Hyperscalers’ Shift to Natural Gas Power Faces Price Risks, Research Finds

Major technology companies including Amazon, Google, Meta, and Microsoft are increasingly investing in natural gas power to fuel their data centers, aimed at supporting growing AI capabilities. However, a report from energy research firm Noreva warns that these firms may soon face significant price fluctuations tied to the fossil fuel.

The demand for natural gas by hyperscalers could lead to price increases of up to threefold in certain U.S. regions, as current supply growth declines and liquefied natural gas exports rise. Noreva’s CEO, Peter Gardett, expressed concerns, stating, “I think everyone in the energy markets has been lulled into a sense that gas prices can’t go up. You just need simple arithmetic to get to a much tighter gas market than you were in just a few years ago.”

In March, Meta announced plans to construct a 7.5-gigawatt natural gas power plant in Louisiana, intended to service its Hyperion data center. Shortly thereafter, both Google and Microsoft revealed plans for gigawatt-scale gas plants in Texas, and Amazon is also pursuing a 7.6-gigawatt facility in the same state. Such moves mark a significant shift for companies that have historically avoided substantial capital investments.

Current natural gas prices vary from approximately $2 to $4.50 per million BTUs, with the Henry Hub in Louisiana priced at nearly $3. However, Noreva forecasts that prices could exceed $10 per million BTUs in some hubs. High fuel costs, representing about half the operational expense of large power plants, could lead to increased operational costs for AI data centers, raising token prices or necessitating connections to the electric grid, thereby driving electricity costs up.

While natural gas prices appear stable in the short term, Gardett notes that the long-standing balance of supply and demand may shift. “That alone wouldn’t change the economics here,” he said, pointing out that integrating the domestic gas market with global markets, combined with the rising demand for AI, plays a critical role in current pricing dynamics.

Hyperscalers have been attracted to cheaper gas prices in Texas, especially in West Texas, where most gas is a byproduct of oil extraction. The lack of infrastructure for transporting this gas has historically limited market access, but recent pipeline developments are changing that. “They’ve finally built some pipelines out there, and a lot of that is headed towards export markets,” Gardett added.

As regional markets evolve and become interconnected, fluctuations in natural gas prices could have widespread effects. Gardett cautioned that significant price disparities could arise, with some areas experiencing gas prices above $10 per million BTUs for extended periods.

The potential for rising natural gas costs adds a complex layer to the ongoing conversation about data center impacts on utility bills, as 80% of consumers express concerns about their financial effects. This discontent might expand to include natural gas expenses as hyperscalers deepen their investments in fossil fuels, a realm where they have minimal prior experience.

Gardett concluded, “On future Alphabet earning calls, you will hear them talk about the correlation between natural gas pricing and Google results, which is strange, but that’s where we are.”