## _wp07108

## Source details

**Canonical URL:** [_wp07108](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07108.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07108.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07108.pdf.json)

---

### Introduction and key findings
- Real average growth of the GDP in 1980-2001: 4.63
- Corbo and Schmidt-Hebbel (2003) estimate that pension reform increased Chile’s average annual GDP growth rate by 0.5 percent in 1980-2001.
- Estimated components of the reform’s effect on GDP growth (Table 1):
  - Saving and investment: 0.13
  - Labor markets:
    - Increase in employment: 0.07
    - Increase in productivity: 0.03
  - Financial development and TFP: 0.2
  - Total*: 0.49
  - *The total is calculated as a compound rate and is therefore not equivalent to the sum of the individual effets.

### Overview of impacts across Latin America
- Reform design and objectives:
  - “Multi-pillar” approach adopted in the 1980-90s aimed to improve old-age retirement security, increase national saving, develop domestic capital markets, and enhance fiscal sustainability.
- Evidence and outcomes by country grouping:
  - Chile: clear success—reforms contributed to increased savings and growth, and to development of capital markets and financial deepening.
  - Other countries: mixed record—reforms more recent and often lacked complementary structural reforms.
- Financial market development and innovations:
  - AUM growth drove increases in bond and equity market capitalization in selected countries.
  - Pension funds contributed to financial innovation and accumulation of “institutional capital”: annuities, mortgage bonds, asset-backed securities, closed-end mutual funds, local rating companies, improved trading and custody.
- Fiscal solvency:
  - Unfunded liabilities (implicit pension debt, IPD) were often reduced substantially in reformed cases, with exception of Argentina.
  - Reductions achieved by phasing out PAYG programs and parametric reforms (increases in retirement ages, contribution rates and periods, reductions in benefits, rationalization of multiple programs).

### Limitations, risks, and outstanding issues
- Coverage and formalization:
  - Coverage ratios have stalled and remain well below OECD levels (GPY, 2005).
  - Coverage higher (>50 percent of economically active population) in Argentina, Chile, Uruguay; lower (<30 percent) in Bolivia, El Salvador, Peru.
  - Individual accounts show only a small, gradual positive incentive effect on formalization.
- Replacement rates and returns:
  - Chile: average current-worker replacement about 60 percent of final salary; projected to decline to just over 40 percent for later cohorts.
  - Real return on pension contributions has been above 8 percent in most countries (see Table 2), but recent yield declines risk lower replacement rates for future retirees.
  - World Bank experience suggests mandated contributions in excess of 20 (10) percent are likely detrimental for middle- and high-income (low income) countries.
- Administrative costs:
  - Colombia fees: 13.3 percent of contributions and 2.4 percent of AUM.
  - Chile fees: around 1 percent of AUM.
  - Scope for centralizing collections, assigning new entrants to lower-fee funds, and other cost-reduction measures.
- Policy recommendations emphasized:
  - Strengthen coverage through basic-income provisions for the elderly (non-contributory safety net).
  - Improve cost effectiveness and risk management of defined contribution programs.
  - Promote voluntary savings, including tax incentives.

### Financial stability issues and portfolio composition
- Concerns and dynamics:
  - Regulatory constraints on asset allocation can lead to distortions and financial instability.
  - Rapid AUM growth has sometimes distorted domestic securities prices and concentrated risk in PF portfolios.
  - Policymakers may consider relaxing limits on foreign investments while managing potential FX market volatility.
- Regional portfolio composition (December 2005, in percent of total):
  - Government Debt: 46.4
  - Financial Institutions: 15.9
  - Non-financial Institutions: 8.5
  - Equities: 10.6
  - Mutual Funds: 2.4
  - Foreign Investments: 15.1
  - Other: 1.1
- Comparison (December 1999, regional, in percent of total):
  - Government Debt: 49.6
  - Financial Institutions: 23.2
  - Non-financial Institutions: 3.5
  - Equities: 12.9
  - Mutual Funds: 3
  - Foreign Investments: 7
  - Other: 1

### Country-specific portfolio shares (December 2005, in percent of total) — Table 3 entries
- Argentina: Government Debt 60.9; Financial Institutions 5.1; Non-financial Institutions 1.8; Equities 13.4; Mutual Funds 8.1; Foreign Investments 8.9; Other 1.8
- Bolivia: Government Debt 70; Financial Institutions 6.8; Non-financial Institutions 13.5; Equities 6.3; Mutual Funds 2.5; Foreign Investments 0.9
- Chile: Government Debt 16.4; Financial Institutions 28.9; Non-financial Institutions 6.8; Equities 14.7; Mutual Funds 2.8; Foreign Investments 30.2; Other 0.2
- Colombia: Government Debt 47.3; Financial Institutions 10.4; Non-financial Institutions 14.4; Equities 11.3; Mutual Funds 2.1; Foreign Investments 10.4; Other 4.3
- El Salvador: Government Debt 81; Financial Institutions 12.7; Non-financial Institutions 6.3; Equities 0
- Mexico: Government Debt 82.1; Financial Institutions 4.2; Non-financial Institutions 11.8; Equities 0.4; Mutual Funds 1.5
- Peru: Government Debt 20.3; Financial Institutions 11.1; Non-financial Institutions 10.7; Equities 36.4; Mutual Funds 2.8; Foreign Investments 10.1; Other 8.7
- Uruguay: Government Debt 59.5; Financial Institutions 36.8; Non-financial Institutions 2.7; Equities 0.1; Mutual Funds 0.9
- Total (regional): Government Debt 46.4; Financial Institutions 15.9; Non-financial Institutions 8.5; Equities 10.6; Mutual Funds 2.4; Foreign Investments 15.1; Other 1.1
- Cross-country notes:
  - Government bond concentration ranges from under 20 percent in Chile and Peru to more than 80 percent in El Salvador and Mexico.
  - Chilean funds lead in foreign asset holdings with 30 percent of AUM invested in foreign assets.
  - Regulatory limits on foreign investments remain a key policy constraint; recent Chile proposals will increase long-term foreign investment limit to 80 percent.

### Pension fund returns (Table 2)
- Annual returns (in percent) by country and measure:
  - Argentina: Nominal historical 14.9; Real historical 9.4; Nominal last 12 months 17.5; Real last 12 months 4.6
  - Bolivia: Nominal historical 13.7; Real historical 9.8; Nominal last 12 months 8.6; Real last 12 months 3.5
  - Chile 1/: Nominal historical 22.9; Real historical 108.6; Nominal last 12 months 4.6
  - Colombia: Nominal historical 20.7; Real historical 8.3; Nominal last 12 months 24.8; Real last 12 months 19
  - El Salvador: Nominal historical 12.4; Real historical 9.3; Nominal last 12 months 5.8; Real last 12 months 1.5
  - Mexico: Nominal historical 16.3; Real historical 7.7; Nominal last 12 months 11.5; Real last 12 months 8
  - Peru 2/: Nominal historical 13.2; Real historical 8.8; Nominal last 12 months 20.2; Real last 12 months 18.4
  - Uruguay: Nominal historical 23.0; Real historical 12; Nominal last 12 months 9.7; Real last 12 months 4.6
  - Notes:
    - 1/ Returns correspond to the fund "tipo C"
    - 2/ Historical returns correspond to the last 120 months and to the fund "tipo 2" or "mixto", which represents 92% of the total.

### Government bond concentration — rationale, regulatory approaches, and country examples
- Rationale for large allocations to government debt:
  - PF acquisition of government bonds smoothed transition financing for residual PAYG obligations.
  - PF investment helped establish yield curves and contributed to acceptance and use of indexed bonds.
  - Early-stage PF managers may be inexperienced in risk management; initial constraints deemed appropriate.
- Regulatory examples:
  - Floors:
    - Uruguay: funds required to invest 40 to 60 percent of assets in government securities.
    - Bolivia: two existing pension funds must invest together a minimum of $180 million annually in government bonds.
  - Ceilings:
    - Argentina and Colombia: funds allowed to hold government bonds up to half their AUM.
- Argentine crisis (2001) consequences:
  - Default, devaluation, and pesoization caused significant losses to PFs.
  - Pension funds’ large holdings of government debt were close to the 50 percent ceiling before the first debt exchange in 2001.
  - As of end-2005, restructured debt accounted for around 40 percent of AUM, the bulk quasi-par bonds not valued at market prices.

### Debt management, captive demand, and market-development consequences
- Increasing PF role in local bond markets raises issues:
  - PFs provide “captive” demand for government debt, lowering sovereign cost of funds with potential undesirable consequences:
    - Fiscal discipline may be lessened.
    - Low yields can adversely affect PF returns and future replacement ratios.
    - Less incentive for governments to issue indexed instruments protecting against inflation risk.
- Recommendation: Ensure debt management policies align with pension fund investment limits and portfolio needs and encourage development of private securities markets.

### Asset price distortions from rapid AUM growth and regulatory limits
- Equity market example (Chile):
  - 1981–1985: PFs could hold only government and corporate bonds, mortgage-backed securities, or bank deposits.
  - 1985: PFs permitted up to 30 percent in equities.
  - Equity prices increased at an average annual rate of over 30 percent until 1993, when the 30 percent cap was reached.
  - Other contributing factors: privatization and GDP growth rates around 7 percent.
- Bond-market distortions examples:
  - Mexico early 2003: external debt (swapped to pesos) yield curve much higher than local bond yield curve due to limits on PF arbitrage.
  - Peru: caps on investments abroad and lack of alternatives allowed dollar-denominated corporate bonds to issue at yields lower than the sovereign despite lower credit quality.
  - Colombia: ceiling on government bond holdings created strong demand for corporate bonds and very low spreads above the sovereign.

### Empirical data on PF share of government debt and market indicators — Table 4 entries (Pension Funds Share of Government Debt Markets, in percent)
- Argentina: 31-Dec-02 = 16.3; 31-Dec-05 = 16.8; 31-Dec-98 = 2.8
- Bolivia: 31-Dec-02 = 15.1; 31-Dec-05 = 36.4; 31-Dec-98 = 63.8
- Chile: 31-Dec-02 = 7.6; 31-Dec-05 = 8.9; 31-Dec-98 = 21.0
- Colombia: 31-Dec-02 = 8.6; 31-Dec-05 = 48.5; 31-Dec-98 = 77.5
- El Salvador: 31-Dec-02 = 13.6; 31-Dec-05 = 36.3; 31-Dec-98 = 24
- Mexico: 31-Dec-02 = 6.4; 31-Dec-05 = 14.8; 31-Dec-98 = 28.7
- Peru: 31-Dec-02 = 5.1; 31-Dec-05 = 33; 31-Dec-98 = 63.8
- Uruguay: 31-Dec-02 = 6.5; 31-Dec-05 = 18.0; 31-Dec-98 = 2.6
- Total: 31-Dec-02 = 14.5; 31-Dec-05 = 0.4

### Policy responses to distortions and market development
- Reforms and measures cited:
  - Chile: privatization of public enterprises; two comprehensive capital market laws to improve pension system and increase available instruments.
  - Peru: regulatory steps to facilitate infrastructure-related securities and structured instruments.
  - Colombia: 2005 law on corporate governance expected to support more equity issuance; PFs investing in private equity.
- For smaller countries:
  - Emphasis on easing constraints on PF investments abroad due to limited scope for local securities market development.

### Pension fund investment in foreign assets — limits, behavior, and optimality
- Common regulatory rationales for restricting foreign investments:
  - Desire to develop local capital markets.
  - Monitoring difficulties and disclosure concerns for overseas investments.
  - Concerns about higher costs and duplication in asset management fees.
  - Concern PFs may exacerbate foreign currency market volatility as they grow.
- Regulatory ceilings on foreign assets (as of December 2005):
  - Chile: up to 30 percent of AUM.
  - Argentina: up to 20 percent.
  - Colombia: up to 20 percent.
  - Mexico: up to 20 percent.
  - Peru: 12 percent.
  - El Salvador: zero.
  - Uruguay: zero.
- Actual allocations:
  - Generally below 10 percent of AUM except Chile and Peru where allocations approach limits.
- Table 6 entries preserved as in source (exact strings):
  - Argentina: "208.920.8"
  - Bolivia: "50/10 2/2.566.8"
  - Chile: "3030.230.6"
  - Colombia: "2010.414.4"
  - El Salvador: "00"
  - Mexico: "201.51.1"
  - Peru: "1210.146.7"
  - Uruguay: "0 45.2"
  - Memo: Canada (2001) 21.4; Denmark (2001) 25; Ireland (2001) 67.8; Netherlands (2001) 65; Spain (2001) 34.3
- Case studies:
  - Chile: overseas investment limits raised from zero in early 1990s to 30 percent in June 2002; PFs slow to take up space due to high domestic returns; rapid reallocation toward foreign assets began in 1997 after two years of large negative local stock market returns; funds currently at regulatory limit.
- Optimal-portfolio studies (selected results):
  - Walker (2005) static mean-variance optimal portfolio for Chile suggests optimal holding of foreign assets around 48 percent of AUM.
  - Jara and others (2005) for Colombia find optimal allocation between 44 and 70 percent of AUM.
  - Footnote: such calculations do not account for human capital, housing assets, risks of low interest rates at retirement, etc., which can significantly affect optimal portfolio estimates.

### Hedging foreign currency exposure and exchange-rate considerations
- Findings on hedging FX exposure:
  - Studies for Chile and Colombia suggest hedging of FX risk is not optimal.
  - For Chile, optimal to borrow local currency and increase FX exposure beyond what is implied by foreign investments.
  - Rationale: foreign currency can be a natural hedge in global equity sell-offs typically accompanied by local currency depreciations.
- PF-driven capital outflows and exchange-rate effects (examples):
  - Chile: foreign assets share rose from 2 percent by end-1997 to 12 percent by end-1999; associated with roughly 20 percent depreciation of the peso.
  - Canada: increase in foreign investment limit from 20 percent in January 2000 to 30 percent in January 2001 led to a tripling of capital outflows and contributed to a 10 percent depreciation of the Canadian dollar in January 2000 through January 2002.
  - Depreciations also associated with deterioration in non-energy commodity prices.
  - Canada response: foreign banks issued Canadian-dollar denominated bonds (“Maple bonds”); issuers swapped proceeds to U.S. dollars, deepening FX swap market and facilitating sovereign debt management.
- Use of PF regulations as FX intervention:
  - Colombia imposed maximum FX exposure of 20 percent of PF assets in 2001 to stem peso depreciation pressures.
  - Chile increased foreign investment allocations to 20-30 percent partly to counter peso appreciation trend.
  - Uruguay: PFs face constraint on total volume of trading of foreign currency during a trading day.
  - Bolivia: central bank reluctance to set regulatory limit on foreign investments partly due to concerns about FX market pressures.

### Debate on using PF regulation for exchange-rate objectives and mitigating factors
- Arguments for using PF restrictions:
  - Hausmann, Rodriguez-Clare and Rodrick (2005) support foreign investment restrictions to fight real exchange rate appreciation and contribute to short-term real exchange rate stability.
  - Zahler (2004) suggests regulating speed at which PFs change holdings of foreign assets.
- Arguments against relying on PF regulation for FX objectives:
  - PFs likely change allocations gradually even in response to large changes in optimal FX holdings (Borrero, 2005).
  - PFs may mitigate rather than magnify FX market volatility under certain structures (Vergara and Betancourt, 2006); PFs have stabilized markets by purchasing assets liquidated by distressed institutions.
  - Herding behavior partly linked to industry-related minimum return regulations; relaxing such regulations could minimize FX volatility (Srinivas, Whitehouse, and Yermo, 2000; Roldos, 2003).

### Underestimation of fiscal effects from making implicit pension debt explicit
- Pre-reform estimates often omitted:
  - impact of loss of contributions to individual accounts, and
  - payment of recognition bonds to those who moved to partially or fully funded systems.
- Omission "added in some cases more than 10 percentage points of GDP to public debt ratios".
- Converting implicit to explicit debt, given different contractual nature, "increased the cost of borrowing for some reformers."

### Determinants of transition fiscal costs and country evidence
- Key determinants of transition fiscal costs:
  - demographics;
  - features of the old system;
  - recognition of previous commitments (recognition bonds);
  - flexibility of choices given to current workers at reform time.
- Selected country observations and numerics:
  - Chile:
    - Operational deficit of old PAYG system around 4 percent of GDP in first 20 years of funded system.
    - Deficit peaking at around 20 years following the reform.
    - Issuance of recognition bonds peaked at 1½ percent of GDP roughly 25 years after reform start.
    - Minimum and non-contributory pensions add around 0.5 percent of GDP to deficit numbers estimated in Corbo and Schmidt-Hebbel.
  - Bolivia:
    - Initial projections: deficit peak at 2.7 percent of GDP in 1998.
    - Actual: indexation and minimum pension in 2001 brought transitional pension deficit to almost 5 percent of GDP in 2002.
  - Colombia:
    - Inclusion of marginal unfunded programs in 2000 database increased transitional deficit by more than 3 percent of GDP.
    - Deficit projected to remain at around 5 percent of GDP until 2010.
  - El Salvador:
    - Recognition bonds initially expected to peak at 1.5 percent of GDP in 2015; more recent projections estimate peak at 2 ½ percent of GDP in 2018.
  - Argentina and Uruguay:
    - Opted to allow public PAYG to co-exist with new funded system for part of population due to high expected transitional costs.
  - Mexico:
    - Opted not to provide recognition bonds; offered lifetime option to claim benefits from old system if higher at retirement.

### Political economy, administrative failures, cyclical shocks, and stress-testing
- Political and design choices raising costs:
  - Argentina: Congress added recognition bonds and permanent additional benefits for those who stayed in old system.
  - Colombia: allowing those aged 35-40 to retire under old system boosted costs.
  - Peru: initial higher contributions to private system incentivized staying in PAYG; reversal in 1997 led to shifts and higher fiscal costs.
  - El Salvador: proposed retirement age increase rejected by congress, contributing to higher costs.
- Administrative weaknesses and fraud:
  - Bolivia: poor administrative controls, strong incentives to retire under old system, wave of early retirement and fraudulent claims amplified deficits beyond initial projections.
- Cyclical downturns and stress-testing results:
  - Chile recession early in reform raised pension deficit by about 1 percent of GDP higher in 1982-84 (relative to potential GDP).
  - PROST model simulations (Zviniene and Packard, 2002):
    - A sharp recession similar to Chile 1981-84 or Argentina 2000-2003 adds around 1 percent of GDP a year to the transitional deficit.
    - Shock modeled: decline in GDP and employment of 4 percent, and a fall in real wages of 3 percent, for five years, then reversed over following five years.
    - For Argentina and Uruguay (maintained PAYG with employee and employer contributions), recession adds ~1 percent of GDP a year, with larger employment impact in Uruguay.
    - For El Salvador and Mexico (eliminated contributions to public sector), endogenous response less perceptible; GDP impact drives deficit increase—effect larger in El Salvador.

### Impact on explicit public debt and borrowing costs
- Estimated range of impact on explicit debt from pension reform across selected Latin American countries: between 6 percent of GDP (Peru) to over 20 percent of GDP (Bolivia).
- Markets place greater weight on explicit debt; converting implicit to explicit debt "seem to have affected negatively the reformers’ perceived solvency and increased credit risk premier and the cost of debt."
- Selected country estimates shown:
  - Argentina: 11.8 percent
  - Bolivia: 21.8 percent
  - Colombia: 16.2 percent
  - El Salvador: 12.5 percent
  - Peru: 5.6 percent
  - Uruguay: 8.9 percent

### Conclusions and policy implications (selected)
- Overall outcomes:
  - "Pension reforms in most Latin American countries have had a positive effect on macroeconomic performance and capital market development."
  - Reforms "established a much stronger multi-pillar framework for pension systems in the region."
- Policy lessons and recommendations:
  - Strengthen first and third pillars (safety net aspects, including minimum pensions, and complementary/voluntary savings).
  - Improve mandatory second pillar efficiency via measures targeted to improve cost effectiveness and competition among private pension funds.
  - Gradual relaxation of portfolio investment limits, coordinated with sovereign debt management, legislation, and capital markets regulation:
    - Requires active coordination between debt managers, pension, and securities regulators.
    - Introduce instruments that fit pension fund needs (e.g., indexed bonds); early adoption of legislation to facilitate private sector securities (mortgage-backed securities, infrastructure bonds, private and public equity, structured notes).
    - Better corporate governance legislation to broaden safe corporate bond investment options.
  - Prevent excessive concentration of pension funds' portfolios to avoid default risk and temptation to exploit captive demand for government bonds.
  - In small countries with limited local private securities markets, consider opening up to investments abroad.

### Limits on pension funds' investments abroad and fiscal policy interaction
- Recommended approach:
  - Relax limits on pension funds' investments abroad gradually.
  - Regulations on foreign assets and currency exposure should be dictated by diversification needs of PFs and risk-adjusted return opportunities, rather than FX market considerations.
  - Avoid opportunistic use of PF regulations to achieve real exchange rate objectives and reduce FX volatility.
  - Policy focus: foster deepening of FX and financial markets and relax industry-related minimum performance requirements to reduce herding.
  - Raise disclosure and transparency requirements of other financial intermediaries closer to levels required for PFs to reduce excessive focus on PFs as drivers of asset prices.
- Fiscal consequences summary:
  - While reforms improved fiscal solvency in many cases, higher-than-expected fiscal deficits and explicit debt have constrained macroeconomic policies and at times compromised macroeconomic stability.
  - Political economy constraints have in some cases added 2 percentage points of GDP to the transitional fiscal deficit.
  - Combined with a likely one percent more under a cyclical downturn, this combined pressure could constrain fiscal policy and prompt calls for reversal of pension reforms.
  - Public debt has risen by more than 10 percentage points of GDP in some reformers, raising issues of debt sustainability and increasing sovereign funding costs.

### Recommended fiscal and institutional measures
- Build fiscal space before implementing pension reforms (GPY, 2005; World Bank, 2006).
- Develop and use actuarial models to monitor and update impact of pension reforms on fiscal projections continuously.
- Conduct stress-testing of these projections as a crisis-prevention measure.
- Make actuarial and fiscal projections publicly available periodically to strengthen the fiscal policy framework and facilitate discussion of contingent policy measures before deviations become too costly.

*Source — _wp07108 (excerpt provided).*

### 1. Total Estimated Effects of Pension Reform on GDP Growth..................................... 4

### 1. Total Estimated Effects of Pension Reform on GDP Growth..................................... 4

### Introduction and key findings
- Real average growth of the GDP in 1980-20014.63
- Corbo and Schmidt-Hebbel (2003) estimate that pension reform increased Chile’s average annual GDP growth rate by 0.5 percent in 1980-2001.
- The estimated components of the reform’s effect on GDP growth (Table 1):
  - Saving and investment: 0.13
  - Labor markets:
    - Increase in employment: 0.07
    - Increase in productivity: 0.03
  - Financial development and TFP: 0.2
  - Total*: 0.49
  - *The total is calculated as a compound rate and is therefore not equivalent to the sum of the individual effetcs.

### Overview of impacts across Latin America
- Pension reforms adopted in the 1980-90s followed a “multi-pillar” approach and aimed to improve old-age retirement security, increase national saving, develop domestic capital markets, and enhance fiscal sustainability.
- Evidence and outcomes:
  - Chile: clear success—reforms contributed to increased savings and growth, and to development of capital markets and financial deepening.
  - Other countries: mixed record—reforms are more recent and often lacked complementary structural reforms.
- Financial market development:
  - AUM growth drove increases in bond and equity market capitalization in selected countries (Figure 1).
  - Catalan, Impavido and Musalem (2000) indicate cases where pension funds causally contributed to securities market growth.
  - Pension funds contributed to financial innovation and accumulation of “institutional capital” (Walker and LeFort, 2000): annuities, mortgage bonds, asset-backed securities, closed-end mutual funds, local rating companies, improved trading and custody.
- Fiscal solvency:
  - Unfunded liabilities (implicit pension debt, IPD) were often reduced substantially in reformed cases (Zviniene and Packard, 2002), with the exception of Argentina (Figure 2).
  - Reductions achieved by phasing out PAYG programs and parametric reforms (increases in retirement ages, contribution rates and periods, reductions in benefits, rationalization of multiple programs).

### Limitations, risks, and outstanding issues
- Coverage and labor formalization:
  - Coverage ratios have stalled and remain well below OECD levels (GPY, 2005).
  - Coverage higher (>50 percent of economically active population) in Argentina, Chile, Uruguay; lower (<30 percent) in Bolivia, El Salvador, Peru.
  - Individual accounts show only a small, gradual positive incentive effect on formalization.
- Replacement rates and returns:
  - Chile: average current-worker replacement about 60 percent of final salary; projected to decline to just over 40 percent for later cohorts.
  - Real return on pension contributions has been above 8 percent in most countries (see Table 2), but recent yield declines risk lower replacement rates for future retirees.
  - World Bank experience suggests mandated contributions in excess of 20 (10) percent are likely detrimental for middle- and high-income (low income) countries.
- Administrative costs:
  - Fees are relatively high in many countries and absorb a large share of contributions and lower net returns.
  - Examples: Colombia fees are 13.3 percent of contributions and 2.4 percent of AUM; Chile fees around 1 percent of AUM.
  - Scope for centralizing collections, assigning new entrants to lower-fee funds, and other cost-reduction measures.
- Policy recommendations emphasized in the paper:
  - Strengthen coverage through basic-income provisions for the elderly (non-contributory safety net).
  - Improve cost effectiveness and risk management of defined contribution programs.
  - Promote voluntary savings, including tax incentives.

### Financial stability issues and portfolio composition
- Concerns:
  - Regulatory and other constraints on asset allocation can lead to distortions and financial instability.
  - Rapid AUM growth has sometimes distorted domestic securities prices and concentrated risk in PF portfolios.
  - Policymakers may consider relaxing limits on foreign investments while managing potential FX market volatility.
- Portfolio concentration (Figure 3 and Table 3):
  - Regional averages (December 2005, in percent of total):
    - Government Debt: 46.4
    - Financial Institutions: 15.9
    - Non-financial Institutions: 8.5
    - Equities: 10.6
    - Mutual Funds: 2.4
    - Foreign Investments: 15.1
    - Other: 1.1
  - December 1999 (regional pie shown for comparison):
    - Government Debt: 49.6
    - Financial Institutions: 23.2
    - Non-financial Institutions: 3.5
    - Equities: 12.9
    - Mutual Funds: 3
    - Foreign Investments: 7
    - Other: 1
  - Country-specific portfolio shares (December 2005, in percent of total) from Table 3:
    - Argentina: Government Debt 60.9; Financial Institutions 5.1; Non-financial Institutions 1.8; Equities 13.4; Mutual Funds 8.1; Foreign Investments 8.9; Other 1.8
    - Bolivia: Government Debt 70; Financial Institutions 6.8; Non-financial Institutions 13.5; Equities 6.3; Mutual Funds 2.5; Foreign Investments 0.9
    - Chile: Government Debt 16.4; Financial Institutions 28.9; Non-financial Institutions 6.8; Equities 14.7; Mutual Funds 2.8; Foreign Investments 30.2; Other 0.2
    - Colombia: Government Debt 47.3; Financial Institutions 10.4; Non-financial Institutions 14.4; Equities 11.3; Mutual Funds 2.1; Foreign Investments 10.4; Other 4.3
    - El Salvador: Government Debt 81; Financial Institutions 12.7; Non-financial Institutions 6.3; Equities 0
    - Mexico: Government Debt 82.1; Financial Institutions 4.2; Non-financial Institutions 11.8; Equities 0.4; Mutual Funds 1.5
    - Peru: Government Debt 20.3; Financial Institutions 11.1; Non-financial Institutions 10.7; Equities 36.4; Mutual Funds 2.8; Foreign Investments 10.1; Other 8.7
    - Uruguay: Government Debt 59.5; Financial Institutions 36.8; Non-financial Institutions 2.7; Equities 0.1; Mutual Funds 0.9
    - Total (regional): Government Debt 46.4; Financial Institutions 15.9; Non-financial Institutions 8.5; Equities 10.6; Mutual Funds 2.4; Foreign Investments 15.1; Other 1.1
- Cross-country variation:
  - Government bond concentration ranges from under 20 percent in Chile and Peru to more than 80 percent in El Salvador and Mexico.
  - Chilean funds lead in foreign asset holdings with 30 percent of AUM invested in foreign assets.
  - Regulatory limits on foreign investments remain a key policy constraint; recent Chile proposals will increase long-term foreign investment limit to 80 percent.

### Pension fund returns (Table 2)
- Annual returns (in percent):
  - Argentina: Nominal historical 14.9; Real historical 9.4; Nominal last 12 months 17.5; Real last 12 months 4.6
  - Bolivia: Nominal historical 13.7; Real historical 9.8; Nominal last 12 months 8.6; Real last 12 months 3.5
  - Chile 1/: Nominal historical 22.9; Real historical 108.6; Nominal last 12 months 4.6
  - Colombia: Nominal historical 20.7; Real historical 8.3; Nominal last 12 months 24.8; Real last 12 months 19
  - El Salvador: Nominal historical 12.4; Real historical 9.3; Nominal last 12 months 5.8; Real last 12 months 1.5
  - Mexico: Nominal historical 16.3; Real historical 7.7; Nominal last 12 months 11.5; Real last 12 months 8
  - Peru 2/: Nominal historical 13.2; Real historical 8.8; Nominal last 12 months 20.2; Real last 12 months 18.4
  - Uruguay: Nominal historical 23.0; Real historical 12; Nominal last 12 months 9.7; Real last 12 months 4.6
  - 1/ Returns correspond to the fund "tipo C"
  - 2/ Historical returns correspond to the last 120 months and to the fund "tipo 2" or "mixto", which represents 92% of the total.

*Italic: Source — _wp07108 - 1. Total Estimated Effects of Pension Reform on GDP Growth..................................... 4*

### 17. Although this suggests a possibly unhealthy concentration of risk, there are several

### _wp07108 - 17. Although this suggests a possibly unhealthy concentration of risk, there are several

### Government bond concentration and rationale for large allocations
- PF acquisition of government bonds has helped smooth the transition to a funded system by enabling governments to finance the residual obligations of the old PAYG systems (Campbell and Feldstein, 2001).
- PF investment in government bonds can help develop local bond markets by establishing a yield curve and contributing to the acceptance and use of indexed bonds (Mathieson and others, 2004).
- Early-stage PF managers likely to be relatively inexperienced in risk management; therefore constraining PF portfolio choices initially is considered appropriate.

### Regulatory approaches and country examples
- Many governments have imposed floors on holdings of government bonds:
  - Uruguay: funds required to invest 40 to 60 percent of their assets in government securities.
  - Bolivia: the two existing pension funds must invest together a minimum of $180 million annually in government bonds.
- Most countries have also placed ceilings on pension funds’ exposure to government to encourage diversification:
  - Argentina and Colombia: funds allowed to hold government bonds up to half their AUM.

### Argentine crisis and effects on PF portfolios
- The 2001 Argentine crisis highlighted risks of concentrated sovereign exposure: default, devaluation, and pesoization of deposits and local bonds caused losses to pension funds that are difficult to quantify but likely significant.
- Pension funds’ large holdings of government debt were close to the 50 percent ceiling before the first debt exchange in 2001; restructured debt remains on pension funds’ books, some booked at “technical” values not fully reflecting the haircuts of the second exchange.
- Footnote data: As of end-2005, restructured debt accounted for around 40 percent of AUM, the bulk of which are quasi-par bonds that are not valued at market prices.

### Debt management, PFs as captive demand, and market-development consequences
- Increasing role of pension funds in local bond markets raises debt management and local securities market development issues:
  - PFs provide “captive” demand for government debt, lowering cost of funds for sovereigns with potential undesirable consequences.
  - Potential consequences listed:
    - Fiscal discipline may be lessened by easier access to bond financing.
    - Low yields would tend to adversely affect PF returns and the replacement ratios obtained by future pensioners.
    - Lesser incentives for governments to accommodate PF demand for indexed instruments to protect against inflation risk.
- Recommendation: Ensure debt management policies are broadly consistent with pension fund investment limits and portfolio needs, and concerted efforts to encourage development of private securities markets.

### Asset price distortions from rapid PF AUM growth and regulatory limits
- Rapid growth of PF AUM helped develop local instruments and markets, but supply of new securities has sometimes been slow to respond, leading to asset price distortions in equity and bond markets.
- Chile example (equities):
  - 1981–1985: PFs could only hold government and corporate bonds, mortgage-backed securities, or bank deposits.
  - 1985: PFs permitted to allocate up to 30 percent of portfolio in equities.
  - Equity prices increased at an average annual rate of over 30 percent until 1993, when the 30 percent cap was reached.
  - Other contributing factors included privatization and GDP growth rates that reached around 7 percent.
- Bond-market distortions:
  - Mexico early 2003: yield curve on external debt (swapped to pesos) was much higher than local bond yield curve, reflecting limits on PFs' ability to arbitrage yield differentials.
  - Peru: caps on investments abroad and lack of investment alternatives allowed local corporate borrowers to issue dollar-denominated bonds at yields lower than the sovereign despite lower credit quality.
  - Colombia: ceiling on holdings of government bonds created artificially strong demand for corporate bonds, resulting in very low spreads above the sovereign.

### Empirical data on PF share of government debt and market indicators
- Table 4 (Pension Funds Share of Government Debt Markets, in percent) entries as reported:
  - Argentina: 31-Dec-02 = 16.3; 31-Dec-05 = 16.8; 31-Dec-98 = 2.8
  - Bolivia: 31-Dec-02 = 15.1; 31-Dec-05 = 36.4; 31-Dec-98 = 63.8
  - Chile: 31-Dec-02 = 7.6; 31-Dec-05 = 8.9; 31-Dec-98 = 21.0
  - Colombia: 31-Dec-02 = 8.6; 31-Dec-05 = 48.5; 31-Dec-98 = 77.5
  - El Salvador: 31-Dec-02 = 13.6; 31-Dec-05 = 36.3; 31-Dec-98 = 24
  - Mexico: 31-Dec-02 = 6.4; 31-Dec-05 = 14.8; 31-Dec-98 = 28.7
  - Peru: 31-Dec-02 = 5.1; 31-Dec-05 = 33; 31-Dec-98 = 63.8
  - Uruguay: 31-Dec-02 = 6.5; 31-Dec-05 = 18.0; 31-Dec-98 = 2.6
  - Total: 31-Dec-02 = 14.5; 31-Dec-05 = 0.4
- Mexico: External and Domestic Yield Curves (February, 2003) and (September, 2003) depicted divergence and later convergence due to market exploitation.
- Table 5 (Peru: Yields on Dollar-denominated Sovereign and Corporate Bonds issued in 2003-2005) presents corporate bond coupons, sovereign YTM, spreads in basis points, issue dates, maturity dates, coupon types, and currency for TRANSMANTARO, ELECTROANDES S.A., DUKE ENERGY EGENOR; specific numeric entries preserved in source table.

### Policy responses to distortions and market development
- Early adoption of capital market reforms and easing constraints on PF investments abroad can ameliorate distortions.
- Country examples of reforms:
  - Chile: privatized public enterprises; approved two comprehensive capital market laws to improve pension system and increase instruments available.
  - Peru: regulatory steps to facilitate infrastructure-related securities and structured instruments.
  - Colombia: 2005 law on corporate governance expected to support more equity issuance; PFs investing in private equity.
- For smaller countries, emphasis could be on easing constraints on PF investments abroad due to limited scope to develop meaningful local securities markets.

### Pension fund investment in foreign assets: limits, behavior, and optimality studies
- Common reasons for restricting PF foreign investments:
  - Desire to develop local capital markets and spur domestic investment.
  - Difficulty of monitoring PF investments abroad due to disclosure requirements.
  - Concern about higher costs (duplication in asset management fees) of overseas investments.
  - Concern that PFs may exacerbate volatility in foreign currency markets as they grow and herd.
- Regulatory ceilings on foreign assets (as of December 2005):
  - Chile: up to 30 percent of AUM.
  - Argentina: up to 20 percent.
  - Colombia: up to 20 percent.
  - Mexico: up to 20 percent.
  - Peru: 12 percent.
  - El Salvador: zero.
  - Uruguay: zero.
- Actual allocations often much lower than regulatory limits and generally below 10 percent of AUM except Chile and Peru where allocations approach limits.
- Table 6 (Foreign Investment Limits and Allocations) entries preserved as in source:
  - Argentina: Limit 20; Actual 8.9? (source table entries: "Argentina 208.920.8" — preserve exact strings from source)
  - Bolivia: "50/10 2/2.566.8"
  - Chile: "3030.230.6"
  - Colombia: "2010.414.4"
  - El Salvador: "00"
  - Mexico: "201.51.1"
  - Peru: "1210.146.7"
  - Uruguay: "0 45.2"
  - Memo: Canada (2001) 21.4; Denmark (2001) 25; Ireland (2001) 67.8; Netherlands (2001) 65; Spain (2001) 34.3
- Chile case study on gradual liberalization:
  - Limits on overseas investments raised from zero in the early 1990s to 30 percent in June 2002.
  - PFs slow to take up increased room due to high domestic asset returns; rapid reallocation toward foreign assets began in 1997 after two years of large negative local stock market returns; funds currently at the regulatory limit (per source figure).
- Optimal-portfolio studies:
  - Walker (2005) static mean-variance optimal portfolio for Chile suggests optimal holding of foreign assets around 48 percent of AUM.
  - Jara and others (2005) for Colombia find optimal allocation between 44 and 70 percent of AUM.
  - Footnote: such calculations do not take into account human capital, housing assets, risks of low interest rates at retirement, etc., which can significantly affect optimal portfolio estimates.

### Hedging foreign currency exposure and exchange-rate considerations
- Less agreement on benefits of hedging foreign currency exposure:
  - Studies for Chile and Colombia suggest hedging of FX risk is not optimal.
  - For Chile, optimal to borrow local currency and increase FX exposure beyond what is implied by foreign investments (per source).
  - Rationale: foreign currency is a natural hedge against global equity fluctuations because global equity sell-offs usually counterbalanced by local currency depreciations.
- Experience of PF-driven capital outflows and exchange-rate effects:
  - Chile: increase in share of foreign assets from 2 percent by end-1997 to 12 percent by end-1999 was associated with a roughly 20 percent depreciation of the peso.
  - Canada: increase of foreign investment limit from 20 percent in January 2000 to 30 percent in January 2001 led to a tripling of capital outflows and contributed to a 10 percent depreciation of the Canadian dollar in January 2000 through January 2002.
  - In both cases, depreciations were also associated with deterioration in non-energy commodity prices.
  - Subsequent Canadian response: foreign banks issued bonds denominated in Canadian dollars (“Maple bonds”); issuers swapped proceeds to U.S. dollars, deepening FX swap market and facilitating sovereign debt management.
- Use of PF regulations as FX intervention:
  - Colombia imposed maximum FX exposure of 20 percent of PF assets to stem peso depreciation pressures in 2001.
  - Chile increased foreign investment allocations to 20-30 percent partly to counter appreciation trend of the peso.
  - Uruguay: PFs face a constraint on total volume of trading of foreign currency during a trading day.
  - Bolivia: central bank reluctance to set regulatory limit on foreign investments partly attributed to concerns about potential FX market pressures.

### Debate on using PF regulation for exchange-rate objectives and mitigating factors
- Arguments for using PF regulations to achieve exchange rate objectives:
  - Hausmann, Rodriguez-Clare and Rodrick (2005) support foreign investment restrictions to fight real exchange rate appreciation, achieve long-term growth goals, and contribute to short-term real exchange rate stability.
  - Zahler (2004) suggests regulating speed at which PFs change holdings of foreign assets, while recognizing tighter regulations could prevent rapid reallocation responses needed to protect PF returns.
- Arguments against relying on PF regulation to counter FX volatility:
  - PFs likely change allocations gradually even in response to large changes in optimal FX holdings (Borrero, 2005).
  - PFs may mitigate rather than magnify FX market volatility under certain market structures (Vergara and Betancourt, 2006); PFs have played stabilizing roles during market turbulence by purchasing assets liquidated by distressed financial institutions (IMF, 2004).
  - Herding behavior is partly linked to industry-related minimum return regulations; relaxing such regulations could minimize FX volatility (Srinivas, Whitehouse, and Yermo, 2000; Roldos, 2003).

*Italic: Source—_wp07108 (excerpt provided).*

### 35. Most pension reforms improved fiscal solvency, but many of them underestimated the

### _wp07108 - 35. Most pension reforms improved fiscal solvency, but many of them underestimated the

### Underestimation of fiscal effects from making implicit pension debt explicit
- Pre-reform estimates often did not fully account for:
  - the impact of loss of contributions to individual accounts, and
  - the payment of recognition bonds to those who moved to partially or fully funded systems.
- This omission "added in some cases more than 10 percentage points of GDP to public debt ratios".
- Converting implicit to explicit debt, given different contractual nature, "increased the cost of borrowing for some reformers."

### Determinants of transition fiscal costs (country variation)
- Key determinants include:
  - demographics;
  - features of the old system;
  - recognition of previous commitments (recognition bonds);
  - flexibility of choices given to current workers at reform time.
- Table 7 (qualitative ranking across countries) highlights high/medium/low impacts across dimensions such as demographics, retirement age, years of contribution for entitlement, benefit parameters, payroll tax reduction, members transferring to new system, recognition bonds, minimum pension, non-contributory pension, and other guarantees.

### Country examples and observed transitional fiscal deficits
- Chile:
  - "operational deficit (expenditures minus revenues of the old PAYG system) was on average around 4 percent of GDP in the first 20 years of operation of the funded system",
  - "deficit peaking at around 20 years following the reform",
  - issuance of recognition bonds "peaked at 1½ percent of GDP roughly 25 years after the beginning of the reform",
  - "Minimum and non-contributory pensions increase add around 0.5 percent of GDP to the deficit numbers estimated in Corbo and Schmidt-Hebbel".
- Bolivia:
  - initial projections: deficit peak at "2.7 percent of GDP in 1998" and decline thereafter,
  - actual outcome: indexation and introduction of a minimum pension in 2001 brought "the transitional pension deficit to almost 5 percent of GDP in 2002".
- Colombia:
  - inclusion of marginal unfunded pension programs in a 2000 database "increasing the transitional deficit by more than 3 percent of GDP",
  - deficit "projected to remain at around 5 percent of GDP until 2010".
- El Salvador:
  - cost of recognition bonds initially expected to peak at "1.5 percent of GDP in 2015", more recent projections estimate peak at "2 ½ percent of GDP in 2018".
- Peru:
  - figure referenced showing fiscal cost deviations (see Figure 8); historical recession effects noted.
- Argentina and Uruguay:
  - opted to allow public PAYG to co-exist with new funded system for part of population, citing high expected transitional costs due to relatively older populations and mature PAYG programs.
- Mexico:
  - opted not to provide recognition bonds; offered lifetime option to claim benefits from old system if higher at retirement.

### Political economy, design choices, and implementation failures that raised costs
- Political decisions altered initial reform designs:
  - Argentina: Congress added recognition bonds and a permanent additional benefit for those who stayed in the old system.
  - Colombia: allowing those aged 35-40 to retire under old system boosted costs.
  - Peru: initial higher contributions to private system incentivized staying in PAYG; reversal in 1997 led to shifts and higher fiscal costs.
  - El Salvador: proposed retirement age increase rejected by congress, contributing to higher costs.
- Options given to “middle-age” groups caused unexpected shifts:
  - Uruguay (more than 40 years old) and El Salvador (36 to 50/55 years old): distrust led to higher-than-expected shifts to private system and higher deficits.
  - Mexico: option to retire under old system if funded-system returns deliver lower pensions poses risk of increased government claims.

### Administrative weaknesses and fraud
- Bolivia: poor administrative controls, strong incentives to retire under old system, wave of early retirement and fraudulent claims amplified deficits beyond initial projections.

### Cyclical downturns and stress-testing
- Cyclical shocks amplified transitional deficits:
  - Example: Chile recession early in reform raised pension deficit by about "1 percent of GDP higher in 1982-84" (relative to potential GDP).
  - Simulations using the World Bank’s PROST model (Zviniene and Packard, 2002) indicate:
    - A sharp recession similar to Chile 1981-84 or Argentina 2000-2003 adds "around 1 percent of GDP a year to the transitional deficit".
    - The shock modeled: "a decline in GDP and employment of 4 percent, and a fall in real wages of 3 percent, for five years, then reversed over the following five years."
    - For Argentina and Uruguay (maintained PAYG with employee and employer contributions), recession adds ~1 percent of GDP a year, with larger employment impact in Uruguay.
    - For El Salvador and Mexico (eliminated contributions to public sector), endogenous response less perceptible; GDP impact drives deficit increase—effect larger in El Salvador.

### Impact on explicit public debt and borrowing costs
- Making pension liabilities explicit has raised measured public debt:
  - Estimated range of impact on explicit debt from pension reform across selected Latin American countries: "between 6 percent of GDP (Peru) to over 20 percent of GDP (Bolivia)."
- Markets place greater weight on explicit debt, and converting implicit to explicit debt "seem to have affected negatively the reformers’ perceived solvency and increased credit risk premier and the cost of debt."
- Selected country estimates shown in Figure 11 (Primary Fiscal Balance and Public Debt with and without pension reform) include highlighted numbers:
  - Argentina: "11.8 percent"
  - Bolivia: "21.8 percent"
  - Colombia: "16.2 percent"
  - El Salvador: "12.5 percent"
  - Peru: "5.6 percent"
  - Uruguay: "8.9 percent"

### Conclusions and policy implications (selected)
- Overall outcomes:
  - "Pension reforms in most Latin American countries have had a positive effect on macroeconomic performance and capital market development."
  - Reforms "established a much stronger multi-pillar framework for pension systems in the region."
- Policy lessons and recommendations:
  - Strengthen first and third pillars (safety net aspects, including minimum pensions, and complementary/voluntary savings).
  - Improve mandatory second pillar efficiency via measures targeted to improve cost effectiveness and competition among private pension funds.
  - Gradual relaxation of portfolio investment limits, coordinated with sovereign debt management, legislation, and capital markets regulation.
    - Requires active coordination between debt managers, pension, and securities regulators.
    - Introduce instruments that fit pension fund needs (e.g., indexed bonds); early adoption of legislation to facilitate private sector securities (mortgage-backed securities, infrastructure bonds, private and public equity, structured notes).
    - Better corporate governance legislation to broaden safe corporate bond investment options.
  - Prevent excessive concentration of pension funds' portfolios to avoid default risk and temptation to exploit captive demand for government bonds.
  - In small countries with limited local private securities markets, consider opening up to investments abroad.

*Italic: Source — _wp07108 - 35. Most pension reforms improved fiscal solvency, but many of them underestimated the*

### 52. Limits on pension funds investments abroad should be relaxed gradually, and

### 52. Limits on pension funds investments abroad should be relaxed gradually, and

### Investment limits, FX exposure, and market development
- Limits on pension funds' investments abroad should be relaxed gradually.
- Regulations on foreign assets and currency exposure should be dictated by:
  - the diversification needs of the pension funds, and
  - the risk-adjusted return opportunities available to the pension funds—
  - rather than by FX market considerations.
- Opportunistic use of pension fund regulations to achieve real exchange rate objectives and reduce foreign exchange volatility should be avoided.
- Policy focus should be on fostering the deepening of FX and financial markets.
- Some pension fund regulations, such as industry-related minimum performance requirements, could be relaxed to reduce herding behavior.
- Raising the disclosure and transparency requirements of other financial intermediaries closer to the high levels required for pension funds would:
  - remove the excessive focus on funds as drivers of most trends in asset prices, including the exchange rate.

### Fiscal consequences of pension reforms
- While fiscal solvency was improved with most reforms (owing to parametric changes and/or phasing out of the old PAYG systems), higher-than-expected fiscal deficits and explicit debt have constrained macroeconomic policies and at times compromised macroeconomic stability.
- Political economy constraints during the debate and/or implementation of reforms have, in some cases, reversed or made insufficient the originally envisaged parametric changes to the old PAYG regimes.
- This has in some cases added 2 percentage points of GDP to the transitional fiscal deficit.
- If added to a likely one percent more under a cyclical downturn, this combined pressure could constrain fiscal policy and prompt calls for a reversal of pension reforms.
- Public debt has risen by more than 10 percentage points of GDP in some pension reformers, raising issues of debt sustainability and increasing sovereign funding costs—since market participants seem to give more weight to explicit than implicit pension debt.

### Policy recommendations
- Build fiscal space before implementing pension reforms (citing GPY, 2005, World Bank, 2006).
- Develop and use actuarial models to monitor and update the impact of pension reforms on fiscal projections on a more or less continuous basis.
- Conduct stress-testing of these projections as a crisis-prevention measure.
- Make these actuarial and fiscal projections publicly available periodically to:
  - strengthen the fiscal policy framework, and
  - facilitate discussion of contingent policy measures before deviations become too costly.

*Source: _wp07108 - 52. Limits on pension funds investments abroad should be relaxed gradually, and*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07108.pdf_