Module XX·Article IV·~5 min read
Performance Attribution Reporting
Documentation and Communication
Turn this article into a podcast
Pick voices, format, length — AI generates the audio
Performance Attribution Reporting: the anatomy of return
Performance Attribution is the decomposition of the portfolio’s total return into components for understanding the sources of alpha and beta. Attribution answers the questions: “Where did the return come from?”, “Which decisions were successful?”, “What can be improved?”. This is a key tool for assessing skill vs luck and for improving the investment process.
Philosophy of Attribution Analysis
- Accountability — understanding the sources of successes and failures
- Process Improvement — identifying systematic errors
- Fee Justification — demonstrating value added for clients
- Resource Allocation — directing the team’s efforts to effective strategies
- Risk-Adjusted Thinking — understanding which risks are compensated by the market
Brinson-Fachler Attribution Model
The classic Brinson-Fachler model splits the excess return into three components:
| Component | Formula | Interpretation |
|---|---|---|
| Allocation Effect | $(W_p - W_b) \times (R_b - R_{total_benchmark})$ | Added value from sector over/underweighting |
| Selection Effect | $W_b \times (R_p - R_b)$ | Added value from security selection within sectors |
| Interaction Effect | $(W_p - W_b) \times (R_p - R_b)$ | Combined effect of allocation and selection |
Where:
$W_p$ = portfolio weight, $W_b$ = benchmark weight,
$R_p$ = portfolio return, $R_b$ = benchmark return
Example Brinson Attribution
| Sector | Portfolio Weight | Benchmark Weight | Portfolio Return | Benchmark Return | Allocation | Selection | Interaction | Total |
|---|---|---|---|---|---|---|---|---|
| Technology | 35% | 25% | 18% | 15% | +0.5% | +0.75% | +0.3% | +1.55% |
| Healthcare | 15% | 15% | 8% | 10% | 0% | -0.3% | 0% | -0.3% |
| Financials | 20% | 25% | 12% | 11% | -0.25% | +0.25% | -0.05% | -0.05% |
| Energy | 5% | 10% | -5% | -8% | +0.4% | +0.3% | -0.15% | +0.55% |
| Other | 25% | 25% | 6% | 6% | 0% | 0% | 0% | 0% |
| TOTAL | 100% | 100% | 11.2% | 9.5% | +0.65% | +1.0% | +0.1% | +1.75% |
Fixed Income Attribution
For fixed income portfolios, attribution is more complex and includes:
| Component | Description | Example |
|---|---|---|
| Income Return | Income from coupons | +3.5% |
| Treasury Effect | Impact of changes in risk-free rates | -1.2% |
| Spread Effect | Impact of changes in credit spreads | +0.8% |
| Currency Effect | Currency revaluation | -0.3% |
| Selection Effect | Selection of individual issuers | +0.4% |
| Total Return | +3.2% |
Factor Attribution
The modern approach is the decomposition of return by risk factors:
| Factor | Exposure (Beta) | Factor Return | Contribution |
|---|---|---|---|
| Market (MKT) | 1.05 | 8.0% | +8.4% |
| Size (SMB) | 0.15 | 2.0% | +0.3% |
| Value (HML) | -0.20 | -3.0% | +0.6% |
| Momentum (MOM) | 0.25 | 5.0% | +1.25% |
| Quality (QMJ) | 0.30 | 4.0% | +1.2% |
| Residual (Alpha) | — | — | +0.45% |
| Total | +12.2% |
Multi-Period Attribution
For analysis over long periods, linking methods are used:
- Carino Method — geometric linking of periods
- Menchero Method — preserves the additivity of attribution effects
- GRAP (Geometric Return Attribution Program) — GIPS-compliant standard
Attribution Report: Structure
-
Executive Summary
- Total excess return for the period
- Top 3 contributors and detractors
- Key insights in 3–5 bullet points
-
Asset Allocation Attribution
- Table by asset classes
- Waterfall contributions chart
- Commentary on major allocation decisions
-
Security Selection Attribution
- Top 10 contributors (individual securities)
- Bottom 10 detractors
- Sector-by-sector selection analysis
-
Factor Attribution
- Factor exposures vs benchmark
- Factor contributions
- Residual alpha (unexplained returns)
-
Currency Attribution
- Currency exposures
- Hedging effectiveness
- Currency contribution to total return
-
Time Series Analysis
- Rolling 12-month attribution
- Consistency of alpha sources
- Hit rate analysis
Hit Rate Analysis
| Decision Type | Decisions Made | Positive Outcome | Hit Rate | Average Win | Average Loss |
|---|---|---|---|---|---|
| Sector Allocation | 48 | 31 | 65% | +0.3% | -0.2% |
| Stock Selection | 156 | 89 | 57% | +1.2% | -0.8% |
| Duration Calls | 12 | 7 | 58% | +0.5% | -0.4% |
| Currency Hedging | 24 | 15 | 63% | +0.2% | -0.15% |
Attribution Tools
| Platform | Strengths | Price |
|---|---|---|
| Bloomberg PORT | Data integration, multi-asset | $$ |
| FactSet | Fixed income attribution, flexibility | $$ |
| MSCI Barra | Factor models, risk attribution | $$ |
| StatPro | GIPS compliance, multi-currency | $ |
| Style Analytics | Peer comparison, style analysis | $ |
Best Practices in Attribution Reporting
- Consistency — use the same method each period
- Granularity — sufficient detail for action items
- Linking — correct linking of multi-period attribution
- Benchmark Alignment — attribution relative to the correct benchmark
- Currency Separation — separating currency effects from asset returns
- Transaction Costs — accounting for impact of trading in selection effect
- Residual Monitoring — analysis of unexplained returns
- Forward-Looking Use — applying insights to improve the process
Typical Errors in Attribution
- Wrong benchmark — attribution to an inappropriate benchmark is meaningless
- Ignoring costs — gross attribution hides transaction cost drag
- Over-interpretation — noise mistaken for signal
- Survivorship bias — exclusion of sold positions from analysis
- Timing issues — improper timing of transaction recognition
- No action — attribution without subsequent process changes
CIO Recommendations
- Attribution is a mirror — use it for honest self-assessment, not for excuses
- Focus on process — good decisions > good outcomes in the short term
- Share openly — the team should see the full picture
- Learn from detractors — losses teach more than gains
- Benchmark matters — selection of the benchmark determines everything else
§ Act · what next