• Yield compression: Moving to personal tax brackets significantly lowers net returns for high-income investors.

• Loss of tax efficiency: This structural shift destroys the core appeal of hassle-free, tax-advantaged passive income.

• Strategic pivot: Paring down local REITs to reallocate capital into AI dividend stocks and Singapore data center REITs.

⚠️Disclaimer: Word of Caution!

Please DO NOT take this educational post as a buy or sell signal. When it comes to investing, it is important to have your own judgement. Despite my detailed analysis, mistakes may occur, and blindly following could lead you to make similar errors and financial losses. Furthermore, I AM NOT a licensed financial advisor. I’m merely sharing my experiences and opinions only.

Additionally, please note that I hold positions in the discussed stock, and my view may be biased as a result.

On 19th March 2026, Malaysia has officially ended the preferential 10% withholding (WHT) tax rate on REIT dividends, effective for 2026. As soon as this announcement was made, the REIT index fell 3.45% to a three-month low.

Below are the key changes starting 2026 onwards:

• Individual Investors: No withholding tax deduction; REIT income must be declared in annual tax filings and taxed at individual rates.

• Foreign Investors: Taxed at a final rate of 30% on chargeable income.

• Non-Resident Corporations: Subject to a final 24% withholding tax.

Why the sudden shift? According to Finance Minister II - Datuk Seri Amir Hamzah Azizan, Malaysian REITs have matured into a widely accepted asset class, with market capitalization grew from RM1 bn to RM57 bn over the past 21 years. As such, these REIT should no longer require continued fiscal support.

In my opinion, the reason investors like REITs is because:

1. REIT’s business model is stable and predictable. Thus, it offers stable stream of income in the form of distribution or dividend.

2. Attractive yield of between 5% to 6% despite the 10% withholding tax

This is what drove the market capitalization growth. By shifting to personal tax rates, this new ruling directly penalizes high-income earners.

Here’s an illustration based on the dividend I received from my current REIT holdings (Sunway REIT and IGB REIT):

Old Tax: 10% WHT

With this old tax rule, my yield on cost is at 8.7%. This rate is higher than most banks’ fixed deposit rates in Malaysia despite after deducting the 10% withholding tax.

New Tax: Individual Tax Rates

With this new rule, the total gross dividend (RM1,030) that I receive from both Sunway REIT and IGB REIT will be added to my chargeable income.

Looking at the table above, my REIT investment loses its edge the moment my chargeable income exceeds RM50,000. This is because my yield-on-cost drops below what I enjoyed under the old tax rule.

One might argue that even with maximum tax rate of 30%, my yield-on-cost of 6.7% is still attractive. This is true but…

Furthermore, my 8.7% yield is a luxury born from buying during the pandemic period. With the Bursa Malaysia REIT Index currently offering an average yield of about 5.36%, new high-income investors stepping in today will face significantly lower after-tax yields.

My Insights

Altering a tax structure simply because market capitalization has grown feels short-sighted, but as an investor, I can only control my reaction to this structural shifts.

This ruling makes local REITs unattractive due to compressed yields and the added administrative headache of tracking distribution received for tax season.

As such, I am adapting my portfolio with the following plans:

• Paring down Malaysian REITs: I will slowly reduce my stakes, as the current sell-off may persist until the market normalizes the new after-tax valuations.

• Re-allocating capital: Cash will be redeployed into higher-priority opportunities:

  • Tier 1: AI-related dividend stocks

  • Tier 2: Singapore REITs (specifically those with data centers in their portfolio)

  • Tier 3: Diversification play (e.g. insurance, healthcare, etc.)

Hopefully the government can undo this new tax rule. This way, I do not need to sell Sunway REITs given how fundamentally strong this REIT is.

Until my next post! 🙂

⚠️Disclaimer: Word of Caution!

Please DO NOT take this educational post as a buy or sell signal. When it comes to investing, it is important to have your own judgement. Despite my detailed analysis, mistakes may occur, and blindly following could lead you to make similar errors and financial losses. Furthermore, I AM NOT a licensed financial advisor. I’m merely sharing my experiences and opinions only.

Additionally, please note that I hold positions in the discussed stock, and my view may be biased as a result.

• Focus Investing: Concentrate on 5–10 high-conviction stocks instead of spreading across too many names.

• Probability Framework: Assign Bull / Base / Bear probabilities and update them as new information comes in.

• Expected Value Signal: Use EV-to-Price ratio to judge whether to buy aggressively, buy moderately, hold, or sell.

• Dynamic Position Sizing: Allocate based on bear-case probability — lower risk = higher allocation.

• Trim & Rebalance: Adjust position sizes when they exceed your calculated threshold.

Recently, I read this book: The Warren Buffett Way, 3rd Edition by Robert G. Hagstorm and learned one interesting idea around portfolio management. I’m thinking about trying it.

The book talks about the idea of focus investing and how one should use probabilities to remove unknowns and act where the odds are best. What does this mean?

Basically, I should increase my position when I’m confident that a stock in my portfolio is only facing short-term headwinds, the underlying business fundamentals remains solid, and there are more bullish signals than bearish ones.

Problem is, how should I quantify these using probabilities? I asked ChatGPT (I had to… hahaha!) but before going further into the mechanics, I’d like to state some assumptions made:

• This is a portfolio management method. This means it should apply to stocks that are already in my portfolio

• Thus, I should know its current intrinsic value for both bull and bear cases

With the above assumptions cleared, here’s what I learned:

#1: ⚖️Decide on position sizing

Focus investing means I should only focus on few stocks (5 to 10) where businesses I understand the most instead of having 50 stocks.

Divide these holdings with equal percentage of capital allocations. For example, if my starting capital is $50,000 and I plan to only manage 5 stocks in my portfolio, then each stock will be allocated with $10,000.

The allocation size will be adjusted accordingly based on the probabilities of the stock (explained in later steps). Simple.

#2: 🎲Determine the base-case probabilities

This is basically a starting point where I should assign what is the probability for bear, neutral and bull case scenario. The total of each scenario should equal 100. Different sector or business type can have very different sets of base-case probabilities. Why?

This is because each sector or business has different qualities. Financial data software companies like FactSet (FDS) typically has highly recurring revenue. Its sticky enterprise workflows provide them with moderate pricing power. So, its base-case probabilities can be as follows (a rough estimates):

• 🟡Base = 55 (neutral), this refers to how the stock should perform normally. For FDS, its sticky enterprise workflow will provide them with steady financial performance but slow growth in earnings.

• 🟢Bull = 25, this refers to how likely the stock will perform out of its norm (base). For FDS, the implementation of AI could drive more new customers sign-up and also higher spending from existing customers.

• 🔴Bear = 20, this refers to how likely the stock will under-perform its norm (base). For FDS, there are competition coming from various AI platform that can scrape financial data but its sticky enterprise workflow and institutional fund’s demand for high quality data will likely prevent its existing customers from cancelling their subscriptions.

#3: 🎰Adjust the base-case probabilities

The base-case probabilities described above are just a starting point. As the company releases more announcement such as earnings that talks about their business performance and outlook, I should adjust it accordingly.

This means I should constantly lookout for any bullish or bearish signals and +/- the probabilities determined in step #2. Keep in mind that the total of these probabilities should always equal to 100.

For example, FDS’ bull and bear signals are as follows:

With these, all 3 scenarios will get adjusted by subtracting from the base and +/ - into the bull and bear case. Here’s the updated probabilities:

Noticed that I have subtracted a total of 25 points from the base scenario because that is the total points from the signals noted above. Thereafter, I add 15 and 10 points respectively into bull and bear case.

I’m oversimplifying here since it’s just a concept. I’ll refine this further once implemented.

#4: 📊Calculate the Expected Value (EV)

With the adjusted base-case probabilities determined, I can then compute the expected value by doing a weighted average of the three outcomes. Below is the formula:

EV = (P_Bull x IV_Bull) + (P_Bear x IV_Bear) + (P_Base x IV_Base)

Where;

• P_Bull/ Bear/ Base = Probability of Bull/ Bear/ Base case

• IV_Bull/ Bear/ Base = Intrinsic value of Bull/ Bear/ Base case

Using FDS as an example again, I need to have 3 intrinsic value. One for each scenario (e.g. bull, bear and base). If you read my book on dividend investing, I’d like to use relative valuation.

For FDS, its 10-yr average PE is at 31.5x and I can use this as the IV for base scenario. For me to identify the IV for both bull and bear, I just need to use standard deviation on the 10-yr PE trends.

This means the IV for bull should be the +1 SD which is 37x while the IV for bear should be -1 SD which is 25x. Below is the calculated EV:

#5: 🚦Translate EV into Buying Signal

My next step is how should I interpret this EV? Because it does not tell me if I should buy aggressively, buy moderately, hold or sell.

ChatGPT suggested that I should establish some Rule-of-Thumb based on the ratio of EV and the current share price. Below is the suggested rule bands:

At the time of writing this journal, FDS’s EV-to-Price ratio is at 1.78x which suggests that I should buy aggressively.

#6: ⚖️Adjust Position Size

The next question is how much should I buy? In step #1, I established that my max allocation size per stock is $10,000 because I want to hold 5 stocks with the starting capital of just $50,000. This means 20% allocation per stock.

ChatGPT suggested below formula to determine the % I should allocate per stock based on the probabilities determined above:

Position Size = Max Allocation Size x (1 - Bear Case Probability)

Using FDS as an example, the % of capital I should allocate to buy this stock is 14%. This is because:

20% x (1 - 0.3) = 14%

Where;

• 20% = max allocation size per stock

• 0.3 = bear case probability for FDS

This kind of makes sense. The lower the bear case probability, the higher my allocation would be. Also, this strategy is conservative as it takes into account of the bear cases and does not allow me to max out the 20% allocation.

One of my principles is “never go all-in” because I might not have enough ammo to take advantage when there is a market downturn.

#7: ✂️When to Sell/ Exit?

With the allocation size established, it’ll now be easier for me to determine when should I trim or exit my position. This allocation size will become the threshold and it will be adjusted either quarterly or annually.

As long as the stock in my portfolio exit the threshold, I will start trimming or exit. For example, if FDS position in my portfolio reaches the max allocation size of 20%, I should trim it down to 14%. This is because the 14% is my ideal allocation size as calculated in Step #6.

If FDS has new bullish or bearish signals, I will update the bull, bear & base probabilities and recalculate the allocation size. From here, I will be able to determine whether I should add, trim or exit my position.

Conclusion

This is an interesting approach to portfolio management. It requires a lot of work to constantly monitor below items:

1. Bullish and bearish news of each stock currently owned

2. Bull, bear and base case probabilities

3. Position size adjustments

I’m thinking whether these can be automated by building an AI agent to help with some of these items.

I’ll try implementing this method manually while thinking of a way to automate this. That’s all for now, thanks for reading up to here. On to the next one…🙂

Disclaimer:

The information provided in this blog post is for informational purposes only and should NOT be construed as financial advice. Investing in stocks and ETFs involves risk, and there is no guarantee of profits. Past performance is not indicative of future results. It is important to conduct thorough research or consult with a qualified financial advisor before making any investment decisions. The author is NOT a financial advisor and is sharing his personal experiences and opinions only.

Additionally, please note that the author holds position in the discussed stocks, and his view may be biased as a result.

• Challenges: The DCF model's accuracy is highly dependent on the quality of input data, making it complex and less reliable.

• DCF result: The intrinsic value calculated is either too high or too low when compared with stock price

• Preferred method: Relative valuation is still the preferred method due to its simplicity

There are two ways to value a stock. One is to use the Discounted Cash Flows (DCF) Model and another is to use the Relative Valuation approach. The latter has been my go to valuation method. Simply because of 2 reasons:

• It is an objective way of valuing a stock

• It is quick and simple to use

You can grab a copy of my book here if you would like to learn more about relative valuation method. I explained about it in great detail there.

Conversely, the DCF model involves numerous variables that warrant consideration. Essentially, it's a case of 'garbage in, garbage out,' meaning the model's accuracy is directly influenced by the quality of the input data. This has always been my belief.

However, the primary reason I invest in stocks is to gain returns through dividends or capital gains. These returns are achievable only if the company’s business is able to grow its income and cash flows. That’s why it's undeniable that the valuation of a company is the sum of its future cash flows discounted to present value. This is the big concept of DCF model.

But the challenging part is the integrity of data input into the DCF model. There are 5 data fields requires in below 2 major sections within DCF model:

#1: Cash Flows Projections

As the name suggests, this refers to the company's cash flows. A key decision here is whether to use free cash flows or operating cash flows or net profit after tax.

Ideally, I should use free cash flows because they represent the balance amount a company can distribute as dividends, directly aligning with my investment objective.

However, not many companies can generate increasing free cash flows year over year. Companies in capital-intensive and cyclical sectors often experience volatile free cash flows. For example, those in property development, construction, energy, etc. As such, it is crucial to find companies with stable cash flows for this model to work.

Apart from this, there are 2 more decision points that I have to make.

Number of Years for Projection

How many years should I project the cash flows? Too short, the intrinsic value will be very low. Too far ahead, the intrinsic value will be too high. What is the reasonable number?

I don’t really have an answer here but I have seen many uses 10 years of projection. Some uses 20 years to project the cash flows. I guess ultimately it depends on one’s judgement on whether the business can continue to survive for the next 10, 20 or even 30 years.

But who can tell what will happen exactly in future, not to mention 10 years ahead?

Growth Rates

Choosing the right growth rates for projecting cash flows is crucial for this DCF model. Using high growth rates can inflate intrinsic value, and the reverse is also true. In the past, I relied on the 10-year average CAGR of revenue, net profit, and operating cash flows. However, I find this approach potentially misleading.

Past growth rates do not account for current business challenges that directly impact future growth. Further, companies go through different growth stages at different time period.

For example, a company might have grown at 30% annually 10 years ago when it was small and just starting out. However, a decade later, the same company might not sustain such growth as its business matures.

I still couldn’t find a solution for this part. I can only rely on below external sources:

• Guidance provided by Management

• Analyst’s estimated growth rates in their reports

• Websites such as Finviz, Simply Wall St, Seeking Alpha, etc.

Regardless, there are no such thing as “accurate” growth rates since it involves future outlook. Unless I have a crystal ball that can see the future, I tend to be conservative in determining the growth rates.

#2: Discount Rate

This is used to determine the present value of the projected cash flows. It represents the opportunity cost of capital, considering the risk associated with the investments & time value of money.

In other words, the discount rate reflects the return rate that investors require for an investment, balancing risk and potential reward. Higher discount rate indicates higher risk, resulting in lower present value and vice-versa.

To determine this discount rate, I use the “Capital Asset Pricing Model (CAPM)”. The formula is as follows:

Discount Rate (%) = Risk-Free Rate + (Beta * Equity Risk Premium)

CAPM ensures projected cash flows are appropriately discounted to reflect the risks & potential returns. It provides a systematic approach to determine the discount rate by incorporating the below 3 components:

Risk-Free Rate

This represents the return on an investment with zero risk. Typically, it is the 10-year government bonds’ interest rate or yield. Each country has different government bonds & therefore, different interest rate.

A simple google search would be able to find these figures. However, the challenge is that bond yield changes on daily basis. It is not fixed.

Equity Risk Premium

This is the additional return that investors expect from investing in stocks over the risk-free rate. Again each country has its own average equity risk premium. There’s no way I can reliably determine this figure on my own.

This is where I need to rely on the research of others such as the one in this website: link.

Stock’s Beta

This measures the volatility of a stock relative to the overall market. A higher beta indicates more risk and potentially higher returns, while lower beta indicates less risk. Similarly, there’s no way I can determine this on my own. Normally, I obtain these data from yahoo finance.

Challenges of DCF Model

Looking at all the above 5 data fields (projection years, growth rate, risk-free rate, equity premium rate & stock beta), it is impossible for me to derive the intrinsic value of a stock reliably.

I find myself constantly playing around these 5 data fields, especially the growth rates, whenever I get an intrinsic value that is either too high or too low than its current share price. There’s no way to confirm that the intrinsic value calculated is reasonable. One can only take it as it is.

Hence, the reason why I seldom use DCF model. If you would like to have the DCF Model file, you can download it by clicking here. Remember to download as Microsoft Excel file instead of Google Sheet.

Disclaimer:

The information provided in this blog post is for informational purposes only and should NOT be construed as financial advice. Investing in stocks and ETFs involves risk, and there is no guarantee of profits. Past performance is not indicative of future results. It is important to conduct thorough research or consult with a qualified financial advisor before making any investment decisions. The author is NOT a financial advisor and is sharing his personal experiences and opinions only.

• Unfocused portfolio: Current investments span various sectors making it difficult to track due to limited time

• Streamlining Research Strategy: Focus on sectors that align with long-term secular trends as they have the highest growth potential such as AI and Cybersecurity

• Diversification plan: Diversification is still important to avoid over-concentration risks. ETFs is a good tool to provide instant de-risk

Lately, I've been reflecting on how to maximize my limited free time for researching potential stock investments. I manage two portfolios: one focused on dividend growth and another on capital growth.

For my dividend growth portfolio, I concentrate on the markets in Malaysia, Singapore, and recently, Hong Kong. The main reason is that dividends in these markets are generally tax-free, except for Malaysian REITs, which have a 10% withholding tax.

My capital growth portfolio, on the other hand, is focused on the US market. This is because many US-listed companies have great businesses and long growth runways.

Maintaining these two portfolios is no easy task. I need to constantly monitor each position whenever earnings results are released. At the time of writing this journal, I have positions in the following sectors:

• Breweries

• REITs (retail, industrial, logistics & data centers)

• Banks

• Industrial materials

• Semiconductor

• Healthcare

• Cybersecurity

• E-commerce

• Artificial intelligence

• Communication services

• IT hardware & peripherals

Understanding the latest developments in all these sectors can be quite challenging and time consuming.

Howard Marks once said:

“Second-level thinkers know that, to achieve superior results, they have to have an edge in either information or analysis or both.”

This quote resonates with me. How can I achieve superior results if I’m constantly trying to understand the various sectors' latest developments instead of just focusing on what matters most?

It’s literally “Jack of all trades, master of none”. My analysis won't have an edge if I keep shifting focus between various sectors instead of concentrating on a few key ones.

That’s why I came to a realization that I need to streamline my research and focus on sectors that have the highest potential growth. This way, I can make better-informed decisions and maximize my returns.

Investing in Big Trends to Maximize Returns

Simply Wall Street in its weekly article identified several secular trends:

• Healthcare technology advancements

• Aging population and growing middle class consumers

• Acceleration in clean energy transition

• Artificial intelligence, data and cloud infrastructure

• Cybersecurity growing importance, etc.

These secular trends provide me with a guide to where my investment should be for the next 10 to 20 years. The idea is that companies within these secular trends are more likely to grow in terms of their earnings.

Now, I don't need to focus on every sector as above but rather on those with the highest potential for profits. For instance, artificial intelligence (AI) and cybersecurity are sectors I believe have the most impact on all other sectors and a long runway for growth because they act as enablers. They provide the tools, systems, and protections needed for other sectors to evolve and thrive.

Additionally, I’m choosing these sectors because of my IT background, being a CISA-certified professional. This makes it easier for me to understand the terminology and nuances when researching companies within these sectors.

My next problem is, how do I separate these into dividend and capital growth? Since I maintain 2 portfolios. I realized that I don't need to worry about whether an investment is for dividends or growth. Instead, I should view my portfolio as a whole and adjust according to the sectors I have chosen to focus on.

For example, within the AI value chain, there are various segments such as semiconductors, data centers, and networks. My research should identify companies within this value chain and classify them as either dividend or growth investments at later stage.

No doubt, most companies within this AI value chain are likely to be US-based. However, I believe other countries like Malaysia and Singapore also have companies within this value chain.

For instance, in Singapore, there are Keppel DC REIT and Mapletree Industrial Trusts, which invest in data centers. While in Malaysia, there are companies such as Time Dotcom and Tenaga National which provide connectivity and electricity to data centers, respectively.

What About Diversifications? Use of ETFs…

Yes, it is important to diversify my investments. I will definitely face huge fluctuations in my returns if I only focus on 1 particular sector.

Thankfully, there are various types of ETFs that can help me to diversify easily. So, I’ll be allocating some portion of my capital into ETFs as a way to diversify into other sectors that I won’t be focusing my research on.

This approach will help me make better use of my limited time and ensure my portfolio is well-aligned with long-term growth trends. I'm excited to start implementing this strategy and see how it transforms my investment journey this year (2025).

Perhaps, I will do an update to record how my portfolio has transformed later in the future.

Disclaimer:

The information provided in this blog post is for informational purposes only and should NOT be construed as financial advice. Investing in stocks and ETFs involves risk, and there is no guarantee of profits. Past performance is not indicative of future results. It is important to conduct thorough research or consult with a qualified financial advisor before making any investment decisions. The author is NOT a financial advisor and is sharing his personal experiences and opinions only.

Additionally, please note that the author holds position in the discussed stocks, and his view may be biased as a result.