The question “how much should I allocate to alternatives?” is asked constantly and answered confidently, usually with a percentage range and a chart showing what large institutions do.
The honest answer is that the question is underspecified. The right size depends on things that vary enormously between investors, and on at least one input that the standard analysis systematically distorts.
Why the usual analysis overstates the case
Portfolio optimisation frameworks take expected return, volatility, and correlation as inputs and produce an allocation. Feed in private market data and the models generally recommend substantial allocations.
The difficulty is with two of the three inputs.
Private market valuations are periodic and appraisal-based rather than continuous and market-determined. A portfolio marked quarterly by a valuation process reports lower volatility than the same economic exposure marked continuously by a market — not because the exposure is less risky, but because the measurement responds more slowly and with less noise.
The same mechanism suppresses measured correlation with public markets. If private assets are marked with a lag, their reported movements do not line up with public market movements, and the correlation calculation reflects the lag rather than the underlying relationship.
Both effects push in the same direction: they make private assets look like better diversifiers with more attractive risk-adjusted returns than the underlying economics support. An optimiser fed those inputs will recommend a larger allocation, and it will do so for reasons that are artefacts of measurement.
This does not mean alternatives have no diversification value. It means the quantitative case is weaker than the models say, and an allocation justified primarily by optimiser output rests on inputs that do not mean what they appear to mean.
Why institutional allocations are not a template
Large endowments and pensions do hold substantial alternatives allocations, and this is frequently cited as validation.
Four things make their position different, and an investor lacking any of them is not in the same situation:
Horizon. A permanent endowment has no liquidation date. Most investors do.
No forced liquidation. They are not going to need the capital at a bad moment because of a job loss, a medical event, or a retirement date.
Staff. Dedicated teams performing manager selection, operational due diligence, and monitoring continuously.
Access. Relationships with managers who do not accept small commitments, and often cannot accept new investors at all.
That last point deserves emphasis. Manager dispersion in private markets is wide — considerably wider than in most public strategies. The returns commonly cited for the asset class are heavily influenced by managers that most investors cannot reach. An investor with access to the available rather than the desirable end of the manager universe should not assume the asset class averages apply to them.
What actually determines the outcome
Three things, roughly in order of importance:
1. Manager selection. Given the dispersion, which managers were used matters more than how much was allocated. A large allocation to weak managers is worse than a modest allocation to strong ones.
2. Diversification within the allocation. Across managers, across strategies, and across vintage years. An investor with three fund positions has manager risk, not asset class exposure — and the outcome will be determined by whether those three happened to be good.
3. The size of the allocation. Real, and third.
This ordering has a practical implication that is rarely stated: an investor who cannot diversify should probably allocate less, not more. The instinct to concentrate when capital is limited — to make the few positions count — is precisely backwards, because concentration in illiquid, opaque, high-dispersion assets is where the worst outcomes come from.
The constraints that should bind first
Before any percentage is considered:
Liquidity needs. How much capital might genuinely be required, over what period, in what circumstances — including bad ones. This is a hard constraint, not a preference. See liquidity budgeting.
Capital call capacity. In drawdown structures, committed capital must be funded on the manager’s schedule. The liquidity held against calls is dry powder earning something else — a real cost that is usually excluded from return comparisons.
Tax situation. K-1s, multi-state filings, and delayed reporting are an annual cost in fees and irritation, recurring for the life of every position.
Access and eligibility. Accredited status, qualified purchaser status, and minimums determine what is actually available.
Diligence capability. Every private position requires work. An investor who cannot do it, and is not paying someone who can, is relying on whoever recommended the investment.
Behavioural tolerance. Long periods with no distributions, negative early reported returns from the J-curve, and no ability to exit. Investors who will find this uncomfortable should hold less regardless of what a model says.
Building an allocation that is actually diversified
For investors with sufficient capital:
Commit across vintages rather than in one year, since vintage effects are large and not predictable in advance.
Spread across strategies — credit and equity behave differently, and real assets differently again.
Use enough managers that no single one determines the outcome.
Consider secondaries for immediate diversified exposure and a shortened J-curve, understanding that the buyer pays for maturity someone else funded.
For investors without that capital, the honest options are:
Registered diversified vehicles — interval funds, non-traded BDCs, non-traded REITs — which deliver diversified exposure at low minimums with 1099 reporting. They cost more and offer capped liquidity, and for an investor who would otherwise hold two concentrated positions they are usually the better answer.
A smaller allocation, well diversified, rather than a larger one concentrated.
No allocation, which is a legitimate answer. Alternatives are not required, and an investor whose situation does not support illiquidity is not missing something by declining.
The measurement problem, revisited
One final practical point. Once an allocation exists, it will be reported at NAV, and that NAV will move less than public markets do.
This produces a well-known effect during public market declines: private allocations appear to hold value, and their share of the portfolio rises mechanically. Rebalancing back to target then means selling private assets — which cannot easily be done — or buying more public assets, which requires cash.
Investors should anticipate this rather than discover it. An allocation that is at target in calm conditions may be above target precisely when flexibility is most wanted, and the illiquid portion is the part that cannot be adjusted.
This guide is educational and general; it is not investment advice. Allocation decisions depend on individual circumstances; consult a qualified adviser.
Frequently Asked Questions
How much should an investor allocate to alternatives?
There is no general answer, and any specific percentage presented as a standard should be treated sceptically. The appropriate size depends on liquidity needs, time horizon, tax situation, access to competent managers, and the ability to diversify within the allocation. An investor who cannot diversify across managers and vintages is taking manager risk rather than asset class exposure.
Why do institutional allocations to alternatives look so large?
Large endowments and pensions have very long horizons, no requirement to liquidate on a schedule, dedicated staff to select and monitor managers, and access to managers that individuals generally cannot reach. Those conditions justify allocations that would be inappropriate for an investor lacking any of them.
Do optimisation models overstate the case for alternatives?
They can. Private market valuations are periodic and appraisal-based, which reduces measured volatility and correlation relative to what the underlying economic exposure would show if marked continuously. Feeding those inputs into a mean-variance framework mechanically increases the recommended allocation.
What matters more than the allocation percentage?
Manager selection and diversification within the allocation. Dispersion between managers in private markets is wide, so the outcome of a private markets programme depends more on which managers were used, and how many, than on whether the allocation was a somewhat larger or smaller share of the portfolio.
Can an investor build a private markets programme with limited capital?
With difficulty in traditional funds, because minimums and the need to diversify across managers and vintages require substantial capital. Registered vehicles such as interval funds and non-traded BDCs provide diversified exposure at low minimums, which is a genuine solution with its own costs.
Sources
- Global Investment Performance Standards (GIPS) maintained by CFA Institute
- ASC 820, Fair Value Measurement
- Institutional Limited Partners Association (ILPA) reporting guidance
