Problems in benchmarking alternative investments

Problems in benchmarking alternative investments

Published September 28, 2026

Problems in benchmarking alternative investments

On this page

Learning Outcome Statement

Describe problems that arise in benchmarking alternative investments.

On the exam: You receive a fact pattern about an alternative investment and its proposed benchmark, then describe specific mismatches and explain how they distort performance evaluation.

A benchmark earns its place by matching the economic experience you want to judge, not merely by publishing a return. An alternative investment can look unusually smooth or strong because its assets are hard to trade, its values are estimated, its capital structure is levered, or its reported return is net of fees. Call the portfolio return $R_P$ and the benchmark return $R_B$, then compare them only after checking that exposures and reporting conventions line up.

The diagram traces how an asset's features reach the investor report and then meet a proposed benchmark. Follow that chain before judging the reported return: the valuation process can smooth observations, while leverage, liquidity, and fees shape what the investor actually experiences. A benchmark should represent those same conditions closely enough for the comparison to say something useful about performance.

TestAlternative investmentCandidate benchmark
StrategyPrivate credit, property, or hedge fundPublic index or peer set
RiskConcentrated or nonlinear exposureBroad listed-market exposure
LeverageFund-level borrowingDifferent borrowing profile
LiquidityLocked or limited dealingFrequent market trading
ValuationEstimated or stale marksObservable market prices
FeesReturn may be net of feesReturn may be gross of fees

Read across the rows. The benchmark must match the alternative investment's strategy, risk, leverage, liquidity, valuation basis, and fee basis. A mismatch can turn a return difference into a reporting difference.

The active return, $R_A$, is the arithmetic difference between the portfolio return and benchmark return.

$$ R_A = R_P - R_B $$

The portfolio return adds to $R_A$. The benchmark return subtracts from it. If $R_P$ is net of fees and $R_B$ is gross, the subtraction includes the fee gap. This measures the difference. It does not attribute the difference to a manager or appraise skill. A misspecified benchmark can corrupt both conclusions.

Worked example

You are handed a private real-estate fund report and a proposed public REIT total-return index. The fund uses property-level borrowing, permits limited withdrawals, values properties with infrequent manager estimates, and reports investor returns net of fees. You need to decide whether the index can support manager appraisal and describe the problems it creates.

Case factStated condition
Portfolio strategyPrivate property ownership
Portfolio riskConcentrated property exposure
Portfolio leverageProperty-level borrowing
Portfolio liquidityLimited withdrawals
Portfolio valuationInfrequent manager estimates
Portfolio feesInvestor return net of fees
BenchmarkUnlevered public REIT index
Benchmark riskBroad listed-market exposure
Benchmark pricingFrequent market prices
Benchmark feesIndex return gross of fees

Run those facts through this decision path. A path with any mismatch sends the benchmark for revision. First compare ownership and exposure, then check how borrowing changes sensitivity to market moves. Next ask whether withdrawal limits and infrequent estimates delay or smooth reported results. Finally, line up fee treatment; only then can you interpret the return difference as a meaningful comparison.

Loading diagram…
  1. Compare the fund's ownership and risk exposure with the index's ownership and risk exposure. You do this because a benchmark must represent the same economic experience before it can measure the manager. This produces a strategy and risk mismatch.
  2. Compare the fund's property-level borrowing with the index's unlevered exposure. You do this because different financing sensitivity can make borrowed risk look like manager skill. This produces a leverage mismatch.
  3. Compare the fund's withdrawal terms and valuation process with the index's market pricing. You do this because limited withdrawals and infrequent manager estimates can make losses appear later and volatility appear lower. This produces a liquidity and valuation warning.
  4. Compare the fund's net-of-fees return with the index's gross-of-fees return. You do this because the fee treatment must match before you interpret the return difference. This produces the fee component embedded in $R_A$.
  5. Reconcile the mismatches before appraising the manager. You do this because $R_A$ is an observed difference, not a direct measure of skill. This produces a benchmark decision rather than a manager-skill score.
MismatchReturn-gap componentAppraisal risk
StrategyDifferent economic exposureSelection skill overstated
LeverageDifferent financing sensitivityRisk taking looks skillful
Liquidity and valuationSmoothing or stale pricesRisk or timing looks better
Fee basisNet versus gross differenceFees look like manager failure

Reject the index as the sole appraisal benchmark, although you may retain it as market context rather than a like-for-like measure of manager skill.

Changed condition

Keep every fact above, but change the valuation convention. The fund now uses independent appraisals rather than manager estimates. Predict the direction of the appraisal problem before reading the table.

ComparisonBeforeAfter
Valuation processManager estimatesIndependent appraisals
Smoothing riskHigherLower
Benchmark decisionReject as sole benchmarkStill reject as sole benchmark

The valuation mismatch falls, so stale-price and smoothing concerns fall with it. The strategy, leverage, liquidity, and fee mismatches remain, so the index still cannot support a clean manager appraisal. If reported returns change, $R_A$ changes too, but arithmetic alone does not repair the benchmark. Before reading on, write the remaining mismatches.

Common error

You may accept a readily available public-market index or peer index as a valid benchmark because the index provider publishes a return. The choice feels right because you can subtract that public number easily, and the result appears objective. But the portfolio may differ from the index in strategy, leverage, liquidity, valuation frequency, composition, or fee basis, so even a strong active return cannot support a reliable appraisal.

Go deeper

When you compare a manager with a peer or database benchmark, you are comparing the manager with a sample assembled by a provider. That provider decides which funds enter, when their histories begin, and how their weights change. The common distortions are:

ProblemDatabase effectAppraisal consequence
Survivorship biasFailed funds disappearPeer returns look too high
Backfill biasSuccessful histories added laterEarly results look too strong
Selection biasManagers choose inclusionSample is not representative
OpacityHoldings and methods unclearExposures cannot be checked
Changing compositionMembers and weights changeComparisons lose continuity

Use a peer return only after you can verify its membership, history, holdings, and weights. If a provider removes failed funds or adds a fund after it has succeeded, the surviving record can make the manager look weak even when the manager matched the full opportunity set. If the peer list changes during your measurement period, a single reported return may combine different portfolios, so you cannot tell whether the difference came from the manager or the sample.

Illiquidity creates a second problem because the last observed price can remain unchanged when no trade occurs. A private fund marked quarterly may look less volatile than a public asset with the same economic exposure, simply because the private fund records fewer and smoother observations. When you compare those returns, the apparent gap may reflect pricing frequency rather than investment skill. A misspecified benchmark can also make you credit selection or timing for a difference caused by leverage, liquidity, valuation, or fees. You can calculate the attribution correctly and still reach an unreliable appraisal conclusion.

Check yourself

Practice

Which statement best explains why a private fund's reported volatility can appear artificially low relative to a public-market index?

Which statement best explains the bias combination created when a database removes failed funds and adds successful fund histories after performance is known?

Revision snapshot

PointCandidate statement
Formula$R_A=R_P-R_B$ measures arithmetic active return
Benchmark ruleMatch strategy, risk, leverage, liquidity, valuation, fee basis
Return qualityIlliquidity, stale pricing, smoothing, infrequent observations impair comparability
Peer dataSurvivorship, backfill, selection, opacity, changing composition distort comparisons
Convenient indexA published return does not make a valid benchmark
Exam conclusionBenchmark misspecification leaves attribution and appraisal unreliable, even when arithmetic is correct

Unlock the full lesson and continue reading

Signup to continue reading this lesson and unlimited access questions, flashcards, AI notes, and more

Scan to download app

Scan to download
UNLOCK FREE ACCESS

Practice Questions: Problems in benchmarking alternative investments

  1. 1. Meridian Asset Consulting is reviewing the annual performance of a US small-cap equity fund for a pension client. The fund is managed against a broad US total-market index, although its investment mandate permits only securities within the US small-cap opportunity set. The pension committee wants performance measurement to reflect the manager's investment decisions rather than the timing of contributions determined by the plan sponsor. The sponsor made one external contribution during the year. The fund was valued immediately before the contribution, and no other external cash flows occurred. Exhibit 1 provides all values needed to calculate the annual return. Returns are not affected by taxes, and the fund values are after transaction costs but before investment management fees. The consultant also conducts a sector attribution analysis using beginning-of-period weights. Exhibit 2 reports sector weights and returns for both the fund and its current benchmark. The consultant uses the Brinson–Fachler model, under which allocation is measured relative to the total benchmark return. Selection and interaction effects are reported separately. During a committee meeting, one trustee argues that the current broad-market index should be retained because it is widely recognized. Another trustee proposes using the median return of a database of small-cap managers. The consultant notes that the database median cannot be purchased or replicated and is affected by changes in the managers included in the database. The consultant instead considers an investable index representing the securities available under the fund mandate. Finally, the committee asks whether the manager has generated active return through security-specific skill after controlling for systematic factor exposures. Exhibit 3 reports the fund's annualized active return, tracking error, and residual risk. The consultant explains that the information ratio uses tracking error, whereas the appraisal ratio uses residual risk. All risk measures were estimated over the same period as the annualized active return.

    • 2. Northbridge Pension Plan is reviewing the performance of an externally managed domestic equity fund. The investment policy statement gives the manager discretion to select securities from the domestic mid-cap value universe. The manager is not responsible for determining the timing or amount of contributions and withdrawals initiated by the plan sponsor. Accordingly, the investment committee wants a return measure that isolates the manager's investment decisions from sponsor-controlled external cash flows. During the most recent quarter, the sponsor made a large contribution immediately after the portfolio was valued at the end of Month 1. Exhibit 1 provides the valuations and cash-flow timing. All portfolio values are stated after investment management fees except where otherwise indicated, and no other external cash flows occurred during the quarter. The committee asks its performance analyst to calculate the quarter's time-weighted return by valuing the portfolio at the time of the external cash flow and geometrically linking the resulting subperiod returns. The analyst also performs sector attribution. Exhibit 2 contains the complete portfolio and benchmark sector data; there were no other sectors. The analyst uses the Brinson–Fachler framework, under which each sector's allocation effect is calculated as the difference between portfolio and benchmark sector weights multiplied by the difference between that sector's benchmark return and the total benchmark return. Selection and interaction effects are calculated separately. The committee next reviews annualized statistics in Exhibit 3. One member states that the information ratio should use the fund's total standard deviation because this measure captures all uncertainty experienced by beneficiaries. The analyst responds that the appropriate denominator depends on whether performance is assessed relative to the risk-free asset or relative to the designated benchmark. Finally, the committee questions whether its current broad domestic equity benchmark remains appropriate. The manager's mandate has consistently required investment in domestic mid-cap value securities, and the manager has not used large-cap or growth securities as strategic allocations. Investable, rules-based broad-market, large-cap value, and mid-cap value indices are all available, transparent, and measurable. The committee wants the benchmark that provides the most valid assessment of decisions within the manager's assigned opportunity set.

      • 3. A portfolio earns a cumulative return of 21.00% over two years. The calculated annualized geometric return is closest to:

        • A. 9.53%
        • B. 10.00%
        • C. 10.50%
        Show answer

        Answer: B. 10.00%

        10.00% - A 21.00% cumulative return gives an ending wealth factor of 1.21 after two years. - The annualized geometric return is \(1.21^{1/2}-1=\) 10.00%, which compounds back to 21.00% over two years. 9.53% - This is the continuously compounded annual rate: \(\ln(1.21)/2 \approx 9.53\%\). - It is not the annualized geometric return, which must be expressed as an effective annual rate. 10.50% - Dividing the cumulative return equally between two years gives 21.00% ÷ 2 = 10.50%. - That simple average ignores compounding: \(1.105^2\) exceeds the given ending wealth factor of 1.21.

      • 4. An asset manager markets one pooled fund as “GIPS compliant,” although the manager does not claim firm-wide compliance. Which conclusion is most accurate?

        • A. The firm's compliance claim applies only to the pooled fund
        • B. The pooled fund may claim compliance independently of the firm
        • C. The firm must claim compliance on a firm-wide basis
        Show answer

        Answer: C. The firm must claim compliance on a firm-wide basis

        The firm must claim compliance on a firm-wide basis - Under the GIPS standards, the investment management firm—not an individual pooled fund—claims compliance. - Because the manager does not claim firm-wide compliance, it cannot market the pooled fund as “GIPS compliant.” The firm's compliance claim applies only to the pooled fund - A firm cannot limit a GIPS compliance claim to the pooled fund while leaving the rest of the firm outside the claim. - The manager’s fund-specific marketing statement therefore does not establish a valid compliance claim. The pooled fund may claim compliance independently of the firm - The pooled fund is a product managed by the firm, not an entity that can make an independent GIPS compliance claim. - Describing the fund itself as “GIPS compliant” does not substitute for compliance claimed by its manager.

      • 5. An evaluator should most likely emphasize a time-weighted return when which condition applies?

        • A. External cash flows are controlled by the client
        • B. Portfolio valuations are available only at year-end
        • C. The manager determines all contribution dates
        Show answer

        Answer: A. External cash flows are controlled by the client

        External cash flows are controlled by the client - Time-weighted return removes the effect of external cash-flow timing by linking returns for the periods between those flows. - Because the client, not the manager, controls the flows, this measure better isolates the manager’s investment performance. Portfolio valuations are available only at year-end - Accurate time-weighted returns generally require valuations at external cash-flow dates to calculate returns for the periods between flows. - Year-end-only valuations do not establish when cash flows occurred or allow their timing effects to be removed accurately. The manager determines all contribution dates - When the manager controls contribution timing, that timing is part of the manager’s decisions rather than a client-imposed effect to remove. - A money-weighted return is more relevant because it reflects the impact of those contribution decisions on the investor’s return.

      • 6. A firm has a discretionary, fee-paying portfolio that has not been included in any of its composites. Under the GIPS standards, the firm should most likely:

        • A. Include the portfolio in at least one composite
        • B. Exclude the portfolio if it is immaterial to firm assets
        • C. Report the portfolio's performance separately outside any composite
        Show answer

        Answer: A. Include the portfolio in at least one composite

        Include the portfolio in at least one composite - GIPS requires every actual, fee-paying discretionary portfolio to be included in at least one composite. - Including this portfolio prevents the firm from selecting only favorable results for its composite performance. Exclude the portfolio if it is immaterial to firm assets - Immateriality does not exempt an actual, fee-paying discretionary portfolio from composite inclusion. - The requirement applies even if the portfolio represents a small share of firm assets. Report the portfolio's performance separately outside any composite - Separate reporting does not satisfy the requirement to include this portfolio in at least one composite. - Its performance may be reported separately, but that does not replace composite inclusion.

      • 7. A portfolio earned a cumulative time-weighted return of 12.00% over 18 months. The calculated annualized return is closest to:

        • A. 7.85%
        • B. 8.00%
        • C. 12.00%
        Show answer

        Answer: A. 7.85%

        7.85% - Annualize the 18-month cumulative return by compounding over 12 months: \((1+0.12)^{12/18}-1 \approx 7.85\%\). - The exponent 12/18 converts the 18-month growth factor into a one-year growth factor. 8.00% - Multiplying \(12.00\%\times12/18\) produces 8.00%. - That arithmetic annualization ignores the compounding required for a time-weighted return. 12.00% - 12.00% is the cumulative return over the full 18 months, not an annual return. - Reporting it unchanged omits the 12/18 adjustment for the measurement period.

      Flashcards: Problems in benchmarking alternative investments

      1. 1. Validity of Using Manager Universes as Benchmarks

        Show answer

        Finishing in the top half of all managers in the same style might be a good performance objective, it has a number of drawbacks as a benchmark: - Apart from being measurable it fails all of the other properties of a valid benchmark - Have to rely on compiler's classifications - Subject to survivor bias

      2. 2. Benchmark Properties

        Show answer

        - Specified in advance - Appropriate - Measurable - Unambiguous with clearly defined securities and weights - Reflective of the manager's current investment opinions - Accountable: manager and client agree it is applicable - Investable SAMURAI

      3. 3. The M2 Measure

        Show answer

        Average incremental return over a market index of a hypothetical portfolio that is created by combining the account with borrowing or lending at Rf. Measure the value added or lost relative to the market if the portfolio had the same risk

      4. 4. Hedge Fund Index Issues

        Show answer

        - Biases in index creation - Value weighted often become skewed to best performer - Relevance of past data - Survivorship bias - Stale price bias - Backfill bias

      5. 5. Information Ratio

      6. 6. Performance Attribution

        Show answer

        - Macro Attribution - Conducted at the fund sponsor level - Micro Attribution - Carried out at the investment manager level

      7. 7. Fixed-Income Attribution

        Show answer

        Factors that contribute to the total return of a fixed income portfolio are the External IR Effect and the Management Effect. External IR Effect - Is estimated using term structure analysis - Manager has no control over external IR environment - Can be seperate into expected (forward curve) and unexpected (what actually happened) Management Effect - Interest rate management effect - Sector/quality management effect - Security-selection effect - Trading effect

      8. 8. Portfolio Return Components

        Show answer

        A portfolio can be broken up into three components: Market, Style and Active Management P = M + S + A S = B - M\ A = P - B

      9. 9. Performance Data Quality Issues

        Show answer

        - Price of illiquid and infrequently priced assets generally not reliable in nature - Matrix pricing often used for thinly traded fixed income (price based on similar) - Account should be valued on trade date rather than settlement

      Everything you need for CFA prep

      Get full Oncourse access with lessons, practice questions, flashcards and AI study tools.

      GET STARTED FOR FREE
      75% OFFLimited time offer
      GET 75% OFF