Privatizing Social Security in Latin America Indermit Gill December 9, 2004 Based on Keeping the Promise of Social Security in Latin America By Indermit S. Gill, Truman Packard and Juan Yermo, with Todd Pugatch Note: Note: The The views views expressed expressed here here are are those those of of Indermit Indermit Gill Gill and and should should not not be be attributed attributed to to the the World World Bank. Bank.
Three propositions Reforms often driven by short term concerns, not to address long-term structural changes Not without longer term benefits: individualization But this feature has not been the headline item Reforms often driven by loss of faith in government, not reflection on its proper role Instrument for reducing reliance on govt.: privatization But dependence on government remains heavy Reforms often driven by narrow interests, not by broad consensus on social security System was supposed to be popular: participation But coverage of social security remains low 12/16/2004 Presentation for AARP 2
Outline of Discussion The Reforms What and how The Backdrop Demographics and poverty The Analysis Theory and evidence The Implications Policy and politics 12/16/2004 Presentation for AARP 3
The Reforms What and How?
The What? Structural reforms in 12 countries in Latin America Chile in 1980s, 7 in the 1990s, 4 more since 2000 Notable exceptions: Brazil and Venezuela Multiple tiers instead of reliance on single system Phase out or scale down old PAYG tier, replacing it with a minimal pension Introduce a fully-funded tier that is privately managed Consider incentives for voluntary retirement savings New approach popular in the region and outside Briefly on the menu even in Brazil in 1997 Popular in Eastern Europe and Central Asia US reform debate in 2001, perhaps again in 200 12/16/2004 Presentation for AARP 5
One of the common features of reforms is a privately managed tier. C h ile P eru C o lom b ia A rgen tina U ruguay M exico B o livia E l S alvad or Year of reform 1981 1992/1993 1994 1994 1996 1997 1997 1998 Contributionrelated PAYGO system? Total payroll tax rate, prereform (% ) Total payroll tax rate, postreform (% ) Participation of new workers? Participation of self employed? Fees & insurance premia (% of wage) Pay-out options Minimum return on investment? Minimum contributory pension? Social assistance pension? closed remains remains remains remains closed closed closed 33 18 17.8 42 40 20 19 11.8 20 20.5/22 a 33.8 46 b 40 26 24 13.5 mandatory voluntary voluntary voluntary c voluntary d mandatory mandatory mandatory voluntary voluntary voluntary mandatory mandatory voluntary voluntary voluntary 2.31 3.73 3.49 3.28 2.68 4.48 2.50 3 Annuity or scheduled withdrawal relative to average yes Annuity or scheduled withdrawal relative to average yes (only for affiliates born before 1945) Annuity or scheduled withdrawal relative to average Annuity or scheduled withdrawal relative to average Annuity only Annuity or scheduled withdrawal A nnuity only Annuity or scheduled withdrawal relative to unregulated unregulated g relative to average average yes yes yes yes no yes yes no no yes yes no yes no 12/16/2004 Presentation for AARP 6
. and later reforms are more likely to make participation in this mandatory Year of reform 1995/2000 h 2000, as yet unimplemented Costa Rica Nicaragua Ecuador Dominican Republic 2001, as yet unimplemented Contribution-related public PAYGO remains closed remains closed system? Total payroll tax rate, pre-reform (%) 22 17 9.25 Total payroll tax rate, post-reform (%) 26 21.5 Varies, but no more than 20 20 Participation of new workers? mandatory mandatory mandatory mandatory Participation of self employed? voluntary voluntary mandatory mandatory Fees & insurance premia (% of wage) (j) 2.5 4.0 2.0 Minimum return on investment? unregulated unregulated relative to average relative to average Minimum contributory pension? yes yes yes Social assistance pension? yes yes yes yes 2001 i 12/16/2004 Presentation for AARP 7
Governments can define pension benefits, contributions, and/or incentives Nature of instrument Mainstay: Pooling Mainstay: Saving* Mandatory Mandatory Voluntary Common name First pillar Second pillar Third pillar Main function Insure against poverty in old age, lower income inequality Smooth consumption over life cycle Smooth consumption over life cycle Main role of Government Principal riskbearer Defines Benefits Defines Contributions Defines Incentives Government Worker Worker Financial instrument Unfunded pay-asyou-go Funded: individual accounts Funded: tax-preferred individual accounts 12/16/2004 Presentation for AARP 8
The reforms differ in extent of choice within mandatory pensions Choice of system Choice of system Mandatory Contributions FF PAYG FF PAYG PAYG FF FF PAYG Bolivia Colombia Argentina Costa Rica Chile Peru Uruguay Dominican Republic El Salvador Mexico 12/16/2004 Presentation for AARP 9
Points to note Big government stays: only one country in which payroll tax was actually lowered Tip of iceberg Chile different in other aspects too System costs high: mandatory management and insurance fees Between 2 and 5 percent of wages, or about 15-20 percent of contributions Workers have options: Old system closed to younger workers in some, not in others But default system was the new, not the old 12/16/2004 Presentation for AARP 10
The Argentine How? Old system gets into trouble in the 1980s, Ministry of Finance puts together a team in 1991 Supported by UNDP and World Bank Inclined toward Chilean model Argentina commits to structural pension reform in 1992 in standby agreement with IMF Pensioners against reform high costs, exclusively private, loss of acquired rights for people below 45 years Business and finance community big supporters of reform June 1992 draft law presented Mixed model, obligatory for those below 45 years Compromises with unions necessary September 1993 law passed by slimmest of margins 12/16/2004 Presentation for AARP 11
The Bolivian How? Old system gets into trouble in the 1970s, Ministry of Finance proposes structural reform in 1992 Supported by USAID Inclined toward Chilean model New President de Lozado endorses reform in 1993 Min. of Enterprise Capitalization entrusted with task World Bank and IDB support reform Public relations campaign starts in 1995 50% of shares of SOEs credited to pension funds Pensioners and trade unions object to reform November 1996 law passed, without much debate Post reform changes May 1997, Bonosol pension of $250 p.a. paid to elderly Stopped in 1997, replaced by less generous Bolivida in 2001 12/16/2004 Presentation for AARP 12
Points to note Fiscal balance goals Ministries of finance generally proponents of reforms Social ministries often opposed to them Financial deepening goals Banking and finance always champions Foreign investors reward such reforms Development goals World Bank often in favor given positive longer term fiscal, capital and labor market implications Reform sometimes even defined as structural, not parametric Many promises fiscal, financial, labor, and social 12/16/2004 Presentation for AARP 13
The Backdrop Demographics and Poverty
The Backdrop Rising Life Expectancy Latin Americans living longer and longer lives Can expect the risk of inability to earn while living to become an increasingly frequent loss Falling Headcount Poverty After setback in 1980s, poverty began falling again Can reasonably expect the risk of old age poverty to become an increasingly rare loss But High Inequality and Volatility Latin America has high and rising inequality And considerably excess volatility of incomes 12/16/2004 Presentation for AARP 15
Life expectancy in Latin America has risen by ten years over last three decades 90 80 70 Rising Life Expectancy & Falling Fertility Rates 6 5 Age 60 50 40 30 4 3 2 Fertility Rate (%) 20 10 0 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 1 0 Life Expectancy at Birth Total Fertility Rate 12/16/2004 Presentation for AARP 16
Latin Americans still alive at 60 can expect to live for another 15-20 years 17.0 16.0 15.0 Life Expectancy of Men at 65 14.0 13.0 12.0 11.0 10.0 2000 2010 2020 2030 2040 2050 2060 2070 C hile Argentina B oliv ia C olom bia Ecuador Mexico Peru Uruguay Venezuela 12/16/2004 Presentation for AARP 17
Poverty in Latin America has fallen by half over the last three decades Declining Poverty in Latin America, 1970-2000 (Source: Sala-i-Martin, 2002) 70 60 50 Headcount Ratio (Percent) 40 30 20 10 0 Latin America Asia Africa Total 1970 1980 1990 1998 12/16/2004 Presentation for AARP 18
But inequality stays high in Latin America, even rises, over last three decades. High and Rising Inequality in Latin America, 1950-2000 (Source: Bourguignon and Morrison, 2002) 60 50 Gini Coefficients (Population Weighted) 40 30 20 10 0 Africa Asia Latin America East Europe Developed Countries 1950 1970 1992 12/16/2004 Presentation for AARP 19
Informality is an aspect of inequality in Latin America: coverage ranges between 10% and 60% 70.0% Proportion of EAP affiliated with pension system or in formal sector 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% Chile 2000 Peru 1999 Colom bia 1999 Argentina 2002 Mexico 2001 Bolivia 2000 El Salvador 1998 Costa Rica 2000 Nicaragua 1999 Male 63.6% 12.8% 19.8% 35.2% 45.1% 11.8% 24.2% 23.2% 8.9% Female 60.9% 9.1% 26.9% 37.2% 46.6% 8.3% 27.6% 22.5% 15.3% Total 62.7% 11.2% 22.3% 36.0% 45.7% 10.3% 25.5% 22.9% 11.0% 12/16/2004 Presentation for AARP 20
The Problem in Latin America: Low and Stagnant Coverage After a decade of reform to multi-pillar systems with mandated savings, coverage remains low Results are contrary to expectation that mandating and individualizing savings would expand coverage Redesign the instruments, or rethink the notion of coverage? 12/16/2004 Presentation for AARP 21
The Analysis Theory and Empirics
What we did Regional study of pension reforms co-authored with Truman Packard (WB) and Juan Yermo (OECD), with Todd Pugatch (WB), sponsored by the Office of the Chief Economist, LAC Region of World Bank Examined multi-pillar reforms using fiscal, financial, and social criteria Employed simple theory, household surveys, administrative data, and actuarial simulations 12/16/2004 Presentation for AARP 23
Briefly, the theory Simple economics of insurance Three options: market insurance, self-insurance, and selfprotection (Ehrlich and Becker, 1972, 2000) Rare losses better insured using market insurance (pooling), frequent losses better through self-insurance (saving) Relevant risks related to old age are Loss of ability to earn while living saving becomes more appropriate as loss becomes more frequent Poverty in old age pooling becomes more appropriate as loss becomes rarer Appropriate roles of governments are Provide instrument for insuring across states of the world: pooling the losses associated with poverty in old age Regulate instruments transferring incomes across stages of life: saving to offset loss of earning ability while living 12/16/2004 Presentation for AARP 24
Briefly, the main results Theoretical: a mixed picture Move towards individualization of social security appropriate given demographic changes, but Unclear if huge mandate is appropriate/sustainable Fiscal: a mixed picture System deficits lowered, build-up of implicit debt slowed, but Strain of transition costs and making debt explicit may have compromised fiscal stability Financial: a mixed picture Dedicated pension providers deepen markets, but Pose a regulatory challenge as evidenced, e.g., in high fees Social: a mixed picture Within-system equity improvements, but Little or no gains in coverage, so overall equity gains small 12/16/2004 Presentation for AARP 25
The reforms reduced buildup of governments implicit pension debt 400% IPD in 2001 IPD in 2020 IPD in 2030 IPD in 2050 350% Government's Implicit Pension Debt as Percentage of GDP 300% 250% 200% 150% 100% 50% 0% Reform No reform Reform No reform Reform No reform Reform No reform Reform No reform Reform No reform Reform No reform Reform No reform Chile Peru Colombia Argentina Uruguay Mexico Bolivia El Slavador 12/16/2004 Presentation for AARP 26
Reforms reduce implicit transfers from poor to richer workers 4 (Percentage Point Difference Between Wealthier and Poorer Workers in Internal Rates of Return Earned from National Retirem ent Security System ) Reform No Reform 3 2 Percentage Points Difference 1 0-1 -2-3 Men Women Men Women Men Women Men Women Men Women Men Women Men Women Men Women Chile Peru Colombia Argentina Uruguay Mexico Bolivia El Slavador 12/16/2004 Presentation for AARP 27
But 2/3rds of median country s fund portfolio consists of government bonds Private Pension Fund Portfolios Consist Mainly of Government Paper 100 90 80 Percent of Total Portfolio 70 60 50 40 30 Costa Rica El Salvador Mexico Argentina Bolivia Uruguay Colombia Chile Peru 20 10 0 Government Securities Financial institutions Corporate Bonds and Equities 12/16/2004 Presentation for AARP 28
Payroll taxes generally rise. Except in Chile and Uruguay, Government Got Bigger in Every Country's Labor Market 50 45 Social Security Contributions, as Share of Wages 40 35 30 25 20 15 10 5 0 Chile Peru Colombia Argentina Uruguay Mexico Bolivia El Salvador Costa Rica Nicaragua Dominican Republic Total payroll tax rate, pre-reform (%) Total payroll tax rate, post-reform (%) 12/16/2004 Presentation for AARP 29
And costs to workers can be high. Between 15-20 Percent of Worker Contributions to Private Funds Still go to Management (December 2002) 40 35 30 25 20 15 10 5 0 Argentina Bolivia Chile Colombia El Salvador Mexico Peru Uruguay Average Administration Fees/W orker Contribution 12/16/2004 Presentation for AARP 30
Sometimes with higher charges for lower wage (younger and poorer) workers Figure 2: Chile - Cumulative cost ratio for different salaries, 1990-2000 29.0 28.5 Cumulative cost ratio (%) 28.0 27.5 27.0 26.5 26.0 25.5 25.0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Cheapest AFP - 3*avg sal Cheapest AFP - avg sal Cheapest AFP - min sal 12/16/2004 Presentation for AARP 31
And coverage remains stalled 70.0 (Contributors to Retirem ent Security Sy stem, Percentage of EAP) 60.0 50.0 % of EAP 40.0 30.0 20.0 10.0 0.0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 Argentina Boliv ia C hile C olom bia El Salv ador Mexic Uruguay Costa Rica Ecuador Nicaragua Brazil 12/16/2004 Presentation for AARP 32
So only a tenth to two-thirds of the aged in Latin America get pensions. Population 65 or older receiving pension income 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% Chile 2000 Peru 1999 Colombia 1999 Argentina 2002 Mexico 2001 Bolivia 2000 El Salvador 1998 Costa Rica 2000 Dominican Republic 1997 Male 57.1% 32.2% 21.3% 69.6% 32.3% 19.2% 12.4% 42.9% 18.7% Female 27.6% 7.9% 9.9% 64.0% 10.5% 5.6% 5.8% 23.9% 10.6% Total 41.4% 19.5% 15.2% 66.2% 20.0% 11.9% 8.8% 33.2% 14.6% 12/16/2004 Presentation for AARP 33
The Implications Policy and Politics
Why the mixed results? Not worker myopia, but. Not lack of information, but rational choice High mandatory contributions. Payroll tax higher, not lower, after reforms; Only one case of lowered payroll tax (Chile) High costs. High commissions and insurance fees (e.g., Peru) Transition costs can compromise macro stability (e.g., Bolivia).and doubtful benefits. Reliance on government remains, and hence persistence of policy risk (e.g., Argentina) 12/16/2004 Presentation for AARP 35
Chile: Government costs of topping-up projected to rise 100% Proyección del flujo de pensiones (PM crece a 2%, salarios al 2%, rentabilidad 5%) 80% % de afiliados 60% 40% 20% 0% 2005 2015 2025 2035 Pensiones menores a PM sin GE Pensiones menores a PM con GE Pensiones mayores a PM Chilean Chilean Superintendencia Superintendencia of of Pensions Pensions 12/16/2004 Presentation for AARP 36
Peru: Governments have found it hard to regulate private funds 100,0 30 Return on Equity (%) 90,0 80,0 70,0 60,0 50,0 40,0 30,0 20,0 10,0 0,0 Fees/Net Contributions Operational Expenses/Net Fees Return on Equity 1998 1999 2000 2001 2002 25 20 15 10 5 0 Fees/Net Contributions and Expenses/Net Fees (%) 12/16/2004 Presentation for AARP 37
Bolivia: Countries that managed PAYG systems poorly may not manage transitions well either Cash Flow Gap: Bolivia pension-related deficit as percentage of GDP 4.0 2.0 0.0 1998 1999 2000 2001 2002 actual projected in 1997 12/16/2004 Presentation for AARP 38
Points to conclude The main achievement of the LAC reforms is individualization of old age income security Appropriate given the demographic changes Observed as stemming of fiscal bleeding, but can be seen as inappropriateness of pooling The importance and extent of privatization may have been exaggerated Has not meant reduced presence of government payroll tax rates went up, not down Has not meant that government promises to pensioners are much lower government bonds biggest part of portfolios The main indicator of the weakness of reforms is stalled participation Coverage rates have not gone up, and Poverty prevention tiers have not got attention they deserve 12/16/2004 Presentation for AARP 39