Repository navigation
XtrackersMsciWorldXwr
Data as of 30 Sep 2026 for MSCI World index returns, 31 Aug 2026 for holdings, country, sector and ES3 returns. Not financial advice. Jev scores are model advice, not facts.
Verdict: a reasonable equity holding for a 5-year horizon (Jev 0.61), but a split on whether it should be the core holding (Jev 0.47). In the last 10 years of data, lump sum beat staged entry in every window tested. Staging is a regret-control choice, not a return choice. There is no valuation data, so "now is a good entry" is not supported (Jev 0.18).
The Xtrackers MSCI World UCITS ETF (XWR) began trading on SGX on 13 October 2026 and trades in SGD. It is a cross-listing of an existing Irish fund, not a new fund (Dollars and Sense). It tracks the MSCI Total Return Net World Index, which covers large and mid-cap companies in developed markets, about 85% of free-float market cap, with no emerging markets or small caps (Beansprout). It is physical and accumulating, with a current all-in fee (TER) of 0.12% and a maximum of 0.19% (Beansprout).
Proxy rule. XWR has no SGX history, and its ISIN and share class are not confirmed. Where XWR facts are missing, this memo uses the USD-base Xtrackers MSCI World UCITS ETF 1C (IE00BJ0KDQ92), launched 22 July 2014, physical optimised sampling, 1,454 holdings (justETF). Its tracking difference is about -0.05% a year since 2015 (Tracking Differences). Index returns come from MSCI itself, not the fund.
| Top 5 holdings (31 Aug 2026) | % |
|---|---|
| NVIDIA | 5.46 |
| Apple | 5.10 |
| Microsoft | 3.94 |
| Amazon | 2.80 |
| Alphabet A | 2.19 |
Top 10 holdings total 26.60%. By country: US 70.31%, Japan 5.73%, UK 3.67%, Canada 3.45%, other 16.84%. By sector: technology 35.49%, finance 18.21%, industrials 9.66%, healthcare 9.05%, other 27.59% (proxy fund, justETF).
| Statement | Evidence | Jev |
|---|---|---|
| Low cost and tax-efficient structure | TER 0.12% against 0.28% for the STI ETF; Irish domicile gives 15% US dividend withholding at fund level, not 30% (Beansprout) | 0.67 yes |
| Broad developed-market diversification | 1,454 holdings across 23 developed markets (justETF, MSCI) | 0.65 yes |
| Strong long-run index returns | 15.24%, 21.40%, 11.89%, 12.81% annualised for 1, 3, 5, 10 years to 30 Sep 2026 (MSCI) | 0.80 yes |
| Statement | Evidence | Jev |
|---|---|---|
| US large-cap tech concentration is material | US 70.31%, technology 35.49%, top 10 holdings 26.60% (justETF) | 0.80 yes |
| Equity drawdown of 20% or more is material | MSCI World fell 18.14% in 2022; worst drawdown 57.82% from Oct 2007 to Mar 2009 (MSCI) | 0.89 yes |
| Non-SGD currency exposure is material | Price is in SGD but NAV is in USD, and the holdings trade in USD, JPY, GBP, CAD and others; it is not SGD hedged (Phillip Nova) | 0.88 yes |
| New SGX line liquidity and spread risk is material | Two days of trading, no observed spreads, unproven liquidity (Beansprout) | 0.70 yes |
| Item | Detail | Source |
|---|---|---|
| TER | 0.12% current, up to 0.19% | Beansprout |
| Brokerage and spread | Not found. The SGX line has no observed spread. Endowus's 0.30% single-fund fee relates to funds, not SGX ETFs, and I did not confirm Endowus offers XWR | gap |
| Lot size | 1 share | Beansprout |
| SRS | Expected to be eligible from listing, subject to platform support | Beansprout |
| Tax layer | Treatment | Source |
|---|---|---|
| Singapore, individual | Foreign-sourced income received by resident individuals is exempt, except through a Singapore partnership | IRAS |
| Inside the fund | Irish domicile reduces US dividend withholding to 15% from 30% | Dollars and Sense |
| US estate tax | Irish-domiciled UCITS ETFs generally fall outside US estate tax because situs follows fund domicile. The sources are secondary and do not address Singapore residents | Taxes for Expats, StashAway |
| Item | Figure |
|---|---|
| Payout | None. The ETF is accumulating, so dividends are reinvested net of withholding taxes (Beansprout) |
| TER at 0.12% | about S$240 a year |
| TER at the 0.19% ceiling | about S$380 a year |
| Brokerage | Unknown |
Backward-looking, using the MSCI World USD net index (the fund's tracking difference is about -0.05% a year) and trailing annualised returns to 30 Sep 2026 (MSCI). Not a forecast. No brokerage, TER or SGD effects are deducted.
| Horizon | Annualised return | S$200,000 became |
|---|---|---|
| 1 year | 15.24% | S$230,480 |
| 3 years | 21.40% | S$357,838 |
| 5 years | 11.89% | S$350,741 |
| 10 years | 12.81% | S$667,584 |
Jev cannot run a backtest (0.04), so the arithmetic below uses MSCI World USD net calendar-year returns for 2016 to 2025 (MSCI): 7.51, 22.40, -8.71, 27.67, 15.90, 21.82, -18.14, 23.79, 18.67, 21.09. Jev judges each result. Staged cash earns nothing, which is a simplification. One index, 10 observations, hindsight only.
Lump sum vs equal annual tranches, S$200,000, windows ending 2025
| Window | Lump sum | Staged | Jev: staging beat lump sum |
|---|---|---|---|
| 3 years (from 2023) | S$355,766 | S$295,114 | 0.02 no |
| 5 years (from 2021) | S$354,777 | S$306,270 | same test |
| 10 years (from 2016) | S$630,639 | S$416,396 | same test |
Rolling windows. Across 8 three-year windows (starts 2016 to 2023) none were negative; the worst was 4.9% a year (start 2020). Across 6 five-year windows (starts 2016 to 2021) none were negative; the worst was 6.1% a year (start 2018). Jev: entry timing was decisive for a 5-year holder, 0.28 no. Staying invested mattered more.
Buying at the start of a loss year vs the start of the next year. Each row invests S$200,000 on 1 January of the entry year and holds to 31 December 2025. In each pair, the later entry skips the loss year and every later year is identical.
| Pair | Loss year skipped | Entry | Years held | Value at end 2025 | Annualised |
|---|---|---|---|---|---|
| 1 | 2018 (-8.71%) | Start of 2018, before the loss | 8 | S$479,237 | 11.54% |
| 1 | 2018 (-8.71%) | Start of 2019, after the loss | 7 | S$524,961 | 14.78% |
| 2 | 2022 (-18.14%) | Start of 2022, before the loss | 4 | S$291,230 | 9.85% |
| 2 | 2022 (-18.14%) | Start of 2023, after the loss | 3 | S$355,766 | 21.17% |
The gap in each pair (S$45,724 and S$64,536) is the loss year itself and nothing else. It shows that avoiding a loss year helps, which is true with hindsight. It is not a signal, because nobody knows at 1 January whether the year will be a loss year. Jev: usable entry-timing signal, 0.07 no.
Now. Strong recent returns (3-year 21.40% against 10-year 12.81%) and no valuation data (no price-to-earnings or CAPE in cited sources). Jev: now is a good lump sum entry, 0.18 no. The honest position is that timing cannot be read from this evidence.
ES3 tracks the Straits Times Index with full replication, 30 holdings, expense ratio 0.28%, semi-annual dividends, 1-share board lot, and is included in CPFIS OA (State Street). Distribution yield is about 3.26% at 15 Oct 2026 (State Street), from a search summary, unverified.
| Horizon | ES3 (SGD, to 31 Aug 2026) | MSCI World (USD, to 30 Sep 2026) | S$200,000 in ES3 | S$200,000 in MSCI World | ES3 minus MSCI World |
|---|---|---|---|---|---|
| 1 year | 39.38% | 15.24% | S$278,760 | S$230,480 | +S$48,280 |
| 3 years | 26.22% | 21.40% | S$402,174 | S$357,838 | +S$44,337 |
| 5 years | 18.17% | 11.89% | S$460,857 | S$350,741 | +S$110,116 |
| 10 years | 11.37% | 12.81% | S$587,100 | S$667,584 | -S$80,484 |
Mismatches: ES3 is in SGD to 31 Aug, the index is in USD to 30 Sep; ES3 is a fund net of fees, the index is gross of fund costs; and before 22 June 2026 the ES3 fund tracked on a price-return basis (State Street). Jev on acceptability was a split (0.50), and on handling it chose "publish with mismatches stated" (0.78).
ES3 concentration: financials 61.46%, DBS 29.26%, OCBC 18.98%, UOB 9.35%, three banks 57.59% (State Street). Jev: concentration is a material risk, 0.89 yes. Jev: ES3's 1 to 5 year lead will persist, 0.20 no, so it probably reflects a recent bank-led rally. Jev: ES3 is a credible alternative to XWR for a Singapore investor, 0.58 yes. The SGD base removes currency risk for an SGD investor, but it adds single-country and single-sector risk.
| # | Statement | Noul | Read |
|---|---|---|---|
| 1 | Low cost and tax-efficient structure | 0.67 | Strength |
| 2 | Broad developed-market diversification | 0.65 | Strength |
| 3 | Strong long-run index returns | 0.80 | Strength |
| 4 | US tech concentration material | 0.80 | Risk |
| 5 | Drawdown of 20% or more material | 0.89 | Risk |
| 6 | Non-SGD currency exposure material | 0.88 | Risk |
| 7 | New SGX line liquidity risk material | 0.70 | Risk |
| 8 | Reasonable as a core holding | 0.47 | Split |
| 9 | Worth recommending (5 years) | 0.61 | Lean yes |
| 10 | USD 1C fund as a labelled proxy | 0.67 | Acceptable |
| 11 | Jev can be the backtest engine | 0.04 | No |
| 12 | Staging beat lump sum | 0.02 | No |
| 13 | Entry timing decisive for 5-year holder | 0.28 | No |
| 14 | Entry after a loss year is a usable signal | 0.07 | No, hindsight only |
| 15 | Now is a good lump sum entry | 0.18 | No |
| 16 | ES3 comparison acceptable with mismatches | 0.50 | Split |
| 17 | ES3 concentration material | 0.89 | Risk |
| 18 | ES3 credible alternative | 0.58 | Lean yes |
| 19 | ES3 lead persists | 0.20 | No |
| 20 | Handling of mismatch (choice) | 0.78 | Publish with caveat |
| 21 | Verdict wording (choice) | 0.96 | Lump sum history |
| 22 | 3-year return unusually strong vs 10-year | 0.74 | Supported |
| 23 | History supports waiting for a loss year | 0.20 | No (my decision, not Jev's) |
- XWR's ISIN and share class (the proxy rule applies until confirmed).
- Brokerage cost and spread on SGX, and whether Endowus offers XWR.
- Equity valuation data (price-to-earnings, CAPE), so no entry-level view.
- ES3's September factsheet, to align dates with the index.
- US estate tax position from a primary source or a cross-border tax adviser.
- ES3's distribution yield from the factsheet itself.
My position, in two points. Point 2 is my decision, not a Jev result.
-
Recent returns are strong against the long run. The index returned 21.40% a year over 3 years, against 12.81% a year over 10 (Jev 0.74, supported).
-
I will revisit this ETF after a calendar year with a negative index return. Section 7 shows a higher annualised return for entry after a loss year than for entry at the start of it: 14.78% against 11.54% after 2018, and 21.17% against 9.85% after 2022. This is my decision, not a Jev decision.
| Year | Return | Source |
|---|---|---|
| 2008 | -40.71% | Investing in the Web, net, citing the MSCI factsheet |
| 2011 | -5.54% | same |
| 2015 | -0.87% | same |
| 2018 | -8.71% | same |
| 2022 | -18.14% | same |
