Selected semiconductor stocks (>$50B market cap) by next-FY forward P/E — May 2026

# Company Fwd P/E EPS Growth PEG
1 SK Hynix 4.3× +23.2% 0.18
2 Samsung 5.1× +18.1% 0.28
3 Micron Technology 5.4× +69.5% 0.08
4 NXP Semiconductors 17.4× +20.9% 0.83
5 NVIDIA 17.5× +34.8% 0.50
6 MediaTek 19.3× +59.1% 0.33
7 TSMC 21.8× +25.0% 0.87
8 Infineon 22.9× +55.0% 0.42
9 Broadcom 23.4× +56.3% 0.42

Uses next-fiscal-year forward P/E (not current-year). N.M. = not meaningful when EPS growth is negative or near zero. Point-in-time snapshot — May 2026. These figures will change. Source: StockAnalysis / Finhub / MarketScreener analyst consensus. For current valuations visit each company's page on StockAnalysis: Micron · NVIDIA · TSMC · SK Hynix · Samsung

Why are memory chip companies trading at single-digit P/E?

The first thing to notice about the May 2026 semiconductor valuation data is which companies sit at the bottom of the forward P/E ranking. SK Hynix, Samsung, and Micron are primarily memory semiconductor companies — manufacturers of DRAM (the RAM in your laptop) and NAND flash (the storage in your phone and SSD). These are commodity-like products with minimal brand differentiation.

Unlike logic chips — NVIDIA's AI accelerators, Broadcom's networking chips — memory chips trade on price. When global inventory swells past demand, prices fall hard. Earnings collapse, P/E ratios spike. When supply tightens and demand recovers, earnings rebound and P/E compresses. That is the memory semiconductor cycle, and it repeats roughly every 3–5 years.

The 2022–2024 period was a severe downturn: a post-pandemic inventory correction caused memory prices to fall 50–70% from their peaks. Earnings troughed across the sector. By 2025–2026, analyst consensus began pointing toward a sharp earnings recovery — hence the low forward P/E values, which divide today's share price by projected future earnings rather than past earnings.

A low forward P/E in memory chips does not mean the market has overlooked something. It means the market already expects earnings to recover. Whether that recovery arrives on schedule — and at the magnitude analysts project — is an entirely separate question.

What forward P/E and PEG ratio mean in plain language

Two metrics from the data table are worth understanding before drawing any conclusion from them.

Forward P/E is a company's current share price divided by its next-year estimated earnings per share (EPS). A forward P/E of 5x means you are currently paying five times what the company is expected to earn per share next year. Lower is generally considered "cheaper," but only if those earnings forecasts materialise. The key caveat: analyst forecasts for cyclical companies are frequently revised, sometimes dramatically.

PEG ratio (price/earnings-to-growth) adjusts for growth by dividing the forward P/E by the expected EPS growth rate. It addresses the fact that a high-growth company rationally commands a higher P/E than a slow-growth one. A PEG below 1.0 is often read as a company growing faster than its current valuation implies.

Two companies from the table illustrate the difference:

As shown in the table above, this data uses next-fiscal-year P/E — one earnings forecast period further out than the conventional one-year forward P/E. This captures a more advanced stage of the recovery cycle and produces lower P/E ratios for companies with strong projected earnings growth.

Why acting on a valuation screen alone is risky

A valuation table is a snapshot of consensus estimates, not a roadmap. Three risks matter particularly in the semiconductor context.

Cycle extension. Memory downturns can last longer and go deeper than consensus expects. Micron traded at a low forward P/E for much of 2022 and 2023 while its share price continued declining. "Cheap" can become cheaper when earnings recovery is delayed by supply discipline failures or softer-than-expected end demand.

Forecast variance. EPS growth projections for memory companies carry wide error bars. A 69% projected recovery could land at 30% or 100% — both are within the historical range of analyst accuracy for cyclical commodity businesses. The PEG ratio shifts considerably depending on which outcome materialises. Past performance of analyst forecasts is not a reliable indicator of future forecast accuracy.

Concentration risk. Buying one or two semiconductor names creates concentrated exposure to a single subsector of a single industry, denominated in currencies other than MYR, traded on exchanges with their own regulatory environments, tax treatments, and settlement mechanics.

The appropriate response to an interesting valuation screen is not a single-stock trade. It is portfolio-level thinking about whether a sector belongs in your long-term asset allocation.

The Malaysian investor's access barriers to foreign semiconductor stocks

Before a Malaysian retail investor can hold SK Hynix or Samsung, there is a setup problem that most valuation discussions skip entirely.

SK Hynix and Samsung trade on the Korea Stock Exchange (KRX) in Korean Won (KRW). Micron, NVIDIA, AMD, and Broadcom trade on US exchanges in USD. TSMC trades on the Taiwan Stock Exchange (TWSE) in TWD, though its ADR (ticker: TSM) is USD-denominated and listed on the NYSE. (StockAnalysis — TSMC)

To hold any of these directly as a Malaysian resident, you need:

None of these barriers are insurmountable. But they add friction, ongoing cost, and risk variables that a first-time global investor may not have modelled into their expected return.

How global equity unit trusts bridge the gap

A unit trust fund registered with the Securities Commission Malaysia (SC) with a global equity or technology mandate can hold the same underlying companies — SK Hynix, Samsung, Micron, NVIDIA, TSMC — without requiring the individual investor to open a foreign brokerage account or hold foreign currency.

The structure works as follows: the fund is denominated in MYR. Investors subscribe and redeem in ringgit. The fund manager — operating under the SC-approved mandate and investment guidelines — purchases the underlying foreign securities on the pooled fund's behalf and manages the portfolio.

What this means for a Malaysian investor:

The investor does not select which semiconductor stocks the fund holds — that decision sits with the fund manager operating within the approved mandate. What the investor chooses is the fund category: global equity, technology sector, or a specific geographic focus.

Investment in unit trust funds carries market risk. The value of units may go up or down. This is not a recommendation to invest in any specific fund. Refer to the relevant Master Prospectus and Product Highlights Sheet before investing.

What to look for in a global technology unit trust

This is not a fund recommendation — matching a specific fund to your financial situation requires a documented Suitability Assessment with a licensed UTC consultant. What follows is an educational checklist of questions to ask when reading any fund's published documents.

All of this information is in the fund's Master Prospectus and Product Highlights Sheet, which your licensed UTC consultant should walk through with you before any subscription.

The parent investor's real question: cycles vs. compounding

A semiconductor analyst tracks quarterly earnings, DRAM spot prices, and cloud hyperscaler inventory levels. Their job is to anticipate earnings revisions before the market does. That is a full-time discipline requiring sector-specific expertise and tolerance for being wrong at exactly the wrong moment.

A Malaysian parent building an education fund or retirement portfolio has a different job: structure a consistent, long-horizon investment that compounds over 10–20 years, absorbs multiple market cycles without requiring tactical decisions every quarter, and does not depend on any single bet being correct.

The right question for that parent is not "is SK Hynix cheap today?" It is: "Will the global semiconductor sector — chips for AI systems, electric vehicles, 5G infrastructure, cloud computing, and consumer electronics — be larger and more integral to the global economy in 2036 than it is in 2026?"

On a 10–15 year horizon, that is a more tractable question than predicting whether the memory cycle recovery lands in Q3 2026 or Q1 2027.

A global equity or technology unit trust with monthly auto-debit (Dollar-Cost Averaging, or DCA) participates in this long-term growth thesis without requiring the investor to time cycles. When fund NAV is lower — as it may be during a sector downturn — a fixed monthly contribution buys more units. When markets recover, those additional units appreciate. The cycle becomes a structural advantage rather than a timing problem.

That is the parent investor's version of the semiconductor trade: not a bet on a single cycle bottom, but consistent participation across multiple cycles, compounded across a decade-plus horizon.