Aggregate positions by economic issuer, connected group, factor, currency, and other relevant dimensions, using look-through holdings where available. Report both exposure weights and scenario loss contributions. Keep overlapping classifications separate, disclose incomplete look-through, and test targeted disturbances without treating them as predictions.
Count economic dependencies, not account rows
Several securities can represent claims on the same business, while several funds can own the same securities. A portfolio containing many tickers may therefore retain substantial dependence on one issuer or industry. Begin by identifying the economic entities behind instrument names. Distinguish the legal issuer, parent group, guarantor, derivative counterparty, and underlying reference asset where these are different entities.
The SEC's diversification guidance notes that narrowly focused mutual funds or ETFs do not necessarily provide broad diversification. For stress analysis, the relevant question is what common disturbance could affect the holdings together. That disturbance may be an issuer failure, a sector earnings shock, a funding interruption, or a shared factor move. Each requires an exposure view suited to the mechanism.
Build separate views of concentration
Capital concentration measures how much portfolio value is assigned to a holding or group. Sensitivity concentration measures how much a factor affects value. Loss concentration measures which positions contribute most under a specified scenario. A large cash holding can dominate capital weight while a smaller derivative dominates a particular sensitivity. None of these measures should be substituted silently for another.
Use issuer, sector, country, currency, maturity, and counterparty views as complementary lenses. These categories overlap. A bond can simultaneously belong to a bank issuer, a financial sector, a country, and a currency; adding all four category totals would count the same position repeatedly. Aggregate within a defined classification and reconcile each view independently to its stated exposure perimeter.
Look through pooled holdings with dated evidence
A fund's latest disclosed holdings may lag the portfolio valuation date. Preserve the disclosure date, coverage percentage, and any cash or derivative treatment. Multiply the investor's fund value by each disclosed underlying weight, while keeping unknown or undisclosed holdings visible. Do not assume that yesterday's fact sheet establishes today's exact exposure, especially for an actively traded or leveraged vehicle.
Choose either the wrapper or the underlying positions for a given economic-value aggregation. If both appear as assets in the same total, the fund is counted twice. Retain the wrapper as a parent reference for reconciliation, with children representing its economic allocation. When the disclosed weights do not sum to the expected amount, investigate rather than automatically rescaling away a potentially important gap.
Work a hypothetical hidden-overlap example
Consider a synthetic USD 1,000,000 portfolio with USD 100,000 held directly in Issuer Q, USD 300,000 in Fund A, USD 200,000 in Fund B, and USD 400,000 elsewhere. Assume Fund A has a 40% exposure to Q and Fund B has 25%, with all remaining holdings unrelated for this isolated example. Look-through adds USD 120,000 and USD 50,000 to the direct exposure.
The combined Q exposure is USD 270,000, or 27% of portfolio value. An illustrative 30% fall in Q, with all other assets unchanged and no fund-level complications, produces USD 81,000 of loss. Looking only at the direct position would show USD 30,000 and miss USD 51,000. These invented numbers demonstrate overlap accounting, not a prediction about any issuer or an assumed historical loss.
| Hypothetical route to Q | Underlying exposure USD | Loss at -30% USD |
|---|---|---|
| Direct holding | 100,000 | 30,000 |
| Fund A: 300,000 x 40% | 120,000 | 36,000 |
| Fund B: 200,000 x 25% | 50,000 | 15,000 |
| Combined | 270,000 | 81,000 |
Investigate connected groups and common drivers
The Basel Committee's large-exposures framework discusses single-name concentration and connected counterparties. Its banking rules are not portfolio limits adopted by this guide. The transferable analytical idea is to examine whether apparently separate claims can fail through the same dependency. Common ownership, guarantees, financing providers, or a major shared customer can matter even when instrument identifiers are different.
Factor concentration requires a separate inquiry. Several unrelated companies may depend on the same commodity price, borrowing condition, or consumer demand source. A collection of sovereign and corporate bonds can share long-duration exposure across many issuers. Use factor mapping to supplement issuer aggregation, and make the distinction clear: common sensitivity is not the same as legal connectedness or common default exposure.
Compare weights with scenario loss contributions
A position's share of total stressed loss depends on both its size and its scenario response. Show monetary contributions alongside percentage shares. When gains elsewhere offset losses, a contributor's share of net loss can exceed 100%; that is not necessarily an arithmetic error. A separate gross-loss view can make concentration easier to interpret when the net portfolio result is close to zero.
Test whether rankings persist across distinct scenarios. The largest contributor in an issuer shock may differ from the largest in a rate or currency shock. Avoid using a single loss ranking as a universal concentration score. Diversification in one mechanism does not guarantee diversification in another, and an apparent offset should be reviewed for basis, maturity, and counterparty differences before being treated as dependable.
Expose uncertainty instead of manufacturing precision
Maintain an unknown bucket for incomplete fund holdings, unidentified parents, and exposures that cannot be mapped credibly. Report its size and the mechanisms it could conceal. Where a broad proxy is used, label the result as estimated look-through rather than verified holdings. A detailed percentage with an old disclosure date should not look more authoritative than a plainly stated limitation.
The final analysis should distinguish observed concentration, modeled factor dependence, and unresolved overlap. Keep classification changes versioned so a falling concentration measure cannot be mistaken for a genuine portfolio change when it merely reflects a remapping. These diagnostics support understanding of conditional vulnerability. They do not prescribe diversification targets, recommend transactions, or claim conformity with banking exposure limits or other regulatory requirements.
Review checkpoints
- Many tickers can conceal a small number of economic dependencies.
- Look through funds without counting both wrappers and underlying values.
- Keep capital, sensitivity, and scenario-loss concentrations distinct.
Source notes
Continue the analysis
Mapping Portfolio Holdings to Risk Factors
Create an auditable bridge from securities and contracts to signed factor exposures, consistent units, and reconciled scenario results.
Coverage and Missing Positions in a Stress Test
Measure what a stress calculation includes, expose unsupported holdings, and interpret partial losses without silently treating missing risk as zero.
Designing Equity Shocks for a Portfolio
Build equity stresses with market, sector, and issuer components while controlling overlap, beta uncertainty, and nonlinear exposures.
Explain a scenario loss without inventing a unique cause
Reconcile position and factor contributions, expose interactions and ordering choices, and distinguish attribution from causal explanation.
First edition, 16 October 2026. Hypothetical values are teaching inputs, not observed portfolio or provider results.