Maxscend released 2Q25 results. Revenue was RMB948mn in 2Q25, up 25% QoQ from a low base in 1Q but down 13% YoY, reflecting continued softness in end-market demand and intensified competition. During the period, GPM fell to 27% (-4ppt/-14ppt vs. 1Q25/2Q24) in 2Q25, missing BBG consensus estimates of 34%, due to 1) ongoing price pressure, 2) impact from depreciation, and 3) lower-than-expected fab utilization. Net loss widened to RMB101mn in 2Q (NLM: -11%) from -RMB47mn in 1Q (NLM: -6%). We cut our 2025/26E revenue forecasts by 21%/22% to reflect weaker-than-expected demand and persistent pricing pressure. We also lower GPM projections by 9ppts/7ppts to 32%/37% in 2025/26E, factoring in near-term margin headwinds from elevated depreciation. Maintain HOLD with TP of RMB81.5.

Revenue declined on soft demand, while product development progress stayed intact. China’s smartphone shipment grew 3.3% YoY in 1Q25 on government subsidies but decreased 4%/4% YoY/QoQ in 2Q as demand fell. The weak recovery in market demand (2025 smartphone shipment to grow by 0.6%/1.9% YoY, per IDC/Counterpoint) and slower- than-expected industry consolidation weighed on Maxscend’s performance. 1H25 sales dropped 25% YoY, although RF modules now accounted for 44% of total revenue, up from 36%/42% in FY23/24. This shift demonstrates the company’s progress in product development (i.e., L-PAMiD, equipped with in-house MAX-SAW, reflecting a fully localized supply chain). We expect module revenue to exceed 50% of total sales in 2027E, aligning with industry modularization and localization trends. However, without a strong rebound in demand, we expect Maxscend’s overall revenue to decrease by 9% in 2025E.

Margin headwinds persist. Maxscend booked a D&A expense of RMB323mn in 1H25, up 22% YoY from RMB265mn in 1H24. Although depreciation growth may have peaked, absolute expense is expected to remain elevated in the near term as the company ramps up its 12-inch IPD line (1H25 capacity: 5kwm). We think improved fab utilization across the 6- inch and 12-inch lines should help ease margin pressure over time. We expect overall GPM to be 32% in 2025, given ongoing margin erosion as fab ramps up.

Maintain HOLD. We see 2025-26E as transitioning years for Maxscend. Our new TP is based on 45x 2026E EV/EBITDA (vs. prior 45x 2026E P/E), slightly higher than 0.5-SD below 5-year historical avg. of 44x, reflecting ongoing business model shift toward fab-lite/IDM. While Maxscend is well- positioned in the RFFE market and China’s localization, we think its margin recovery may take another year. Upside catalysts: stronger-than-expected recovery, accelerated localization. Potential downside risks: slower capacity ramp-up, R&D delays.