Morgan Stanley doubles down on its cautious view on North America hardware "memory payers"

Overview

This report discusses the "memory payer" trade in the North America IT hardware sector during the memory price-upcycle. Morgan Stanley believes that the recent relative outperformance of hardware OEMs, consumer electronics makers, and channel partners does not change the medium-term risk: memory contract prices are still rising rapidly, CSP demand is more price-inelastic, while traditional hardware demand is more elastic, which may ultimately squeeze hardware makers' gross margins and trigger demand destruction in 2H26.

Core views

The core view is to stay the course and remain cautious. The report argues that the market is overemphasizing softer DRAM spot prices, memory-optimization technologies such as TurboQuant, and the pullback in memory stocks, while overlooking the fact that most memory transactions are determined by long-term agreements/contract prices and that supply-allocation risk remains. Hardware stocks are currently in the second stage of a three-stage trade: the short-term relief rally may continue, but the third stage is expected to unfold in 2H26, characterized by further price increases, gross-margin compression, demand destruction, earnings cuts, and valuation compression.

Analysis framework

The report combines relative stock performance, memory contract prices versus spot prices, supply-chain checks, AlphaWise work, CIO surveys, VAR surveys, OEM shipment data, and valuation/earnings revision data to assess how memory-cost inflation flows through to hardware makers' margins and demand. The focus is not on a single-company valuation, but on the theme trade of 'memory suppliers benefiting, hardware payers under pressure.'

Methodology notes

• Theme-trade frameworkThree-stage memory payer trade

  The report divides the trade into three stages: first, memory price increases cause hardware stocks to underperform; second, a rebound appears on demand pull-forward and better-than-expected commentary; third, further price increases lead to gross-margin compression, demand destruction, and earnings cuts.

  The current phase is judged to be the middle of stage 2, and the report believes the real negative catalysts are more likely to concentrate in 2H26.

• Price-system analysisSpot price versus contract price split

  The DRAM spot market only represents a smaller, lower-end portion of demand, while hardware makers' cost pressure is driven more by contract prices.

  The report emphasizes that 2Q DRAM contract prices are expected to be about +58% Q/Q and NAND contract prices about +70% Q/Q, so weaker spot prices do not prove that memory tightness has eased.

• Demand elasticity analysisPrice transmission and demand destruction

  CSPs are less sensitive to memory prices, while demand in traditional hardware markets such as PCs, servers, storage, and smartphones is more elastic.

  When device prices rise 30%, 50%, 70%, or more, historical elasticity assumptions may break down, and hardware revenue and profit may come under greater pressure than the market expects.

• Survey and supply-chain validationCIO/VAR/ODM checks

  Use CIO budgets, VAR customer feedback, and ODM shipment changes to verify hardware demand pressure.

  The CIO survey shows 2026 hardware budget growth expectations of only 1.0% Y/Y, and VAR feedback suggests that 30%-60% of customers may cut spending on PCs, servers, and storage because component-cost inflation is pushing up end prices.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Hardware OEMs / memory payer cohort

  Mainly pressured asset

  Strengths

  In the near term, it may still benefit from demand pull-forward, better-than-expected management commentary, improving earnings-revision breadth, and short covering.

  Weaknesses

  It needs to absorb price increases in key components such as DRAM, NAND, and CPUs, while traditional hardware demand is relatively elastic.

  Comparison

  Compared with memory suppliers, it occupies the cost-payer side; stock performance has shown a strong negative correlation with memory stocks.

  Risks

  2H26 may see gross-margin compression, demand destruction, earnings cuts, and valuation compression.

• Memory suppliers

  Relative beneficiary asset

  Strengths

  Supported by CSP demand, GPU rental prices, and token growth, with better support for both prices and margins.

  Weaknesses

  If memory demand rolls over or contract prices peak earlier than expected, the cycle's durability weakens.

  Comparison

  The report notes that hardware stocks and memory stocks have recently been correlated at -0.88, making memory-stock moves an important factor for hardware payers.

  Risks

  If DRAM/NAND spot or contract prices fall earlier than expected, it would weaken the cautious case on hardware.

• HPQ, LOGI, GPRO, CRCT

  High-conviction Underweight exposure

  Strengths

  If near-term demand remains resilient or costs can be passed through smoothly, the share prices may continue to find support.

  Weaknesses

  High exposure to consumer-end, low price points, and price-sensitive demand.

  Comparison

  Compared with more commercial or larger-scale hardware buyers, these names face a greater demand-destruction risk.

  Risks

  Without negative data before 2H26, the underweight view may be premature.

• NTAP

  High-conviction Underweight exposure

  Strengths

  Has almost no consumer-end exposure.

  Weaknesses

  As a relatively smaller memory buyer versus peers, it has weaker purchasing leverage.

  Comparison

  Unlike consumer-exposed underweights, NTAP's risk comes more from procurement scale in the supply chain.

  Risks

  If enterprise storage demand proves more resilient or cost pressure can be passed through, the risk may be lower than the report suggests.

• DELL

  Underweight but with higher short-term debate

  Strengths

  AI-server expectations are stronger, and the company has more tools to protect itself; investors may grant it more benefit of the doubt before 2H26.

  Weaknesses

  It still faces rising memory costs and the risk of elasticity in hardware demand.

  Comparison

  Compared with other underweight names, DELL's recent stock performance is stronger and harder to validate in the near term.

  Risks

  If AI server orders and cost pass-through remain stronger than expected, the underweight view may stay under pressure.

• Apple, Inc.

  Relative exception / larger-scale buyer

  Strengths

  Large shipment scale; the report says it may be one of the few exceptions in supply-allocation risk.

  Weaknesses

  Still potentially affected by component prices and end-demand elasticity.

  Comparison

  Compared with PCs and smaller hardware makers, it has stronger supply-chain purchasing power.

  Risks

  If high-end smartphone demand weakens or component costs cannot be passed through, the advantage may shrink.

Key data

• memory payer recent performanceOutperformed by 13 percentage points since February 1After trailing by 11 percentage points in C4Q; the report sees this as a stage-2 relief rally.
• hardware stocks and memory stocks correlation-0.88Since September 1, 2025, hardware stocks have shown a strong negative correlation with memory stocks.
• 2Q26 DRAM contract prices+58% Q/Q to start, with some deals at +50%-70% Q/QThe report believes contract prices better reflect the actual cost pressure on hardware makers than spot prices do.
• 2Q26 NAND contract prices+70% Q/Q to startTrendForce estimates indicate NAND prices continue to rise sharply.
• spot DRAM versus contract priceStill about 25% higherEven if spot prices soften temporarily, spot buyers still have to pay a premium above contract prices.
• notebook ODM shipments1Q26 down 13% Y/Y; 2Q26 expected down 15% Y/YViewed as evidence that the more price-elastic hardware market is already weakening.
• 2Q gross margin expectation gapConsensus expects an average 30bps Y/Y expansion; MSe is for a 40bps Y/Y decline2Q gross-margin guidance is the key catalyst for validating the cautious view.
• 2H26 revenue/EPS expectation gapMSe is 5%-10% below consensusThe report thinks market expectations for 2H revenue and margins are too optimistic.
• 2026 hardware budget growthCIO survey expects +1.0% Y/YDown 60bps from the 1.6% in the 3Q25 survey, the weakest reading outside the COVID period.
• VAR customer spending feedback30%-60% of customers may cut spending on PCs, servers, and storageBecause component-cost inflation is pushing up end prices.
• coverage basket valuationMedian consensus P/E is about 12.2xThe report says hardware stocks have fallen about 2 turns since November 2025, but EPS remains near multi-year highs.

Impact & implications

In terms of investment implications, the report leans toward continuing to short or underweight hardware payers that are sensitive to memory costs, rather than chasing the recent rebound. In the near term, absent negative earnings revisions, hardware stocks may continue to outperform; but as 2Q gross-margin guidance, cloud capex disclosures, memory contract-price revisions, and 2H26 demand data arrive, the report expects the theme to shift toward downward revisions and valuation compression.

Risks

• The cautious call may be too early; in the absence of clear negative data points, hardware stocks could keep outperforming, and the 1Q earnings season and 2Q guidance may also end up better than the market fears.
• Short-term mission-critical IT demand may be less elastic, extending the hardware spending cycle and leading to revenue and earnings revisions that are stronger than the report expects.
• If memory demand comes under pressure and DRAM/NAND spot or contract prices roll over earlier than expected, component-cost inflation and demand-destruction risk would ease, and hardware demand and margins could experience a cyclical recovery.

What to watch

• 2Q26 hardware OEM gross-margin guidance, especially whether high memory costs have already created Y/Y gross-margin pressure.
• The direction of DRAM and NAND contract-price revisions, and whether spot-price changes truly flow through to most contract deals.
• Memory-stock performance versus hardware-stock relative performance, given their recent strong negative correlation.
• Whether the 1Q earnings season, 2Q guidance, and cloud capex disclosures change the view on demand pull-forward and supply tightness.
• Whether 2H26 revenue, margins, and EPS consensus estimates begin to be revised down.
• Whether VAR, CIO survey, and ODM shipment data continue to show weakening demand for PCs, servers, and storage.
• Whether CPU price inflation and further hardware price increases of more than 25% actually materialize.