True Exposure Decomposition
Break every position into underlying bets before calling a portfolio diversified
- Difficulty
- Moderate
- Time to result
- ~days to results
- Steps
- 6
- Confidence
- 93%
A portfolio can contain many trades while remaining exposed to a single underlying thesis. The method begins by listing positions and describing what must happen for each to work. Each position is then translated into its principal economic drivers, such as dollar strength, falling rates, credit expansion, or rising growth. Positions sharing a driver are grouped and treated as one combined bet for concentration analysis. The portfolio is then stressed against failure of its dominant thesis, including behavior outside the assumptions of mathematical models. The result is a truer picture of diversification based on risk sources rather than ticker symbols, asset-class labels, or the number of positions held.
Origin
Extracted from Masters in Business, where Sherman described the lesson he wished he had understood earlier about imperfect models and hidden portfolio correlation.
Core principles
- 01Models are tools rather than perfect answers
- 02Position count does not equal diversification
- 03Different trades can express the same underlying thesis
- 04Uncontrolled events will exceed model assumptions
- 05Diversify by economic exposure, not security label
How to run it
- 1
Inventory the Positions
List every security, derivative, fund, and tactical trade in the portfolio. Include indirect and leveraged exposures.
Pro tip Use current economic exposure rather than original purchase value.
Watch out Do not omit hedges or small derivatives that can become material under stress.
- 2
State Each Thesis
Write the market condition or causal belief that makes each position profitable. Make the thesis specific enough to test.
Pro tip Phrase the thesis independently of the security name.
Watch out Vague labels such as diversification or quality conceal the actual bet.
- 3
Identify Risk Drivers
Translate each position into major drivers such as rates, currency, growth, inflation, volatility, or credit. Note directional and nonlinear exposures.
Pro tip Use scenario analysis alongside historical correlation.
Watch out Historical correlations may change abruptly during a crisis.
- 4
Group Equivalent Bets
Combine positions that depend on the same driver even when they belong to different asset classes. Calculate the portfolio's aggregate exposure to each thesis.
Pro tip Ask what would hurt several positions at the same time.
Watch out Counting securities rather than shared drivers exaggerates diversification.
- 5
Stress the Dominant Thesis
Model what happens if the portfolio's largest shared assumption fails. Include outcomes beyond the normal range of the model.
Pro tip Use both plausible and severe scenarios.
Watch out A model's inability to represent an event does not make the event impossible.
- 6
Rebalance by Exposure
Reduce, hedge, or offset unintended concentrations until the portfolio contains genuinely different sources of risk and return. Recheck exposures as markets and position sensitivities change.
Pro tip Set concentration limits for economic drivers as well as individual holdings.
Watch out A hedge may fail if it depends on the same liquidity or funding conditions as the original position.
In the wild
An investor owns several currency, commodity, equity, and rate positions and believes the portfolio is diversified because the tickers differ. Decomposition shows that each trade benefits primarily from a stronger US dollar. The positions are therefore one concentrated macro thesis rather than several independent bets.
→ The investor can reduce duplicated exposure and introduce positions driven by genuinely different conditions.
Common mistakes
Counting Positions as Diversification
Several securities can all fail together when they depend on the same economic driver.
Treating Models as Complete Answers
Models simplify reality and cannot eliminate events or behavior outside their assumptions.
Using Only Historical Correlations
Relationships observed during normal markets can converge sharply when liquidity disappears or a common thesis breaks.
Is it for you?
Best for
It is best for investors holding multiple securities, strategies, or asset classes that may share correlated risk drivers.
Not ideal for
It is not ideal as a substitute for security-level due diligence or liquidity analysis.
From the transcript
“mathematical models are not perfect. They are tools. They're not that albeit end game to everything.”
“the thesis you have and you realize all your bets are correlated, they're all the same thesis.”
“sometimes you think you talk yourself and this idea that you have all these trades on but they're all just being long the dollar right…”
From the episode
Jeffrey Sherman on Trading in Academics for Fund Management
Jeffrey Sherman