Allegro MicroSystems, Inc.

▲ Bull Case

• ◆Margin recovery is structural, not cyclical. Non-GAAP gross margin expanded from 41% trough to 50.6% exit Q4 FY2026; non-GAAP op margin improved from -3% to ~17% Q4 FY2026. Consensus FY2028 non-GAAP op margin of 24% supported by R&D leverage (23%→18%), SG&A leverage, and mix shift to higher-margin TMR and data-center products.
• ◆xEV + data-center represent durable secular growth pockets. EV content multiplier 2–4x ICE; ALGM holds #2 global magnetic-sensor position with 5–7 year TMR lead; data-center current sensing reached 14% of industrial revenue Q4 FY2026 (from <5% prior year)—structural new TAM.
• ◆Capital discipline maintained through cycle with optionality. R&D held at 23% of revenue through downcycle; net debt deleveraged $248M→$137M in 12 months; FY2026 FCF $125M (vs. $22M FY2025) underwrites Sanken absorption, opportunistic buyback, and bolt-on acquisitions.

▼ Bear Case

• ◆Valuation prices MPWR-class outcomes, not auto-analog peer outcomes. At 8.6x EV/Revenue NTM (~3x peers at 3.0–3.6x), justifying current price requires terminal non-GAAP op margin 28%+ and 5-yr revenue CAGR 17–20%—both above historical peaks. Any margin shortfall triggers compression to $25–30/share.
• ◆Extreme cycle risk plus Crocus ROIC overhang. FY2025 proved 31% revenue decline possible; current revenue near pre-cycle peak leaves thin cushion. Crocus earning ~5% pre-tax ROIC after three years (vs. ~10% WACC); needs FY2028–FY2029 ramp; impairment risk compounds downside.
• ◆Sanken ownership, governance, and China complex unresolved. Sanken still owns 32.5% (largest holder); Polar Semiconductor wafer-pricing transparency imperfect; 27% of revenue China-origin with tariff overhang; insider selling (former CEO $1.4M May 2026) without offsetting insider buying through recovery.

“Street consensus unanimously bullish (Strong Buy, 10+ Buys vs. 0 Sells, average PT $54.42 with Jefferies at $62), but the debate is not 'buy vs. sell'—it has been voted on. Core debate is 'can fundamentals catch up to valuation or has recovery been over-priced?' Specifically: (1) Can Q4 FY2026 50.6% non-GAAP gross margin hold through industrial softness or auto price-downs? (2) Is Q4 14% data-center share of industrial a step-function or cyclical bump? (3) Does TMR revenue ramp from ~$80M to $200M+ by FY2029 to validate $420M Crocus deal? (4) Is Sanken overhang open-ended or accelerating? Margin durability is highest-stakes debate—each 100 bps of terminal op margin shortfall equals ~$3.50/share.”

• ◆Q1 FY2027 earnings (August 2026): Full-year guide raise; data-center share confirmation ≥15% of industrial revenue
• ◆Q2 FY2027 earnings (October 2026): Data-center share ≥20% confirms step-function thesis
• ◆Investor day / capital allocation update (fall 2026): Sanken disposal timeline acceleration
• ◆Major TMR / XtremeSense customer win: Large EV OEM design-in (Mercedes, BYD, Stellantis class)
• ◆Sanken full-divestiture announcement: Removes structural overhang and supports multiple expansion

• ◆Auto cycle reversal with renewed destock: FY2025 proved 31% revenue decline possible; current revenue near pre-cycle peak
• ◆Margin recovery stall or regression: Each 100 bps of terminal op margin shortfall equals ~$3.50/share downside
• ◆Infineon TMR competitive catch-up within 3 years: Largest moat erosion vector; could compress 5–7 year lead
• ◆China tariff escalation (25%+): Direct impact on 27% of revenue; immediate margin compression
• ◆Crocus intangibles impairment writedown: $130M potential write-off signals failed M&A bet (-8 to -10% sentiment impact)
• ◆Sanken disorderly exit or governance event: Removes overhang but at cost; signals Polar JV relationship breakdown