CORNELL SCORES THE $300 BITCOIN TAX BREAK AS A MONEYMAKER — $859 MILLION MORE OVER TEN YEARS — THE COFFEE TAX WAS ONLY EVER COLLECTING PAPERWORK
Only a third to a half of owners report the gains at all. The coffee tax was never revenue — it was paperwork with a penalty attached.
Under current law, spending Bitcoin is a taxable event at any size. Buy a $5 coffee with a coin that has gone up, and you owe a capital-gains calculation, a cost-basis lookup, and a line on your return. Cornell’s Tech Policy Institute just ran the fiscal math on Senator Lummis’ fix — Section 2 of S. 2207, which would exempt personal transactions under $300 with a $5,000 yearly cap — and found the exemption MAKES the Treasury money: roughly $859 million more over ten years. The mechanism is the honest part. The cost of an exemption depends on what the government actually collects, not what is owed on paper — and the report finds only 32 to 56 percent of digital-asset owners report gains at all. At a 40 percent reporting rate, the real revenue at stake is $1.50 per $100 spent, not the $3.75 of the full-compliance fantasy. Meanwhile the rule’s biggest effect was never revenue — it was behavior. In the report’s own words, the simplest way to avoid the burden is not to spend the asset at all. Congress’s own scorekeeper already reached the same shape of answer — the JCT put the bill at plus $600 million — Cornell just came in higher. What stands in the way: the House’s PARITY Act wants the break for stablecoins only, and Lummis leaves the Senate in January. The cleanest pro-payment fix in the tax code is on her clock.
KEY RECEIPTS
The report: “Fiscal Effects of a De Minimis Exclusion for Personal Digital-Asset Payments,” Cornell Brooks Tech Policy Institute, Sept 4 — a federal revenue analysis of Section 2 of S. 2207.
The bill: personal digital-asset transactions under $300 exempt from capital-gains recognition, $5,000 annual cap on excluded gains; business and investment uses excluded (S. 2207, Lummis).
The score: ~$859 million ten-year NET revenue GAIN. Congress’s Joint Committee on Taxation separately scored the bill at roughly +$600 million over 2025–2034 — Cornell’s estimate lands above the official scorekeeper’s.
Why: only 32–56% of U.S. digital-asset owners report gains. At a 40% reporting rate, the actual revenue at stake is ~$1.50 per $100 of qualifying volume — not the $3.75 full compliance would imply.
The behavioral finding, verbatim: “The simplest way to avoid those burdens is not to spend the asset at all.”
Precedent: U.S. tax law already gives foreign-currency transactions a de minimis exemption for exactly this reason.
The fight: the House PARITY Act draft would limit the break to payment stablecoins; the Bitcoin Policy Institute is leading a coalition to keep Bitcoin in. Lummis leaves the Senate in January 2027 — the window is hers.
Related: The sheriffs stand down on the Clarity Act.
The bill: personal digital-asset transactions under $300 exempt from capital-gains recognition, $5,000 annual cap on excluded gains; business and investment uses excluded (S. 2207, Lummis).
The score: ~$859 million ten-year NET revenue GAIN. Congress’s Joint Committee on Taxation separately scored the bill at roughly +$600 million over 2025–2034 — Cornell’s estimate lands above the official scorekeeper’s.
Why: only 32–56% of U.S. digital-asset owners report gains. At a 40% reporting rate, the actual revenue at stake is ~$1.50 per $100 of qualifying volume — not the $3.75 full compliance would imply.
The behavioral finding, verbatim: “The simplest way to avoid those burdens is not to spend the asset at all.”
Precedent: U.S. tax law already gives foreign-currency transactions a de minimis exemption for exactly this reason.
The fight: the House PARITY Act draft would limit the break to payment stablecoins; the Bitcoin Policy Institute is leading a coalition to keep Bitcoin in. Lummis leaves the Senate in January 2027 — the window is hers.
Related: The sheriffs stand down on the Clarity Act.
They taxed the coffee and collected paperwork.
The cap is still twenty-one million.
The cap is still twenty-one million.
Cornell Brooks Tech Policy Institute report (exec. summary) + CryptoBriefing · Fri Sep 4 2026