Nvidia's Profit Margins Are Safe for Now, Luria Says

Key Concepts

• Semiconductor Cycle: The cyclical nature of the chip industry, characterized by periods of high demand and capital expenditure followed by potential corrections.
• Hyperscalers: Large-scale cloud computing providers (e.g., Google, Microsoft, Amazon, Meta) that drive the majority of demand for AI-focused semiconductors.
• Capital Expenditure (Capex) Intensity: The massive financial investment required by hyperscalers to build AI infrastructure, which impacts market liquidity and stock volatility.
• Valuation Multiples: The ratio of a company's stock price to its earnings (P/E ratio), used here to determine relative safety in a volatile market.
• Pricing Power: The ability of semiconductor manufacturers (like Nvidia) to maintain high margins due to a lack of viable alternatives for their customers.

---

1. Market Sentiment and the "Semiconductor Crack"

Gil Luria of DA Davidson characterizes the current market reaction to semiconductor earnings as a "small crack" rather than a "red flag." While companies like Broadcom reported strong growth—50% revenue growth and 143% AI revenue growth—investors have become hyper-sensitive. Luria notes that spectacular results from companies like Intel, AMD, and Dell have inflated investor expectations, leading to severe consequences for stocks that do not "beat" earnings by a significant margin.

2. Valuation Discrepancies and Investment Strategy

Luria argues that the market is currently pricing semiconductor stocks based on two conflicting timelines regarding the AI cycle:

• Long-term Cycle (Peak 2030): Intel, AMD, and various optical/nuclear stocks are trading as if the AI boom will continue unabated until 2030.
• Short-term Cycle (Peak 2025): Nvidia and Micron are trading as if the cycle will peak as early as next year.

Actionable Insight: Luria suggests that investors should favor companies with less demanding valuations to mitigate risk. He highlights the following P/E ratios as indicators of relative safety:

• Micron: 10x
• Nvidia: 22x
• Broadcom: 35x
• AMD: 40x
• Intel: 100x

3. The Role of Hyperscalers and Capital Demands

The sustainability of the semiconductor boom is tied to the spending habits of hyperscalers. Luria notes that these companies are reinvesting the majority of their operating cash flow into AI infrastructure.

• Capital Squeeze: With Google raising $80 billion in capital—on top of massive demands from SpaceX, Anthropic, and OpenAI—the sheer volume of capital being pulled from the market is creating volatility.
• Return on Investment (ROI): Luria defends the hyperscalers' spending, noting that they are growing revenue at an accelerating rate while maintaining margins. He points to the $100 billion+ run rate of AI products as evidence of "real economic value" being generated.

4. Pricing Power and Margin Stability

A critical question is whether hyperscalers will eventually push back against rising chip prices. Luria contends that hyperscalers currently lack leverage:

• Dependency: Hyperscalers remain almost entirely reliant on Nvidia chips. While they are attempting to develop internal chips (often in partnership with Broadcom) or ramp up AMD production, these alternatives are in the "early stages."
• Margin Outlook: Because there are few alternatives, semiconductor gross margins remain in the mid-70% range. Luria believes these margins are "relatively safe" as long as AI adoption continues to diffuse into the broader economy and consumers continue to engage with AI-driven products (e.g., subscriptions and ad-supported chat).

5. Synthesis and Conclusion

The semiconductor sector is currently navigating a high-volatility environment driven by extreme investor expectations and massive capital requirements. While the "cycle" is inevitable, its duration remains uncertain—potentially lasting until 2030.

Main Takeaways:

• Volatility is expected: The combination of massive capital raises by hyperscalers and IPO activity will continue to impact stock performance.
• Valuation matters: Investors should prioritize companies with lower P/E ratios (like Micron and Nvidia) over those priced for perfection (like Intel).
• Dependency sustains margins: Until hyperscalers successfully diversify their hardware supply chains, semiconductor manufacturers will retain significant pricing power, keeping margins high.
• Economic validation: The long-term health of the sector depends on the continued integration of AI into business productivity and consumer behavior, which justifies the current high levels of capital expenditure.