35631 World Bank Social Safety Nets Primer Notes 2006 No. 25 Social Safety Nets in OECD Countries The focus of the note is on non-contributory social Level of SP and SSN Spending programs for low-income households or other vul- The OECD countries redistribute a large share of nerable groups in OECD countries. These programs, their GDP through social protection (SP) programs, typically referred to as social safety net (SSN) pro- about 19% of GDP in the European Union, and 9% grams in developing countries, are labeled welfare of GDP in the US. On average, 85% of this spend- programs in the US and social assistance programs ing is associated with social insurance programs: in the European Union. This note covers 28 countries pensions for old-age, disability or dependents, and belonging to the OECD, and refers to an in depth contingency for temporary loss of work due to un- reviewofSSNprogramsintheUSandnineEuropean employment, illness or maternity. Non-contributory Union countries prepared for a course on "Social social assistance programs account for 15% of the Safety Nets in OECD Countries." The accompanying total ­ with high variance across countries. In the course materials have been developed by a team EU-15, spending on social assistance programs aver- from the Urban Institute (for the US) and the Uni- ages 3% of GDP. In the US, means-tested welfare versity of Maastricht (for nine European Union coun- programs account for 2.2% of GDP; adding the cost tries). The material on US welfare policies also draws associated with providing health insurance for the on Lindert (2005), and the review of reforms in OECD low income households ­ Medicaid ­the total bill of countries from Abt (2003). the welfare programs goes up to 4.4% (in 2000, see Lindert, 2005). Table 1. Social Protection Spending in OECD Countries, 2001 % of GDP All Old Age Survivors Disability Active LM Unemploy- Family Housing Others ment Australia 10.1 4.1 0.2 2.2 0.1 1.0 2.4 0.1 - Austria 19.1 10.3 2.7 2.3 0.1 0.8 2.4 0.1 0.4 Belgium 18.5 8.6 2.6 2.2 0.7 2.2 1.9 - 0.3 Canada 8.4 4.8 0.4 0.8 0.4 0.8 0.9 - 0.3 Czech 12.5 6.7 0.9 3.0 0.1 0.2 1.5 0.1 - Republic Denmark 15.4 6.5 - 2.7 0.2 3.0 1.5 0.7 0.8 Finland 15.6 7.1 0.9 3.1 0.3 2.0 1.7 0.3 0.3 France 17.4 10.4 1.5 1.7 0.4 1.6 1.5 - 0.4 Germany 15.9 10.8 0.4 1.6 0.3 1.2 1.1 - 0.5 Greece 16.5 12.6 0.8 1.6 - 0.4 1.1 - - Hungary 13.3 7.4 0.3 2.5 0.3 0.4 1.9 0.5 0.1 Iceland 8.4 4.1 0.6 1.8 - 0.2 1.2 0.1 0.4 Ireland 7.9 2.4 0.8 1.3 0.4 0.7 1.4 0.5 0.5 Italy 18.7 12.6 2.6 2.0 0.2 0.6 0.6 - - Japan 9.1 6.4 1.2 0.6 0.1 0.5 0.3 - 0.2 Korea 2.4 1.1 0.2 0.3 0.1 0.2 - - 0.5 EmilTesliuc prepared this note as part of the materials used in theWorld Bank training course,"How the Rich ProtectTheir Poor: Social Safety Nets in OECD Countries." All Old Age Survivors Disability Active LM Unemploy- Family Housing Others ment Nether- 13.7 5.7 0.7 4.0 0.4 1.3 0.7 0.4 0.6 lands New Zea- 11.6 4.7 0.1 2.8 0.1 1.1 2.1 0.6 0.1 land Norway 11.6 4.6 0.3 3.9 - 0.4 1.9 0.2 0.3 Poland 18.0 8.5 2.1 5.4 - 1.0 0.9 - - Portugal 13.3 7.6 1.5 2.4 0.1 0.9 0.7 - 0.2 Slovak 12.1 6.5 0.2 2.1 0.2 0.5 1.4 0.1 1.1 Republic Spain 13.2 8.1 0.6 2.3 0.4 1.3 0.3 0.2 - Sweden 14.6 6.8 0.6 3.8 0.2 1.0 1.8 - 0.4 Switzer- 18.3 11.6 1.6 3.0 0.1 0.5 1.1 - 0.5 land United 13.7 7.7 0.6 2.2 - 0.3 1.5 1.5 - Kingdom United 8.0 5.2 0.8 1.1 0.2 0.3 0.1 - 0.3 States Source: OECD (2004) Benefits andWages The Architecture of SSN Systems of the social insurance system in continental Europe, Spending on social protection programs is larger, relative to Anglo-Saxon countries, transformed SSN in both absolute and relative terms, in the OECD interventions into residual programs, less important compared to developing countries. Within OECD in terms of the volume of spending or number of countries, however, there is considerable heteroge- beneficiaries, but very important to ensure that neity in terms of the level and composition of the the net is tightly woven for the poor or vulnerable social protection spending, and the role of the social households. safety nets. In spite of this heterogeneity, one can distinguish two models: Whatfactorsseemtoexplainthedivergentpaths followedbyAnglo-Saxonvs.Europeancountries The SSN-dominated, or Anglo-Saxon model, in terms of social protection spending? found in the US, UK, Australia, New Zealand, and AlesinaandGlaeser(2004)lookatthedifferencesbe- Canada. In these countries, there is less reliance tween the size of the welfare state (publicly financed on social insurance instruments (pensions, health, education, health and social protection) in the US unemployment insurance), and greater use of tar- and the EU, and investigate a wide range of factors geted support to deserving poor. that might be associated with the different levels of spending. Surprisingly, their research finds no cor- TheresidualSSN-model,foundespeciallyincon- relation between economic factors postulated as tinental Europe. Specific for continental Europe is important by different theories, such as differences the substantial role played by social insurance in- in pre-tax inequality; efficiency of the tax system, struments. In this second model, the net is woven social mobility, and the propensity to spend on the based on universal/generous family benefits and welfare state. Much of the gap in spending seems residual means-tested programs for the poor (the to be explained by differences in political institu- net of last resort). tions (type of political representation, federalism vs. national states, the system of checks and balances), While the relative share of non-contributory spend- racial fragmentation, or beliefs about the nature of ingisgreaterinAnglo-Saxoncountries,thecoverage poverty (laziness vs. lack of opportunity). and the extent of redistribution through social pro- tection programs are greater in continental Europe. Type of Programs Higher spending levels and coverage translate in For the program practitioner concerned about a larger reduction in poverty and inequality in conti- particular type of program or policy, OECD coutries nental Europe compared to Anglo-Saxon countries. provide a large number of examples or potential ThesizeandroleoftheSSNprogramsdependsheav- models(Table2).Whileprogramrulesdiffersubstan- ily on the other components of the social protection tially from country to country, most OECD countries system; to some extent, there is some degree of sub- operate the following types of social transfers: stitutability between pensions, unemployment ben- efits and social assistance programs. The expansion Minimum income programs: government pro- winning, care and other domestic responsibilities vided minimum income financial assistance for and a lower overall earnings potential. They take people without resources. Typically, these pro- the form of cash benefits or tax reductions. grams top up family or individual income up to a certainthresholdsetnationallyorregionally.These Employment-conditional benefits for able-bod- last-resort safety nets can have a major impact ied individuals: In an attempt to ensure that at on both the extent and the intensity of financial least some incentive to work is maintained, many poverty. At the same time, the conditionality on OECD countries allow benefit recipients to work resources, including employment income, reduces a certain number of working hours without stop- any short-term gains of attempts to escape pov- ping eligibility or taxing the full amount of the erty by pursuing other income sources. income earned while re-entering employment. These programs take the form of cash benefits, tax Housing benefits: support for housing-related credits or reduced rates on social contributions. costs of various forms: cash benefits, other finan- cialassistance(e.g.low-interestloans)andbenefits Childcare benefits: financial support for families in-kind (e.g. subsidized housing). with small children requiring care is available in a multitude of different forms: direct cash benefits, Familybenefits(universalormeans-tested):ben- tax breaks, or subsidies paid to the providing in- efits directed towards families with children, with stitutions. Support may be available to all children the existence of a family being the main eligibility of a certain age, or may be conditional on having criterion. children in certain types of childcare such as those provided by approved institutions or specially Benefits for lone-parents attempt to mitigate qualified individuals. the high poverty risk for lone-parent households, triggeredbyreducedopportunitiestosharebread- Table 2. Non-contributory SocialTransfer Programs Guranteed Housing Family Benefits Lone-parent Employ- Childcare Benefits Minimum Benefits Benefit ment Income conditional Benefits Universal Means- Non-paren- Parental tested tal Care Care Australia Y Y - Y Y Y Y Y Austria Y GMI Y - - - - Y Belgium Y - Y - - Y Y - Canada Y GMI - Y Y Y Y - Czech Re- Y Y - Y - - - Y public Denmark Y Y Y - FB - Y - Finland Y Y Y - FB Y - Y France Y Y Y Y Y Y Y Y Germany Y Y Y - T Y Y Y Greece - Y Y - - - - - Hungary Y Y Y - FB - - Y Iceland Y Y Y Y Y - - - Ireland Y GMI Y - Y Y - Y Italy - Y - Y - - - - Japan Y GMI - Y Y Y - - Korea Y GMI - - Y - Y - Luxem- Y GMI Y - T - Y Y bourg Netherlands Y Y Y - T Y Y - New Zea- - Y - Y Y Y - - land Guranteed Housing Family Benefits Lone-parent Employ- Childcare Benefits Minimum Benefits Benefit ment Income conditional Benefits Universal Means- Non-paren- Parental tested tal Care Care Norway Y Y Y - Y - Y - Poland Y Y - Y CCB - - Y Portugal Y - - Y T - - - Slovak Y Y Y Y - - - Y Republic Spain Y - - Y T - - - Sweden Y Y Y - Y - - - Switzerland Y GMI Y - - - - - United Y Y Y - - Y Y - Kingdom United - Y - Y - Y Y - States Notes:"Y"indicates that the specific benefit or tax credit exists in this country.Where no specific housing or lone-parent benefit is available,"GMI"(guaranteed minimum income),"FB"(family benefit) or"CCB"(childcare benefit) indicate that housing or lone-par- ent specific provisions exist as part of these schemes."T"indicates different tax provisions or specific tax allowances for lone parents where no other benefits are available. Source: OECD (2004) Benefits andWages For each type of program, OECD countries exhibit of determination of the qualifying income threshold strong variation in terms of design features. Here, (national vs. regional), generosity of the system (the we illustrate this heterogeneity for one type of pro- level of benefit per member), and elements taken in gram, guaranteed minimum income. As apparent in consideration when administering the means-test. Table 3, GMI programs differ substantially in terms Table 3. Key Design Features of GMI Programs Maximum amount, % APW Threshold Features of the means test Household Spouse Children Income Benefit Benefits head disregards withdrawal excluded rate Australia - - - - - - - Austria 19 12 6 N - 1 Family Belgium 23 8 4-10 N Y 1 Family Canada 16 12 4-5 R - 0.75 - Czech Repub- 24 17 13-17 N - - - lic Denmark 31 31 10 N Y 1 - Finland 16 11 9-12 N Y 1 - France 20 9 0-9 N Y 1 Family/Hous- ing Germany 11 9 5-10 N Y 1 Family Greece - - - - - - - Hungary 16 - - NG - 1 - Iceland 31 25 - R - 1 Family/Hous- ing Ireland 24 16 3 NG - 1 Family Italy 16 9 7-8 R Y 1 - Japan 24 13 6-7 R Y 1 - Korea 16 10 9-10 N Y 1 - Maximum amount, % APW Threshold Features of the means test Household Spouse Children Income Benefit Benefits head disregards withdrawal excluded rate Luxembourg 36 18 3 N Y 1 Family Netherlands 30 13 - N - 1 Family/Hous- ing New Zealand - - - - - - - Norway 30 7 4-16 NG - 1 - Poland 21 - - N - 1 - Portugal 20 20 10 N Y 1 Family Hous- ing Slovak Re- 17 12 8 N - 1 - public Spain 27 3 3 R - 1 Family Sweden 16 11 7-12 NG - 1 - Switzerland 22 12 6 NG - 1 - United King- 14 8 10 N Y 1 Housing dom United States 5 4 4 N Y 1 EITC Note: Average Productive Wage (APW) is the average wage of a blue-collar worker in the manufacturing sector in each country. The income threshold can be determined nationally (N), under national guidelines (NG), or regionally (R). Source: OECD (2004) Benefits andWages Factors thatTriggered the Need for SSN Reform growth of single parent families and single youth VariouspressuresacrossOECDcountrieshavedriven families, have led to increases in the dependency social protection policy reform in a few common on social assistance. directions. Traditionally, social safety nets were de- signed to protect against old age, short-term (or a Funding and delivery tensions between central low level of long-term) unemployment, disability andlocalauthorities. Invariouscountries,regional and low income. Recently, however, several over- governments are demanding more authority over arching factors have substantially affected social the financing and administration of social protec- protection policy regimes, including: tion benefits. This has created problems in many areas where the regions with the strongest tax Increasing long-term unemployment, partially bases have the least need for social protection ser- as a result of severe labor market change. The vices,leavingpoorerregionswithgreaterfiscaldif- globalization of production caused a reduction ficulties to cover the needs of their populations. in the amount of unskilled jobs within the OECD. Consequently, much of the population adversely These changes do not affect only developed coun- affected by these job losses have come to rely on tries. Aging and the dissolution of traditional kinds safety nets. Unlike previous economic downturns, of family support system are increasingly affecting however, this structural change in the economy the middle-income countries and, to a lower extent, has led to longer-term claims on unemployment low-income countries. In confronting these issues, benefits. Time limits on social insurance in many developing countries can learn from the successful countries have led to greater dependence on so- experiences in OECD countries. cial assistance. Policy Issues on the Social Safety Net Agenda in Population aging in many countries has led to OECD Countries increasing dependency on not only social insur- The aforementioned pressures have bread a wide ance, but targeted social assistance benefits for realm of debates among policymakers and stake- the low-income elderly. However, the greatest holders about how social protection policy might challenges in terms of aging remain in the future, be reformed in response. The most common discus- as OECD populations continue to live longer and sions across OECD countries have involved the two birth rates drop. following issues: Changing family structures, particularly the work incentives, and their relation to social protection reform, and demographic and family single parents are not required to work in order to structure shifts. receive benefits until their children have reached a certain age. In the certain U.S. states, General social exclusion or "new poverty," as a result of Assistance, the cash benefit of last resort, is simply changing conditions on the labor market, and de- notofferedtoable-bodiedsingleadultsorcouples mographic and family structure shifts. without children, under the assumption that they should be active in the labor market. Work Incentives Increasing unemployment has led to a discussion Levels of benefits: some countries offer low cash of the role of social assistance--primarily for the benefits, such as Texas (US) and Greece; other low able bodied unemployed--in aggravating jobless- cash with additional tied benefits, such as the UK; ness through the creation of disincentives to work. and others, such as Australia, offer relatively high Among the several issues spurring social policy re- cash benefits. form, this debate has been the most prominent, by far.Policymakersinmanycountrieshavevoicedcon- Tax structures: beneficiaries in many countries cerns about the theoretical "welfare trap" whereby who re-enter the workforce find that their "effec- recipients of SA benefits find it in their interest not tive marginal tax rate"--the amount they lose in to enter the labor market, increasing pressure on benefits upon labor market re-entry--can be very the social protection system, while reducing overall high in comparison to their new earnings. productivity. Differences between social insurance and social In theory, the greater the benefits provided to the assistance: some OECD countries, such as the US, unemployed person, the lower their incentive to make the differences of social insurance and social (re)enter the workforce. As such, debates have con- assistance very clear, so as to incentivize use of cerned the levels, duration, and requirements of the SI system, and, consequently, the active labor benefitstotheunemployed,aswellasotherfeatures market. Benefits such as Unemployment Insur- of the system that might contribute to long-term ance (UI) and Disability insurance (SSDIB) are paid unemployment. At the same time, however, these out at much higher rates than social assistance debateshaveoccurredwiththechangedcontextsof benefits such as Temporary Assistance for Needy demography and family structure. Increasing single Families (TANF) Supplemental Security Income for parent and single youth households have pushed the Disabled (SSI). Cultural forces have also distin- issues of childcare support and incentives to marry guished between the two types of benefits, and a into the policy debate around work incentives. significantstigmaisattachedtobeingareceiverof SA benefits. Australia, on the other hand, only of- Major issues in the debate about work incentives ferssocialassistancebenefits.Assuch,thereareno include: advantages to being "paid in" in terms of benefit levels, nor are any social stigma forces at work. Duration of benefits: social insurance benefits often have time limits, after which they may drop Availability of training opportunities: some to lower, social assistance levels. Until recently, so- countries, such as Sweden, offer intensive job cialassistancelevelswerealmostalwaysindefinite. training and insertion programs to the unem- Traditionally, however, a few OECD countries have ployed. Other countries are more lenient in their had time limits on SA benefits for young people, labor market interventions. Arguments in favor of including Austria, Denmark, and Italy. reforming social assistance benefits are often ac- companied by new training opportunities, most of Conditions for forfeit of benefits: rules by which which allow the beneficiary to continue receiving recipients lose their benefits vary widely. Some benefits. countrieshavetraditionallyallowedforsomemem- bers of the household to work, providing benefits It is important to note that these issues, which have for other individuals, rather than the household been put forth as evidence of disincentives to work, unit. Other countries allow recipients to perform are largely untested. One major study did find a rela- part-timeworkwithoutlosingtheirclaims. Others. tionship between increased duration of benefits and However, do not allow for any part-time work to long-term unemployment. Nevertheless, some con- accompany benefit receipt. sensushasemergedwithinandacrosscountriesthat reform along these lines would ease dependency on Job search requirements: most countries implic- social assistance, and push more benefit recipients itlyassumethatbenefitrecipientsareactivelylook- into the active labor market. ing for work, whereas other employ stringent tests todeterminejobsearchactivity.Inafewcountries, Policy Responses SurroundingWork Incentives Sweden, cutting the most. However, the relative In response to increasing concerns about work in- generosity of countries compared to one another centives, many OECD countries have reformed their has changed little. Almost uniformly, countries social protection systems, and social assistance, in have attempted to reduce benefit replacement ra- particular, in order to develop more efficient work tios(benefits/earnings*100)inanattempttomake incentives. These reforms have generally taken the the earnings benefits of work more salient. form of positive support to encourage labor market participation, or cuts in the existing support system, Reduced duration of benefits: various countries in order to discourage use of and prevent long-term have reduced the duration of benefits to recipi- dependence on social assistance. Positive supports ents. In most cases, this has involved only social to work include: insurance, which end sooner, dropping the re- cipientsbenefitstoSAlevels. TheUS,however,has Extending possibilities for part-time work: oc- placed strict time limits on even SA benefits. casional work is recognized as a gateway into full- time jobs. By not canceling benefits for recipients More rigorous tests of active work-seeking: Fin- whoenterpart-timework,manycountrieshopeto land and the Netherlands, among other countries, ease the transition into the labor market. Canada, have added stricter enforcement mechanisms to the UK, and the US state of Minnesota are among their requirements that benefit recipients be look- those who have instituted this reform. ing for work. Changing benefit delivery from the household Social Exclusion to individuals: a policy whose intentions are simi- SomeOECDcountries,particularlythoseincontinen- lartoallowingforpart-timework. Incountriesthat tal Europe, have raised the issue of"social exclusion" deliverbenefitstothehousehold,individualmem- in their approach to policy reform. Social exclusion bers can be discouraged from working for fear of has been defined as"a multi-dimensional socio-eco- the risks to the rest of their household. Countries nomic conception of deprivation," as compared to suchasAustraliaandCanadanowallowotherindi- poverty, "a lack of adequate income or insufficient viduals in the household to receive benefits while material resources to satisfy basic needs." Using so- only the worker loses hers. cialexclusionratherthanpovertyasabasisfordefin- ingneedusuallymeanstargetingmorerecipientsfor Employment conditional benefits: in a direct social assistance, as means-testing shifts from only attempt to incentivize work, many countries have income and material resources to the "access to full offered various benefits to those in the labor participation in the community." market. These policies take the form of one-time cash bonuses (Ireland, UK); or wage supplements ThereexistsageneralconsensusacrosstheOECDthat to low earners (US, UK), which can take the form socialcohesionisnecessaryforeconomicgrowth. As of regular payments or annual tax credits; or such, policymakers have made the concept of social temporary terms of complete benefit receipt dur- exclusion integral to their social assistance policy re- ing full-time work (Canada). Some countries have forms. The debate around the issue has centered on instituted these programs for the first-time, while three issues related to social assistance policy: others have eased eligibility levels or extended their benefit levels. In addition those previously The relative nature of social exclusion: unlike mentioned, Australia, Italy and New Zealand have poverty, which is traditionally defined as abso- also instituted these programs, each in the form of lute--one is either poor or not poor--social ex- tax credits. clusion is a relative condition, which takes into accountthestandardoflivingoftheentiresociety. Morehelpinthejobsearch:theUK,Switzerland, Asaresult,somehavearguedthatsocialassistance and the US have reformed their systems to offer benefits need to place less emphasis on absolute more effective assistance to job-seekers. Such poverty lines, and more on social integration. interventions have included making labor market information more available or improving job train- The mechanisms that produce exclusion: ex- ing. clusion may involve the curtailment of rights or differential treatment of certain individuals by the Methods to discourage social assistance depen- range of social institutions within a given commu- dency include: nity. Some policymakers have argued that overly generous social assistance can produce exclusion Reduced benefit amounts: as a result of wide- through the development of an"underclass"men- spreadfiscalpressures,almostallcountrieshavere- tality, with little emphasis on the personal respon- duced their benefit levels, with Denmark, Finland, sibility to find work. Others have used this concept to argue that social assistance needs to be more generous so as to provide standards of living that allow individuals to more fully participate in the community. Thedynamicsofexclusion:debatehassurround- ed the trends that provoke exclusion in the future, which can include unemployment, parenthood and family structure, or other circumstances. These debates have been fueled by an increasing public policy emphasis on the importance of social capital. As policymakers focus on the value of social networks and relationships, they seek to identify the phenomena that separate social groups and de- crease these endogenous sources of development. Policy Responses Surrounding Social Exclusion There exists considerable debate over what consti- tutes satisfactory social inclusion. Some countries believe that full-time paid employment is enough, whereas others are more concerned with covering a range of needs, including adequate housing, health- care and access to social institutions. Strategies for reducing social exclusion include many of the"posi- tive" policies discussed in the previous section on work incentives, and combine support to low-wage workers, such as employment conditional benefits, and active labor market interventions, such as train- ing programs and insertion schemes. TheWorld Bank Social Safety Nets Primer series is intended to provide a practical resource for those engaged in the design and implementation of safety net programs around the world. Readers will find information on good practices for a variety of types of interventions, country contexts, themes and target groups, as well as current thinking on the role of social safety nets in the broader development agenda. World Bank, Human Development Network Social Protection, Social Safety Nets http://www.worldbank.org/safetynets