1. Financial Performance of Pension Funds
Pensions for the Future
FIAP, Lima, Peru, May 2008
Pablo Antolin
DAF/FIN OECD, Private Pension Unit

2. OECD – WB Private Pensions Partnership
Preliminary study comparing “Investment Performance of Privately Managed Pensions Funds” across several Latin America, CEE and OECD countries (private partners: BBVA, ING and VB).
2nd stage: comparative analysis of performance using sophisticated benchmarks

3. Study so far
1st, describes privately managed pension funds around the world and their regulatory environment, comparing them according to total assets under management and asset allocation. It also describes the different approaches used to evaluate assets and to report investment returns.
2nd, it discusses potential problems with the data reported by pension funds and regulators on investment returns.

4. Study so far
3rd, preliminary assessment of aggregate investment performance across countries using relatively standard risk-adjusted measures: Sharpe ratios and attribution analysis (Walker).
4th, Ex-post comparison of investment performance of privately managed pension funds on a risk-adjusted basis.
5th, analysis of administrative expenses (report on costs).

5. PrMg pension funds around the world
Mandatory-voluntary; occupational-personal; DB- DC

6. Regulatory investment environment

7. Total assets and asset allocation

8. Assets (% of GDP)
Source: OECD, Global Pension Statistics

9. Basic Statistics on Real Returns

10. Real Investment Return (average 2000-05)

11. Average Annual Returns and their SD

12. Potential problems with the data reported on investment performance
Data available from pension funds and regulators suffer of several biases that render any comparison attempt across countries hazardous:
Different valuation methodologies
Differences in expenses charged
Differences in the legal environment (e.g. investment restrictions)
The data suffer also of some biases.

13. Potential biases with the data reported
The data suffer also of some biases.
“lagged aggregate weights
“weighted average share values”
“non-overlapping synchronized portfolio, benchmark and risk free returns”
“clear portfolio composition separation between different asset classes” and
“survival bias”.

14. Exploratory analysis of performance
Calculations of the Sharpe ratio with respect to the short-term and long-term, both local and foreign, riskless assets returns.
Attributions analysis
Pension funds have obtained a positive risk premium, i.e. have added value.
Few countries (e.g. Hungary) added value for long and short riskless instruments, as well as local and foreign. Most countries added value with respect to short-term instruments, either local or foreign.

15. Comparing ex-post returns
Comparing ex-post returns with the best performing portfolio with the same risk level (i.e. same standard deviation), with and without constraints suggests
Obviously, pension funds have done worst that the best portfolio ex-post. But not by much.
Investment restrictions do not seem to have been binding in some countries (best performing portfolio high percentage of domestic bonds).

16. Conclusion
Availability of adequate data important limitation for conducting this study, especially the evaluation of the risk-adjusted investment performance of pension funds.
Need for an effort to develop international standards for reporting pension fund financial performance data that could support international comparisons and more in depth performance evaluation.
Structural problem: most private systems have less than 10 years of history, which imposes a restriction for examining the long-term nature of investment performance.

17. Conclusions 2
Sharpe ratio and attribution analysis show that, for those countries with enough information and data to adjust returns accordingly, privately managed pension funds have obtained a risk premium against short-term investment alternatives.
Pension funds have generally underperformed with respect to the hypothetical portfolio with the highest (mean) return for a given level of risk.
The analysis suggests that in most countries investment restrictions have had a negative impact on performance.

18. Next steps
The next steps of the research agenda will focus on developing system and country specific benchmarks that would allow assessing the impact of pension design and different regulatory environments on the investment performance of privately managed pension funds across the world