Five big questions emerging from the OBBB

The cross-cutting variable is the foreign entity of concern restriction, which Lubershane calls the biggest question mark because it sits on top of everything else: the investment and production tax credits for deployed renewable and storage projects, and the 45X manufacturing credit. His working definition is that you cannot buy things originating in or controlled by Chinese companies, and he flags immediately that lawyers would find that gloss inadequate. The impact then splits by technology for supply-chain reasons rather than legal ones. Wind is largely fine because non-Chinese turbine components are readily available, though he notes wind has other problems in the bill. Solar projects are mostly fine for a slightly perverse reason: years of tariff pressure had already pushed the industry toward India and other parts of south and southeast Asia, so the supply chain was messy in a way that now works in its favor. The real exposure is upstream in manufacturing. Ingot and wafer production is, in Lubershane’s view, the single most China-concentrated step left in all of clean energy, and the threshold ramp is what turns that from tolerable into a problem: a new US cell plant could probably qualify today on Chinese wafers, because ingot and wafer together are around a third of cell cost against a 50 percent non-restricted requirement starting in 2026, but the requirement reaches 85 percent by 2029.

Batteries are harder, and Lubershane thinks stimulating battery manufacturing was the most important thing the IRA was trying to do toward making clean energy manufacturing robust against geopolitical risk. The arithmetic is unforgiving for nickel-manganese-cobalt cells, where cathode active material can be roughly half the cost of a cell and graphite anode powder another 10 to 15 percent, both heavily concentrated in China, against a 60 percent non-China requirement from 2026. Non-Chinese supply exists but is thin, so the likely results are more expensive US cells and genuine bottlenecks if many plants try to ramp at once. What makes him confident anyway is not the tax code but the political economy of where the plants went. He cites Ford proceeding with a $3 billion Michigan battery plant licensing lithium iron phosphate technology from CATL, the Chinese battery giant, in a county Trump won by 56 percent, feeding EV assembly in other reliably Republican districts. His reading is that the IRA’s architects were deliberately trying to insulate the manufacturing rollout from political risk, and that constituencies of this kind, more than the tax code, are what make a program durable.

The wind and solar question is the one they most explicitly cannot answer, because it was open at the time of recording. The bill lets projects that commence construction by the end of 2026 escape a restrictive in-service deadline, which under historical precedent includes safe harboring by spending 5 percent of project cost on equipment. An executive order issued immediately afterwards directs Treasury to revisit those definitions, presumably more aggressively, and the 45-day window had not closed. Lubershane’s assessment is pointed: this is the biggest uncertainty and it is not really because of the bill. He expects nothing much to happen while everyone waits, then, if the rules resemble past practice, a rush by large developers with the balance sheets to buy inverters and modules and the pipeline visibility to know they will be used, which advantages the best-capitalized players. His quantification of what is at stake is the most useful number in the episode: wind and solar at roughly $25 per megawatt-hour with credits sit below the marginal cost of running existing gas plants, so they get built on avoided fuel cost alone, while at the $40 to $45 they would cost without credits, they have to compete as an additional resource rather than a substitute for fuel. He is careful that this does not mean the market disappears, and careful again that it is not clear how much the credits matter to near-term build given how much power buyers want. Kann adds the sharper worry, which Lubershane accepts: a world where some projects qualify and some do not may damage market scale more than clean removal would, because the industry can reprice against certainty but not against a coin flip.

The last three questions get shorter answers that mostly deflate the premise. On nuclear, geothermal and carbon capture keeping their credits longer than wind and solar, Lubershane says it barely changes his five-year view, because those resources arrive later anyway and are added for different reasons rather than in competition with solar. On hydrogen, the deadline extension from end-2025 to end-2027 is real relief, but the three pillars rules still require clean power that is new, deliverable and hourly matched, and that power is exactly what data centers are bidding up, so Kann’s estimate is more than a couple of projects and well short of a hundred. On EVs, Lubershane stays bullish on the consumer side precisely because buyers at 10 percent market penetration are unusual people who are probably not very price sensitive, and less bullish in the near term on the commercial side because fleet operators are the opposite. The through-line back to batteries is political economy rather than tax policy: hydrogen never got off the ground under Biden because qualification took too long, so it never built a constituency to defend itself, while battery plants did.