Letter to Batch 43 of the Early Retirement Masterclass

It has been an honor and a privilege to conduct a 5-Day Early Retirement Workshop for you.

Three months ago I wrote to Batch 42 that the Straits Times Index at 5,212 was a generational milestone, and that a rising index compresses dividend yields. Both statements have held up. Neither went far enough. The index has kept running, the leadership on SGX has shifted decisively towards growth, and the hunting ground for income investors in Singapore has narrowed to the point where your own portfolio exercise produced only twelve names.

The Index Is at a Record, But the Real Money Was Made Outside It

The STI closed at 5,801.96 on 4 September 2026, a fresh all-time high and roughly 35% above where it stood a year earlier. It returned 13.1% in the first half of 2026 alone. By any historical standard for this market, that is an exceptional run, and the banks did most of the heavy lifting inside the index.

Now look at what happened outside it. In the first half of 2026, AEM Holdings returned 512%. UMS Integration returned 132%. Nanofilm returned 113%. Frencken returned 106%. Against an STI that gained 13.1%, the semiconductor supply chain not only outperformed; it made the index look like a savings account.

Every one of those businesses is tied to the same story: AI capital expenditure flowing into semiconductor equipment and precision manufacturing. The market spent the first half of the year repricing that exposure, and it repriced it hard. As of May 2026, AEM traded at around 42 times earnings versus a historical average near 14, UMS at around 38 times versus 16, and Frencken at around 29 times versus 11.

I want to be direct with you about what this means, because it would be dishonest to gloss over it. If you had anchored on dividend yield in January 2026, you missed all of it. None of these counters would have passed the screens we ran in class. Their yields are small or absent, their earnings are cyclical, and their current multiples price in an earnings recovery that has not fully arrived. That is a real cost of the discipline you have just learned, and pretending otherwise would insult your intelligence.

The question is what to do about it, and my answer is nothing. You cannot build a retirement income stream out of counters that pay you almost nothing and require you to be right about the AI capital expenditure cycle. The market may well be correct that these businesses earn into their multiples. It may also be wrong. Either way, the exposure does not do the job you need done.

Singapore Is No Longer a Comfortable Market for Dividend Investors

When I started writing Tree of Prosperity, assembling an SGX portfolio yielding 6% required patience but not much cleverness. Dozens of counters with decent balance sheets paid above 5%. That is no longer true, and I think you should hear it plainly rather than discover it slowly.

The banks' re-rating is the clearest illustration. OCBC sits in your portfolio at $24.61, yielding 3.84%. That is a good business, well-capitalized, with room to continue returning capital. It also pays you less than 4% now, even though the same franchise yielded well above 5% not long ago. The market has repriced quality income upward. What remains cheap on a yield basis is, more often than not, cheap for a reason.

This is the structural shift. Singapore equities have become a market where good income is expensive and cheap income is fragile. The gap in between is where your work has to happen now, and it takes more effort than it used to.

REITs Remain the Exception, and That Is Where the Bargains Are

While the STI was setting records, the FTSE ST All-Share REIT Index fell 8.2% over 2026 to early September. The two moved in opposite directions, and that divergence is the single most useful fact in this letter.

The iEdge S-REIT Index now offers an average distribution yield of 6.3% to 6.4%. The sector trades at roughly 0.86 times price-to-net-asset value, a 14% discount to stated book value. With the 6-month T-bill at about 1.70%, the yield spread is close to 460 basis points. Historically, spreads above 400 basis points have been attractive entry points for investors who can tolerate the associated volatility.

The weakness is driven by rates, not operations. Markets are pricing at least one more Federal Reserve hike by the end of 2026, and the Singapore 10-year government bond yield has risen about 0.53 percentage points over the past year to 2.38%. Rates are the weather that REITs live in, and the forecast turned hostile.

What did not turn hostile is the cash. Most S-REITs grew distributions year on year in the first half of 2026, with several posting double-digit gains. Keppel DC REIT and OUE REIT both posted double-digit DPU growth. Prices fell while distributions rose. That combination is precisely what expands the yield available to a buyer, and it is why REITs occupy five of the twelve slots in your portfolio.

Three checks before you add to any REIT position. First, the proportion of debt that is fixed rather than floating, because a REIT with 70% fixed-rate debt is insulated in a way that one with 70% floating exposure is not. Second, the interest coverage ratio and gearing indicate whether the distribution can survive a hostile refinancing. Third, whether the discount to book is a mispricing or a judgment about the book itself. A REIT at 0.7 times NAV because the market disbelieves the valuation is not a bargain; it is a disagreement you have to win.

Why Your Final Portfolio Has Only Twelve Stocks

You made a batch decision I want to put on the record, because it was more rigorous than what I usually see.

You set a portfolio yield target above 6%, and then refused to reach it the easy way.

The easy way was Deep Value. Every screen we ran surfaced counters trading below net cash. Anchun International at 30.5 cents against roughly 67 cents of cash and short-term deposits per share. Fuxing China at three to four times earnings and 0.13 to 0.16 times book, among the statistically cheapest names on SGX. HL Global, where cash per share exceeds the share price. Asia Enterprises, Nippecraft, Koyo, VibroPower. On a spreadsheet, these are the cheapest things in the market.

You eliminated all of them, and your rejection notes were consistent: severe illiquidity, controlling-shareholder influence, no formal dividend policy, and earnings that were project-driven, one-off, or negative. The line one of you wrote on Anchun was the sharpest thing produced in the whole exercise. You are paying 30.5 cents for something holding roughly 67 cents of cash per share, but you need management to eventually unlock it, and cheap can stay cheap for a very long time.

Cutting the Deep Value bucket removed most of the high-yield candidates in one stroke. Twelve counters survived, with a projected portfolio yield of 6.21%. The five-year backtest on that basket returns 19.91% annualized with a standard deviation of 12.02%, a Sharpe ratio of 1.41 against a 3% risk-free rate, and a maximum drawdown of 12.21% in October 2023. Those are respectable numbers for a portfolio built to pay you, not to beat an index.

Twelve is fewer names than I would normally want. Be honest with yourself about what that means: with twelve holdings, a single suspended distribution costs you roughly 8% of your income. The mitigation is that your cash flows come from genuinely different places, including banking, suburban retail, healthcare property, data centers, fiber infrastructure, energy, steel, regional beverages, and two Hong Kong listings. Diversification is about the drivers, not the count. Even so, treat twelve as a starting point. Add names as your capital grows and as your watchlist throws up better entry yields.

Putting It Together

Batch 43 enters a market where the exciting money is in counters that do not pay you, and the counters that do pay you have been marked down because of the interest rate cycle. That is an uncomfortable position for a new dividend investor, and I would rather you understood it now than felt it later.

On the evidence, it is also a reasonable time to buy income. A sector yielding 6.3% at a 14% discount to book, with distributions still growing, is not a market that hates your strategy. It is a market that has temporarily marked it down.

The framework you learned holds. Buy businesses with durable earnings and a record of returning cash. Buy them when the yield is attractive relative to the alternatives. Check that the balance sheet can survive a bad refinancing. Spread the income across sectors so that no single rate decision can break it. Review the portfolio thoughtfully, not obsessively.

Persist

The hardest part of this journey is not the analysis. It is the first few years, when the dividends are small next to your salary, and the compounding has not yet become visible. A $200,000 portfolio at 6.21% pays about $12,400 a year, or a little over $1,000 a month. That does not replace an income. It buys your groceries. It is still the beginning of what will eventually replace your income.

You will also be tempted, repeatedly, by the AEMs of the next few years. Someone in your circle will have made 500% while you collected 6%, and they will mention it. Hold two facts in your head when that happens. Their return is unrealized until they sell, and you have no idea what they did with the other 90% of their capital. Your portfolio is built to pay you whether the AI cycle continues or breaks, and that is a different objective, not an inferior one.

Measure yourself against your income, not against the index. Count the dividends received this year against the dividends received last year. That number is the one that determines when you can stop working, and it is the only scoreboard that has ever mattered in this course.

Keep buying. Keep reading the annual reports. Keep the watchlist current and the cash ready for the days when the market offers you a better entry yield than it does today. Those days come around more often than you would think.

I am proud of the work every one of you put into Batch 43, and the Day 5 portfolio was among the most disciplined I have seen. I look forward to hearing where your journeys take you. As always, my door remains open.

Good luck, and invest wisely.

 

Christopher Ng Wai Chung

Tree of Prosperity

19 September 2026

Are local racists leaving money on the table in the stock market?

 

It's understandable why the government is concerned about local racism. Some hurtful comments were made about ethnic Indian Singaporeans in response to the disaster in Nepal. I'm another case: criticism of Singapore's investments in Air India attracted another round of racist comments about Indians in general. 

As an investor, I think we should not politicize a company's business decisions. 

• If you believe that SIA can turn around Air India's fortunes, then you should buy the stock. 
• If you are indifferent like me, stay out of the SIA share market. 
• If you are super-confident that SIA would lose money, then muster the courage to short the counter instead of politicizing this issue. 

India has a rising middle class and superb GD growth, and this should get some investors excited.

Today, I want to ask a provocative question. 

Are local racists leaving too much money on the table in the markets? 

And to illustrate this further, I drew a scatterplot of all the stocks in my portfolio, plotting Return on Equity (ROE) against the Price-to-Book Ratio (PB), using data from Yahoo Finance. Then I fit a least-squares regression line through the data points. Points furthest above the line may indicate drastic undervaluation.

So the most undervalued stock is Laopu Gold, which will be the subject of a future blog article an it;s my latest acquisition.

But note that a close second is CapitaLand India Trust, which, in my opinion, has superb performance at a steep discount to its market price.

I did further research by invoking a quick stock research skill I built into my Claude Co-work, and this is what it had to say:

Stock Name: CapitaLand India Trust

Ticker: CY6U.SI

Current Price: SGD 0.950

Analyst Target Price: SGD 1.33

Fundamentals: PE Ratio - 4.7 | PB Ratio - 0.77 | Current Yield - 9.8% (8.3% run-rate — a 1.44c advanced distribution put three payouts in the trailing window) | ROE - 15.3% | RSI 180 days - 46.8 |

Business: A Singapore-listed business trust owning Indian real estate — 22.0 million square feet of completed IT business parks, industrial and logistics facilities, and data centers across Bangalore, Chennai, Hyderabad, Pune, and Mumbai.

Revenue Source: 1H 2026 total property income of S$137.6 million (INR 9,923 million) is almost entirely rent from IT business park tenants, with a growing data center and industrial slice, supplemented by interest on S$417.9 million of receivables from six forward-purchase development assets.

SWOT Analysis: 

Strengths - Net property income margin widened to 78.1% from 76.1%, driven by 24% positive rental reversions and 91% committed occupancy, lifting distributable income by 8% to S$64.2 million, while the Navi Mumbai data center Tower 1 was fully leased and handed over to a hyperscaler in July, with full income from August. 

Weaknesses - The rupee fell 12% against the Singapore dollar, converting 13% DPU growth in INR into just 1% in the currency in which unit holders are actually paid, and only 53% of borrowings are hedged back into INR. 

Opportunities - There is 4.9 million square feet of development potential inside the existing IT parks plus 6.4 million square feet across the forward-purchase pipeline, funded from S$1.1 billion of debt headroom, with a July INR 5.5 billion drawdown alone expected to add 1.6% to DPU. 

Threats - Gearing of 38.0% is at the high end for an SGX-listed trust, and with 25.5% of borrowings floating, Indian rate moves hit distributions quickly — the units already sit at their 52-week low of S$0.95, down from a S$1.28 high.

Typical Dividend Months: March and September

From this quick examination, I can conclude that everything under management's control has been decent, with high rental reversions. But the PE ratio is single-digit, and dividend yields are close to 10%. But everything seems to be ruined by movements in the Indian Rupee. But even though analysts have a high target price, the REIT's six-month momentum is negative.

Of course, this doesn't mean there's actual racism behind such low valuations, but I think the weakening Rupee would still make this REIT a tempting buy.

At this point, I already own this REIT. It's generated some capital losses, but after accounting for dividends received, I'm still up on my purchase price.

Same rules apply.

If you think the Rupee's weakness could push the REIT lower, then stay out of the markets. 

But if you believe that there's somehow a thin veneer of racism that is similar to what you've noticed in social media comments about SIA, and it's a reason for this discount, then you may want to risk some capital to see if you are correct.

What Should I Do With $100,000?

 

Three general approaches, compared on the same basis: expected return, effort, and reversibility

$100,000 is large enough to move the needle on a Singaporean's finances, but where it should go depends on what stage of life you're in and how much ongoing work you're willing to put in to keep it compounding. There are three general approaches worth comparing: put it into yourself, put it into the world's stock markets through a low-cost fund, or put it into SGX income assets that pay you cash every quarter. Each has a different return profile, time horizon, and level of ongoing effort required.

1. Grow human capital: a credential that raises your income

For a younger person, the highest expected return on $100,000 is often not in a portfolio at all. It is in raising your own earning power, because a salary increase compounds over the 20 to 30 working years still ahead of you, in a way that a lump sum invested at market returns cannot match on the same time horizon.

The INSEAD MBA is a useful reference point because the numbers are public. INSEAD's own financing page puts full tuition at €109,860, with total cost of attendance (tuition plus a year of living expenses) at roughly €140,000, or about S$200,000 at current exchange rates. $100,000 does not cover that on its own. What the salary outcomes show is why people still borrow the difference or draw down savings to make up the gap: INSEAD's latest employment report puts median base salary post-MBA at €100,000 (roughly S$145,000), a 111% increase over pre-MBA salary, plus a median signing bonus of €28,900. [Source: MBA Crystal Ball, INSEAD Employment Report 2026.]

This is why the approach specifically favors younger people. A 28-year-old who doubles their base salary captures that uplift, compounding through raises and promotions, for another three decades. A 50-year-old capturing the same percentage increase has a much shorter runway to compound it, and by then may have less appetite for a year without income. The expected return here is real but not guaranteed the way a bond coupon is: it depends on the specific program, the specific industry you're targeting, and your ability to convert the credential into an actual offer. Not every credential pays off like an INSEAD MBA. A CFA, a specialist diploma, or a technical certification can produce a similar effect at a fraction of the cost, which is often the more sensible use of $100,000 for someone not aiming at a top-tier full-time MBA specifically.

2. Globally diversified UCITS ETFs on the London Stock Exchange

For $100,000 that isn't going toward a credential, the next question is how much ongoing effort you want to put in. A globally diversified UCITS ETF, domiciled in Ireland and traded on the London Stock Exchange, is close to the lowest-effort option available: buy, hold, rebalance occasionally, and let the index do the work.

Two commonly used building blocks are the Vanguard FTSE All-World UCITS ETF (VWRA, accumulating) and the iShares Core MSCI World UCITS ETF (IWDA), both giving exposure to well over a thousand companies across developed and emerging markets in a single trade. Ireland-domiciled UCITS funds are the standard choice for non-US investors, specifically because they avoid US estate tax exposure and benefit from a more favorable dividend withholding tax treaty than a US-domiciled ETF held directly.

On returns, be careful with the headline number. VWRA's average annual return since its 2012 inception has been about 13.4%, but that period captured an unusually strong bull run in US equities. [Source: stockanalysis.com, VWRA historical performance.] For planning purposes, a more conservative long-run assumption of 6 to 8% a year is the more defensible number, in line with long-run global equity history before the 2010s. At 7% per year, $100,000 grows to roughly $197,000 in ten years and $387,000 in twenty years, before any further contributions. This is a total-return vehicle, not an income vehicle: there is no cash paid out to live on unless units are sold, and it carries the full volatility of global equities. Currency exposure (the fund is USD-denominated; your costs are in SGD) is a real variable too.

3. Dividend stocks: banks, REITs, business trusts and SDRs on SGX

The third approach is building a portfolio for cash income rather than total return, using SGX-listed banks, REITs, business trusts, and Singapore Depository Receipts (SDRs). At current levels, a 5% blended yield is achievable without reaching for yield traps: DBS trades around a 4.6% trailing yield, OCBC around 5.9%, and UOB around 5.0%. [Source: StashAway Singapore, DBS/OCBC/UOB 2026 outlook.] S-REITs typically yield 4 to 6%, a function of the regulatory requirement to distribute at least 90% of taxable income, and business trusts, backed by contracted, often regulated cash flows, can run 5 to 8%. SDRs provide access to a small basket of Thai blue-chips in SGD, but the segment is newer and less liquid, and is better treated as a satellite than a core income holding.

The arithmetic is simple: $100,000 at a 5% blended yield is about $5,000 a year in additional income, paid mostly in quarterly installments. What the arithmetic does not show is the ongoing work. Unlike the ETF approach, this is not a buy-and-forget portfolio. It requires continuing to check that the dividend is covered by free cash flow rather than just the headline yield, watching balance sheet risk and refinancing schedules, and reinvesting distributions with the same discipline. That work does not stop once the first $100,000 is deployed. It continues for as long as you keep building your portfolio, which, for most people pursuing financial independence, means years, not a single transaction.

Comparing the three

Approach	Expected return / income	Ongoing effort	Best suited to
Human capital (credential)	High but uncertain; compounds through career, not the portfolio	High upfront (1-2 years), then none	Younger people early in their career
Global UCITS ETF (LSE)	~6-8% total return p.a. (conservative planning assumption)	Very low: buy, hold, rebalance	Investors who want low effort and full market exposure
SGX dividend portfolio	~5% cash yield, roughly $5,000/year on $100,000	Continuous: monitoring, reinvestment, position sizing	Investors building a spendable income stream toward financial independence

The honest answer

Personal finance is personal. The right split across these three depends on your age and how many working years are left to compound a salary increase, how much of your net worth is already tied up in CPF and property, and whether you actually have the temperament to keep doing the ongoing work that option three demands. A 26-year-old with most of their net worth still ahead of them gets more out of option one than a 55-year-old does. Someone who wants to set it and forget it is better served by option two than by a dividend portfolio they won't maintain.

I've run option three for over two decades. It was not the lowest-effort choice, nor was it guaranteed to outperform a global index fund over that period. What it gave me was a growing stream of cash income I could see and use, the discipline of checking dividend coverage rather than chasing yield, and a fit for a temperament that wanted to stay actively involved in the portfolio rather than hand the decision to an index. That is a personal fit and not a universal recommendation. The framework above is what I'd use to help someone else find theirs.

The Ten Years That Fund Your Retirement Open at 30 and Close at 40

What MOM's own retrenchment data says about a window most Singaporeans assume is still open

In the Early Retirement Masterclass preview, the slide that draws the least argument is the one that deserves the most.

The model is three numbers. You need roughly $500,000. You have about ten years to build it. You save $2,000 a month. People will argue with the $500,000, usually by telling me their expenses are lower. They will argue much harder with the $2,000. Almost nobody argues with the ten years, because ten years sounds like a generous allowance rather than a binding constraint.

It is the binding constraint. Singapore's own labour statistics say the window is narrower and harder-edged than the model lets on, and that the cost of discovering this late is not linear. Lose three years at the front, and you do not need 30 per cent more savings. You need closer to double.

The window opens at 30, not at 25

Nobody saves $2,000 a month out of their first job.

The years from 25 to 30 go to finding out what you actually are professionally, servicing a study loan, getting married, and paying the first tranche on a flat. This is not a discipline failure. It is the normal shape of a Singaporean's twenties, and any model that assumes serious saving capacity at 25 is describing somebody else's life.

Real surplus starts around 30. That's the window opening, and it is the least controversial part of the argument.

The window closes at 40, and the data is not subtle about it

The closing edge is where most people's assumptions break. The common belief is that the 40s are the peak earning decade, therefore the peak saving decade. The first half of that is true. The second half depends on staying employed, and MOM's numbers show that this is precisely where the risk arrives.

From the Labour Market Report 2025 (Tables 3.6 and 5.1):

Age band	Retrenched per 1,000 resident employees, 2025	Back in a job within 6 months
Under 30	3.0	70.8%
30 to 39	7.0	68.9%
40 to 49	9.8	62.5%
50 to 59	11.1	45.2%
60 and over	4.9	37.8%

Read the two columns together, because separately each one understates the problem.

A worker aged 50 to 59 was retrenched at roughly 3.7 times the rate of a worker under 30. That is the first column. The second column is worse: fewer than half of those retrenched in that age band were back in a job within six months, against seven in ten of the under-30s. So the same event, a retrenchment, costs a 52-year-old roughly twice the runway it costs a 28-year-old, and it arrives nearly four times as often.

One caveat on the last row, because the number looks reassuring and is not. Retrenchment incidence falls for the 60-and-over group partly because the pool has already shrunk. Many in that cohort have left the labour force rather than survived within it. A low retrenchment rate among those still standing is not evidence that standing is easy.

Your degree is the part of the profile that raises the risk

The same report breaks retrenchment down by qualification, and the direction surprises most rooms I teach.

Highest qualification	Retrenched per 1,000 resident employees, 2025
Post-secondary, non-tertiary	2.9
Below secondary	3.4
Secondary	3.5
Diploma and professional qualification	6.0
Degree	11.7

Degree holders were retrenched at more than three times the rate of the secondary-educated. This is not a story about capability. It is a story about cost and structure. Graduate roles sit higher on the payroll, cluster in the sectors that restructure first, and are easier to consolidate when a firm decides that four functions can be done by two people with better tools.

If your plan for the 40s rests on the idea that qualifications buy stability, it runs against the data. The qualification bought you the earnings that make the ten-year window possible. It does not buy you the years.

The gap that nobody budgets for is 50 to 65

Here is where the Singapore-specific arithmetic bites.

From 1 July 2026, the statutory retirement age is 64, and the re-employment age is 69 (MOM). Those provisions restrict an employer from dismissing you on grounds of age below those thresholds. They do not oblige anyone to hire you, and they do nothing at all about retrenchment, which is a business-grounds exercise rather than an age-grounds one. The protection covers the job you hold. It does not cover the job you need to find at 53.

Meanwhile, CPF Life payouts begin at 65 by default. A worker retrenched at 52 with a 45 per cent chance of re-entry within six months is looking at a stretch of up to thirteen years that must be funded from somewhere other than salary and CPF Life. That stretch is what the $500,000 exists to cover. That is why the target isn't negotiable downwards just because someone feels their expenses are modest.

What losing years actually costs

The base model is $2,000 a month for ten years, totalling $ 240,000. Note what that implies before going further, because it is the assumption people skip.

$2,000 a month for 120 months is $240,000 of your own money. The other $260,000 must come from investment returns. Solve for it, and the model implies roughly 14 per cent a year. That is a demanding number, and it deserves interrogation rather than assumption, which is a separate discussion. But hold it constant for a moment and ask what happens when the window shortens.

Years left to save	Monthly savings required	Your own money over the period
10	$2,000	$240,000
8	$2,930	$281,000
7	$3,620	$304,000
5	$5,900	$354,000
3	$11,360	$409,000

Assumption: same implied return, same $500,000 target. Only the runway changes.

Three years of delay takes the requirement from $2,000 to $3,620, an 81 per cent increase in monthly savings to buy back a 30 per cent reduction in time. Five years takes it to $5,900. The relationship is not linear because compounding does its heaviest lifting in the years you no longer have, and no amount of later intensity replaces early duration.

This is why the ten years matter more than the $500,000 or the $2,000. Both of the other numbers are levers you can adjust. Time is the one input that only moves in one direction.

If you are already past 40

Most people reading this are, and I am not going to pretend the window is still open when it is not. Assume the ten years are behind you, then answer four questions honestly before deciding anything.

1. What is your actual remaining runway, in years, to the point where your income becomes unreliable? Not to 65. To the age at which your industry stops hiring people who look like you. For most PMET roles, that is closer to 50 than to 60, and the table above is the evidence.

2. What is your real monthly surplus, measured over the last twelve months rather than estimated? Take the bank statements, not the intention. The table above tells you what the number needs to be for your remaining runway. If the gap is large, you have found the binding constraint and it is on the savings side.

3. If the gap cannot be closed by saving more, which of the three variables are you actually moving? There are only three: the target, the time, or the return. Lowering the target means permanently auditing expenses down, because every recurring dollar of expense raises the capital required by roughly 300 dollars at a 4 per cent withdrawal rate. Extending the time means accepting a later date rather than a worse plan. Raising the return means taking on the work and the risk that go with it. Pick deliberately, because doing nothing is the same as picking the third option by accident and hoping.

4. What happens to the plan if you are retrenched next year? Run it. If the answer is that the plan fails, the plan was not a plan; it was a projection that assumed the one thing the data says you cannot assume.

The honest position for someone at 45 is that the plan is now shorter, tighter, and more dependent on the return on savings than it would have been at 32. That is not a reason to skip it. It is a reason to stop treating the timeline as an abstraction and start putting real numbers against it.

That third variable, the return, is the one most people have never seriously worked on, and it is what the Early Retirement Masterclass spends its time on. Not because returns are the whole answer, but because for anyone who has already spent part of the window, it is the only lever with enough leverage left in it.

---

Data source: Ministry of Manpower, Labour Market Report 2025, Tables 3.6 and 5.1. Retirement and re-employment ages under the Retirement and Re-employment Act, as amended, with effect from 1 July 2026. The savings calculations above are arithmetic based on the stated assumptions, not a forecast.

A fun experiment with "alternative investments"

 

Just for fun, I've decided to get into alternative investments. In this case, "alternative investments" means buying a box of Riftbound boosters and holding it for several years to see whether the price appreciates.

Riftbound is currently one of the hottest new CCGs in the market, with a major tournament actually happening over this weekend in Singapore. It is based on League of Legends, I believe. I was taught how to play this game a few months ago and found that the gameplay is quite simple, but the competitive strategy can go really deep.

At the moment, Riftbound is still an affordable game with a decent starter box at about $25. Players tend to be more about enjoying the game rather than collecting rare and premium cards, although there are cards costing over $1,000 USD, just like Magic: The Gathering or Pokémon. What's important to me is that I can still buy the first Origins set in English at a reasonable price, something I couldn't do with MTG in the 1990s. 

So I guess this blog would be a fun way to track the price of my box of boosters compared to (1) DBS stock and (2) Bitcoin. I see Riftbound as the equivalent of paying for an altcoin. It certainly does not have the stability of Pokémon cards, or the longevity of Magic: The Gathering. If the game dies out in a year, I'll probably suffer a small loss.

While current players disagree with me on the longevity of Riftbound, I've been around to see better games fail in the market, like Legend of the Five Rings and Versus. So I'm effectively betting against myself.

Just for the record, I paid $360 for a box of Origin boosters. 

A DBS stock is $78.65 today. 

BTC is trading at $79,680 when I wrote this article.

For folks who also play the game, I have no intention of opening my box. I have a "pauper" attitude toward collectible card games: get a set of playable starters, find a character my kids and I have an affinity for, then search for free or cheap common cards to build a fun deck to play at home. Collectible cards can teach children strategy, and trading cards can teach them a little about speculation.

 

The Best Time to Leave Singapore to Study Overseas

 

Why O-levels, not A-levels, is the right exit point, and what it actually costs

Ruixue Jia and Hongbin Li's The Highest Exam: How the Gaokao Shapes China makes an argument that applies directly to Singapore, even though the book is about China. Their claim: the best age for a Chinese student to leave for the US is around 15, neither earlier nor later. Leave too early, and the child has not yet absorbed the discipline instilled by China's exam-driven system. Leave too late, and the child has spent so many years optimizing for a single test that they struggle to function in an American classroom built on discussion, initiative, and open-ended work. Fifteen is the point at which the discipline is already in place, and the capacity to unlearn a narrow test-taking mindset remains intact.

Singapore does not have the Gaokao. It has the O-levels. The structural logic is the same. The right exit point for a Singaporean student who wants to study overseas is neither before O-levels nor after A-levels. It is right after O-levels, at around 16.

Why O-levels are Singapore's equivalent inflection point

Singapore's education system compresses a similar kind of discipline into the years before O-levels: PSLE streaming, six years of secondary content, a national exam that determines the next track. A student who has gone through this and done reasonably well has already built the study habits, content discipline, and exam stamina that Jia and Li credit to the gaokao system. That is the raw material the book says is necessary for a move abroad to work.

Leaving earlier, at PSLE, does not clear that bar. A 12-year-old has not yet been tested on six years of secondary-level content, and the discipline the book describes has not yet been developed. The risk is a student who is neither disciplined by the Singapore system nor yet capable of self-directing in a more open one.

Leaving later, after A-levels, misses the window in a different way. By 18, the student has spent two more years inside a system that rewards the same narrow test-optimizing behavior, this time at higher stakes (H1/H2/H3 subject combinations, a single dominant score). The adaptation runway the book argues for, the two years of exposure to an open academic culture before the high-stakes stretch of a full degree, is gone. The student goes from one exam-maximizing system straight into another country's university, with no transition period in between.

O-levels sit at the point where the discipline is built, but the runway to adapt is still available. That is the argument, not a preference.

The complication the book does not have to deal with: National Service

This is where the Singapore case diverges from the China case, and where a generic transposition of the book's argument breaks down. For half of the affected population, the exit point is not a matter of free choice in timing. It is constrained by law.

Male Singapore citizens and PRs must register for NS at 16.5, whether they are overseas or not. MINDEF grants deferment for full-time studies up to A-level, IB, polytechnic diploma, or equivalent qualifications. It does not grant deferment for university studies. A son who leaves after O-levels to do two years of overseas high school (the US junior-senior years, UK sixth form, or an overseas IB) can get that deferment. What he cannot do is defer NS again once that pre-university stage ends. He has to return, enlist, and serve before starting an overseas degree program, or apply to defer his university admission until after his Operationally Ready Date.

In practice, this means a son on this path enters university at around 20 or 21 rather than 18 or 19. That is not actually worse than the timeline for a son who stays local and does JC (enlistment after JC also lands him at university around 20 to 21), so the O-level exit point does not cost a boy time relative to the local track. What it costs is continuity and reversibility. He has to leave an overseas academic and social environment midstream, return to Singapore for two years of NS, and then decide whether to return to the same country and system or pivot. Some overseas boarding schools and universities accommodate this; many international admissions processes are not built around a two-year interruption after acceptance.

A daughter faces none of this. She can go straight from O-levels through two years of overseas high school into an overseas degree without interruption, fully capturing the book's adaptation window. This is a real asymmetry in the decision, and it should change how a family plans for a son versus a daughter rather than being treated as a footnote.

What it costs

The book's argument is about timing and psychology. It says nothing about money because, in the China case, the constraint the authors are writing about is admission and adaptation, not affordability, for the families they are studying. In Singapore, cost is the binding constraint for most families weighing this decision, and the gap between staying and leaving is large.

Path	Two years pre-university	Degree	Rough total (SGD)
Local: JC + NUS/NTU	Fees for citizens are heavily subsidised, effectively negligible	~S$82,000 (tuition + living at home, 4 years)	~S$80,000–90,000
UK: sixth form + UK degree	International boarding sixth form, S$45,000–90,000 per year (S$90,000–180,000 for two years)	~S$257,000 (Warwick-level, 3 years, tuition + living)	~S$350,000–450,000
US: boarding school + US degree	International boarding school, S$70,000–105,000 per year (S$140,000–210,000 for two years)	~S$480,000 (MIT-level, 4 years, tuition + living); mid-tier private universities run lower	~S$500,000–700,000+

These are estimates, not quotes, and the range depends heavily on which school and university a student gets into. The point is the order of magnitude: a full overseas track from O-levels through graduation costs somewhere between five and eight times the local track, and that gap has to be funded by the family, since Singapore's MOE subsidy applies to local institutions, not to a private boarding school or an overseas university's home-country fee schedule. Government-funded scholarships (PSC, SAF, MOE teaching scholarships) can close this gap, but they come with a service bond, usually five to six years, which is itself a constraint on later career choices.

Advantages of leaving after O-levels

The case for this timing, beyond the discipline-and-adaptation logic above, comes down to four things. 

• First, two years is enough runway to adjust to an unfamiliar academic style (essay-based, discussion-driven, less structured than the O-level syllabus) before the stakes of a full degree. 
• Second, a student applying to competitive overseas universities from an overseas high school builds a more legible profile for holistic admissions: teacher recommendations from the target system, grading that the admissions office understands natively, and two years to build extracurricular depth, for which Singapore's JC timetable leaves little room. 
• Third, the student escapes the narrower H1/H2/H3 subject-combination logic of the JC system two years earlier, with more room to explore before committing to a major. 
• Fourth, two years of immersion builds language fluency, social capital, and local references that a direct university application from Singapore does not.

Disadvantages

Cost is the first and largest. The second is what the family gives up locally: the JC-to-NUS/NTU cohort is also a professional network Singaporeans draw on for the rest of their careers, and a daughter who leaves after O-levels forgoes it entirely; a son forgoes two years of it before NS reconnects him to a local cohort, but a different one from his secondary school peers. Third, the decision is less reversible than switching streams locally. If the adaptation does not go well (the discipline built by O-levels turns out not to transfer, which happens), the family has already committed a large, largely non-refundable sum. Fourth, some Singapore government scholarship pathways (PSC, SAF) are structured around completing A-levels locally; leaving after O-levels can preclude eligibility for them, unless the family is deliberately pursuing a different overseas scholarship route.

What should actually decide this, family by family

The book's framework, applied to Singapore, says O-levels is structurally the right exit point. Whether it is right for a specific family depends on four things that need honest answers before committing, not after:

1. Can the family fund S$350,000–700,000 without compromising retirement or other children's education, or does the plan depend on winning a bonded scholarship?
2. For a son: is the family prepared for the NS interruption (return at 18, serve, then resume overseas), and have the target school and university admissions processes been checked to confirm they accommodate a two-year gap?
3. Has the child actually demonstrated the discipline the O-level result is supposed to signal, or was the result carried by tuition and structure that will not exist overseas?
4. Does the target university's admissions process actually reward two years of overseas high school more than it would reward a strong A-level or IB result presented directly from Singapore? For some universities and courses it does. For others, a direct application at 18 gets a comparable or better outcome at a fraction of the cost.

Where all four answers are favorable, O-levels is the right time to leave, for the same structural reason Jia and Li give for age 15 in China: the discipline is already built, and the runway to adapt is still there. Where the funding is not secure, where a son's NS timeline has not been mapped against the target school, or where the O-level result was propped up rather than earned, the JC-to-local-university path remains the better bet, not because it is safer in the abstract, but because it does not require betting six figures on an unverified premise.

How to think about the Return on Education for degree programs in Singapore

 

The Polytechnic semester has ended, but I have to constantly remind myself that, thanks to blog readers and friends who also work at a Polytechnic, there are now young and impressionable readers of this blog, and it behooves me to make it a little more useful and wholesome for everyone.

So this week, I'll be sharing a powerful framework to think about the value of an education that everyone can benefit from.

This concept, known as Return on Education (ROE), is useful for comparing the value of different qualifications. It is useful for students in tertiary education to assess the value of a university degree, so they can compare a 4-year degree at a local university with other educational options.

I'm actually going to use my own students in this example.

According to a web search on salary data for Temasek Polytechnic Information and Digital Graduates, the median income is $2,950. Suppose the student is an elite graduate with a high GPA > 3.9 and is considering a Business+Computer double degree at NTU, which offers the highest starting pay of $6,950 after four years.

We use the formula to calculate ROE (ignoring the effects of NS):

(Degree Median Income / Diploma Median Income) ^ (1/degree duration) - 100%

We can compute the ROE as ($6,950/$2,950)^(1/4) - 100%, which is 23%.

23%, according to the Gaokao book, is exceptionally high. This is the equivalent of a China Gaokao student qualifying for a program at Beijing, Tsinghua, or Fudan University. 

Let's consider another scenario. 

If you have a Diploma in Humanities, your starting pay is even higher than that of an Infocomm graduate. It is $3,200. If you are aiming for a Humanities degree from NUS, the salary is $4,300, but it is also a direct honors program that lasts 4 years.

We can compute the ROE as ($4,300/$3,200)^(1/4) - 100%, which is 7.6%.

A 7.6% ROE is objectively low, as the average ROE for China and the US is expected to be consistently 10% for a professional degree.

And this is not even the worst-case scenario for young people contemplating the value of local degrees. 

An A-level graduate is worth about $3,600 in the industry. If this person is contemplating a Music degree at Yong Siew Toh Conservatory, they would also need to spend 4 years, with a starting salary of about $3,820.

We can compute the ROE as ($3,820/$3,600)^(1/4) - 100%, which is 1.4%.

I think there is a decent chance that if I teach my students maths this way, I might lose my job, but the numbers clearly show that not all local degrees are built the same way, and we need a reasonable framework to think about the value of an education instead of merely appealing to a person's love of learning when trying to pick a university degree.

In the first case, where the ROE is 23%, I can conclude that no investment products outperform investing in this NTU program. My students who qualify for the program should even be happy, even, to take out a loan to get that degree. This also means that achieving sky-high GPAs in a Polytechnic and competing the heck out of your peers is a rational move.

In the second case, where the ROE is about 7%, the choice is almost on par with an investment in equity markets. A rational parent can even reason that it's better to put the money in a diversified equity portfolio for their child and push them into the industry earlier in life. I teach law at a polytechnic to working adults, so I do not have visibility into humanities education in a polytechnic, but I predict that with an ROE of 7+%, the humanities environment would be more chill. ( Certainly less cut-throat unless someone figures out the ROE for law school.)

The final case is so irrational that voluntarily putting the funds into your CPF yields better returns, but this is where a deeper understanding of society is needed to explain the existence of music schools in Singapore. With an ROE below 2%, the qualification serves as a means for the truly wealthy to signal their indifference to their economic needs. At this stage, we stop applying economic principles and instead apply Pierre Bourdieu's Sociology of Distinction to our analysis.

Currently, both my kids are fairly advanced piano players, with my son regularly playing solo fr his primary school, but even I know this might be unsustainable as they get older. But I don't mess around when parents signal to me that their kid is some ABRSM Grade 8 and gunning for a music degree. 

In this society, it's a pretty substantial flex.

Why I did not date until my passive income hit $600 a month

 

Someone wanted me to elaborate on my idea that a guy shouldn't date until he has a source of passive income. 

So maybe I share a little bit of my story when I was starting out with the FIRE movement. 

In my twenties, there wasn't a FIRE movement, but I was acutely aware of what was likely to happen to me in my 50s due to the lack of folks in their 50s in an American MNC, so I started saving like crazy because the outsourcing wave had just started and companies were farming off their IT departments to benefit from labor arbitrage. In those days, ETFs did not exist, so my funds were mainly in a Templeton Global Equity Fund, which I eventually liquidated and transferred to a CDP portfolio of dividend stocks. 

Once I started collecting dividends, it became very natural to want a certain amount of money on average every month. I settled on $600 a month because it was my allowance during my university days. At that time, my benchmark-yielding stock was SPH, which yielded about 6%, so I just needed $120,000 in my CDP across various stocks and REITs that yielded about 6% to meet my target. 

Prior to hitting this target, I didn't really date during my single days because I didn't think I could be a good provider if I couldn't at least replicate my University allowance through an investment portfolio. At the very least, I just needed my investment income to pay for dates.

Looking back, now that I am in my 50s, a lot of the things I predicted have come true! 

Folks of my generation are losing their jobs to retrenchment exercises, and ageism practically guarantees that they will spend their days doing SkillFutures courses and watching their savings dwindle before finding a job that pays half as much.

So recently, I've doubled down on this idea. 

I think Gen Z is in a transitional phase where guys are still expected to play the role of provider, but women are becoming more financially independent, and jobs are no longer as stable as before. Right now, some kind of new masculinity is still a work in progress, so guys still need to man up and play a provider role. And, I suspect in practice, throughout a marriage, both spouses may have to endure periods of unemployment throughout their working lives, but guys stand to take a larger hit to their egos if they can't play the role of a provider in a family unit.

To make things more interesting, I'm going to borrow some software engineering concepts from this excellent book, which I'm currently obsessed with, Designing Data-Intensive Applications, which is rapidly becoming the go-to book for casual-vibe coders looking to level up to senior-level engineering skills. 

The book distinguishes between a fault and a failure. 

In an array of hard disks in a RAID-5 arrangement, one hard disk failure is a fault. It can be replaced with no real system outage because RAID-5 is fault-tolerant. Just pull the hard disk out and put a new one in, and you're good to go. However, if two hard disks fail in a RAID-5 array, we have a failure because recovery is impossible without replacing hardware, followed by tape recovery.

So you can recover from a fault painlessly, but generally, failures can be events that you might not recover from.

Taking this back to family economics, in an ideal world, both spouses work, but if one spouse gets retrenched, it is just a fault in the system, and the family can recover so long as one spouse can hold the fort while the other finds a job. 

But in practice, traditional norms ensure that if the wife loses her job, it's a fault that can be overcome. But if the husband loses the job, it's a failure. We see that many divorces are triggered by a lack of economic means on the husband's part or his inability to keep up with the wife's capabilities.

So if this observation is true, then the husband must develop fault tolerance internally.

The most basic approach is for the husband to generate multiple sources of income. This can be created with rental property, blue-chip stocks, song royalties, or a side hustle. The best source of side income should be passive, because you still have a day job to take care of. 

So I think a young man should ask himself how much he will need to eke out a bare-bones existence living with his parents. Calculate the minimum amount of expenses you need. I think you can squeeze by on $500 a month. 

Now, to generate this amount a month using a portfolio that yields 4%, multiply the number by 300. So, to generate $500 a month at a 4% yield, you will need $150,000 (300 x $500). The numbers look daunting at first, but if you can get $30,000, your monthly income will be raised by $100 a month, making the next $100 much easier to achieve with the salary raises you made along the way.

Of course, imagining that you can live on $500 a month is not enough. 

We have to test it in a business continuity planning exercise.

While holding onto your job, you might want to just completely save your entire salary and set aside monthly expenses of just $500 just to see if you can survive on that.

Should young men who read this article follow my lead and refrain from entering the dating market until they have a passive income stream?

With AI disrupting the workplace, I think even if the young man refuses to ever enter the dating market, he needs a source of income if he finds that his skills are no longer valuable in the job market. 

And young women, too, because the income from Gen Z guys is hardly reliable these days.

The Four Types of Singaporean Investor: Why the "Safest" One Might Be the Riskiest

Ask ten Singaporeans how they're investing for the future, and you'll get ten different answers: a crypto wallet here, an ETF portfolio there, a stack of bank and REIT counters, or a shrug and "I'm just focused on my career right now." Look closely, and discounting the sophisticated investors who like to flex their alternative assets, and those answers collapse into four recognizable archetypes, each with its own relationship to risk, time horizon, and definition of "winning". Three of them know they're taking a risk. The fourth doesn't realize it's taking one at all, which is exactly what makes it dangerous.

1. The Speculator: Chasing the Fast Buck

At one end sits the get-rich-quick trader, drawn to crypto tokens, contracts for difference, and leveraged options, hoping to compress decades of returns into months. This group tends to be younger, chronically online, and quick to mistake volatility for opportunity. The wins get broadcast on Telegram groups and TikTok; the losses are quietly absorbed.

The numbers suggest this crowd is larger than it looks and more nuanced than the stereotype implies. A 2025 Coinbase x MoneyHero survey of over 3,500 respondents found that 61% of Singapore retail investors now hold some cryptocurrency, yet average allocations were a conservative 6-12% of their portfolios, and 58% described themselves as long-term holders versus 22% who identified as active traders. In other words, most people who own crypto in Singapore aren't the archetype: they've bought a small position and left it alone. The true speculator is a narrower, louder subset: the trader who treats derivatives and leveraged tokens as a primary income strategy rather than a small satellite position.

It's not that speculation is inherently irrational. SGX itself runs a substantial derivatives franchise, and options and futures serve real hedging purposes for sophisticated investors. The trouble is that for the retail speculator, position sizing and risk management are usually the first casualties of the chase for a fast buck. A trade that would be a reasonable 2% hedge for an institution becomes a 50% bet on a single altcoin for a 24-year-old trying to skip the queue to financial freedom.

2. The Builder: Quietly Compounding Through Low-Cost ETFs

A second, more disciplined group has emerged over the last decade: the savvy careerist who treats investing as a payroll deduction rather than a hobby. They dollar-cost average into low-cost, broadly diversified ETFs (a Straits Times Index tracker, a global equity fund, sometimes a REIT ETF for local income flavor) and largely ignore the noise in between.

This is no longer a niche habit. SGX-listed ETF assets under management hit S$16.3 billion by Q3 2025, up 40% year-on-year, and the SPDR STI ETF (ES3) alone pulled in roughly S$387 million of net inflows over the year, at a total expense ratio of just 0.28% a year. CPF and SRS investors have become one of the biggest forces behind this growth, funneling forced or tax-deferred savings into the same handful of low-cost, broad-based funds month after month.

Their edge isn't stock-picking skill; it's a stable income, a long horizon, and the discipline to automate the decision so emotion never gets a vote. This is the group financial educators spend the most time trying to grow, because it's the one strategy that scales to the average person without requiring either luck or genius: you don't need to correctly call the next hot sector; you just need to keep buying the whole market and get out of your own way.

3. The Income Seeker: Living Off Dividends

Then there's the old-school investor, typically further along in their career or already retired, who built a portfolio of blue-chip dividend payers (banks, REITs, telcos) specifically to generate a spendable income stream. For this group, share price appreciation is almost beside the point; what matters is whether the dividend cheque covers the month's expenses.

The appeal is easy to understand: a basket of well-run Singapore REITs is currently yielding in the 5.5-6%+ range, comfortably ahead of the roughly 3-3.5% yield on 10-year Singapore government bonds, and paid out quarterly or semi-annually like clockwork. For someone who has already accumulated capital and simply wants it to pay them a salary, that's a compelling proposition.

It's a philosophy suited to a low-growth, income-hungry stage of life, though it carries its own blind spot: chasing yield can concentrate a portfolio in a handful of rate-sensitive sectors, and a payout that looks safe on a dividend calendar can still get cut when the underlying business (a hospitality trust in a downturn, a retail landlord facing an anchor tenant's exit) hits a rough patch. A high yield is sometimes the market's way of pricing in a risk the investor hasn't priced in yet.

4. The Ignorant Masses: Betting Everything on a Single Career

The fourth group barely considers itself "investing" at all, which is exactly the problem. This is the mass of Singaporeans who pour their financial energy entirely into a single career, climbing the corporate ladder, chasing promotions and bonuses, while treating that income as the only asset that matters. It feels safe because it's familiar and within their control, but it's really a concentrated, undiversified bet on one employer, one industry, and their own continued health and employability, with no hedge if any of those three falters.

2025 gave this group an uncomfortable reality check. Singapore recorded 14,490 retrenchments, up from 12,930 in 2024, with the incidence rate climbing to 6.3 per 1,000 employees. PMETs (the professionals, managers, executives and technicians who make up exactly the "safe career" crowd) were hit hardest, with a retrenchment rate of 10.1 per 1,000, up from 8.6 the year before and above pre-recessionary averages, concentrated in financial services, info-comms and professional services. A retrenchment, an industry downturn, or a health scare exposes just how little of their financial life was ever actually diversified, because the career was never a separate asset from their income; it was the only asset.

Same Spectrum, Different Blind Spots

These four types map onto a spectrum of risk awareness, not risk tolerance. The speculator takes on visible risk in pursuit of outsized reward, and at least knows it: nobody puts money into a leveraged token thinking it's a sure thing. The builder and the income seeker each manage risk through a defined strategy suited to their life stage: accumulate broadly while young, tilt toward income as retirement nears. The career-only saver, by contrast, carries risk they don't even recognize as risk, which is precisely why it's the hardest of the four to fix: you can't diversify away from a danger you don't believe exists.

The healthiest financial life for most Singaporeans probably borrows a little from the middle two archetypes: steady, diversified accumulation in the working years via low-cost ETFs, gradually tilting toward dividend-paying income as retirement nears, while treating both blind speculation and blind faith in a single paycheque as two versions of the same mistake: concentrating your entire financial future in one bet and calling it safe because it's familiar.

Happy National Day ! Let's talk about something that stresses you

 

This National Day, I want to talk about what I know about education systems.

When scholars from China are asked about books that shaped the country's evolution, they often point to books about Lee Kuan Yew. So this National Day, I'm going to talk about this book entitled The Highest Exam, which really taught me how to think about Singapore and the way we structure Singapore society.

For a start, the Chinese have been taking Imperial exams for thousands of years; this is a timeline that is way longer than that of the Chinese Communist Party or Singapore, for that matter, so countries with a large population of ethnic Chinese would naturally try to structure society using what the book describes as a "centralized hierarchical tournament".

The version of this tournament in China is the gaokao, one of the most gruesome exams in the world. Singapore's PSLE is also a centralized hierarchical tournament, but even some mainland influencers consider making 12-year-old kids go through our version of gaokao to be exquisitely cruel and would remind PRCs thinking of settling down in Singapore that our academic system is no cakewalk either.

From the Chinese perspective, the Gaokao is a great system because it measures everyone based on a single, transparent standard. In a corrupt society where money can buy favors, this is probably a better alternative than the US system, which prefers a multi-dimensional, holistic assessment criteria that can be gamed with money. Consequently, the proportion of urban elites in places like Beijing and Tsinghua University is smaller than in places like Harvard and Yale, which welcome legacy admissions.

Of late, the Chinese have been expressing some regret over their system. 

This is because the Chinese now feel that the latest Fields medallist, Prof Wang Hong, was only able to win the award after she left Chinese academia and settled in France. The academic environment in China may mean that she would be constantly overlooked, and some of her research might actually be stolen by her superiors, as she has stayed.

But the Western system has its own hobgoblins. Look no further than Cambridge Professor of Sociology Jason Arday, the DEI wunderkind, who is also a plagiarising fantasist.

Looking at the lenses through which the Gaokao matters, I can now see why the PSLE makes sense to us. Even though we are not (that) corrupt, a tournament gives everyone some kind of a hedge in case corruption rears its ugly head in the future. It's also a powerful way for the government to determine what is rewarded in society in the future.

How can we improve the system?

Right off the bat, eliminating the T-Score in favor of Achievement Levels does nothing for parents or kids because the PSLE remains a centralized, hierarchical tournament. I might even argue that A levels would be even better if they were T-scores, as that would provide greater transparency in determining who gets into top professions.

Reforms would be better if we examined the ITE and Polytechnics,  and disabused this notion that investing in Poly and ITE is like "throwing good money after bad," as allegedly said to the Education Minister in the 1990s by Tony Tan.

If the median GPA of a polytechnic graduate is higher than that of a university A-level graduate, then it makes sense to reduce A-level intake and increase polytechnic intake for that degree course. I'm seeing decent ITE students performing in a polytechnic, and perhaps a similar adjustment can take place for those few elite diploma programs.

So any alternative to the A-level system should not be another centralized tournament like the IB (which is like rich people paying to avoid A-levels by picking an easier alternative); it should be a brutal tournament based on a single score, with exams designed locally. The most direct road to a middle-class lifestyle and a profession should remain a tough road of pain and struggle.

On the other hand, Poly and ITE can have their own assessment systems, and universities need to constantly adjust their admissions based on alumni performance. Continuous assessment, project work, and skills development, rather than a winner-takes-all exam, should be a valid option for smart, ambitious Singaporeans to achieve a middle-class lifestyle. Our universities should also consider alternatives to the PhD. Keep two research universities, but some kind of new accreditation system should award doctorates for practical inventions, patents, and even for producing jobs for the economy.

Finally, I don't think any educational reform can address the major lack of social and cultural capital among Polytechnic students compared to JC students right now. This has always been a problem hidden in Singapore. Despite many attempts to make our society more inclusive and equal, we can only measure things like salaries and personal wealth.

Parents are not dumb. If my child does well in both JC and Poly and eventually gets into a University, I will still choose the option that exposes them to more peers who may end up as doctors, lawyers, or businessmen. And even with a strong social network, some environments simply allow a person to develop better "taste".  

Policy makers have a tough nut to crack.

The curious case of Colin Lau's Early Retirement

 

When the video about Colin Lau came out, a lot of friends forwarded it to me. Some folks even mentioned that there's now someone to give me a run for my money, even though there are folks like AK71 and Investment Moats who probably have a safer financial margin than me.

I thought it might be useful to blog about the strengths of Colin Lau's approach to financial independence and its inevitable weaknesses. I'm not coming from a position of being a critic; I actually think that Colin Lau's work is admirable, and it is a sad waste that the media took a while to recognise his effort doing charity work for the Philippines.

So here's what I think are the key takeaways:

a) You do not need to be rich to be financially independent

Colin's technique is hyper-frugal but also relied on the economic situation during the Great Financial Crisis of 2007-2009. In 2007, at 35, Colin bought a S$87,000 three-room flat with about 64 years left on the lease, paid in full, with no mortgage, since he felt loans "lose a lot of money." He rented out a room for S$900/month, which paid off the flat's cost within eight years. He still collects that rent today, and the flat has since appreciated in value.

b) You don't even need to buy stocks if you can minimise your personal expenses

He saved 80–90% of his income during his working years ("Everybody was spending money like water, I was spending like a test tube"). He now lives on under S$150/month, with total monthly cash flow of about S$2,500 (rental income plus insurance payouts bought when younger). He's saving the surplus for future eldercare, since he's a bachelor with no children. His philosophy: don't pinch pennies on small stuff; instead, cut the big-ticket items like housing.

This is a number that no one in my family can live on, not even my kids, because they have enrichment. Amazingly, Colin does not employ equities because all it takes is a 5% dividend stock portfolio of $36,000 to generate $150 cash flow every month on average.

c) You still need to do a bit to take care of your health

A serious health scare in June 2025 left him hospitalised for 62 days with 15 surgeries and a S$146,000 bill that is covered fully by insurance and subsidised ward class, so he paid nothing out of pocket. I'm actually surprised that he is willing to pay for some kind of insurance plan. 

Nevertheless, I think his planning and possible C/B2 class ward stay are the reasons why he's still alive today.  I can't help but feel quite sorry for him after looking at this episode.

Of course, this video raises a lot of questions as to whether someone can find a tiny little flat to live in without paying for a home mortgage. I doubt it's possible if you want a remaining lease of 60+ years, but if you are in your 60s and want something with less than 40 years, it might be possible. Also, I suspect the government may allow low-cost rental housing for singles in the future, so it's possible to build a dividend portfolio to pay off the rent for a one-room flat. Whatever it is, Gen Z would need to find a creative solution to replicate the same results.

Finally, there's no way to sugarcoat this, but Colin Lau is making the same mistake as the other ultra-frugal FIRE thought leaders in Singapore, all of whom are frugal single men.

What is he going to do to deal with loneliness when he gets older?

I think charity work is an important component of having a meaningful life, but being frugal to the point of remaining single might mean a lonely existence when you enter your 70s or 80s. There are some things that having a family and children can do for you - if you need to go for an operation, at least your kids (if raised well) will be there for you.

You can Google the number of deaths from loneliness; it kills at three times the rate of heart attacks. And men tend to be even lonelier than women as they get older.

But no FIRE blogger really wants to blog about the endgame when they hit their 70s; they just want to showcase their frugality porn on the web.

How an investor can benefit from the Feynman Technique

 

I still get a lot of questions about why I teach. 

It's always a variation of the question, like, "If you have such a successful formula for investing, why do you want to share it with others?" 

For quite a while, I argued that by teaching the subject deeply, I am forced to progressively simplify the concept and understand it based on first principles; this is further honed by serious students with genuine questions who expect answers because they pay top dollar to attend my Masterclass.

Now, with AI and teaching an actual Data Analytics class, things have gotten crazier.

It's no longer about deepening a person's understanding when a lecturer teaches a subject. The lecturer can build something in real time to test the theories they learn and see whether it holds up in reality.

So here's a concrete example. 

One of the things my students have to do is linear regression; they might need to figure out which attributes result in better fitness, like lower blood pressure. For every topic I teach, I ask myself how to make it relevant to me and incorporate it into my Portfolio Manager Windows program, which I vibecoded using Claude Code.

So I tried to create a simple regression that looked at just stocks in my portfolio and determined whether a PE ratio and dividend would affect investment gains after a year. And I can see that a low PE ratio resulted in superior performance, but the high dividend stocks in my own portfolio would actually drag down returns after a year.

Upon closer examination, the reason is less dramatic; the outperformers in my portfolio are the AI Tech stocks on SGX like UMS, AEM and Frencken. With the exception of Frencken, I owned these counters when they were still dividend stocks a number of years ago.

Another thing I make my students do is to conduct K-means clustering. It employs an algorithm to divide a data sample into K parts. It's like dividing the class into 5 groups using AI based on the data fed to it, so that insights can be gleaned from this categorisation.

So when I teach this topic, I try to get my students to use their intuition to make a case for how the categorisation is done ( groups like "Fit Daddies" or "Bored Tai Tais" using gym data). It is not an easy task because it's highly subjective.

So after teaching, I began to eat my cooking and tried to cluster the stocks in my portfolio using stock ratios, and it somehow grouped banks with stable REITs, the Tech-based growth stocks, and high-yielding REITs with less prestigious sponsors. More interesting is a cluster of orphan stocks that include First Shipping Trust, Global Investments and, probably unjustified, Capitaland India Trust. 

Where analysis can be done, data can be ingested in a built-in AI chatbot. 

So, I've gone even further and given my software chatbot capabilities to answer questions about what I should do with my portfolio. In this example, the AI tells me to ditch ESR REIT for Sabana REIT.

My journey continues; this article is to get folks to look out for a future video on whether AI can replace Financial advisors with the good folks of The Financial Coconuts. I suspect the video might not do that well because it is easily one of the most technical talks I've ever done.