Sold An ETF and Bought 4 Stocks

I recently sold my position in the UOBAM Ping An FTSE ASEAN Dividend Index ETF (UPD) after holding it for less than a year.  I collected one distribution, sold at a small profit, and gave up the next payout.  When I first bought this ETF, I thought I was diversifying.  Looking back, I think I was actually di-worsifying.

Why I bought UPD

The idea was quite simple.  I wanted to diversify beyond Singapore, and at the same time, I wanted to increase the yield of my portfolio.  UPD ETF seemed to do both.  It was a dividend-focused ETF covering five ASEAN markets, namely Singapore, Thailand, Indonesia, Malaysia and the Philippines, and it was targeting a distribution of at least 6% a year for 2026 and 2027.

On paper, that sounded attractive.  I already had a large Singapore dividend portfolio.  Hence adding some ASEAN exposure and collect a decent amount of income along the way sounds sensible, especially when it included some larger economies in the region, and it allows me to gain exposure to the energy sector, which was lacking in my existing portfolio.  So I bought it.

Then I looked underneath the ETF.  The more I looked at UPD, the more I realised that "ASEAN diversification" was not quite the same thing as the diversification I actually needed.  The biggest issue was also the sector exposure.  As of August 2026, financials made up about 64% of UPD, while energy was another 13%.  So although the ETF contains companies from five different countries, it is still heavily concentrated in the financial sector.  This is also the one largest sector I already own separately in my SG dividend portfolio and MY dividend portfolio respectively.

That made me stop and think.  I already own DBS, OCBC, UOB and Maybank directly.  So when I thought I was buying ASEAN diversification, but in reality, I was buying more banks.

The Discomfort And Lack Of Conviction Led To The Sale

The other thing I noticed was what was actually happening in the different markets.  Singapore banks had a spectacular run.  Thai banks also performed reasonably well.  Malaysia was much less impressive. My Malaysian bank holdings had been underperforming, and that made me less enthusiastic about owning even more Malaysian financial exposure through an ETF.  Indonesia and the Philippines were where I became more uncertain.

I do not claim to have a deep enough understanding of either market to make a sweeping judgement about their economies.  But I was becoming increasingly uncomfortable with the economic and market outlook there.  Indonesia still has respectable headline growth forecasts, but there are plenty of things I do not fully understand about its domestic economy, currency, policy direction and financial markets, and the recent unrest in the country seen on social media made me wary of the situation. The IMF's current forecast is still around 5% growth for 2026, so this is not me saying Indonesia is heading for some economic disaster.  It is simply that I did not feel comfortable enough with the risks to justify owning more of it just because the dividend yield looked attractive.

The Philippines gave me a similar feeling, but with more obvious near-term concerns.  The IMF's September 2026 assessment described the Philippine outlook as challenging.  Growth slowed sharply in the second quarter, the IMF expects 2026 growth of only 3.4%, and it highlighted downside risks from weaker investment and confidence, the property downturn, higher oil costs and other external factors.  Again, that does not mean the Philippines is a bad investment.  It just mean I did not want to make a long-term investment decision based on the assumption that everything would eventually work itself out.  Overall, the uncertainty I had of these economies made me uncomfortable to own this ETF at this juncture, hence I decided to sell into strength before the recent XD date.

Rebalancing the Pillars: Tilting Towards Non-Bank, Non-REIT Weight

After selling, I was thinking where to channel the funds.  Banks are great at this point in time, but I already had substantial exposure to them.  REITs are underperforming at this point in time, and valuations are not stretched.  However, the rising interest rates macro-environment does not favour REITs at this time. 

When looking at my portfolio's core structure, I noticed a distinct imbalance.  Across my top pillars, my asset distribution looked heavily skewed.  The 7 largest pillars in my portfolio consists of 3 Banks (OCBC, DBS and UOB), 3 REITs (CICT, PWLR and AAR) and 1 non-bank, non-REIT (STE).  Personally, I think from an overall portfolio management perspective, my non-bank, non-REIT segment was clearly underweight.  To correct this structural tilt and diversify my cash flow generation away from financials and property trusts, I decided to channel the proceeds from UPD directly into building up my positions in Kimly, Riverstone, and HRNetGroup. These counters help bridge the gap and add operational diversity to my core holdings.  Besides the 3 mentioned above, I also channeled part of the funds into a recently added Keppel DC REIT.

Kimly: Boring On Purpose

Kimly runs coffeeshops, food stalls and outlet management.  It operates 86 food outlets and 176 stalls, and its customers eat there whether the economy is booming or not.  That is the point.  I wanted income from daily-life spending, not from credit or rental markets. 

It is not a growth story.  First-half FY2026 revenue rose just 1.3%, but earnings grew 10.6% on better margins, and the FY2025 dividend of 2 cents was about 75% of earnings, so there is room to keep paying.  Management is also buying outlets in mature housing estates, which is a slow, property-backed way to compound.

What I am watching: management itself calls the operating environment challenging, citing higher logistics and production costs, and interest expense has crept up from about SGD 4.4Mil in FY2024 to about SGD 6.1Mil as it funds property purchases. 

Riverstone: The AI Angle Without Paying An AI Multiple

Riverstone makes cleanroom and healthcare gloves.  The cleanroom segment is the interesting part, where these gloves are used in semiconductor and other controlled manufacturing.  Brokers expect cleanroom growth to be driven by AI infrastructure demand and new customers, and gloves are a small part of customers' production cost, but timely supply is essential, which helps its pricing power.

I am not buying it for the AI story alone, though.  It has a long record of returning cash, paying MYR 0.24 for FY24, a payout above 100% of earnings.  It is also not a stock that has already run.  1Q2026 revenue fell 15.1% and net profit fell 27.1% year-on-year, so I am buying after a miss rather than after a rally.

What I am watching: Competition from Chinese suppliers and forex risks.  Special dividends are a bonus, not something I count on.

HRnetGroup: A Cash Pile With A Business Attached

HRnetGroup is a recruitment and staffing firm across Asia.  What drew me in is the balance sheet, consisting of SGD 332.1 million in cash, T-bills, credit-linked notes (CLNs) and gold, with zero borrowings.  Excluding that cash, management puts the group's P/E at roughly 7x, so investors are paying a modest price for the operating business.

The dividend looks well supported.  The interim dividend rose 10% to 2.2 cents, a trailing yield of about 5.9%, and free cash flow of around SGD 46 million covers a payout of about SGD 44 million.  It also gives me exposure to the hiring cycle, which is different from the cycles my banks and REITs follow.

What I am watching: the 1H payout was 103% of net profit, so earnings need to keep recovering, and the recruitment industry is still facing headwinds.  This is a cyclical business, so patience matters here.

Keppel DC REIT: A REIT, But A Different Engine

Yes, this is another REIT, and I have said my REIT exposure is already large.  But my existing REITs are tied to retail and office rents, healthcare and regional property.  Keppel DC REIT is tied to digital infrastructure, which is a different demand driver and gives me indirect exposure to AI-related growth (which has demonstrated more resilience than Mapletree Industrial Trust for now).

The numbers back that up.  1H26 DPU grew 11.3% to SGD 0.05714, helped by the Tokyo Data Centre 3 acquisition and rental reversions of about 10%.  Its portfolio WALE is 6.7 years, and gearing is about 34%, which leaves room to grow without stretching the balance sheet. 

What I am watching: occupancy slipped to 92.5% from 95.6% because of one expired lease at the Cardiff data centre, and distributable income grew faster than DPU because new units were issued to fund acquisitions.  

What Is Next

As my cash flow improves next year, I am watching the newer SGX-listed, Ireland-domiciled UCITS ETFs, in particular XUS (S&P 500) and XWR (MSCI World).  Ireland-domiciled funds face a 15% US withholding tax on dividends, versus 30% for US-domiciled funds held by Singapore investors, which is a meaningful edge over time.  I have not set a timeline, but this looks like a cleaner way to get broad developed-market exposure than chasing yield in markets I lacked conviction.  But if it happens, it will be slow, and in small quantities.

The Lesson

Diversification is not about owning more things.  It is about owning things that behave differently for the right reasons.  UPD ETF gave me more holdings, but not necessary better or more suitable ones, and I would prefer to admit my lack of conviction now after a year than hold it for five years and ignore it further.  Whether this is the right move, investment wise, remains to be seen.  But at least for now, I am more comfortable with this arrangement for my portfolio.

This is my own portfolio and thinking, not a recommendation.  Do your own research.  Barista FIRE, here I come...!

Comic Version: https://www.instagram.com/p/DeQI9XfEk1_/?obrf=ZDdwMm9qOHl1MGEy