Consolidating Social Protection and Labor Policy in Tunisia Building Systems, Connecting to Jobs Policy Note Report No. 103218-TN Policy Note Consolidating Social Protection and Labor Policy in Tunisia Building Systems, Connecting to Jobs Maghreb Department Middle East and North Africa Region December 2015 CONTENTS FIGURES, BOXES AND TABLES ........................................................................................................IV ACKNOWLEDGEMENTS .................................................................................................................VI ABBREVIATIONS ....................................................................................................................... VIII EXECUTIVE SUMMARY ................................................................................................................. IX CHAPTER 1 INTRODUCTION: FROM PROGRAMS TO SYSTEMS FOR SOCIAL PROTECTION AND LABOR POLICY IN TUNISIA ........................................................................................................................... 1 Objectives..................................................................................................................................................................2 Social Protection and Labor Institutions and Expenditure ........................................................................................3 A New Social Contract ...............................................................................................................................................5 CHAPTER 2 ACTIVATING LABOR MARKETS STRATEGICALLY................................................................. 7 Employment Trends ..................................................................................................................................................8 Quality of Jobs .........................................................................................................................................................11 Public Sector Employment ......................................................................................................................................14 Labor Regulations ...................................................................................................................................................14 Active Labor Market Programs ...............................................................................................................................15 Conclusions .............................................................................................................................................................18 CHAPTER 3 INTEGRATING SOCIAL SECURITY FOR BETTER JOBS .......................................................... 19 Pensions ..................................................................................................................................................................19 Health Insurance .....................................................................................................................................................26 Unemployment Benefits..........................................................................................................................................29 Conclusions .............................................................................................................................................................30 CHAPTER 4 INVESTING IN PRODUCTIVE SOCIAL SAFETY NETS ............................................................ 31 Benefit Incidence of Energy and Food Subsidies .....................................................................................................31 Impact Assessment of Potential Reforms on Welfare and Growth .........................................................................33 Performance of Social Safety Net Programs ...........................................................................................................34 Impact Assessment of Targeting Reforms Using Proxy-Means Testing ..................................................................39 Conclusions .............................................................................................................................................................40 ii CHAPTER 5 POLICY IMPLICATIONS: TOWARD A CONSOLIDATED SOCIAL PROTECTION AND LABOR SYSTEM IN TUNISIA ......................................................................................................................... 41 A Pending Agenda ...................................................................................................................................................41 Conclusion: Building Systems, Connecting to Jobs ..................................................................................................45 KEY REFERENCES ...................................................................................................................... 46 iii FIGURES, BOXES AND TABLES FIGURE ES- 1 POLICY FRAMEWORK: BUILDING CONSOLIDATED SOCIAL PROTECTION AND LABOR SYSTEMS IN TUNISIA ............................................................................................................................................................ XIV FIGURE 1 POLICY FRAMEWORK: BUILDING CONSOLIDATED SOCIAL PROTECTION AND LABOR SYSTEMS IN TUNISIA 3 FIGURE 2 PUBLIC SOCIAL EXPENDITURE (SOCIAL PROTECTION AND LABOR PLUS OTHER EXPENDITURES), 2013-2015 .............................................................................................................................................................................4 FIGURE 3 PUBLIC SOCIAL EXPENDITURE IN MIDDLE- AND HIGH-INCOME COUNTRIES, 2013 (PERCENT OF GDP) .......4 FIGURE 4 MAIN SOCIAL PROTECTION AND LABOR-RELATED INSTITUTIONS IN TUNISIA, 2015 ....................................6 FIGURE 5 EVOLUTION OF UNEMPLOYMENT RATES, 2006 – 2014 ................................................................................8 FIGURE 6 UNEMPLOYMENT RATE BY REGION ...............................................................................................................9 FIGURE 7 DISTRIBUTION OF UNEMPLOYED PERSONS ...................................................................................................9 FIGURE 8 COMPOSITION OF UNEMPLOYED POPULATION BY SKILL LEVEL ACROSS REGIONS, 2013 ..........................10 FIGURE 9 SKILLS AS A PERCEIVED BARRIER TO FIRM GROWTH BY TYPE OF FIRM, REGION, AND SECTOR .................11 FIGURE 10 SHARE OF JOBS BY SECTOR ........................................................................................................................12 FIGURE 11 NUMBER OF JOBS BY SECTOR ....................................................................................................................12 FIGURE 12 NET JOB CREATION BY SECTOR ..................................................................................................................13 FIGURE 13 BENEFICIARIES OF PROGRAMS FINANCED BY THE NATIONAL EMPLOYMENT FUND ................................15 FIGURE 14 SATISFACTION WITH PUBLIC EMPLOYMENT SERVICES BY EDUCATIONAL LEVEL ......................................16 FIGURE 15 SATISFACTION WITH PUBLIC EMPLOYMENT SERVICES BY INCOME LEVEL................................................17 FIGURE 16 BENEFICIARY MOTIVATIONS FOR ACCESSING PILOT ALMP AMONG LOW-SKILLED WORKERS .................17 FIGURE 17 DISTRIBUTION OF JOBS FOUND BY TYPE AND REGION SIX MONTHS AFTER PROGRAM EXIT ...................18 FIGURE 18 WAGE DISTRIBUTION AMONG PENSION BENEFICIARIES, 2011-2012 .......................................................22 FIGURE 19 PENSION BENEFITS DISTRIBUTION BY AGE AND GENDER, 2011-2012 ......................................................23 FIGURE 20 PROJECTED DEFICIT, CNRPS AND RSNA 2010-2018 (PERCENT OF GDP)....................................................24 FIGURE 21 HEALTH FINANCING COVERAGE RATE (HEALTH INSURANCE AND OTHER SCHEMES) (PERCENT).............27 FIGURE 22 HEALTH INSURANCE EXPENDITURE, 2007-2011 ........................................................................................28 FIGURE 23 DISTRIBUTION OF HEALTH SERVICES USED, CNAM, 2008-2011 ................................................................28 FIGURE 24 BENEFIT INCIDENCE OF SPECIFIC TYPES OF ENERGY AND FOOD SUBSIDIES (PERCENT) ...........................32 FIGURE 25 INCIDENCE OF CASH TRANSFERS AND AMG-1 BY GOVERNORATE, 2011-2014.........................................35 FIGURE 26 INCIDENCE OF HEALTH CARDS (AMG-2) BY GOVERNORATE, 2011-2014 ..................................................35 FIGURE 27 COVERAGE RATE OF CASH TRANSFER PROGRAM, PNAFN, (TOP PANEL) AND SUBSIDIZED HEALTH CARD PROGRAM, AMG-2, (BOTTOM PANEL) BY QUINTILE (PERCENT) .......................................................................37 FIGURE 28 DISTRIBUTION OF PNAFN BENEFICIARIES ACROSS QUINTILE (PERCENT) ..................................................38 FIGURE 29 SHARE OF BENEFICIARIES OF CASH TRANSFERS (UNCONDITIONAL) IN THE FIRST QUINTILE (LOWEST INCOME GROUP), INTERNATIONAL COMPARISONS (PERCENT) ........................................................................38 FIGURE 30 COMPARISON OF THE DIFFERENT SCENARIOS FOR SUBSIDIES AND TARGETING OF SOCIAL ASSISTANCE REFORM .............................................................................................................................................................39 FIGURE 31 ACTUAL DISTRIBUTION (LEFT) VERSUS SIMULATED DISTRIBUTION USING PMT (RIGHT) OF PNAFN INCIDENCE BY REGION (PERCENT) .....................................................................................................................40 FIGURE 32 LEVEL OF VULNERABILITY BY POPULATION GROUP AS OF 2015, TUNISIA ................................................42 BOX 1 TOWARDS A JOBS STRATEGY IN TUNISIA ............................................................................................................7 TABLE 1 COVERAGE RATES AND PUBLIC PENSION SPENDING, SELECTED MIDDLE- AND HIGH-INCOME COUNTRIES 20 TABLE 2 ASSESSMENT OF PENSION SCHEMES IN TUNISIA, 2011-2012 .......................................................................21 TABLE 3 FINANCIAL FLOWS AND DEFICIT PROJECTED, CNRPS AND CNSS (RSNA) 2010-2018 ....................................24 iv TABLE 4 ESTIMATED MACROECONOMIC IMPACTS OF ENERGY SUBSIDY REFORMS OVER FIVE YEARS......................33 TABLE 5 TARGETING SCENARIOS FOR HOUSEHOLD COMPENSATION FOR REFORMING ALL ENERGY SUBSIDIES ......34 TABLE 6 MAIN SOCIAL SAFETY NETS PROGRAM IN TUNISIA, 2014 .............................................................................35 TABLE 7 AVERAGE BENEFIT AMOUNT AND GENEROSITY OF PNAFN ..........................................................................37 TABLE 8 KEY CHALLENGES AND OPTIONS FOR REFORM .............................................................................................43 v ACKNOWLEDGEMENTS This policy note is based on the World Bank’s Governance in Social Sectors Technical Assistance Program (2012-2015), mobilized to support Tunisia’s post-revolution transition phase, and is a companion to the Tunisia Systematic Country Diagnostic Report (2015). The note is primarily intended for and the result of collaboration with key stakeholders involved in reform from the Government of Tunisia, civil society, labor unions, and the private sector. The authors extend their gratitude to the Government of Tunisia for their hospitality and coordination as part of this work, notably, representatives from the Ministry of Investment and International Cooperation, Ministry of Social Affairs, Ministry of Employment and Vocational Training, Ministry of Economy and Finance, Ministry of Industry, and Ministry of Commerce, as well as civil society and private sector groups including the General Union of Tunisian Workers, Tunisian Union of Industry, Commerce and Crafts, the Tunisian Union of Social Solidarity, and the many helpful government and civil society actors at the regional levels whose insights have enriched this work. The World Bank team comprised Heba Elgazzar (Senior Economist, Social Protection & Labor), David Robalino (Manager and Lead Economist, Jobs), Rene Antonio Leon Solano (Senior Social Protection Specialist, Social Protection & Labor), Diego Angel-Urdinola (Senior Economist, Social Protection & Labor), Michael Weber (Economist, Jobs), Abdel Rahman Lahga (Assistant Professor, University of Tunis, and Consultant), Mohamed Ali Marouani (Associate Professor, Sorbonne University, and Consultant), Isabelle Huynh (Senior Operations Officer, ICT), Peter McConaghy (Financial Sector Specialist, Finance & Markets), Jade Salhab (Senior Private Sector Development Specialist, Trade & Competitiveness), Montserrat Pallares-Miralles (Social Protection Specialist, Social Protection & Labor), Carine Clert (Lead Social Protection Specialist, Social Protection & Labor) and Gustavo Demarco (Lead Economist, Social Protection & Labor). Helpful contributions were provided by Jean-Luc Bernasconi (Lead Country Economist, Macroeconomic & Fiscal Management), Antonio Nucifora (Lead Country Economist, Macroeconomic & Fiscal Management), Willian Stebbins (Senior Communications Officer, External Communications), Ruslan Yemtsov (Lead Economist, Social Protection & Labor), Daniela Marotta (Senior Economist, Macroeconomic & Fiscal Management), Paolo Verme (Senior Economist, Poverty), Jose Cuesta Leiva (Senior Economist, Poverty), Masami Kojima (Lead Energy Specialist, Energy & Extractives), Mohamed El Shiaty (Operations Officer, Trade & Competitiveness), Gloria La Cava (Senior Social Scientist, Social Development), Tobias Lechtenfeld (Economist, Social Development), and Setareh Razmara (Lead Social Protection Specialist, Social Protection & Labor). The team appreciates the support provided by Francoise Mukamana (Program Assistant, Social Protection & Labor), Astania Kamau (Team Assistant, Social Protection & Labor), Besma Saadi Refai (Team Assistant, Tunis Country Office), Narjes Jerbi (Program Assistant, Tunis Country Office), and Wassim Turki (Project Coordination Consultant). vi The team also sincerely thanks the peer reviewers for their useful feedback and comments: Anita Schwarz (Lead Economist, Social Protection & Labor), Philip O’Keefe (Lead Economist, Social Protection & Labor), Pia Schneider (Lead Social Sectors Evaluator, Independent Evaluation Group), and Maureen Lewis (Visiting Professor, Global Human Development Program, Georgetown University). This work benefited from invaluable counsel provided by Yasser El-Gammal (Practice Manager, Social Protection & Labor) and Eileen Murray (Country Manager, Tunisia) and the overall guidance of Simon Gray (former Country Director, Maghreb) and Marie Francoise Marie-Nelly (Country Director, Maghreb). vii ABBREVIATIONS AMG Assistance Médicale Gratuite ANETI Agence Nationale de l’Emploi et du Travail Indépendant CNAM Caisse Nationale d’Assurance Maladie CGSP Contrôle General des Services Publics CNRPS Caisse Nationale de Retraite et de Prévoyance Sociale CNSS Caisse Nationale de Sécurité Sociale CRES Centre de Recherche et d’Etudes Sociales CSO civil society organization FDI foreign direct investment GDP gross domestic product ICA Investment Climate Assessment IMF International Monetary Fund INS Institut National de la Statistique MENA Middle East and North Africa MOE Ministry of Education MOH Ministry of Health MOSA Ministry of Social Affairs MVTE Ministry of Vocational Training and Employment PISA Programme for International Student Assessment PMO Prime Ministry’s Office PMT Proxy-means testing PNAFN Programme National d’Aide aux Familles Nécessiteuses SPL Social protection and labor TIMSS Trends in International Mathematics and Science Study TND Tunisian dinar UGTT Union Générale des Travailleurs Tunisiens USD United States dollar UTICA Union Tunisienne de l’Industrie, du Commerce et de l’Artisanat CURRENCY EQUIVALENTS (Exchange Rate Effective as of December 4, 2015) 1 TND = 0.49 USD 1 USD = 2.03 TND GOVERNMENT FISCAL YEAR January 1 – December 31 viii EXECUTIVE SUMMARY Tunisia today represents a paradox: despite political progress since its 2011 revolution, wide economic and social disparities persist, threatening stability. The events of 2011 revealed what had been festering for years: growing inequity. Four years and four interim governments later, a new government has been formed as of 2015, charged with developing the country’s first mid- term, five-year development plan since 2010. However, the challenges facing Tunisia remain formidable. On the one hand, Tunisia has generally shielded itself from the political risks and deep civil strife witnessed in other “Arab Spring” countries such as Libya, Yemen, Syria, and to an extent, Egypt. However, ongoing, unpredictable security threats continue to exist in Tunisia, exacerbated by substantial regional spillovers. Existing social and economic exclusion among lower- and middle-income households remains an impediment to stabilizing these threats. In the absence of bold measures, Tunisia’s short-term economic outlook remains particularly dim, with real GDP growth estimated at approximately 1 percent for 2015.1 While Tunisia has several social protection and labor programs in place, its ability to respond to increasing social needs is compromised by inefficiency, fragmentation, and inequity. Against this backdrop, the objective of this note is to evaluate the e ffectiveness of Tunisia’s main social protection and labor programs and identify options for reform through a systems- based approach. The note is based on a review of recent evidence on the efficiency and equity of key programs. Importantly, it contributes new analysis in three main areas of systems building: (i) financing and sustainability, (ii) institutions and governance, and (iii) service delivery. The note argues that without significantly improving institutional coordination on financing and delivery, Tunisia’s social protection and labor system will be ill equipped to strengthen economic and social inclusion. Labor, Skills, and Social Insurance Persistent unemployment has exacerbated poor social conditions for large swathes of Tunisians, most notably in its western and southern hinterlands. Although unemployment declined from 18.9 percent in 2011 to 15 percent as of late 2014, the decrease is mainly due to continuing increases in public sector recruitment. Poverty was estimated at 15.5 percent as of 2010, ranging from 9.1 percent in the Greater Tunis area to 32.3 percent in the center-west region of Tunisia2. 1 World Bank (2014). Investment Climate Assessment: Enterprises’ Perception in Post-Revolution Tunisia, 2014. Washington DC: World Bank. 2 National Institute of Statistics, Tunisia (2012). Measurement of poverty, inequalities and polarization in Tunisia 2000-2010. Tunis: National Institute of Statistics, Ministry of Investment, Development and International Cooperation. ix While long-standing investment barriers have constrained private sector-led job growth in Tunisia, supply-side factors have also played a role. Key factors include poor implementation of labor insertion programs and the lack of practical job skills among low- and high-skilled populations. Structural constraints on the demand and supply sides have particularly affected the bottom 40 percent of income distribution.3 On the demand side, a lack of targeted, sectoral policies to improve productivity and job creation in key value chains are chief barriers. On the supply side, constraints include the prevalence of skills mismatches, inefficient intermediation, rigid labor market regulations, and the wide disparity between public sector and private sector employment benefits. Tunisia’s private sector also faces a relatively high tax wedge, with the pension system facing a growing, unsustainable deficit. At the same time, the informal sector remains large, with many workers poorly protected from risks and shocks. However, renewed social dialogue on pensions and emerging evidence from recent pilot employment programs offer alternative options to support the most vulnerable. New public- private partnership implementation arrangements at the local level can boost job placement for the most vulnerable and semi-skilled.4 Analytic tools developed for the Tunisian context to assess the impact of alternative pension and unemployment benefit reforms on the labor market could be used to pave the way for improving the quality of jobs. Overall, key labor market and social insurance challenges facing Tunisia include:  Lagging job creation in the private sector and persistent informality: Excessive investment climate barriers, a lack of targeted job strategies, and a high tax wedge impinge on the creation of sustainable, inclusive jobs. Among the working population, nearly 30-45 percent of the employed labor force either do not have contracts or do not contribute to social security. Exacerbating this informality is the poor access to finance at the household level, where nearly two-thirds of Tunisian adults lack bank accounts or have partial access to formal financial services.  Inadequate skills among the unemployed: While unemployment has largely fallen on youth, women, and university graduates, the vast majority (67 percent) of the unemployed are low skilled and lack university degrees. There is also a disconnect between skills provided by the educational and vocational training system on one hand and the demands of a rapidly evolving private sector on the other. Tunisia performs relatively poorly in relation to comparable middle-income countries on educational outcomes as measured by the Trends in International Mathematics and Science Study (TIMSS( and the Programme for International Student Assessment (PISA), and the data 3 World Bank (2015). Towards a Job Strategy in Tunisia, as part of the Tunisia Country Systematic Diagnostic. Washington DC: World Bank. 4 World Bank (2015). Implementation Completion and Results Report for Tunisia State and Peacebuilding Fund (SPF) Participatory Service Delivery for Reintegration Project. Report No: ICR00003568. Washington DC: World Bank. x from the most recent Investment Climate Assessment (ICA) cites the relevance of skills among the key barriers to hiring in Tunisia.  Insufficiently targeted active labor market programs and lagging on-the-job experience in critical sectors: Active labor market programs (ALMPs) in Tunisia have not sufficiently boosted job placement even in sectors that have created jobs, resulting in a placement rate of just 20 percent on average. ALMPs face implementation capacity constraints in the public sector, poor governance in terms of monitoring and accountability of service providers, and a lack of programs catering to low- and semi-skilled workers. Public employment services often lack sufficient job counseling and placement matching for job openings that exist, which has contributed to a delayed school-to-work transition.  Impinging deficits facing pensions and low social insurance coverage: The public and private pension funds are fragmented in terms of administration, identification, and policies. Together these funds currently face significant deficits that are expected to reach at least 2 percent of GDP by 2018, threatening benefits and coverage rates. Only 37 percent of the population contributes to pensions, half of the population is covered by health insurance, and no unemployment benefits exist for workers who lose their jobs. The social dialogue process in Tunisia, launched in 2013 to address pensions and unemployment insurance options, has stalled due to political economy challenges and a lack of a common vision among social partners comprising the government, labor, and trade unions. Resilience, Risks, and Productive Social Safety Nets While poverty has been steadily decreasing in Tunisia, large pockets of vulnerability exist that can be better tackled through more productive social safety net programs. Key priorities include improving identification systems, targeting of universal price subsidies and the existing national cash transfer and health care programs, as well as reforming the design and delivery of productive mechanisms. The latter include enhancing the role of public-private partnerships for productive social safety net (SSN) programs, notably ALMPs for low-skilled, low-income households in viable sectors and a restructuring of public works programs to better respond to Tunisia’s current demographics and aspirations. To address lagging social indicators, the role of social safety nets could also be adapted to incentivize the poor to invest in human capital, namely nutrition, education, and health, particularly in Tunisia’s western and southern regions. Overall, key social safety net challenges facing Tunisia today include:  Inefficient targeting and sustainability of social safety nets: Tunisia’s existing cash transfer and health card programs account for an estimated 0.4 percent of GDP, already reaching levels spent by countries such as Brazil, but with far less accurate targeting of the poor. Only 12 percent of Tunisia's poor (based on a threshold of US$2 per day) benefit xi from the national cash transfer program for needy families.5 Among those receiving benefits, which account for nearly 23 percent of all Tunisians, only 40 percent are in the lowest income quintile representing the poor (considered using a threshold of approximately US$2 per day).  A lack of graduation mechanisms: Limited mechanisms exist among current social safety net programs for enhancing access to work, training, or basic social services for the poor and vulnerable who are out of school. Recertification and verification procedures are not clearly defined, and exit strategies for various categories of beneficiaries and their dependents are unclear.  Inequity among universal price subsidies and hampered fiscal space for social protection, labor, and other productive investments: Universal price subsidies represent a tremendous burden, with fuel subsidies accounting for approximately 5 percent of GDP alone (as of 2013, which has since fluctuated due to global oil prices). The effects on productivity and household welfare vary by type of energy product, but maintaining the current system siphons off much-needed resources to boost private sector-led job growth and reinforce the sustainability of social protection systems. Building Systems, not Programs: Policy Recommendations for the Short to Long Term Key recommendations comprise a gradual, systems-building framework over the short to long term (Figure ES-1). Reforms fall into three types, namely (i) enhancing implementation and delivery platforms, (ii) strengthening institutional capacity and governance; and (iii) improving financial sustainability. Over the short term, enhancing implementation and delivery platforms can pave the way for broader coordination. Understanding how benefits are delivered, distributed, duplicated, or entirely omitted across levels of vulnerability is critical to consolidating the system as a whole. Unemployment has largely fallen on youth and women among the bottom 40-60 percent by income, with the vast majority (67 percent) of the unemployed being low-skilled populations in underserved regions. As of 2013, nearly 30-45 percent of the employed labor force either do not have contracts or do not contribute to social security, particularly among the poorest 40 percent. Finally, among the poorest 20 percent, only 12 percent of Tunisian's benefit from the national cash transfer program. 5 World Bank (2013). Vers une meilleure équité: les subventions énergétiques, le ciblage et la protection sociale en Tunisie (Towards greater equity: energy subsidies, targeting and social protection in Tunisia). Washington DC: World Bank. xii Enhancing delivery platforms can be achieved by:  Developing public-private implementation systems for skills development and income generation mechanisms in key sectors with high employment potential for out-of-school, low-skilled, vulnerable populations, as part of a private sector-led jobs strategy;  Introducing transparent targeting and monitoring systems for existing social safety net and active labor market programs;  Developing platforms for digital payment mechanisms to replace postal transfers and enhance financial literacy among households; and  Strengthening coordination of and promoting access to essential public services (notably nutrition, education, health, and water in Tunisia’s lagging regions) for social safety net programs. Over the medium term, strengthening institutional capacity and governance would help lay the foundation for aligning social inclusion and job growth. ALMPs, pensions, and social safety nets as currently designed and implemented are not reaching vulnerable groups in part due to institutional fragmentation in terms of identification and eligibility. Financial inclusion and access to social protection services remains weak, most notably in lagging regions where the delivery of services is hampered by inadequate investments in skilled human resources and equipment. ALMPs are also not adequately capturing and targeting firms’ skills needs in key sectors and regions. In 2013, Tunisia launched the creation of a unique identification system and introduced interoperability between information systems for key social protection and labor programs. Stronger institutional capacity and governance within the system in Tunisia can be attained by:  Rolling out unified social registry information, targeting, and unique identification systems;  Consolidating pension administration (CNSS and CNRPS) as well as health insurance mechanisms (CNAM and AMG); and  Strengthening the social dialogue policy framework and evidence base for developing public- private partnerships for ALMPs, reform labor contracts, and redefining wage-setting mechanisms that more actively promote productivity. Over the long term, although improving the financial sustainability of the overall system may take time, building the evidence base and consensus can be achieved in the meantime. A high tax wedge and pension deficits impinge on the quality of jobs and protection from shocks, and consumption subsidies crowd out spending on other vital investments that could spur revenue generation. Pressures on fiscal space could be supported in the medium to long term by redirecting savings from reforming inefficient consumption subsidies and in the long term through comprehensive pension reform. Savings could also be used to enhance mechanisms that promote access to services and coverage of social safety nets. xiii Tunisia can therefore improve financial sustainability of its system by:  Developing supplemental resource generation models and greater portability for pension benefits (CNSS and CNRPS) that explicitly tie benefits to contributions and identify subsidized pension benefits;  Aligning investment code and labor regulations with targeted schemes for unemployment benefits to improve the quality of jobs among the vulnerable self-employed and small and medium enterprises (SMEs) in lagging regions; and  Reinvesting universal subsidies into public or private investments that stimulate job growth and improve the effectiveness of targeted social protection programs. Looking ahead, Tunisia’s forthcoming National Development Plan for 2016-2021 provides an important platform for financing a unified social protection and labor reform agenda. Overall, consolidation in Tunisia entails enhancing social protection and labor delivery systems over the short term and realigning policies on benefits, eligibility, and financing schemes over the medium to long term. Ultimately, modernizing its system will help Tunisia better cope with new shocks and, importantly, reap the social benefits of its hard-fought political transition. Figure ES- 1 Policy Framework: Building Consolidated Social Protection and Labor Systems in Tunisia Source: Authors, adapted from World Bank (2013), World Development Report: Jobs, and World Bank (2011), World Development Report: Conflict, Security and Development. World Bank: Washington DC. xiv CHAPTER 1 INTRODUCTION: FROM PROGRAMS TO SYSTEMS FOR SOCIAL PROTECTION AND LABOR POLICY IN TUNISIA Tunisia’s historically modest, yet steady, social and economic trajectory now faces deep challenges in terms of persistent unemployment and widening regional disparities. The immediate challenge for Tunisia is to ensure economic and social stability in a situation where the short-term economic outlook remains uncertain. The need to bridge economic and social policies is ever-present as Tunisia struggles to reset its growth path and ensure that the most vulnerable benefit. Despite advancing politically with a new constitution and elections since 2011, economic progress has lagged. Economic challenges have worsened over the course of 2015 with outbreaks of violent attacks, further weakening the economic and social outlook over the medium term. Real gross domestic product (GDP) growth stood at only 2.3 percent as of end-2014 and was expected to be similar, at 2.6 percent, in 2015 given the lag in investment and leveling off of consumption,6,7 but has likely deteriorated to approximately 1 percent owing to violent attacks along the coast during June 2015. The fiscal deficit is estimated to be 4.8 percent of GDP as of end 2014, down from 6.2 percent in 2013. The decrease is largely due to a marginal reduction in energy subsidy spending from 7.8 percent in 2013 to 7.2 percent in 2014. Although unemployment declined somewhat from 18.9 percent in 2011 to 15 percent as of late 2014, the decrease has mainly been due to continuing increases in public sector recruitment, which is unsustainable going forward. As such, the wage bill has increased from 12.5 percent of GDP in 2013 to 12.7 in 2014, further hampering sustainability. In addition, unemployment remains well above the pre-revolution rate of 13 percent, and broad regional disparities in unemployment persist. Unemployment reached as high as 30 percent in poorer regions such as Gafsa and Kasserine in late 2014. Poverty was estimated at 15.5 percent in 2010, but ranges widely from 9.1 percent in the Greater Tunis area to 32.3 percent in the center-west region of Tunisia.8 Tunisia’s challenges call for a development plan built on a new approach to tackling challenges through a well-coordinated, consolidated social protection and labor system. 6 World Bank (2015). Tunisia Economic Monitoring Note, Spring 2015. Washington DC: World Bank. 7 World Bank (2015). Tunisia Systematic Country Diagnostic. Washington DC: World Bank. 8 National Institute of Statistics, Tunisia (2012). Measurement of poverty, inequalities and polarization in Tunisia 2000-2010. Tunis: National Institute of Statistics, Ministry of Development and International Cooperation. 1 Objectives Against this backdrop, the objective of this note is to evaluate the effectiveness of Tunisia’s main social protection and labor programs and identify options for reform through a systems- based approach. The note is based on a review of recent evidence on the efficiency and equity of key programs. Importantly, it contributes new analysis in three main areas of systems building: (i) financing and sustainability, (ii) institutions and governance, and (iii) service delivery. The note argues that without significantly improving institutional coordination on financing and delivery, Tunisia’s social protection and labor system will be ill equipped to strengthen economic and social inclusion. The note is structured as follows. Chapter 1 assesses key institutional and expenditure trends; Chapter 2 presents the labor market context, drawing on jobs diagnostics;9 Chapter 3 outlines key social insurance trends in terms of pensions and health and unemployment insurance; Chapter 4 discusses social safety nets and consumption price subsidies in terms of welfare impacts; and finally, Chapter 5 summarizes key challenges facing the system as a whole and describes key reforms needed over the short to long term from a systems-building perspective. The framework for the assessment is based on promoting equity through jobs and inclusive social policies (Figure 1). The World Development Reports for 2011 and 2013 highlight the role of these policies within the context of fragile, political transition. In addition, the World Bank Social Protection and Labor Strategy emphasizes that well-coordinated social protection and labor systems are central to this objective and contribute to realizing three main policy aims:10 (i) strengthening opportunity for all (such as through human capital and productive employment); (ii) strengthening equity (particularly in mitigating chronic poverty); and (iii) building resilience for households (in terms of protecting against unforeseen risks and shocks). In the case of Tunisia, several recent evaluations have taken stock of the broader country context and economic and social programs, demonstrating that much of the challenges to its social protection and labor programs remain today.11 This paper adopts a new framework that calls for emphasizing that single policies alone will likely fail to realize their full gains without fully leveraging the financing, targeting, and service delivery of other programs. 9 World Bank (2015). Towards a Jobs Strategy in Tunisia, as part of Tunisia Systematic Country Diagnostic. Washington DC: World Bank. 10 World Bank (2012). World Bank Social Protection and Labor Strategy 2012-2022. Washington DC: World Bank 11 World Bank (2014). Tunisia Country Program Evaluation FY05-FY13. Washington DC: World Bank. 2 Figure 1 Policy Framework: Building Consolidated Social Protection and Labor Systems in Tunisia Source: Authors, adapted from World Bank (2013), World Development Report: Jobs, and World Bank (2011), World Development Report: Conflict, Security and Development. Washington DC: World Bank. Social Protection and Labor Institutions and Expenditure Tunisia’s main social protection and labor programs include active labor market programs (ALMPs), social insurance, and social assistance. Tunisia spent approximately 6.4 percent of GDP on social protection and labor programs during the 2013-2015 period (Figure 2). However, weaknesses in governance, coverage, and sustainability threaten the accelerating shared prosperity and poverty reduction noted, particularly in lagging regions. The level of public social expenditure in Tunisia is comparable to that of other middle-income countries such as Mexico and Chile, but the mix and effectiveness of its programs differ widely (Figure 3). Public expenditure on social sectors (excluding subsidy transfers) amounts to almost 19 percent of GDP. By way of comparison, public social expenditures amounted to an average of 28 percent of GDP among middle- and high-income countries in 2013, although pension expenditure makes up a larger share of this spending in high-income countries than in middle-income countries. 3 Figure 2 Public Social Expenditure (Social Protection and Labor plus other Expenditures), 2013-2015 Sources: World Bank staff calculations. Data from the Ministry of Finance, October 2013; data of the CNAM, CNRPS and CNSS. Note: subventions = universal price subsidies on energy and basic commodities as of 2013; subsidy expenditures have decreased in 2014 and 2015 and expected to rise to 2013 levels after 2015. Note: Subsidies (total) = energy (explicit), basic products, and transportation subsidies. Health (total) = health and CNAM. Employment = active employment programs and employment services. Education = all levels. Social Assistance = cash transfer programs and health care cards of the Ministry of Social Affairs, end 2012. Other social services = services/programs of the Ministries of Social Affairs; Youth; and women and children. Figure 3 Public Social Expenditure in Middle- and High-Income Countries, 2013 (percent of GDP) Source: World Bank staff calculations, OECD Social Expenditure and Education Statistical Databases. Note: Social expenditures in 2013 for most sectors and 2010 for education (the most recent year). Public social expenditure total expenditure = expenditure on services and social insurance (retirement, health, minimum income allocation, social services), that is an OECD average of 22 percent of GDP, plus spending on education, that is a OECD average of 6 percent of GDP, for a total of 28 percent of GDP. 4 The political economy context of social protection and labor in Tunisia has contributed to fragmented policies. Today, social protection and labor policies and programs in Tunisia are fragmented and often at odds; thus, better alignment will help improve growth and inclusion (Figure 4). These policies and programs are managed by a myriad of agencies and ministries, often with conflicting eligibility criteria and lacking sufficient coverage because of lags in information on beneficiaries and expenditures. Given the close linkages between labor regulations, active labor market programs, social insurance, social safety nets, and subsidies, there is a need to consolidate Tunisia’s social protection system for greater efficiency and effectiveness. A New Social Contract Since 2012, Tunisia has put in place measures that pave the way for reforming social protection and labor policies to more effectively reach the poor and promote labor insertion. In May 2012, the government launched a social dialogue process that reached a significant milestone in January 2013 with the signing of a new “Social Contract.” The Social Contract agreement lays out critical areas of reform for several key themes, including social protection and labor reform as well as economic development in lagging regions. A draft law was also prepared in 2014 for the creation of a new tripartite national social dialogue council, which can help alleviate some of the political economy challenges. The council, with a rotating presidency, would have tripartite representation by the government (coordinated by the Ministry of Social Affairs), labor unions (represented by Union Generale des Travailleurs Tunisiens (UGTT) and other bodies), and the private sector (represented by Union Tunisienne de l'Industrie, du Commerce et de l'Artisanat (UTICA) and other bodies). Going forward, Tunisia will need to develop a broader social protection and labor policy strategy that improves the efficiency and equity of these programs, which is the focus of the remainder of this note. 5 Figure 4 Main Social Protection and Labor-related Institutions in Tunisia, 2015 Source: World Bank staff. Note: CSO = civil society organizations; TBC = to be confirmed (draft law prepared as of 2014). Other relevant institutions include the Ministry of Health and the Ministry of Education. 6 CHAPTER 2 ACTIVATING LABOR MARKETS STRATEGICALLY Despite earlier trends in growth, broad social disparities in employment persist, with a need to revitalize labor markets in Tunisia. For decades Tunisia was heralded as an economic engine; however, Tunisia’s growth masked broad disparities, crony wealth-generation, and deeply rooted social exclusion. However, annual job growth stood at only 2.5 percent, catering mainly to higher- skilled, urban groups and excluding youth, women, and vulnerable populations in lagging regions. A comprehensive jobs analysis highlights current and future trends in detail (Box 1). This chapter outlines key supply- and demand-side trends, with particular emphasis given to the labor force and supply-side challenges. Box 1 Toward a Jobs Strategy in Tunisia A recent analysis3 of jobs challenges in Tunisia shows that removing the macroeconomic and fiscal constraints alone will likely not be enough to address Tunisia’s job challenge, particularly among the bottom 40 percent of the population. Specifically:  Growth, jobs, and productivity: Assuming that labor force participation rates remain at current levels, demographics alone will push around 47,000 new entrants into the labor market on average per year. Tunisia’s lagging economy has been creating only around 36,400 jobs per year , largely through micro- firms and in low-productivity activities. Under an optimistic scenario assuming that the economy grows at only 4 percent per year, the unemployment rate would decline by only one percentage point over the next 5 years.  Access to jobs: Among workers living in households in the poorest 10 percent, the average unemployment rate is more than 30 percent, but below 15 percent among the richest 10 percent. In the poorest households, less than 20 percent of workers are in formal wage employment in the private or public sectors (depending heavily on agricultural and commercial self-employment), compared to 75 percent among the richest.  Investments and jobs: Optimization of the distribution and quality of jobs will not automatically result from increased investment alone. Simulations show that a 4 percentage points’ increase in the investment rate may create 76000 jobs (net). However, in the absence of reforms to guide the allocation of investments, most jobs would be created in sectors with low value added per worker, notably in construction, trade, non-metal industries, and transportation services.  Toward an integrated jobs strategy: The starting point of a jobs strategy for Tunisia is to define specific social objectives in terms of jobs to inform policies and programs at three levels: (i) facilitating the creation of new jobs through private sector investments , taking into account regional and population disparities in terms of labor market outcomes; (ii) upgrading production technologies and increasing the productivity of jobs in economic activities that are already underway, with a focus on small-scale or subsistence entrepreneurs in lagging regions; and (iii) connecting individuals to jobs by facilitating labor market transitions, from inactivity or unemployment into jobs or from low- to high-productivity jobs. Source: World Bank (2015). Towards a Job Strategy in Tunisia, as part of the Tunisia Systematic Country Diagnostic. Washington DC: World Bank. Employment Trends Since the 2011 revolution, labor market outcomes in Tunisia have worsened, particularly in interior regions. After jumping from 13 percent in 2010 to 18.9 percent in 2011, unemployment gradually decreased to 15.2 percent in 2014 (Figure 5). While some of the decrease was due to the initial recovery of selected labor-intensive sectors, the majority has been due to successive waves of public sector recruitment.12 Unemployment is highest in the western and southern governorates, headed by the southern governorates of Tataouine (37 percent) and Tozeur (25 percent) and the central governorates of Sidi Bouzid and Kasserine (24 and 23 percent, respectively) (Figure 6). In absolute terms, Tunis has the highest number of unemployed persons at 12 percent of the national total, of which 83 percent are low skilled workers (Figures 7 and 8). Figure 5 Evolution of Unemployment Rates, 2006 – 2014 30 25 Féminin, 21.5 Unemployment rate (%) 20 15.1 Total, 15.2 15 12.5 Masculin, 12.7 10 11.5 5 0 2006 2007 2008 2009 2010 2011 Q4 2012 Q4 2013 Q4 2014 Q1 Source: Tunisia National Statistics Institute (INS), Labor Force Surveys and Quarterly Data, 2014. Unemployment is more pronounced among women, youth, and low-income groups, although the majority has only completed secondary education. Unemployment rates are highest among youth (35.7 percent), women (21.5 percent), and university graduates (31.9 percent, with women at 41.9 percent). Despite recent improvements, Tunisia’s labor force participation rate remains low (51.7 percent), particularly among women (28.2 percent).13 In absolute terms, the majority of the unemployed remain low-skilled workers (67 percent), although in selected governorates the trend is reversed, such as Kibili and Monastir where only one third of the unemployed are low skilled. This trend may be explained by migration patterns, the dominance of agriculture (in Kibili) and low-value added services (in Monastir), or other factors related to the type of investments made in these regions. Evidence also suggests that social norms and the role 12 Tunisia Labor Force Statistics, National Statistics Institute (INS). Tunis: INS. 13 Key Indicators of the Labor Market (KLIM) Database, International Labor Organization, 2015. 8 assigned to women within the family remain critical determinants of women’s participation in the labor force in the region.14 Figure 6 Unemployment Rate by Region Figure 7 Distribution of Unemployed Persons Unemployment rate by region, 2010 versus 2013 (%) Tataouine 37.0 Tozeur 24.8 Sidi Bouzid 24.4 Kasserine 23.4 Gafsa 22.3 Gabes 21.7 2013 Kibili 20.8 2010 Siliana 20.5 Medenine 20.4 Beja 19.7 Jendouba 19.1 Menouba 18.7 Tunis 18.6 Kairouan 18.2 Ben Arous 16.8 Mahdia 16.4 Sousse 14.0 Ariana 12.6 Kef 10.8 Sfax 10.7 Bizerte 10.7 Nebeul 10.0 Zaghouan 8.2 Monastir 7.8 0.0 10.0 20.0 30.0 40.0 Unemployment rate (%) Source: World Bank staff calculations. INS Labor Force Quarterly Data, 2007-2014. 14 World Bank (2012). Jobs for Shared Prosperity: Time for Action in the Middle East and North Africa. World Bank: Washington DC. 9 Figure 8 Composition of Unemployed Population by Skill Level across Regions, 2013 Source: World Bank staff calculations. INS Labor Force Quarterly Data, 2007-2014. Note: low-skilled: secondary education or below; high-skilled: university education or above. The incomplete school-to-work transition and poor labor mobility are associated with a low quality of education and a skills mismatch. Tunisia performs relatively poorly against comparable middle-income countries on TIMSS and PISA. In addition, there appears to be some mismatch between the skills required by the labor market and those acquired by students through formal schooling. Findings from the most recent Investment Climate Assessment (ICA) revealed that the workers’ skills and education are the second leading constraint to doing business in Tunisia, with 39 percent of firms sampled citing this factor as a major barrier, second only to political instability at 56 percent.15 Skills are generally perceived as a relatively greater barrier outside of Tunis than in the capital (Figure 9), which is likely associated with the type of firms, sectoral differences, and the demand for certain skills that also varies by region and type of skill. It is then no surprise that the rates of underemployment are exceptionally high among university graduates. Underemployment reaches 30 percent for technicians and 36 percent for graduates with humanities degrees. 15 World Bank (2014). Investment Climate Assessment: Enterprises’ Perceptions in Post -Revolution Tunisia, 2014. World Bank: Washington DC. 10 Figure 9 Skills as a Perceived Barrier to Firm Growth by Type of Firm, Region, and Sector Source: World Bank staff calculations, Tunisia Investment Climate Assessment/Enterprise Survey, 2102. Note: Shows percent of firms citing the factor as a “severe” or “very important” constraint. N=600. Quality of Jobs Services account for the majority of employment in Tunisia, with limited employment in sectors with high-value added. As of 2014, the majority of Tunisians (51 percent) were employed in the services sector, followed by manufacturing (19 percent) and agriculture and non- manufacturing industries (15 percent each) (Figure 10). In the services sector, public sector employment (education, health, and general administration) accounted for almost half of employment, equivalent to nearly 670,000 civil servants (Figure 11). Between 2007 and 2014, the public sector witnessed a 17 percent jump in the number of jobs. During the same period, the mechanical and electrical sector demonstrated the highest rate of job growth (50 percent job growth), followed by real estate services (37 percent), commerce (25 percent), mining and energy (24 percent), and construction and chemicals (21 percent each) (Figure 12). By contrast, important sectors have seen job losses, namely agriculture (11 percent loss) and unclassified manufacturing industries (10 percent loss). Unclassified general job loss of 61 percent was also witnessed. Other job losses were seen for hotels and restaurants (2 percent loss) and textiles (nearly 1 percent). 11 Figure 10 Share of Jobs by Sector Source: World Bank staff calculations. INS Labor Force Quarterly Data, 2007-2014. Figure 11 Number of Jobs by Sector Source: World Bank staff calculations. INS Labor Force Quarterly Data, 2007-2014. 12 Figure 12 Net job creation by sector Source: World Bank staff calculations. INS Labor Force Quarterly Data, 2007-2014. It is worth nothing that the quality of jobs has been low overall, both in terms of informality and productivity. Up to half of the workforce in Tunisia is informal, leaving workers vulnerable to shocks. Approximately 30 percent of jobs in the private sector lack social insurance. The majority of these workers are either in informal wage employment (35 percent) or in self- employment (30 percent). Overall, 45 percent of the labor force does not have an employment contract or benefits of any kind,20 translating into high informality. Exacerbating this informality is the poor access to finance at the household level, where nearly two-thirds of Tunisian adults lack bank accounts or have partial access to formal financial services.16 Coupled with informality, low productivity is among the most salient features of the Tunisian labor market. The growth of output per worker was around 2.5 percent on average over the past decade, below most comparable countries in MENA and emerging economies in the European Union. It is estimated that the vast majority of the labor force, at 77 percent, is employed in low-productivity jobs, notably in sectors such as agriculture, textile, diverse manufacturing, commerce, construction, and public infrastructure. High-productivity service sectors absorbed only 7.7 percent of total employment. Lagging job creation in Tunisia stems from a deteriorating investment climate, cronyism, and relatively high labor costs. At the core of the country’s economic impasse is a protected 16 World Bank (2015). Overview of Financial Inclusion in Tunisia, as part of Global Findex Database. World Bank: Washington DC. 13 regulatory environment resulting in lack of competition and a large bureaucratic burden, inhibiting labor legislation, a financial sector hampered by governance failures, policies regulating services that limit competition, and industrial and agricultural policies that introduce distortions and deepen regional disparities.17 Privileges captured by a few politically connected firms under the old regime are associated with many of the policy distortions that have weakened private sector and job growth.18 The tax wedge in Tunisia, at approximately 35 percent, is high relative to other middle-income countries, encouraging young, small, and low-productivity firms to opt out of the formal sector.18 Public Sector Employment The dominance of public sector employment crowds out the pool of workers for the private sector. Over 90,000 new employees joined the public sector in 2011 and 2012 alone, the majority of them through the regularization of contract workers, thus doubling annual recruitment numbers. Along with promotions and salary increases, this recruitment drive led to a 44 percent increase in the wage bill between 2010 and 2014, which is higher than the 28 percent increase between 2006 and 2009. Benefits and wages are vastly more generous in the public sector than in the private sector. For example, on average, a 50 percent higher wage premium exists in the public sector relative to the private sector for a worker with a secondary education. The average wage replacement rate for pensioners in the public sector is nearly 90 percent, as compared to 60 percent in the private sector. These premiums help to explain queuing for public sector employment, especially amongst skilled youth.19 Labor Regulations The structure of Tunisia’s labor regulations in terms of contracts and wages may create disincentives for firms to create jobs. While some provisions in the labor code, such as work- time arrangements, are relatively flexible, other entitlements, such as maternity benefits, are below internationally accepted International Labor Organization (ILO) standards. No clear provisions exist for dismissing individual workers for economic or performance reasons, with the exception of complex tribunal procedures that most employers avoid, and collective dismissal is restricted by stringent criteria.20 The minimum wage in Tunisia is close to 25 percent of value added per worker and appears aligned with the country’s average level of labor productivity. The wage distribution in Tunisia shows that the majority of workers in the private sector earn close to the minimum wage. Preliminary simulations suggest that, all else held equal, the minimum wage policy in Tunisia may be exacerbating youth unemployment and informality. Going forward, 17 World Bank (2014). The Unfinished Revolution: Bringing Opportunity, Good Jobs and Greater Wealth to All Tunisians. Development Policy Review. Washington DC: World Bank. 18 World Bank (2014). Jobs or Privileges: Unleashing the Employment Potential of the Middle East and North Africa. World Bank: Washington DC. 19 World Bank (2014). Public Sector Size and Performance Management: A Case-Study of Post-Revolution Tunisia. Policy Research Working Paper Number 7159. World Bank: Washington DC. 20 World Bank (2014). Labor Policy to Promote Good Jobs in Tunisia. World Bank: Washington DC. 14 mechanisms for setting the minimum wage will need to be revisited to better reflect social and economic policy objectives. Active Labor Market Programs Active labor market programs (ALMPs) in Tunisia include on-the-job training, skills development, and wage subsidy programs (Figure 13), complemented by public employment services (PES). The National Employment Fund (FNE) is the country’s main source of financing for ALMPs, including wage subsidies and regional development programs that finance public works, benefiting 405,000 individuals during the period 2011–2014. Programs to promote insertion into wage employment consist primarily of wage subsidies for on-the-job training, and benefit a large number of high-skilled individuals (about 114,000 beneficiaries in 2011).21 These programs have been undergoing reforms since 2013.22 Tunisia also spent about 380 million TND on labor-intensive public works programs targeting the poorest segments of the population and reaching approximately 180,000 beneficiaries. A recent assessment of these programs23 shows that many of public works sites are never completed and, in many cases, wage payments are given to workers who do not show up for work. Figure 13 Beneficiaries of Programs financed by the National Employment Fund Source: Administrative data from the Ministry of Vocational Training and Employment, 2011, latest available data. Note: AMAL program was replaced in 2013 by new wage subsidy and training programs.22 21 The fund has been financed from general revenues, proceedings of state privatization operations, and contributions from private donors as well as from income, payroll, and sales taxes. 22 Reforms undertaken in 2013 consolidated job insertion programs financed by the National Employment Fund, from five programs into three programs: two supporting training and wage subsidies, and a first-time employment program [Programme d’encouragement à l’emploi (PEE)]. As of 2015, implementation remains slow. 23 World Bank (2013). Review of Tunisia Employment Fund 21-21. (Background Paper). 15 The majority of ALMPs implemented by the Agence Nationale de l’Emploi et du Travail Indépendant (ANETI) are wage subsidy programs that primarily target high-skilled youth. Most wage subsidy programs provide beneficiaries with a stipend and in-class or on-the-job training, as well as cover social security contributions. However, these programs are poorly targeted, fragmented, and duplicative, which hinders their effectiveness and efficiency. To address these challenges, the Ministry of Employment consolidated and streamlined all wage subsidy programs into two new ones, to be implemented though results-based PPPs with eligible private providers and civil society organizations: a “training voucher” (Chèque pour l’amélioration de l’employabilité) and a “wage voucher” (Chèque d’appui à l’emploi). However, the Ministry of Employment has faced significant delays in operationalizing these reforms. In fact, to date, the wage voucher has yet to be launched, while the training voucher is only now currently being piloted. This is primarily due to the limited capacity of the Ministry of Employment and ANETI regarding designing and managing the transition to the new consolidated and results-based system. On average, job insertion rates for Tunisia’s ALMPs were approximately 20 percent in 2011. PES managed by regional Ministry of Vocational Training and Employment (MVTE) bureaus are not perceived as being effective even where job opportunities exist, particularly among low- skilled and low-income groups (Figures 14 and 15). Overall, only 16 percent of PES users report being satisfied with the quality of services offered, ranging from 11 percent among those with no education to 22 percent among the high skilled.24 Higher-income job seekers often do not use these services and rely instead on informal networks, explaining their lack of dissatisfaction with services. No major differences in satisfaction appear across gender, where only 13 percent of females and 18 percent of males report being satisfied with services. Figure 14 Satisfaction with Public Employment Services by Educational Level 25.0% 21.7% Percent of beneficiaries rating satisfactory services (%) 20.0% 16.0% 16.1% 14.8% 15.0% 11.3% 10.0% 5.0% 0.0% None Primary Secondary University Total or above Source: World Bank Tunisia Governance in Service Delivery Survey, 2012. N=820. Shows percent reporting “3” or above on a five-point scale, 1=fully dissatisfied; 5=highly satisfied. 24 World Bank (2015). Raising the Bar: Institutionalizing Participatory Monitoring of Service Delivery in Tunisia. Background Paper. Washington DC: World Bank. 16 Figure 15 Satisfaction with Public Employment Services by Income Level 35.0% 32.6% Percent of beneficiaries rating 30.0% satisfactory services (%) 25.0% 25.0% 21.7% 20.0% 16.1% 16.1% 15.0% 11.1% 10.0% 5.0% 0.0% Poorest 2 3 4 Richest Total Source: World Bank Tunisia Governance in Service Delivery Survey, 2012. N=820. Shows percent reporting “3” or above on a five-point scale, 1=fully dissatisfied; 5=highly satisfied. Emerging, new approaches to ALMPs shed light on ways to further improve employability. Recent evidence from a pilot direct employment program in southern Tunisia, where unemployment rates are nearly double the national average, indicates that skills development is important for the majority of beneficiaries (Figure 16).25 Beneficiaries that received on-the-job skills training and work experience most closely overseen by employers were more likely to find wage employment and/or self-employment following the program. A preliminary assessment shows that among beneficiaries who found jobs after the program, 65 percent found wage employment and 35 percent became self-employed (Figure 17). Figure 16 Beneficiary Motivations for Accessing Pilot ALMP among Low-Skilled Workers Source: World Bank Participatory Service Delivery Project Monitoring and Evaluation System, Beneficiary Assessment, 2014. N=6100. 25 World Bank (2015). Implementation Completion and Results Report for Tunisia State and Peacebuilding Fund (SPF) Participatory Service Delivery for Reintegration Project. Report No: ICR00003568. Washington DC: World Bank. 17 Figure 17 Distribution of Jobs found by Type and Region Six Months after Program Exit Source: Tunisia SPF Project Monitoring and Evaluation Database, 2014. Sample size = 183. Overall, emerging ALMP pilot programs offer insights into how to enhance the effectiveness of ALMPs going forward to catalyze job insertion. Critical success factors include (i) strategic targeting of ALMPs and direct employment interventions to job-creating sectors; (ii) allowing flexibility in the design of the programs to meet local needs; (iii) ensuring comprehensive program monitoring; (iv) supporting beneficiaries with follow-up services through intermediation with local associations, training centers, municipalities, entrepreneurship services and private sector networks; and (v) partnering with or contracting out to local civil society and private sector actors during design and implementation. Conclusions For labor outcomes to improve in Tunisia, developing a comprehensive jobs strategy would lay the foundation for short- and medium-term reforms to promote sustained employment. In 2013, Tunisia launched the preparation of a 2013-2017 National Employment Strategy that was still not finalized as of 2015. However, a renewed jobs strategy would include measures to strengthen the supply and demand side jointly and in a targeted way. As part of this, the modernization of labor policies will help improve the competitiveness of the Tunisian labor force. In the short-term, strengthening delivery and evaluation of ALMPs for the most vulnerable will be advantageous. Key medium- to long-term reforms include aligning contracts, wages, dismissal rules, and public sector employment policies with international standards. 18 CHAPTER 3 INTEGRATING SOCIAL SECURITY FOR BETTER JOBS Consolidating Tunisia’s social protection and labor system entails better connecting citizens to social security benefits throughout the life cycle in a more equitable, sustainable way. Tunisia currently does not have a national unemployment insurance program or policies to provide services to workers who lose their jobs in the private sector. With an informality rate of nearly 45 percent, there are wide differences in social benefits between the public and private sector. Social insurance in Tunisia mainly includes pensions, maternity benefits, disability benefits, and emergency household loans. Since 2013, Tunisia’s key stakeholders have committed to revitalizing social dialogue26 needed to implement a new vision for social security. The focus of this chapter is on assessing options for pensions and health insurance to contribute to this process, with a discussion of future considerations for unemployment benefits. Pensions Social insurance schemes in Tunisia are contributory and include two main schemes for public and private sector workers. Public sector workers are covered by the Caisse Nationale de Retraite et de Prévoyance Sociale (CNRPS, or National Pension and Social Contingency Fund), including civil servants, the military, security forces, contractual workers, local authorities, and employees of State-owned enterprises. Private sector workers are covered by the Caisse Nationale de Sécurité Sociale (CNSS, or National Social Security Fund). The pensions system dates from 1898, when a pay-as-you-go (PAYG) system was established only for civil servants. Today, the system is highly fragmented, spanning different pension schemes for various categories of workers. The number of contributors as of 2012 stood at approximately 1.5 million for CNSS and 0.7 million for CNRPS. CNRPS, created in 1985, covers risks associated with old age, invalidity, survivorship, and sickness for civil servants. Family allowances, maternity and work injuries are fully paid by the government and State-owned enterprises. The total contribution rate for CNRPS is 15.7 percent, based on 8.95 percent on the part of the employer (government) and 6.75 percent on the part of the employee. CNSS covers most risks, but benefits vary across the eight sub-regimes for different groups of workers and by fixed and open-ended contracts (which are limited). These regimes include (i) salaried (non-rural) workers; (ii) salaried rural workers; (iii) improved salaried rural workers; (iv) non-salaried workers; (v) Tunisian workers abroad; (vi) students; (vii) lower-income earners; and (viii) writers and artists not covered by any other regime. The administrative boards of these funds comprise tripartite representation by the State, employers (namely represented by UTICA), 26 World Bank (Kuddo, A, Weber M and Robalino D) (2015). Balancing Regulations to Promote Jobs: From Employment Contracts to Unemployment Benefits. Washington DC: World Bank. 19 and workers (namely represented by UGTT). The private social insurance sector (insurance companies and mutual benefit organizations) remains very limited and primarily spans complementary and optional health insurance. For CNSS, the contribution rate is 12.5 percent, split between 7.76 percent on the part of the employer and 4.73 percent on the part of the employee. Coverage by the mandatory pension systems in Tunisia is one of the highest in the MENA region but lags behind other comparable economies (Table 1). Only 50 percent of the current labor force in Tunisia contributes to either of the two main schemes (CNRPS and CNSS as represented by RSNA, its main scheme). After taking smaller schemes into account, the total coverage rate is approximately 70 percent of the labor force. Approximately 37 percent of the working age population (aged 15-64 years) comprises active contributors to pension schemes. In Tunisia, as in Egypt and Morocco, the statutory retirement age is 60 for both men and women, which is similar to the retirement age found in most countries in MENA of 60 years for men and 55 years for women, but lower than the age found in Malta of 65 years for both men and women. Eight countries in MENA have integrated pension systems, while in seven countries pension schemes for private sector workers and civil servants are still separated, as is the case in Tunisia. Table 1 Coverage Rates and Public Pension Spending, Selected Middle- and High-Income Countries Total Working First Second Number of Pension Labor Force Age Active Active Active Schemes Spending Year (thousands) Population Coverage Coverage Year Contributors Covered (as a % (2) (thousands) Definition: Definition: (thousands) of GDP) (3) (1)/(2) (1)/(3) (1) Czech Republic 2007 4,968 5,208 7,363 95.4% 67.5% ns 9.20 2010 Portugal 2005 5,088 5,529 7,104 92.0% 71.6% ns 11.60 2009 Sweden 2005 4,263 4,800 5,906 88.8% 72.2% ns 8.20 2010 Poland 2008 14,366 17,650 27,246 81.4% 52.7% ns 11.90 2010 Korea, Rep. 2011 19,886 24,898 36,703 79.9% 54.2% ns, cs, ss 1.60 2005 Tunisia 2011 2,551 3,896 7,437 65.5% 34.3% ns, ss 5.11 2012 Turkey 2008 15,100 25,763 49,480 58.6% 30.5% ns, cs 6.30 2010 Chile 2010 4,635 8,037 11,745 57.7% 39.5% ns 5.50 2010 Egypt 2009 14,614 26,536 52,400 55.1% 27.9% ns, cs, ss 3.00 2010 Malaysia 2010 6,400 11,970 22,776 53.5% 28.1% ns 3.75 2012 Jordan 2010 821 1,553 3,649 52.9% 22.5% ns, cs, ss 2.00 2005 Morocco 2011 3,314 11,386 21,247 29.1% 15.6% ns, cs 2.70 2011 Mexico 2010 13,780 49,617 73,215 27.8% 18.8% ns 2.40 2012 Vietnam 2010 10,585 51,137 61,188 20.7% 17.3% ns 2.50 2004 Source: World Bank Pension Database, 2012. Most recent year shown. Note: n/a: not available. ns: national scheme, cs: civil servants scheme, ss: special scheme, os: other scheme. Among all contributors, more than 70 percent are part of CNRPS or RSNA, and the rest of the pension programs cover less than 700,000 people. Table 2 below shows key characteristics of the design and performance of the mandatory pension programs in Tunisia.27 Pension spending in 2011 already represented 5.1 percent of GDP, revenues represented less than 4 percent, and reserves, 3 percent. 27 Revenues from contributions are estimated from the total number of active contributors multiplied by the average wage and by the contribution rate of each program. Pension expenditures are the average pension multiplied by the number of beneficiaries. 20 Table 2 Assessment of Pension Schemes in Tunisia, 2011-2012 Source: World Bank staff calculations.28 Note: CNRPS = Caisse Nationale de la Retraite et de la Prevoyance Sociale (pension fund for civil servants). RSNA, RSA, RTNA, RTA, RSAA, RTFR = schemes managed under CNSS (Caisse Nationale de la Sécurité Sociale) for private sector workers : Régime des Salariés Non Agricoles (formal sector, agricultural); RTNA = Régime des Travailleurs Non-Salariés Secteur Non Agricole (self-employed, non-agricultural); RTA = Régime des Travailleurs Non Salariés Secteur Agricole (self-employed, agricultural); RSAA = Régime des Salaries Agricoles Amélioré (pension scheme for formal, agricultural enterprises); RTFR = Regime des Travailleurs à Faible Revenu (low-income self-employed). 28 World Bank (Pallares-Millares M and DeMarco G) (2012). Tunisia Pensions Study (Background Paper). Wages and pensions benefits in the public sector are on average 30-50 percent higher than in the private sector. The current distribution of average wages and average pension benefits by age and gender for the two pension schemes (CNRPS and CNSS as represented by its main scheme, RSNA) are shown in Figures 18 and 19. Projected benefits for both categories have been modeled directly as benefit ratios or relative to the average wage, in the case of the disabled, and as a ratio of new old age pension, in the case of survivors. Benefit ratios for both the new disabled and survivors are set to remain constant over time. The average length of service of new retirees is assumed to be constant during the simulation period. While wages are generally similar by gender in the public sector, both wages and pensions can be up to 15 percent higher for men than for women in the private sector, in some cases even more. More analysis is needed to understand these differences in the future. Figure 18 Wage Distribution among Pension Beneficiaries, 2011-2012 a. Wage distribution in CNRPS 25000 20000 Average Wage Women 15000 10000 5000 Men 0 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 Age b. Wage distribution in CNSS (RSNA) 9000 8000 7000 Men 6000 Average Wage 5000 4000 3000 2000 Women 1000 0 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 Age Source: World Bank staff and MOSA PROST input file for Tunisia. Figure 19 Pension Benefits Distribution by Age and Gender, 2011-2012 a. Average Pension in CNRPS 8000 Men 7000 6000 Average Pension 5000 4000 3000 Women 2000 1000 0 50 53 56 59 62 65 68 71 74 77 80 83 86 89 92 95 98 Age b. Average Pension in CNSS (RSNA) 6000 Men 5000 4000 Average Pension 3000 2000 1000 Womens 0 50 53 56 59 62 65 68 71 74 77 80 83 86 89 92 95 98 Age Source: World Bank staff and MOSA PROST input file for Tunisia. By the year 2018, under reasonable assumptions and with no reform, the deficit of both schemes would represent almost 2 percent of GDP, and reserves in both schemes would be completely depleted. Figure 20 and Table 3 present the overview and details, respectively, of financial flows and reserves projected through 2018. CNRPS faces a deficit given that the combination of its revenues and investment returns do not cover the expenditures. RSNA was also showing a deficit in 2014, but still has enough reserves to finance expenditures. Unless reforms take place, by the year 2018 both funds will likely continue to be in need of increased levels of external financing. However, a reliance on public budget transfers has increasingly become unsustainable. 23 Figure 20 Projected Deficit, CNRPS and RSNA 2010-2018 (percent of GDP) Source: World Bank staff calculations based on data provided by CRES for the baseline year (2010). Table 3 Financial Flows and Deficit Projected, CNRPS and CNSS (RSNA) 2010-2018 CNRPS (public sector) CNSS (RSNA) (private sector) 2010 2014 2018 2010 2014 2018 Revenues (TND millions) 1537.1 2014.5 2542.1 989.7 1288.9 1680.5 Investment returns (as percent of 0.4 0 0 4.6 1.5 0 revenues) Expenditures (TND millions) 1664.8 2515.5 3701.8 1198.9 1511 2393.5 Administrative costs (as percent of 2.6 2.3 1.9 5 5.2 4.3 expenditures) Total current balance (TND millions) -127.7 -501.1 -1159.8 -209.2 -222.1 -713 Final reserves (TND millions) 27.2 0 0 1450.8 125.4 0 As percent of GDP 0.04 0.00 0.00 2.31 0.14 0.00 Deficit/Surplus projected (TND millions) -100.5 -501.1 -1159.8 1241.6 -96.7 -713 As percent of GDP -0.16 -0.56 -0.9 1.98 -0.11 -0.55 GDP (TND millions) 62813 89482 128867 62813 89482 128867 Source: World Bank staff calculations based on data provided by CRES for the baseline year (2010). Overall, both pension schemes are financially unsustainable as expected income flows are not enough to cover expected expenditures. At 5.1 percent of GDP, total pension expenditures in Tunisia (for all schemes of CNRPS and CNSS) are high for the countries’ level of income and its demographic structure.29 Countries with a similar population structure and coverage rates as Tunisia spend less than 4 percent of GDP. 29 See Pallares-Miralles M, Romero C and Whitehouse E (2012). International Patterns of Pension Provision II : A Worldwide Overview of Facts and Figures. Washington DC: World Bank. 24 Moreover, expenditures have been growing rapidly and there is already an aggregated deficit of 0.2 percent of GDP deficit in CNRPS and 0.3 percent of GDP in RSNA. These deficits may be underestimated due to possible underestimation of reported wages by firms, but nonetheless, the deficit remains large. How this deficit will change over time will depend on, among other factors, the evolution of the dependency ratios, the growth of wages, and indexation policies. In the case of the CNSS, the deficit is relatively low and could be managed in the short term more readily, but nonetheless remains a challenge in the medium to long term. In the case of the CNRPS, the deficit is expected to increase more rapidly beyond 2018, reaching up to 1.3 percent of GDP by 2020. Challenges facing Tunisia’s pension schemes are multifaceted. While the challenge of the private sector is largely to improve coverage and equity of pensions, the public sector scheme is primarily faced with unsustainability. First, there is a fundamental misalignment between contribution rates and benefits paid at various retirement ages for both schemes. As such, the underlying principle of defined-benefit pension systems financed through a pay-as-you-go regime no longer hold in the case of Tunisia, unless parametric reforms are implemented soon. Second, with sluggish economic growth and over-absorption of workers into the public sector, relying on public transfers to cover deficits makes it even more unsustainable. Finally, the existing financing system and revenue generation policies of both pension schemes leave them ill equipped to cope with increasing fiscal pressures. The challenges are exacerbated by the lack of coordination and fragmented governance between the public and private sectors in terms of benefit levels, contribution policies, and beneficiary identification. The choice of reforms to address these challenges will need to be carefully assessed in terms of impacts and tailored to Tunisia’s social dialogue climate. The climate remains complicated by widely divergent views and underlying social value judgements on the roles of workers, employers, and the state in financing. At the same time, there is emerging consensus on the need for fundamental measures that can be undertaken to improve governance and portability and to assess the collective impact of potential reforms. A combination of parametric and structural reforms would help address these challenges gradually over the short to long term. To avoid exacerbating fiscal challenges, Tunisia may want to first strengthen the financial sustainability of both CNRPS and CNSS and improve coverage among workers in the private sector, including vulnerable workers through social pension schemes. Following these measures, reforms might be undertaken to consolidate the system in the long term by improving portability and reducing fragmentation between regimes within CNSS, fixed and open-ended contracts, and the public and private sectors. As of 2015, Tunisia had already identified potential parametric measures to try to address sustainability in the public sector, including an increase in the retirement age. Other short- term measures include rationalizing entry into public sector employment and voluntary early retirement. Over the mid-term, the reform of certain State-owned enterprises (SOE) might also entail these types of measures. Over the long term, underlying financing and revenue generation schemes should be assessed and examined in terms of more explicitly linking financing to 25 benefits, including policies related to financing through taxation, capitalization, and private markets. However, in order to guarantee a longer surplus in the systems, recommended reforms should include a combination of measures as part of a coherent, short- to long-term reform package so as to avoid jeopardizing long-term outcomes. For instance, only increasing the retirement age without decreasing accrual rates will likely render the schemes even more unsustainable. Given that reform options carry their respective trade-offs, careful evidence-based analysis of a package of both short- and long-term measures and their net effect as a whole would best inform future policy decisions. Health Insurance Despite sustained health improvements, Tunisia’s health care costs and the impact on workers and households represent a growing challenge. Costs have been rising and regional disparities in health have widened.30 Health care in Tunisia is financed through a mix of public and private sources. Between 1980 and 2013,31 total health expenditure grew by over 100 percent from 3.2 percent to 7.1 percent of GDP. The system has been financed through a mix of sources since 2007, with the following levels in 2013: general governmental revenues (26.0 percent of total health spending), CNAM (33.3 percent), private insurance (4.4 percent), and household out-of- pocket payments (36.3 percent). Public spending on health decreased from 2.7 percent of GDP in 1995 and 1.8 percent in 2013. Private expenditures accounted for 41 percent of total health expenditures in 2013, approximately 87 percent of which were co-payments at the point of use and 13 percent of which represented health insurance premiums. This level of out-of-pocket spending has remained constant since reforms were undertaken to consolidate CNAM in 2007, despite the need to reduce the burden on households. To date, only 50 percent of the population is covered by the national health insurance fund, although an additional 34 percent receive limited health care benefits through a non- contributory social safety net program, Assistance Medical Gratuite (AMG) (Figure 21). Nearly 11 percent of the population remains uninsured. Health insurance is provided by a public fund, Caisse Nationale d’Assurance Maladie (CNAM), created in 2007 following the merger of various schemes. CNAM is financed by mandatory contributions from employees and public/private employers; the contribution rate is 6.75 percent for employees and nearly 15 percent for employers. CNAM provides health insurance through social security contributions for 3.3 million civil and private sector employees and their dependents. 30 World Bank (Arfa C and Elgazzar HA) (2013). Consolidation and Transparency: Transforming Tunisia’s Health Care for the Poor. Universal Health Coverage Studies Series (UNICO) UNICO Studies Series No. 4. Washington DC: World Bank. 31 Latest available data. Source: World Health Organization (2015). National Health Account Estimates for Tunisia. Geneva: World Health Organization. 26 Social safety net beneficiaries receive health insurance cards that cover basic services through Ministry of Health public providers (hospitals and health centers), managed by the Ministry of Social Affairs. One program provides basic free services for poor households (AMG-1) and another provides subsidized services for low-income households (AMG-2). Despite these programs, beneficiaries often pay for primary care services and pharmaceuticals through out-of- pocket payments due to long queues, inadequate quality, or a lack of medications in stock, particularly outside of the capital and main cities. The fragmentation between CNAM, AMG-1, and AMG-2 further widens regional and income disparities in access. Figure 21 Health Financing Coverage Rate (Health Insurance and other Schemes) (percent) 60% 50% Coverage rate (percent) 50% 40% 34% 30% 20% 11% 10% 6% 0% Social security Complementary Medical Uncovered (CNAM for CNSS mutuals assistance CNRPS) (subsidized and free health card) Source: National survey on health services access and utilization in Tunisia, 2006, Institute of Public Health, Tunisia, and Ministry of Social Affairs, 2015. Rates remain similar between 2006 and 2015. During the period 2011-2015,32 CNAM expenditures amounted to 1.5 million TND, or 2 percent of GDP, having increased at an average annual rate of growth of 19 percent during 2007-201133 (Figure 22). CNAM purchases health services from public and private providers, but there is no formal prospective contracting system in place. Instead, services are retrospectively reimbursed on a fee-for-service basis. The use of private services has been increasing, with 31 percent of all members using private services in 2011, up 20 percent from a 26 percent utilization rate in 2008 (Figure 23). 32 Latest available data is for 2011-2012 and is expected to have remained the same for 2015. 33 Latest available data. 27 Figure 22 Health Insurance Expenditure, 2007-2011 1600 Total 1400 1200 Active workers' Expenditure (million TND) scheme 1000 800 600 400 Occupational 200 health Social schemes 0 2007 2008 2009 2010 2011 Source: World Bank staff, CNAM, 2012. Social schemes include schemes for students and low-income households. Figure 23 Distribution of Health Services Used, CNAM, 2008-2011 Source: World Bank staff, CNAM, 2012 Out-of-pocket spending remains a burden on households, particularly in the private, informal sector and among AMG beneficiaries. Approximately 12 percent of all households faced catastrophic health care payments in 2006,34 which is higher than the international recommended maximum of 5 percent. Household expenditures have risen due to the increasing preference for private services owing to the poor quality of public services, spending on pharmaceuticals, and significant co-payments that are incurred even under State-subsidized schemes. 34 Latest available data. 28 Provider payment policies have also contributed to increasing health care costs and contributions on the part of workers, employers, and public transfers. Thirty percent of hospital expenditure is financed through historically based global budget transfers from the Ministry of Finance and 70 percent through fee-for-service billing and co-payments through CNAM, while public primary care is funded directly through governmental revenues. Although CNAM has agreements with certain private clinics and pharmacies to deliver services, contracting criteria are based on basic commercial certification rather than on service delivery or provider performance. Between 2008 and 2010, the government adopted a reform to roll out a basic system of performance-based budgeting in several sectors including health, known as Gestion du budget par objectif (GBO). Budgets are to be allocated according to performance indicators for primary health care, regional hospitals, university hospitals, and functional agencies. As of 2011, implementation was put on hold and is expected to resume following the establishment of a permanent government in 2015. Reducing out-of-pocket payments and fragmentation between CNAM and AMG regimes can be achieved by strengthening efficiency and incentives within the financing model. In the long term, merging AMG with CNAM and providing explicit subsidies through per-capita transfers would broaden the pool among CNAM beneficiaries, reducing risk overall and expanding coverage for both public and private services contracted by CNAM. Unemployment Benefits Alongside pensions and health insurance, the design of unemployment benefits is a key pillar of social dialogue in Tunisia. Protecting income and access to health care for limited-duration unemployment spells is needed to shield against shocks and stimulate job growth. While Tunisia has only had very limited schemes, mainly in the form of severance pay for selected cases, its 2013 Social Contract calls for tripartite dialogue to institute a national scheme in the medium to long term. Building the evidence base in the short term on potential policy impacts would be a key input to guiding this dialogue. Potential reforms for the Tunisian context will need to take into account fiscal impacts, labor outcomes, and coherence with the level of and eligibility for other benefits, notably pensions, health insurance, and social safety net programs. Some of these programs currently assist de facto unemployed citizens either directly or indirectly. Therefore, coordination and identification will be important. Unemployment insurance, benefits, or services can take on multiple designs, be portable, be sourced through different contributory, taxation, or individual savings accounts, other arrangements, or a combination thereof. Several key factors will need to be evaluated with social dialogue partners in Tunisia.35 Important policy parameters to assess for an unemployment benefits scheme include technical 35 World Bank (Arvo K, Robalino D and Weber M) (2015). Balancing Regulations to Promote Jobs: From Employment Contracts to Unemployment Benefits. Washington DC: World Bank. 29 and social value parameters such as (i) the replacement rate, which defines the unemployment benefit as a share of the previous wage of the unemployed while insured; (ii) the vesting period, which defines for how many months an individual needs to contribute to the unemployment insurance scheme to be eligible for unemployment benefit reception; (iii) the maximum number of months the unemployed can receive the unemployment benefit; and (iv) societal tendencies and preferences, including the profile of the unemployed, current coping mechanisms, and behavioral impacts of various benefits. Unemployment benefits should be assessed in terms of impacts such as (i) labor shares including the share of unemployed and the informal sector, (ii) wages, (iii) fiscal impacts, and (iv) the ratio of insurance coverage of the unemployed. Forthcoming collective work by tripartite actors aims to develop evidence-based options for reforms. Conclusions Pension benefits and health insurance programs in Tunisia generally provide better coverage for civil servants than for public sector workers. However, existing programs have increasingly become unsustainable due to limited fiscal space and an inflated wage bill, particularly in the public sector. At the same time, most public sector jobs are largely limited to larger cities and more highly skilled workers. The quality of jobs in the private sector remains weak and little incentive exists to seek formal private sector employment. The lack of a coordinated social security system has therefore left lower-skilled workers and households in lagging regions particularly at risk, exacerbating inequity. Building consensus and evidence of potential impacts of reforms on poverty and social and labor outcomes will help define the contours of these reforms as part of Tunisia’s social dialogue platform. 30 CHAPTER 4 INVESTING IN PRODUCTIVE SOCIAL SAFETY NETS Although Tunisia has had social safety net (SSN) programs in place since the 1980s, its expenditures are minimal in comparison to universal price subsidies on energy and basic commodities. These expenditures remain largely inequitable and inefficient, siphoning away financing that is needed to accelerate growth and strengthen social protection and labor systems in Tunisia. During 2013-2014, social transfers accounted for approximately 7.3 percent of GDP in Tunisia, comprising 6.9 percent of universal price subsidies and 0.4 percent of social assistance. Among subsidies, approximately 1.8 percent of GDP was allocated to subsidies on basic commodities (including cereals, tomato paste, oils, and paper); 0.4 percent allocated to transportation; and the remaining 4.7 percent of GDP to energy (primarily electricity, liquid petroleum gas, diesel, gasoline). Energy subsidies include both explicit subsidies to prices and implicit subsidies to State-owned enterprises (SOEs) involved in the energy value chain. 36,37 With energy prices having dropped by mid-2015, public expenditure on total subsidies fell from its levels of 6.9 percent of GDP in 2014 and 6.5 percent of GDP by end-2015 to approximately 5.1 percent in 2016.38 Nonetheless, the inequity and lack of sustainability of the subsidy system comprises efforts to reach the most vulnerable through programs that are more productive. Benefit Incidence of Energy and Food Subsidies Despite high spending, universal subsidies generally benefit the most well off disproportionately.39 The distribution of benefits is determined a priori by the consumption of these products. While food subsidies also benefit the poor, energy subsidies are particularly inequitable. For electricity, the highest quintile consumes about 29 percent of the total benefits versus 13 percent among the lowest quintile (Figure 24). For petrol and diesel, the shares of benefits for the higher-revenue quintile amount to 67 percent for petrol and 59 percent for diesel respectively. For LPG, 15 percent of the benefits go to the lowest quintile, while 21 percent go to the highest. Overall, this distribution shows that energy subsidies tend to be pro-rich. 36 World Bank (2013). The need for better governance of public enterprises in Tunisia. Report 78675-TN. Washington DC: World Bank. 37 Implicit subsidies in this analysis have been calculated based on the price-gap methodology of analysis which takes into account the differences between the overall price and the retail price. 38 International Monetary Fund (IMF) (2015). Tunisia 2015 Article IV Consulation, Sixth Review Under the Stand-By Arrangement - Staff Report. IMF Country Report No. 15/285. Washington DC: IMF. 39 World Bank (2013). Vers une meilleure équité: les subventions énergétiques, le ciblage et la protection sociale en Tunisie (Towards greater equity: energy subsidies, targeting and social protection in Tunisia). Washington DC: World Bank. 31 Figure 24 Benefit Incidence of Specific Types of Energy and Food Subsidies (percent) Source: Household Budget and Consumption Survey (EBCM), 2010, INS. World Bank staff calculations. Note: Values were adjusted based on fuel prices from April 2013. "Q" means quintile. For all energy subsidies, the direct benefit per capita was 94 TND per capita per year on average in 2013, of which 60 percent consisted of LPG subsidies and 29 percent of electricity subsidies. For all energy subsidies, the fifth quintile (highest income) benefits from 139 TND in subsidies per capita per year, versus 62 TND per capita per year among the first quintile (lowest income). All direct benefits from energy subsidies amount to 162 million TND per year for the 20 percent of the population with higher incomes, that is to say almost 627,000 households. The 20 percent of the population with the lowest incomes, namely 400,000 households, consume 118 million TND of energy subsidies per year. 32 Impact Assessment of Potential Reforms on Welfare and Growth The impact of reforming energy subsidies on growth is expected to be positive, largely because of the positive effect on investment (Table 4). The effect consists of an average annual increase of 0.12 or 0.21 percent due either to the elimination of subsidies on LPG and petrol or the elimination of subsidies on all products, respectively. However, this is an estimate to be considered with caution because it assumes all savings would be reinvested directly into the economy. It does not take into account the allocation of savings to finance other measures, such as tax measures or other transfers to help mitigate the effects of higher prices of selected products on vulnerable households. Reforming subsidies would create the fiscal space needed to stimulate investment, and including mitigating measures for households can promote equity. Investment is expected to increase by 12.5 percent on average over the period under review following a reform of all energy subsidies. This gain will result from an average decrease of 48 percent of the public deficit. The increase in investments can be done directly via public investment, or indirectly via a reduction of the public deficit, which benefits private investment. In addition, the effects are more sensitive to reforms of diesel and electricity than to those of LPG gas and petrol. Table 4 Estimated Macroeconomic Impacts of Energy Subsidy Reforms over Five Years INDICATOR IMPACT (PERCENT) REFORM OF ALL ENERGY REFORM OF LPG AND PETROL SUBSIDIES SUBSIDIES TOTAL SUBSIDY EXPENDITURES (PERCENT) -100 -34.9 PUBLIC DEBT (PERCENT) -8 -2.7 INVESTMENT (TOTAL) (PERCENT) 12.5 4.5 EXCHANGE RATE (PERCENT) -1.8 -1.1 GROSS DOMESTIC PRODUCT (GDP) (PERCENTAGE 0.21 0.12 POINTS) PUBLIC DEFICIT (PERCENT) -47.9 -16 DEMAND FOR LABOR (PERCENT) -0.41 -0.1 UNEMPLOYMENT RATE (PERCENTAGE POINTS) 0.34 0.1 HOUSEHOLD CONSUMPTION (PERCENT) -3.7 -1.3 Source: World Bank staff calculations. Social Accounting Matrix (SAM) 2005, INS. Note: The 2005 SAM is the most recent, dated September 2013. The impact represents the average of the impact over five years in relation to the situation without reform (2014 to 2018). All subsidies = reform of LPG, petrol, diesel (gas oil), fuel and electricity subsidies. In the absence of mitigation measures, household consumption is expected to decrease following the lifting of energy subsidies and an increase in prices. The effect of an average of -1.3 percent is driven by a reform of LPG and petrol subsidies and increases to 3.7 percent if all subsidies are reformed. Over five years, the effect per year (compared to the previous year) increases gradually if diesel and electricity are added. The effect varies with a decrease of -2.8 percent during the first year to reach a decrease of -4.6 percent during the fifth year compared 33 to the previous year. This effect reflects the decline of production in certain sectors, such as water and electricity, land transport and transport industries, agriculture and fisheries, food industries, the textile industry, and telecommunication. This decline can in turn cause a rise in consumer prices. Reducing LPG subsidies may also lead to an increase in the poverty rate by 1.8 percent, based on a poverty threshold of 1,025 TND per capita (or almost US$2 per capita). For petrol, a negligible impact is anticipated, due to its relatively limited use. The savings resulting from the reform of subsidies may be used to finance targeted transfers to households and sectors. The larger the number of eligible households, the more the use of these savings for investment or tax measures to strengthen growth is reduced. For purposes of illustration, removing all subsidies over a period of 12 months would require a total budget of 145 million TND as of 2013 to cover 20 percent of the population with the lowest incomes (Table 5). These transfers can be channeled by adapting Tunisia’s existing social safety net system, which is discussed in the next section. Table 5 Targeting Scenarios for Household Compensation for Reforming All Energy Subsidies TARGETING SCENARIOS BY QUINTILE NUMBER OF HOUSEHOLDS BUDGET* (MILLION TND) Targeting up to 20 percent of the population 398,803 145 Targeting up to 40 percent of the population 832,620 278 Targeting up to 60 percent of the population 1, 301,396 406 Targeting up to 80 percent of the population 1, 816,745 528 Source: EBCM 2010, INS. World Bank staff calculations. Note: Estimates based on total suppression of subsidies within a period of 12 months. "Q" means quintile; *means as of April 2013. The indirect effects of diesel are not included. Performance of Social Safety Net Programs Social safety net (SSN) programs in Tunisia currently comprise non-contributory social assistance programs for poor households through unconditional cash transfers (UCT) and free health cards (AMG) (Table 6). Benefits are provided through the National Assistance to Needy Families Program, Program National d’Aide Aux Familles Nécessiteuses (PNAFN), managed by the Ministry of Social Affairs (MOSA). Households received a cash transfer of 100 TND per month in 2013 and 130 TND per month in 2015, accounting for approximately 20 percent of the poverty line used to determine eligibility, which is 585 TND (approximately US$ 1.25 per day). In addition, MOSA provides in-kind services such as social workers, childcare services, adult learning, and disability assistance. Although poorer regions such as Tozeur and Siliana have a high share of the local population receiving cash transfers and subsidized health cards (Figures 25 and 26), targeting efficiency is weak overall, with at least 60 percent of benefits leaking to the non-poor (defined as US$2 per day). In 2012, the Government launched a program to 34 consolidate and improve targeting of SSN programs, within the framework of broader social protection and labor (SPL) system consolidation.40 Table 6 Main Social Safety Nets Program in Tunisia, 2014 Program Budget in 2013: Share of Number of Population coverage (million TND) expenditure beneficiaries rate (percent) (households) (percent) National Program of Assistance to Needy Families (PNAFN) Cash transfers 292 69 235.000 9 Health cards, free services (AMG- 39 9 1) Health cards, discounted services 70* 17 578.000 22 (AMG-2) Disability Assistance 12 3 181.000** 1.8 Social Services 9 2 n. a. n. a. TOTAL 422 100 813.000 33 Source: Ministry of Social Affairs. Note: * Indicative. Size of household = approximately 4. n. a. = not applicable. ** Number of individuals. AMG = Assistance Medicale Gratuite (free health care programs). Social Services = services of social workers, assistance to children and adult learning. Figure 25 Incidence of Cash Transfers and AMG-1 by Figure 26 Incidence of Health Cards (AMG-2) by Governorate, 2011-2014 Governorate, 2011-2014 Source: World Bank staff calculations; Ministry of Social Affairs data. 40 World Bank (2013). Project Appraisal Document to the Republic of Tunisia for a Tunisia Social Protection Reforms Support Project. Washington DC: World Bank. 35 Eligibility for PNAFN is determined through categorical criteria assessed by local commissions overseen by MOSA. Commissions of local authorities, who now include representatives of CSOs, determine eligibility based on self-presentation and according to a list of categorical criteria. Eligibility criteria include (i) self-declared (unverified) household revenue falling below the poverty line, defined in PNAFN as 585 TND per year (US$1.25 per day), up from approximately 400 TND per year (US$0.75 per day) before the revolution; (ii) household size; (iii) number of household members with a disability and/or chronic health condition; (iv) household living conditions, such as dwelling and assets; and (v) the inability for the head of household to work due to an impairment (physical or mental). Results from analysis of benefit incidence and targeting efficiency using the Tunisia Household Consumption and Budget Survey (HCBS) for 2005-2010 show that leakage in PNAFN is relatively high at 40 percent. The system has been prone to accusations of corruption and a lack of transparency and equity and in 2011, MOSA began the process of reforming the system. Based on 2011 data from the PNAFN database (and complemented by data from household budget surveys in italics):  51 percent of beneficiaries are female (heads of households);  57 percent are over the age of 60 (with an average age of 58 years);  65 percent live in the 14 primarily western governorates comprising Tunisia’s lagging regions;  The average household size of 5 persons; and  Nearly 60 percent of beneficiaries (heads of households) have not completed primary education. Nearly 13 percent of the population among the 20 percent with the lowest incomes (who are largely under the 1,025 TND high threshold of poverty per capita per year, namely nearly US$2 per capita) are covered by social assistance. Figure 27 illustrates the rate of coverage of the PNAFN and AMG2. AMG2 covers only 45 percent of the population of the first quintile. More than 42 percent of the poorest (the first quintile) do not benefit from health coverage nor from cash assistance. In terms of benefits granted, or "generosity" of the program, the cash transfer represents 21 percent of the average expenditures for low-income households (Table 7). 36 Figure 27 Coverage Rate of Cash Transfer Program, PNAFN, (top panel) and Subsidized Health Card Program, AMG-2, (bottom panel) by Quintile (percent) Source: World Bank staff calculations, EBCM, 2005 INS. Latest available data. Table 7 Average Benefit Amount and Generosity of PNAFN Quintile Q1 Q2 Q3 Q4 Q5 Total Lowest Highest income income Transfer average benefit amount (TND per 113 133 182 246 193 148 capita) Generosity (in percent of total consumption) 21 14 14 13 5 9 Source: World Bank staff calculations, EBCM, 2005 INS. Latest available data. The distribution of the beneficiaries of social assistance shows that 40 percent of all benefits accrue to the lowest quintile, while 60 percent accrue to higher quintiles (Figure 28). PNAFN generally covers a considerable proportion of the low-income population (80 percent have a level of consumption that is lower than the national average). However, leakage to the non-poor (lowest quintile) remains relatively high. 37 Figure 28 Distribution of PNAFN Beneficiaries across Quintile (percent) Source: World Bank staff calculations, EBCM, 2005 INS. Latest available data. In terms of the non-eligible share of beneficiary households, the targeting of PNAFN benefits is less efficient as compared to that of other similar programs in countries such as Turkey (Figure 29) and Argentina. The coverage rate of the poorest population (of the first quintile) in Tunisia seems to be low compared to Argentina, the Dominican Republic, and Sri Lanka. In these countries, the coverage rate varies between 22 and 51 percent, against 12.6 percent in Tunisia. At the same time, the generosity of the PNAFN is one of the highest within middle- and low- income countries. These programs include unconditional cash transfers based on targeting mechanisms such as the categorical methods, proxy-means testing (PMT), community methods, geographical methods, self-targeting methods, or a combination of these.41 Figure 29 Share of Beneficiaries of Cash Transfers (unconditional) in the First Quintile (lowest income group), International Comparisons (percent) Source: ATLAS Database, World Bank, 2013. Note: The most recent data is shown for 2005-2010. The unconditional cash transfer programs do not entail conditions for inclusion of beneficiaries, as is the case with Tunisia for the PNAFN. 41 For more information on conditional and unconditional cash transfers and targeting, see World Bank (2008). For Protection and Promotion: The Design and Implementation of Effective Social Safety Nets. World Bank. Washington DC: http://go.worldbank.org/10PXFKKZD0 38 Impact Assessment of Targeting Reforms Using Proxy-Means Testing A comparison of Tunisia’s current PNAFN distribution with what it would be were PMT adopted shows that PMT can improve targeting (Figure 30). The cost-benefit ratio of the existing PNAFN is nearly 0.4 versus 0.77 for the simulated PMT program. The PMT approach allows for broader coverage of the eligible population and a better exploitation of budgetary resources. For example, the application of PMT to grant aid would ensure the delivery of 78 percent of the budget to the individuals with the lowest incomes versus 31 percent currently. Similarly, the PMT approach would allow for the reallocation of social transfers to the most disadvantaged regions (Figure 31). Figure 30 Comparison of the Different Scenarios for Subsidies and Targeting of Social Assistance Reform Source: EBCM 2005, INS. World Bank staff calculations. Note: Indirect effects are excluded. "Q" means 20 percent of the population (quintile). 39 Figure 31 Actual Distribution (left) versus Simulated Distribution using PMT (right) of PNAFN Incidence by Region (percent) PNAFN PNAFN-PMT 0-5 percent 6-10 percent 11-15 percent 16-20 percent 21-25 percent 26-30 percent 31-35 percent 36-40 percent Source: EBCM 2005, INS. World Bank staff calculations Conclusions Building on its well-established programs, Tunisia’s social safety net system can be strengthened by reducing the reliance on inequitable universal subsidies, improving identification and targeting, and building graduation mechanisms. Reforms will be needed to redirect consumption subsidies while protecting households from the shocks of rising costs of living through cash transfers and other forms of direct, targeted assistance. During the transition period, the government has taken steps to strengthen targeting and establish comprehensive information systems on beneficiaries and services. However, a clear strategy for subsidy and safety net reform is needed that also addresses access to labor markets and human capital- related services for safety net beneficiaries in the medium to long term. 40 CHAPTER 5 POLICY IMPLICATIONS: TOWARD A CONSOLIDATED SOCIAL PROTECTION AND LABOR SYSTEM IN TUNISIA In its shift from exclusive to inclusive growth, Tunisia’s social protection and labor system will play a central role in promoting social inclusion for the most vulnerable and connecting citizens to better jobs. The role of the public sector remains important in mobilizing financing and ensuring that critical services are better targeted to lower-income and vulnerable regions. It will also be vital in working more closely with the private sector as the main engine of job growth in the future by ensuring that labor and social security conditions become more attractive. While existing social protection programs provide more generous benefits for the middle- to upper- income groups, the challenge is ensuring that they are able to cope with the growing needs of the most vulnerable. Enhancing financial sustainability and the equity of service delivery remain key challenges to realizing this goal. Developing a consolidated social protection and labor system through short- to long-term strategies would boost social inclusion in three important ways. Such a system would (i) boost access to different public services for the poorest; (ii) facilitate labor market participation for low- and semi-skilled, vulnerable groups; and (iii) spur growth by enhancing purchasing power in lagging regions. Other country experiences such as that of Brazil, Chile, India, and Indonesia show that consolidation of key programs takes time, but important steps in the short term can lay the foundation for new trajectories. A Pending Agenda Since 2012, Tunisia has begun laying the foundations for a more effective, well-coordinated system. The development of the Social Contract framework represents an important step toward developing a coherent policy agenda. On the delivery side, the launch of the design of a unique social identification system will help pave the way for greater coordination on targeting and benefits across different levels of vulnerability (Figure 32). Given Tunisia’s challenges, the pending agenda focuses on modernizing the system more comprehensively. Key recommendations comprise three main types, namely: (i) enhancing implementation and delivery platforms; (ii) strengthening institutional capacity and governance; and (iii) improving financial sustainability. Specific recommendations are summarized in Table 8. Over the short term, enhancing service delivery platforms can help improve implementation and efficiency to reach the most vulnerable. Understanding how benefits are delivered, distributed, duplicated, or entirely omitted across levels of vulnerability will be central to consolidating the system as a whole. Unemployment has largely fallen on youth and women, who are among the bottom 40-60 percent by income, with the vast majority (67 percent) of the 41 unemployed being low skilled in underserved regions. Nearly 30-45 percent of the employed labor force either do not have contracts or do not contribute to social security, particularly among the poorest 40 percent. Finally, among the poorest 20 percent, only 12 percent of Tunisian's benefit from the national cash transfer program. Enhanced delivery platforms can be achieved by:  Developing public-private implementation systems for skills development and income generation mechanisms in key sectors with high employment potential for out-of-school, low-skilled, vulnerable populations, as part of a private sector-led jobs strategy;  Introducing transparent targeting and monitoring systems for existing social safety net and active labor market programs;  Developing platforms for digital payment mechanisms to replace postal transfers and enhance financial literacy among households; and  Strengthening coordination of and promoting access to essential public services (notably nutrition, education, health, and water in Tunisia’s lagging regions) for social safety net programs. Figure 32 Level of Vulnerability by Population Group as of 2015, Tunisia Source: World Bank staff in consultation with representatives of the Government of Tunisia 42 Table 8 Key Challenges and Options for Reform CHALLENGES CONSTRAINTS OPTIONS FOR REFORM LABOR ACTIVATION MECHANISMS • Low productivity and lack of • Sluggish job creation particularly for most vulnerable • Develop multisectoral and regional job creation strategy targeting sufficient job growth (notably • Wage-setting mechanisms and contracting regulations both semi- and high-skilled skilled labor in key value chains among SMEs and self-employed) exacerbate inequity between private and public sector • Introduce more competitive wage-setting and contracting • Low-skilled labor comprising 70 • Lagging implementation and targeting of new, post- regulations for private sector within 2013 Social Contract framework percent of unemployed 2013 consolidated active labor market programs based on rigorous evaluation of potential impacts • High informality at 30-45 percent (ALMP) managed by MFPE (cheque d’emploi, cheque • Expand performance-based contracting (PBC) with firms and private d’employabilité) and civil society providers for design and delivery of ALMPs based on • Low quality of and weak transition to work particularly prior pilot projects involving vocational skills for vocational trades SOCIAL SECURITY • High tax wedge at 30 percent • Fragmented administration and multiplicity of regimes • Link benefits to financing more explicitly and increase transparency impinging on benefit level, job (CNSS) of redistributive mechanisms creation, and equity • Lack of portability and transparent financing • Consolidate CNSS regimes and enhance portability • Lack of unemployment benefits mechanisms across regimes • Design and implement well-defined, well-monitored UB for job loss (UB) for income protection and • Lack of harmonized social benefits across public and (pilot followed by evaluation and roll-out) access to services private sectors • Consolidate CNAM and AMG and develop explicit benefits • Increasing deficits at 2 percent of package/subsidy policies for the poor GDP by 2018 and inequity within • Narrow gap between key benefits in private and public sectors CNSS, CNRPS, CNAM, and AMG (open-ended vs. various fixed contracts, wages, pension benefits) SOCIAL SAFETY NETS • Weak targeting to the poorest • Vague categorical targeting criteria, poorly monitored • Design, pilot, evaluate, and implement new targeting protocol (only 40 percent of benefits implementation, and weak capacity at local level (under preparation since 2014) in coordination with other ministries accruing to poorest 20 percent) and local civil society • Lack of policies for exit, • Lack of clear terms, conditions, and graduation • Introduce recertification, exit, and mechanisms to promote access to recertification, and graduation mechanisms linked to local labor market primary health and education and water services • Switch payment modality from post to e-banking options to improve financial inclusion, savings, and graduation SOCIAL PROTECTION AND LABOR DELIVERY SYSTEMS • Lack of policy coherency (benefit • Develop unified social protection and labor benefits • Institute social dialogue council law (under preparation as of 2014) levels, eligibility criteria, and financing reform and unified social protection and labor policy strategy financing sources, incentives) • Implement unified identification and management • Consolidate institutional mandates across MFPE, MOSA and regional • Lack of interconnected databases information system development funds between MOSA and ANETI at • Roll-out unique identification and management information system central and local levels at central and regional levels (under development since 2014) Source: Compiled by World Bank staff and Tunisian administration. Over the mid-term, strengthening institutional capacity and governance would help lay the foundation for aligning social inclusion and job growth. ALMPs, pensions, and social safety nets as currently designed and implemented are not reaching vulnerable groups in part due to institutional fragmentation in terms of identification and eligibility. Financial inclusion and access to social protection services remains weak most notably in lagging regions, where the delivery of services is hampered by inadequate investments in skilled human resources and equipment. ALMPs are also not adequately capturing and targeting firms’ skills needs in key sectors and regions. Tunisia has started to create a unique identification system and introduce interoperability between information systems for key social protection and labor programs as of 2013. Stronger institutional capacity and governance within the system in Tunisia can be attained by:  Rolling out unified social registry information, targeting, and unique identification systems;  Consolidating the pension administration (CNSS and CNRPS) as well as health insurance mechanisms (CNAM and AMG); and  Strengthening the social dialogue policy framework and evidence base for developing public- private partnerships for ALMPs, reform labor contracts, and redefining wage-setting mechanisms that more actively promote productivity. Over the long term, although improving the financial sustainability of the overall system may take time, building the evidence base and consensus for transformative reforms can be achieved in the meantime. A high tax wedge and pending pension deficits impinges on the quality of jobs and protection from shocks, and consumption subsidies crowd out spending on other vital investments that could spur revenue generation. Pressures on fiscal space could be supported in the medium to long term by redirecting savings from reforming inefficient consumption subsidies and in the long term through comprehensive pension reform. Savings could also be used to enhance mechanisms that promote access to services and coverage of social safety nets. Tunisia can therefore improve the financial sustainability of its system by:  Developing supplemental resource generation models and greater portability for pension benefits (CNSS and CNRPS) that explicitly tie benefits to contributions and identify subsidized pension benefits;  Aligning investment code and labor regulations with targeted schemes for unemployment benefits to improve the quality of jobs among the vulnerable self-employed and small and medium enterprises (SMEs) in lagging regions; and  Reinvesting universal subsidies into public or private investments that stimulate job growth and improve the effectiveness of targeted social protection programs. Conclusion: Building Systems, Connecting to Jobs Looking ahead, Tunisia’s forthcoming National Development Plan for 2016-2021 provides an important platform for financing a unified social protection and labor reform agenda. Overall, consolidation in Tunisia entails enhancing delivery systems over the short term and realigning policies on benefits, eligibility, and financing schemes over the medium to long term. Close involvement of the private sector and civil society at the regional level will be critical to identifying and targeting only the most viable opportunities for long-term inclusion and job growth. Ultimately, modernizing its social protection and labor system will help Tunisia better cope with new shocks and, importantly, reap the social benefits of its hard-fought political transition. 45 KEY REFERENCES World Bank (2012). Tunisia Pensions Study. Background Paper. Washington DC: World Bank. World Bank (2013). Consolidation and Transparency: Transforming Tunisia’s Health Care for the Poor. Washington DC: World Bank. World Bank (2013). Vers une meilleure équité: les subventions énergétiques, le ciblage et la protection sociale en Tunisie (Towards greater equity: energy subsidies, targeting and social protection in Tunisia). Washington DC: World Bank. World Bank (2013). Tunisia Social Protection Reforms Support Project, Project Appraisal Document (P144674). Washington DC: World Bank. World Bank (2014). Labor Policy to Promote Good Jobs in Tunisia: Revisiting Labor Regulation, Social Security, and Active Labor Market Programs. Directions in Development. Washington DC: World Bank. World Bank (2014). The Unfinished Revolution: Bringing Opportunity, Good Jobs and Greater Wealth to All Tunisians. Development Policy Review. Washington DC: World Bank. World Bank (2014). Tunisia Country Program Evaluation FY05-FY13. Washington DC: World Bank. World Bank (2015). Balancing Regulations to Promote Jobs: From Employment Contracts to Unemployment Benefits. Washington DC: World Bank. World Bank (2015). Implementation Completion and Results Report for Tunisia State and Peacebuilding Fund (SPF) Participatory Service Delivery for Reintegration Project. Report No: ICR00003568. Washington DC: World Bank. World Bank (2015). Overview of Financial Inclusion in Tunisia. Global Findex Database. World Bank: Washington DC. World Bank (2015). Raising the Bar: Institutionalizing Participatory Monitoring of Service Delivery in Tunisia. Background Paper. Washington DC: World Bank. World Bank (2015). Towards a Jobs Strategy in Tunisia, in Tunisia Systematic Country Diagnostic. Washington DC: World Bank. World Bank (2015). Tunisia Systematic Country Diagnostic. Washington DC: World Bank 46