What it means
A fund that beat its benchmark by 2.6% has told you almost nothing about why. Attribution analysis takes that excess return apart and assigns each slice to a decision the manager actually made, so a board can see whether the gain came from being overweight in technology or from picking better technology stocks.
The standard framework splits excess return into allocation, selection and an interaction term. Allocation measures the value added by weighting sectors differently from the benchmark, selection measures the value added by holding better stocks within each sector, and interaction captures the overlap between the two.
It matters because investment committees and trustees are hiring judgement, not outcomes. A manager who beat the benchmark purely by being overweight in one sector that happened to rally has demonstrated a different, and less repeatable, capability than one who consistently picked the stronger names in every sector.
The same logic extends well outside fund management. Marketing teams attribute revenue growth to channels, sales leaders attribute variance to price, volume and mix, and finance teams use bridge analyses that are conceptually identical, splitting a movement into named, additive drivers.
The important nuance is that the components must add up to the actual excess return, and that the choice of benchmark drives everything. Change the benchmark and the same portfolio can move from skilled to lucky, which is why the benchmark is negotiated before a mandate begins rather than after the results arrive.
In practice
Real-world examples.
Example
A pension trustee board reviews a manager who beat the benchmark by 3.1% and finds that 2.9% came from a single overweight position in energy during an oil price spike. The board keeps the manager but caps future sector deviations, because the record shows a bet rather than a process.
Example
A multi-asset fund attributes a weak year to currency rather than to holdings, with unhedged exposure to a falling foreign currency costing 1.8% while stock selection added 0.6%. The result is a policy change requiring 50% of overseas equity exposure to be hedged.
Example
A retail bank applies the same technique to net interest income, splitting a $4,000,000 increase into $2,500,000 from higher volumes, $1,800,000 from wider margins and a negative $300,000 from a shift towards cheaper products. The board can then see that margin, not growth, carried the year.
Think of it
“Attribution analysis explains where your returns came from-which decisions added or subtracted value.
Formula
Calculation
Allocation Effect = (Portfolio Weight - Benchmark Weight) x (Benchmark Sector Return - Total Benchmark Return)
Selection Effect = Benchmark Weight x (Portfolio Sector Return - Benchmark Sector Return)
Interaction = (Portfolio Weight - Benchmark Weight) x (Portfolio Sector Return - Benchmark Sector Return)
Take a simple two-sector portfolio. The benchmark holds 50% technology returning 12% and 50% utilities returning 4%, so the total benchmark return is (0.50 x 12%) + (0.50 x 4%) = 8.0%. The portfolio holds 70% technology returning 13% and 30% utilities returning 5%, so its return is (0.70 x 13%) + (0.30 x 5%) = 9.1% + 1.5% = 10.6%. Excess return is 10.6% - 8.0% = 2.6%.
Allocation, technology = (0.70 - 0.50) x (12% - 8%) = 0.20 x 4% = 0.8%
Allocation, utilities = (0.30 - 0.50) x (4% - 8%) = -0.20 x -4% = 0.8%
Total allocation = 1.6%
Selection, technology = 0.50 x (13% - 12%) = 0.5%
Selection, utilities = 0.50 x (5% - 4%) = 0.5%
Total selection = 1.0%
Interaction, technology = 0.20 x 1% = 0.2%
Interaction, utilities = -0.20 x 1% = -0.2%
Total interaction = 0.0%
Adding the three components gives 1.6% + 1.0% + 0.0% = 2.6%, which exactly matches the excess return, so most of the outperformance came from sector weighting rather than stock picking.Case study
Seen in the real world.
This is an illustrative example featuring a fictional organisation. The Marden Foundation, an endowment, was delighted when its equity manager delivered 14.2% against a benchmark of 11.0%, and the investment committee prepared to increase the mandate. Before doing so, the committee commissioned an attribution analysis covering the previous four years.
The analysis showed that in three of those four years the entire excess return had come from allocation, and specifically from a persistent overweight in small companies. Selection had contributed a cumulative -0.4%, meaning that within each sector the manager's individual stock choices had been slightly worse than the index.
In this fictional case the committee's conclusion was uncomfortable but useful. It was paying an active stock picking fee for what was essentially a size tilt, and it replaced half the mandate with a low-cost small company index fund, keeping the manager for the portion where the committee genuinely wanted a discretionary view.
Watch out
Common mistakes.
- Judging a manager on total return alone, which rewards whoever happened to hold the fashionable sector rather than whoever made the better decisions.
- Running attribution against a benchmark that does not reflect the mandate, which produces effects that are really just a mismatch of definitions.
- Ignoring the interaction term or quietly folding it into selection, which flatters or penalises the stock picking record without saying so.
Questions
People also ask.
Does attribution analysis prove skill?
Not on its own, because a single period can be luck, so committees look for the same effect repeating across several years and market conditions.
Can it be used outside investing?
Yes, and the same additive logic underpins sales variance analysis, marketing channel attribution and profit bridges in management accounts.
What if the components do not add up to the excess return?
That normally signals a data problem such as mismatched valuation dates, unrecorded cash flows or transaction costs excluded from one side of the comparison.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%