JP Morgan Just Revealed a Massive Data Center Gap—These 3 Stocks Will Fill It

Key Concepts

• Data Center Buildout: The massive physical infrastructure expansion required to support AI computing, including power, cooling, and electrical distribution.
• Hyperscalers: Large-scale cloud providers (e.g., Microsoft, Amazon, Meta) driving the demand for AI infrastructure through massive capital expenditure (capex).
• "Picks and Shovels" Strategy: Investing in the companies that provide the essential tools, hardware, and infrastructure for an industry (AI) rather than the end-product companies themselves.
• Liquid Cooling: A specialized thermal management technology required to cool high-density AI server racks.
• Backlog: The volume of contracted work that has been committed to but not yet completed or delivered.

---

1. The State of the AI Infrastructure Market

The video argues that the AI investment story is in its early stages, supported by a JP Morgan report indicating that 60% of planned data center capacity for 2027 has not yet broken ground. Furthermore, 7% of current projects face delays due to supply chain bottlenecks, regulatory permitting hurdles, and energy availability issues. While hyperscalers are "doubling down" on capex, the physical construction sector is struggling to keep pace with demand, creating a significant backlog that ensures long-term growth for infrastructure providers.

2. Featured Companies and Market Roles

Eaton (ETN)

• Role: Provides the "electrical guts" of data centers, connecting the power grid to server racks.
• Performance: Revenue reached $7.45 billion (up from $6.38 billion YoY); adjusted EPS was $2.81.
• Key Metric: Data center orders in the "Electrical Americas" segment grew 240% year-over-year.
• Strategic Moves: Acquired Boyd Thermal to enter the liquid cooling market and is collaborating with Nvidia on the "GB200 NVL72" (referred to as the Beam Ruben DSX platform) for AI factories.

Quanta Services (PWR)

• Role: Specializes in the physical infrastructure required to deliver power to data center sites.
• Performance: Reported a $44 billion backlog (up 27.5% YoY). Projects 15–20% annual EPS growth through 2030.
• Market Opportunity: The company identified a $2.4 trillion addressable market for their services through 2030.
• Investor Sentiment: Strong institutional buying (buying 50% more than selling) suggests high confidence in the company’s ability to execute on its massive backlog.

Vertiv (VRT)

• Role: A leader in thermal management and liquid cooling solutions, essential for high-heat AI GPUs.
• Performance: Up over 169% in the last 12 months. Raised full-year sales guidance to $13.5–$14 billion.
• Strategic Moves: Aggressive M&A activity, including the acquisitions of Strategic Thermal Labs and Thermo Key, allowing them to scale from chip-level cooling to facility-wide heat rejection.
• Outlook: While some analysts have turned cautious (assigning "hold" or "sell" ratings due to high valuation), others maintain bullish targets, citing the company's critical role in solving the environmental and thermal challenges of AI data centers.

---

3. Key Arguments and Perspectives

• Volatility vs. Growth: The host and guest argue that volatility in these stocks is a result of investors rotating capital, not a lack of demand. They contend that while hyperscalers (Microsoft, Meta) are spending heavily, the "picks and shovels" companies offer a "cleaner" exposure to the actual dollars being spent on construction.
• Analyst Conservatism: The speakers suggest that analysts have consistently underestimated the growth of AI-linked companies, often failing to keep pace with the rapid fundamental improvements reported by firms like Micron (MU) and the featured infrastructure providers.
• Risk Management: Chris Marott advises that for stocks with massive run-ups (like Vertiv), it is prudent for retail investors to "trim" positions to take risk off the table while maintaining exposure to the long-term growth trend.

4. Synthesis and Conclusion

The AI story is transitioning from a speculative phase to a massive, multi-year physical infrastructure buildout. The primary bottleneck is no longer just chip availability, but the ability to power and cool the massive data centers required for AI. Investors are encouraged to look past the volatility of the hyperscalers and focus on the companies responsible for the electrical and thermal backbone of the industry. The consensus is that the "backlog" of projects provides a clear, multi-year runway for growth, making these infrastructure providers essential components of an AI-focused portfolio.