Lesson 6.1 compared strategies one at a time. Real investors don't buy one strategy. They build an allocation: a set of commitments across strategies and managers, sized to meet a return target without taking more risk than their board will accept.
This case study follows one fictional pension plan through that decision. LP-side analysts and consultants do this work directly. Managers' product and IR teams see it from the other side, because it explains why an allocator might want one of their funds and not another. The numbers are illustrative throughout.
The situation
The Calder Valley Teachers' Retirement System is a fictional US public pension plan. Its board has approved a $500m private credit allocation. The investment committee wants steady income, some upside, and diversification away from the plan's large public bond portfolio.
After meeting several managers, including the fictional multi-strategy manager Hollin Bay Capital, the staff propose four strategies:
- Senior direct lending (Module 1) as the core, for steady cash income.
- Asset-based finance (Module 3) for exposure to asset pools rather than single companies.
- Opportunistic and junior credit (Module 2) for higher returns.
- Real estate debt (Module 4) for exposure to property, secured by buildings.
The proposed allocation
The following example is illustrative. All yields and expected losses are illustrative assumptions, not market figures. Opportunistic and junior credit uses lesson 6.1's mezzanine expected loss of 2.4%.
| Strategy | Allocation | Yield | Expected loss | After losses |
|---|---|---|---|---|
| Senior direct lending | $250m | 10.0% | 0.6% | 9.4% |
| Asset-based finance | $100m | 9.0% | 0.4% | 8.6% |
| Opportunistic / junior | $75m | 13.0% | 2.4% | 10.6% |
| Real estate debt | $75m | 9.5% | 0.8% | 8.7% |
| Blended | $500m | 10.175% | 0.86% | 9.315% |
Working out the blend
A blended figure is a weighted average: each strategy counts in proportion to its allocation. The weights are $250m ÷ $500m = 50%, then 20%, 15% and 15%.
Blended yield:
50% × 10.0% + 20% × 9.0% + 15% × 13.0% + 15% × 9.5% = 5.0% + 1.8% + 1.95% + 1.425% = 10.175%
Blended expected loss:
50% × 0.6% + 20% × 0.4% + 15% × 2.4% + 15% × 0.8% = 0.30% + 0.08% + 0.36% + 0.12% = 0.86%
Return after expected losses:
10.175% − 0.86% = 9.315%
The same result in dollars, for one year:
| Strategy | Income | Expected loss | After losses |
|---|---|---|---|
| Senior direct lending | $25.000m | $1.500m | $23.500m |
| Asset-based finance | $9.000m | $0.400m | $8.600m |
| Opportunistic / junior | $9.750m | $1.800m | $7.950m |
| Real estate debt | $7.125m | $0.600m | $6.525m |
| Total | $50.875m | $4.300m | $46.575m |
$46.575m ÷ $500m = 9.315%. A common mistake is to take a simple average of the four yields, (10.0 + 9.0 + 13.0 + 9.5) ÷ 4 = 10.375%. That treats the $75m opportunistic sleeve as if it were as large as the $250m core.
The committee might also test a change. Moving $50m from senior direct lending to opportunistic credit ($200m and $125m) raises the blended yield to 10.475% and the expected loss to 1.04%, for a return after losses of 9.435%. The yield goes up 0.30 points, but only 0.12 points of that survive expected losses.
What the average leaves out
The 9.315% is a useful starting point, but it's an expected figure, before fees, in an average year. An investment committee would push on five things.
Concentration
The opportunistic sleeve is 15% of the money but contributes 0.36 of the 0.86 points of expected loss, about 42%. The riskiest sleeve drives losses far more than its size suggests.
Concentration also shows up inside each sleeve. A $75m commitment to one opportunistic fund may end up in a handful of positions. And if Calder Valley gave all four sleeves to Hollin Bay Capital, which runs funds in each strategy, the plan would have one manager's team, processes and business risk behind its whole allocation. Many allocators spread commitments across managers and across vintage years (the years funds start investing) for that reason.
Liquidity
All four strategies are illiquid. Calder Valley can't sell out quickly if its needs change, and selling fund interests on the secondary market can mean accepting a discount. There's also timing: drawdown funds call capital over time (PE101 2.2), so the $500m won't be invested on day one, and loans repay and distributions return money before the plan is fully invested. Allocators commonly plan commitments over several years to reach and hold a target level. If public markets fall, private credit can become a larger share of the plan than intended while its values adjust more slowly.
Correlation in a recession
The expected losses are averages across good and bad years, and they're treated here as if each strategy behaves independently. In a recession they tend not to. Defaults rise and recoveries fall across corporate, consumer and property borrowers at the same time (PC101 4.1).
As a simple illustration, if every sleeve's expected loss doubled in a bad year, the blended loss would be 1.72% and the return after losses 8.455%. Real stress can be worse and uneven: in lesson 6.1's recession case, mezzanine's loss rose to 6.4%, well over double. The four strategies were chosen partly for diversification, but in a severe downturn diversification can help less than the average suggests.
Manager selection
The table uses one yield and one loss rate per strategy, as if every manager were average. In practice, results can differ widely between managers in the same strategy, particularly in opportunistic and specialist strategies where outcomes depend on a few decisions. Choosing managers, and checking their underwriting, workout record and alignment, can matter as much as choosing strategies. That's the job of manager due diligence.
Fees
Everything above is before fees and fund costs. Management fees, incentive fees (PC303 3.3) and fund expenses come out of the 9.315%, and they vary by strategy and manager. Every percentage point of fees and costs would reduce the return to Calder Valley from 9.315% to 8.315%. Allocators compare strategies net of fees, not on gross yield.
The committee's decision
A sensible committee wouldn't approve the allocation on the blended number alone. It might ask staff for the return after losses in a recession case, the expected fees for each sleeve, a plan for spreading commitments across managers and vintage years, and a pacing plan showing how long it would take to reach $500m invested. It might also set limits, for example a cap on the opportunistic sleeve, and review them once the portfolio is built.
That's the real lesson of the course: each strategy is a different way of taking credit risk, and the value of an allocation depends as much on how those risks combine as on the average yield.
Key terms
- Allocation: A set of commitments across strategies and managers, sized to meet an investor's goals.
- Weighted average: An average in which each item counts in proportion to its size.
- Blended yield: The weighted average yield across an allocation.
- Concentration: Exposure to a small number of strategies, managers or positions that dominates the risk.
- Correlation: The tendency of different investments to move, or lose money, together.
- Vintage year: The year a fund starts investing, used to spread commitments over time.
- Pacing plan: A schedule of future commitments designed to reach and hold a target allocation.
Key takeaways
- Blended figures are weighted by allocation: Calder Valley's blended yield is 10.175%, expected loss 0.86% and return after losses 9.315%.
- The 15% opportunistic sleeve contributes about 42% of expected loss, so risk is more concentrated than the dollar split suggests.
- Averages hide bad years: losses tend to rise together in a recession, and doubling every sleeve's loss would cut the return after losses to 8.455%.
- Liquidity, manager selection and fees all shape what the plan actually earns.
- An allocation should be judged on how its risks combine, not just on its average yield.
This lesson is for educational purposes only and is not investment advice.