Grid War: Data Centers vs. Citizens — Who Should Control America's Energy Future?

Let's follow the incentives and see where they actually lead. The core economic mechanism here is allocation by price versus allocation by political decree. When you cap data center energy use or force utilities to prioritize one group, you override the price signals that tell us where electricity is most valuable. That doesn't make the demand go away—it just hides it, leading to shortages, rationing, or wasted investment. If we let markets work, high demand from data centers will drive up prices, which signals both consumers and producers to adapt. Households and businesses will invest in efficiency, utilities will expand supply, and capital will flow to where it's most productive. Yes, there are distributional concerns—low-income households need protection from price spikes. But that's best handled with targeted subsidies, not blunt rationing.History is clear: when governments try to pick winners—whether it's Soviet bread lines or California's rolling blackouts—the result is inefficiency and, often, greater inequality. The real tradeoff is between dynamic growth and static allocation. If we want innovation, we need to let prices do their job. So, no, regulators should not impose arbitrary caps, and utilities should not override market signals. The question is not who deserves the electricity, but how we grow the pie so there's more for everyone.

.

Everyone in this room is nodding along to "let markets work." That makes me nervous. The Economist just invoked California's rolling blackouts as a cautionary tale about government interference. But here's what actually happened in California in two thousand one — it was deregulation that created the crisis. Enron manipulated those markets. Prices spiked artificially. Households got crushed. The market didn't protect anyone. It was gamed. So when my opponent says "let prices do their job" — I want to ask, whose prices? Set by whom? Governed how? And look, the framing here is a false binary. Either caps and rationing, or pure market allocation. But there's a third option the room isn't discussing: public utility obligation. Electricity isn't a luxury good. It's infrastructure. We don't let markets decide who gets water pressure or road access. Why is electricity suddenly different the moment a data center moves in? Here's the uncomfortable fact — data centers are, uh, already negotiating sweetheart deals with utilities. Special rates. Priority access. Not because the market rewarded them, but because they have leverage that residential customers simply don't have. That's not a free market. That's a rigged one. The consensus tonight is drifting toward "markets good, regulation bad." But the last time we trusted that consensus in energy markets, people died in Texas in twenty twenty one when the grid failed. So I'll ask The Economist directly — why are you so comfortable agreeing with everyone else on this one? .

. .

The Contrarian raises real cases — California 2001, Texas 2021 — but let me be precise about what those actually demonstrate, because the mechanism matters enormously. California's crisis wasn't caused by markets. It was caused by a hybrid — retail prices were capped while wholesale prices were deregulated. That's not a market. That's a price control sitting on top of a market, creating exactly the arbitrage opportunity Enron exploited. The lesson isn't "markets fail." It's "half-measures create perverse incentives." Texas is more complex. ERCOT's grid failure was partly a weatherization problem — a regulatory failure to mandate winterization standards. PLAUSIBLE INFERENCE: a properly structured market with full cost pricing, including reliability obligations, would have priced that risk in. Now, the sweetheart deals point — that's actually my strongest argument, not the Contrarian's. When data centers negotiate special rates, that's not a free market. That's regulatory capture. The solution isn't more political allocation — it's transparent, competitive pricing where everyone, residential and commercial, faces the true cost of power. Here's the mechanism question I want answered: The Contrarian invokes public utility obligation. Fine. Who sets the allocation? A regulator. Which regulator? Appointed by whom? Accountable to what incentive structure? Because the history of utility regulation in America is, uh, not a story of equity — it's a story of incumbent protection and rate-payer capture. Tell me the mechanism, not just the intention.

The Economist wants a mechanism. Fair enough. But notice what just happened — they spent three paragraphs defending markets by explaining why the markets we actually have aren't real markets. California wasn't a real market. Texas wasn't a real market. The sweetheart deals aren't a real market. At some point, you have to ask — if every market failure is explained away as "not a true market," are we debating theory or reality? Here's my mechanism: the Federal Energy Regulatory Commission already has authority to set interconnection standards and reliability obligations. States already run public utility commissions. The infrastructure exists. The question is whether we use it, or whether we keep waiting for the perfect free market that has never actually existed in American energy. And look, the Economist says transparent competitive pricing where everyone faces the true cost of power. I actually don't disagree with that in principle. But right now, uh, that's not what's happening. Right now, a hyperscaler can negotiate a fifteen-year fixed rate deal with a utility, lock in cheap power, and leave residential customers absorbing the grid upgrade costs. That's the actual market we're operating in. The consensus in this room is drifting toward "regulation is the problem." But the Public Utility Holding Company Act of nineteen thirty five — passed after private energy markets collapsed and left millions without power — that was regulation. And it worked for decades. So I'll ask The Economist again — why are you so comfortable defending a market that, by your own admission, doesn't actually exist yet? .

Let me

pull the data on that. The data actually cuts both ways here, and intellectual honesty requires I say so. VERIFIED: The U.S. Energy Information Administration confirms residential electricity prices rose eleven and a half percent in twenty twenty five, outpacing inflation. Bloomberg reports wholesale electricity costs up two hundred sixty seven percent near data center clusters. Vermont Law School's research confirms traditional ratemaking does allow cost-shifting onto residential consumers. The Contrarian is right about that mechanism — it's real and it's documented. But here's the other VERIFIED finding: PG&E's CEO reports that data center growth actually cut their rates eleven percent since twenty twenty four — each new gigawatt of load shaving roughly one percent off customer bills. So the outcome depends entirely on the rate structure, not on whether data centers exist. That's my point. The Contrarian is diagnosing a real problem — cost-shifting — but prescribing the wrong cure. The cure isn't caps or political allocation. It's fixing the rate structure so data centers pay full cost-reflective pricing, including grid upgrade costs. That's a market mechanism, not a market rejection. And to answer the question directly — I'm not defending the market that exists. I'm arguing for the market that should exist, because the alternative — regulatory allocation — has a documented track record of protecting incumbents, not consumers. So here's my question back: if FERC and state PUCs already have authority and the problem persists, why do you trust the same institutions to fix what they've already failed to prevent?

That's actually a sharp question. And I'll give it a straight answer. You're right that FERC and state PUCs have failed to prevent cost shifting. But notice what The Economist is doing — using institutional failure as an argument against regulation, while simultaneously arguing we need better rate structures, which are... regulations. You can't have it both ways. The argument isn't "trust the existing institutions blindly." It's that the answer to captured regulation is better regulation, not no regulation. We didn't respond to the FDA's failures by abolishing drug safety standards. We reformed them. And here's what the data The Economist just cited actually proves — the outcome depends entirely on rate structure. Their words. Which means this is fundamentally a policy design question, not a markets versus government question. The framing of this entire debate has been a false binary. Now, the PG&E example — rates dropped eleven percent near data centers. Great. But that's one utility, one region, one rate structure. Meanwhile Vermont Law School documents cost shifting happening systematically elsewhere. You can't cherry-pick the success case and ignore the pattern. Here's the deeper problem. The Economist is arguing for a market that should exist. I'm arguing about the grid that does exist, serving real people right now. Uh, theoretical market design is a luxury. Keeping the lights on in Phoenix in July is not. The consensus tonight keeps retreating to "fix the incentives" without ever saying who fixes them, how, or what happens to the people getting squeezed in the meantime. So why is The Economist so comfortable defending a solution that, by their own admission, requires institutions they just told us can't be trusted? .

That's a fair hit, and I'll take it directly. You're right that I can't simultaneously distrust regulatory institutions and rely on them to implement better rate structures. That's a genuine tension in my argument, and I acknowledge it. But here's where I push back on the resolution. The Contrarian's answer is "better regulation." Fine. I agree. But "better regulation" is not a mechanism — it's a destination. The question is what design principles get us there. And that's where market signals become essential, not optional. Here's the concrete mechanism I'm proposing: cost-reflective pricing, meaning data centers pay the full marginal cost of the grid infrastructure their load requires. Not negotiated sweetheart deals, not socialized grid upgrades. Full cost internalization. That's not theoretical — it's, uh, standard utility economics. The EPRI framework for high load factor customers already outlines how this works. Now, to the people getting squeezed in Phoenix in July — that's a real and urgent concern. The answer is direct income support or targeted rate subsidies for low-income households, not suppressing the demand signal that tells us we need more generation capacity. Because if you cap data center energy use without expanding supply, you don't lower prices — you just shift who gets rationed politically. The Contrarian and I actually agree on more than this debate's framing suggests. The disagreement is narrow but important: I want price signals to drive the allocation, with redistribution on the income side. They want allocation decisions made politically, with markets playing a secondary role. So tell me — when a regulator decides Phoenix residents get priority over a data center, what's the mechanism that prevents that same regulator from deciding the politically connected data center gets priority next cycle?

The Economist just conceded a genuine tension in their argument — and then immediately pivoted to asking me about regulatory capture. That's a tell. When your opponent starts asking your questions for you, they're worried about the answer. So here it is. The mechanism that prevents a regulator from favoring the politically connected data center next cycle is the same mechanism that prevents a utility from doing it right now — transparency, public rate cases, judicial review, and democratic accountability. Imperfect? Absolutely. But The Economist's alternative is a market where the politically connected already win, just through lobbying and contract negotiation instead of regulatory hearings. And look — we've now reached a remarkable moment in this debate. The Economist has conceded that cost shifting is real and documented. They've conceded that the current market doesn't work. They've conceded that institutions are unreliable. And yet the conclusion is still... trust the market. That's not logic. That's faith. Here's what I actually think is happening. The consensus in this room has quietly shifted from "markets are efficient" to "markets are the least bad option." That's a much weaker claim. And it's one I'm, uh, more sympathetic to than the opening argument suggested. But "least bad" still requires us to answer — least bad for whom? Because the twelve percent of grid consumption that data centers will consume by twenty twenty eight isn't abstract. It's megawatts that someone else doesn't get. And right now, the people who don't get them are the ones without fifteen-year fixed rate contracts and lobbyists in Washington. The Economist wants price signals. Fine. But price signals only work when everyone can actually respond to them. A family in Phoenix can't relocate their home to a cheaper grid zone. A hyperscaler can.

The Contrarian

just made their strongest point of the debate, and it deserves a direct response. The mobility asymmetry is real. VERIFIED. A hyperscaler can site a data center in Wyoming where power is cheap.

Here's what this debate actually

Here's what this debate actually revealed. The Contrarian and I agree on the diagnosis — cost-shifting is real, sweetheart deals are real, the current system is rigged. Where we diverge is the cure. My opponent wants democratic accountability and regulatory oversight to fix what regulatory capture already broke. That's not cynicism — it's a mechanism question. The same institutions that allowed hyperscalers to lock in fifteen-year fixed-rate deals will be the ones setting the caps and the priorities. Changing the label from "market allocation" to "regulatory allocation" doesn't change the underlying incentive structure of who has leverage. Here's the economic choice this debate presents. You can allocate electricity by price — imperfect, gameable, but self-correcting when designed with full cost-reflective pricing. Or you can allocate it by political decision — also imperfect, also gameable, but slower to correct and more vulnerable to entrenchment. The family in Phoenix can't relocate. That's true. But the answer to that is direct income support and low-income rate protection — not suppressing the price signal that tells investors we need more generation capacity in Phoenix. Data centers consuming twelve percent of U.S. electricity by twenty twenty eight isn't a crisis. It's a signal. Build more power. Price it honestly. Protect the vulnerable directly. Incentives don't care about your intentions. They respond to structure. Design the structure well — cost-reflective pricing, transparent rate cases, targeted subsidies — and you get both innovation and equity. Design it badly, and you get rationing with a democratic veneer. The economy is not a morality play. It is a system. And right now, we have a badly designed one. The answer is better design — not abandoning the only mechanism that actually scales.

proved — and it's not what either of us argued. The Economist came