NVDA Cheaper Than Semis: How to Play the Gap

Nvidia's forward price-to-earnings ratio sits 41% below the semiconductor sector median, creating a rare valuation spread for the AI infrastructure cycle's dominant hardware supplier.

Lead

Nvidia (NVDA), the company that effectively defined the AI hardware market, now trades at a meaningful discount to the very semiconductor sector it propelled. With a trailing price-to-earnings ratio near 30x against a semiconductor industry average above 68x, and a forward multiple of roughly 22x compared to a sector median of 38.78x—a gap of more than 41%—the world's most valuable chipmaker has become, on conventional valuation metrics, among the cheapest major names in the Nvidia semis cohort. The inversion is as stark as it is counterintuitive, and it is forcing institutional investors to reassess positioning heading into the second half of 2026.

What Happened

The spread has opened steadily across 2026 as the semiconductor rally broadened sharply beyond Nvidia. The VanEck Semiconductor ETF (SMH), which counts NVDA as its largest holding at roughly 14% of assets, has surged approximately 73% year-to-date through late June 2026. NVDA itself has gained approximately 12% over the same period—a striking divergence for the name anchoring that benchmark.

The drivers of that gap tell the story of where capital has rotated. AMD has surged more than 114% in 2026, carrying a forward price-to-earnings multiple of roughly 58x. Intel has rallied approximately 130%. Broadcom (AVGO) trades at more than 70x trailing earnings. Capital that once concentrated almost entirely in NVDA as the sole liquid AI hardware proxy has dispersed across the full semis supply chain—memory, equipment, advanced packaging, and competing accelerator platforms—leaving Nvidia's multiple stranded while peers' expanded.

NVDA's P/E has compressed sharply from its 2023 peak above 131x, and the 12-month average ratio of roughly 46x underscores just how aggressive the de-rating has been. Meanwhile, AMD and Broadcom—with materially lower revenue and thinner margins—continue to command premiums that Nvidia no longer carries.

The Financial Backdrop

The paradox sharpens against Nvidia's operating results. For fiscal first quarter 2027, ended April 26, 2026, the company posted record total revenue of $81.6 billion, up 85% from a year earlier. Data center revenue reached $75.2 billion, a 92% year-over-year gain driven by hyperscaler demand for Blackwell-architecture accelerators, with gross margins holding near 75%.

For Q2 fiscal 2027, Nvidia guided revenue to $91.0 billion, plus or minus 2%—another double-digit sequential acceleration. Full fiscal year 2026 revenue reached $215.9 billion, up 65%. Nvidia projects combined Blackwell and next-generation Rubin architecture revenue at roughly $1 trillion across fiscal years 2025 through 2027, a ramp that would represent the fastest large-scale revenue build in semiconductor history.

At roughly 22x forward earnings against 85% revenue growth, Nvidia's price-to-earnings-growth ratio sits well below one, a threshold historically associated with undervaluation in high-growth technology hardware. The consensus analyst price target sits near $299, implying approximately 48% upside from current levels around $211.

Risk Factors Behind the Discount

The market's restraint on NVDA's multiple reflects a set of real concerns that justify some premium compression, even if not the full magnitude of the current discount.

GPU compute rental rates have softened materially. Spot pricing for the flagship B200 processor fell from $6.11 per hour on May 30 to $4.22 per hour by late June—a contraction exceeding 30%—raising questions about near-term utilization absorption and whether hyperscaler capacity builds will generate sustained pricing power or temporary oversupply. Export controls remain a structural overhang. U.S. restrictions on advanced chip exports to China have curtailed one of Nvidia's largest potential markets, and ongoing regulatory and copyright litigation adds uncertainty to the forward investment case. Manufacturing yield risks tied to advanced packaging for Blackwell accelerators continue to attract investor scrutiny, particularly given the revenue concentration in data center. Any delivery schedule slip carries outsized quarterly consequence.

Finally, the sector's broadening rally itself creates a mechanical headwind: as capital flows into smaller-cap semis names, NVDA's index weight dilutes rather than amplifies relative performance.

Positioning Around the Spread

Three approaches address the valuation gap between NVDA and the broader Nvidia semis sector.

Direct NVDA exposure captures the multiple convergence most directly. If the forward P/E re-rates toward the sector median of 38.78x—even partially—while earnings power continues compounding at current growth rates, the total return potential is substantial. Near current prices, the risk-reward on a 12-month basis appears asymmetric relative to where the stock has historically traded. Broad semiconductor ETF exposure via SMH or SOXX provides diversified participation in the sector rally while retaining Nvidia as its largest weight. This approach sacrifices the specific valuation arbitrage but limits single-stock execution risk and has meaningfully outperformed NVDA alone in 2026. Relative value structures pairing NVDA long against higher-multiple peers—AMD at 58x forward earnings, Broadcom above 70x trailing—allow investors to play multiple convergence with reduced directional market exposure. This approach carries basis risk if the divergence persists or widens before normalizing.

Outlook

Nvidia enters the second half of 2026 as the AI infrastructure cycle's primary revenue engine—guiding to $91 billion in quarterly revenue at 75% gross margins—yet priced at a discount to a semiconductor sector it has materially outgrown on nearly every fundamental measure. The B200 rental rate compression and export control environment represent genuine risks that warrant some valuation haircut. But a 41% discount to the sector forward multiple, sustained against 85% revenue growth, is a spread that historically closes. The most likely catalysts are a Q2 fiscal 2027 beat-and-raise, stabilization in spot GPU compute pricing, and incremental clarity on the fiscal 2028 Rubin platform ramp.