How to Compare Two Similar ETFs
Step 1: Are they the same exposure?
Before anything else, verify the funds are comparable at all. Two "world" funds may differ by emerging markets or small caps (Chapter 20); two "S&P 500" funds are genuinely identical; two "technology" funds may share 60% of holdings (Chapter 42). Check: index full name including suffixes (capped variants, Chapter 40), holdings count, and top-10 overlap. If exposures differ materially, you're choosing a strategy, and Chapter 20/45's logic applies before any of this chapter's.
Step 2: Disqualifiers first
Cheap eliminations that save the rest of the work:
- Can you buy it? PRIIPs (EU retail), wrapper eligibility (PEA, ISA platform lists), broker availability by ISIN.
- Tax status in your country: UK reporting-fund status, German equity ratio for Teilfreistellung, Irish domicile versus alternatives, PFIC exposure for US taxpayers (Chapters 50–55).
- Share class: accumulating versus distributing, hedged versus unhedged (Chapters 9, 49). A "better" fund in the wrong class isn't better.
Anything failing here is out, regardless of its numbers.
Step 3: The decisive number
Multi-year tracking difference, three to five years, annualized, against the same benchmark variant (net-return for UCITS equity funds), with the year-by-year spread for consistency (Chapter 35). This is where the decision is actually made, and it is the step comparison sites omit.
Read it as: what did owning this fund cost against the index it promises, and how reliably. A fund at −0.05% with a −0.08 to −0.02 range beats one at −0.04% averaged from −0.30 to +0.20: the first is craft, the second is luck with a good mean.
Step 4: Explain the gap
Never accept a TD difference you can't attribute; unexplained edges tend to evaporate. The four usual explanations (Chapters 36, 50, 51, 11):
- Domicile: Irish versus Luxembourg on US-heavy exposure ≈ 0.2%/yr.
- Replication: synthetic's gross-dividend route on US indices ≈ another 0.15–0.20%.
- Securities lending: program size, revenue split, indemnification: worth 1–30bp by segment.
- Sampling and execution craft: relevant for broad or illiquid indices (Chapter 37).
If the cheaper-TER fund tracks worse, one of these explains it, and the explanation tells you whether the advantage is structural (persists) or circumstantial (may not).
Step 5: Viability
- Fund size: comfortably above closure-risk thresholds for its segment (Chapter 67); a €40m thematic fund and a €40m core fund carry different risks.
- Age and track record: enough history for step 3 to mean something.
- Provider commitment: is this a flagship range or a peripheral experiment? Providers rationalize peripheries.
Closure isn't catastrophic (you receive market value), but it forces a taxable sale at someone else's schedule: worth avoiding for a decades-long holding.
Step 6: Your own layer
The fund's numbers are only Layer 1 (Chapter 56). Now price your access:
- Spread and venue at your broker, at the hour you'd trade (Chapter 57).
- Savings-plan eligibility and fee if you invest monthly: frequently the largest single term for small savers, and frequently decisive.
- Currency line availability to avoid FX.
- Wrapper fit if it sits inside ISA/PEA/pension platforms with their own charges.
Step 7: Decide, and write one paragraph
Record: the two candidates, the deciding factor, the numbers as of a date, and what would change your mind. Reason: comparison tables reshuffle constantly, and without a written basis you will re-litigate this decision annually at a cost (Layer 2 plus tax) that exceeds anything the switch could earn. The paragraph is the mechanism that converts twenty minutes into a permanent decision.
Where the tie-breakers live
When steps 1 to 6 genuinely tie (it happens, and often with the largest flagship funds):
- Prefer the accumulating class in accumulation phase (Chapter 9), unless your tax regime says otherwise.
- Prefer the fund you can hold without FX and inside your savings plan.
- Prefer larger and older at equal quality.
- Prefer the simpler structure if you'd lose sleep over the other (physical over synthetic, low lending over aggressive lending): sleep has value, and Chapter 64 will show it has financial value.
- Then stop. Beyond this point you are optimizing noise (Chapter 56's 0.05% threshold), and the returns to further analysis are negative.
A worked example
Sofia compares two S&P 500 UCITS funds for a €60,000 long-term holding.
- Same exposure: identical index, no capping variants. ✓
- Disqualifiers: both Irish, both accumulating available, both on her broker, both plan-eligible. ✓
- TD (5-yr, net-return benchmark): Fund A −0.19%; Fund B −0.04%. TERs: A 0.05%, B 0.07%.
- Attribution: B is synthetic (gross-dividend route, Chapter 11); A is physical Irish (15% withholding drag). Structural, not luck.
- Viability: both multi-billion, 8+ years. ✓
- Her layer: identical spreads and venues; both free in her plan. Neutral.
- Decision, written: "Chose B (synthetic) on 5-yr TD −0.04% vs −0.19%, attributable to the US dividend-withholding route; read the swap checklist (Ch. 38): unfunded, daily reset, four counterparties, published collateral, and accepted. Would revisit if TD converges for two consecutive years or if the counterparty panel narrows to one."
Her sister, running the same seven steps, stops at step 4 and chooses A because she declines derivative exposure on principle. Both decisions are correct, both are written, and neither will be reopened by a comparison site's sort order.
Key takeaways
- Seven steps: same exposure → disqualifiers → multi-year TD on the same benchmark variant → attribute the gap → viability → your own cost layer → written decision.
- TD is the decision metric; TER is an input. The ranking reverses often enough that fee-sorted lists are actively misleading.
- Never accept an unexplained tracking advantage: attribute it to domicile, replication, lending or craft, and note whether it's structural or circumstantial.
- Your layer (spread, venue, plan eligibility, FX, wrapper) can outweigh everything at Layer 1, particularly for monthly savers.
- The written paragraph is the deliverable: it converts one afternoon of work into a decision you don't repeat, which is where most of the value actually comes from.
FAQ
How different does TD have to be before I switch existing holdings?
For a taxable account, the switch must overcome realized tax plus Layer 2 costs: usually meaning gaps below ~0.15%/yr aren't worth crystallizing. Direct future contributions to the better fund instead.
Both funds have positive tracking difference. Is one lying?
Probably neither: net-return benchmarks assume harsh withholding, and lending adds income (Chapters 36, 51). Check they use the same benchmark variant, then compare normally.
What if the funds track different indices from different providers?
Then you're comparing exposures first (Chapters 20, 39), and only after choosing the index does this chapter's procedure apply.
Is provider reputation worth anything in the comparison?
Indirectly: it correlates with reclaim competence, lending governance and range longevity. But it shows up in the numbers you're already reading, so use it as a tie-breaker, not a premium to pay.
Can I skip all this for a flagship fund?
Largely yes, and that's a legitimate use of your attention: the giants cluster within a few basis points. Run the full procedure once, on the fund that will hold most of your money, and default thereafter.
Sources & further reading
- Issuer tracking-difference disclosures and annual reports; independent UCITS TD aggregations (dated; this site's maintained table)
- Index variant definitions (price/gross/net) from MSCI, FTSE Russell, S&P DJI
- Provider documentation on domicile, replication and lending programs (Tier-1B, cross-checked)
- Broker fee schedules and savings-plan terms (your layer's evidence)
- This site's Chapters 9, 11, 20, 35–38, 40, 49–51, 56–57, 67