normal Enrollment No PLWS and Child (0-23 M) Counter ENROLLMENT No IN PROGRAMME PLWS and Child (0-23 M) Yes BEHAVIORAL CHANCE COMMUNICATION • Healthy Diet AWARENESS SESSION • Hygiene and Sanitation LHWS/ LHSS • Breast Feeding • Consumption of SNF • Population Growth • Immunization Normal Distribution SNF TRANCHE CATEGORY Counter Treatment COMPLIANCE No Compliance MONITORING Counter • Awareness Session • Immunization • SNF/Treatment Yes Bank Counter BVS & CASH TRANSFER Process End Source: Nashonuma Project Document, December 2020 51 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s 4.3 PROVIDE other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme centered approach. This sub-section will outline the progression its beneficiaries. Cash benefits are widely used in of payment provision by BISP, highlighting social protection, and since the COVID-19 pandemic, improvements over time. Emphasizing efficiency there has been an emphasis on digitizing payments and transparency, BISP has successfully transitioned to ensure their safe, secure, swift, and convenient to digital methods. Plans to further enhance delivery. A substantial amount of money is channeled payment provision, empowering beneficiaries with from governments to their people in the form of expanded choices, are also discussed. social protection cash benefits; globally, nearly a quarter of all adults receive payments, whether as The provision of benefits and services is a key public sector wages, public sector pensions, or social stage of the delivery chain, and there has been benefits such as subsidies, unemployment benefits, a shift toward digitizing this process (Figure 21). or payments for educational or medical expenses The payment of benefits is a core phase in the according to the World Bank’s 2017 Global Findex recurring implementation cycle and, often, is one of surveys (Lindert et al. 2020). the main points of contact between a program and FIGURE 21 PROVISION OF PAYMENTS RECURRING 1 2 3 4 5 6 7 CYCLE 8 9 Provision of Benefits and/ 7 or Services BISP has improved its payment mechanism over system, and the BVS. BISP is currently developing a time, moving toward using digital methods to payment mechanism that will allow beneficiaries increase efficiency and transparency. BISP uses to pick payment service providers14 BISP has been a the same system for its UCT and CCT payments. In leader in experimenting with advanced technology the quest to better its payment process, BISP has solutions designed to tackle the diverse and complex cycled through five different payment mechanisms challenges related to a population with different over the years. These mechanisms include Pakistan cultural backgrounds and geographic locations. Post Money Orders, the Smart Card payment system, the Mobile Banking system, the Debit Card 14 https://bisp.gov.pk/ProjectDetail/ZDZjYzY3ZDAtZTQ2OS00NGRhLTliNmItMzJmMzdiYTY3ZDE0 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 52 PAKISTAN POST or managed by a partner financial institution (such as UBL Bank or telecom operators). The QR code on The first payment modality was money orders.15 the Smart Card could be copied, making it vulnerable Money orders were mailed through the Pakistan to fraud. The second pilot used a branchless banking Postal network to the beneficiaries’ doorsteps.16 model. Beneficiaries received free mobile phones, However, relying on a manual delivery system SIM cards, and a personal identification number invited corruption and leakages. Women also felt (PIN) code that could be used to withdraw their cash obliged to pay a financial bribe (baksheesh) to the transfers from designated franchises after showing postmen for delivering the cash grant. Additionally, a valid CNIC. However, the beneficiaries often in areas lacking house-to-house postal delivery handed over their PIN codes to family members (such as the province of Balochistan and some parts or franchises, which led to fraud. There were also of Khyber Pakhtunkhwa), recipients had to travel incidents of loss, damage, theft, and even the sale long distances to reach the nearest post office. of the BISP-provided mobile phones. These pilots failed mainly because of low levels of TECHNOLOGY-BASED SOLUTIONS financial literacy among the beneficiaries and low cell phone penetration rates. The lack of familiarity BISP piloted two technologies to assess their with digital technology in the early 2010s, especially viability before scaling them up. The BISP Smart Card among those in the bottom two quintiles of Pakistani based on QR technology was piloted in 2010 with households, translated into an increasing number support from NADRA. BISP beneficiaries collected of beneficiaries being defrauded out of their cash their cash benefit from various franchises operated grants. These mechanisms had to be abandoned. 15 A certificate (directive to pay) issued by the Government of Pakistan for a specified sum of money, to be paid upon demand to the intended recipient. It is a pre-funded instrument generally preferred by recipients without a formal bank account who can withdraw cash upon presentation of any government-issued document verifying their identity (CNIC). 16 Pakistan Post is a state enterprise that is Pakistan’s largest postal operator, with over 13,419 post offices across the country, servicing at least 50 million people each year. FIGURE 22 EVOLUTION OF BISP PAYMENT SYSTEM Pak Post Benazir Smart Card Mobile Banking Money Order April 2010 Dec 2010 2008 A Pilot launched in 4 Districts A Pilot in 5 Districts Initial payment mode Beneficiary-centric Social Protection Accounts (Being Piloted) Benazir Debit Card Biometric Verification System Feb 2012 Piloted in March 2015 Rolled out in 124 Districts National Rollout from June 2016. across the country. Refined Payment Model through new Banking contracts from 10th Oct. 2019 Source: BISP 53 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program Generic ATM cards and the Benazir Debit Card (BDC) the One Woman One Account initiative, which 7. Social protection delivery s other sectors, such as heal were employed from 2012 onwards but, again, offered beneficiaries an optional savings account benefits, and legal services challenges arose due to beneficiaries’ low levels of to encourage them to save; however, this initiative 8. The dual challenges of incl the continuous improveme literacy and unfamiliarity with digital technology. failed. Third, the program made agent networks centered approach. Beneficiaries could use these cards to withdraw (POS) BVS enabled in addition to ATMs. At the same cash from any ATM in Pakistan or from designated time, all BVS devices were geo-tagged and equipped point-of-sale (POS) terminals of BISP’s partner banks with live fingerprint detection (LFD) technology across the country. While the BDC was convenient and electronic payment devices. NADRA specified for beneficiaries, it was not without its challenges. standard protocols for using such devices to ensure Most partner banks issued generic debit cards (i.e., that payments were being made to beneficiaries. To an ATM card without the name and/or picture of make the system more responsive, BISP’s payment the beneficiary) with a generic passcode of 1234, complaint management system was integrated with with instructions to change the passcode before the complaints management systems of all partner using it. However, most women never changed the banks so that BISP could track all payment-related password and/or gave their card to “middlemen” to queries and complaints (see Section 4.4.1.). Finally, withdraw cash on their behalf in exchange for a small BISP specified minimum quality control measures “commission.” The generic nature of the BDCs also and reporting standards for banks, one of which was led to a mix-up of BDCs and passwords, making the API integration to enable real-time reporting. transaction impossible. Women routinely misplaced their BDCs, requiring them to apply for replacements Nevertheless, the current BISP payment process has which was a lengthy process. While the system several limitations. First, the current disbursement continued to operate in some parts of the country process allows only one financial institution to till recently, it was an expensive undertaking. operate in a particular area (with a limited option for beneficiaries to receive their payments through telecommunication companies and POS agents). BIOMETRIC VERIFICATION SYSTEM (BVS) This limits interoperability as well as the ability of beneficiaries to choose to withdraw their money BISP currently uses biometric information from from the most convenient or preferred financial beneficiaries collected by NADRA via a thumb institution. Second, the limited availability of impression when the CNIC is issued. Beneficiaries withdrawal platforms provides more opportunities are notified by text as soon as their cash transfer for disbursement agents to indulge in exploitation funds are credited into their accounts. To collect and rent seeking, compromising the delivery of their money, a beneficiary presents her CNIC at a the cash transfer to the targeted beneficiaries. designated BVS-enabled POS terminal at a BISP Third, since there is no automated reconciliation partner bank. After the beneficiary’s thumb print is system and each disbursement region is handled verified by the BVS system in real time, the payment by a single financial institute, BISP must do separate is released. This system is easy to use and ensures reconciliations for each payment partner. This results that the cash transfer is withdrawn and retained by in delays in the posting of accounts, the payment the beneficiary herself. It requires the beneficiary to of refunds, and the settlement of complaints. Also, be physically present at the POS terminal, although the current disbursement mechanism is based on this condition may be difficult for some women BISP paying its payment partner banks up front, given household responsibilities and social norms. which creates challenges of liquidity and funds Despite these concerns, the BVS payment system management as well as scheduling issues for BISP. has more than 90 percent withdrawal rates. BISP has made significant changes to the BVS payment system design to further reduce inefficiencies and leakages. In 2019, BISP took a series of actions to increase accessibility and financial inclusion and improve administration. First, BISP divided the county into three clusters, instead of the previous six or seven clusters, to streamline system administration. Second, BSIP launched THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 54 FIGURE 23 BISP PAYMENT WORKFLOW BISP 1 2 3 4 Payment Generation Approval Cheque Preparation Cheque Endorsement BISP generates the Approval for the BISP Intimation sent to Cheque is endorsed payment list payment is taken the NBP from AGPR BANKS BENEFICIARY RS 5 6 7 Bank Account Posting Entitlement Payment Payment Withdrawal Cheque deposited to Beneficiary Beneficiary withdraws each bank for a set of Entitlement is Paid the money from beneficiaries assigned bank Internal A/c Posting POS Source: Background Paper on IT System of NSER 55 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program Both UCT and CCT disbursements are made transfers. They cannot make deposits or use this 7. Social protection delivery s other sectors, such as heal via the same payment mechanism. While the account for any other purpose. Moreover, they must benefits, and legal services 8. The dual challenges of incl Kafaalat UCT has no precondition other than withdraw their cash transfer in full. These accounts the continuous improveme centered approach. the verification of identity, both CCT programs were approved by the State Bank of Pakistan only have attached preconditions, as discussed in the for BISP transactions. This payment system allows previous section. Nevertheless, the cash transfers beneficiaries to withdraw their cash transfer only under these programs are disbursed at the same through the LMAs, which limits their choices. time as the UCT,17 despite following different Another negative is that this system exacerbates processes of verification and compliance. During the existing low level of financial literacy among BISP each payment cycle, the Cash Transfer (CT) Wing beneficiaries, which could be mitigated if they were of BISP follows these steps: (i) funds planning; (ii) allowed to become more familiar with the formal setting UCT payment parameters; (iii) calculating financial system by using other outlets. Beneficiaries payment amounts; (iv) generating the UCT payment now have the choice to withdraw their money from list; (v) submitting the UCT payment list to the CT their LMA account or to transfer it to a savings Wing for pre-audit and approval by the Finance and account at the same bank that was provided by the Accounts Wing and the BISP Secretary; (vi) disbursing BISP program as part of its payment mechanism. As the funds; (vii) notifying the beneficiaries, and (viii) soon as the funds are credited to the beneficiaries’ reconciling funds. accounts, they are notified by text. BISP has developed payment software for Funds reconciliation refers to the process of the smooth processing of payments, but it is assessing how much money has been disbursed housed under the MIS Wing, which often creates to whom and when. For this purpose, the MISs of inefficiencies. The MIS Wing maintains the master the banks and BISP are integrated so that BISP can list of beneficiaries, but it can only pay them once monitor the withdrawal of each disbursement in real it receives detailed instructions from the CT Wing time. BISP carries out a final reconciliation with each before each quarterly disbursement. CT Wing sets payment agency at the end of each quarter. This parameters such as the amount to be paid to allows the BISP to identify areas where withdrawal each beneficiary as well as the date from which rates are less than 100 percent, so that it can take it must be paid, and the MIS Wing then transfers proactive measures at the local level to ensure that the specified amounts. After the disbursement, the beneficiaries withdraw their funds. When funds are MIS Wing shares a report with the CT Wing. BISP not withdrawn, they remain in the beneficiary’s LMA follows a standardized sequence for generating for nine months before getting credited back to and disbursing payments as follows: (i) it pays all BISP. Moreover, if a beneficiary wants to withdraw a beneficiaries within each selected geographical area payment from a district where she does not live, she (province or district); (ii) it pays all beneficiaries using first has to go the BISP tehsil office to inform them. either electronic or manual payment modalities; This rule applies only to those BISP beneficiaries who (iii) it prepares the quarterly UCT payment list; and are being served by the Habib Bank, which enacted (iv) it maintains a historic record. The finalized UCT this rule to prevent fraud at the POS. payment list is sent to the CT Wing. However, it would be better and more efficient if the CT Wing could create these lists without having to rely on the Technology Wing. Banks disburse the cash transfers through special virtual accounts created specifically for BISP beneficiaries. Accounts are called Limited Mandate Accounts (LMA) and beneficiaries can use this account only for withdrawing their BISP cash 17 The UCT used to be provided on a quarterly basis, but this was amended to semi-annual pay-outs during the pandemic but has since reverted to being paid quarterly. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 56 BOX 11 USING THE BISP PAYMENT SYSTEMS IN SHOCK RESPONSE Once a target population has been identified for a shock response, BISP uses the BVS and digital payment systems to make cash transfers directly to those households. However, in the context of climate shocks, such as the most recent floods, these systems can break down because of telecommunication problems or the inability of the recipients to physically access these payment points. Since BISP has access to real-time information on when and where cash withdrawals are made, it is possible to plan alternatives in a timely manner. After the flood response, although payments were disbursed quickly, cash withdrawal rates remained low in areas that were inaccessible and/or where the local infrastructure had been badly damaged. By triangulating data on payment withdrawals with location data in NSER, it was possible to quickly identify areas where withdrawal rates were low even weeks after disbursement. Based on reports from BISP field offices on accessibility, boats were arranged for the payment service providers, who set up temporary camps in the relevant areas until all beneficiaries were paid, thus improving the efficiency and effectiveness of the crisis response. The disbursement of the BTW and Nashonuma from the BISP. Their awareness of potential scams and cash transfers are reliant on a detailed compliance fraud—such as texts that claim to be from the BISP mechanism. The quarterly verification of children’s and that ask for personal information—is also limited. school enrollment and attendance records is a BISP has tried to address these issues by constantly massive undertaking (See 4.4.2). Verification often communicating with beneficiaries about how they takes a long time, which means that the payment can identify scammers and protect themselves from of the BWT transfer is not perfectly aligned with the such fraud. It is also in the process of designing a timing of the UCT disbursements. This can lead to CCT financial literacy curriculum for beneficiaries. payments being pushed into the next payment cycle. Often, an amount accumulated over several quarters The experience of BISP’s multiple payment is paid out to the beneficiary in a lump sum, creating mechanisms has invariably pointed toward an information asymmetry where beneficiaries are one common factor, which is the preference of not always aware of the total amount that they are beneficiaries for human contact whether with BISP due. This can be exploited by payment agents who officials, bank POS agents, or the unlawful ATM may withhold a percentage or charge the beneficiary middlemen. Despite a strong effort by the BISP to a fee when they withdraw their cash. In contrast to educate beneficiaries about their rights regarding cash the BTW, the Nashonuma payments are processed grants, cases of unlawful agents charging commissions immediately as a beneficiary can fulfill all the co- or fees (sometimes referred to as the “social tax”) are responsibilities during her visit to the program’s rampant. Anecdotal evidence suggests that women facilitation center. (See 4.4.2) often pay 10 to 20 percent of their benefit to the payment agents to access their cash transfer because Overall, the evolution of BISP’s payment system of the inherent power imbalance in which they are was driven by its commitment to transparency completely dependent on the payment agent to give and better management and administration them their cash. In some cases, if women refuse to pay rather than ensuring that beneficiaries have a this commission, they are asked to come again the good experience with the system. Digitalization next day, thus incurring more travel and opportunity has been instrumental in ensuring transparency costs. A pending task is for BISP to delve deeper into the and compliance as manifested by a reduction in the problem and assess how a human-centered approach number of complaints and of the incidence of fraud. could be used to improve the situation. When such However, it has also highlighted the low financial and complaints are filed through official channels, they technological literacy of the target population. Most are resolved on a case-by-case basis, but BISP has not women beneficiaries are illiterate and rely on their been able to adopt a systemic approach that penalizes children to read texts and other communications POS agents for exploiting and harassing beneficiaries. 57 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; Aware of these shortcomings, BISP is piloting a new This new disbursement system is being introduced and can improve program 7. Social protection delivery s payment system with the aim of creating a more in the context of a broader global trend toward other sectors, such as heal benefits, and legal services open beneficiary-centric environment. BISP is beneficiary choice and modern digital infrastructure 8. The dual challenges of incl the continuous improveme testing the concept of providing beneficiaries with for social protection payments. Countries as centered approach. fully mandated bank account, in contrast to limited diverse as India, Zambia, and Colombia are moving mandate account, which will offer multiple options for toward giving beneficiaries more choices of financial withdrawing cash, including bank branches, ATMs, and accounts. Combined with the use of shared digital mobile money or POS agents. The payments will be infrastructure like RAAST in Pakistan, APBS in India, eventually facilitated by RAAST, an instant payment or GEWEL in Zambia, these changes represent a system owned and operated by the State Bank of global trend toward a more modern government- Pakistan. RAAST is a payment gateway, which enables to-person (G2P) infrastructure (Hobson et al. 2022). end-to-end digital payments among individuals, Increasing beneficiaries’ choices has the potential to businesses, and government entities, offering the increase competition, reduce costs, and improve the potential to reduce delays in payment, while reaching customer experience. More important, it recognizes a wider range of financial institutions. As the number beneficiaries as customers with complex financial of participating financial institutions grows, BISP’s lives, rather than as passive recipients. Kenya’s grievance redress mechanism, funds reconciliation, experience in moving toward the electronic payment and other administrative activities may become more of social benefits and enhancing access to financial complex and will require careful management. services provides valuable insights for BISP’s efforts to improve its payment system (Box 12). BOX 12 KENYA’S INUA JAMII PAYMENT SYSTEM Despite being a lower-middle-income country, Kenya has, for many years, been recognized as a pioneer in digital financial inclusion. In 2013, the Kenyan government initiated the electronic payment of social benefits in a move toward establishing a more modern and efficient payment system. In 2018, the Inua Jamii social protection program restructured its payment system to offer beneficiaries a choice of payment service providers (PSPs), which were all commercial banks. This move was intended to promote competition among providers and increase convenience for beneficiaries. A study was conducted to evaluate the effectiveness of the payment system from the perspective of beneficiaries, including their opinions on its convenience and competition. Additionally, the study examined whether these digital government-to-person (G2P) payment programs had increased financial inclusion in Kenya, recognizing that Kenya already had high financial inclusion rates due to the widespread use of M-Pesa digital wallets. The study found that recipients strongly supported making payments through financial accounts. Most respondents considered this to be a good system, with some preferring the commercial bank channel and others expressing a preference for direct payments through wallets. The study also found strong support among beneficiaries for being given a choice of PSP where feasible. However, the single payer G2P model was also found to be effective, depending on local conditions. From a gender perspective, the study found a high level of gender parity in the Inua Jamii program, with 90 percent of women claiming it was no more difficult for them than for men to open or operate an account. However, women were less likely than men to report knowing that they were able to change PSPs and were less likely to take advantage of the limited added functions offered by their accounts. Despite this, both men and women were equally likely to claim that the new choice-based payment model was superior, and women across all cohorts were no more likely than men to dislike the system. Source: Gleb et al 2023 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 58 4.4 MANAGE This sub-section discusses the case management and and exit decisions focusing on its evolution and grievance redressal mechanism of BISP, monitoring of continuous improvement. conditionalities for the CCTs, monitoring of payments, 4.4.1 Case Management and Grievance Redress Mechanism Grievance redress mechanisms (GRMs) are an BISP has a comprehensive case management system important component of the delivery of social (CMS) that is available to existing beneficiaries, protection as it gives the public a voice and can those who have exited the program, and potential create an effective feedback loop. A GRM provides applicants as well as to community members. In 2012, beneficiaries and the public with a voice in the BISP launched its original case management system administration and performance management (CMS) (Figure 24). On receiving a complaint, the CMS of a social protection program. Ideally, the GRM created a computer-generated complaint ID to track it. could provide more than one channel to ensure The system only dealt with complaints related to CNIC that the maximum number of people can provide updates, duplicate CNIC clearance, address updates, feedback to program administrators and seek and eligibility-related appeals. The CMS was developed resolution of their complaints. Common grievances by NADRA and remained operational till 2017. The CMS can involve unclear program guidelines, a lack of paved the way for 1.8 million females to become BISP awareness of the program because of insufficient beneficiaries to date. In 2015, when BISP payments were outreach, possible inclusion and exclusion errors, an being made through debit cards, the program’s MIS unsatisfactory package of benefits, problems with developed a second version of the CMS, the Payment the payment of benefits or the provision of services, Complaints Management System (PCMS), to handle mistreatment by frontline staff, or the inefficacy of payment-related complaints. Whereas the original CMS the GRM itself. When aggregated, grievance data help operated through the exchange of emails and excel program administrators and decision-makers judge sheets between the field offices, headquarters, and the efficiency and effectiveness of the GRM based on banks, the PCMS desktop application is now being used the experiences of beneficiaries and other citizens. at the BISP headquarters and in all tehsil and district It is essential that all stakeholders understand how offices. Initially, it handled only complaints related to grievances will be collected and resolved (Lindert the debit card and the loss and replacement of PIN et al. 2020). numbers. Later, more types of complaints were added after the debit cards were discarded and biometric verification became the mode of payment.18 18 BISP Presentation, January 2023. 59 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program FIGURE 24 KAFAALAT CASE MANAGEMENT (GRM) 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme centered approach. UCT Eligible Beneficiaries NSER CASE MANAGEMENT Pending Missed Out Eligibility Beneficiary HHS Appeals Data Updates UCT CASE MANAGEMENT Quality of Payment Communication Monitoring Service Claim of Resolution and Reporting Grievance UCT ACTIVE Beneficiary Master File There are various ways to register complaints and In 2022, BISP strengthened its PCMS by integrating a wide range of communication channels. These it with partner banks. All complaints are monitored options include: (i) mailing complaints to the BISP at BISP’s headquarters. Complaints are categorized headquarters located in Islamabad; (ii) calling as high, medium, and low in severity and are assigned the national toll-free hotline;19 (iii) submitting an expected turn-around times (TATs) accordingly. online complaint through the BISP website, which is According to BISP’s contracts with the banks, the exclusively available to vendors and contractors; (iv) TATs for each type of complaint range from 24 hours making an in-person visit to the BISP headquarters to four days. BISP could impose penalties on banks or any of the 16 zonal offices (located throughout the for TAT breaches since October 2022.20 BISP has provinces/regions), 140 district offices, or over 400 adopted an exceptions policy21 for those beneficiaries tehsil offices across the country, and (v) submitting a who cannot be verified biometrically (such as older complaint through BISP social media accounts such women whose thumb lines have faded or amputees as X (formerly known as Twitter). Most beneficiaries with both thumbs gone). These people will now be prefer to visit an office to lodge their complaints provided with debit cards with their name, CNIC instead of calling the helpline, even if they are aware number, and photograph on the card to enable them of the helpline number and how to use it because to withdraw their payments transparently.22 Broadly they feel more comfortable expressing their concerns speaking, the PCMS handles complaints related to in person to another human being. the BVS, payments, and the quality of service. Most of the complaints received are related to the BVS. In 19 The toll-free number is 0800-26477, which is extensively advertised by the BISP. 20 BISP Presentation, January 2023. 21 Barring these exceptions, BISP has discontinued using debit cards and replaced it with the BVS. 22 CRISP Stakeholder Engagement Plan, January 2021. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 60 addition, there are various subtypes of complaints The grievance resolution authority has been within these three categories, resulting in over 30 decentralized to improve the GRM. The grievance/ different types of complaints that can be processed complaint resolution authority has been devolved through the PCMS. The PCMS also offers beneficiaries from the Head Office to the tehsil offices to accelerate various other services such as the ability to review resolution time and to provide beneficiaries with their transaction history, dispute a blocked status, a more familiar and accessible environment. update next of kin and contact information, submit Beneficiaries already visit tehsil offices to determine appeals for transgender beneficiaries, report the their eligibility and to enroll themselves in the BISP death of a beneficiary, and generate reports for BISP program, and now they can also lodge grievances headquarters and field staff (Figure 25). related to targeting, enrollments, or payments, or make a general complaint. FIGURE 25 BISP PAYMENTS COMPLAINTS MANAGEMENT SYSTEM Note: User interface from BISP’s Complaint Management System 61 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; All BISP offices are connected online with the CMS BISP discontinued cash transfers for those that had and can improve program 7. Social protection delivery s system to be able to process these complaints. The exited the programs because beneficiaries were other sectors, such as heal benefits, and legal services BISP operates three complaints systems: (i) the case questioning why they were no longer eligible for 8. The dual challenges of incl the continuous improveme management system (CMS) for enrollment in the benefits.23 centered approach. Kafaalat UCT; (ii) the BTW CMS introduced specifically for BTW enrollment complaints; and (iii) the PCMS Action needs to be taken to increase awareness introduced for complaints related to payments of the GRM. This applies to all beneficiaries but is of the UCT and CCTs. All BISP staff can access all particularly important for vulnerable groups. The three complaints systems online. The websites for GRM needs to be known and accessible to less the three systems enable BISP operational teams well-educated people and persons with disabilities. to receive and address beneficiaries’ complaints Information about the GRM needs to be widely and enable existing and potential beneficiaries to circulated in dedicated media campaigns using print, understand the procedures involved in lodging electronic, and social media, so that citizens and appeals and complaints, including appeals by organizations in all project areas are aware of its households who have either not been surveyed or existence and can use it to report their issues and whose poverty score is above the eligibility cut-off complaints. score, but they consider themselves to be “poor” and eligible for BISP benefits. Moreover, several aspects of BISP’s grievance redress system are fragmented, and coordination The CMS/GRM system has been designed to be among staff working across the system is limited. inclusive. Anyone can check their eligibility status, Different databases exist for each initiative although report an exclusion error, lodge a complaint or appeal, BISP keeps a master database of beneficiaries. and seek other information. For this purpose, BISP However, when it comes to complaints, each BISP has set up a centralized call center that is available wing has a separate database that is maintained by across Pakistan from 8.00 am to 5.00 pm. The staff the MIS Wing. There are no rules that standardize provides information about new initiatives, registers responses to grievances or advise program officials complaints, and resolves some complaints on the spot on how to tackle different kinds of complaints and such as those regarding CNIC updates and appeals appeals. Timelines have been set for resolving some on eligibility. The status of all pending complaints but not all types of complaints. The resolution of is available on BISP’s internal website with case IDs complaints is often neither timely nor transparent, for tracking purposes. Active complaints remain on which has led to low confidence in the process (ADB the website until they are resolved. Between mid- 2021). Moreover, tehsil offices are manned by only April 2012 when the CMS became fully functional and three people on average who attend to multiple October 31, 2020, BISP received almost three million tasks including complaints (ADB 2021). complaints, out of which 98 percent were processed within 72 hours of receipt. To address these issues, BISP is enhancing its GRM by integrating the separate MISs. BISP is developing However, not all beneficiaries know about the a UCMS with a single log in for users who can then GRM. According to a third-party beneficiary select the avenue to submit their complaint. With feedback operational review carried out by the the development of the UCMS, all complaints data OPM in 2020, only 14-37 percent of beneficiaries will be stored in one grievance redress database. BISP had any knowledge of BISP’s grievance redress can learn lessons from the experiences of countries system. Only 3-5 percent of beneficiaries were with advanced GRM systems such as Türkiye (Box 13) aware of BISP’s helpline, even though beneficiaries and Jordan (Box 14). As the UCMS is developed, it will have access to mobile phones. The percentage of need to be integrated with the banks and aligned beneficiaries who had ever had a grievance ranged with the new payment model. from 10-27 percent. It is also important to note that the number of grievances also increased after the 23 Out of the current 9 million beneficiaries, slightly over 2 million are considered to be continuing beneficiaries as they were also captured in the previous survey. The remaining beneficiaries are new and were not previously included in the program. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 62 BOX 13 THE GRIEVANCE REDRESS MECHANISM IN TÜRKIYE The grievance redress mechanism in Türkiye uses a combined centralized and decentralized model to deliver social assistance. It includes a range of channels to ensure that all recipients can submit grievances at their convenience. These channels are: • An integrated GRM module in the Integrated Social Assistance System (ISAS) can be used for any type of grievance related to social assistance programs. Recipients can submit an appeal at the Social Assistance and Solidarity Foundation (SASF) in their district using a standard form provided by SASF. These appeals are then sent to ISAS for evaluation during the District Board of Trustees’ weekly decision meetings. The Board is required to resolve the matter and notify the recipient by text within a month in accordance with a regulation issued by the General Directorate of Social Assistance (GDSA). • The Alo 144 hotline is a call center that provides information and support to recipients, including information on the status of applications and payments. It is one of the main information channels used by all social assistance programs. It can be used to coordinate between SASFs about grievances and to direct citizens to a particular SASF. Pre-COVID-19, the hotline received over one million calls on average per year. • The CIMER (Presidency Communication Center) is an online portal for all demands, appeals, proposals, or complaints from citizens. Recipients can use this channel to direct grievances to the Ministry of Family and Social Services (MFSS) or to the GDSA. The law on information (Law no. 4982) and the law on petition (Law no. 3071) require that these grievances should be resolved within 30 business days by the relevant public authority. • The Ombudsman’ Institution is an independent complaint mechanism that accepts appeals through an online portal or in paper form, but it takes much longer than other channels for complaints to be resolved. • The E-Government (e-GOVT) portal is linked with the ISAS and enables recipients to check for updates on their application or to submit queries. In 2016, 9,000 service queries per day were submitted through the e-GOVT portal. Source: World Bank (2022) It is critical to monitor the effectiveness of the GRM. resolved could be analyzed as part of a review of The KPIs include monthly reporting on the number the efficacy of the GRM. Soliciting the views of the of grievances received, the number that have been stakeholders for whom the GRM was designed would resolved, and the number that remain outstanding. increase the credibility of the mechanism and the The trends and time taken for grievances to be program itself. 63 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; BOX 14 and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl JORDAN’S NATIONAL AID FUND the continuous improveme centered approach. The National Aid Fund (NAF) is the cornerstone of Jordan’s social protection system, providing assistance to vulnerable Jordanian families (excluding refugees) and facilitating their access to other resources in the social support network. As the primary safety net provider for the Government of Jordan, NAF implements various poverty reduction programs, providing cash transfers to 29 percent of the population. NAF has a widespread network of 42 branch offices and 16 sub-branch offices in remote areas. Previously, NAF relied on a manual Complaints and Handling Mechanism (CHM) that could be accessed at public service desks at its headquarters and branches located throughout Jordan. The responsibility of these desks was to receive paper documents containing complaints and grievance applications, process the referrals to relevant units, and provide feedback from NAF staff who worked on the cases. The complaints mainly consisted of grievances about being excluded from a particular program and requests to be included in NAF’s assistance programs. However, these complaints and applications were not documented or saved for reporting and tracking purposes. NAF has made substantial changes to its CHM to improve communication with its beneficiaries and provide a centralized system to receive queries and feedback through multiple channels. Currently, the CHM includes a Call Center, an online feedback form on the NAF website, and social media platforms, all of which are linked to NAF’s broader MIS database. To enhance the performance of its call center, NAF conducted an evaluation to identify areas for improvement. The evaluation found that the NAF call center is effective in assisting beneficiaries, but there are challenges in terms of its accessibility, awareness of its existence, and its responsiveness and cohesion in resolving complaints. Beneficiaries who reach the call center are highly satisfied, but many are unaware of its existence and purpose. The call center’s effectiveness is due to the digitization of NAF systems, which enables it to respond more swiftly. However, there is a need for improved workflows and institutional sustainability through links to the larger national social protection framework. Source: World Food Program (2022). External Evaluation: Complaints Handling Mechanism (CHM) Evaluation at the National Aid Fund (NAF)/ Jordan. February 2022 4.4.2 Monitoring Conditionalities for the Benazir Taleemi Wazaif and Nashonuma CCTs Institutional coordination—both vertical and actors. The central ministries typically oversee horizontal—is critically important for monitoring CCTs, manage their information systems, including beneficiaries’ compliance with conditionalities. the compliance verification modules, decide on One of the factors that makes compliance monitoring any consequences for non-compliance, and feed complex is the number and diversity of actors compliance information back to the payments involved, including frontline program personnel, department for payroll processing. Subnational teachers, health care workers, subnational agencies, (regional or local) agencies typically serve as and central agencies, including the social, education, conduits for transmitting information, overseeing the and health ministries. Monitoring CCTs requires monitoring process, and interacting with the service considerable vertical collaboration between the providers. Various approaches govern these vertical central ministries, subnational agencies, and local collaboration arrangements (Lindert et al. 2020). THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 64 BENAZIR TALEEMI WAZAIF Most schools maintain manual records, which can EDUCATION CCT sometimes be incomplete. The provincial education departments have hesitated to make teachers Collecting school data to monitor compliance with responsible for sharing the attendance data, the conditions of the CCT is a time-consuming task, despite BISP offering to compensate the efforts. and BISP’s experience has been mixed. BISP tried to Facing similar issues, other countries implementing coordinate with provincial education departments, CCTs have distributed responsibilities between but even so, obtaining regular data on student education authorities and program staff in a variety admission and attendance has been challenging. of arrangements (Box 15). BOX 15 ROLES IN THE VERIFICATION COMPLIANCE WITH CCT CO-RESPONSIBILITIES: COUNTRY EXAMPLES Countries that operate CCTs are aware that the monitoring of conditions is information intensive and time sensitive. In the case of education, knowing how information flows from schools to program managers is key so that the data can be used to determine payment, need for follow-up, and penalties. Worldwide, the distribution of operational roles and responsibilities among various actors (particularly the workload of teachers vs. program staff) differs depending on the country context. These arrangements have evolved in some countries, particularly with advances in the use of information technology by schools. At the start of Colombia’s Familias en Accion CCT, beneficiary mothers received forms from the municipal program liaison, had the forms completed by the teacher, and returned the forms to the liaison who input the data into the MIS. The information was consolidated by program staff (at the regional and national level) and used to generate payment lists. Now, teachers input attendance data for beneficiary children directly into a module of the program MIS, based on parameters supplied by program managers. Program staff verify and consolidate the information as before. Bolsa Familia in Brazil has always checked compliance on school attendance conditions through consolidated computerized data provided by the Ministry of Education to the Ministry of Social Development. The process used by the previous CCT in Mexico and the PATH CCT in Jamaica are similar. In Mexico, state-level program staff prepared the required forms, which were distributed to schools by state education authorities. Teachers filled out the forms and returned them for certification to the state education authorities. CCT program staff at the state level input the data into the MIS to be consolidated and used to generate payment lists by national level program staff. In Jamaica, the national program staff generate the required forms and program staff at the parish level distribute to schools and collect the completed forms. Parish level program staff then input the data into the MIS for consolidation and generation of payment lists at the national program level. In Chile, verification of compliance with the education condition for the Single Family Allowance (Subsidio Unico Familiar – SUF) is done by municipal staff who check school enrollment documentation provided by beneficiaries once a year (the Single Family Allowance-SUF in Chile) and digitally record and upload documentation. 65 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; Compliance monitoring for the education CCT has The activities include: and can improve program 7. Social protection delivery s undergone significant changes since its inception. other sectors, such as heal benefits, and legal services Given the less than satisfactory performance of the • Update the CM module: This module is updated 8. The dual challenges of incl the continuous improveme agencies initially recruited to monitor compliance, with the latest data on beneficiary children and centered approach. the BISP currently employs a team of 900 field their co-responsibilities based on data from officers to carry out monitoring and compliance schools to ensure that it is accurate. verification. • Produce beneficiary children school The NSER Wing is responsible for monitoring concentration maps: This involves producing compliance with the BTW conditionalities. The maps that illustrate the geographical BISP sets targets for compliance monitoring for distribution of beneficiary children and their each tehsil field office, and compliance monitors schools. These maps can be used to identify (contractual employees) visit schools in a continuous areas where there may be a higher likelihood of cycle using geo-tagged android tablets to record non-compliance. attendance data on the children of BTW beneficiaries. A small monitoring unit (about 30 people) at the • Plan logistics and train personnel: This activity BISP headquarters operates an MIS to monitor the entails planning the logistics of the monitoring progress of the compliance monitors. process, such as the allocation of personnel, materials, and equipment required, and training Compliance monitoring is a continuous process personnel including the compliance monitors in that begins after a family enrolls in the BTW. The the use of the tools and materials that they will main objective of compliance monitoring is to need. assess the extent to which beneficiary families are meeting their co-responsibility to send their children • Communicate with local beneficiaries about to school in return for receiving cash transfers. There the compliance monitoring process: This are two types of co-responsibility: admission and activity involves communicating with beneficiary attendance. Compliance monitoring aims to: (i) families at the local level to explain and engage identify those BTW children whose families have them in the monitoring process. fulfilled their co-responsibilities and verify their eligibility for the cash transfer; (ii) identify any BTW children whose families have not complied with their BISP has learned that when transitioning to co-responsibilities and trigger appropriate alerts; digital solutions for compliance, it is important to (iii) apply consequences for non-compliance in such consider infrastructure and the digital capabilities cases; (iv) follow up with BTW families whose children of key stakeholders. Although the BISP adopted a are not complying to encourage their compliance; biometric sign-in system for students entering and and (v) suspend beneficiary children and their leaving school, this was not universally welcomed. families from the program if they fail to fulfill their Technology could not prevent students from attendance co-responsibility for three consecutive signing in and then immediately leaving the school. compliance quarters. Compliance monitoring In some cases, the system also caused significant ensures that only eligible BTW children receive cash delays because of long lines at the school entrance transfers and fosters greater compliance with co- in the mornings. Vigilance on the part of teaching responsibilities among beneficiaries. staff was necessary but could be challenging to manage due to teacher absenteeism or collusion. As Compliance monitoring of BTW involves a series mentioned earlier, most schools have not converted of preparatory activities that are essential for to digital attendance methods. One reason might be both manual and automated processes. These that the teaching staff lack the skills to use digital activities require meticulous planning, preparation, technology or simply prefer traditional paper-based and execution to ensure that the monitoring process mechanisms. In addition, internet access is not runs smoothly and achieves its objectives. consistently available throughout the country, with some areas having only limited or patchy coverage. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 66 NASHONUMA HEALTH AND antenatal and post-natal care by pregnant and NUTRITION CCT lactating women, the routine immunization of their children, and the consumption of at least 90 percent Compliance monitoring is organized at the of the specialized nutrition foods provided by the facilitation centers established in tehsil health program. Compliance monitoring takes place on a facilities. The enrollment data on program quarterly basis aligned with the quarterly payment beneficiaries (who are pregnant and lactating of all BISP cash transfers and is simplified because all mothers who are beneficiaries of the Kafaalat UCT services and payment take place in the facilitation program) are uploaded into an MIS. The payment center during the visit of the beneficiary. has several conditionalities including the use of 4.4.3 Monitoring of Payments At present, BISP uses the services of two partner The payment cycle dashboard is maintained and financial institutions to disburse its cash benefits. tracked by the Cash Transfers Wing including The relationship is governed by a comprehensive fraudulent transactions and performance agreement detailing the arrangements including, benchmarks, but not beneficiaries’ experience. real-time reporting by the partner banks to the BISP The frequency, nature, and number of complaints of cash withdrawals by beneficiaries, reconciliation about service quality, payment process, and of withdrawn amounts by the banks and the BISP, the other challenges faced by beneficiaries are not channels available to beneficiaries to withdraw their systematically focused upon through beneficiary money, the performance of the banks at payment feedback or satisfaction surveys, by either the sites, and what bank infrastructure is available to banks or M&E department. The MISs focus on support the BISP’s requirements. It is of paramount improving payment administration rather than importance to monitor information as it flows actively pursuing indicators related to beneficiary between the BISP to the bank(s) and back to the requirements In addition, the payments complaints BISP in a continuous, uninterrupted cycle to ensure management system (PCMS) has not been integrated all performance related KPIs are tracked. All data into the dashboard, which points to a lack of focus related to BISP payments at every POS or ATM is on beneficiaries’ needs. This might be one of the relayed to the BIPS MIS automatically in real time. reasons for the continued concerns surrounding The biometrically enabled terminals are geofenced the quality of service of the point-of-service agents. and geotagged and cannot be moved from the designated point-of-sale sites for control purposes. The steps involved in the payment monitoring cycle are: (i) the generation of lists of beneficiaries and the payments that they received by the CT Wing with the help of the Technology Wing; (ii) the verification of the generated lists conducted by the Finance and Accounts Wing according to predefined parameters; (iii) the use of a payment cycle dashboard to track the entire payment cycle value chain; (iv) the use of a fraud alert dashboard to monitor any abnormal transactions detected by data on geofencing and geotagging, the number of attempts at biometric verification, and the number of out-of-district withdrawals and withdrawals outside normal operating hours, among others; and (v) funds reconciliation by the Finance and Accounts Wing with the help of the Technology Wing. 67 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; and can improve program 4.4.4 Exit Decisions 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl the continuous improveme One of the major decisions in beneficiary Before exiting these ineligible beneficiaries from the centered approach. management is about moving beneficiaries out of program, the BISP conducted a process of verification the program. Well-functioning beneficiary rosters and validation. After this exercise, the BISP found tend to be highly dynamic tools: they are constantly about 50 percent of existing beneficiaries were no incorporating new beneficiaries through the longer eligible, or about 2 million beneficiaries. The enrollment phase and moving out those who have BISP did not formally notify these exited beneficiaries graduated from needing support as well as those of their discontinued status by letter or text. Instead, who, due to a change or correction, no longer meet a web portal has been made available where the program’s criteria. There are normally four steps beneficiaries can check their status. Beneficiaries involved in the process of moving beneficiaries off can also text their CNIC to the number 8171 on their the roster: an exit trigger, criteria validation, an exit mobile phones to check their status. decision, and beneficiary notification. Exit triggers are changes in a beneficiary’s basic information As the NSER transitions to a dynamic registry, BISP or compliance metrics (resulting from the update will have to develop a strategy on exit decisions. The process and conditionality monitoring respectively) NSER plans to ensure that every household updates that indicate that a beneficiary no longer falls within its data at least once every four years. With regular the program’s eligibility parameters. data updates, BISP will be able to monitor changes in socioeconomic circumstances, which may lead to The BISP made its first beneficiary exit decisions currently eligible households not qualifying for the when it transitioned from using the parliamentary program. At present, BISP does not have a strategy on targeting system to using poverty scorecards. The exit decisions—how often they should be made; what transition involved identifying and selecting eligible should trigger the decisions; how the decisions will be beneficiaries of the Kafaalat program based on their communicated; whether existing beneficiaries have data in the NSER. In this process, the BISP had to a right to appeal; and, whether existing beneficiaries exit beneficiaries who no longer met the eligibility will be eligible for other types of support. BISP is criteria as identified by the poverty scorecard considering introducing a Hybrid Social Protection system. The beneficiaries selected through the Scheme (Box 16) that will incentivize long-term savings previous parliamentary selection system were to be among households that are no longer eligible for BISP completely phased out by the summer of 2011 if they programs. This can become part of the exit strategy. were found to be ineligible under the new scorecard criteria. However, since they might still be vulnerable, In the specific context of the BTW program, a child the government decided to let those households will be removed from the program under certain appeal their status if their score was between 16.17 conditions (such as non-compliance with the and 20 and if they had other characteristics such condition of enrolling in and attending school) as an elderly person or a person with disability in resulting in the cessation of their cash transfers. The their household and a certain number of children in BTW separation policy classifies a beneficiary child as school. This allowed for an acceptable purging of the being either graduated, suspended, or exited (Table parliamentary rolls since the same categories were 2). Once all the children in a BTW family have exited, used in the parliamentary targeting. More than 50 graduated, or been suspended, the family will be percent of existing BISP beneficiaries were found to removed from the program. However, if the family has have PMT scores above the 16.17 cut-off. The BISP then another child who becomes eligible for the program faced the considerable task of communicating these after their removal, they may re-enter the program. decisions to the affected households. In 2021, the BISP had to make its second set of beneficiary exit decisions for the Kafaalat UCT program, following the update of the NSER. As a result of the update some households were found to have poverty scores higher than the cut-off of 32. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 68 BOX 16 LEVERAGING SAVINGS AND FINANCIAL INCLUSION TO SUPPORT BENEFICIARIES BEYOND CASH TRANSFERS Introducing a Hybrid Social Protection Scheme would be a way for BISP to continue supporting ex- beneficiaries. The Hybrid Social Protection Scheme will give low-income households an opportunity to participate in a savings scheme that offers fiscal incentives in form of matching contributions from the BISP at a certain percentage. It will leverage NSER and the country’s financial infrastructure, including payment systems, for targeting and delivery. The main goal of this scheme is to reduce the fiscal burden on the government by ensuring that beneficiaries require little or no government assistance in the event of a shock or financial hardship. The Hybrid Social Protection Scheme is expected to benefit from the trust that the public has in BISP, which is likely to play a significant role in the success of the program. BISP plans to carry out a pilot in districts where financial literacy training is already being provided, before deciding to roll out the scheme on a larger scale. In preparation for the pilot test, BISP has conducted surveys and focus group discussions to assess the ability and willingness of potential participants to save, as well as their preferred methods for doing so, such as using mobile wallets or banks. The scheme aims to promote financial inclusion by leveraging the new payment model. Source: Guven et al 2021 TABLE 2 SEPARATION CRITERIA FOR BENAZIR TALEEMI WAZAIF EXITED CHILD GRADUATED CHILD SUSPENDED CHILD a. Exceeded 7 years in the BTW A beneficiary child who Three consecutive attendance Program for primary and successfully completes non- compliance quarters. secondary education. primary education. This means that the child is registered in the program for b. Exceeded 3 years in the BTW b. A beneficiary child who at least one year before being Program for higher secondary successfully completes considered for suspension. education. secondary education. c. Reaches the age of 15 years c. A beneficiary child who for primary education. successfully completes higher secondary education. d. Reaches the age of 22 years for secondary education. e. Reaches the age of 25 years for higher secondary education. 69 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 3. Delivery systems evolve over time, in a nonlinear fashion, and their s 4. Efforts should be made to “keep it simple” and to “do simple well,” fro 5. The “first mile”—people’s direct interface with administrative function link in the delivery chain; improving it may take systemic change but efficiencies and mitigate the risk of failures on the frontlines. 6. Social protection programs do not operate in a vacuum, and thus th not be developed in silos; synergies across institutions and informat and can improve program outcomes. 7. Social protection delivery systems can contribute more broadly to go other sectors, such as health insurance subsidies, scholarships, socia benefits, and legal services. 8. The dual challenges of inclusion and coordination are pervasive and the continuous improvement of delivery systems, through a dynami centered approach. MONITORING AND EVALUATION KEY MESSAGES BISP has a dedicated M&E unit responsible for assessing and evaluating the delivery systems and operational processes and measuring the impact of the programs. Impact evaluations conducted by private firms have helped to build confidence and trust in BISP by providing transparent and objective assessments of the program’s performance. BISP emphasizes monitoring and evaluation (M&E). The M&E activities of the BISP are mainly focused According to the BISP Act of 2010, M&E is one of the key on three key areas: (i) assessing the delivery systems responsibilities of the BISP board and management. In of all BISP programs, (ii) evaluating the effectiveness line with this provision, BISP has a dedicated M&E unit of their operational processes; and (iii) measuring responsible for monitoring its operational activities their impact. Monitoring is carried out to track the to promote effective program implementation. The progress of program implementation and related M&E unit is guided by a manual which outlines various processes for improving procedures and methods, processes, tools, and methodologies that should be evaluate operational efficiency, and document and used for the purpose. The manual includes guidelines share lessons learned. Evaluation, on the other hand, for using MISs for internal management purposes as is used to measure output and outcome indicators well to support impact evaluations by third parties against a program’s goals, activities, and anticipated using predefined indicators. results, assess its impact, and identify any evidence of shortcomings in its effectiveness and reliability to inform subsequent design changes. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 70 BISP’s monitoring function is divided into internal that program resources are being used efficiently and monitoring and external monitoring. Internal effectively. However, in BISP’s case, these evaluation monitoring involves collecting information related reports are often not made public. to the day-to-day operations of the program to compare against performance indicators and BISP’s commitment to M&E extends beyond process operational work plans to monitor the operational evaluation and spot checks to include impact progress of different delivery stages. The Board evaluations. Impact evaluations are conducted by develops an annual strategic roadmap based on the private firms under the guidance and supervision of targets and outcomes agreed upon by the Board the M&E Wing and focus on assessing the program’s and the Prime Minister, who serves as the patron effectiveness in achieving its intended outcomes. of the organization. The roadmap outlines the key There have been several evaluations conducted performance indicators (KPIs) for the program, which since ISP’s inception. Like the process evaluations, have been selected in collaboration with donors and the impact evaluations are conducted using stakeholders. The M&E Wing of BISP then monitors indicators chosen by BISP in collaboration with its key the departments which are implementing the tasks stakeholders. The impact evaluations measure the in the roadmap. External monitoring is conducted by program’s impact on beneficiaries’ well-being in the external consultants who use spot checks for process short and long run as well as measuring its outputs monitoring, beneficiary assessments, operational and outcomes against the benchmarks defined in the audits, and community scorecards to measure the roadmap. Through impact evaluations, BISP has been efficiency of the program’s operations. able to gather valuable insights into its effectiveness in providing support to vulnerable populations, to BISP has established itself as a strong performer identify areas that need improving, and to make the in terms of process evaluations and spot checks. necessary adjustments. In addition, the evaluations Since its inception, many process evaluations have have helped to build confidence and trust in the been conducted,24 which have helped to identify program among beneficiaries and stakeholders by weaknesses in the program’s implementation, which providing transparent and objective assessments of in turn have enabled it to make improvements. For the program’s performance. example, if an evaluation revealed that beneficiaries in a certain area were experiencing delays in receiving As mentioned earlier, the performance of their payments, BISP could take action to correct the staff working directly with potential or actual problems causing the delay. The evaluations make beneficiaries needs to be monitored. In Pakistan, the program more transparent and accountable these staff report to headquarters and work in BISP’s and help to build trust among beneficiaries and decentralized local offices. Looking at examples from stakeholders. Process evaluations and spot checks other countries, including Brazil, these individuals are internationally considered to be best practices work for local governments (Box 17). Either way, it is in program management as they help to ensure that necessary to ensure that they are performing their programs are being implemented as intended and jobs well. 24 These include (i) regular process evaluation and spot checks throughout the NSER survey 2010-11; (ii) regular process evaluation and spot checks through- out the NSER update 2016-21; (iii) operational reviews of and beneficiary feedback on the BWT education CCT since its inception in 2012; (iv) ongoing oper- ational reviews of the Nashonuma health and nutrition CCT; (iv) four rounds of impact evaluations using the same panel of households between 2010 and 2019; and (v) external reviews and beneficiary surveys over several years funded by the UK’s Foreign, Commonwealth, and Development Office (FCDO). All of the above produced quarterly and annual reports as well as weekly reports when needed. 71 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; BOX 17 and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services 8. The dual challenges of incl MANAGEMENT TOOLS TO PROMOTE QUALITY the continuous improveme centered approach. IN DELIVERY SYSTEMS: LESSONS FROM BRAZIL How well the various steps along the delivery chain are implemented is critical to any program’s effectiveness. The performance of staff working directly with potential or actual beneficiaries is critical in this regard. In some countries (such as Brazil and Chile), these individuals work for local governments. In Pakistan, they work in the BISP’s decentralized local offices and report to a central agency. In either situation, it is important to monitor the activities of these staff, and to ensure that they are performing their jobs well. Brazil has experience with using two instruments designed for this purpose – the decentralized management index and the strategy of active search. In Brazil, the decentralized management index is used to monitor performance of municipalities in managing the social registry and monitoring beneficiaries’ compliance with the conditions related to a CCT. The national government also transfers funds to cover the administrative costs of the municipality using the index. The index is constructed of several variables, including: • The number of valid SR registrations relative to the estimated number of families included in the SR in that locality (coverage). • The number of valid and updated registrations in the previous two years relative to all valid registrations in that locality (updating). • The number of child beneficiaries on whom there is information on their school attendance relative to the total number of children of beneficiary families in the area (extent of compliance monitoring). Brazil also has the active search strategy, which combines poverty maps and administrative data from the social registry to identify the under-registration of certain vulnerable groups (by locality). The national government in Brazil has since financed specific plans to remedy this under-registration and to measure the results. Source: Hellmann 2015; Leite et al. 2017 Although BISP has a dedicated wing for M&E, the one analyst, and one M&E coordinator. However, monitoring function remains fragmented and lacks some major M&E activities are allocated to the CT some elements needed to be fully effective. The Wing (for monitoring the Kafaalat UCT Program), the M&E Wing was established in 2009 and is responsible NSER Wing, and the CCT Wing (for monitoring the BTW for designing and implementing the M&E framework and Nashonuma CCT programs), reducing the role of as well as for validating the deliverables of service the M&E Wing. Also, while the M&E Wing tracks and providers. Reporting directly to the BISP Secretary, the collects data on the KPIs, its efforts are undermined M&E Wing is headed by a director, who is supported by by the absence of a comprehensive M&E Framework two assistant directors, one performance monitoring and by its own lack of enforcement authority. specialist, one evaluation specialist, one statistician, THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 72 4. Efforts should be made to “keep it simple” and to “do simple well,” from th 5. The “first mile”—people’s direct interface with administrative functions—is link in the delivery chain; improving it may take systemic change but will efficiencies and mitigate the risk of failures on the frontlines. 6. Social protection programs do not operate in a vacuum, and thus their de not be developed in silos; synergies across institutions and information s and can improve program outcomes. 7. Social protection delivery systems can contribute more broadly to govern other sectors, such as health insurance subsidies, scholarships, social ene benefits, and legal services. CONCLUSIONS AND 8. The dual challenges of inclusion and coordination are pervasive and pere the continuous improvement of delivery systems, through a dynamic, int RECOMMENDATIONS centered approach. KEY MESSAGES BISP can strengthen itself by building close links with early warning systems, developing a comprehensive communications strategy, incorporating human-centered design to make systems more easily accessible, integrating grievance redress mechanisms, incorporating tools to ensure performance of the increasing number of field staff, and deepening collaboration with provinces. Developing a national vision for NSER would enable it to deliver benefits and services that go well beyond the BISP programs. Given the high frequency of climate and manmade shocks, NSER could step up to make Pakistan’s social protection system more adaptive. This will require better linkages with early warning systems and disaster management authorities. Automating the process of updating the databases to ensure that information updated in NADRA is automatically updated in the NSER would save people a follow-up visit to the NSER office. BISP and NADRA data can be used to create more transparency by creating an interactive data portal where aggregated information is readily available. To promote research and inquiry, BISP can facilitate sharing of its administrative data with other research and academic institutions. 73 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th 6. Social protection programs not be developed in silos; A well-performing delivery chain supports the delivery—clients and administrators can go through and can improve program 7. Social protection delivery s effective and efficient delivery of benefits and each phase of the delivery chain at a reasonable other sectors, such as heal benefits, and legal services services to the intended population. Effective cost in terms of time and money. Well-performing 8. The dual challenges of incl the continuous improveme delivery systems are essentially inclusive. They reach delivery systems ensure effectiveness and efficiency centered approach. the intended population, often overcoming the throughout the delivery chain, from outreach to challenge of including vulnerable groups and those routine oversight, and are supported by effective and who face specific access barriers. Delivery systems efficient information systems, client interfaces, and that function well also promote efficient program institutions (Figure 26). Programs deliver services and benefits FIGURE 26 effectively and efficiently and promote GOAL the inclusion of specific groups with ASSESSMENT OF access barriers. DELIVERY SYSTEMS ASSESS ENROLL PROVIDE MANAGE RECURRING 1 2 3 4 5 6 7 CYCLE 8 9 OUTREACH ENROLLMENT PROVISION OF BENEFICIARIES Intended Population (IP), including Eligible applicants are onboarded BENEFITS & COMPLIANCE, UPDATING Vulnerable Groups (VG) understands efficiently, with minimal leakage to SERVICES AND GRIEVANCES program and is willing to apply ineligible population Enrolled Information is kept up to INTAKE & REGISTRATION DECISION ON PACKAGE beneficiaries date, free of EFC, responsive IP and VG are applying efficiently and their Benefits and service packages are receive to clients evolving needs, and information is recorded accurately accurately determined appropriate promoting desired behaviors ASSESS NEEDS & CONDITIONS services and Applicants are accurately profiled and benefits categorized according to service standards Adopting more dynamic ways of updating the mechanisms considers the planned changes in the NSER would increase the efficiency, accuracy, and payment system; incorporating tools to ensure the adaptiveness of the delivery of social protection performance of the high number of field staff; and in Pakistan. For example, the NSER is establishing deepening collaboration with the provinces. registration desks at the tehsil level for beneficiaries to register and update their information in real time and Establishing a comprehensive national vision for setting up mechanisms for two-way data exchange with the NSER and improving its data-sharing processes partner organizations. Going forward, key directions would greatly benefit the government. The social for the BISP delivery system include strengthening registry represents a significant investment by several aspects of the NSER, including links with early the country and is a valuable national asset. The warning systems; reenforcing communication effort; government could explore opportunities to use the incorporating a human-centered design approach; registry to deliver benefits and services beyond the ensuring that planned integration of grievance redress current BISP programs. BISP has taken positive steps THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 74 in this direction by developing data-sharing protocols governments would also increase coordination and with partner organizations, including provincial create a better understanding of local needs, thereby governments. To further enhance these efforts, BISP ensuring more efficient delivery. This will not only make could focus on streamlining and automating its data it possible for the government to respond rapidly once exchange processes, making them more efficient for a shock hits but will also enable policy makers to focus both BISP and its partner organizations. A two-way on building resilience among vulnerable populations data exchange will enable the NSER to become a more even before shocks hit. dynamic registry, better equipped to serve the needs of its beneficiaries. Strengthening the relationship between the NSER and NADRA would make it easier for Increasing the involvement of the NSER in both BISP beneficiaries to report changes in their ex ante and ex post planning will help Pakistan’s circumstances. In this adaptation, NADRA would social protection system be more adaptive. BISP’s report to the NSER the changed social status of a investment in delivery systems, particularly in the person or household already on the NSER’s verified NSER, has made its social protection system more list, enabling the NSER to update its database. If these adaptive, as evidenced by its success in responding changed attributes were to influence the family’s to two recent crises: the COVID-19 pandemic and PMT score, the NSER could automatically re-calculate the floods of 2022. However, better links need to be the PMT. NADRA would only share the data of those established between the NSER and early warning families who consented to sharing their details with systems (EWS), disaster management authorities, and the NSER. Currently, people have to visit NADRA provincial governments. Establishing a mechanism registration centers to get their records updated whereby the NSER can be sent a trigger by the country’s and then visit the NSER’s tehsil offices for the same early warning systems could significantly reduce BISP’s purpose. The adapted system would be more response time, one of the most important factors in efficient and less time-consuming for beneficiaries shock response. Better coordination and data sharing because they would only have to visit a NADRA with disaster management authorities would help center and their NSER record would be automatically tailor the recovery efforts to the needs of the affected updated (Figure 27). Almost 100,000 people visit population while also helping to sequence and NADRA centers daily. If even 15 to 20 percent of them prioritize interventions (Walker and Johnson 2023). A fall within the ambit of the NSER, a huge number of strong partnership between BISP and the provincial updates would happen automatically. This would FIGURE 27 PROPOSED EXPANDED DATA INTEGRATION BETWEEN NADRA AND THE NSER NADRA DATA PROVISIONING Citizen approaches NADRA for • Change in particulars is • NSER is updated with update in particulars (Birth, recorded at NADRA database. changes recvd. from Marital Status, Disability etc.) NADRA • Updated record is forwarded to BISP if CNIC is found in • PMT is re-calculated NSER verified list 75 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan one or many programs and to the The Sourcebook structures itself ar 1. There is no single blueprin common elements constit 2. Quality of implementation affect the entire system, re 3. Delivery systems evolve ov 4. Efforts should be made to 5. The “first mile”—people’s d link in the delivery chain; im efficiencies and mitigate th FIGURE 28 DEPICTION OF A KNOWLEDGE DISCOVERY PLATFORM 6. Social protection programs not be developed in silos; and can improve program 7. Social protection delivery s other sectors, such as heal benefits, and legal services Dashboards 8. The dual challenges of incl the continuous improveme centered approach. Data Mining Data Sources Social Protection Web Reports (NADRA, NSER) BI Infrastructure Access to Provinces, DMAs and other private SP initiatives motivate beneficiaries to keep their NADRA record to extract data that could then be verified through a up-to-date, helping both the NADRA and the NSER cloud MIS. Currently, BISP is working on a prototype databases to remain current. for the cloud MIS that could be used by those using data from the NSER. This prototype could be Unleashing the potential of the data held by NADRA pilot tested in collaboration with the provinces to and the NSER and converting it into a knowledge evaluate its efficiency and draw lessons. discovery platform would require BI software with data visualization and data analytics features.25 Sharing BISP’s wealth of data with academics BI software could be used by NADRA, NSER, or and researchers could create both transparency provincial databases to collect fresh data in real time as well as a feedback loop that can further and allow users to select data according to their area improve its operations and impact. Other than of interest. Provincial social protection programs the NSER, BISP collects multiple types of data—such and disaster relief management authorities could as administrative data for each of its programs, be given access to this knowledge discovery spot check data, and payments data—which can platform (Figure 28). End-users could drill down be triangulated with other databases to further within a particular geographical area (such as a improve the scope and design of BISP programs. BISP province, district, tehsil, union council, or mauza/ can also make such data available to researchers deh) and apply filters related to socioeconomic and academics after anonymizing it to promote attributes (such as gender, marital status, age, evidence-based policymaking. Such openness and family size, disability, education, or employment). transparency will also help build trust among other After a thorough analysis of this data, the finalized stakeholders. eligibility criteria for a specific program could be used 25 In business intelligence software, the aggregate data are a statistical representation of the available information based on combining multiple measure- ments. It is a summarized view of data from a specific perspective and does not include any individual records. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 76 Prioritizing inclusion by strengthening outreach and of participating financial institutions is likely to require communication efforts will be essential as BISP goes a more complex grievance redress mechanism. The forward with plans to make the NSER a dynamic planned unified grievance redress mechanism for all registry. Effective communication is crucial for social programs within BISP is expected to improve the way protection systems at each delivery stage. BISP’s shift complaints are handled by streamlining the entire toward on-demand the NSER registration will make grievance redress process. It would be beneficial for outreach and communication efforts even more BISP to anticipate this increased complexity when critical, especially for disadvantaged communities developing the UCMS. and individuals located in remote regions. Pro-active search methods such as mobile registration vehicles Promoting the performance and effectiveness of its are being used but will require more planning and field staff will be a priority for BISP. Reflecting the need resources to scale up. The BTW program’s compliance for a larger on-the-ground presence to implement the monitors could be encouraged to implement two CCTs as well as the dynamic the NSER, the BISP has outreach and community activities, already identified increased its field staff. This presents a management as part of their responsibilities. Involving local challenge for BISP’s headquarters. BISP could consider notables in BISP’s outreach efforts would be helpful learning from the experiences of other countries that because of their familiarity with the preferences and have faced similar challenges.. constraints of current and potential beneficiaries. Outreach efforts could inform beneficiaries about Pursuing opportunities to collaborate with the the program’s grievance redress mechanism given provinces at multiple stages of the delivery chain that there is only limited awareness of its existence will help BISP strengthen Pakistan’s social protection among beneficiaries. programs. The provinces have introduced and scaled up their social protection programs requiring more Using human-centered design methods that coordination and collaboration between provincial explore how people interact with the system and federal agencies to prevent fragmentation and throughout the delivery chain would benefit BISP. duplication. High priority areas include the two CCT Such methods would help to ensure that the system subprograms for which the provinces are responsible continues to meet the needs of its target population. for providing services. To ensure the efficiency of Developing journey maps that present an end-to- spending, programs at the federal and provincial end visualization of the client experience including levels need to be complementary and reenforcing, their expectations, behaviors, and emotions can help not overlapping. The statement issued at the end to identify challenges that people may encounter of the first National Social Protection Conference in along the delivery chain, which in turn can inform Pakistan held in May 2023 is promising because of the improvements to improve the client experience. focus on coordination between BISP and provincial authorities on social protection programming and Anticipating potential problems with the new implementation.26 There are also opportunities for the payment system would help with the development provinces to leverage BISP’s delivery systems for their of the UCMS. The transition to a new payment own programs as has already happened for natural system and the anticipated increase in the number disasters (Sindh) and the COVID-19 response (Punjab). 26 The statement emphasizes coordination between federal and provincial entities where federal programs should be designed and implemented in consul- tation with the provincial entities. For this purpose, a Coordination Committee will be formed with representatives of federal and provincial governments and development partners. Data sharing between federal and provincial entities will also be further strengthened. Similarly, role of provincial entities in update of NSER will also be explored. 77 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan REFERENCES • ADB. 2021. Proposed Results-Based Loan and Grant and Administration of Grant Islamic Republic of Pakistan: Integrated Social Protection Development Program. Asian Development Bank. https://www.adb. org/sites/default/files/project-documents/45233/45233-007-rrp-en.pdf. • Agha, Sohail, and Emma Williams. 2016. Quality of Antenatal Care and Household Wealth as Determinants of Institutional Delivery in Pakistan: Results of a Cross-Sectional Household Survey - Reproductive Health. BioMed Central. July 19, 2016. https://doi.org/10.1186/s12978-016-0201-5. • Ali, S. M. 2023. “Evolution of Social Protection Systems in Pakistan.” Unpublished Background Paper. World Bank • Ambler, Kate, and Alan de Brauw. 2019. “Household Labor Supply and Social Protection: Evidence from Pakistan’s BISP Cash Transfer Program.” International Food Policy Research Institute. March 2019. https:// ebrary.ifpri.org/. • Baird, Sarah, Francisco H.G. Ferreira, Berk Özler, and Michael Woolcock. 2014. “Conditional, Unconditional and Everything in Between: A Systematic Review of the Effects of Cash Transfer Programmes on Schooling Outcomes.” Journal of Development Effectiveness 6 (1): 1–43. https://doi.org/10.1080/19439342.2014.890362. • Bastagli, Francesca, Jessica Hagen-Zanker, and Georgina Sturge. 2016. Rep. Cash Transfers: What Does the Evidence Say? Overseas Development Institute. https://odi.org/en/publications/cash-transfers-what-does- the-evidence-say-a-rigorous-review-of-impacts-and-the-role-of-design-and-implementation-features/. • Benazir Income Support Program (BISP). 2019. “BISP Unconditional Cash Transfers Operations Manual”. BISP internal manuscript. Unpublished. • “Benazir Taleemi Wazaif.” n.d. Benazir Income Support Programme. Accessed June 1, 2023. https://www.bisp. gov.pk/Detail/YzNlY2Q2ZGYtNjIwZS00MjNiLWFhMmEtZGM5NWNkMjZhMjQ3. • “Benazir Nashonuma.” n.d. Benazir Income Support Programme. Accessed June 1, 2023. https://www.bisp. gov.pk/Detail/YjAyMjI5ZDQtMTVkOC00YTNlLWE5NjctMjA1NTYwN2JhOTE3. • Cheema, Iftikhar, Simon Hunt, Sarah Javeed, Tanya Lone, and Sean O’Leary. 2016. “Benazir Income Support Programme: Final Impact Evaluation Report.” Www.Opml.Co.Uk. Oxford Policy Management. June 2016. https://www.opml.co.uk/files/Publications/7328-evaluating-pakistans-flagship-social-protection- programme-bisp/bisp-evaluation-wet-cash-transfer.pdf?noredirect=1. • Cheema, Iftikhar, Maham Farhat, Michele Binci, Rimsha Asmat, Sarah Javeed, and Sean O’Leary. 2020. “Benazir Income Support Programme: Final Impact Evaluation Report.” Www.Opml.Co.Uk. Oxford Policy Management. March 2020. https://www.bisp.gov.pk/SiteImage/Misc/files/BISP_EvaluationReport_Ver%20 without_FINAL.pdf • Dodlova, Marina, Anna Giolbas, and Jann Lay. 2018. “Non-Contributory Social Transfer Programs in Developing Countries: A New Dataset and Research Agenda.” Data in Brief 16 (February): 51–64. https://doi. org/10.1016/j.dib.2017.10.066. • Ersado, Lire, Amer Hasan, Koen Martijn Geven, Ashi Kohli, Juan Baron, May Bend, and S. Amer Ahmed. 2023. Pakistan – Human Capital Review: Building Capabilities Throughout Life. World Bank, Washingrton DC. World Bank. 2022c. Charting a Course Towards Universal Social Protection: Resilience, Equity, and Opportunity for all. Open Knowledge Repository. World Bank, Washington DC. https://openknowledge.worldbank.org/ entities/publication/84ba2380-624c-553a-b929-2882e72c7468 • Field, Erica M., and Elisa M. Maffioli. 2021. “Are Behavioral Change Interventions Needed to Make Cash Transfer Programs Work for Children? Experimental Evidence from My.” National Bureau of Economic Research. February 2021. https://www.nber.org/system/files/working_papers/w28443/w28443.pdf. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 78 • “GDP per Capita (Current US$) - Pakistan.” World Bank Open Data. World Bank. Accessed June 1, 2023. https:// data.worldbank.org/indicator/NY.GDP.PCAP.CD?locations=PK. • Gelb, Alan, Anit Mukherjee, and Brian Webster. 2023. “Models of Social Payments through Inua Jamii.” Center For Global Development | Ideas to Action. Center for Global Development. January 19, 2023. https://www. cgdev.org/publication/models-social-payments-through-inua-jamii. • Gentilini, Ugo, Mohamed Bubaker Alsafi Almenfi, TMM Iyengar, Yuko Okamura, John Austin Downes, Pamela Dale, Michael Weber, et al. 2022. “Social Protection and Jobs Responses to Covid-19.” Open Knowledge Repository Beta. World Bank, Washington, DC. February 2, 2022. https://openknowledge.worldbank.org/ entities/publication/fa7a2f3c-efbd-5950-bfac-4b2b4bfc8cad. • “Global ID Coverage, Barriers, and Use by the Numbers : Insights from the ID4D-Findex Survey.” n.d. World Bank, Washington, DC. Accessed June 1, 2023. https://documents.worldbank.org/en/publication/ documents-reports/documentdetail/953621531854471275/global-id-coverage-barriers-and-use-by-the- numbers-insights-from-the-id4d-findex-survey. • “Government of Pakistan Planning Commission Annual Plan 2007-08.” 2007. Government of Pakistan Planning Commission. June 2007. https://www.pc.gov.pk/uploads/annualplan/2007-2008.pdf. • Government of Pakistan. 2016. “Communications Strategy and Action Plan: Waseela-e-Taleem CCT Program”. Internal BISP report: unpublished. • Grosh, Margaret, Phillippe Leite, Matthew Wai-Poi, and Emil Tesliuc. 2022. “Revisiting Targeting in Social Assistance: A New Look at Old Dilemmas.” Open Knowledge Repository. World Bank, Washington DC. March 31, 2022. https://openknowledge.worldbank.org/entities/publication/a6b0063a-4805-5542-89a9- f6da877b5e37. • Guven, Melis, Himanshi Jain, and Clement Joubert. 2021. Social Protection for the Informal Economy: operational lessons for developing countries in Africa and beyond. Washington, Dc: World Bank. https:// documents1.worldbank.org/curated/en/946341635913066829/pdf/Social-Protection-for-the-Informal- Economy-Operational-Lessons-for-Developing-Countries-in-Africa-and-Beyond.pdf • Haque, Nadeem ul, Ayesha Nazuk, Robin Burgess, Imran Rasul, Shaheera Jalil Albasit, Henna Ahsan, Nasir Iqbal, et al. 2021. The State of Poverty in Pakistan. Pakistan Institute of Development Economics. 2021. https://file.pide.org.pk/uploads/rr-050-the-state-of-poverty-in-pakistan-pide-report-2021-68-mb.pdf. • Hellmann, Aline Gazola. 2015. “How Does Bolsa Familia Work?: Best Practices in the Implementation of Conditional Cash Transfer Programs in Latin America and the Caribbean?” Inter-American Development Bank. September 2015. https://publications.iadb.org/publications/english/viewer/How-Does-Bolsa-Familia- Work-Best-Practices-in-the-Implementation-of-Conditional-Cash-Transfer-Programs-in-Latin-America- and-the-Caribbean.pdf. • Hobson, Emma Wadie, Craig Kilfoil, and Andrea Martin. 2022. Rep. Digitizing Cash Transfers to Remote Rural Populations: Challenges and Solutions from the Experience of Zambia. Vol. 27. Washington, DC: World Bank. https://openknowledge.worldbank.org/server/api/core/bitstreams/4762c2c6-b2ce-5c58-b3ba- 904df40b5dd3/content. • International Development Strategies. N.d. “Tracer Report: Spot checks in Pilot Phase Districts.” NSER Internal report. Unpublished. • Ivaschenko, Oleksiy, Jesse Doyle, Jaekyun Kim, Jonathan Sibley, and Zaineb Majoka. 2019. “Does ‘Manna from Heaven’ Help? The Role of Cash Transfers in Disaster Recovery-Lessons from Fiji after Tropical Cyclone Winston.” Wiley Online Library. Disasters. October 3, 2019. https://onlinelibrary.wiley.com/doi/10.1111/ disa.12411. • Khan, Shanza N., and Sarah Qutub. 2010. Rep. The Benazir Income Support Programme and the Zakat Programme: A Political Economy Analysis of Gender. Overseas Development Institute. https://citeseerx.ist. psu.edu/document?repid=rep1&type=pdf&doi=b919aec179280d694a882b50852e6f053041be00. 79 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan • “Labor Force Participation Rate, Female-Pakistan.” 2023. World Bank Open Data. World Bank. April 25, 2023. https://data.worldbank.org/indicator/SL.TLF.CACT.FE.ZS?locations=PK&name_desc=true. • Leite, Phillippe, Tina George, Changqing Sun, Theresa Jones, and Kathy Lindert. 2017. “Social Registries for Social Assistance and Beyond: A Guidance Note and Assessment Tool.” World Bank. July 2017. https:// documents1.worldbank.org/curated/zh/698441502095248081/pdf/117971-REVISED-PUBLIC-Discussion- paper-1704.pdf. • Lindert, Kathy, Tina George Karippacheril, Ines Rodriguez Caillava, and Kenichi Nishikawa Chavez, eds. 2020. “Sourcebook on the Foundations of Social Protection Delivery Systems.” World Bank, Washington DC. https://openknowledge.worldbank.org/server/api/core/bitstreams/b33300fe-008d-5579-8895- 1b7e45119cd6/content. • Loeser, John, Berk Ozler, and Patrick Premand. 2021. What Have We Learned about Cash Transfers? World Bank. May 10, 2021. https://blogs.worldbank.org/impactevaluations/what-have-we-learned-about-cash- transfers. • Mott MacDonald Limited. 2014. “Spot Checks and Beneficiary Feedback, National Cash Transfer Programme- Pakistan.” BISP. May 30, 2014. https://www.bisp.gov.pk/SiteImage/Misc/files/GrievanceRedressReport.pdf. USAID. 2012. Audit of USAID/Pakistan’s Support to the Benazir Income Support Program. USAID. May 21, 2012. https://oig.usaid.gov/sites/default/files/2018-06/g-391-12-006-p.pdf. • Nayab, Durr e, and Shujaat Farooq. 2020. “Unconditional Cash Transfer and Poverty Alleviation in Pakistan: BISP’s Impact on Household Socioeconomic Well Being.” Pakistan Institute of Development Economics. Pakistan Institute of Development Economics. 2020. https://file.pide.org.pk/pdfpideresearch/pv-23- unconditional-cash-transfer-and-poverty-alleviation-in-pakistan-bisps-impact-on-households- socioeconomic-wellbeing.pdf. • Nikitin, Denis, Xiaohui Hou, and Cem Mete. “BISP 2009 Beneficiary Survey Report”. Online. http://catalog.ihsn. org/index.php/catalog/3538/download/49906 • Oxford Policy Management. 2016. “Benazir Income Support Programme: Evaluation of the Waseela-e- Taleem Conditional Cash Transfer Programmer 2016.” Oxford Policy Management. https://www.bisp.gov.pk/ SiteImage/Misc/files/WET-Report.pdf. • Pakistan Bureau of Statistics. 2019. “Employment Trends 2018.” 2019. Pakistan Bureau of Statistics. https:// www.pbs.gov.pk/sites/default/files/labour_force/publications/lfs2020_21/tables/Table_14.pdf. • Pakistan Today. 2011. “BISP-A Commitment Falling Short,” June 25, 2011. http://archive.pakistantoday.com. pk/2011/06/25/bisp-a-commitment-falling-short/. • Shahzad, Asif. 2023. Pakistan Posts Highest-Ever Annual Inflation; Stampedes for Food Kill 16. Thomson Reuters. April 1, 2023. https://www.reuters.com/markets/asia/pakistan-posts-highest-ever-annual- inflation-stampedes-food-kill-16-2023-04-01/. • Smith, Gabrielle Ruth and Thomas Vaughan Bowen. 2020. “Adaptive Social Protection: The Delivery Chain and Shock Response.” World Bank, Washington DC. http://documents.worldbank.org/curated/ en/799281603376140118/Adaptive-Social-Protection-The-Delivery-Chain-and-Shock-Response • UNICEF. 2019. “UNICEF Pakistan: Annual Report 2018.” 2019. UNICEF. July 2019. https://www.unicef.org/ pakistan/media/1966/file/UNICEF%20Pakistan%20Annual%20Report%202018.pdf. • Ali, S. M. 2023. “Evolution of Social Protection Systems in Pakistan.” Unpublished Background Paper. World Bank • Walker, Thomas, and Kelly Johnson, eds. 2023. “Responsive by Design: Building Adaptive Social Protection Systems in South Asia.” World Bank, Washington, DC. Accessed June 1, 2023. https://documents1.worldbank. org/curated/en/099192003212311865/pdf/P17869108018a10c50822c05d3eea25a5e9.pdf. THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 80 • World Bank. 2007. “Social Protection in Pakistan: Managing Household Risks and Vulnerability.” 2007. World Bank, Washington DC. https://documents1.worldbank.org/curated/en/493331468059090991/pdf/354720PK. pdf. • World Bank. 2016. “Pakistan: Building Equality for Women on a Foundation of Identity.” World Bank. World Bank, Washington, DC. February 4, 2016. https://www.worldbank.org/en/news/feature/2016/02/04/pakistan- building-equality-for-women-on-a-foundation-of-identity. • World Bank. 2018. “The State of Social Safety Nets 2018.” World Bank. April 12, 2018. https://www.worldbank.org/en/ topic/socialprotectionandjobs/publication/the-state-of-social-safety-nets-2018#:~:text=The%20State%20of%20 Social%20Safety%20Nets%202018%20examines%20global%20trends,and%20Equity%20(ASPIRE)%20database. • World Bank. 2020a. “The Human Capital Index 2020 Update: Human Capital in the Time of COVID-19.” 2020a. Open Knowledge Repository. World Bank, Washington, DC. September 16, 2020. https://openknowledge. worldbank.org/entities/publication/93f8fbc6-4513-58e7-82ec-af4636380319. • World Bank. 2020b. Pakistan Human Capital Index 2020. World Bank. October 2020. https://databankfiles. worldbank.org/public/ddpext_download/hci/HCI_2pager_PAK.pdf. • World Bank. 2021a. Poverty and Equity Brief South Asia Pakistan. World Bank, Washington DC. October 2021. https://databankfiles.worldbank.org/public/ddpext_download/poverty/987B9C90-CB9F-4D93-AE8C- 750588BF00QA/AM2021/Global_POVEQ_PAK.pdf. • World Bank. 2021b. “What Have We Learned about Cash Transfers?” 2021. World Bank Blogs https://blogs. worldbank.org/impactevaluations/what-have-we-learned-about-cash-transfers. • World Bank. 2022a. “Deep Dive into the Ecosystem for the Delivery of Social Assistance Payments: Turkiye Case Study.” 2022. World Bank, Washington DC. 2022. https://documents1.worldbank.org/curated/ en/099010011072245762/pdf/P17316601c8b0401b0ab340214079f623b6.pdf. • World Bank. 2022b. Pakistan: Flood Damages and Economic Losses Over USD 30 Billion and Reconstruction Needs Over USD 16 Billion - New Assessment. World Bank. October 28, 2022. https://www.worldbank.org/en/ news/press-release/2022/10/28/pakistan-flood-damages-and-economic-losses-over-usd-30-billion-and- reconstruction-needs-over-usd-16-billion-new-assessme. • World Bank. 2022c. Charting a Course Towards Universal Social Protection: Resilience, Equity, and Opportunity for all. Open Knowledge Repository. World Bank, Washington DC. https://openknowledge.worldbank.org/ entities/publication/84ba2380-624c-553a-b929-2882e72c7468 • World Bank. 2022d. Integrating Information Education and Communication Tools to Strengthen Disaster Preparedness and Resilience among Social Assistance Beneficiaries: Lessons for Indonesia and Beyond. Washington, DC: World Bank. Accessed June 3, 2023. https://openknowledge.worldbank.org/server/api/core/ bitstreams/56eadeb0-3c69-5c8d-8a4b-a6d9002164fe/content#page=9&zoom=100,0,0. • World Food Program. 2022. External Evaluation: Complaints Handling Mechanism (CHM) Evaluation at the National Aid Fund (NAF)/ Jordan • World Without Poverty (WWP). N.d. “Introduction to Brazil’s Unified Registry”. Online. BSM_activesearch_ targeting_english.pdf (wwp.org.br) • https://assets-global.website-files.com/5fff9800b9f86a19055527f4/62b8798943e285687abcacdd_ffdws_ ehaas_20220602_0.pdf 81 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan ANNEX 1 DESIGN OF THE NSER DATABASE The NSER database captures information on The NSER is unique in Pakistan since other registries every household in the country. The basic unit of are designed to register either individuals or families. measurement is the household, comprised of a The data stored in the NSER database can be broadly head and other member(s), and the data comprises classified into the categories shown in Figure 1.A. information on the attributes of each household. FIGURE 1.A NSER DATABASE CATEGORIES GEOGRAPHIC HEAD HOUSEHOLD ASSETS ROSTER Names & ID Residence Province Identity Relationship Agricultural District/Tehsil Name Age & gender land Livestock Union Council Contact Info Marital Status Life Standards Village Education Transport Address Occupation Lat/Long Disability The geographical information of each household is consistency and promote data accuracy as spelling described in a sorted manner. Descriptions include errors can occur when recording the names of places. the household’s province, division, district, tehsil, Each household is geo-tagged, and this information union council, village, and physical address. Each of is available as its longitude and latitude. these indicators is assigned a numeric ID to ensure THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS: Leveraging Digital Technology for Adaptive Social Protection in Pakistan 82 THE EVOLUTION OF BENAZIR INCOME SUPPORT PROGRAMME’S DELIVERY SYSTEMS Leveraging Digital Technology for Adaptive Social Protection in Pakistan