What Happened
Index Investing · Markets · Malaysia · Singapore · Comparison Briefevery level read 9 Sep 2026 · figures perish; the date is the point
A record from 2014, and a record from last week
Four Indices, Read on One Day — and One of Them Is Twelve Years Behind
These four benchmarks are often described together as markets at or near record highs. Read on a single afternoon — 9 September 2026, all four from one table so that they are comparable to each other — three of them are, and one is not, by an order of magnitude. The S&P 500 is 2.24% below a high set last month. Singapore's Straits Times Index is 1.69% below one set earlier this month; it was printing fresh records on 3 and 4 September. The FTSE Bursa Malaysia KLCI is 9.61% below a high set in July 2014, and it has gained 2.04% this year against the STI's 23.32%. The interesting part is not the gap. It is that FTSE Russell, which computes the KLCI, has published a document explaining that the index no longer represents the market it is named after — and is rebuilding it, starting in December.
- 9.61%how far the FBM KLCI sits below its all-time high — 182.23 index points
- Jul 2014when that high was set. The other records on this page are weeks old
- 11.4×how much more Singapore's STI has gained this year than Malaysia's KLCI
- 30 → 50constituents the KLCI is being rebuilt to, in two steps from 21 December 2026
- 4indices on this page, and all four are down over the past month
One reading, one afternoon Market figures perish faster than anything else this site publishes, so every level here carries the same date and the same source.
| Standing | What is recorded | Where |
|---|---|---|
| Confirmed | Levels on 9 September 2026, all four from one table in one read: S&P 500 7,641.90, Dow 52,468, FBM KLCI 1,714, STI 5,730 | Trading Economics — a data aggregator, not an exchange |
| Confirmed | Year to date: S&P +11.63%, Dow +9.16%, KLCI +2.04%, STI +23.32% | the same table |
| Confirmed | All-time highs: S&P 7,816.70 in August 2026; STI 5,828.50 in September 2026; KLCI 1,896.23 in July 2014 | the aggregator’s three country pages |
| Confirmed | The STI printed 5,776 on 3 September and 5,780 then 5,803 on 4 September, crossing 5,800 for the first time. It has given up 1.69% since | the aggregator’s dated news stream |
| Confirmed | FTSE Russell, on the KLCI: “the representativeness of the FBMKLCI has come under increasing scrutiny”, with “a long-term decline in the proportion of total market capitalization captured by the index” | FTSE Russell’s own consultation, March 2026, read in full |
| Confirmed | And on why: financials have “consistently accounted for more than 30% of the index, limiting diversification and underrepresenting faster-growing sectors such as Technology and Energy” | the same document |
| Confirmed | The remedy proposed: 30 to 50 constituents, a 10% company-level cap, and market-capitalisation coverage rising from about 60% to 70% | the same document |
| Confirmed, not verifiable here | The expansion proceeds in two phases: 21 December 2026 at half weight, 21 June 2027 at full. Reported as the first major revamp since 2009 | Malaysian business press. Neither official announcement was reachable |
| Not carried | The Dow’s all-time high; and every composition figure — sector weights, top-ten concentration, price-to-earnings ratios, dividend yields, individual constituent weights. None was verifiable at a source that answered, so none appears on this page | — |
Timeline
July 2014 to June 2027 Short on purpose. It starts at the oldest record still standing and carries only what was read this session.
- Jul 2014The FBM KLCI closes at 1,896.23, a high it has not matched since. Nothing else on this timeline had happened yet
- 30 Mar 2026FTSE Russell publishes a consultation on the FTSE Bursa Malaysia index series. Its stated objective is to “improve market representation and reduce sector concentration”. Responses are invited by 24 April
- Aug 2026The S&P 500 sets its all-time high at 7,816.70. In the same month, the outcome of the Malaysian consultation is reported: the KLCI will go to 50 constituents, its first major revamp since 2009
- 3–4 Sep 2026The Straits Times Index prints 5,776, then 5,780, then 5,803, crossing 5,800 for the first time. Its all-time high for the month is recorded at 5,828.50
- 9 Sep 2026Every level on this page is read, within a few minutes, from one table. All four indices are down over the past month; three are within about two per cent of a record and the fourth is 9.61% below one from 2014
- 21 Dec 2026Phase one. Twenty new constituents enter the FBM KLCI at half their final weight, and a 10% cap applies to every company in it
- 21 Jun 2027Phase two. The new constituents reach full weight. From that date the KLCI is a 50-stock index, and comparing its future with its past means comparing two different things
The Argument
Two of these records are weeks old. One is twelve years old Derived here, from the levels above.
Set the four side by side and the usual sentence about markets at record highs breaks in one place. The S&P 500 is 2.24% below its high and the Straits Times Index 1.69% below its own, both set within the last five weeks; on the STI the record is days old, which is why the index is simultaneously at a record and slightly down. The FBM KLCI is 9.61% below its high — 182.23 index points — and the high is from July 2014. Twelve years and two months. The year-to-date figures say the same thing in a different unit: the STI has gained 23.32% this year and the KLCI 2.04%, a ratio of 11.4 to one, between two neighbouring, bank-heavy, dividend-paying South-East Asian markets that a portfolio would happily treat as the same bet.
- 2.24 · 1.69 · 9.61per cent below the record: S&P, STI, KLCI
- daysold, the STI's record; the KLCI's is from 2014
The index provider says so itself FTSE Russell, Consultation on enhancing the FTSE Bursa Malaysia Index Series, March 2026.
A twelve-year drawdown invites the wrong explanations, so it is worth noting that the strongest available account of it comes from the body that computes the index. FTSE Russell's consultation paper says the KLCI's “representativeness… has come under increasing scrutiny”, with “a long-term decline in the proportion of total market capitalization captured by the index”. The reason it gives is concentration: financials have “consistently accounted for more than 30% of the index, limiting diversification and underrepresenting faster-growing sectors such as Technology and Energy”, and this “reduces the benchmark's ability to reflect Malaysia's shifting economic structure”. That is not a critic's framing. It is a provider explaining, in a document inviting public comment, why it proposes to rebuild its own flagship — from 30 constituents to 50, with a 10% cap on any single company, lifting coverage of the market from about 60% to 70%. The liquidity to support it already exists, the paper argues: the least-traded of the top 50 stocks still trades above the 80th percentile of everything on Bursa Malaysia.
- 30%the share of the index financials have consistently exceeded
- 10%the new cap on any single company
And one number here depends on a choice, not a fact Price return and total return are different measures, and on a high-yielding index the gap is most of a year's dividends.
The Straits Times Index's 2026 is quoted in two ways, and both are correct. A price return of 23.32% counts only where the index level went. A total return, which is the number more often used in the coverage, adds the dividends back — and on an index whose largest holdings are banks, that difference is not decoration. Neither figure is wrong; quoting one while calling it the other is. This page carries the price return, because that is what was verified here, and states the distinction rather than resolving it. The same caution applies in the opposite direction to the KLCI: an index that has gone almost nowhere on price for a decade has not necessarily paid its holders nothing, and this page verified no yield figure for any of the four, so it asserts none.
- two measuresprice return and total return, neither wrong
What Others Add
Four windows on the same four indices Change the window and the ranking changes with it. Over one month they agree; over the year they do not.
| S&P 500 | Dow | FBM KLCI | STI | |
|---|---|---|---|---|
| Level, 9 Sep 2026 | 7,641.90 | 52,468 | 1,714 | 5,730 |
| One day | −0.41% | −0.60% | 0.00% | −0.66% |
| One month | −1.43% | −2.79% | −1.21% | −0.43% |
| Year to date | +11.63% | +9.16% | +2.04% | +23.32% |
| Twelve months | +17.04% | not read | +7.77% | +31.82% |
| Below its record | 2.24% | not read | 9.61% | 1.69% |
| Record set | Aug 2026 | not read | Jul 2014 | Sep 2026 |
Two things fall out of that table that the year-to-date column alone would hide. The first is that all four are down over the past month — the Dow most, at 2.79% — so over a short window these markets are one weather system, whatever the annual figures suggest about them being four. The second is the row of blanks. The Dow's record and its twelve-month change are simply absent, because no source that answered this session carried them, and the honest thing is a blank rather than a number borrowed from somewhere unchecked. That gap is worth naming for what it says about the rest: every level on this page comes from one data aggregator, because every index provider and every exchange refused. S&P Dow Jones Indices returned 403, Bursa Malaysia returned 403, SGX serves nothing without script, and the Federal Reserve's own database would not complete a connection. A page about benchmarks that cannot reach a single benchmark administrator should say so, and this is it saying so.
- 8providers, exchanges and data services that refused this session
- 4 / 4indices down over the past month
Conclusion
Which of these suits which reader Not a ranking. Four differently built instruments, and what each is actually good at.
S&P 500
For the reader who wants one line for “American listed equity”
- Best at: breadth. It is the reference against which American equity is measured, which is a use no other index on this page has
- The trade-off this page can evidence: it fell 1.43% over the past month, in line with the other three
- Not evidenced here: its concentration and valuation, both widely discussed and neither verified this session
Dow
For the reader who wants a headline, and knows what it is not
- Best at: recognisability. It is the number that reaches people who read no market coverage at all
- The trade-off: it is thirty companies weighted by share price, which is a rule almost no other major index uses
- This page carries no record level for it, because none could be verified
FBM KLCI
For the reader who wants Malaysian large caps — and can wait out a rebuild
- Best at: being about to become a better instrument. From 21 December it is a 50-stock index with a 10% cap, adding sectors it has never held
- The trade-off: its own provider says the current construction underrepresents technology and energy, and that is what the last twelve years reflect
- And a caution: after June 2027 its past and its future are not the same index
STI
For the reader who wants South-East Asian banks and will read the small print on returns
- Best at: this year. Nothing else on this page is close, on any window from one month to twelve
- The trade-off: the number you have seen quoted may include dividends, and this page's does not. Check which one you are looking at
- A record only days old means a record is not the same as safety; it is 1.69% below one already
Read the date before the number
Every level here is from 9 September 2026 and will be wrong tomorrow. That is not a weakness of this page; it is the nature of the subject, and a market figure quoted without a date is not a fact but a memory.
A record is a date, not a state
Three of these four are “near record highs”. One of those records is from last month, one from days ago, and one from July 2014. The phrase covers all three and distinguishes none of them.
When the provider files a consultation, read it
The clearest account of why the KLCI has gone nowhere is a sixteen-page PDF from FTSE Russell inviting public comment. Index methodology documents are dull and they are where the actual explanations live.
Price return and total return are not interchangeable
On an index whose largest holdings are banks, the difference between the two is most of a year's dividends. This page carries the price return because that is what it verified, and says which it is.
What could not be checked is not here
No sector weight, no price-to-earnings ratio, no dividend yield and no record level for the Dow. All are in circulation; none survived this session's sourcing, so none is on the page, hedged or otherwise.
And none of this is advice
The aggregator publishes quarterly forecasts beside every figure quoted here. They were read and are deliberately not carried. This page records where four indices stood on one afternoon and how each is built.