The Machine Test

Section 2 of the Common Cents Act (H.R. 3074), passed by the House of Representatives on July 14, 2026, authorizes the Secretary of the Treasury to change the composition of the five-cent coin to a coin with an inner layer of zinc and an outer layer of nickel, subject to testing and evaluation that the composition reduces production cost and, to the greatest extent practicable, has a minimal adverse impact on machines designed to accept coins. This paper examines whether that second condition can currently be evaluated.

To our knowledge, it cannot be evaluated against any public evidence base. No unified inventory of coin validators and their recalibration capability exists in the United States. The machine-population estimates in the federal record diverge by millions of units within single industries. The conversion-cost estimates that dominate the record, 2.4 billion to 10 billion dollars, were solicited by the United States Mint from the industries that would bear the costs, were never independently validated, and were assessed by the Government Accountability Office in 2015 as likely overstated.

Section 2 of H.R. 3074 amends 31 U.S.C. 5112(c) to provide that the five-cent coin may be a coin with an inner layer of zinc and an outer layer of nickel, and that the Secretary may prescribe the composition, subject to testing and evaluation that the composition reduces the cost incurred to produce the coin and, to the greatest extent practicable, has a minimal adverse impact on machines designed to accept coins.

Three features matter analytically. The authority is permissive: nothing compels the change. The test is conjunctive: cost reduction alone does not satisfy the statute. And the machine-impact standard is qualified twice, by "minimal" and by "to the greatest extent practicable." Neither term is defined in the bill.

The standard the statute borrows descends from a 2010 law under which, per the Government Accountability Office, the United States Mint never determined how to quantify either phrase. The agency that would have to apply the test has not established what "minimal" means or where "practicable" ends, and the validation work that exists runs on proprietary methods that cannot be checked from outside the process that produces them.

The federal record contains machine counts and it contains conversion costs, and the two were produced by different processes for different purposes. The counts are self-reported, roughly a decade old, and internally inconsistent by millions of units within single industries. The conversion-cost range of 2.4 billion to 10 billion dollars came from the industries that would bear the cost and was never independently validated; GAO-16-177 judged it likely overstated and found the corresponding savings estimates failed most steps of the government's own cost-estimating guide.

In the one instance we identified in which a government body reported a realized conversion cost after a composition change and compared it to the prior estimate, the United Kingdom after 2012, the reported figure came in approximately 58 percent below the estimate, and even that figure rests on a single Royal Mint memorandum.

Section 2 conditions a coinage change on machine-impact evidence no one collects. Section 5 of the same bill conditions an assessment of rounding harm on transaction evidence no one collects. Statutes that condition action on measured impact require measurement infrastructure. Per-unit evidence infrastructure is buildable. As of this writing, for the machines that accept America's coins, it has not been built.