Will Amkor's gross margin exceed 15% in either Q3 or Q4 2026?

Three reinforcing factors push toward YES: (1) Q2 guide midpoint 15.0% already at the threshold before H2 drivers kick in; (2) management explicit framing 'mid- to high teens' for H2 covers and exceeds the threshold; (3) Arizona depreciation drag deferred to 2027 means H2 2026 is the cleanest pre-drag operating leverage test. The 'either Q3 OR Q4' structure is forgiving — only one quarter needs to clear. Per-quarter probability ~0.60-0.65 with moderate correlation gives a combined ~0.74. Risks: H2 capex weight (~70% of $2.5-3B) compresses margin via accelerated depreciation, and HDFO ramp execution is concentrated in this period.

Tail risk consideration: H2 capex of ~$1.75-$2.1B is a step-up from H1 and adds significant equipment to the depreciation base. If management's 'mid- to high teens' is interpreted as 14-19%, the lower end (14-15%) doesn't clear the threshold and the higher end is needed for a YES. The Q4 2025 GM print was 16.7% (a high-water mark), so 15%+ is precedented for AMKR but only just. Korea new test building completing end-2026 adds Q4 depreciation. The Q1 ensemble miss on the H1 14% market (predicted 0.22, resolved YES) suggests the prior has been too conservative on margin recovery — calibration argues for higher P(YES) here.

Decomposing by quarter: Q3 likely benefits from peak HDFO data center CPU revenue contribution (high-margin advanced packaging), normal seasonal mix improvement, and pricing actions catching up to materials cost. Q4 historically the highest GM quarter (Q4 2025 was 16.7%) due to mix and absorption. The threshold of 15% is well within the demonstrated capability of the business pre-Arizona drag. Both quarters individually have ~0.60-0.65 probability, and they're moderately correlated (~0.6) — the OR-combined probability is in the 0.72-0.78 range. I lean toward 0.76 reflecting the management explicit guidance plus the pattern of H1 beats.

Anchoring on management's explicit 'mid- to high teens' H2 GM guidance and the Q2 midpoint of 15.0% as the launching baseline. HDFO ramp adds high-margin mix in Q3+. The OR structure of 'either quarter > 15%' is materially more forgiving than 'both' or any specific quarter. Calibration lesson from the H1 14% market (predicted 0.22, miss to YES at 14.2%) argues against being too conservative on margin recovery. ~72% probability.

More cautious view: H2 capex weight of 70% of $2.5-3B significantly increases depreciation base into Q4. Korea new test building also adds depreciation late in the year. While the OR structure helps, 'mid- to high teens' could easily mean 14-15% in an unfavorable scenario. The 15% threshold is exactly at management's Q2 midpoint guide — meaning Q3 and Q4 need to maintain or improve over that baseline against rising depreciation. I weight this at ~0.68 — clearly above coin-flip but not aggressively given H2 capex weight.

Three-factor calculus: (1) revenue growth via HDFO ramp adds high-margin mix; (2) pricing actions to offset materials cost should be incremental positive by H2; (3) Arizona drag deferred to 2027 removes the largest depreciation overhang. Against this, H2 capex equipment depreciates as it's placed in service. Net positive — probability of clearing 15% in at least one of two quarters is moderately high. ~73%.

Mgmt says 'mid- to high teens' for H2. Q2 already guided to midpoint 15%. OR structure means only one quarter needs to clear. ~72%.

H2 carries 70% of capex — significant depreciation. But pricing and HDFO mix push the other way. Net moderately above coin-flip. ~69%.

Q4 2025 GM was 16.7% — already proven AMKR can run above 15% in absence of Arizona drag. Arizona drag deferred. ~74% probability one of Q3 or Q4 clears 15%.