New Bill of Rights

Candidate category · Unpolled mandate; regime-coded

Good-government data

Never polled as a mandate; waiting on a mechanism-specific poll clearing a 60% floor. The nearest reading, trust in federal economic data, is 67% of Republicans and 22% of Democrats (Economist/YouGov, 2026), reversed since 2022.

Under consideration since 2026-08-06. A candidate is neither accepted nor rejected — it is waiting on evidence against the full-spectrum test. What we know, what we don't, and what would move it are below.

The proposal

The idea is that the government’s account of itself — what it takes in, what it spends, and what its programs achieve — should rest on the Constitution rather than on statute. Nobody has drafted it. Two shapes are under evaluation: a publication duty — accurate, complete accounts of revenues, spending, population, and program results, on a regular schedule — and an integrity guarantee — official statistics produced by independent agencies, free from political direction as to their substance, conclusions, or timing.1 A third shape, a right of access to records, belongs to the rejected civic-information category.

The evidence

No independent national poll asks whether the Constitution should require accurate, nonpartisan official statistics, or whether statistical-agency heads should be shielded from political removal. We searched for both and found neither, so there is no chart.

The trust readings are fresh and split by party: 40% trust “data reported about the economy by the federal government” — 67% of Republicans and 22% of Democrats, the latter down from 54% in March 2025 (Economist/YouGov, 2026).2 The same pollster measured 53% trust overall in March 2025 and 42% trust against 45% distrust by early August 2025 (Economist/YouGov, 2025; no party breakdown published in either wave).3 In April 2022, under the previous administration, Republicans were more likely to distrust than to trust each of twelve government statistics polled, Democrats more likely to trust each (Economist/YouGov, 2022).4 Each party has held each position.

Dissatisfaction with official information, unlike trust in the numbers, is bipartisan: 76% said information about government expenditures and outcomes is biased — 81% of Republicans, 72% of Democrats, 76% of independents (USAFacts/Harris, 2017; sponsor-commissioned).5 The rest is broad, old, or both: at least 8 in 10 of each party agreed that facts must be “accurate, complete, and transparent” (exact party figures not published) (AP-NORC/USAFacts, 2019);6 53% said open-government data could make officials more accountable — 45% of Republicans and 58% of Democrats (Pew Research Center, 2014);7 and 91% of Republicans and 90% of Democrats favor evidence-based policymaking (Echelon Insights for Results for America, 2026; advocacy-commissioned).8

Congress voted on the statutory version: the Foundations for Evidence-Based Policymaking Act, requiring “credible and accurate” and “objective” official statistics, passed the House 356–17 in 2018 — Republicans 182–17, Democrats 174–0 — and the Senate by unanimous consent.910 A 2025 bill making four statistical-agency heads removable only for cause has a sponsor and five cosponsors, all from one party, and no floor action.1

Where it stands

The bipartisan floor is Unpolled. The nearest data-specific proxy, 45% of Republicans (Pew, 2014), sits at the bottom of the Moderate band and is twelve years old. The tests that apply:

  • Support flips with who holds power. Measured, not feared: 67% of Republicans and 22% of Democrats trust federal economic data now, and the positions were reversed in 2022 (Economist/YouGov, 2022, 2026).24 Any floor measured in one year would be measured against one administration.
  • Depth — untested. No survey has named a cost — enforcement against the executive, agency heads insulated from elected accountability, the expense of collection.
  • Abuse-vs-remedy gap — likely. The abuse half is measured and bipartisan, 81% of Republicans and 72% of Democrats (USAFacts/Harris, 2017);5 49% say federal information is influenced by the president’s political beliefs (AP-NORC/USAFacts, 2024; no party breakdown published).11 The remedy half is unmeasured, and the only remedy bill has single-party cosponsors.1
  • The existential-problem test — open, and deciding. The record since 2025 supplies a new argument: a statistical-agency head was dismissed in August 2025 after a disputed jobs report, at an estimated $20 billion in output lost to the resulting uncertainty;12 in 2019 the Supreme Court found the stated reason for a census question “contrived.”13 Against that: the country has run on statute since 1790 without the counting itself failing.
  • The why-an-amendment test — fails on the present record. The text is not silent: Article I already commands a decennial count and publication of “receipts and expenditures of all public money.”14 Congress can bind itself and did, 356–17, with no member of one party’s caucus voting no and 91% of the other’s voting yes.10 The gap is enforcement: in 1974 the Supreme Court held a taxpayer lacks standing to enforce the publication clause,15 so the duty has no plaintiff — which argues for a standing clause, historically the unpolled or party-coded part of any proposal.

Framings we tested

No framing clears the bar. The two supermajorities — facts must be accurate, complete, and transparent (AP-NORC/USAFacts, 2019);6 evidence-based policymaking should guide government (Echelon for Results for America, 2026, advocacy-commissioned)8 — are statements no one opposes and no constitutional text would change: a different proposal wearing the name. The one framing with a genuinely bipartisan party split, that official information is biased,5 measures the complaint, not the cure. The only mandate-adjacent framing with a party split, open data making officials more accountable, lands at the bottom of Moderate, 45% of Republicans and 58% of Democrats (Pew, 2014).7 The two framings that are the proposal itself have never been asked.

What would move it

In: an independent national poll with party crosstabs asking both mechanisms, forward-looking — a constitutional duty to “publish accurate and complete accounts of what it takes in, what it spends, and the size and condition of the population, on a regular schedule set by law,” and a statistical-agency head “removable only for cause, and never because of the substance, conclusions, or timing of a statistical release,” binding “every future president of either party” — each clearing a 60% floor in the less-supportive party and staying out of the Weak band once the cost is named. A cross-party floor vote on the 2025 bill, or a wide-margin state statistical-integrity amendment in a state of the opposite lean, would corroborate.

Out: those questions returning Weak, or a floor that swaps parties between waves under different administrations, fails regime coding; resolving the existential-problem test as we did for civic information — an improvement in government, not a threat to the country’s future — would fail that test.

Sources

  1. Statistical Agency Integrity and Independence Act of 2025, H.R. 4907, introduced August 2025, would permit removal of the heads of the Census Bureau, BLS, NCES, and BJS only for “proven inefficiency, neglect of duty, or malfeasance” and bar removal based on “the substance, conclusions, or timing” of any statistical release; referred to committee, no floor action; sponsor plus five cosponsors, all Democratic — four original (Aug. 5, 2025) and one added Feb. 10, 2026 — Congress.gov bill page; Congress.gov cosponsor list; The Census Project. ↩ ↩2 ↩3

  2. Trust “data reported about the economy by the federal government”: 40% trust / 52% distrust — 67% R / 22% D, Democrats down from 54% in March 2025; 23% say all or most government statistics are reliable and accurate, down from 36% in March 2025; none reliable: 20% D / 3% R; n=1,073 adult citizens, Feb. 17 – Mar. 1, 2026 — Economist/YouGov. ↩ ↩2

  3. Trust “data reported about the economy by the federal government”: 53% trust (7% completely + 46% somewhat), fielded March 6–8, 2025, n=1,120 adult citizens, ±4, with 36% saying most or all government statistics are reliable — YouGov, “Trust in Government Statistics” topline; 42% trust / 45% distrust, Aug. 1–4, 2025 — Economist/YouGov. ↩

  4. None of twelve government statistics polled was rated trustworthy by a majority; “Democrats are more likely to trust than to distrust each of the statistics” and “Republicans are more likely to distrust than to trust each of the statistics”; n=1,000 adult citizens, April 4–7, 2022 — Economist/YouGov. ↩ ↩2

  5. 76% feel information about government expenditures and outcomes is biased — 81% R / 72% D / 76% I; 90% say data is critical to believing information; 81% say information must come from a nonpartisan source to be seen as factual; fieldwork June 30 – July 5, 2017, online, n=2,521 U.S. general public; commissioned by USAFacts, fieldwork by Harris — USAFacts / Harris, 2017 State of the Facts full report; USAFacts release. ↩ ↩2 ↩3

  6. At least 8 in 10 of each party agree facts must be “accurate, complete, and transparent”; 6 in 10 or more of each say facts must be “based in data”; n=1,032, AmeriSpeak panel, Oct. 15–28, 2019 — AP-NORC / USAFacts, “State of the Facts”. ↩ ↩2

  7. Open-government data could make officials more accountable: 53% — 58% D / 45% R; could help journalists cover government better: 56%; n=3,212, American Trends Panel, Nov. 17 – Dec. 15, 2014 — Pew Research Center. ↩ ↩2

  8. Support for evidence-based policymaking: 91% R / 90% D; n=1,038 likely voters, February 2026; commissioned by Results for America, an evidence-based-policy advocacy organization; fieldwork by Echelon Insights — Results for America. ↩ ↩2

  9. Foundations for Evidence-Based Policymaking Act of 2018, H.R. 4174, Pub. L. 115-435; 44 U.S.C. §3563 requires each statistical agency to “produce and disseminate relevant and timely statistical information,” “conduct credible and accurate statistical activities,” and “conduct objective statistical activities,” with “objective” defined as “accurate, clear, complete, and unbiased” — Congress.gov; govinfo statute compilation. ↩

  10. H.R. 4174 passed the House 356–17 on Dec. 21, 2018, on the motion to suspend the rules and concur in the Senate amendment; per party, Republicans 182–17 and Democrats 174–0, with 36 Republicans and 23 Democrats not voting; the Senate agreed by unanimous consent; signed Jan. 14, 2019 — Clerk of the House, Roll Call 484 XML; Clerk of the House, Roll Call 484; SSA legislative bulletin. ↩ ↩2

  11. 49% say information from the federal government is influenced by the president’s political beliefs; 48% say the same of members of Congress; 10% say policymakers’ decisions are always or often grounded in factual information; n=1,019, July 29 – Aug. 8, 2024 — AP-NORC / USAFacts, “State of the Facts 2024”. ↩

  12. Dismissal of a statistical-agency head in August 2025 following a jobs report the administration disputed; independent analysis estimates the resulting uncertainty shock reduced GDP by roughly $20 billion — CEPR / VoxEU; Cornell Chronicle. ↩

  13. The Court found the stated justification for a census question “contrived” and pretextual — Department of Commerce v. New York, 588 U.S. 752 (2019). ↩

  14. Enumeration Clause: “the actual Enumeration shall be made … within every subsequent term of ten years” (Art. I §2); Statement and Account Clause: “a regular statement and account of receipts and expenditures of all public money shall be published from time to time” (Art. I §9) — U.S. Constitution, Article I, Cornell LII. ↩

  15. A taxpayer challenging the CIA’s statutory exemption from itemized accounting lacked standing; the claim was a “generalized grievance” with no “logical nexus” to taxpayer status — United States v. Richardson, 418 U.S. 166 (1974). ↩

← All candidates