Most energy analysts see rising electricity bills as an inevitable side effect of the green transition or extreme weather. Dylan Patel flips this script entirely, arguing that a massive, self-inflicted modeling error by the regional grid operator PJM has already cost ratepayers $12 billion and threatens to waste even more. This is not a story about the weather or the cost of fuel; it is a forensic accounting of how a broken algorithm and a paralyzed governance structure are bleeding consumers dry while the grid operator plans to double down on the same mistakes."
## The $12 Billion Calculation Patel's central claim is staggering in its specificity: the grid operator's internal model, the 'Reserve Requirement Study,' is fundamentally broken. He writes, "PJM's model includes errors that we estimate have cost all of its 66 million residents a total of $12B between 2025 and 2027 alone." This isn't a vague projection; it is based on a six-month effort by his team to reverse-engineer a system that PJM has treated as a black box. The core of the argument rests on two specific oversights. First, the model ignores the physics of cold air, which makes gas turbines up to 25% more efficient in winter. Second, it fails to credit the billions of dollars in winterization investments mandated after the 2022 Storm Elliott, a blizzard that previously exposed the grid's fragility. As Patel puts it, "PJM underestimates by ~4 gigawatts the existing power plants it already has; owing to a methodology which doesn't account for the higher efficiency of power plants in winter and improved power plant resilience since Storm Elliott."
This framing is powerful because it isolates the problem from the usual political noise. The failure isn't that there aren't enough power plants; it's that the operator doesn't know how many it actually has. The evidence suggests that if the model were accurate, the emergency auction planned for late 2025 would be unnecessary, or at least significantly smaller. Critics might argue that modeling future grid reliability is inherently uncertain and that conservative estimates are a safety feature, not a bug. However, Patel's data suggests the margin of error has shifted from "prudent" to "prohibitive," driving costs up by 20% without a corresponding increase in actual reliability."
PJM has wasted ~$12 billion of ratepayers' money from 2025 to 2027 due to this weak methodology, which dramatically overstated the supply/demand shortfall it faced.
## A Market Designed to Fail The commentary deepens when Patel examines the structural incentives of PJM's capacity market. He argues the system is "structurally anti-growth" because it treats new and existing power plants identically in its auctions. This design flaw means that when the grid operator buys capacity to cover a perceived shortfall, it pays a massive premium to existing plants that don't need the money to stay online. "PJM also runs the only capacity market in the world that does not distinguish between new and existing power plants," Patel notes. "So that premium paid for new power plants is also paid to existing power plants for doing nothing."
The consequence is a transfer of wealth from households to incumbent generators, with little to no new infrastructure actually being built. The article highlights that despite four record-breaking auctions totaling $63 billion, only 4.8 gigawatts of new capacity was procured. This is a stark contrast to the urgency of the situation. The governance structure, which requires a two-thirds majority across five sectors to change rules, has created a "vetocracy of vested interests" that prevents reform. While the Federal Energy Regulatory Commission (FERC) has given PJM a deadline to fix its governance, Patel warns that FERC's powers are largely reactive. The system is stuck in a loop where the only solution offered is to spend more money on the same broken mechanism.
## The Emergency Auction Risk Perhaps the most alarming section concerns the upcoming "emergency auction." With the current model predicting a massive shortfall, PJM plans to sign contracts lasting until 2043. Patel points out a critical flaw: these contracts are being sold without committed buyers. "PJM is doing this with no committed counter-parties," he writes. "If no other counter-party emerges, then once again those left holding the bag will be the residential ratepayers." The plan relies on large datacenters to foot the bill, but these developers are increasingly opting out, building their own power, or leaving the region entirely. "Many datacenter developers have written off PJM regardless, going elsewhere or planning behind-the-meter power configurations."
This creates a scenario where the grid operator is essentially gambling with ratepayer money. If the large loads don't materialize, the cost falls squarely on the 66 million residents in the PJM region. The parallel to the December 2022 North American blizzard is instructive; just as that storm forced a re-evaluation of winter resilience, the current crisis should force a re-evaluation of how capacity is valued. Yet, the administration's response seems to be more of the same: signing long-term contracts based on flawed data. The argument here is that the emergency auction is not a safety net, but a potential trap that could lock in high rates for nearly two decades."
## Bottom Line Patel delivers a devastating critique of PJM's capacity market, backed by a rare, transparent reconstruction of the operator's own flawed models. The strongest part of the argument is the clear link between a specific modeling error—ignoring winter efficiency—and the $12 billion price tag on ratepayer bills. The biggest vulnerability lies in the political reality: even with this evidence, breaking the "vetocracy" of vested interests within PJM's governance may be impossible without external regulatory force. Readers should watch closely for FERC's intervention by the end of September, as the window for correcting this course before the emergency auction begins is rapidly closing.