Nvidia to pay dividends in less than 3 weeks; Here's how much 100m NVDA shares will earn

Nvidia will pay a $0.25-per-share quarterly dividend on October 1, 2026 to shareholders of record before the September 10 ex-dividend date, maintaining an annualized $1.00 payout and a 0.46% yield. Its 6.37% forward payout ratio, roughly $26B of latest-quarter dividends and buybacks, and nearly $193B in net income on about $303B of trailing revenue indicate substantial capacity for further capital returns. The company expects approximately 70% fiscal-2028 revenue growth as AI infrastructure demand exceeds supply, with memory availability rather than demand cited as the key production constraint.

Analysis

The dividend is immaterial to NVDA valuation and should not create a durable post-payment catalyst: the cash yield is too small to attract incremental income demand, while the ex-date effect is mechanical. The relevant signal is capital-allocation flexibility: sustained repurchases can cushion drawdowns, but at a premium multiple they also reduce the accretion benefit versus reinvesting in supply assurance, software, and customer financing. Any stock reaction around the payment date is more likely liquidity/noise than a change in fundamentals.

The binding constraint shifts bargaining power upstream to AI-memory suppliers, especially MU, rather than to NVDA's cloud customers. If high-bandwidth-memory availability remains constrained through the next two quarters, NVDA can preserve pricing and shipment visibility but may defer revenue rather than lose it; hyperscalers with internally funded capex (MSFT, GOOG) are better positioned than more financing-dependent CRWV and ORCL to absorb timing slippage. The second-order risk is that constrained system deliveries push customers toward alternative architectures and custom silicon, increasing the strategic value of ASIC programs even if they do not displace NVDA near term.

Consensus appears to treat supply tightness as unambiguously bullish for NVDA. It is bullish for margins only while allocation favors its platforms and customers accept delivery delays; prolonged shortages can convert booked demand into capex rescheduling, especially if cloud utilization or AI monetization disappoints. Over the next 1-3 months, watch HBM lead times, NVDA system lead times, and hyperscaler capex commentary; over 6-18 months, the key debate becomes whether supply normalization reveals demand elasticity and compresses the infrastructure stack's valuation multiples.

Key Decisions for Investors

• No dividend-driven NVDA trade: avoid buying for the October payment or expecting a meaningful ex-date dislocation; use any dividend-related weakness only as an execution window for an existing AI thesis.
• Initiate a 1-3 month relative-value position long MU / short NVDA in equal dollar beta-adjusted size if HBM pricing and supply commentary remain tight. MU has more direct earnings torque to the bottleneck, while NVDA already embeds scarcity rents; target 10-15% relative upside, with a stop if MU guidance fails to show HBM mix/margin acceleration or NVDA raises supply availability materially.
• Maintain NVDA core exposure only with defined downside protection into the next hyperscaler capex updates: buy 3-6 month NVDA put spreads financed partially by selling upside calls at a level consistent with portfolio trimming. The thesis is not demand collapse, but a multiple reset if delivery timing slips or management lowers near-term revenue conversion despite strong backlog.
• Favor MSFT and GOOG over CRWV for AI-capex exposure through the next two quarters. Large platforms can monetize scarce compute across broad customer bases and fund deployment internally; CRWV is more exposed to financing costs and utilization risk if delayed hardware delivery slows revenue recognition. Reassess if CRWV demonstrates utilization and contracted-revenue conversion that materially outpaces its capital intensity.