Clara Bennett: Every year, some investors end up paying taxes on gains they didn't actually realize — not in any meaningful sense. The asset went up just enough to keep pace with inflation. They sold. They owe. That's the problem indexation says it fixes.
Clara Bennett: And look — it IS a real problem. I want to be clear about that before I get into what the proposed solution actually looks like when you put it in practice.
Clara Bennett: Here's how the mechanics work. Current law taxes nominal gains — the full spread between what you paid and what you sold for. Indexation adjusts the purchase price, the basis, upward by cumulative inflation before that math runs. So you buy at $100,000, inflation's up 10% — your basis becomes $110,000, and your taxable gain just shrank by $10,000.
Clara Bennett: Clean concept. The question is who actually owns the assets where that concept has real dollar weight.
Clara Bennett: Donald Trump is reportedly looking at this right now. Scott Bessent — his Treasury Secretary — is being pressured by conservative advocates and Republican senators to implement it. Early 2026. Actively.
Clara Bennett: The Yale Budget Lab put a number on it: retroactive indexation — covering assets people already hold — costs approximately $1 trillion over ten years. Prospective-only, just new purchases going forward, runs about $170 billion over the same period.
Clara Bennett: $830 billion separates those two options.
Clara Bennett: And here's where the distributional picture becomes — I mean, you can't ignore this. The top 0.1% of earners would get an average tax break of approximately $350,000. The bottom 40% by income? Zero. No benefit.
Clara Bennett: Think about what that gap actually represents. The people who need the inflation adjustment to mean anything are, almost by definition, the people holding the largest, longest-appreciated positions. Assets like Elon Musk's approximately 413 million long-held Tesla trust shares — where analysis estimates his potential savings at hundreds of millions to potentially billions of dollars.
Clara Bennett: That is the real-world shape of a proposal that gets described as correcting an unfairness in the tax code. It may correct one. It creates another.
Clara Bennett: Now — before I just write that off — the academic case deserves a real hearing.
Clara Bennett: A 2023 study in the Canadian Tax Journal ran fifty years of Toronto Stock Exchange data — 1971 through 2020 — and found that taxing nominal gains systematically overstates or understates real income depending on the asset and the holding period. That's not a political argument. That's measurement error baked directly into the tax code.
Clara Bennett: Leonard Burman, in The Labyrinth of Capital Gains Tax Policy, actually sits with both sides of this seriously. His work doesn't hand you a clean verdict — it hands you the real tradeoffs.
Clara Bennett: And there's a theoretical model worth knowing about: taxing indexed gains at ordinary income rates could actually improve investor welfare AND increase total revenue compared to taxing nominal gains at the long-term rate. The math can work.
Clara Bennett: The intellectual case is not invented.
Clara Bennett: But — and this is where the practical picture diverges pretty sharply from the theory — there's a wall. A legal wall that's been there for thirty years.
Clara Bennett: In 1992, the Department of Justice Office of Legal Counsel looked at whether Treasury could index capital gains by regulation alone. The answer was no. Treasury lacks the unilateral authority. Full stop.
Clara Bennett: The Tax Law Center revisited that exact question in May 2026 and landed in the same place — unlawful, administratively chaotic, and likely to get reversed by courts or a future administration. Those aren't soft concerns.
Clara Bennett: Congress would have to do this. And Congress hasn't.
Clara Bennett: Here's what makes that telling — this exact movie played out in Trump's first term. Larry Kudlow, who'd championed executive indexation for years, explored it from the White House. Didn't proceed. Steven Mnuchin was pressed by Americans for Tax Reform to act. Also didn't proceed. The legal ceiling stopped them both.
Clara Bennett: So when Kevin Hassett goes on Fox Business and says voters should expect 'a lot more policy proposals between now and the midterms,' and he's naming capital gains indexation as part of the Republican agenda — what is that, exactly?
Clara Bennett: It could be a genuine legislative push. But it could also be a signal — something you put on the agenda to animate a base heading into an election.
Clara Bennett: That framing — midterms, agenda, proposals — is worth sitting with.
Clara Bennett: Because the gap between the fairness theory and the actual distributional outcome — $350,000 to the top 0.1%, nothing to the bottom 40% — that gap doesn't close just because the intellectual scaffolding is sound.
Clara Bennett: The key is distinguishing a policy designed to fix a real problem from one designed to benefit specific holders of long-appreciated assets while borrowing the language of fairness.
Clara Bennett: Those can look identical from a distance. They don't feel the same when you run the numbers.
Clara Bennett: Here's where the road actually forks — and this is the part worth watching closely.
Clara Bennett: The administration has two paths. Executive action — Treasury rewrites the regulations, redefines 'cost' to include an inflation adjustment, rolls it out. Or legislation — Congress passes a bill, gets a CBO score, goes on record. Those are not equivalent choices. Not legally, not politically, not practically.
Clara Bennett: The executive path is the one that breaks first.
Clara Bennett: The 1992 DOJ OLC opinion said Treasury lacks the unilateral authority. The Tax Law Center's May 2026 analysis lands in exactly the same place — unlawful, administratively chaotic, likely reversed by courts or a future administration. That's not one skeptic raising a hand. That's thirty-plus years of consistent legal conclusion. If Bessent moves by regulation, it gets challenged fast, and the odds in court are not good.
Clara Bennett: Which leaves Congress. And Congress has a long history of not passing this.
Clara Bennett: Tom Emmer introduced H.R. 2017. Devin Nunes introduced H.R. 6444. Neither was enacted. These bills exist — they got numbers, they went to committee, and they stopped. That's the legislative graveyard this proposal keeps returning to. The Bipartisan Policy Center has been tracking the renewed 2026 pressure on Bessent, and the pattern looks familiar — advocates push, Treasury nods, Congress doesn't move.
Clara Bennett: Now — if Congress did move, the design question becomes EVERYTHING.
Clara Bennett: The Yale Budget Lab's numbers make this concrete: retroactive indexation — covering assets already held — runs approximately $1 trillion over ten years. Prospective-only, just new purchases going forward, comes in around $170 billion over the same window. The number changes by a factor of six depending on which version moves. That's the hinge. That single design choice — retroactive or prospective — is what separates a narrow technical fix from a multitrillion-dollar wealth transfer to people who already hold the largest long-appreciated positions.
Clara Bennett: If you're watching this unfold — and I think it's worth watching carefully — that's the number to track. Which version gets drafted. Whether it's retroactive.
Clara Bennett: Which brings me back to Hassett. 'A lot more policy proposals between now and the midterms.' In practice, that framing tells you something. A serious legislative push gets a bill number and a CBO score and a markup date. A donor signal gets TV appearances. Watch which one this becomes — because those two things do not look the same by spring.
Clara Bennett: The word the proposal keeps reaching for is 'real.' Taxing only real gains. Not phantom gains. Not inflation. Real.
Clara Bennett: And that framing does a lot of work — because it sounds neutral, right? It sounds like a measurement problem getting fixed. But ask yourself who actually gets to experience the difference between nominal and real in dollar terms significant enough to matter, and the answer is not ambiguous. The Yale Budget Lab ran the numbers. Top 0.1% — $350,000 average tax break. Bottom 40% — zero. Not small. Not marginal. Zero.
Clara Bennett: So 'real' turns out to be a word with a location. What counts as a real gain — a gain worth correcting for — depends entirely on where you sit in the income distribution. If you're holding 413 million long-held Tesla trust shares, the inflation adjustment on your cost basis is worth potentially billions. If you're in the bottom 40%, the adjustment is worth nothing, because you don't hold the assets the adjustment touches. The math is technically the same for everyone. The benefit is not.
Clara Bennett: That's the irony sitting at the center of this. The fairness language is genuine — the academic case is real, the measurement problem is real — but the policy as designed answers the question 'real for whom?' in only one direction.
