A
AcadiFi
ES
EquityResearch_Sam2026-04-06
cfaLevel IIIPerformance EvaluationPortfolio Management

How does the Brinson performance attribution model decompose portfolio returns into allocation, selection, and interaction effects?

I'm studying performance attribution for CFA Level III and the Brinson model keeps tripping me up. I understand it separates returns into allocation effect and selection effect, but some versions include an interaction effect while others fold it into selection. Can someone do a full numerical walkthrough so I can see exactly how each component is calculated?

149 upvotes
AcadiFi TeamVerified Expert
AcadiFi Certified Professional

The Brinson-Hood-Beebower (BHB) attribution model decomposes the active return (portfolio return minus benchmark return) into three components that explain where the outperformance or underperformance came from.

The Three Components:

  1. Allocation Effect — Did the manager add value by overweighting/underweighting sectors relative to the benchmark?
  2. Selection Effect — Did the manager add value by picking better/worse securities within each sector?
  3. Interaction Effect — The combined effect of simultaneously overweighting a sector AND picking better securities in it (or vice versa)

Formulas:

For each sector i:

  • Allocation_i = (w_p,i - w_b,i) × (R_b,i - R_b)
  • Selection_i = w_b,i × (R_p,i - R_b,i)
  • Interaction_i = (w_p,i - w_b,i) × (R_p,i - R_b,i)

Where:

  • w_p,i = portfolio weight in sector i
  • w_b,i = benchmark weight in sector i
  • R_p,i = portfolio return in sector i
  • R_b,i = benchmark return in sector i
  • R_b = total benchmark return

Total Active Return = Σ(Allocation_i) + Σ(Selection_i) + Σ(Interaction_i)

Worked Example — Clearwater Growth Fund vs. Russell 1000 Growth

SectorPortfolio WeightBenchmark WeightPortfolio ReturnBenchmark Return
Technology35%30%12.0%10.5%
Healthcare20%25%8.0%9.2%
Consumer Disc.25%20%7.5%6.0%
Financials20%25%5.0%4.8%

Step 1 — Total Benchmark Return:

R_b = 0.30(10.5%) + 0.25(9.2%) + 0.20(6.0%) + 0.25(4.8%) = 3.15% + 2.30% + 1.20% + 1.20% = 7.85%

Step 2 — Total Portfolio Return:

R_p = 0.35(12.0%) + 0.20(8.0%) + 0.25(7.5%) + 0.20(5.0%) = 4.20% + 1.60% + 1.875% + 1.00% = 8.675%

Active Return = 8.675% - 7.85% = 0.825%

Step 3 — Allocation Effect by Sector:

Sectorw_p - w_bR_b,i - R_bAllocation
Technology+5%10.5% - 7.85% = +2.65%+0.1325%
Healthcare-5%9.2% - 7.85% = +1.35%-0.0675%
Consumer Disc.+5%6.0% - 7.85% = -1.85%-0.0925%
Financials-5%4.8% - 7.85% = -3.05%+0.1525%
Total Allocation+0.125%

The positive allocation effect (+0.125%) means the manager added value through sector weighting — primarily by overweighting Technology (a strong sector) and underweighting Financials (a weak sector).

Step 4 — Selection Effect by Sector:

Sectorw_b,iR_p,i - R_b,iSelection
Technology30%+1.5%+0.450%
Healthcare25%-1.2%-0.300%
Consumer Disc.20%+1.5%+0.300%
Financials25%+0.2%+0.050%
Total Selection+0.500%

Step 5 — Interaction Effect by Sector:

Sectorw_p - w_bR_p,i - R_b,iInteraction
Technology+5%+1.5%+0.075%
Healthcare-5%-1.2%+0.060%
Consumer Disc.+5%+1.5%+0.075%
Financials-5%+0.2%-0.010%
Total Interaction+0.200%

Verification:

0.125% + 0.500% + 0.200% = 0.825% (matches the active return)

Loading diagram...

BHB vs. Brinson-Fachler:

Some references combine the interaction effect into the selection effect (the Brinson-Fachler variant). The CFA exam typically uses the three-component BHB version but may present either. Read the question carefully to determine which decomposition is expected.

Exam Tip: Always construct the full table sector by sector. The most common errors are: (1) using portfolio weights instead of benchmark weights for the selection effect, and (2) forgetting to subtract the total benchmark return in the allocation formula. Verify that your three components sum to the total active return.

Practice performance attribution in our CFA Level III question bank.

📊

Master Level III with our CFA Course

107 lessons · 200+ hours· Expert instruction

#brinson-attribution#allocation-effect#selection-effect#interaction-effect#performance