02 — Federal Budget, Debt & Entitlements

STAGE 1 — EVIDENCE BRIEF (Part 1: Facts)

Referee panel: Prof. Olivia Brandt (#57, statistics, C) · Prof. June Takahashi (#17, econometrics, C) · Domain specialist Prof. Beatrice Lang (#13, accounting, C)

Ground rules we applied: every figure below was pulled from the linked primary or near-primary source; the year of the data is stated. Where CBO has already been overtaken by events (the Supreme Court's February 2026 tariff ruling), we give both the February baseline and the later update. "Referee calc" marks our own arithmetic. Today is 2026-09-24.

---

A. The 10-year baseline (CBO, February 2026)

1. Deficit, FY2026: $1.9 trillion, 5.8% of GDP; rising to $3.1T (6.7% of GDP) in 2036. 50-year average deficit: 3.8% of GDP. — CBO, Budget and Economic Outlook: 2026–2036 (Feb 11, 2026). [S1] https://www.cbo.gov/publication/61882
2. Debt held by the public: 99% of GDP at end-2025 → 101% (2026) → 120% (2036); passes the 1946 record (106%) in 2030; 175% of GDP by 2056 in CBO's 30-year extension. [S1; S2 Director's statement] https://www.cbo.gov/publication/62050
3. Revenues vs outlays (% GDP): 2026 revenues 17.5%, outlays 23.3%; 2036 revenues 17.8%, outlays 24.4%. 50-year average revenue 17.3%; CRFB's reading of the same CBO tables gives a 50-year average outlay level of 21.2%. [S1; S13a CRFB] https://www.crfb.org/papers/cbos-february-2026-budget-and-economic-outlook
4. Primary deficit (excluding interest): 2.6% of GDP in 2026, 2.1% in 2036. [S1]
5. Net interest: $1.0T (3.3% of GDP) in 2026 → $2.1T (4.6%) in 2036. 50-year average: 2.1% of GDP. [S2; S3 Exec. Summary] https://www.cbo.gov/system/files/2026-02/61882-Executive-Summary.pdf
6. Composition of outlays, % GDP (2026 → 2036; 50-yr avg): Social Security 5.2 → 5.9 (4.5); Medicare 3.3 → 4.2 (2.2); Medicaid/CHIP/marketplace 2.6 → 2.5 (1.4); other mandatory 3.0 → 2.5 (3.2); defense discretionary 2.8 → 2.4 (4.1); nondefense discretionary 3.1 → 2.4 (3.7); net interest 3.3 → 4.6 (2.1). [S3]

Lang's note: every category that is above its 50-year average is an old-age program or interest. Both discretionary categories are below average and projected to fall further.

1. Revenue composition 2026 (% GDP): individual income 8.6; payroll 5.7; corporate 1.3; customs 1.3 (Feb baseline—see §D); other 0.6. [S3]
2. Growth & rates: real GDP 2.2% (2026), ~1.8%/yr thereafter; 10-yr Treasury 4.1% rising to ~4.3–4.4%. [S2, S3]
3. r vs g: CBO projects the average interest rate on federal debt to exceed nominal GDP growth from FY2031 (both ~3.8% nominal that year); by 2056, 4.2% vs 3.5% nominal. CRFB (Mar 9, 2026). [S17] https://www.crfb.org/blogs/cbo-projects-possible-debt-spiral-r-exceeds-g
4. Fiscal gap: Auerbach & Gale (Brookings, Mar 12, 2026) estimate stabilizing debt requires ~$707 billion/yr in today's dollars (≈27% of current individual income tax revenue); under a current-policy scenario (expiring OBBBA provisions made permanent) debt reaches 211% of GDP by 2056. [S18] https://www.brookings.edu/articles/an-update-on-the-federal-budget-outlook
5. CBO Director Swagel: "the fiscal trajectory is not sustainable." [S2]

B. Actual FY2026 to date (CBO Monthly Budget Review, first 11 months)

1. Deficit Oct 2025–Aug 2026: $2.0T, $6B below the same period in FY2025 (but $82B higher after adjusting for calendar timing shifts). [S4] https://www.cbo.gov/system/files/2026-09/61984-MBR.pdf
2. Net interest on public debt: $1.052T (+12%) vs Department of Defense–military: $833B (+5%). Social Security $1.514T (+5%); Medicare $976B; Medicaid $655B (+8%). [S4] → Net interest now exceeds military spending by ~$219B over 11 months.
3. Receipts $4.845T (+3%); individual income tax +8%; customs duties $167B vs $165B a year earlier (+1%), net of IEEPA refunds (~$110B refunded through August). [S4]
4. CBO in August raised its FY2026 deficit projection to ~$2.1T (from $1.9T), chiefly because customs receipts are running ~$250B below February's projection. [S6, reporting CBO] https://fortune.com/2026/08/10/how-big-deficit-national-debt-tariffs-cbo/
5. FY2025 final: outlays $7.0T (23.1% GDP), revenues $5.2T (17.2%), deficit $1.8T (5.8%); net interest $970B (3.2%). [S44] https://www.cbo.gov/system/files/2026-03/61950-federal-budget.pdf

C. The 2025 reconciliation law (P.L. 119-21, "OBBBA")

1. Conventional score (enacted law): +$3.4T deficits 2025–2034 = $4.5T revenue loss − $1.1T spending cuts. CBO, Jul 21, 2025. [S7] https://www.cbo.gov/publication/61570
2. With debt service: $4.1T (incl. $718B interest); $5.0T if the 10 temporary tax provisions are made permanent; debt in 2034 +9.5 pp of GDP (+11.5 pp if permanent). CBO, Aug 4, 2025. [S8] https://www.cbo.gov/publication/61466
3. In the Feb 2026 baseline CBO attributes +$4.7T in 2026–2035 deficits to the act (incl. interest and macro effects). [S1]
4. Dynamic feedback (House version): GDP +0.5% avg over the decade; higher output raises revenue by only $124B, while higher rates add $441B to interest; dynamic deficit increase with debt service $3.4T (for the House-passed bill). CBO, Jun 17, 2025. [S9] https://www.cbo.gov/publication/61486
5. Distribution (2026–2034 avg): bottom decile resources −~$1,200/yr (−3.1% of income), mainly Medicaid/SNAP; top decile +~$13,600/yr (+2.7%), mainly tax cuts. [S10] https://www.cbo.gov/system/files/2025-08/61367-Distributional-Effects.pdf
6. Coverage: ~10 million more uninsured by 2034; ~$1.06T federal health spending reduction (Medicaid work requirements $326B, provider-tax freeze $191B, state-directed payments $149B). [S11, reporting CBO] https://www.aha.org/news/headline/2025-07-21-cbo-projects-obbba-increase-uninsured-10-million-federal-deficit-34-trillion
7. OBBBA also raised the debt limit by $5T to $41.1T; CRFB expects it to bind mid-to-late 2027. Gross debt ≈ $39T (≈$31T public + ≈$8T intragovernmental), May 2026. [S27] https://www.crfb.org/papers/qa-everything-you-should-know-about-debt-ceiling

D. Tariffs as revenue

1. Feb 2026 baseline: higher tariffs cut deficits $3.0T (2026–2035) — the single largest offset to OBBBA in CBO's accounting. [S1]
2. Feb 20, 2026: Supreme Court (6–3) held IEEPA does not authorize tariffs (Learning Resources / V.O.S. Selections). [S28] https://www.wilmerhale.com/en/insights/client-alerts/20260220-supreme-court-strikes-down-ieepa-tariffs-what-now
3. Replacement: temporary Section 122 surcharge from Feb 24, 2026 (expired July 24 by statute); CIT held it unlawful for three plaintiffs on May 7 (stayed on appeal). Section 301 tariffs (10–12.5%) imposed July 24. [S29, S5] https://www.skadden.com/insights/publications/2026/05/us-trade-court-strikes-down-section-122-tariffs
4. CBO (Aug 20, 2026): tariffs now reduce deficits $0.9T less over 2027–2036 than in February ($0.7T primary + $0.2T interest). $166B collected under IEEPA subject to refund; effective tariff rate now ~10% (vs 15% est. Nov 2025; 2% in 2024). [S5] https://www.cbo.gov/publication/62704
5. Tax Foundation (Sep 10, 2026): tariffs raise $1.4T conventional / $1.0T dynamic over 2026–2035; average household cost $820 in 2026; long-run GDP −0.4%. [S31] https://taxfoundation.org/research/all/federal/trump-tariffs-trade-war/
6. CRFB's day-of-ruling worst case (no replacement tariffs): +$2.4T debt through 2036. Subsequent replacements made that too pessimistic; CBO's +$0.9T is the current official figure. [S30] https://www.crfb.org/blogs/scotus-tariff-ruling-could-add-24-trillion-debt

E. Trust funds (2026 Trustees Reports, June 9, 2026)

1. OASI depletion: 2032 (payable 78%); combined OASDI: 2034 (payable 83%); DI solvent through 75 years. [S12] https://www.ssa.gov/news/en/press/releases/2026-06-09.html
2. OASDI 75-year actuarial deficit: 4.42% of taxable payroll (≈1.5% of GDP), up from 3.82% in 2025. Benefit cut at OASI depletion 22%, growing to 38% by 2100. [S12, S13] https://www.crfb.org/blogs/social-security-and-medicare-trustees-release-2026-reports
3. Why it worsened (+0.60 pp): lower ultimate fertility (1.90→1.75: 0.35 pp), lower assumed immigration (0.18 pp), OBBBA's reduction in income-tax revenue on benefits (0.16 pp); partially offset by productivity (+0.10) and mortality (+0.09). [S15 CRR] https://crr.bc.edu/social-securitys-financial-outlook-the-2026-update-in-perspective/
4. Closing the gap by tax alone: immediate +4.42 pp payroll tax (12.4% → 16.82%). Penn Wharton's independent model: deficit 4.65% of payroll, depletion Feb 2035, fertility 1.6. [S15, S16] https://budgetmodel.wharton.upenn.edu/p/2026-06-11-social-security-outlook-june-2026/
5. Medicare HI depletion: 2033 (unchanged); payable 89% (11% cut, rising to 16% by 2040); HI 75-yr deficit 0.56% of payroll (up from 0.42%); 1.38% under the Chief Actuary's alternative. Medicare total 4.1% of GDP (2026) → 6.5% (2050). [S14] https://www.crfb.org/papers/analysis-2026-medicare-trustees-report

F. Process costs

1. Debt limit: GAO estimated the 2011 impasse raised FY2011 borrowing costs ~$1.3B (not counting multiyear effects). [S24] https://www.gao.gov/products/gao-12-701 GAO's 2026 re-analysis of eight impasses (2011–2023) estimates $107–161M in immediate acute-period costs (2011: $47–57M), with $1.3–5.9T of securities "at risk," and recommends replacing the debt-limit process. [S25] https://files.gao.gov/reports/GAO-26-107872/index.html
2. Shutdown (Oct–Nov 2025, 43 days): CBO estimated a 1.0–2.0 pp hit to Q4-2025 annualized growth and $7–14B of permanently lost GDP (range by duration). [S26] https://www.cbo.gov/publication/61823

G. Options menu (CBO, Options for Reducing the Deficit 2025–2034, Dec 2024) — 10-year savings

1. 5% VAT: $2.18–3.38T; raise payroll tax rate: $1.28–2.54T; raise/modify Social Security taxable maximum: $0.73–1.43T; limit itemized deductions: $0.74–3.42T; carbon tax $0.65–0.92T; raise full retirement age: $95B (10-yr; savings back-loaded); chained CPI for COLAs: $278B; reduce benefits for high earners: $48–197B. [S32] https://www.cbo.gov/publication/60557
2. Largest tax expenditures (Treasury, FY2026): employer health exclusion $296B; imputed rent $157B; DC retirement plans $156B; capital-gains preferences $135B. [S33] https://home.treasury.gov/policy-issues/tax-policy/tax-expenditures

H. History & institutions

1. 1983 Amendments: Greenspan Commission (15 members; 12 endorsed). Package: taxation of up to 50% of benefits above $25k/$32k; 6-month COLA delay; coverage of new federal hires and nonprofits; payroll-tax acceleration. The retirement-age increase to 67 came from Congress (Pickle floor amendment, 228–202), not the Commission's agreement. Long-range deficit cut from 1.80% to ~0.58% of payroll. [S34] https://www.ssa.gov/policy/docs/ssb/v46n7/v46n7p3.pdf
2. Simpson-Bowles (Dec 2010): 11 of 18 votes (14 needed); ~$4T savings 2012–2020 at ~2:1 spending:revenue; Cooper-LaTourette budget based on it failed 38–382 (2012). [S35] https://en.wikipedia.org/wiki/National_Commission_on_Fiscal_Responsibility_and_Reform
3. State balanced-budget rules: only end-of-year, constitutional rules enforced by independently elected courts significantly reduced deficits (47 states, 1970–91); adjustment came via spending cuts, not tax increases. Bohn & Inman (NBER w5533, 1996). [S36] https://www.nber.org/papers/w5533 Poterba (NBER w4375) finds no-deficit-carryover rules and TELs speed adjustment to shocks. [S37] https://www.nber.org/papers/w4375
4. Canada 1995 (Martin budget): program spending cut 18.9% by 1997–98, business subsidies −60%, with revenue measures minor; context was a 1994 interest-rate spike. [S42] https://en.wikipedia.org/wiki/1995_Canadian_federal_budget

(Part 2 — contested literature — follows.)

STAGE 1 — EVIDENCE BRIEF (Part 2: Contested evidence)

Brandt · Takahashi · Lang

C1. Does high debt slow growth? (Threshold vs. no threshold)

• Reinhart & Rogoff (2010), 44 countries: relationship weak below 90% debt/GDP; above 90%, median growth ~1 pp lower. [S38] https://www.nber.org/papers/w15639
• Herndon, Ash & Pollin (2014) found selective exclusion, a spreadsheet coding error, and unconventional weighting; corrected mean growth above 90% is 2.2%, not −0.1% — i.e., a modest negative gradient, no cliff. [S39] https://academic.oup.com/cje/article/38/2/257/1714018
• Takahashi: Both camps agree on a mild negative correlation; neither design identifies causality (low growth also raises debt). Any professor citing "90%" as a cliff will be rated Wrong.

C2. r vs g — how much fiscal space?

• Blanchard (AER 2019): when safe r < g (historically the norm), debt can be rolled over without later tax increases; fiscal cost may be ~zero; welfare cost smaller than assumed, but multiple-equilibria risk real. [S19] https://www.aeaweb.org/articles?id=10.1257%2Faer.109.4.1197
• Furman & Summers (2020): replace debt/GDP with real interest ≤ ~2% of GDP as a guideline; explicitly contingent on low real rates, and assume Social Security/Medicare reforms happen anyway. [S20] https://www.brookings.edu/wp-content/uploads/2020/11/furman-summers-fiscal-reconsideration-discussion-draft.pdf
• Mian, Straub & Sufi (AER): r−g itself rises with debt; the relevant condition is r < g − φ. Calibration: the US had "little space" as of 2019 while Japan had "ample." [S21] https://www.nber.org/papers/w29707
• CBO (Feb 2026): r > g from 2031. [S17]
• Brandt, referee calc: nominal net interest 3.3% of GDP minus (~2.5–3% inflation × ~1.0 debt ratio) → real net interest ≈ 0.5–1% of GDP in 2026 — inside the F–S guideline today; CBO's rate path pushes it up steadily. This is our arithmetic, not a CBO statistic; treat as approximate.

C3. Composition of consolidation: spending vs taxes

• Alesina, Favero & Giavazzi (NBER w18336): spending-based adjustments far less costly in output than tax-based. [S40] https://www.nber.org/papers/w18336
• Romer & Romer (AER 2010): exogenous tax increase of 1% of GDP lowers GDP ~2–3% — but tax increases to cut inherited deficits have noticeably smaller effects. [S41] https://www.nber.org/papers/w13264
• The IMF's Guajardo-Leigh-Pescatori critique of expansionary austerity is relevant but we could not retrieve the primary source today (503); flagged, not relied upon.

C4. "Spending problem or revenue problem?"

The data support a narrower claim than either bloc usually makes: revenue (17.5%) is at/just above its 50-year average (17.3%); outlays (23.3%) are ~2 pp above theirs (21.2%) [S1, S13a]. But (a) that "above average" gap is entirely Social Security, Medicare, Medicaid and interest [S3]; (b) the aging of the population mechanically makes a 50-year average revenue level inadequate to fund programs that both parties have refused to cut; (c) revenue would be ~0.5–1% of GDP higher absent the 2017/2025 tax laws (OBBBA alone: $4.5T over 10 years [S7]). A fair statement: the gap is caused by legislated benefit growth meeting legislated revenue decisions; the choice of which to change is values, not arithmetic.

C5. MMT

• MMT (Kelton, Wray, Mosler): inflation, not solvency, is the binding constraint for a currency issuer; taxes drain demand. IGM expert survey (2019): 0 of 38 agreed that currency issuers "should not worry about deficits"; 22 strongly disagreed, 15 disagreed. [S23] https://www.kentclarkcenter.org/surveys/modern-monetary-theory/
• Galbraith's post-Keynesian critique of F–S: rates are largely a policy choice of the Fed, not a loanable-funds outcome. [S22] https://www.ineteconomics.org/perspectives/blog/reconsideration-of-fiscal-policy-a-comment

C6. Projection uncertainty

• CBO vs Penn Wharton: 75-year Social Security deficit 4.42% vs 4.65% of payroll, driven by fertility (1.75 vs 1.6) and longevity assumptions. [S12, S16]
• Tariff revenue: three official/semi-official estimates moved by trillions within 7 months (CBO Feb: −$3.0T deficit; CRFB Feb 20: +$2.4T debt worst case; CBO Aug: −$0.9T relative to Feb). Any 10-year tariff number is a policy-regime bet, not a forecast.

C7. Debt-ceiling cost measurement

GAO's own two estimates of 2011 differ by more than an order of magnitude ($1.3B FY-wide [S24] vs $47–57M acute-period [S25]) because of window and method. Both are GAO; neither is "the" number.

Referee standing instruction: professors must cite by S-number. We will rate every empirical claim in Stage 4.

STAGE 2 — OPENING POSITIONS (Part 1 of 2: Right bloc)

---

Prof. Douglas Wren (#2, public finance, R) — "It's a spending problem, and the spending is on autopilot"

Causal claim. The deficit is structurally a spending problem. Revenue in 2026 is 17.5% of GDP against a 50-year average of 17.3%; outlays are 23.3% against ~21.2% [S1, S13a]. You cannot look at those two lines and call this a revenue shortfall. Every category above its historical average — Social Security (5.2 vs 4.5), Medicare (3.3 vs 2.2), Medicaid (2.6 vs 1.4), interest (3.3 vs 2.1) — is mandatory and on autopilot [S3].

On OBBBA: yes, CBO scores it at $3.4T conventional [S7]. But the static number overstates the damage: the growth effects of permanent expensing and lower marginal rates recover a large part of the cost over time. And the law contained the largest mandatory-spending reduction in a generation ($1.1T).

Remedy. (1) Gradual FRA increase indexed to longevity for workers now under 50; (2) progressive price indexing of initial benefits for the top half of earners; (3) Medicare premium support pilots; (4) site-neutral hospital payment.

Tradeoff I concede. Benefit cuts for today's 40-year-olds are real losses to people who planned around current law, and the savings are back-loaded — CBO's FRA option saves little in year 1–10.

Falsifiable prediction. If I'm right that it's spending-driven, then holding tax law fixed, CBO's revenue-to-GDP projection will stay within 17–18.5% through 2036 while outlays rise — and every update will show that. If revenue projections fall materially below 17% without new tax cuts, I'm wrong about where the problem lives.

---

Prof. Raymond Kessler (#18, macro, R) — "The interest bill has arrived"

Causal claim. The regime Blanchard described in 2019 is over. Net interest ($1.052T in 11 months) already exceeds military spending ($833B) [S4]. We are already in r > g territory, which is why the debt ratio compounds on its own even with a modest primary deficit. And the empirical literature is clear that once debt passes 90% of GDP you lose roughly a point of growth a year [S38] — we are at 101%.

Evidence. CBO interest path 3.3% → 4.6% of GDP [S2]; CBO dynamic score shows OBBBA raised 10-year rates 14 bp and added $441B of interest on baseline debt [S9]. Crowding out is not hypothetical; CBO put a number on it.

Remedy. A statutory primary-balance target: shrink the primary deficit from 2.6% to ~0% of GDP by 2032, locked in with automatic sequestration if missed.

Tradeoff I concede. A ~2.5% of GDP consolidation over six years is contractionary. Romer & Romer's multipliers are large [S41].

Prediction. Absent consolidation, the 10-year term premium and CBO's projected 10-year yield will drift up, not down, across the next three CBO baselines (Jan 2027–Jan 2029). If the 10-year settles below 3.5% with debt above 110%, my crowding-out story is weak.

---

Prof. Tobias Hartley (#4, monetary econ, Lib) — "Tariffs are taxes, and the Fed is not your backstop"

Causal claim. Two illusions are propping up the fiscal story. First, tariffs are a consumption tax on Americans — Tax Foundation puts it at $820 per household this year [S31] — dressed up as foreigners paying. Second, the Supreme Court's IEEPA ruling wiped out the tariff offset that CBO counted against OBBBA; the "$3 trillion" is gone [S1, S28]. What remains is an OBBBA-sized hole.

Third: once r > g and debt >100%, the Fed faces pressure to hold rates down for the Treasury — fiscal dominance. The inflation of 2021–23 showed what happens when fiscal and monetary expansion align.

Remedy. Spending caps (nominal growth ≤ population + inflation) on all non-interest outlays including mandatory; tariffs only by explicit act of Congress; no "revenue" credit for executive tariffs in scoring.

Tradeoff. Caps binding on mandatory programs mean real per-beneficiary cuts in Medicare as the population ages. I accept that; many here won't.

Prediction. If fiscal dominance is a live risk, we'll see 5y5y inflation breakevens drift above 2.75% on a sustained basis within 3 years as deficits stay >6% of GDP. If breakevens stay anchored near 2.2–2.5%, I'm overstating the risk.

---

Prof. Benjamin Ashford (#31, federalism, R) — "49 states already do it"

Causal claim. Institutions, not individual budgets, determine outcomes. Strict state balanced-budget rules reduce deficits — Bohn & Inman found end-of-year constitutional rules enforced by elected courts work, and adjustment comes through spending restraint, not tax increases [S36]. Poterba shows these rules speed adjustment to shocks [S37]. A federal BBA would work the same way — Congress would simply have to do what governors do each spring.

Remedy. A federal balanced-budget amendment with a 3/5 supermajority override and an automatic waiver for declared recessions or war.

Tradeoff. Procyclicality: a BBA forces tightening in downturns unless the escape clause is used — and escape clauses get used.

Prediction. If BBAs discipline spending, states with strict (end-of-year, constitutional) rules should show smaller structural deficits and fuller rainy-day funds over the next cycle than states with weak rules. If they show no difference, my case collapses.

STAGE 2 — OPENING POSITIONS (Part 2 of 2: Left and Center)

---

Prof. Priya Venkataraman (#19, inequality econ, L) — "We chose this deficit, in July 2025"

Causal claim. The debt path is a revenue choice. OBBBA cost $4.1T with interest, $5.0T if made permanent [S8]; CBO's distribution tables show the bottom decile loses ~$1,200/yr while the top decile gains ~$13,600/yr [S10], and ~10 million people lose coverage [S11]. Tax cuts are the entire reason the debt ratio is rising — without 2001/2017/2025, we would not be discussing a crisis. Even the Social Security gap widened 0.16 pp of payroll because OBBBA cut taxation of benefits [S15].

Remedy. (1) Let OBBBA's temporary provisions expire; (2) eliminate the Social Security taxable maximum, which solves Social Security's shortfall; (3) limit itemized deductions to 28% value; (4) minimum tax on billionaires' unrealized gains.

Tradeoff. Top marginal rates would rise to levels not seen since the 1970s for high-wage earners; some avoidance response is certain.

Prediction. If this is a revenue problem, then revenue-raising packages (e.g., allowing OBBBA temporaries to lapse in 2028–29) will reduce CBO's projected 2036 deficit by at least 1% of GDP without measurable growth loss in the subsequent two CBO baselines.

---

Prof. Rosa Delgado-Finch (#7, political economy, Soc) — "Solvency is a political fiction for a currency issuer"

Causal claim. The US cannot "run out" of dollars; the binding constraint is real resources and inflation, not the debt ratio. Japan has carried debt far above ours at a trivial interest cost [S20]. Interest rates are a policy variable set by the Fed [S22], so the "interest bill" is itself a choice. Interest payments are income to American households and pension funds; they're a distributional question, not a solvency one. And mainstream economics is moving toward this view since 2019.

Remedy. Fund Social Security from general revenue when the trust fund depletes (a one-line statutory fix); a federal job guarantee as automatic stabilizer; taxes designed for inflation control and inequality reduction (wealth tax), not "paying for" programs.

Tradeoff. If the Fed does not cooperate, or real capacity binds, the inflation cost is real — 2021–23 was a warning, and I won't pretend it wasn't partly fiscal.

Prediction. If MMT's core is right, then in the next 3 years, with deficits >6% of GDP, the long-run (5y5y) inflation expectation stays anchored below 2.75% and there is no failed Treasury auction. That's the same test Hartley proposed; we disagree on the outcome.

---

Prof. Miriam Castellano (#1, labor econ, C-L) — "Furman-Summers, not panic — but the window is closing"

Causal claim. Debt/GDP is the wrong gauge; the right one is real debt service. Real net interest is currently below 2% of GDP (referee estimate ~0.5–1%) [C2], so this is not a crisis. But CBO has r > g from 2031 [S17], so the window for calm, phased reform is ~5 years. The trust-fund cliff in 2032–34 is the real forcing event, not the debt level.

Remedy. A Social Security package modeled on 1983: the 1983 deal split roughly evenly between revenues and benefit changes, and so should this one. Raise the taxable max to cover 90% of earnings; add a modest rate increase phased over 20 years; reduce benefit growth for the top 30% of earners; raise the minimum benefit.

Tradeoff. Payroll taxes are regressive at the bottom; the minimum benefit only partly offsets.

Prediction. If real interest stays <2% of GDP through 2030 (CBO projections), no fiscal crisis indicators (failed auctions, CDS spikes >100bp) will appear before 2030 even without consolidation.

---

Prof. Gregory Stahl (#29, congressional studies, C) — "The problem is procedural"

Causal claim. Congress knows the options (CBO publishes 76 of them [S32]); it lacks a mechanism to take collective blame. Successful precedent: the Greenspan Commission, which raised the retirement age to 67 and restored solvency for decades [S34]. Failed precedent: Simpson-Bowles (11 of 18, then 38–382 on the floor) [S35] — because it had no guaranteed vote. Meanwhile process failures cost money: the 2011 debt-limit fight cost $1.3B [S24]; the 2025 shutdown $7–14B in permanently lost GDP [S26].

Remedy. (1) Fiscal commission with guaranteed up-or-down floor vote; (2) auto-CR to abolish shutdowns; (3) tie the debt limit to enacted budgets (Gephardt rule) per GAO [S25].

Tradeoff. Fast-track reduces deliberation and minority leverage.

Prediction. If process matters, a guaranteed-vote commission will produce an enacted package where Simpson-Bowles failed — testable if one is created before the 2032 OASI cliff.

---

Prof. Nadia Haddad (#15, trade, C) — "Tariffs are neither a savior nor a nothing"

Causal claim. Customs receipts through August were $167B vs $165B a year earlier [S4] — flat, net of ~$110B in IEEPA refunds. CBO now counts tariffs as reducing deficits $0.9T less than in February [S5]. At a ~10% effective rate, tariffs cannot sustainably raise more than about 1% of GDP, because the base shrinks as rates rise. Tariffs also cost output: −0.4% long-run GDP [S31].

Remedy. If Congress wants tariff revenue, legislate it (e.g., a uniform, WTO-negotiated border rate) so scoring and business planning have a stable base; otherwise treat executive tariffs as zero in baselines.

Tradeoff. Legislated tariffs are more durable — which is good for revenue and bad for anyone who thinks tariffs are welfare-reducing.

Prediction. Net customs receipts in FY2027 will land between 0.6% and 1.0% of GDP. Outside that range I'll revise.

STAGE 3 — CROSS-EXAMINATION (Part 1)

---

Exchange 1: Castellano (C-L) ⟷ Kessler (R) — r vs g (4 rounds)

Castellano → Kessler (R1): Raymond, "we are already in r > g territory" is not what CBO says. CBO has the average rate on debt crossing nominal growth in 2031, at ~3.8% each [S17]. Today the effective rate on the stock is below nominal growth because a large share of the debt was issued at 2020–21 coupons. And your "90% costs a point of growth" line is the Reinhart-Rogoff figure that Herndon-Ash-Pollin corrected: mean growth above 90% was 2.2%, not −0.1% [S39]. There's no cliff at 90%. You've got a real argument about the trajectory; stop dressing it up with a threshold that doesn't exist.

Kessler (R1 reply): On the date, fine — I'll accept CBO's 2031 and amend "already" to "within five years." That doesn't help you: your whole Furman-Summers comfort rests on low real rates, and F–S themselves wrote that if real rates rise to ~2%, the debt ratio "should be lower" [S20]. The 10-year is at 4.1% with inflation near target. On 90%: HAP killed the cliff, agreed, but HAP still shows lower growth at high debt. And Mian-Straub-Sufi — hardly austerians — calibrate the US as having "little space" in 2019, when debt was ~20 points lower [S21].

Castellano (R2): MSS is a fair hit, and I'll take it seriously: their key point is that r−g rises with debt, so Blanchard's r<g isn't a free lunch at the margin. But note what MSS didn't say: that the US faces a crisis. "Little space" means new permanent deficits are costly — which is an argument against OBBBA, frankly, not for Kessler's 2.5%-of-GDP six-year squeeze. Romer & Romer find deficit-motivated tax increases have smaller output costs [S41]; a phased package is what the evidence supports.

Kessler (R2): A phased package that starts after the next election is how we got here. Canada in 1995 didn't phase: program spending fell 18.9% in two years, after a rate spike forced their hand [S42]. I'd rather do it before the market forces it.

Castellano (R3): Canada is a good example of the opposite of your mix claim for us: they cut a discretionary-heavy federal budget and pushed costs to provinces via the CHST [S42]. Our discretionary spending is already below its 50-year average (defense 2.8 vs 4.1; nondefense 3.1 vs 3.7) [S3]. There's nothing comparable to cut quickly. Our cost growth is in pensions and health for retirees, which must be phased for legitimacy.

Kessler (R3): That's the first argument that moves me: I accept that a Canadian-style front-loaded cut isn't available on our budget composition. I'll revise my remedy from "~0 primary balance by 2032" to "primary deficit ≤1% of GDP by 2032, with Social Security/Medicare changes enacted now even if phased." I do not concede the level of urgency — interest now exceeds military outlays [S4], and that's a fact about today, not 2031.

Castellano (R4): Accepted on the revision. On urgency we still differ; my prediction (no crisis indicators before 2030) and yours (rising 10-year yields) are both on the record.

---

Exchange 2: Venkataraman (L) → Wren (R)

Venkataraman: Douglas, your own dynamic evidence cuts against you. CBO's dynamic score of H.R. 1: faster growth adds $124B of revenue — but higher rates add $441B of interest [S9]. The "growth recovers a large part of the cost" claim is not what the official dynamic score found: the dynamic cost was higher than conventional. Second: "revenue is at its average" ignores that the population over 65 is not at its average. You are comparing a 2026 obligation set to a 1976–2025 revenue average.

Wren: On the dynamic score: I'll accept that CBO's dynamic analysis of the House bill shows net feedback worsening the deficit, and I withdraw "large part of the cost." My view is that CBO under-models the long-run expensing effect beyond 10 years — but that's a view, not evidence I can cite. On demographics: yes, aging raises costs. But the choice to fund a growing retiree share entirely from general revenue and borrowing is the spending choice I'm pointing to. Social Security benefits are wage-indexed at initial claim; that's a design choice that grows real benefits generation over generation.

Venkataraman: Then say "the problem is that benefits grow with wages and we won't tax to pay for it," which is a both statement.

Wren: I'll accept "it is a both statement at the level of arithmetic, and a spending statement at the level of what's new relative to history." Not more.

---

Exchange 3: Wren (R) → Venkataraman (L)

Wren: Priya, "tax cuts are the entire reason the debt ratio is rising." CBO's table: Social Security +0.7 pp of GDP over the decade, Medicare +0.9, interest +1.3 [S3]. Revenues actually rise from 17.5% to 17.8%. Even with OBBBA fully repealed, CBO would show a rising debt ratio because the primary deficit is 2.6% of GDP and r approaches g. And "eliminating the cap solves Social Security"? CBO's option scores eliminating/raising the cap at $0.73–1.43T over ten years [S32]. The 75-year gap is 4.42% of payroll — about 1.5% of GDP [S13]. Show me that closes it.

Venkataraman: On "entire": I'll amend to "tax legislation since 2001 is the largest single legislated contributor; aging-driven spending growth is the other." I'm not retreating on the distributional point: the one piece of discretionary fiscal policy in the last year widened the deficit by $4T while taking from the bottom decile [S10]. On the cap: I'll wait for the referees; my recollection is that the Chief Actuary's scoring closes most of the gap.

Lang (referee specialist, procedural note): The Chief Actuary's 2026-basis payroll-tax provisions page is not yet updated (SSA says "working on updating") — the referees could not verify any current share. We'll rate accordingly.

---

Exchange 4: Hartley (Lib) → Delgado-Finch (Soc)

Hartley: Rosa, "mainstream economics is moving toward MMT" — the only systematic evidence is the IGM panel: zero of 38 agreed that currency issuers needn't worry about deficits; 22 strongly disagreed [S23]. And "interest rates are set by the Fed": the Fed sets the overnight rate. The 10-year is at 4.1% [S3] and CBO's dynamic score showed deficits pushing it up 14bp [S9]. If rates were purely policy, that estimate would be zero.

Delgado-Finch: The IGM survey is from 2019, before the pandemic deficits failed to produce the bond crisis the panel implied. But I accept I have no newer survey; I withdraw "moving toward" and restate: "MMT remains a minority view; my claim is that its predictions have fared better than the 2010-era austerity consensus." On the 10-year: long rates are expected future short rates plus a term premium; Galbraith's point is that the first component is policy [S22]. The 14bp is a term-premium estimate from a model — small.

Hartley: 14bp on $30T is $42B a year. "Small" is doing a lot of work. And 2021–23 is precisely the test: fiscal expansion met capacity, and we got 9% inflation.

Delgado-Finch: I conceded in my opening that 2021–23 was partly fiscal. That's consistent with MMT — the constraint was real resources, exactly what we say. Where we disagree is whether the answer is a debt target or an inflation target.

---

Exchange 5: Haddad (C) → Hartley (Lib)

Haddad: Tobias, "the SCOTUS ruling wiped out the tariff offset" is wrong on CBO's own numbers. CBO's August update: tariffs now reduce deficits by $0.9T less over 2027–36 than in February — not $3T less [S5]. Section 301 tariffs replaced much of the IEEPA base. Effective rate ~10%, not 2% [S5]. You and I agree tariffs are a tax paid largely domestically; but the fiscal hole is smaller than you claimed.

Hartley: Fair on the arithmetic: I retract "wiped out" and substitute "cut by roughly 30% relative to February, and now resting on authorities (Section 301) that are also being litigated." My policy point stands: none of that revenue was voted by Congress.

Haddad: On that, we agree entirely.

STAGE 3 — CROSS-EXAMINATION (Part 2)

---

Exchange 6: Kerr (#30, public policy, C-L) → Ashford (R)

Kerr: Benjamin, Bohn & Inman is your best evidence and it's narrower than you use it. Only end-of-year, constitutional rules enforced by independently elected courts worked; prospective rules, statutory rules, and rules enforced by appointed courts showed no significant effect [S36]. Most federal BBA drafts are prospective (Congress adopts a balanced plan) with no clear judicial enforcement. That's the category that failed in the states. Second, the states' mechanism is that they save via rainy-day funds and — something your analogy omits — they can lean on a federal government that is not bound by a BBA in a recession. "A federal BBA would work the same way" is an unsupported extrapolation.

Ashford: The design point is well taken, and I'll amend: the evidence supports strict, ex-post, judicially enforced rules, and I'd write the amendment that way. On the backstop: yes, states can run procyclical budgets partly because Washington doesn't. I concede there's no direct evidence a federal BBA would reproduce state results; my claim is that the institutional logic (make deficits require a supermajority) would bind. That's a prediction, not a finding.

Kerr: And the prediction has a precedent in reverse: the rule that binds would make recessions deeper unless the 3/5 override is routine — at which point it doesn't bind.

Ashford: Then we're arguing about the override threshold, which is progress.

---

Exchange 7: Morrow (#5, economic history, L) → Stahl (C)

Morrow: Gregory, you've credited the Greenspan Commission with raising the retirement age. The SSA Bulletin's legislative history is explicit: the retirement-age increase came from Rep. Pickle's floor amendment (228–202); the Commission itself had split on long-range solutions [S34]. That matters to your argument: the commission's consensus package was the short-run revenue-and-COLA deal; the hardest benefit cut was done by ordinary majoritarian politics on the House floor.

Stahl: You're right and I'm glad it's on the record; I retract. I'd argue it cuts both ways: the commission created the vehicle and the political cover that let a 228–202 floor vote happen. But the specific claim was wrong.

Morrow: And your GAO $1.3B for 2011 — GAO's 2026 report estimates $47–57M for 2011 in immediate costs [S25].

Stahl: Both are GAO. The $1.3B is FY-wide, the 2026 figure is acute-window only [S24, S25]; the referees already flagged it (C7). I'll cite both. Either number is pure waste.

---

Exchange 8: Kessler (R) → Haddad (C)

Kessler: Nadia, "tariffs can't sustainably raise more than ~1% of GDP" — CBO's own February baseline had customs at 1.3% of GDP in 2026 [S3], at a ~15% effective rate. You're asserting a revenue ceiling as though it were a law of nature.

Haddad: February's 1.3% was pre-ruling and assumed no base erosion over time. At today's ~10% ETR [S5] the static arithmetic is: goods imports ≈ 11% of GDP × 10% ≈ 1.1%, before any import compression. So "about 1%" is my estimate, not CBO's; I'll relabel it as an estimate and keep the prediction range (0.6–1.0% in FY2027).

Kessler: Fine — as long as it's labeled.

---

Exchange 9: Delgado-Finch (Soc) → Kessler (R) — bond vigilantes

Delgado-Finch: Raymond, CBO projects the 10-year at 4.1% rising only to ~4.4% through 2036 [S3] even with debt going to 120%. Where are the vigilantes?

Kessler: In the level. 4.4% nominal with ~2% inflation is a real rate of ~2.4% — above projected real growth of ~1.8% [S3]. That is the r>g problem; it doesn't need a spike to compound.

Delgado-Finch: That's a coherent answer and better than "crisis next year." I'd note CBO's projected rates embed CBO's view of Fed policy — the policy variable I keep pointing at.

---

Exchange 10: Klein (#59, gerontology, C) → Wren (R) — retirement age

Klein: Douglas, FRA-to-longevity sounds neutral but CBO's FRA option saves $95B over 10 years [S32] against a 75-year gap of 4.42% of payroll. It's a long-run tool, fine — but longevity gains have been concentrated among higher earners, so a uniform FRA increase is a larger proportional cut for low earners, who claim earlier and die sooner.

Wren: The $95B is right and I conceded back-loading. On the longevity gradient: can you cite it?

Klein: Not from this brief — the referees didn't pull the NAS or SSA studies, so I'll label it as widely reported but unverified here. It's the reason I'd pair any FRA increase with an enhanced minimum benefit and a disability-pathway fix.

Wren: Pairing with a minimum benefit I can support.

---

Exchange 11: Lang (C, referee specialist) → Castellano (C-L)

Lang: Miriam, "the 1983 deal split roughly evenly between revenues and benefits." The SSA Bulletin gives the total effect (long-range deficit 1.80% → ~0.58% of payroll) [S34] but the referees have no verified revenue/benefit split in the sources pulled. Do you have one?

Castellano: Not with a link in this thread. I'll restate as: "the 1983 deal combined substantial revenue measures (benefit taxation, coverage expansion, tax acceleration) with benefit measures (COLA delay, later FRA)." The proportion I withdraw pending a source.

STAGE 4 — REFEREE FACT-CHECK

Brandt · Takahashi · Lang. Every empirical claim from Stages 2–3. Ratings: Supported / Contested / Unsupported / Wrong. We planted nothing; where a claim is plausible but not verified from the sources pulled for this thread, it is rated Unsupported, not Wrong.

# · Speaker · Claim · Rating · Basis
1 · Wren · Revenue 17.5% vs 17.3% avg; outlays 23.3% vs ~21.2% · Supported · S1, S13a
2 · Wren · Every category above its 50-yr avg is SS/Medicare/Medicaid/interest · Supported · S3
3 · Wren · "Structurally a spending problem" · Contested · C4: aging vs revenue choices; S7
4 · Wren · OBBBA growth effects "recover a large part of the cost" · Wrong · S9: +$124B revenue vs +$441B interest; dynamic cost > conventional
5 · Wren · $1.1T cut = "largest mandatory reduction in a generation" · Unsupported · $1.1T is S7; superlative not verified
6 · Kessler · Net interest > military outlays (FY26 YTD $1.052T vs $833B) · Supported · S4
7 · Kessler · "Already in r > g territory" · Wrong · S17: CBO crossing in FY2031
8 · Kessler · Debt >90% costs ~1 pt of growth/yr · Wrong · S39 corrects S38; no threshold; mild gradient
9 · Kessler · OBBBA (House ver.) +14bp 10-yr, +$441B interest · Supported · S9 (note: House-passed version)
10 · Hartley · Tariffs cost ~$820/household (2026) · Supported · S31 (single-model estimate)
11 · Hartley · SCOTUS ruling "wiped out" the $3T tariff offset · Wrong · S5: $0.9T smaller, not $3T
12 · Hartley · Fiscal-dominance risk; 2021–23 partly fiscal · Contested · theory; no source pulled either way
13 · Ashford · Strict state BBAs reduce deficits, via spending · Supported · S36
14 · Ashford · Rules speed adjustment to shocks · Supported · S37
15 · Ashford · A federal BBA "would work the same way" · Unsupported · S36 limits to strict ex-post rules; no federal evidence
16 · Venkataraman · OBBBA $4.1T w/ interest ($5.0T permanent); decile −$1,200 / +$13,600; ~10M uninsured · Supported · S8, S10, S11
17 · Venkataraman · OBBBA widened SS gap by 0.16 pp of payroll · Supported · S15
18 · Venkataraman · Tax cuts are "the entire reason" debt ratio rises · Wrong · S3: SS +0.7, Medicare +0.9, interest +1.3 pp of GDP; primary deficit exists on old-age cost growth
19 · Venkataraman · Eliminating taxable max "solves" Social Security · Unsupported · Chief Actuary 2026-basis scores not published; S32 high end ($1.43T/10 yr ≈ 0.4% GDP/yr) vs gap ≈ 1.5% GDP (S13) implies far short of full closure in the 10-yr window (referee calc)
20 · Delgado-Finch · Japan: very high debt, trivial interest cost · Supported (dated) · S20: 177% debt, 0.2% GDP interest (2020 datum)
21 · Delgado-Finch · Interest rates are a policy variable set by the Fed · Contested · S22 vs S9, S21 (short rates yes; long rates partly market)
22 · Delgado-Finch · Interest is income to American households · Unsupported · partly true; foreign-holder share not pulled
23 · Delgado-Finch · Mainstream moving toward MMT · Unsupported · S23 (0/38 agree, 2019); no newer survey
24 · Castellano · Real net interest < 2% of GDP today · Supported · C2 referee calc ≈0.5–1% (approximate)
25 · Castellano · CBO: r > g from 2031 · Supported · S17
26 · Castellano · 1983 deal split "roughly evenly" rev/benefits · Unsupported · S34 gives total effect only
27 · Stahl · Greenspan Commission raised the FRA to 67 · Wrong · S34: Pickle floor amendment, 228–202
28 · Stahl · 2011 debt-limit fight cost $1.3B · Contested · S24 ($1.3B FY-wide) vs S25 ($47–57M acute) — both GAO
29 · Stahl · Simpson-Bowles 11/18; 38–382 · Supported · S35
30 · Stahl · 2025 shutdown $7–14B permanent GDP loss · Supported · S26 (CBO scenario range)
31 · Stahl · CBO lists 76 options · Supported · S32
32 · Haddad · Customs $167B vs $165B YTD; ~$110B refunds · Supported · S4
33 · Haddad · Tariffs now cut deficits $0.9T less than in Feb · Supported · S5
34 · Haddad · Tariffs can't sustainably raise > ~1% GDP · Contested · S3 Feb baseline had 1.3% (pre-ruling); estimate
35 · Haddad · Tariffs −0.4% long-run GDP · Supported · S31 (model estimate)
36 · Castellano · HAP: >90% mean growth 2.2%, not −0.1% · Supported · S39
37 · Kessler · MSS: US had "little space" in 2019 · Supported · S21
38 · Kessler · F–S: higher real rates → debt ratio should be lower · Supported · S20
39 · Castellano · Deficit-motivated tax hikes have smaller output effects · Supported · S41
40 · Kessler · Canada: program spending "fell 18.9% in two years" · Contested · S42 gives an 18.9% planned reduction by 1997–98; realized outcome not verified
41 · Castellano · Discretionary categories below 50-yr avg · Supported · S3
42 · Wren · SS +0.7, Medicare +0.9, interest +1.3 pp; revenue 17.5→17.8 · Supported · S3, S1
43 · Wren · Cap option $0.73–1.43T; gap ≈1.5% GDP · Supported · S32, S13
44 · Hartley · IGM: 0 of 38 agreed · Supported · S23
45 · Hartley · 14bp ≈ $42B/yr · Supported · consistent with S9 ($441B/10 yr)
46 · Kessler · Real 10-yr ~2.4% > real growth 1.8% ⇒ "that is r>g" · Contested · marginal rate vs average rate on stock; S17
47 · Klein · FRA option saves $95B over 10 yrs · Supported · S32
48 · Klein · Longevity gains concentrated among high earners · Unsupported · not sourced here (self-flagged)
49 · Kerr · Only strict ex-post constitutional rules worked · Supported · S36
50 · Morrow · Pickle amendment 228–202 · Supported · S34
51 · Haddad · Goods imports ≈11% of GDP · Unsupported · not sourced in brief

Tally (51 claims): Supported 30 · Contested 7 · Unsupported 8 · Wrong 6.

Takahashi's summary: the Right's errors clustered on growth feedback and thresholds (#4, #7, #8, #11); the Left's on sufficiency of revenue fixes and a monocausal story (#18, #19, #23); the Center's on history (#27) and measurement (#28). No bloc was error-free.

---

Retractions and amendments on the record

• Wren (#4): "I withdraw 'recovers a large part of the cost.' CBO's dynamic score shows net feedback raised the deficit. I retain the view that 10-year windows understate expensing, without citation." (#5): "Drop 'largest in a generation'; the $1.1T stands."
• Kessler (#7): "Amend to 'r crosses g in 2031 per CBO.'" (#8): "Withdrawn as a threshold; I claim only a modest negative gradient, which HAP also finds." (#40): "Amend to 'planned 18.9% program-spending reduction.'" (#46): "Accept: I compared a marginal rate to growth; the stock's average rate crosses later."
• Hartley (#11): "Retracted; replaced with '≈30% smaller offset, resting on litigated authorities.'"
• Venkataraman (#18): "Amended: 'largest single legislated contributor.'" (#19): "Withdrawn as stated. Restated: 'eliminating the cap closes a large share, magnitude pending a 2026-basis Chief Actuary score.' And I'll say plainly: the referee arithmetic suggests it cannot close the gap alone."
• Delgado-Finch (#23): "Withdrawn." (#22): "Amended to 'largely domestic income'; I'll bring the TIC holdings data next round."
• Castellano (#26): "Proportion withdrawn."
• Ashford (#15): "Restated as a prediction, not a finding."
• Stahl (#27): "Retracted. The Commission provided the vehicle; the House floor raised the age." (#28): "Will cite both GAO figures."
• Haddad (#34, #51): "Relabeled as my estimate; import share to be sourced."
• Klein (#48): "Stands as unverified here; I'll bring the NAS study."

STAGE 5 — STEELMAN

Each bloc writes the strongest version of an opposing case; the owners reply.

---

Left (by Castellano, for the bloc) steelmans the Right

"The Right's best case is not 'taxes are theft' or '90% is a cliff.' It's this: the growth of the federal budget since the 1970s is almost entirely the growth of three open-ended entitlement promises whose real value rises automatically with wages and medical prices [S3]. Revenue has been roughly stable around 17% of GDP for 50 years across very different tax codes, which suggests a political-economy ceiling — Americans have repeatedly refused to sustain higher federal taxation. If that ceiling is real, the only durable fix is to slow benefit growth, and the earlier that starts the gentler it is: 1983 phased the FRA over 44 years [S34]. Meanwhile interest has overtaken defense [S4], and every dollar of interest is a dollar not available for anything the Left values. Spending-based consolidations have historically had smaller output costs [S40]. The humane conservative position is: reform the old-age programs now, protect the poorest retirees, and stop pretending the rich alone can fund a 6%-of-GDP-and-rising pension system."

Right reply (Wren): Accepted as fair — better than some of our own speeches. One correction: we don't concede a ceiling on revenue as an empirical law; we argue it as a revealed preference.

---

Right (by Kessler, for the bloc) steelmans the Left

"The Left's best case: the one lever Congress actually pulled in 2025 was a $4.5T revenue cut [S7] — so talk of 'spending on autopilot' ignores that revenue was put on autopilot downward, three times in 25 years. Revenue sits at its historical average while the population over 65 is at a historical high; holding revenue flat is itself a policy choice to shrink government relative to its commitments. The distributional record is damning: the bottom decile loses ~3% of income, the top decile gains ~2.7% [S10], and the Social Security gap widened partly because of the same law [S15]. Deficit-motivated tax increases have smaller growth costs than the headline multipliers suggest [S41], and base-broadening (tax expenditures of ~$300B for employer health alone [S33]) raises revenue without raising rates. You can't credibly ask a 45-year-old janitor to work to 69 while cutting taxes for the top decile in the same Congress."

Left reply (Venkataraman): Accepted as fair. We'd add: the Left's serious wing accepts benefit-side changes for high earners; the steelman shouldn't imply we won't touch benefits at all.

---

Center (by Stahl) steelmans the socialist/MMT position

"Delgado-Finch's strongest case: the US issues the world's reserve asset in its own currency and cannot involuntarily default. The binding constraint is real capacity and inflation, and the debate over 'debt ratios' distracts from the question that actually matters — what does the spending buy and does it overheat the economy? Japan ran debt far above ours at negligible interest cost for decades [S20]. Trust-fund 'depletion' is an accounting convention: Congress could legislate general-revenue transfers tomorrow, as it already does for Medicare Parts B and D. The 2010s austerity consensus was built partly on an error-ridden spreadsheet [S39]."

Socialist reply (Delgado-Finch): Accepted with one correction: MMT does not say deficits never matter — it says they matter through inflation, which is why the 2021–23 episode is a test we half-failed, and I said so.

---

Right-libertarian (by Hartley) steelmans the Center's institutional case

"Stahl's best case: nobody lacks information — CBO lists 76 options [S32]. The failure is a collective-action problem: each party gains by blaming the other for pain. Commissions work when they pre-commit Congress to a vote (1983 enacted; Simpson-Bowles lacked a guaranteed vote and died 38–382 [S34, S35]). Process weapons — debt limits, shutdowns — impose real costs ($7–14B for one shutdown [S26]) without ever having produced lasting consolidation."

Center reply (Stahl): Accepted as fair. We'd add that the 1983 success was forced by an imminent (months away) cliff — which suggests the 2032 OASI date may be the next genuine window.

---

Center (by Haddad) steelmans the nationalist-Right tariff case (no nationalist-Right speaker opened, so we state it)

"Tariffs are a revenue source that also carries strategic value; a ~10% effective rate raises roughly 1% of GDP [S5] — comparable to the corporate income tax (1.3% [S3]) — and taxes foreign producers' rents to the extent they absorb some of it. If they're going to exist, count them."

Right reply (Wren, for the bloc's tariff-sympathetic members): Accepted as fair; Hartley dissents on the merits but agrees it's accurately stated.

STAGE 6 — REVISED PROPOSALS

Drafted jointly by Castellano, Kessler, Stahl, Wren, Venkataraman, Haddad and Ashford after cross-examination. Scores are cited or marked "no official score."

---

P1. Social Security solvency package ("1983 II"), enacted by 2029

Mechanism. (a) Raise the taxable maximum over 10 years to cover 90% of covered earnings, with benefit credit; (b) index the FRA to cohort life expectancy beyond 67, starting for those born 1975+, paired with (c) an enhanced minimum benefit (≥125% of poverty after 30 years) and (d) progressive price indexing of initial benefits for the top ~30% of earners.
Scoring. No official score for the package. Components: CBO cap options $0.73–1.43T/10 yrs; FRA $95B/10 yrs (back-loaded); high-earner benefit reductions $48–197B [S32]. Target: close the 4.42%-of-payroll gap [S12] ~half revenue, ~half benefits, subject to a Chief Actuary score.
Precedent. 1983 Amendments cut the long-range deficit from 1.80% to ~0.58% of payroll [S34].
Risk. Needs a forcing event; FRA indexing is a larger proportional cut for groups with lower longevity (unverified gradient, #48).

P2. Replace the debt limit with automatic authorization tied to enacted budgets

Mechanism. Borrowing authority deemed approved by any enacted law that changes spending or revenue (a Gephardt-style rule), as GAO recommends [S25].
Scoring. No budgetary score; avoids impasse costs estimated at $1.3B (2011, FY-wide) [S24] or $107–161M acute across 8 episodes [S25].
Precedent. The House Gephardt rule operated 1979–1995 (and intermittently since).
Risk. Removes a (rarely effective) forcing mechanism the Right values.

P3. Fiscal commission with a guaranteed, unamendable floor vote

Mechanism. 16-member bipartisan commission; mandate to stabilize debt held by the public as a share of GDP by 2036 at or below its then level; no pre-excluded category (revenue, SS, Medicare, defense all on the table); fast-track up-or-down vote within 60 days of report if ≥2/3 of members approve.
Scoring. No official score (procedural). Required magnitude: ~$707B/yr in today's dollars (Auerbach-Gale fiscal gap) [S18].
Precedent. 1983 Greenspan Commission (enacted); Simpson-Bowles (no guaranteed vote; failed 38–382) [S34, S35]; BRAC base-closure commissions used the up-or-down model.
Risk. Congress can repeal the fast-track by majority vote; a commission can deadlock as in 2010.

P4. Trim the largest tax expenditures

Mechanism. Cap the employer health-insurance exclusion at the 75th percentile of premiums, indexed; cap the value of itemized deductions at 28%.
Scoring. CBO scores itemized-deduction limits at $0.74–3.42T/10 yrs depending on design [S32]; employer exclusion is the largest expenditure at $296B/yr [S33]; no official score for this exact combination.
Precedent. The ACA "Cadillac tax" on high-cost plans was enacted in 2010 and repealed before taking effect — a cautionary precedent.
Risk. Politically fragile (see precedent); raises effective taxes on middle-income workers with rich plans.

P5. Automatic continuing resolution (end shutdowns)

Mechanism. If appropriations lapse, funding continues at prior-year levels (no inflation adjustment) until enacted; Members' pay held in escrow.
Scoring. No official score (budget-neutral vs. prior year; avoids CBO-estimated $7–14B permanent losses per long shutdown [S26]).
Precedent. Several states operate automatic-continuation rules (not separately sourced here).
Risk. Flat nominal funding erodes real discretionary spending — Left worries about stealth cuts; Right about locking in baseline.

P6. Legislative tariff rule

Mechanism. Tariffs imposed by executive action under emergency/temporary authorities are scored at zero revenue in CBO baselines beyond 12 months unless ratified by Congress; any tariff used as a legislative offset must be statutory.
Scoring. No official score (scoring convention). Shows the fragility: CBO's tariff offset moved from $3.0T to ~$2.1T in six months [S1, S5].
Precedent. SCOTUS IEEPA ruling (Feb 2026) [S28]; refunds of $166B [S5].
Risk. Reduces the executive's trade-negotiating leverage; nationalist Right opposes.

P7. Federal balanced-budget amendment (strict ex-post design)

Mechanism. Constitutional amendment requiring year-end outlays ≤ receipts; 3/5 override; automatic waiver in recessions (two quarters of negative growth) and declared war; judicial enforcement limited to declaratory relief.
Scoring. No official score.
Precedent. State evidence: only strict, ex-post, independently enforced rules worked [S36]; no national precedent.
Risk. Procyclicality; states rely on a federal backstop that would itself be constrained.

P8. Medicare site-neutral payments

Mechanism. Pay the same for the same outpatient service whether billed by a hospital outpatient department or a physician office, phased over 4 years with a rural safe-harbor.
Scoring. CBO has scored versions in its options volumes; the referees did not re-verify a current range — treat as "scored savings, magnitude to be confirmed" [S32].
Precedent. Partial site-neutral policy for new off-campus hospital departments since the Bipartisan Budget Act of 2015 (not separately sourced here).
Risk. Hospital consolidation lobbying; rural hospital margins.

Also discussed but not advanced to a vote: 5% VAT ($2.18–3.38T [S32]) — Left worried about regressivity absent a rebate, Right about a new money machine; general-revenue backfill of Social Security (Delgado-Finch) — had insufficient support outside the Left to draft.

STAGE 7 — RECORDED VOTE

Blocs: Left 34 · Center 33 · Right 32 (moderator #100 abstains). Consensus requires ≥60% in every bloc (L ≥21, C ≥20, R ≥20).

# · Proposal · Left yes (of 34) · Center yes (of 33) · Right yes (of 32) · Consensus?
P1 · Social Security "1983 II" package · 19 (56%) · 29 (88%) · 21 (66%) · No (Left short by 2)
P2 · Replace debt limit (Gephardt-style) · 32 (94%) · 30 (91%) · 14 (44%) · No
P3 · Fiscal commission, guaranteed vote, nothing off the table · 23 (68%) · 30 (91%) · 22 (69%) · YES
P4 · Trim largest tax expenditures · 28 (82%) · 25 (76%) · 13 (41%) · No
P5 · Automatic continuing resolution · 24 (71%) · 29 (88%) · 21 (66%) · YES
P6 · Legislative tariff scoring rule · 30 (88%) · 26 (79%) · 17 (53%) · No
P7 · Federal BBA (strict ex-post) · 2 (6%) · 8 (24%) · 24 (75%) · No
P8 · Medicare site-neutral payments · 26 (76%) · 29 (88%) · 25 (78%) · YES

Consensus reforms: P3, P5, P8. Near-misses: P1 (Left 56%), P6 (Right 53%).

---

Vote explanations

Prof. Priya Venkataraman (#19, L) — P1: NO. "I'm for 80% of this package. I can't vote for longevity-indexing the FRA without a verified longevity-by-income analysis, and the referees rated that claim Unsupported (#48) in both directions. What would change my mind: a Chief Actuary score showing the minimum-benefit enhancement fully offsets the FRA change for the bottom two lifetime-earnings quintiles."

Prof. Douglas Wren (#2, R) — P1: YES; P4: NO. "P1 raises the cap, which I dislike, but it's the first package here that treats benefit growth as part of the problem. P4 raises taxes on middle-class workers through their health plans; the Cadillac-tax precedent says it won't survive anyway. What would change my mind on P4: pairing it with an equal-sized rate reduction."

Prof. Raymond Kessler (#18, R) — P2: NO; P3: YES. "The debt limit is a crude tool, but GAO's own acute-cost numbers ($107–161M over eight episodes) are small next to a $2T deficit. It's the only moment the fiscal path gets national attention. What would change my mind: replacing it with a binding debt-to-GDP or interest-to-revenue trigger that forces a vote."

Prof. Rosa Delgado-Finch (#7, Soc) — P3: NO. "A commission whose mandate is a debt-ratio target builds austerity into the terms of reference. What would change my mind: a mandate stated in terms of inflation and real-resource capacity, or one that required any benefit cut to be matched dollar-for-dollar by revenue from the top decile."

Prof. Miriam Castellano (#1, C-L) — P7: NO; P3: YES. "Bohn-Inman supports strict state rules; states have a federal backstop the federal government doesn't. What would change my mind: evidence from a sovereign with a strict constitutional balance rule that it didn't deepen a recession — I'd look hard at Germany's debt brake episodes."

Prof. Tobias Hartley (#4, Lib) — P6: YES; P7: YES; P2: NO. "Tariffs are taxes; Article I puts taxes in Congress. Half my bloc disagrees on P6 because they see tariffs as strategy, not revenue. What would change my mind on P7: seeing a credible escape clause design that isn't triggered every year in practice."

Prof. Benjamin Ashford (#31, R) — P7: YES; P5: YES. "The auto-CR at flat nominal levels is a quiet spending restraint and ends a costly ritual. What would change my mind on P7: if states with strict rules show no difference from weak-rule states in the next downturn — the prediction I put on record."

Prof. Gregory Stahl (#29, C) — P3: YES; P1: YES. "P3 passing and P1 narrowly failing is the right order: the commission is where P1 gets its missing Left votes, by trading them revenue."

STAGE 8 — VERDICT

Prof. Adelaide Wainwright (#100, political science / institutional design, moderator, abstaining)

Established (evidence is strong; no bloc seriously disputes it)

1. The path is unsustainable on current law. Deficit ~5.8% of GDP in a non-recession year, debt 101% → 120% (2036) → 175% (2056); CBO's director says so plainly [S1, S2]. FY2026 is running ~$2.1T after the tariff ruling [S6].
2. Interest is now a first-tier budget item: $1.052T in 11 months vs $833B for military outlays [S4]; 3.3% → 4.6% of GDP [S2].
3. Where the growth is: every outlay category above its 50-year average is Social Security, Medicare, Medicaid or interest; both discretionary categories are below average [S3]. Revenue is at its 50-year average (17.5% vs 17.3%) [S1].
4. OBBBA widened deficits by $3.4T (conventional), $4.1T with interest; dynamic feedback made it worse, not better [S7, S8, S9]; it was regressive in net resources by decile [S10].
5. Trust-fund dates: OASI 2032 (78% payable), OASDI 2034 (83%), HI 2033; the Social Security gap is 4.42% of payroll and widened 0.60 pp this year, mostly on fertility and immigration, 0.16 pp on OBBBA [S12, S15, S14].
6. Tariff revenue is real but legally fragile: the offset shrank ~30% in six months after Learning Resources [S5, S28]; tariffs are paid substantially at home [S31].
7. Process fights cost money — shutdowns ($7–14B unrecovered GDP per CBO [S26]) and debt-limit impasses (GAO, both estimates [S24, S25]) — and have not produced lasting consolidation.

Contested (reasonable experts disagree on evidence, not just values)

• How much fiscal space r<g still buys. Blanchard/F–S vs Mian-Straub-Sufi; CBO's 2031 crossover is a projection sensitive to rate assumptions [S19–S21, S17].
• Composition of consolidation: spending-based (Alesina et al.) vs Romer-Romer's finding that deficit-motivated tax increases are less costly [S40, S41].
• Whether "spending problem" or "revenue problem" is even the right frame. The panel's best synthesis: benefit formulas legislated to grow meet revenue legislated not to; which side yields is a values choice.
• Whether state BBA evidence transfers to Washington [S36].

Unknown

• The share of the Social Security gap closed by eliminating the taxable maximum on a 2026 basis (Chief Actuary update pending).
• Longevity-by-income gradients for FRA design (not verified in this thread).
• The durability of Section 301 tariffs and any further refunds.
• Whether AI productivity (CBO: ~+0.1 pp/yr [S2]) is under- or over-estimated.

Research agenda

1. Chief Actuary scoring of P1 on the 2026 basis, with distributional detail by lifetime-earnings quintile.
2. A "real interest / GDP" series published officially by CBO so the F–S test can be tracked without referee arithmetic.
3. Ex-post evaluation of state automatic-continuation (anti-shutdown) rules.
4. Tariff incidence studies using 2025–26 customs microdata, including refund pass-through.

Consensus reforms (≥60% in all three blocs)

• P3 Fiscal commission with guaranteed vote, nothing off the table (L68/C91/R69)
• P5 Automatic continuing resolution (L71/C88/R66)
• P8 Medicare site-neutral payments (L76/C88/R78)

Near-misses worth another round: P1 (Social Security package; Left 56%) and P6 (tariff scoring rule; Right 53%).

Where Symposium 1 was wrong or too glib on this domain

I did not re-read Symposium 1's fiscal material for this verdict, so I frame this as a list of glib claims this round disproved; to the extent Symposium 1 made any of them, they are superseded:

• "Tariffs will pay for the tax cuts." After February 2026 that is not a defensible planning assumption; even CBO's pre-ruling $3.0T didn't cover OBBBA's $4.7T [S1].
• "Growth will pay for it." CBO's dynamic score: +$124B revenue vs +$441B interest [S9].
• "90% debt is a cliff." No — a mild gradient [S39].
• "Just lift the cap"/"just tax the rich" solves Social Security. Not supported by any score we could verify; the gap is ~1.5% of GDP [S13].
• "Deficits don't matter for a currency issuer." 0 of 38 top economists agree [S23]; even the socialist speaker conceded the inflation constraint.
• "The Greenspan Commission raised the retirement age." The House floor did [S34].

Moderator's three-sentence verdict: The fiscal path is unsustainable on current law, driven by old-age and health spending growth meeting revenue held near its historical average and recently cut, with interest now exceeding defense. The evidence rules out the easy exits on both sides — growth feedback, tariffs, "just tax the rich," and "deficits don't matter" — and leaves a genuine values choice about the benefits/revenue mix. The panel agreed on process (a guaranteed-vote commission, ending shutdowns) and one health-payment reform, and came within two Left votes of a 1983-style Social Security deal.

Thread closed for Stage 8. Sources S1–S45 as listed in the Evidence Brief.

Independent Audit

Audit Panel — Prof. Laura Pennington (#97, operations research, C) & Prof. Olivia Brandt (#57, statistics, C). Brandt sat on this thread's referee panel and wrote the C2 "referee calc," so she recuses from every Stage 4 judgment and from any finding on her own arithmetic. Pennington leads. Corrections are appended here. Nothing above has been deleted or edited.

1. Source check (18 claims, fetched 2026-09-24)

# · Claim · Cited figure · Verified figure · Status · URL
1 · CBO baseline: deficit · $1.9T (5.8%) in 2026 → $3.1T (6.7%) in 2036; 50-yr avg 3.8% · Same · Confirmed · cbo.gov/publication/61882
2 · CBO baseline: debt; rev/outlays · 101% → 120%; passes the 106% record; revenue 17.5%, outlays 23.3% → 24.4% · Same · Confirmed · same
3 · OBBBA and tariffs in baseline · +$4.7T; tariffs −$3.0T · Same · Confirmed · same
4 · MBR, 11 months · $2.0T deficit; −$6B y/y, +$82B timing-adjusted; interest $1,052B (+12%) vs DoD $833B (+5%); SS $1,514B; Medicare $976B; Medicaid $655B (+8%) · Same · Confirmed · cbo.gov/system/files/2026-09/61984-MBR.pdf
5 · MBR receipts and customs · $4.845T (+3%); customs $167B vs $165B; ~$110B IEEPA refunds · Same · Confirmed · same
6 · Updated FY26 deficit · ~$2.1T; customs ~$250B below Feb · Same (Fortune, reporting CBO) · Confirmed · fortune.com/2026/08/10/how-big-deficit-national-debt-tariffs-cbo/
7 · CBO dynamic score, House H.R. 1 · +0.5% GDP; +$124B revenue; +$441B interest; +14bp; $3.4T with debt service · Same. The score covers the House-passed version (5/22/25). Conventional primary deficit $2.4T; dynamic $2.8T · Confirmed · cbo.gov/publication/61486
8 · CBO distribution · Bottom decile −$1,200 (−3.1%); top decile +$13,600 (+2.7%), 2026–34 · Same · Confirmed · cbo.gov/system/files/2025-08/61367-Distributional-Effects.pdf
9 · CBO tariffs update (8/20/26) · $0.9T smaller offset ($0.7T primary + $0.2T interest), 2027–36; $166B IEEPA; ETR ~10% · Same · Confirmed · cbo.gov/publication/62704
10 · Tax Foundation · $1.4T conventional / $1.0T dynamic; $820 per household; −0.4% GDP · Same (updated 9/10/26) · Confirmed · taxfoundation.org/research/all/federal/trump-tariffs-trade-war/
11 · Trustees 2026 · OASI Q4 2032 (78%); OASDI 2034 (83%); DI solvent; 4.42% vs 3.82% · Same · Confirmed · ssa.gov/news/en/press/releases/2026-06-09.html
12 · CRR decomposition · Fertility 1.90→1.75: 0.35; immigration 0.18; OBBBA 0.16; productivity +0.10; mortality +0.09 · Same · Confirmed · crr.bc.edu/social-securitys-financial-outlook-the-2026-update-in-perspective/
13 · CRFB Medicare HI · 2033; 89% payable; 0.56% (from 0.42%); alternative 1.38%; 4.1% → 6.5% of GDP · Same · Confirmed · crfb.org/papers/analysis-2026-medicare-trustees-report
14 · 1983 Amendments (SSA Bulletin) · 12 of 15 endorsed; Pickle 228–202; "long-range deficit cut from 1.80% to ~0.58%" · 12/15 and 228–202 are correct. **0.58% was the deficit left over after the Commission's package. The law as enacted, with the Pickle amendment, projected a +0.03% of payroll surplus.** Also, 8 of the 12 endorsers recommended a deferred, gradual retirement-age increase · Minor discrepancy · ssa.gov/policy/docs/ssb/v46n7/v46n7p3.pdf
15 · CBO Options 2025–34 · 76 options; FRA $95B; cap $0.73–1.43T; VAT $2.18–3.38T; chained CPI $278B; itemized $0.74–3.42T; high earners $48–197B · Same · Confirmed · cbo.gov/publication/60557
16 · Herndon–Ash–Pollin · >90% debt: 2.2% corrected vs −0.1% published · Same. The cited OUP page showed metadata only, so we verified against the PERI working-paper version · Confirmed (alt. URL) · peri.umass.edu/…/526-does-high-public-debt-consistently-stifle…
17 · IGM MMT survey (2019) · 0 of 38 agree; 22 strongly disagree; 15 disagree · Same (1 no opinion) · Confirmed · kentclarkcenter.org/surveys/modern-monetary-theory/
18 · GAO-26-107872 and CBO shutdown · $107–161M across 8 impasses; $1.3–5.9T at risk; replace the debt limit. Shutdown: 1.0–2.0 pp; $7–14B lost · Same. The page we fetched did not show the 2011 figure of $47–57M, so that number is not independently confirmed · Confirmed (one sub-figure unchecked) · files.gao.gov/reports/GAO-26-107872/index.html; cbo.gov/publication/61823

Tally: Confirmed 17 · Minor discrepancy 1 · Not supported 0 · Could not access 0. The cited HAP URL was unreadable, but we verified the figure through an alternate URL. Lang's note that the SSA Chief Actuary's 2026-basis payroll-tax provisions are "not yet available" is also confirmed.

2. Internal consistency

• Vote math: all 24 yes-count ↔ % pairs recomputed and correct. Consensus labels are correct: P3 (23/30/22), P5 (24/29/21) and P8 (26/29/25) clear every bloc floor. P1 fails because the Left has 19 votes, 2 short of 21. P6 fails because the Right has 17. No mislabels.
• Fact-check tally: 51 rows = S 30 (row 20, "Supported (dated)," counted as S) · C 7 · U 8 · W 6. This matches the stated tally.
• Verdict vs fact-check: no claim rated Wrong comes back as established. Hartley's "≈30% smaller" and P6's "$3.0T → ~$2.1T" compare the Feb window (2026–35) with the Aug window (2027–36). The ratio is approximate, not like-for-like.
• Roster: all 13 speakers and referees match on name, number, field and lean.
• Corrections appended:

1. Brief fact 39 and P1 "Precedent": replace "cut the long-range deficit from 1.80% to ~0.58%" with "the Commission package left a 0.58% gap; the enacted law, including the Pickle retirement-age amendment, projected a +0.03% of payroll balance." This makes the 1983 precedent stronger than stated.
2. Row 27 (Stahl, W): the rating stands, since the Commission did not enact anything. But the record should add the Bulletin's own detail: 8 of the 12 endorsing members recommended a gradual retirement-age increase, so the Commission did supply the idea that Pickle legislated. Stahl's "vehicle and cover" reply is better supported than the thread credits.
3. Row 19 referee calc (Brandt recused; Pennington's finding): the calc compares a 10-year CBO score for raising the cap (≈0.4% of GDP a year) with a 75-year gap (≈1.5% of GDP). The comparison mismatches both the window and the option (raise vs eliminate), so it cannot show "far short of full closure." The U rating stands on the absence of a 2026-basis actuarial score. The words "far short" should be dropped, and Venkataraman's retraction should be read as conditional on this point.
4. Row 4 (Wren, W): the basis is the CBO dynamic score of the House-passed bill over a 10-year window. W holds for "recovers a large part of the cost" within that window. Wren's beyond-window claim is uncited, and U would be the more precise rating for it.
5. Procedure: Lang, a referee, cross-examined Castellano (Exchange 11) and then rated the claim she had challenged (#26). The rating is correct, but a referee who questioned a claim should recuse from rating it.

3. Balance

Wrong ratings: 4 Right (Wren, Kessler ×2, Hartley), 1 Left (Venkataraman), 1 Center (Stahl). The skew toward the Right is earned. Each of those four rulings rests on a source we confirmed: S9's $124B vs $441B, CBO's 2031 r>g crossover, HAP's 2.2%, and CBO's $0.9T. Takahashi's advance warning that anyone citing "90%" as a cliff would be rated Wrong was fair notice, not bias. Two calls lean the other way. Delgado-Finch's "mainstream moving toward MMT" drew U despite the 0-of-38 survey. The referee calc against Venkataraman's cap claim was methodologically loose (correction 3). The "glib claims" list in the verdict splits 3 Right-coded, 2 Left-coded and 1 Center, and the verdict's "values choice" framing follows the evidence. No systematic tilt.

4. Grade: A−

All 18 claims checked are confirmed or nearly so, and the arithmetic is flawless. The deductions are for the 1983 "0.58%" misreading, which carries into P1; an apples-to-oranges referee calc; and a referee rating a claim she had cross-examined.

— Pennington (lead); Brandt (recused on Stage 4 and C2)

Verdict v2 (audit-corrected)

Prof. Adelaide Wainwright (#100, moderator). This post is appended to the record, and no earlier post is edited. It resolves the claims the thread left unverified and the items the audit flagged. It normalizes the Stage 4 tally to the standard scale (Supported / Contested / Unsupported / Wrong, where Unsupported includes "could not verify"). It then restates the verdict with corrected figures. Sources were fetched on 2026-09-24.

1. Resolution of unverified and audit-flagged items

Item · Claim · Result · Figure and source
Brief fact 39 / P1 precedent (audit) · 1983 Amendments "cut the long-range deficit from 1.80% to ~0.58%" · Corrected · 0.58% is the gap the Commission's package left open (it closed 1.22%, later revised to 1.41%). The law as enacted, including the Pickle retirement-age amendment, projected a +0.03% of payroll balance. 8 of the 12 endorsing members recommended a gradual retirement-age increase. SSA Bulletin v46n7
Row 19 (audit + resolution) · Eliminating the taxable max "solves" Social Security; referee calc said "far short" · Corrected · The audit is right that the referee calc compared a 10-yr CBO score for raising the cap with a 75-yr gap, so "far short" is dropped. The Chief Actuary's scores on the 2025 Trustees basis show: eliminating the cap with benefit credit (E2.2) improves the balance by 1.85% of payroll = 48% of the long-range deficit; without benefit credit (E2.1) by 2.55% = 67%. Raising the cap to cover 90% of earnings: 0.69% (18%) with benefit credit and 0.82% (22%) without. SSA OCACT summary. The 2026-basis scores are still "not yet available." The rating stays U: "solves" is not supported, but the realistic range is roughly half to two-thirds, not a small share
Row 22 · Interest is income to American households (amended: "largely domestic") · Now verified (amended form) · Foreign holdings of Treasuries were $9,248.1B (July 2026) per Treasury TIC. Against debt held by the public of ≈101% of GDP (CBO, S1), that is roughly 30% foreign / 70% domestic. The domestic share includes the Federal Reserve. Our calculation is approximate
Row 48 · Longevity gains concentrated among high earners · Now verified · Chetty et al., JAMA 2016: the richest 1% of men live 14.6 years longer than the poorest 1%; for women the gap is 10.1 years. Over 2001–14 the top 5% gained 2.34 yrs (men) and 2.91 (women); the bottom 5% gained 0.32 and 0.04. MIT News summary
Row 51 · Goods imports ≈11% of GDP · Now verified (approx.) · 2025 goods imports were $3,438.4B (BOP basis; BEA). Against nominal GDP of ~$30T, that is ≈11%. GDP was not separately fetched
Row 40 · Canada: program spending "fell 18.9% in two years" · Corrected (with caveat) · 18.9% was a planned reduction (S42). IRPP's retrospective reports a realized cut of about 10% (1993-94 → 1998-99) and program spending falling from 17.5% of GDP (1992-93) to 11.7% (1999-2000). IRPP, Half-Way Home. The IRPP spending aggregate may be narrower than total program spending. The realized decline was real, large relative to GDP, and smaller in nominal terms than 18.9%
Row 26 · 1983 deal split "roughly evenly" revenue/benefits · Still unverifiable · The SSA Bulletin gives package totals, not a provision-by-provision split
Row 5 · "Largest mandatory reduction in a generation" · Still unverifiable · A superlative with no source. Withdrawn by the speaker
Row 12 · Fiscal-dominance risk; 2021–23 partly fiscal · Still unverifiable · Theory. No source was pulled either way
Row 15 · A federal BBA "would work the same way" · Still unverifiable · A prediction. No federal evidence exists
Row 23 · Mainstream moving toward MMT · Still unsupported · The only survey (IGM 2019) runs 0/38 the other way
Audit, S24/S25 · GAO 2011 acute cost $47–57M · Still unverifiable · The text of GAO-26-107872 gives the $107–161M total across 8 impasses. The per-impasse values appear only in a figure, and we could not read an exact 2011 value from it. GAO-26-107872
P8 scoring · Site-neutral: "scored savings, magnitude to be confirmed" · Now verified (cross-thread) · CBO options: ~$39B/10 yrs (all off-campus HOPD) and ~$102B/10 yrs (on-campus subset), scored separately. The BBA 2015 new-off-campus rule is the precedent. Via KFF, confirmed in the domain 03 audit. KFF
P5 precedent · States with automatic-continuation rules · Still unverifiable · Not sourced

2. Rating normalization

This thread already rated unverified claims U. Two rows used C for what was really "not verified": row 12 ("no source pulled either way") and row 40 ("realized outcome not verified"). Both are reclassified C → U. Rows 3, 21, 28, 34 and 46 stay C, because the evidence on them is genuinely split.

· Supported · Contested · Unsupported · Wrong
Stage 4 as published · 30 · 7 · 8 · 6
Normalized scale · 30 · 5 · 10 · 6
After resolutions (rows 22, 48, 51 → S) · 33 · 5 · 7 · 6

Old → new: S30 · C7 · U8 · W6 → S33 · C5 · U7 · W6 (51 claims). Row 40 stays U: "fell 18.9% in two years" is not borne out as a realized figure. Row 4 (Wren) stays W for the within-window claim, per the audit, and his beyond-window expensing claim is noted as U.

3. Corrected verdict

Established

1. The path is unsustainable on current law. The deficit is ~5.8% of GDP in a non-recession year. Debt rises from 101% of GDP to 120% (2036) and 175% (2056). FY2026 is running at ~$2.1T after the tariff ruling.
2. Interest is a first-tier budget item: $1.052T in 11 months vs $833B for military outlays. (Added: roughly 70% of Treasury debt is held domestically, including by the Fed, so most interest is domestic income. That does not change its budget cost.)
3. Where the growth is: every category above its 50-yr average is SS, Medicare, Medicaid or interest. Revenue is at its 50-yr average (17.5% vs 17.3%).
4. OBBBA widened deficits by $3.4T (conventional) and $4.1T with interest. Dynamic feedback (House-passed version) made it worse. It was regressive by decile.
5. Trust funds: OASI 2032 (78% payable), OASDI 2034 (83%), HI 2033. The SS gap is 4.42% of payroll.
6. Tariff revenue is real but legally fragile. The offset shrank by $0.9T. The "≈30%" figure is approximate, because the Feb (2026–35) and Aug (2027–36) windows differ.
7. Process fights cost money (shutdowns $7–14B; debt-limit impasses $107–161M acute across 8 episodes, $1.3B FY2011-wide) and have not produced consolidation.
8. (New) Longevity gains are strongly skewed by income (Chetty 2016: a 14.6-yr gap for men between top and bottom 1%). This is now a verified input for FRA design. It is no longer an open question.
9. (New) 1983 is a stronger precedent than the thread stated. The enacted law, including the retirement-age rise the Commission's majority had recommended, projected a small long-range surplus (+0.03% of payroll). It did not leave a 0.58% gap.

Contested

• How much fiscal space r<g still buys (CBO's 2031 crossover is rate-sensitive).
• Composition of consolidation (spending-based vs Romer–Romer on tax increases).
• The "spending problem vs revenue problem" framing, which is a values choice.
• Whether state BBA evidence transfers to Washington.

Unknown

• The share of the SS gap closed by the cap options on a 2026 basis. Corrected context: on the 2025 basis, elimination closes 48% (with benefit credit) to 67% (without), and raising the cap to 90% closes 18–22%.
• How to offset FRA indexing for low-longevity groups. The gradient is now established; its distributional scoring is not.
• Durability of Section 301 tariffs and further refunds.
• Whether AI productivity (+0.1 pp/yr) is mis-estimated.

Glib-claims list, corrected. "Just lift the cap solves Social Security" remains not supported. But the reason should read: eliminating the cap closes roughly half to two-thirds of the 75-year deficit (2025 basis), not all of it. The earlier referee comparison of a 10-year score with a 75-year gap is withdrawn. "The Greenspan Commission raised the retirement age" remains Wrong as stated (the House floor did). The record now notes that 8 of the 12 endorsers had recommended the idea.

Conclusions that changed.
(a) The cap: the verdict's implied magnitude ("far short") is replaced by 48–67% for elimination. This makes a revenue-heavy SS package more feasible than the thread suggested, though still not sufficient on its own.
(b) P1's 1983 precedent is stronger (full closure on paper, not a residual 0.58% gap).
(c) The longevity gradient behind Venkataraman's P1 objection moves from Unknown to Established. What is still missing is the Chief Actuary's distributional score that she asked for.
(d) P1 component (a), raising the cap to 90% with benefit credit, closes only ~18% of the 2025-basis gap. That confirms the package needs its benefit-side components to reach solvency.

The consensus list as recorded is unchanged.

— A. Wainwright (#100)