Halden Foundation: Required Return, Smoothed Private Equity, and Capital Calls
Trustee Ana Duarte reviews the EUR 500 million Halden Foundation. Spending policy is 5% of a three-year average market value; inflation is 2.5%; all-in fees are 60 bp. The policy portfolio is 45% global equity (expected return 8.5%, volatility 17%), 35% bonds (4.0%, 6%) and 20% private equity (11.0%, 9%). The reported private-equity return series shows first-order autocorrelation of 0.45. The board asks whether shifting to 35% private equity funded from bonds would close the return gap. Private capital calls must be met within 10 business days, and the foundation must fund a 5% spending outflow plus a 2% operating buffer.
1. Compute the foundation's required nominal return and compare it with the policy portfolio's expected return.
Read the worked answer
Required return must cover spending, inflation and fees multiplicatively:
(1.05)(1.025)(1.006) - 1 = 8.27%
Policy portfolio expected return = 0.45(8.5) + 0.35(4.0) + 0.20(11.0) = 3.825 + 1.400 + 2.200 = 7.43%.
So the portfolio is expected to fall about 84 bp short of the required return each year, meaning a slow real erosion of the corpus.
Two refinements before the board treats 8.27% as the hurdle. First, spending is based on a three-year average market value, so in a sharp drawdown spending falls with a lag rather than immediately - the effective required return in the first year of a bear market is higher than the formula implies. Second, if the 8.5% equity and 11.0% private-equity assumptions are arithmetic returns while the required return is geometric, the comparison is biased in the portfolio's favour; the gap is likely wider than 84 bp.
2. Estimate the expected return of a 45/20/35 equity/bond/private equity mix. Should the board rely on the reported private-equity volatility and correlations when evaluating the change?
Read the worked answer
New mix: 0.45(8.5) + 0.20(4.0) + 0.35(11.0) = 3.825 + 0.800 + 3.850 = 8.48%, which clears the 8.27% hurdle by about 21 bp.
Do not take the reported risk numbers at face value. First-order autocorrelation of 0.45 in the return series is the signature of appraisal-based smoothing. Smoothed returns understate volatility and understate correlation with public equities, because the appraisal lags the market. The consequences are systematic: the reported Sharpe ratio of private equity is inflated, and a mean-variance or risk-budgeting exercise fed with these inputs will over-allocate to the asset class and understate total portfolio risk.
The board should use desmoothed estimates or, more simply, apply a haircut to the return/volatility pair and stress-test with correlations closer to equity. The 21 bp margin over the hurdle is far smaller than the estimation error in any of these inputs.
3. Advise the board on whether to raise private equity to 35%, considering liquidity.
Read the worked answer
The arithmetic supports a modest increase, but three constraints argue for a staged approach rather than an immediate move to 35%.
Liquidity. 35% of EUR 500m is EUR 175m, drawn against capital commitments on 10 business days' notice. Annual spending is EUR 25m, and the 2% buffer adds EUR 10m of required liquidity. A portfolio with 35% private equity and only 20% bonds leaves a thin liquid sleeve to meet both spending and an unexpected capital call in a market where the private NAV cannot be sold quickly. A committed subscription line or a larger short-duration sleeve is the practical fix.
Risk. Desmoothed, the private allocation is more equity-like than the 9% volatility suggests, so the diversification benefit is overstated and the effective equity exposure rises well above 45%.
Recommendation. Raise private equity toward 30% in stages over two to three vintages, fund it partly from equity rather than entirely from bonds so that the liquid sleeve is preserved, and require desmoothed risk inputs in the next asset-liability study. If the return gap must close sooner, the board should also revisit the 5% spending rule.