Texas Data Center Moratorium: Gina Hinojosa's Upset Win Puts a Statewide Ban to Its First Real Test

Texas was already the strangest front in the American data center boom before Tuesday. It held close to 30 percent of all the data center capacity planned for the United States in 2028, with 6.5 gigawatts under development, and its grid operator was staring at an interconnection queue of nearly 1,800 projects seeking almost 474 gigawatts — roughly five times the state's record peak demand, with data centers accounting for about 90 percent of the queue. Then, over the summer, the state's Republican governor, Greg Abbott, started pulling the emergency brake himself: on June 10 he ordered regulators to make data centers pay the full cost of the grid infrastructure they need rather than shifting it to household bills; on August 3 he halted new grid connections pending an audit of every project in the queue; on September 14 he ordered water-use reporting enforced; and on September 21 he told the Texas Commission on Environmental Quality to issue no permits sought by data centers until the audits are done. The audit is due to report by December 10. What Abbott built, step by step, was a de facto moratorium — imposed by executive directive, with no fixed end date and no statute behind it.

On Tuesday, Texas voters replaced him with the candidate who had called that freeze a scam. Democrat Gina Hinojosa, who argued that Abbott would lift his own freeze on November 4, the day after the election, won on a promise of an immediate moratorium on data centers lasting until the legislature can meet and pass real protections: a voice for local communities, safeguards for household bills, and an end to data center tax incentives. Her victory was one of five Democratic gubernatorial gains in a midterm in which 26 candidates across 21 states ran on anti-data-center platforms, but Texas is the one that matters most to the industry, because it is where the most capacity is actually being built. The question now is not whether Texas pauses. It is whether a pause becomes a law.

That distinction is exactly where the prediction market draws its line, and it is why I find this market so interesting. On Polymarket, the question "Texas enacts a data center moratorium" resolves to Yes only if the state enacts a law or constitutional provision establishing a statewide moratorium. An executive freeze, however sweeping, does not count; neither does merely suspending tax incentives. As I write this on the evening of November 4, traders price a law by June 30, 2027 at just 16.5 percent — up from 13.5 percent the day before, as the election result sank in — and by December 31, 2027 at 21.5 percent. Even by the end of 2028, the price is only 34.5 percent. The market, in other words, is saying: a governor can freeze permits with a letter, but a legislature has to pass a ban, and that is a much harder thing to do.

I think the market is roughly right about the difficulty and slightly too low on the odds. The 90th Texas Legislature convenes on January 12, 2027 and sits until May 31, giving Hinojosa a real session in which to convert her mandate into a statute, and a governor who has just won on this exact issue can also call special sessions if the regular one stalls. Against that, a full statewide ban is the most extreme option on the table, and the path of least resistance in Austin has always been conditions rather than prohibition: make developers pay for their grid upgrades, publish their water draw, give neighbours a say, strip the tax breaks. Those measures would change the economics of the buildout without ever triggering a market resolution. And the industry's counter-argument has not gone away either — that Texas invited this boom, that the campuses bring jobs and tax revenue, and that a ban simply pushes the same electricity and water demand into the next state over.

So my prediction is deliberately narrow, and it follows the market's own definition: Texas will enact a law establishing a statewide moratorium on new data centers by December 31, 2027. I put it at 30 percent — above the market's 21.5 percent, because a governor elected on this single promise, with a session ahead of her and the power to call more, is a stronger force than traders seem to credit, but well short of even odds, because the legislature still has to choose a ban over the easier politics of conditions. If I am wrong, the most likely way is the quiet one: Texas regulates data centers hard, bills stay protected, communities get their hearings, and no statute ever uses the word moratorium. That would be a victory for Hinojosa's voters in substance and a loss for this prediction in form — which is precisely why the distinction between a freeze and a law is worth watching. Texas spent 2026 discovering that it could stop the machines with a letter. In 2027 we find out whether it can stop them with a law.