Monday, 24 November 2025

KPJ: Surge in the P/E and P/B multiples (2023-2024)

Summary:  Surge in the P/E and P/B multiples (2023-2024) coincides

with KPJ achieving its highest ever PBT Margins and ROE.  


KPJ Healthcare: 10-Year Financial Summary (Adjusted for Latest Capital Structure)

All monetary values in RM '000, except per-share data. 

Per-share data is adjusted for NOSH of 4,363,857.








This adjusted dataset allows for a proper analysis of KPJ's valuation 

metrics (like P/E and P/B ratios) over the entire 10-year period, as 

if the company's current share capital had always been in place.


Calculated Financial Ratios for KPJ Healthcare

























Key Observations:

Profitability (PBT & Net Margin): There is a clear trend of 

significantly improving profitability from the low in 2020, with 
margins in 2023-2024 reaching their highest point in the last decade.
Return on Equity (ROE): The ROE has followed a similar trajectory
to the profit margins, showing strong improvement in recent years. 
The exceptionally high ROE in 2015 (38.14%) is an outlier and is 
directly caused by the very low Adjusted NTA (RM 0.0923) in that
 year, which significantly inflates the ratio. This low NTA is consistent
 with the major capital change that occurred.
Dividend Payout (DPO): The DPO ratio has become more stable 
and higher in recent years (around 51-55%), indicating a commitment
to returning more cash to shareholders as profits have grown. The 
very low DPO in 2015 (11.65%) is consistent with the capital 
restructuring event that likely occurred around that time.

KPJ Healthcare: Valuation & Profitability Metrics (Adjusted)














Key Observations:

Valuation (P/E Ratio): 

- The P/E ratio was extremely high in 2020 (83.76) 

because earnings (EPS) fell sharply during that period, while the share 

price did not drop as precipitously.

- The current P/E (33.74 for 2024) is at the high end of the historical range, 
reflecting market expectations for continued strong earnings growth.
- The period from 2021 to 2023 shows a fascinating trend: as earnings (EPS)
recovered and grew strongly, the share price initially lagged (leading to 
a low P/E of 17.58 in 2022) before surging ahead (resulting in a higher 
P/E of 33.74 in 2024).

Valuation (P/B Ratio):

- The P/B ratio has consistently been above 2.0 for most of the last decade, 
indicating the market values KPJ at a significant premium to its book value.
- The current P/B of 4.71 is the highest in the last 10 years (excluding
the  2015 outlier), signaling very strong market confidence in the 
company's future profitability and asset utilization.
- The 2015 P/B of 11.27 is an outlier, directly caused by the abnormally
low Adjusted NTA (RM 0.0923) following the capital restructuring, which
artificially inflated the ratio.

Profitability vs. Valuation:

- The recent surge in the P/E and P/B multiples (2023-2024) coincides 
with KPJ achieving its highest-ever PBT Margins and ROE, justifying 
the market's higher valuation to a large extent.

Is NVIDIA a buy, a hold or a sell at today's price?

Summary: NVIDIA's phenomenal business prospects are fully appreciated by the market. Future returns will be driven by the hard work of earnings growth rather than multiple expansion. For a long-term investor, it remains a foundational holding for exposure to AI, but new capital might find better risk-reward opportunities on market dips.



Based on a synthesis of the financial data, competitive position, and market dynamics, here is an assessment and forward-looking return expectations for NVIDIA.

Arguments for HOLD

Reasons Against a Strong BUY:

  1. Priced for Perfection: At a P/E of ~53x trailing earnings, the stock embeds extremely high expectations. Any stumble in execution, a slower-than-expected adoption curve for AI, or a significant margin contraction would likely trigger a sharp correction. There is very little room for error.

  2. Law of Large Numbers: Growing revenue at >100% from a base of $130 billion is mathematically improbable. The current valuation requires sustained hyper-growth for several more years. As the company matures, growth will decelerate, and the multiple will likely contract simultaneously (a phenomenon known as "multiple compression").

  3. Intensifying Competitive Threats: The bear case is not trivial. Competitors like AMD are gaining traction, and the largest customers (Microsoft, Google, Meta, Amazon) are all developing their own custom AI chips to reduce dependency. This could erode NVIDIA's pricing power and market share over the long term.

  4. Cyclicality Risk: Semiconductors are inherently cyclical. While the AI boom may extend the cycle, a global economic slowdown or a reduction in enterprise AI spending could lead to a painful inventory correction.

Reasons Against a SELL:

  1. Unparalleled Dominance and Moat: The combination of its hardware, the critical CUDA software ecosystem, and its full-stack platform creates a defensive moat that is arguably the strongest in the tech sector. This is not a company that will be displaced easily or quickly.

  2. Massive Total Addressable Market (TAM): The AI revolution is in its early innings. Applications in enterprise software, robotics, automotive, and biology represent a TAM of over $1 trillion. NVIDIA is the primary infrastructure provider for this shift.

  3. Proven Execution and Financial Fortress: The management team has executed flawlessly. The company has a pristine balance sheet with massive cash flow, allowing it to invest aggressively in R&D and return capital to shareholders, insulating it from short-term shocks.

Conclusion for HOLD: For new investors, waiting for a more attractive entry point (a market pullback or a period of consolidation) would be prudent. For existing investors, selling a company with such a strong competitive position and growth runway is likely a mistake. The optimal strategy is to hold and let the company's earnings growth eventually justify the valuation.


5-Year Total Return and Annualized Return Expectations

Predicting returns for a stock like NVIDIA is speculative, but we can build a realistic model based on fundamental drivers: Earnings Growth and Multiple Change.

Let's use the following Base Case scenario:

  • Starting Point (FY2025):

    • EPS = $3.39

    • P/E Ratio = 52.8x

  • 5-Year Assumptions:

    1. EPS Growth: Growth will decelerate but remain very strong.

      • Years 1-2: ~30% annual growth (down from ~150%).

      • Years 3-5: Growth decelerates to ~20% annually.

      • This results in a 5-year EPS CAGR of approximately 22%.

    2. P/E Multiple Contraction: As growth normalizes, the market will assign a lower multiple. A P/E of 30-35x is more typical for a high-quality, mature growth company.

      • We assume the P/E contracts to 32x by Year 5.

Base Case Calculation:

  1. FY2030 EPS: $3.39 * (1.22)^5 = $3.39 * ~2.70 ≈ $9.15

  2. FY2030 Share Price: $9.15 (EPS) * 32 (P/E) = $292.80

  3. Total Return: (($292.80 / $178.88) - 1) * 100 = +63.7%

  4. Annualized Return (CAGR): (1.637)^(1/5) - 1 = ~10.4% per year

Scenario Analysis:










Final Conclusion

  • Going Forward: Investors should expect volatility. The path to these returns will not be smooth.

  • Realistic Expectation: The Base Case annualized return of ~10% is a realistic expectation from today's price. This would be a strong absolute return, roughly doubling the historical market average, but it is a far cry from the 80%+ annualized returns of the recent past. The era of easy, explosive gains is likely over.

  • Primary Risk: The biggest risk is the Bear Case scenario, where growth decelerates rapidly and the P/E multiple contracts sharply, leading to a flat or negative return over five years.

Summary: NVIDIA is a "Hold" because its phenomenal business prospects are fully appreciated by the market. Future returns will be driven by the hard work of earnings growth rather than multiple expansion. For a long-term investor, it remains a foundational holding for exposure to AI, but new capital might find better risk-reward opportunities on market dips.







































































Recommendation: HOLD

This is not a strong "BUY" at current levels, but it is absolutely not a "SELL." The recommendation is nuanced and highly dependent on an investor's existing position, risk tolerance, and time horizon.

This is not a recommendation.  Please do your own research and decide accordingly.


Additional Notes:

Benjamin Graham:

Purchase of "GROWTH STOCKS" at GENEROUS PRICES is speculation..

(Should set aside a sum for this separate from their money in investing.


Investment and speculation  (Security Analysis, Ben Graham.):

1. Graham defined investment thus: 
An INVESTMENT OPERATION is one which, upon THOROUGH ANALYSIS, promises SAFETY OF PRINCIPAL and a SATISFACTORY RETURN. Operations NOT meeting these requirements are speculative.

The difference between investment and speculation, when the two are thus opposed, is understood in a general way by nearly everyone; but it can be difficult to formulate it precisely. In fact something can be said for the cynic's definition that an investment is a successful speculation and a speculation is an unsuccessful investment.

The failure properly to distinguish between investment and speculation was in large measure responsible for the market excesses and calamities that ensued, as well as, for much continuing confusion in the ideas and policies of would-be investors.

2. Graham's addition criterion of investment: An investment operation is one that can be justified on BOTH QUALITATIVE and QUANTITATIVE grounds.

Investment must always consider the PRICE as well as the QUALITY of the security.






How to value NVIDIA

 Valuing NVIDIA is one of the most fascinating and challenging tasks in the market today. It's a company that has transcended its original industry and become a foundational player in the most significant technological shift of the decade: Artificial Intelligence.

There is no single "correct" value, but we can build a robust valuation by looking through different lenses. The core challenge is balancing its astronomical growth and current dominance against the inevitable uncertainties of competition, market evolution, and technological disruption.

Here is a multi-faceted approach to valuing NVIDIA.

1. The Narrative & Qualitative Foundation

Before any numbers, you must understand the story. The valuation is entirely predicated on the belief that:

  • The AI Boom is Real and Long-Lived: This is not a short-term fad. AI is being integrated into every layer of the global tech stack.

  • NVIDIA is the "Picks and Shovels" King: While others build AI applications (the "gold miners"), NVIDIA sells the critical hardware (GPUs) and software (CUDA) needed to do the work. This is a less risky, highly profitable position.

  • The Moat is Durable: The combination of cutting-edge hardware, the entrenched CUDA software ecosystem, and its full-stack platform (from chips to DGX systems to software libraries) creates a competitive advantage that is extremely difficult to replicate.

2. Valuation Methodologies & Calculations

Using the data we have (FY2025 = Jan 2025), let's apply different models.

Lens A: Relative Valuation (Comparing to the Market)

This is the most common approach. We'll use the adjusted EPS we calculated.

  • Current Trailing P/E: As of FY2025, the P/E is ~53x ($178.88 / $3.39).

  • Comparison: This is high compared to the overall market (S&P 500 ~20-25x) but arguably cheap for a company growing earnings at >100%.

The core of the relative valuation is a Forward P/E analysis.

  1. Estimate Future Earnings: Analysts' estimates vary, but let's assume a conservative 30% EPS growth for FY2026 (a significant slowdown from the ~150% of FY2025). This is a critical and debatable assumption.

    • FY2026 EPS Estimate = $3.39 * 1.30 = $4.41

  2. Apply a Forward P/E Multiple: What multiple is justified?

    • Bull Case (Optimistic): If the growth story remains intact, a P/E of 40-50x could be justified.

      • Valuation = $4.41 * 45 = ~$198 per share

    • Base Case (Cautiously Optimistic): A P/E of 30-35x as growth normalizes.

      • Valuation = $4.41 * 32.5 = ~$143 per share

    • Bear Case (Pessimistic): Growth slows faster than expected or competition intensifies, leading to a P/E of 20-25x.

      • Valuation = $4.41 * 22.5 = ~$99 per share

This simple model shows a wide range, entirely dependent on the growth and multiple assumptions.

Lens B: Discounted Cash Flow (DCF) - The Intrinsic Value

A DCF model values a company based on the present value of its future cash flows. Given the volatility, a 2-Stage DCF is appropriate.

  • Stage 1 (High Growth - Next 5 years): Assume Free Cash Flow grows at a high but decelerating rate (e.g., 50% -> 25%).

  • Stage 2 (Terminal Growth - Perpetuity): Assume a modest, perpetual growth rate (e.g., 3-4%), in line with the long-term global economy.

Using the Free Cash Flow of $60,853M from FY2025 and making a set of reasonable (but optimistic) assumptions:

  • High-growth period: 5 years

  • Initial FCF Growth: 40%, declining to 15%

  • Discount Rate (WACC): 10% (reflecting high risk)

  • Terminal Growth: 3.5%

A back-of-the-envelope DCF calculation under these conditions can yield an intrinsic value per share in the range of $150 - $180. This aligns broadly with the Forward P/E analysis.

Lens C: Market-Based & Sum-of-the-Parts

  • Total Addressable Market (TAM): NVIDIA itself estimates its TAM in Data Centers at ~$1 Trillion by 2030. If NVIDIA can capture even 10-15% of this TAM, its revenue could be $100-150B, supporting a much higher market cap.

  • Comparables: It's almost in a class of its own. Comparisons are difficult, but it trades at a premium to other semiconductor giants like TSMC and AMD, which the market accepts due to its superior growth profile and platform moat.

Synthesis: The Bull vs. Bear Debate

The Bull Case (Justifying a High Valuation):

  • Unmatched Dominance: >80% market share in AI accelerators.

  • Software Moat: CUDA is the defacto standard, creating immense customer lock-in.

  • Expanding TAM: AI is moving into new domains: automotive, robotics, healthcare, etc.

  • Rising Guidance: The company consistently beats and raises its own forecasts, suggesting visibility is strong.

  • Valuation is Actually Reasonable: A PEG Ratio (P/E / Growth) of less than 1 (e.g., P/E of 50 with growth of 100%) is considered "cheap" for a growth stock.

The Bear Case (Risks that Could Derail the Valuation):

  • Valuation Bubble: The stock is priced for perfection. Any stumble in growth will lead to a severe multiple contraction (as we saw in the P/E CAGR calculations).

  • Intense Competition: AMD, Intel, and custom silicon from cloud giants (Google's TPU, Amazon's Trainium) are real threats.

  • Customer Concentration: A significant portion of sales goes to a few large cloud companies, who are also their competitors.

  • Cyclicality: The semiconductor industry is historically cyclical. An industry downturn is a matter of "when," not "if."

  • Technological Disruption: A new, more efficient AI architecture could make the GPU obsolete.

Conclusion: How to Value NVIDIA

You don't value NVIDIA on its past performance but on your belief in the duration and magnitude of the AI growth cycle.

  1. Base Case Fair Value: Based on a synthesis of the models above, a fair value range in the current environment appears to be $140 - $170 per share. At ~$179 (the given FY2025 price), the market is pricing in a very strong, but not impossible, growth trajectory.

  2. It's a Bet on the Future: Buying NVIDIA at this level is a conviction that the AI-driven demand for its products will continue to surprise on the upside for years to come, justifying its premium multiple.

  3. The Key Metric to Watch: The most important indicator won't be the next quarter's revenue, but the guidance for future quarters and any signs of gross margin pressure. The narrative is everything.

In essence, valuing NVIDIA is less about a precise DCF calculation and more about assessing the probability and longevity of the AI paradigm shift. The stock is a high-risk, high-reward asset that embodies the market's collective belief in that future.