Context This is a follow-up to Samsung and SK Hynix Are 90.8% of Korean Semis. It also connects to Korea theme ETF rebalance flow, ETF flow is leading the Korean market, the HBM hub, the AI PCB and substrate hub, and the Exclusive Analysis hub.
TL;DR
- Samsung Electronics and SK Hynix remain the engine of Korean semiconductors. But for relative alpha over the next two months, the better candidates may be second-line equipment, substrate, inspection and AI storage names where ETF ownership is still lower and flow is just beginning to build.
- A scan of 1,137 Naver ETFs, Korean semiconductor Top 50 constituents, and local price/flow data highlights Korea Circuit, HPSP, TES, VM and FADU.
- Korea Circuit is the cleanest balance of low ETF exposure, institutional accumulation and substrate catalysts. HPSP has the best flow quality. TES and VM have strong fundamental triggers but have already moved very fast. FADU is the high-risk convex option.
1. Method
This is not a generic semiconductor watchlist. It is a screen for Korean semiconductor Top 50 stocks that could outperform Samsung Electronics and SK Hynix over a two-month horizon.
| Input | Coverage |
|---|---|
| Universe | Korean semiconductor Top 50 |
| ETF data | 1,137 Naver ETFs and 75,001 constituent rows |
| Price and flow | Thesis OS local database, as of June 12, 2026 |
| Windows | 2-week, 1-month and 3-month returns plus foreign/institutional flow |
| ETF exposure | Estimated ownership from ETF NAV and constituent weights |
The ETF exposure figure is a proxy, not official PCF data. It is useful for comparing ownership pressure, but it does not directly measure creation/redemption flow.
2. Baseline: Samsung and SK Hynix
| Stock | 2W | 1M | 3M | Estimated ETF exposure | ETF count | 1M foreign + institution |
|---|---|---|---|---|---|---|
| Samsung Electronics | +7.7% | +15.6% | +69.7% | KRW 51.5tn | 229 | -KRW 15.2tn |
| SK Hynix | -6.1% | +17.2% | +125.1% | KRW 52.1tn | 214 | -KRW 15.3tn |
These two stocks remain core exposure. But they already sit inside more than 200 ETFs each, with estimated ETF ownership above KRW 50tn. That makes them the market beta. The alpha question is different: where can incremental flow have more price impact?
3. Compressed Candidate List
| Rank | Stock | Segment | Main evidence | View |
|---|---|---|---|---|
| 1 | Korea Circuit | Substrate | ETF exposure only KRW 85.6bn, 1M foreign+institution +KRW 66.6bn, distributed institutional buying | Best balance |
| 2 | HPSP | Equipment | 1M foreign+institution +KRW 197.2bn and 2W +KRW 196.5bn | Best flow quality |
| 3 | TES | Equipment | 1M return +67.8%, SK Hynix equipment contract | Strong but hot |
| 4 | VM | Equipment | Low ETF exposure, strong 2026E growth, institutional accumulation | Under-owned equipment beta |
| 5 | FADU | Fabless / AI storage | 2W pullback, heavy foreign buying, SSD controller contracts | High-risk convexity |
4. Stock Notes
Korea Circuit
Korea Circuit is the most balanced candidate. It is not yet crowded in ETF ownership, but the price and institutional flow have already started to move. The thesis is not just “substrates.” It is SOCAMM plus FC-BGA plus Broadcom AI switch exposure. Prime Economy reported that Meritz expected FC-BGA utilization to rise from 55% in 2025 to 70% in 2026 and 85% in 2027, and forecast SOCAMM revenue growth of 301% in 2027. Prime Economy
The risk is that the move has already begun: 1M return +36.9% and 3M return +152.6%. It is a watch / buy-on-pullback candidate.
HPSP
HPSP has the cleanest flow. Foreign investors and institutions are both buying, and the 2-week flow almost matches the 1-month flow, which means the accumulation has not faded. The business quality is also clear: 2026E OPM is 53.4%.
The issue is price. The stock is up 40.5% over two weeks and trades at 50.9x 2026E PER. It needs a pullback or a stronger 2Q confirmation before aggressive entry.
TES
TES has the clearest near-term catalyst. DataTooza reported a KRW 21.2bn semiconductor equipment contract with SK Hynix, equal to 6.04% of recent sales, and noted that announced supply contracts in 2026 year-to-date were up 81.1% year over year. DataTooza
This is why TES screens well on fundamentals. But the stock is already up 67.8% in one month. It is a buy-on-pullback candidate, not a chase candidate.
VM
VM is an institutional accumulation story rather than a foreign-flow story. Foreign investors have sold, but institutions have bought aggressively. The 2026E setup is attractive: revenue +118.6%, operating profit +281.0%, OPM 29.8%, and PER around 30.0x.
Chosun Biz reported that Daol Investment & Securities viewed VM as undervalued among Korean semiconductor equipment names and set a KRW 140,000 target price. Chosun Biz
The risk is short-term overheating. It needs a pullback and evidence that foreign selling is fading.
FADU
FADU is different from the others. It is not a substrate or equipment name. It is an AI storage and SSD controller option.
The catalyst density is high. The Bell reported that FADU signed a KRW 50bn enterprise SSD controller contract with an overseas NAND maker, equal to 54.1% of the prior year’s revenue, and that announced orders had exceeded KRW 216.3bn after the contract. The Bell
The flow is powerful but not clean: foreign investors bought KRW 460.4bn over one month while institutions sold KRW 296.2bn. FADU is therefore an event option, not a core position.
5. Ranking
For two-month relative alpha:
| Rank | Stock | Why |
|---|---|---|
| 1 | Korea Circuit | Best mix of ETF under-ownership, institutional flow and substrate catalysts |
| 2 | HPSP | Best flow quality, but needs a better entry |
| 3 | FADU | Pullback plus foreign accumulation, high-risk convexity |
| 4 | TES | Strongest beta, but too hot near term |
| 5 | VM | Institutional accumulation and low ETF exposure |
For fundamental catalyst strength:
| Rank | Stock | Why |
|---|---|---|
| 1 | TES | Contract evidence and SK Hynix equipment linkage |
| 2 | VM | Large 2026E earnings step-up and new customer/process option |
| 3 | Korea Circuit | SOCAMM + FC-BGA dual catalyst |
| 4 | HPSP | High quality but needs fresh earnings proof at a high multiple |
| 5 | FADU | Strong contracts, but high valuation and institutional selling |
Final View
Samsung Electronics and SK Hynix remain the center of Korean semiconductors. But if the question is which names can beat them over the next two months, the answer likely sits in second-line names where ETF ownership is still lower and flow is improving.
The cleanest single name is Korea Circuit. The highest-quality flow is HPSP. The strongest fundamental catalysts are TES and VM. The most convex option is FADU.
The common rule is simple: do not chase vertical moves; wait for pullbacks, check flow persistence, and demand evidence that the theme is becoming revenue.