Hanmi Semiconductor: Overvalued, Maintain Underweight, Raise Target Price to KRW 200,000
AI summary card
Hanmi Semiconductor: Overvalued, Maintain Underweight, Raise Target Price to KRW 200,000
J.P. Morgan acknowledges underestimating memory capex cycle intensity and raises target price, but maintains Underweight rating due to valuation premiums far exceeding peers and HBF opportunities being premature.
- Maintain Underweight rating, raise target price from KRW 150,000 to KRW 200,000
- Current stock price trades at 62% premium to backend equipment peers, significantly above historical average of 41%
- Lower FY26-27 EPS estimates by 9-17%, reflecting order reductions and share loss
- FY27-28 operating profit forecasts are 15-31% below market consensus
- Prospects for High Bandwidth Flash (HBF) applications remain unclear; pricing is too early
- Micron (MU) orders expected to replace SK hynix as key short-term growth driver
Report interpretation
Overview
J.P. Morgan released an earnings review for Hanmi Semiconductor (042700.KS). While acknowledging prior underestimation of the strength and duration of the memory capex upcycle and raising the target price from KRW 150,000 to KRW 200,000, the firm maintains an 'Underweight' rating. The report argues that despite improved industry conditions, Hanmi's current valuation premium versus global backend equipment peers has significantly exceeded the historically reasonable range. Additionally, the market is pricing in long-term tailwinds such as High Bandwidth Flash (HBF) too early, combined with earnings forecasts significantly below market consensus, resulting in an unattractive overall risk-reward profile.
Core views
Valuation and Rating Logic: The report notes that Hanmi Semiconductor currently trades at a premium of approximately 62% compared to global backend equipment peers, whereas the average premium since the second half of 2023 was only 41%. Based on assumptions of a 45x FY27/28E P/E ratio and a 45% peer premium, the target price was raised to KRW 200,000 to reflect an extended capital expenditure cycle. However, the current share price of 284,500 KRW has already discounted fundamental drivers. Core reasons for maintaining the Underweight rating include excessive valuation stretching, lack of empirical support for HBF opportunities, limited EPS upside from extended TCB lifecycle, and a 15-31% downside gap in earnings forecasts for 2027-28 compared to market consensus. Earnings Forecast Adjustments and Order Structure: Following Q1 earnings impact, the report lowered FY26-27E EPS forecasts by 9-17%. This primarily reflects reduced TCB order volumes in H2 2025, a slight decline in SK hynix's internal market share from 55% to 50% (with competitors securing similar orders), and a significant reduction in replacement demand. Due to the slow pace of TCB ordering by SK hynix year-to-date (approximately 100k wpm capacity equivalent), Micron's (MU) orders are expected to become the key growth engine from H2 2026 through 2027, accounting for over 50% of TCB sales during that period. Looking ahead to 2028, as HBM total capacity reaches approximately 1 million wpm, the TCB market size is expected to reach $1 billion, though sales growth may decelerate from 44% in 2027 to 13%. Uncertainty Regarding New Technologies and Applications: Regarding the market's focus on High Bandwidth Flash (HBF) opportunities, the report deems current pricing premature. Although the company plans to demonstrate HBF TCB equipment to NAND manufacturers by mid-2026, and the higher number of NAND manufacturers implies a potentially large market, product specifications (stack layer count), end-use applications, and customer roadmap adoption remain unclear. Responses from various memory vendors regarding mass production timing also vary (ranging from 2027 to 2030). Furthermore, JEDEC may relax stacking height limits for HBM5 to 20-Hi, increasing uncertainty regarding the Hybrid Bonding technology route. The report does not factor into its forecasts any expectation of Hybrid Bonding replacing TCB equipment within its forecast horizon.
Analysis framework
The report adopts a 'Relative Valuation + Industrial Cycle Validation' analytical framework. First, it assesses the rationality of current valuations by comparing Hanmi Semiconductor's P/E premium against global backend packaging equipment peers, benchmarking against historical premium centers rather than relying solely on absolute valuation. Second, in terms of fundamental analysis, it employs a bottom-up order tracking method, breaking down TCB equipment demand into specific order rhythms for SK hynix, Micron, and Chinese storage vendors, calibrated with HBM capacity expansion plans to refine revenue forecasts. Finally, it evaluates the probability of long-term growth story realization through technical roadmap cross-validation (such as JEDEC standard changes and HBF customer validation progress), thereby distinguishing between 'realized cycle dividends' and 'unsubstantiated concept premiums'.
Methodology notes
Relative Valuation Premium Analysis
Beyond absolute P/E ratios, the core focus is on observing the valuation premium rate of the target company relative to 'global backend equipment peers.' By comparing the current premium (62%) with the historical average (41%), the report determines whether the stock price has deviated from the reasonable range supported by fundamentals. This approach is commonly used for segment leaders to prevent blindly assigning high multiples during periods of strong industry cyclicality.
Equipment Orders Linked to Capacity Model
For specialized semiconductor equipment like TCB, demand depends directly on capacity construction at specific downstream process nodes. The report derives equipment purchase volumes by analyzing SK hynix and Micron's HBM capacity plans (wfpm) and capex pacing, rather than relying solely on macro industry growth rates. This micro supply-demand matching is key to anticipating short-term performance inflection points for equipment vendors.
Pricing Mismatch Between Concept Introduction and Mass Production Phases
The report emphasizes distinguishing between 'vendor-driven' and 'customer-driven' technology stages. For new technologies like HBF, when only vendor sampling exists without a clear customer mass production roadmap, it belongs purely to a concept speculation phase; only when end applications are clear and standardized can they be included in earnings forecasts. This methodology helps investors identify which positives are overly priced and which require catalysts.
Asset mapping & comparison
Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).
- Hanmi Semiconductor (042700.KS)Analyzed Subject: TCB Equipment Leader, Benefited by HBM Expansion but Constrained by Valuation
- Strengths
- Dominant position of TCB equipment in HBM packaging; benefited by extended memory capex cycle; Micron order ramp provides near-term support
- Weaknesses
- Valuation premium far exceeds historical average; uncertain HBF commercialization timeline; slight decline in SK hynix internal share; earnings forecasts significantly below consensus
- Comparison
- Premium of 62% over global backend equipment peers (historical average 41%); FY27/28E P/E ratios are 66x/59x, while peers trade at 34x/28x
- Risks
- HBM ASP increases missing expectations; HBF standardization delays; Samsung Electronics TCB orders falling through; weak purchasing by Chinese storage vendors
Key data
- Target Price200,000 KRWBased on 45x FY27/28E EPS and 45% peer premium, up from previous 150,000 KRW
- FY26E EPS Adjustment3,192 KRWDecreased by 8.6% from previous forecast
- FY27E EPS Adjustment4,275 KRWDecreased by 17.1% from previous forecast
- Valuation Premium62%Current premium over backend equipment peers; historical average is 41%
- FY27-28E OP Difference-15% ~ -31%Magnitude by which J.P. Morgan operating profit forecasts fall below market consensus
- 2028E TCB TAM$1 billionExpected market size when HBM capacity reaches 1 million wfpm
Impact & implications
The report argues that although the memory capex upcycle is real and longer-lasting than expected, this has largely been reflected in the stock price. For investors, this means limited room for further valuation expansion driven by industry beta in the short term, while facing dual risks of earnings misses and valuation regression. The market needs to see substantial progress in HBF applications or order breakthroughs from new clients (e.g., Samsung Electronics) to support the current premium level. At current prices, the risk-reward ratio points to an unfavorable area; we recommend waiting for clearer catalysts or a valuation pullback before considering entry.
Risks
- HBM ASP increases miss expectations, reducing motivation for storage vendor capacity expansion
- HBF market expansion slower than expected or JEDEC standardization delayed
- Failure to secure TCB orders from Samsung Electronics (SEC)
- TCB equipment procurement by Chinese storage vendors falls below expectations
- JEDEC relaxing HBM5 stacking height limits accelerates Hybrid Bonding substitution for TCB
What to watch
- Pace of recovery for SK hynix HBM4-level TCB orders
- Status of Micron (MU) order backlog and subsequent delivery confirmation
- Customer feedback on HBF TCB demo equipment planned for mid-2026
- Updates on JEDEC standards regarding HBM5 stacking heights
- Changes in operating margin spread between HBM and traditional DRAM