Baird downgrades Rambus stock rating on DRAM shortage concerns By Investing.com

Baird downgraded Rambus to Neutral from Outperform and set a $120 price target, implying downside from the current $141.31 share price. The firm warned that DRAM shortages and rising DRAM prices could slow registered DIMM growth through 2027, partially offsetting demand tailwinds from x86 and agentic AI. Rambus also recently missed first-quarter 2026 expectations, reporting EPS of $0.63 versus $0.64 consensus and revenue of $180 million versus $189.71 million expected.

Analysis

The key issue is not the near-term miss; it is that Rambus sits in the wrong part of the memory cycle. When DRAM is scarce, the value pool migrates upstream to memory makers and the fastest AI-linked allocation lanes, while unit-driven infrastructure suppliers get squeezed by mix, lead times, and customer deferrals. That creates a classic “good technology, bad cycle” setup where gross margin can stay elevated but revenue leverage disappears.

The second-order risk is that the AI buildout is helping the industry in aggregate while hurting RMBS’s specific product cadence. If OEMs and module partners are forced to prioritize HBM and premium server configurations, the addressable volume for more commoditized DIMM-related demand can weaken for several quarters, even if end-market compute demand is healthy. In other words, AI demand is not a clean offset; it can actually intensify the mismatch between industry shortages and RMBS shipment growth.

This looks like a months-long earnings-revision air pocket rather than a one-day headline trade. The stock’s valuation leaves little room for lower unit growth, so any sign of slower registered module orders or delayed x86 refresh activity should compress the multiple further. The main bullish counterpoint is that if memory shortages ease faster than expected or AI server demand broadens beyond hyperscalers into enterprise refreshes, the narrative could re-rate quickly—but that is a 2H26 story, not an immediate catalyst.

Key Decisions for Investors

• Short RMBS into strength over the next 1-3 months; use rallies toward prior support as better entry points. Risk/reward favors downside if guidance commentary confirms slower unit growth, with 10-15% downside plausible before valuation support emerges.
• Pair trade: long HBM-exposed memory beneficiaries / short RMBS for the next 2 quarters. The trade expresses the view that capital and margin pool stay concentrated in the tightest part of the supply chain rather than in memory-adjacent IP names.
• Consider buying RMBS put spreads 3-6 months out to capture multiple compression while limiting premium bleed. Best structure is a defined-risk bearish view on lower shipment expectations rather than a binary earnings short.
• For longs in the broader semicap / AI hardware basket, avoid adding RMBS as a levered AI proxy until evidence shows volume acceleration, not just margin resilience. The setup is more vulnerable to estimate cuts than many investors will assume.
• Set a tactical cover trigger if DRAM lead times normalize or management reframes the demand mix toward enterprise x86 refresh. That would be the earliest sign the current negative revision cycle is ending.