If you're middle income, your best relationship manager is yourself.

 

Three incidents over the past month pointed to the same conclusion: for middle-income Singaporeans, nobody in the financial industry is structurally set up to give you their best attention, and you no longer need them to. The tools to run your own money are cheaper and better than they were even two years ago. What you have to supply is the discipline and the willingness to check numbers. 

Three incidents

One. An ex-classmate of mine, a successful businessman, turned up at a networking event underdressed. A relationship manager made a backhanded remark about it. The comment was petty, but it was also information: a part of the industry sorts people by appearance, which is a poor proxy for either wealth or judgment, and an even poorer basis for allocating service.

Two. A financial advisor posted a story about a private banker who reluctantly accepted a client’s S$3 million account, then told the client to build it up to S$5 million to maintain the relationship. Treat it as marketing copy, because that is what it is. The reason it works as marketing copy is that it inverts the relationship in a way people recognize: the client becomes responsible for growing his assets to keep his banker interested, rather than the banker being responsible for growing the client’s assets.

Three. A relative of mine, an ex-banker, told me REITs had been performing badly and that my approach was therefore wrong. He then showed me his custodian account: a diversified UCITS global ETF on the LSE, in a 60/40 portfolio, presented as self-evidently superior.

The third conversation is the one worth unpacking, because he was half right and I was half right, and neither of us could settle it by tone of voice.

What the numbers actually say

I ran the comparison afterward. All figures below are total returns in Singapore dollars, using a 2.5% risk-free rate, for the five- and ten-year periods to 26 September 2026. The proxies are the STI ETF (ES3) for local blue chips, the Lion-Phillip S-REIT ETF (CLR) for S-REITs, the MSCI World ETF (URTH) for global developed equities, and a 60/40 mix of ACWI and AGG to shape his portfolio.

Five years

Portfolio	Annualised return	Volatility	Sharpe ratio
STI ETF	16.1%	11.2%	1.22
S-REIT ETF	−1.1%	12.7%	−0.28
MSCI World	10.4%	16.8%	0.47
60/40 global	5.6%	11.5%	0.27

Ten years

Portfolio	Annualised return	Volatility	Sharpe ratio
STI ETF	10.3%	13.0%	0.60
S-REIT ETF	2.6%	14.7%	0.01
MSCI World	12.3%	18.3%	0.54
60/40 global	7.4%	12.0%	0.41

Three things fall out of this.

First, my relative was right about REITs, and I was wrong to lump them in with the rest of the local market. S-REITs have delivered roughly nothing for a decade and have lost money over five years, with equity-like volatility during that time. Rising rates did most of the damage, and leverage did the rest. If you hold S-REITs, hold them for the income and the eventual rate cycle, and be honest that the last five years were bad.

Second, the local market as a whole has been strong, largely due to the banks. On a risk-adjusted basis, an investor who simply held the STI ETF for five years outperformed a global 60/40 portfolio by a wide margin. That result is period-dependent. Roll the window back ten years, and the gap narrows sharply, which is the point: five-year Sharpe ratios are not a permanent ranking of strategies.

Third, the 60/40 portfolio my relative was proud of returned 5.6% a year over five years with a Sharpe ratio of 0.27. It is a reasonable, defensible portfolio. It is not genius, and nobody should present it as such.

The wider lesson is the one that matters for this article. Neither of us could have resolved that disagreement in conversation. It took 10 minutes to process the price data. Anyone who tells you what your portfolio should look like, and cannot show you the numbers behind the claim, is giving you an opinion dressed as advice.

The middle-income bind

The middle is the worst place to sit in the current service model.

If you have modest savings, the people willing to serve you are usually paid by commission on what they sell you. The advice is not necessarily bad, but the incentive is tied to the product, not to your outcome.

If you have just enough to qualify for a relationship manager or the entry tier of private banking, you are the smallest client in the book. You are competing for attention with people who have ten times your assets, and you will lose that competition on most days.

The cost is easy to quantify. A 1% annual advisory or platform fee on a S$300,000 portfolio is S$3,000 a year. The same money in an index ETF charging 0.12% costs S$360. Over twenty years, that difference compounds into a meaningful part of a retirement.

Why doing it yourself got easier

From a tentative listing date of 13 October 2026, SGX will list four Xtrackers UCITS ETFs, traded in Singapore dollars. The one that matters most for a core holding is the MSCI World ETF (ticker XWR), with a total expense ratio of 0.12% per year, and the S&P 500 ETF (XUS) at 0.03%.

Why this is useful for a Singapore investor:

•           You can buy it on SGX through CDP, so the shares sit in your own name rather than with an overseas custodian.

•           It is expected to qualify for SRS investment, subject to eligibility and your platform supporting it. Confirm with your SRS operator before you plan around it.

•           As Irish-domiciled UCITS funds, US dividends inside the fund are taxed at 15% rather than the 30% a Singapore investor faces on US-listed holdings, and they sit outside US estate duty.

Two caveats that the launch publicity will not emphasize. Trading in Singapore dollars does not eliminate currency risk, because the underlying assets are still foreign; it only changes the currency in which you see the price. And bid-ask spreads on a newly listed SGX ETF are unproven, so compare the spread against the London or US listing before assuming the local one is cheaper overall. DWS delisted its Singapore ETFs in 2020, which is worth remembering when choosing a holding you intend to keep for 20 years.

For those who want to own individual stocks, AI has genuinely lowered the cost of research: extracting figures from an annual report, building a first-pass comparison of peers, checking whether cash flow covers dividends. It removes the clerical work. It does not supply judgment, and it will state a wrong number as confidently as a right one. The workflow only improves if you verify the output against the filings.

What being your own relationship manager actually requires

Firing the middleman is not free. You take on the work, and the work is specific:

1.         Write your policy down. Target allocation, what you will buy, and what would make you sell.

2.         Know your costs. Expense ratios, brokerage, custody, and spreads, totaled once a year.

3.         Measure against a benchmark. Return, volatility, and Sharpe against an STI ETF and a global index ETF. If you cannot beat both after costs, hold them instead.

4.         Review after results, not after headlines. Once or twice a year is enough for a portfolio you own for income.

5.         Keep a decision log. Write down why you bought. It is the only way to tell skill from luck later.

That is a few hours a year. It is less than most people spend choosing a phone plan, and it is the difference between being sold a portfolio and owning one.

Nobody at a bank is going to care about your money as much as you do. The useful response is not resentment; it is to make yourself competent enough that their attention stops being the thing you need.

---

This article is for education only and is not financial advice. Return figures are computed from adjusted daily price data to 26 September 2026 and will change with the measurement period.