Meta Signs 144-MW Texas Solar PPA With Apex Clean Energy

Artificial intelligence has turned electricity into the constraint that now dictates where and how fast the digital economy can grow. As of today, 16 September 2026, data centre operators are absorbing available grid capacity faster than utilities can build it, and the scramble for firm renewable supply has moved from a sustainability footnote to a core operational bottleneck.

Against that backdrop, Meta has signed a 144-MW corporate power purchase agreement with developer Apex Clean Energy to underwrite a new utility-scale solar project in Texas, according to reporting from Renewables Now. The deal is structured to match a slice of Meta’s data-centre electricity consumption in one of the toughest wholesale markets in the country.

What follows here is a working framework for how these tech-backed contracts decide which solar projects actually get built in deregulated markets, and the grid risks hiding beneath the headline megawatts.

How corporate offtakers act as the financial backstop for Texas solar

Strip away the press release, and the reason Meta’s signature matters comes down to money that will not move without it.

Texas runs on ERCOT, the Electric Reliability Council of Texas, a deregulated wholesale market with no guaranteed utility off-take and no regulated rate base to lean on. A solar developer building into ERCOT is exposed to merchant price risk: the project sells power at whatever the market pays, which can swing violently and even turn negative during congestion. Lenders will not finance construction against that kind of uncertainty.

Wholesale price volatility, including extended periods of negative pricing during high-renewable penetration, is not unique to ERCOT; European markets have hit sub-zero prices with enough frequency that project finance models built on merchant revenue assumptions are being stress-tested in ways developers did not anticipate five years ago.

A corporate PPA or virtual PPA (VPPA) solves the problem by locking in a predetermined price, typically across a 10-to-25-year term. That revenue certainty is what allows developers to underwrite construction debt and secure tax-equity investment. Historically, roughly 67% of corporate solar PPAs have been sited in ERCOT and the neighbouring SPP market, precisely because a creditworthy buyer is the only thing that makes those merchant-exposed projects bankable.

That is the real function of a Meta signature. The company is not simply buying electrons; it is providing the balance-sheet credibility that lets utility-scale solar exist in a market that otherwise punishes it.

The pattern is visible in Meta and Apex’s prior work together, including the 195-MW Angelo Solar project in Tom Green County. Since 2014, Apex has delivered 3,408 MW across 16 utility-scale wind, solar, and storage projects in Texas.

For anyone tracking infrastructure assets, the takeaway is blunt: developer pipelines live or die on landing Fortune 50 buyers before a single panel is bolted down.

The hyperscaler squeeze on the US power purchase market

That single Texas deal sits inside a demand wave that is reshaping who gets to buy clean power at all.

Corporate PPA volumes have surged on the back of data centres, and the concentration is stark. In 2025, Big Tech, meaning Meta, Amazon, Google, and Microsoft, accounted for 49% of global corporate PPA capacity. Meta and Amazon alone contracted 20.4 GW of corporate clean-energy deals that year, including nuclear.

The US picture is even more lopsided toward the technology sector. Data centres drove roughly 60% of US corporate PPA deals in 2024, a year that set a record with 28 GW contracted across 184 agreements. Momentum held into 2025, with US corporate capacity climbing even as global volumes dipped.

• 2024: A record 28 GW across 184 US corporate PPAs, with data centres accounting for about 60% of deal volume.
• 2025: US corporate capacity rose to roughly 29.5 GW, and more than 80% of domestic offtake volume came from the IT sector.

What this concentration tells you as an investor is that buyer power has collapsed into a handful of names. A “two-speed PPA market” has emerged: hyperscalers transact at a scale and creditworthiness that smaller corporate buyers simply cannot match, leaving mid-sized firms struggling to secure renewable capacity for their own sustainability targets.

The pattern of hyperscaler energy procurement reshaping entire regional power markets extends well beyond ERCOT, with Microsoft’s 40 GW renewables push demonstrating how Big Tech balance sheets are now setting the pace for utility buildout in multiple grid jurisdictions simultaneously.

The practical read is that developers with Big Tech relationships get funded first, and corporate sectors without that access may quietly miss their ESG mandates for lack of available supply.

Additionality, curtailment, and the growing scrutiny on net-zero claims

Now for the part the glossy renewable announcements tend to skip: the physics of ERCOT does not care about a corporate net-zero slide.

Meta pursues 100% renewable matching for its regional operations, but West Texas grid congestion complicates the claim. In 2024, more than 8 TWh of wind and solar were curtailed in ERCOT, meaning the grid physically refused to take that power. Because most PPAs settle “as-produced,” curtailed energy is neither delivered nor paid for, which strips developers of expected revenue and, critically, reduces the volume of Renewable Energy Certificates (RECs) generated for the buyer.

Fewer RECs means a weaker claim to 100% renewable status. Lenders now price this in, stress-testing curtailment assumptions of 3% to 6% for West Texas nodes before committing capital.

The grid bottleneck constraining ERCOT solar delivery is part of a wider infrastructure ceiling that also threatens the nuclear capacity additions being fast-tracked across the western US, where interconnection queues and transmission investment shortfalls create the same bankability problems for a different generation technology.

Buying a year’s worth of RECs is not the same as powering a data centre with clean electricity in the hours it actually runs. During peak curtailment, the certificates on paper and the electrons on the wire can tell completely different stories.

The scrutiny is no longer academic. A recent US Senate probe examined Meta’s operation of a gas-powered AI data centre in Louisiana, sharply questioning how the company reconciles climate-leadership messaging with fossil-fuelled infrastructure. A shareholder proposal has separately argued that Meta’s reliance on RECs exposes it to regulatory and litigation risk as its actual energy emissions grow.

For energy investors, the lesson is that a large PPA no longer buys immunity from climate criticism. Basis risk and congestion directly threaten both the environmental credibility and the financial returns of these high-profile contracts.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.

Gauging the limits of the corporate renewables boom

The tension running through this deal is straightforward: tech giants command near-limitless capital, and regional grids like ERCOT impose hard physical limits that capital alone cannot buy through.

Proposed changes to Scope 2 greenhouse-gas accounting could tighten additionality rules, potentially excluding legacy assets and pushing buyers away from REC-heavy strategies toward location-matched or hourly carbon-free energy. Should that shift land, it would narrow the eligible buyer pool and pressure returns on projects financed the old way.

Corporate PPAs remain the primary engine for new US solar development, and Meta’s latest commitment reinforces that. But the megawatts on a contract only decarbonise the grid if the grid can actually deliver them. Transmission and interconnection have to catch up for these investments to work as advertised.

Transmission and interconnection constraints have become the binding limit on how many contracted megawatts actually reach the grid on schedule, with interconnection queue backlogs in ERCOT and other US markets now running years longer than the construction timelines of the projects waiting in them.

Past performance does not guarantee future results, and these forward-looking regulatory scenarios are speculative and subject to change based on policy and market developments.