Two exchange-traded funds can have different names, issuers and index providers and still own many of the same companies. Holdings overlap measures how much of one fund's portfolio is also in another's. It is a way to see, before comparing anything else, whether two funds are really different.
How overlap is measured
Every fund reports its positions and how much of the fund each one makes up, as a percent of net assets. Overlap is worked out from those weights in two ways.
Weighted overlap is the main figure. For every security both funds hold, take the smaller of its two weights, then add those up:
Weighted overlap = sum over common holdings of min(weight in fund A, weight in fund B)
The smaller weight is used because that is the part of the position the two funds genuinely share. If one fund holds a company at 6% and the other at 4%, they share 4 percentage points of exposure to it; the extra 2 points belong to the first fund alone. The result runs from 0% (no securities in common) to 100% (the same securities at the same weights). Because it uses each fund's own weights, it is the same whichever fund you start from.
Count-based shares look at the number of holdings rather than their size: how many of fund A's holdings fund B also holds, as a share of fund A's holdings, and the same the other way round. These two shares are usually different, because a small fund's holdings can all sit inside a large fund's list without the large fund being mostly the same as the small one.
To know that two lines refer to the same security, holdings are matched by identifier. The ISIN (International Securities Identification Number) is tried first, then the CUSIP, a nine-character code used for US and Canadian securities (a US ISIN contains the CUSIP). Only as a last resort, when one fund gives no identifier, are holdings matched by issuer name after removing punctuation and words such as "Inc" or "Ltd". Name matching is not used for bonds, because one issuer can have dozens of different bonds.
A worked example
The two funds below are made up for illustration. Each has six holdings and the weights add to 100%.
| Holding | Fund A | Fund B | Smaller weight |
|---|---|---|---|
| Alpha Corp | 25% | 10% | 10% |
| Beta Ltd | 20% | 30% | 20% |
| Gamma Inc | 20% | 5% | 5% |
| Delta plc | 15% | not held | |
| Epsilon SA | 10% | not held | |
| Zeta Co | 10% | not held | |
| Eta Ltd | not held | 25% | |
| Theta Inc | not held | 20% | |
| Iota Corp | not held | 10% |
Three securities are held by both funds: Alpha, Beta and Gamma.
- Weighted overlap = 10% + 20% + 5% = 35%.
- Count-based shares: 3 of Fund A's 6 holdings (50%) are in Fund B, and 3 of Fund B's 6 holdings (50%) are in Fund A.
- By weight, the shared securities make up 25% + 20% + 20% = 65% of Fund A but only 10% + 30% + 5% = 45% of Fund B.
The count-based shares say the funds are half the same. The weighted overlap gives a fuller picture: only about a third of the money in each fund is in shared exposure, because the funds weight the common holdings very differently. Fund A has a quarter of its assets in Alpha Corp; Fund B has a tenth.
Why it matters for diversification
Diversification works by spreading money across holdings that do not all rise and fall together. Two funds with high overlap own much the same companies in similar proportions, so their returns tend to behave alike. Holding both may look like spreading money across two products while adding little that is new.
Overlap also shows what you pay for. If two funds overlap heavily, the shared part of the portfolio carries two sets of fees, and the funds can be compared on cost, size and distributions as close substitutes. A low overlap shows the funds give genuinely different exposure, and then other differences, such as the index followed or the countries covered, explain what each adds.
You can measure the overlap between up to four funds with the ETF comparison tool.
Limits of the measure
Overlap is a useful first check, but it is narrow, and the figure is only as good as the holdings behind it.
- It is a snapshot. On this site holdings come from each US fund's latest public Form N-PORT report. Funds publish the report for the last month of each quarter, about 60 days after that month ends, so the holdings can be several months old. Two funds' reports may also be for different month ends.
- Weights drift. Prices move and funds trade every day, so today's weights differ from the reported ones. Index funds drift slowly; actively managed funds can change a lot between reports.
- Some positions are left out. Here, cash and money market vehicles, repurchase agreements, derivative contracts and short positions are excluded, because their reported value is not comparable exposure.
- Large funds may be incomplete. For a fund with more than 2,000 positions only the largest 2,000 are stored, so the figure shown is a lower bound and the page says how much higher it could be.
- Leveraged, inverse and other derivative-based funds get much of their exposure from swaps and futures, which are not counted, so their overlap with ordinary funds understates how closely they track the same market.
- Name matching can be wrong. It is used only when an identifier is missing, and the comparison page says how many holdings were matched each way.
- Low overlap does not mean low correlation. Two large-company growth funds that hold different technology companies can show little overlap yet move together, because their holdings respond to the same things. Overlap is about what is held, not how it behaves.
- It says nothing about fees, tax, currency or risk. Two funds with identical holdings can differ in cost, in how their distributions are taxed, in whether they hedge currency, and in how closely they track their index.
- ASX funds are not covered. ASX-quoted funds do not publish a standard machine-readable holdings report like the US Form N-PORT, so overlap is not available for them here.
Overlap is general information about how two portfolios compare. It does not take your circumstances into account and is not a recommendation about any fund.