Americans pay a tax Congress never votes on: the cost of proving they followed the rules. It runs to 6.9 billion hours and more than $530 billion a year — nearly two percent of GDP, spent not paying taxes but filing them — and the New American Accord repeals most of it, by a method no reform commission has proposed: delete the rules that exist only to police other rules, and make the government that writes the remaining obligations carry their complexity itself.
The scale is easiest to see from one ledger. A diner owner hiring her fourth cook pays a payroll service, a year-end W-2 fee, an accountant for quarterly filings, and her own Sunday afternoons — roughly a thousand dollars a year and many hours before a single customer is fed. Multiply her by thirty-three million small businesses and add the average household's thirteen hours and $290 per return, and you have the national figure. None of it is the price of a complex economy. It is the price of a complex code — changed by Congress nine thousand times since 2000, roughly once a day, each change spawning a form, each form an industry.
The Accord's repeal works in two moves. The first deletes complexity at the source: one levy on all compensation replaces the FICA wage caps, the employer-premium administration, and the benefit-exclusion tests that generate most payroll arithmetic; the preference bestiary — the deductions, deferrals, and exclusions that make April a research project — is gone under a single ruled clause, because the floor now delivers openly what the preferences promised covertly. The second move automates what survives: the employer reports who worked and what they earned; a public rail computes, withholds, deposits, files, and issues year-end forms at cost. Whoever writes the rules owes the citizens a cheap way to follow them — a principle conservative in the strictest sense, and self-disciplining besides: a Congress that must carry the burden of its own paperwork writes less of it.
Honesty requires the full accounting, and it has three entries.
First, the code's weight is relocated before it is reduced. The deletions land exactly where today's hours live — wage-cap arithmetic, exclusion tests, preference planning, filing-to-receive. The additions — the VAT's credit-invoice chain, the wealth registry — land on the state itself, on large asset holders, and on business-to-business chains built for offsetting credits. The design intent is explicit: small traders below a published registration threshold stay outside the VAT chain entirely, so the diner gains the payroll simplification without inheriting a bookkeeping regime.
Second, an industry shrinks, and the Accord says so rather than mumbling. Preparers, payroll processors, enrolled agents, and the deferral-optimization bar exist because compliance is hard; making it easy is a managed reallocation the Accord itself causes, and the same bridge it offers every displaced sector applies — wage insurance, Skills Wallet priority — along with a candid market observation: much of the profession redeploys to work clients actually want to buy, advisory and audit and CFO service, once the compelled purchase ends.
Third, Congress loses its favorite instrument: the invisible subsidy. For a century, steering-by-deduction let legislators spend without appearing to. Under the Accord every subsidy is visible spending in a register, priced and scored. That is a feature wearing the costume of a loss — but every constituency told for decades that its deduction was sacred will experience it as the loss first, and the Accord owes them the argument, not a footnote.
The return on all this is the oldest kind: time. Hundreds of billions of dollars of hours convert back into the work the rules exist to protect — the griddle, the payroll met, the fourth cook trained. A government serious about enterprise measures itself by how little it costs an honest person to comply.
Sources: NTUF 2025 (6.93B hours, $477B+); Tax Foundation ($536B, ~1.8% GDP); NTUF filer averages (13 hrs, $290); typical payroll pricing (~$40/mo + $6/employee). VAT small-trader threshold: design intent, parameter pending — confirm status before publication.