The Return of the Energy Transition

The first third of the post is a concession, and most of the voices in it are not his. He reports Dan Yergin calling the whole endeavor troubled, quotes Matt Yglesias arguing in the New York Times that an uncompromising climate left has made it impossible to win the majorities decarbonization would need, and quotes Bill Gates saying the doomsday framing is diverting resources from things that would help more. Lubershane’s own contribution is to accept three of the underlying claims: hydrocarbons are extraordinary resources and substitutes for them are extremely hard, the energy system has enough inertia that forcing rapid change creates losers who will resist, and energy addition matters as much as energy transition. He says outright that he has not always fully wrestled with the consequences of those. What he refuses to drop is a fourth, that climate change is still a problem that needs solving, and he makes that point by quoting the sitting US energy secretary rather than an advocate. The result is the quadrilemma, plus a timing claim he declines to turn into a fifth axis: net-zero by 2050 was never realistic, and he will say only that an 80 to 90% cut takes three or four more decades, or seven or eight.

Priority one is gas, argued from the record rather than from hope. The largest aggregate emissions reduction achieved so far came from gas displacing coal in the US power sector, and he thinks the usual telling understates gas, because gas plants absorbed most of retiring coal’s duty while renewables mostly displaced marginal gas. That complementarity is why he treats gas as renewables’ partner rather than their rival, from grid-balancing generators sited where power is used through to industrial heat, where for a plant with access to ultra-cheap solar he says the ideal setup could well be solar, thermal storage and a gas boiler, so that the solar and storage need not be overbuilt to cover occasional lulls. Geopolitically he wants gas in the mix because he calls expanding its role one of the best ways to give a hydrocarbon superpower a stake in the outcome. Then the catch, and it is quantitative. A methane leak rate of just 2 to 3% materially undermines gas’s advantage over coal, and 2 to 3% is his estimate of the current US average from wellhead to power plant. The load in the phrase “responsible gas” is carried by the adjective.

Nuclear and solar are where he argues against a received view, in opposite directions. On nuclear he grants the three standard objections, cost, schedule and the variance around both, then relocates them: the engineering, supply chain, safety and security requirements make nuclear more dependent than any other resource on doing the same thing repeatedly, which makes it a collective action problem rather than a technology problem. His three conditions follow from that, a very limited number of designs, a deliberately staged pipeline of projects, and proportionate regulation, and none of them is something a competitive market produces, so he concludes that some central planning among government, industry and utilities is necessary. The same reasoning makes him cool on small modular reactors: their cost advantage is a theory running against the known economies of scale in large power plants and in most other heavy industries, only building a lot of them would settle it, and the largest potential buyers each appear to have backed a different design. On solar he attacks the opposite error, that renewables have already won. Comparing levelized costs is the wrong comparison, because intermittent output is not firm capacity; the comparison he thinks honest is renewables’ levelized cost against the marginal fuel cost of burning coal or gas, and on that basis, unsubsidized, fossil generation still wins even in the best American locations. Storage helps in most scenarios and always adds cost. Renewables are also geographically constrained and land-hungry, and he says advocates should concede that with more gravity than they have. Yet solar is still the only resource he gives a fighting chance of reaching the cost of energy embodied in a lump of coal, a line he quotes from his own recent series, which is why the priority is to keep pushing it rather than to declare it finished.

The last two priorities concern the physical substrate. Doubling the grid is what he calls a conservative estimate of what electrification and data centers will require, and it has to happen at roughly half the cost, which points him at technology that lifts the throughput of existing lines and at power electronics. But transmission lands in the same place as nuclear, as a collective action problem whose benefits are diffuse and whose costs are local, and the compromise he thinks it requires is accepting continued investment in oil and gas pipelines alongside the grid. Supply chains then close the loop, because on his framing the energy transition is mostly a mineral transition; copper is the bottleneck he worries about most; and China’s position runs past minerals into the whole manufacturing stack. His conclusion there is unusually blunt for a piece about Western industrial capacity: neither America nor Europe can rebuild that alone, so they should work closely with Korean and Japanese firms while making bigger bets on processes that route around Chinese supply chains. He ends by refusing the obvious reading of his own list. Priorities are temporary stances rather than maps, and a five point plan is something to be wary of.