Social Safety Nets in the Western Balkans: Design, Implementation and Performance ABCDE Albania Conference June 2010 Boryana Gotcheva and Ramya Sundaram World Bank, Europe Central Asia Region Social Protection Team
Outline Economic and Policy Context. Scope and objectives of social assistance programs. Spending: what do countries spend on social assistance. Performance: targeting accuracy, coverage and generosity. Main recommendations and conclusions: implications for safety nets in the region. 2
Immediate crisis context Countries in the Western Balkans entered the crisis with safety nets in place. Expenditures on social assistance were protected, in the face of significant cuts in other public expenditures. Safety nets continued to protect those already receiving benefits. However, the automatic stabilizer role did not materialize: the number of beneficiaries has increased little, and only with significant lag. 3
Medium term challenges Households cope with the crisis by reducing essential expenditure Demographic: aging population, falling birth rates Negative long-term human development consequences Higher pension spending, working age population declines in the future High, persistent unemployment and informal employment Low contribution to social insurance; pressure for more spending on social assistance, particularly challenging for Albania The Western Balkan countries are also faced with the challenge of adapting their safety nets to EU country models. No acquis communautaire in social assistance 4
Framework for analysis of social assistance Social Protection Social Insurance Labor Market Social Assistance 5
Objectives and structure of social All non-contributory public transfers: assistance Last resort social assistance: protect poor and vulnerable. Family and child allowances: protect children, boost fertility, protect incomes and jobs. Disability allowances: income replacement for unable to work. War veterans benefits: protection of disabled during regional military conflicts; survivors and in some cases rewarding war veterans (region specific objective) 6
Wide variation in spending on social protection (pensions); relatively low spending on social assistance Social Protection Spending (% GDP) Kosovo 08 Georgia 07 Tajikistan 08 Armenia 08 Turkey 07 Azerbaijan 08 Kyrgyzstan 08 Albania 08 Estonia 06 Romania 07 Latvia 08 Russia 06 FYR Macedona 08 Moldova 08 Montenegro 08 Lithuania 08 Bulgaria 08 Serbia 08 Croatia 08 Ukraine 08 Social Assistance Labor Market Social Insurance 0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 7
Bosnia 08 Serbia 08 Albania 08 Kosovo 08 Montenegro 08 FYR Macedona 08 Croatia 08 Belarus 08 Ukraine 08 Russia 06 Romania 07 Estonia 06 Lithuania 08 Poland 07 Moldova 08 Bulgaria 08 Kyrgyzstan 08 Georgia 07 Armenia 08 Latvia 08 Azerbaijan 08 Turkey 07 Spending structure differs from other ECA: higher on disability (Albania) and war veterans 4,0% Social Pension War Veteran Benefits Disability Benefits Family and Child Allowance Last-Resort Social Assistance 3,0% 2,0% 1,0% 0,0% Western Balkans Other ECA 8
Very diverse ratios between spending on categorical and on means-tested social benefits Means-tested Categorical 3,50 3,00 2,50 2,00 2,82 1,50 1,00 0,22 0,55 0,75 1,29 1,44 0,50 0,00 0,75 FYR Macedonia 0,70 0,51 0,33 0,42 0,52 Montenegro Kosovo Albania Serbia BiH 9
Expenditure on Last-resort Social Assistance has typically declined 2,00% 1,80% 2000 1,60% 1,40% 1,20% 1,00% 2004 2008 0,80% 0,60% 2000 2003 2008 2008 0,40% 0,20% 0,00% 2008 2005 2008 2008 2000 2008 2004 2009 2008 2000 2007 2000 2008 Albania Bosnia and Herzegovina FYR Macedonia Kosovo Montenegro Serbia Croatia Georgia Armenia Latvia Western Balkans Comparators 10
Performance of social assistance We use household surveys to assess Coverage: percent of poorest quintile who receive benefits. Targeting accuracy: percent of benefits going to the poorest quintile. Generosity (Adequacy): average transfer amount as a fraction of average consumption for beneficiary households in poorest quintile. Unit transfers as a fraction of minimum wage 11
Armenia FB Prog Kosovo SA *Russia CA FYR Macedonia SFA *Kyrgyzstan UMB Albania NE Georgia TSA Poland SW benefits *Azerbaijan TSA Romania GMI Montenegro FMS/MOP Bulgaria GMI Croatia S. Allowance Ukraine XP program Lithuania S. Benefit Hungary Regular SA Estonia MT Benefits Serbia MOP Latvia GMI + dwelling BiH CSW *Uzbekistan SA for low Kazakhstan TSA Coverage of last-resort social assistance 45 40 35 30 25 20 15 10 5 0 Coverage of the Poorest Quintile (%) 12
Romania GMI Ukraine XP program Bulgaria GMI Serbia MOP Kosovo SA Croatia S. Allowance Estonia MT Benefits Kazakhstan TSA Lithuania S. Benefit Hungary Regular SA Georgia TSA Albania NE *Azerbaijan TSA Armenia FB Prog Montenegro FMS/MOP Poland SW benefits FYR Macedonia SFA *Kyrgyzstan UMB BiH CSW *Uzbekistan SA for low Latvia GMI + dwelling *Russia CA Targeting accuracy of last-resort social assistance 90 Percent of Total Benefits Received by the Poorest Quintile 80 70 60 50 40 30 20 10 0 13
*Russia CA FYR Macedonia SFA BiH CSW Montenegro Latvia GMI + dwelling Albania NE Kazakhstan TSA Georgia TSA Poland SW benefits Lithuania S. Benefit Bulgaria GMI Romania GMI Hungary Regular SA Armenia FB Prog Estonia MT Benefits Croatia S. Allowance Kosovo SA *Kyrgyzstan UMB Serbia MOP Ukraine XP program Elite capture: last-resort social assistance 18 16 14 12 10 8 6 4 2 0 LRSA % of Total Benefits Received by the Richest Quintile 14
Generosity of last-resort social assistance Last-resort social assistance Average Transfer Value per capita (Beneficiary Households Only), % of minimum wage Total Poorest Quintile Albania NE 4.4 4.1 Bosnia-Herzegovina CSW 13.1 14.7 Kosovo SA 7.2 7.2 FYR Macedonia SFA na na Montenegro FMS/MOP 45.4 43.5 Serbia MOP 12.6 12.9 Serbia CA 5.3 5.6
What Are the Implications for Social Policy in the Western Balkans?
Design and implementation features contribute to mixed outcomes of last-resort social assistance Low coverage, large exclusion due to Strict rules of eligibility low thresholds for access presence of exclusionary filters Rigorous enforcement of these rules (verification of self-reported income, recertification, audits) Little pro-active outreach to deserving poor 17
Design and implementation features contribute to mixed outcomes of last-resort social assistance Transfer amounts of last-resort programs are fairly small Equivalence scales (rate of increase in benefits) are not in favor of large families Ceilings on number of eligible recipients in one family Irregular indexation of eligibility thresholds and benefits levels 18
Recommendations for improving equity, efficiency, and transparency of last-resort programs Programs build on the existing foundation Strengthen and standardize eligibility criteria using a continuous scoring formula instead of exclusionary filters Introduce regular indexation of benefits and eligibility thresholds to price inflation Strengthen benefits administration (registries) 19
Recommendations for improving equity, efficiency, and transparency of last resort programs Improve implementation by Reduce number of documents required, improve outreach, and improve screening of welfare characteristics during home visits Improve the links between the last-resort and other programs and services Social care services Activation services and other support services, connecting the poor to job pools, and removing other obstacles to work 20
Thank you! 21