Brinson Attribution and a Performance Fee Claim
A UK equity fund manager reports a return of 11.9% for the year against a benchmark return of 10.2%, an excess return of 1.7%. The manager claims this demonstrates stock-selection skill. The fund's sector weights and returns are: Financials benchmark 25%, fund 15%, benchmark return 9.0%, fund 9.0%; Industrials 15%/20%, 13.0%/13.0%; Consumer staples 15%/10%, 7.0%/5.0%; Energy 10%/15%, 19.0%/19.0%; Health care 12%/10%, 10.0%/7.0%; Technology 8%/15%, 17.5%/23.0%; Utilities 5%/5%, 4.0%/4.0%; Real estate 5%/5%, -3.0%/-3.0%; Other 5%/5%, 8.0%/8.0%. The fund has no currency exposure. The manager is paid a performance fee above benchmark. The trustee asks for an attribution analysis and an assessment of whether the fee is justified.
1. Calculate the Brinson attribution of the excess return into allocation, selection and interaction effects. Show your workings.
Read the worked answer
All effects are in basis points of fund value. Benchmark return is 10.20%. Allocation = Σ(wf - wb)(rb - 10.20%). Financials: (-10%) × (9.0 - 10.2) = +12.0bp. Industrials: (+5%) × (13.0 - 10.2) = +14.0bp. Staples: (-5%) × (7.0 - 10.2) = +16.0bp. Energy: (+5%) × (19.0 - 10.2) = +44.0bp. Health care: (-2%) × (10.0 - 10.2) = +0.4bp. Technology: (+7%) × (17.5 - 10.2) = +51.1bp. Utilities, real estate and other have zero weight differences. Total allocation = +137.5bp.
Selection = Σ wb(rf - rb). Staples: 15% × (5.0 - 7.0) = -30.0bp. Health care: 12% × (7.0 - 10.0) = -36.0bp. Technology: 8% × (23.0 - 17.5) = +44.0bp. Other sectors have zero stock-selection difference. Total selection = -22.0bp.
Interaction = Σ(wf - wb)(rf - rb). Staples: (-5%) × (-2.0%) = +10.0bp. Health care: (-2%) × (-3.0%) = +6.0bp. Technology: (+7%) × (+5.5%) = +38.5bp. Total interaction = +54.5bp. Total excess return = 137.5 - 22.0 + 54.5 = 170bp, matching the reported 1.7%.
2. Assess the manager's claim of stock-selection skill, including the limitations of the single-period attribution.
Read the worked answer
The attribution does not support the claim. Selection is -22bp, so stock selection within sectors detracted slightly. If interaction is assigned to selection, selection is still only +32.5bp, less than one-fifth of the excess return. The outperformance came overwhelmingly from allocation: overweight energy and technology and underweight financials and staples added 137.5bp. The manager's sector bets, not stock picking, drove the result.
There is a nuance: if the manager is bottom-up, sector weights are a by-product of stock selection, so the allocation effect may partly reflect stock picking. But the data show that within most sectors the fund matched the benchmark return exactly or underperformed. The trustee should not pay a stock-selection fee on this evidence. The result is also a single period, before risk adjustment, fees and transaction costs. It may be luck rather than skill. A proper assessment needs three to five years of attribution, information ratios, factor exposures and style consistency.
3. Recommend how the trustee should structure the manager's performance fee and monitoring.
Read the worked answer
The fee should reward excess return over a clearly specified benchmark, but only after risk and over a longer period. A suitable structure would use a three-year rolling measurement period, a high-water mark, a cap on the fee, and symmetric malus or clawback for underperformance. The fee should be linked to an information ratio or another risk-adjusted measure, not raw excess return. If the mandate is genuinely stock selection, the trustee should separate the allocation and selection effects and avoid paying for sector bets that could be replicated cheaply.
Monitoring should include quarterly attribution, risk exposures, sector concentrations and style drift. The trustee should review whether the manager's sector weights are deliberate top-down bets or a by-product of stock selection, and whether the benchmark is appropriate. The performance fee should not be paid on a single year of strong allocation. The trustee should also compare net-of-fee returns with peers and passive alternatives.