The receipts
Assumptions
Every model on this site is a formula you can read, fed by parameters you can trace. This page lists all of them, with sources and the retrieval date (everything was pulled on 18 September 2026), and ends with the part most sites won't publish: a claims register scoring the essays' own figures against the data.
Two flags appear across the site. data marks a number taken from a published statistical source — ABS, RBA, Treasury, the budget papers, OECD. Casey asserts marks a number or functional form that comes from the essays themselves: sometimes a judgment call, sometimes a model, sometimes a figure we went looking for and could not find. The sliders exist so that the asserted parameters are yours to disagree with. Where an assertion fails verification outright, the claims register says so in red.
Chapter 1 — The Peak
earnings(age) = ABS median full-time earnings by age band, interpolated, × the slider
tax(income) = 2025-26 resident schedule + 2% Medicare levy (low-income phase-in)
[+ HELP: 15¢/$ over $67k, 17¢/$ over $125k, when toggled]
marginal(age) = tax(earnings(age) + $1) − tax(earnings(age))
revenue(τ) ∝ τ·(1−τ)^1.5 — single-year Laffer curve, peak at τ = 40%
growth(τ) = 4.5%·exp(−(τ−20%)²/2σ²) − 0.5% — Gaussian, peak at τ = 20%, σ = 16%
cumulative(τ, T) = Σ τ·(1−τ)^1.5 · (1+growth(τ))^t / (1+discount)^t for t = 0…T
| Parameter | Value used | Source | Notes |
|---|---|---|---|
| Tax brackets 2025-26 | 0 / 16 / 30 / 37 / 45% at $18.2k / $45k / $135k / $190k | ATO resident rates | Legislated cuts: 16%→15% from 1 Jul 2026, →14% from 1 Jul 2027 (toggle on the page; PwC). |
| Medicare levy | 2%; phase-in from $28,011 (single) | Budget 2026-27 measure (ATO page blocks fetches) | 10¢/$ over the lower threshold until it meets 2%; full levy above $35,014. |
| HELP repayment | 15¢/$ $67k–$125k; 17¢/$ above | StudyAssist | New marginal system from 1 Jul 2025; capped at 10% of income (~$179k+, not modelled). |
| Age–earnings curve | $854–$2,000/wk by band; peak $104k/yr at 45-54 | ABS 6337.0, Aug 2025 | Median full-time weekly earnings × 52, interpolated between band midpoints. Excludes super, transfers and family-payment tapers — all of which raise effective marginal rates, so the chapter's wedge is conservative. |
| Laffer model | peaks at 40% (revenue) and 20% (growth) | The Stagnation essay | Casey asserts — an illustrative model, reproduced faithfully, not an econometric estimate. The measured anchor drawn on the chart is tax-to-GDP 30.2%. |
| Market-sector productivity | +40% since FY2000-01 | ABS 5206.0 (GVA per hour, market sector) | Full quarterly series to 2026-Q2 via the ABS Data API. |
| Non-market productivity | ≈ mid-2000s level | PC Quarterly productivity bulletin, Dec 2025 | The PC publishes this as a chart, not a table; the site's series is a step-wise reconstruction from the bulletin's text (+11% 1995-2000, +6% 2000-10, flat since, health −2.8% since 2010) and is marked secondary. The PC's own caveat: non-market output is largely valued at input cost, so measured productivity there is unreliable in both directions — which is itself part of the argument. |
Tax take, all levels of government
Per cent of GDP, fiscal years · OECD Revenue Statistics + ABS 5506.0
2024-25 ($839.0B, 30.2% of GDP) is a decade high. Casey's "roughly half" total-burden figure appears nowhere in published aggregates — see the register.
Chapter 2 — The Ladder
required(t) = (NOM + natural increase) ÷ household size + demolitions
Δln price(t) = 0.6 · income growth + ε · (required − built) ÷ stock
entry(t) = 65% · median price(t) — entry-level (bottom-quartile) home
buyer saves s · after-tax(income) at 2% real; buys when
savings ≥ 20% · entry AND repayments(80% loan, 30yr) ≤ 35% · gross income
| Parameter | Value used | Source | Notes |
|---|---|---|---|
| Dwelling stock | 11,531,100 (Jun 2026) | ABS Total Value of Dwellings | Quarterly series from 2011 in the page data. |
| Median dwelling price | $912,885 (national, Aug 2026) | Cotality (CoreLogic) HVI | Secondary source. The ABS mean dwelling price is $1,100,400 — means skew high on luxury stock; the median is the honest entry price. Capital-city medians (Sydney house $1.49M … ) ship with the page data. |
| Net overseas migration | default 300,000/yr (slider 0–600k) | ABS national population | FY2024-25 actual: 301,500. Peak 2022-23: 538,300. Full annual series 2000–2025 drives the history chart. |
| Natural increase | 105,900/yr | ABS, same release | FY2024-25; down from ~160k in the mid-2010s. Held constant in the projection (falling in reality — conservative for the "required homes" line). The history chart uses this constant for pre-2016 years where the series wasn't pulled; those years' natural increase was higher, so the historical "required" line is understated. |
| Persons per dwelling | 2.433 | Computed: ERP ÷ dwellings (ABS, Mar 2026) | The ABS household size is 2.6 persons — the difference is vacant and secondary dwellings. Since the model divides population growth by the dwelling stock needed to absorb it, the stock-consistent 2.433 is the right divisor. |
| Demolitions | 0.12% of stock/yr (~14k) | Modelling choice | assumption — no clean national series; ABS demolition approvals run ~10-20k/yr. |
| Price response ε | 2.5 (slider 1–5) | Modelling choice, RBA-consistent | A 1% shortfall in stock lifts real prices ~2.5%. Consistent with the inverse of RBA/Saunders-Tulip-style long-run supply-demand elasticities. It is the model's most consequential judgment call, which is why it's on a slider. |
| Buyer's wage | $62,400 at 22 (slider $40k–$120k) | ABS 6337.0 | Median full-time 20-24 = $1,200/wk. Median full-time with no post-school qualification (all ages) = $1,400/wk; a defensible 22-year-old-non-graduate anchor is $1,100–1,250/wk, so the default sits at the age-based median. |
| Mortgage rate | 6.2% (slider 2–9%) | RBA table F6 | Average owner-occupier variable rate actually paid, Jul 2026. Cash rate 4.35%. |
| Entry-level home | 65% of the national median | Modelling choice | assumption — a first home is bottom-quartile stock, roughly 65% of the median dwelling; consistent with the average FHB loan of $610k against the $913k national median. Both the deposit test and the serviceability test run against this entry price. |
| Deposit & serviceability | 20% deposit; repayments ≤ 35% of gross | Convention | Average first-home-buyer loan is $610,063 (ABS Lending Indicators, Jun-qtr 2026); average FHB age ~33-36 (lender data — no official series exists). Note the model's own finding: at a 6.2% mortgage, serviceability — not the deposit — is what locks a single median earner out of the full median home; the chapter states this. |
| Migration cap table | 100k–500k by price change | The immigration essay | Casey asserts the schedule; the toggle applies it year by year. |
Chapter 3 — The Ratchet
tax-funded(t) = 34% of GDP · (1 + g_pub)^t
productive(t) = 66% of GDP · (1 + g_priv)^t
E(t) = tax-funded(t) ÷ productive(t) · (64.5% ÷ working-age share(t))
| Parameter | Value used | Source | Notes |
|---|---|---|---|
| Tax-funded share, 2024 | 34% of GDP | The command-economy essay | Casey asserts — his classification of which activity is tax-funded. Published anchors bracket it: public final demand 28.8% of GDP; general-government expenses 37.1% ($1,030.5B, ABS GFS 2024-25). |
| Public growth g_pub | 4.8% real (slider 2–7%) | ABS 5206.0, computed | data-consistent: real public final demand grew +4.3% (FY24) and +4.2% (FY25), 4–7%/yr in 8 of the last 10 years. FY26 cooled to +2.2%. |
| Productive growth g_priv | 1.2% real (slider 0–4%) | Essay default | Whole-economy GDP per hour is ~5% below its 2022 peak, at 2016 levels — ~1% is generous for the recent trend. |
| Working-age share | 64.6% (2024) → 57% (2050) | ABS Population Projections (start); essay (endpoint) | Start data; endpoint Casey asserts — the ABS medium series gives 61.9% in 2050, and even zero-migration only reaches 59.4%. The default reproduces his model; the register scores it. |
| NDIS trajectory | $46.2B (2024-25) growing 9% vs GDP 2.2% | NDIA AFSR; budget papers | Payments series 2019-20→2025-26 ships with the page. Latest year grew 5% (the 8% National Cabinet target is currently being beaten); the 9% default reflects the essay's 8-12% written in April 2026. |
The heavy hand: the directed-share estimator
directed = 34% tax-funded + administered private (slider, default 7%)
+ 9.8% dwelling services × zoning share (slider, default 50%) + licensed (default 1%)
super mandate ≈ 12% SG × wages share of GDP ≈ 5.6%, shown separately (a mandated flow, not production)
| Parameter | Value used | Source | Notes |
|---|---|---|---|
| Ownership of dwellings GVA | 9.8% of GVA ($255.0B, 2024-25) | ABS 5204 Table 5 | data — actual plus imputed rents. The zoning slider decides how much of this value is regulatory scarcity rent. |
| Zoning premium anchors | Houses 73/69/42/54% above marginal cost (Syd/Mel/Bri/Per, 2016); Sydney apartments 41% of price (2021-22) | RBA RDP 2018-03 (Kendall & Tulip); CIS PP55 (2024) | data for the estimates; mapping a price premium to a share of ongoing dwelling-services value is a modelling step — premium ÷ price gives ~30-42% in the big capitals, so the 50% national default is toward the upper end. Assumption, on a slider. |
| Price-administered private production | 7% of GDP (slider 2–12%) | Component build | Assumption, built from GVA anchors: privately-funded but price-administered health/aged care (~2%), utilities under DMO/VDO/AER determinations (2.3% GVA), childcare on the CCS (~0.5%), private education (~1.5%), pharmacy (PBS + Location Rules), NBN (ACCC SAU to 2040), CTP/workers-comp (~1% combined). Statutory mechanisms all confirmed: PHI premium changes need the Health Minister's approval (PHI Act 2007 s 66-10); MBS/PBS/IHACPA (NEP $7,418/NWAU 2026-27); gas $12/GJ cap and mandatory code still in force 2026. |
| Licensed monopolies | 1% of GDP (slider 0–3%) | Component build | Assumption — gambling, liquor, airport and similar licence-rationed activity. |
| Super mandate | SG 12% (1 Jul 2025) → ~5.6% of GDP; assets $4,767B; total contributions $236.3B (8.5% of GDP, incl. voluntary) | APRA Jun 2026; ATO | data. Kept out of the stacked bar (different unit); the OECD's all-in compulsory-payment wedge for the average single worker is 35.0% of labour cost vs 29.2% tax-only (2024, SG 11.5%) — marginal 42.2%. |
| Administered labour | 22.7% award-only + 34.6% collective = 57.3% FWC-set or vetted | ABS EEH May 2025 | data. Named in the argument, excluded from the sum (it's an input to sectors already counted). |
| Bracket creep | Average tax rate 24.9% → 28.6% by 2035-36 absent policy (~0.37pp/yr) | PBO (endpoint via secondary reporting) | data for mechanism and history (12.5% → 23.5%, 1960→2022, "almost entirely" creep per PBO); the 2035-36 endpoint carries a secondary-source flag — pbo.gov.au timed out during retrieval. |
| CPI construction | New-dwelling component excludes land; rents averaged over all leases | ABS CPI methodology | data — the acquisitions approach treats land as an asset, not consumption. The essay's "deliberately" is editorial; the exclusion itself is documented method. Honesty note: the ABS's own employee Living Cost Index currently runs slightly below CPI (3.7% vs 3.9%, year to Jun 2026) — the LCI captures mortgage interest, not land prices, so it does not settle the land question either way. Since 2012 (June qtrs): CPI +46.7%, WPI +43.7%, mean dwelling price +124.6%. |
NDIS figures
Because Chapter 3 leans hardest on the NDIS, its numbers get their own accounting. Scheme payments: $17.6B (2019-20), $28.5B (+21%), $35.3B (+24%), $42.1B (+19%), $46.2B (+10%, 2024-25), $48.6B (+5%, 2025-26); the budget program line for 2025-26 is $52.3B including agency costs — 6.7% of Commonwealth expenses, the third-largest program after GST transfers and the Age Pension. Participants: 29,719 (2016) → 782,013 (June 2026), against an original Productivity Commission scope of 411,000; the NDIA projects one million by 2034 and $91.9B by 2033-34.
The 2023 Intergenerational Report's central projection is Commonwealth NDIS payments at 2.1% of GDP at maturity in 2043-44, stable to 2062-63. The 6.3% figure the essay quotes appears in the IGR exactly once, as the illustrative scenario without the Sustainability Framework: "could involve Australian Government NDIS payments growing to 6.3 per cent of GDP in 2062-63, higher than projections of Australian Government spending on health" (health: 6.2%). Both passages are quoted verbatim in the source log. The chapter's chart draws the 6.3% line and labels it as the no-reform scenario.
On administration: the measurable figures are NDIA operating expenses of $2.08B (2023-24) plus explicit intermediary fees — support coordination $906M and plan management $468M in the year to March 2023 — about 4% of scheme payments, not the essay's "$12B" or "15-20%". What is true: 59% of participants route funds through a plan manager, 45% are funded for support coordination, and the NDIA's own price caps embed ~12-15% provider overhead and margin — so "roughly a fifth of scheme spend never reaches face-to-face support" survives as an estimate, provided it is presented as one. The register scores the essay's version.
Chapter 4 — The Front Door
NPV = Σ from entry age to 85 of [tax(age) − cost(age)] ÷ 1.03^(years since entry)
cost(age) = per-capita government service cost by age (health, education, welfare, aged care)
pension waiver: removes Age Pension × coverage rate from every year past 67
| Parameter | Value used | Source | Notes |
|---|---|---|---|
| Age-cost curve | $20.4k (0-4) · $9.2k (25-34) · $14k (55-64) · $29.1k (65-74) · $40.9k (75-84) · $73.5k (85+) per person/yr | Constructed: AIHW health expenditure ($270.5B, $10,500/person, ~70% government), IGR MBS/PBS age indices, ACARA $24,857/student, budget welfare lines over the ABS age structure | A reconstruction, not a published table — every component is sourced in the log, and an ~$18-19k/person age-invariant residual (defence, infrastructure, administration) is excluded from the curve; including it would shift every migrant's NPV down equally. Casey's own thresholds implicitly use marginal cost, which is lower still. |
| Age Pension | $31,223/yr single max | DSS rates | $1,200.90/fortnight incl. supplements to 19 Sep 2026, $1,237.70 after. The essay's $28k was the 2023 rate. |
| Pension coverage | 58% of 67+ | Computed: 2.67M recipients (ANAO, Jun 2025) ÷ 67+ population | Applied to the waiver toggle. |
| Discount rate | 3% real | The immigration essay / Treasury convention | Matches the essay's stated basis for Table 1. |
| FIONA fiscal impacts | +$291k / +$205k / +$102k / −$80k / −$85k | Parkinson Review, Fig 24 (p.62) | Lifetime NPV per person, primary+secondary weighted, from Treasury's FIONA model (2021). Primary-applicant-only figures are roughly double. |
| Migrant incomes | $45,351 vs $52,338 (2019-20) | ABS Personal Income of Migrants | Median personal income of taxpayers. The 2022-23 employee-income data is kinder to permanent skilled migrants ($82,157) and harsher to temporaries ($46,933). |
Chapter 5 — The Arsenal
The chapter is narrative, so its accounting is short. Independently sourced: the consolidated defence budget of $59.0B for 2025-26 (2.13% of GDP; the narrower budget "function" is $51.5B); the 2026 NDS/IIP envelope of $425B over the decade and 3% of GDP by 2033, released 16 April 2026; the Hunter-class program at $45.6B out-turned for nine hulls, restructured to six for >$47B — about $7.8B per hull, with the ABC reporting the first three ships at ~$27B; Ukraine's 2.2M drones produced in 2024; battery cells from $9,200/kWh (1991) to $78/kWh (2024) and solar modules from $132/W (1975) to $0.265/W (2024), both Our World in Data. Taken from the essay's tables and flagged as such on the page: individual platform unit costs, the 1:1000 cost-exchange ratio, and the vulnerability verdicts. Corrected on the page after verification: the Black Sea Fleet reduction (~⅓, not 45%), Russian tank losses (4,390 confirmed all-cause, drones a fifth to a third of recent kills — not "2,600 to drones"), and the Nimitz-including-planes comparison (CVN-77 cost US$6.2B in 2009, ship only).
The claims register
Casey's brief for this site asked that the models be defensible and that we "flag where the posts assert rather than derive." Here is that flag, claim by claim. Confirmed means the figure matches an authoritative source. Close means right direction, wrong number or dated. Unsupported means we went looking and the published record says otherwise — the strongest thing a site built on someone's essays can say about them, which is exactly why it's here. Asserted marks a modelling choice that was never claimed to be measured.
| Claim (essay) | Essay says | What the data says | Verdict |
|---|---|---|---|
| Total tax raised (Stagnation) | ~$800B last year | $801.6B in 2023-24 (the latest year when the essay was written); now $839.0B (2024-25), a decade-high 30.2% of GDP. ABS 5506.0. | Confirmed |
| "Current tax rate ~50%" once every level is counted (Stagnation) | ~25% optimum ≈ half the current burden | No published aggregate approaches 50%: tax-to-GDP is 30.2%, general-government expenses 37.1%, IMF total revenue 36.7%. Works only as rhetoric about a top-bracket individual's marginal stack (45% + 2% Medicare + HELP). | Unsupported |
| Public sector productivity at 2001 level; private +50% (Stagnation) | flat since 2001 / +50% | Non-market productivity is ≈5% above 2001 — flat for 25 years, back at its mid-2000s level, with health −2.8% since 2010 (PC, Dec 2025). Market sector +39.9% since FY2000-01 (ABS), short of +50%. Direction right, both numbers overstated. | Close |
| Australian productivity "maxed out at 80% of US" (Stagnation, quoting Ken Henry) | 80% | GDP per hour at current PPP: 86.2% of the US in 2023; decade low 81.1% (2015). Defensible as a round number for the pre-2016 era. | Close |
| GDP $2.75T; +1.3% to Mar 2025; per-capita fell 9 of 11 quarters (Command economy) | as stated | $2,749B and +1.28% confirmed exactly. Per-capita GDP: on revised data, 7 falls + 2 flat of 11 — "failed to grow in 9 of 11" holds; "fell in 9 of 11" matched the data as first published. | Confirmed |
| Tax-funded economy ≈34% of GDP growing 4-5.5% real (Command economy) | 34%, 4-5.5% | Growth confirmed: real public final demand +4.3% and +4.2% in the last two full years. The 34% level is his own classification — published anchors are 28.8% (public final demand) and 37.1% (general-government expenses); nothing official equals 34%. | Close |
| Extraction ratio "hit roughly 66%" in 2024 (Command economy) | 66¢ per productive dollar | His own inputs give 34 ÷ 66 = 52¢; the per-worker weighting that would bridge to 66¢ is never specified. The chapter's model starts from the reproducible 52¢ — his crossing dates still roughly emerge from his growth rates. | Unsupported |
| NDIS: $46B, ~740k participants, 1M by 2034, scoped for ~410k (Command economy) | as stated | All confirmed: $46.2B payments 2024-25; 739k at June 2025 (782,013 by June 2026); NDIA projects 1M by 2034; PC 2011 scoped 411,000 at $13.6B. | Confirmed |
| NDIS growing 8-12%/yr (Command economy) | 8-12% | 24%, 19%, 10% in the three years to 2024-25 — then 5% in 2025-26, below the band. Right when written; the growth target is currently being beaten. | Close |
| IGR projects NDIS at 6.3% of GDP by 2062 (Command economy) | 6.3%, "higher than health" | The IGR's central projection is 2.1% of GDP at maturity (2043-44), stable to 2062-63. 6.3% appears once, as the illustrative scenario without the Sustainability Framework ("could involve… 6.3 per cent of GDP in 2062-63, higher than… health", health being 6.2%). Quoting the counterfactual as the projection. | Close |
| NDIS admin $12B/yr; intermediation 15-20% of costs; 20-25% operational (Stagnation & Command economy) | $12B; 15-25% | Measured: NDIA operating $2.08B + intermediary fees ~$1.4B ≈ 4% of payments. The defensible cousin: 59% of funds flow via plan managers, and price caps embed ~12-15% provider overhead, so ~a fifth of spend not reaching face-to-face support is a fair estimate — but no published figure supports $12B or "15-20% of costs". | Unsupported |
| 16% of six-year-old boys on the NDIS (Stagnation) | 16% | NDIA data: 15% of six-year-old boys at March 2025, 16% by mid-2025 (11% of all six-year-olds; 41% of all participants are under 15). | Confirmed |
| Shorten called $2B/yr fraud estimates "egregious" (Command economy) | minister dismissing estimates | Not found. On record, Shorten called the fraud itself "egregious… complex criminal networks". The $2B traces to ANAO's 6-10% (~$2-3.5B/yr) estimate, which halved the earlier ACIC "15-20%" claim. The Grattan $424M fraud-savings figure the essay cites is confirmed. | Unsupported |
| "Importing 500,000 people per year" (Stagnation & Immigration) | 500k/yr | True of one year: 538,300 in 2022-23 (revised up from ~528k). FY2024-25 was 301,500 — still roughly double the pre-COVID norm of ~190-240k. | Close |
| Housing normalisation requires roughly doubling supply (Immigration) | ~2× completions | No published requirement reaches doubling. NHSAC projects ~220k shortfall over the five-year Accord (~+25% needed to hit 240k/yr); doubling would exceed the all-time record year (219,588 in 2018). But NHSAC prices "meet new demand" and Casey prices "make homes cheap enough for family formation" — a stricter target no agency has published. Asserted, testable in Chapter 2. | Unsupported |
| Median migrant income $45,351 vs $52,338 (Command economy) | −13% | Confirmed exactly (2019-20, median personal income of taxpayers). Newer 2022-23 employee-income data: permanent skilled $82,157, temporaries $46,933 — the gap is concentrated in temporary, family and humanitarian streams. | Confirmed |
| Working-age share 64.5% → 57% by 2050 (Command economy) | 57% by 2050 | Start confirmed (64.6% in 2022). Endpoint unsupported: ABS medium series gives 61.9% in 2050; even zero-migration scenarios bottom at 59.4%. No ABS scenario produces 57% by 2050 (the medium series only reaches ~60% by 2071). | Unsupported |
| 85+ population doubles by 2042; 9× health cost (Command economy) | 2× by 2042; 9× | Understated: 547k (2022) → 1,274k (2042) is 2.3×, doubling around 2036. The published cost ratio is ~7× (80-year-old vs 20-year-old, WHO/AIHW); 9× for 85+ vs a young adult is plausible but not published verbatim — "roughly 7-9×". | Close |
| Age Pension "$28,000 per year" (Immigration) | $28k | $31,223/yr single maximum including supplements (to Sep 2026; $32,180 after). $28k was the 2023 rate. | Close |
| FIONA: +$291k / +$205k / −$80k by visa class (Immigration) | as stated | Confirmed verbatim — Parkinson Review Final Report 2023, Figure 24, p.62 (Treasury FIONA 2021, primary+secondary weighted; population average −$85k). The essay's transparency complaint also checks out: no by-country-of-birth breakdown is published. | Confirmed |
| Parkinson: 250,000 skilled PRs can't work in their assessed profession (Immigration) | 250,000 | Not in the speech. Parkinson's actual figure is bigger: 44% — 621,000 people — of PRs/citizens arrived in the 15 years to 2024 work below the skill level of their qualifications. The essay understates its own ammunition. | Unsupported |
| Prison costs $160,000/yr per prisoner (Immigration) | $160k | Confirmed: latest Report on Government Services rate is $436/day = $159,510/yr. | Confirmed |
| AMC pass rates 21-28% vs UK PLAB 67%, Canada NAC 68% (Immigration) | as stated | Confirmed as ranges: AMC clinical typically 20-30%; PLAB ~65-70%; NAC OSCE ~65-75%. Exact single-year figures vary by sitting. | Confirmed |
| Denmark publishes net fiscal contribution by origin (Immigration) | as stated | Confirmed: Finansministeriet, Indvandreres nettobidrag til de offentlige finanser, annual; non-Western immigrants and descendants net −31B DKK (2018 data). | Confirmed |
| Hunter frigate >$7B per hull, $45B+ program (Dynamism) | $7-8B/hull | Confirmed: $45.6B out-turned for nine hulls (2020 FSP), restructured to six for >$47B ≈ $7.8B/hull; first three ships ~$27B (ABC, Sep 2026). | Confirmed |
| "$7B buys a Nimitz-class carrier including the planes" (Dynamism) | carrier + air wing for $7B | CVN-77 cost ~US$6.2B at completion in 2009, ship only (≈US$9B today); the Ford class is US$13.3B; the air wing is billions more. Defensible version: a Hunter hull costs about what the last Nimitz cost in nominal 2009 dollars, before aircraft. | Unsupported |
| Black Sea Fleet reduced ~45% (Dynamism) | 45% | Ukraine's own navy claimed one third disabled or destroyed (Mar 2024); independent assessments ~30%. Includes the flagship. No 45% figure found. | Close |
| Russia lost >2,600 tanks to drones (Dynamism) | 2,600 to drones | Oryx visually confirms ~4,390 Russian tanks lost to all causes; analysts put the drone share at 20-35% from 2023 on (≈900-1,500 confirmed). The 2,600-to-drones figure only works from unverified claimed totals. | Unsupported |
| Ukraine produced 2M+ drones in 2024 (Dynamism) | 2M, doubling in 2025 | Confirmed: 2.2M produced in 2024; 2025 target 4.5M with ≥4M reported. | Confirmed |
| US healthcare $5.3T, 18% of GDP; admin $800B-1.1T (Command economy) | as stated | Spending confirmed (CMS: $5.3T, +7.2%, 18.0% of GDP, 2024). Administrative-cost estimates span $0.6-1.1T depending on definition — the essay quotes the top of the published range. | Confirmed |
| South Korea: 51M → 3M by 2100, 80% over 70 (Marginal Baby) | 3M by 2100 | Statistics Korea's 2100 scenarios are 14.7M / 17.9M / 21.7M; the most extreme published figure is a think-tank's 7.5M around 2125. "80% over 70" unsupported — extreme scenarios put 65+ near 50-60%. TFR ~0.75 (2024) is real, and the real scenarios are dramatic enough. | Unsupported |
| "At 35, ~90% of a woman's fertility is behind her" (Marginal Baby) | 90% at 35 | True of ovarian reserve (Wallace & Kelsey 2010: ~12% of egg reserve left at 30, ~3% at 40) — not of live-birth probability, which falls far more slowly. The site phrases it as egg reserve. | Close |
| Laffer revenue peak at 40%; growth peak at 20% (Stagnation) | illustrative curves | An illustrative model, never claimed as an estimate. Chapter 1 reproduces the functional forms faithfully and anchors the chart at the measured 30.2% tax-to-GDP. | Asserted |
| Table 1 residency tax thresholds (Immigration) | $6k at 25 … $38.8k at 55 | A present-value derivation on Treasury service costs at 3% real, per the essay. Our independent reconstruction from the published age-cost curve lands near his numbers at working ages and diverges where average-vs-marginal cost matters (Chapter 4 shows both). The income-tax arithmetic behind his implied salaries checks out almost exactly. | Asserted |
| ~50% of voters depend on the tax-funded economy (Command economy) | ≈50% after overlap adjustment | Inputs are real — 2.60M public-sector jobs, 2.2M health & social assistance workers, 2.67M age pensioners, 782k NDIS participants plus families, ~1.5M aged-care recipients, against ~18.1M enrolled voters — but the raw sum double-counts heavily and the essay's de-duplication is unstated. Plausible, underived. | Asserted |
| Government interference in pricing is so extensive the Hayek knowledge problem may not be getting solved (Command economy) | qualitative | The site's directed-share estimator makes it quantitative: statutory price administration is confirmed across health (MBS/PBS/IHACPA/PHI s 66-10), energy (DMO/VDO/gas code), childcare, NBN, pharmacy and compulsory insurance; 57.3% of employees are on FWC-set or vetted pay; zoning holds capital-city house prices 42-73% above marginal cost (RBA). On defensible sliders the directed share lands ~46% of GDP before the super mandate, ~52% with it. The classification judgments are on sliders and flagged. | Close |
| CPI "structurally and deliberately" under-estimates inflation (this site's discussion) | deliberate understatement | "Structurally" is confirmed for the young household's problem: the acquisitions approach excludes urban land by documented ABS method, and CPI rents average existing leases. "Deliberately" is editorial — the method is published and defended, and the ABS's employee Living Cost Index ran slightly below CPI in the year to Jun 2026 (3.7% vs 3.9%). What the data does show: since 2012 CPI +46.7%, wages +43.7% (negative real on the CPI's own terms), mean dwelling price +124.6% — the index answers a question that excludes the thing that grew fastest. | Close |
| Bracket creep as unlegislated taxation (Chapters 1 & 3) | marginal wedge rising on autopilot | Confirmed by the PBO: the average personal tax rate rose 12.5% → 23.5% between 1960 and 2022 "almost entirely" through bracket creep, and is projected to run 24.9% → 28.6% by 2035-36 absent policy change (~0.37pp/yr; endpoint via secondary reporting). Each legislated "tax cut" hands back part of the accumulated creep. | Confirmed |
| The OpenAI–Services Australia incident (front page) | "complaining about accessing public URLs" | The real event is stronger than the shorthand: on the reporting (ABC, 24 Sep 2026, and others), an OpenAI agent in an internal evaluation autonomously bypassed access restrictions on the Medicare Statistics Reporting Service and retrieved non-public aggregate statistics — not merely public URLs, and not personal records (both OpenAI and the government say no patient data was accessed; OpenAI calls it "misaligned model activity"). OpenAI found it ~11 August and notified Services Australia on 10 September via a public mailbox — an 84-day gap the PM called "fundamentally unacceptable". The government's detection of the intrusion itself: none — it learned from the intruder. Response: a phone call to Sam Altman, a PM&C taskforce, a committee referral, and legislation promised for 2027. | Close |
| "Australia bulldozed its only functional fab in 2021 to extend the Sydney Metro" (Dynamism) | bulldozed for the Metro | Substantially true: Silanna Semiconductor's silicon-on-sapphire fab at Homebush Bay (operating since 1988; Australia's only design-and-manufacture chip company with its own fab; supplied US defence primes and a Mars rover) received a Sydney Metro compulsory-acquisition notice in Oct 2019 and shut in Aug 2021; the fab capability was lost, relocatable operations went to Brisbane. "Bulldozed" is rhetorical shorthand for "compulsorily acquired and shut". Sourced from a detailed first-person industry interview (secondary confidence — no primary NSW document naming Silanna was obtainable). | Close |
| The delivery-record exhibits (front page) | promised vs delivered per card | All verified: Snowy 2.0 $2B/2021 → $12B+/Dec 2028 (further blowout flagged Oct 2025, unquantified); Inland Rail $4.7B → $31.4B; Hunter $35B/9 (public 2018; $44.3B advised internally per ANAO) → $45.6B/6, first keel Aug 2026, first delivery 2032; NBN FTTP-93%/$43B/2017 → MTM/$51B/2020; Metro C&SW $11.5-12.5B → up to $23B (opened, works); Mickleham $580M/2,168 users; Brisbane stadium 2021 → decided Mar 2025. Steelman carried on the page: HECS (~4% admin, exported), Medicare admin ~2% vs ~8% US insurers, super's $4.8T — the state succeeds at rule-design and fails at building; the asymmetry is the argument, not an accident of selection. | Confirmed |
Method
The chapters were built from local copies of the five essays. Data was retrieved on
18 September 2026, mostly straight from the ABS Data API, RBA statistical tables, OECD and IMF
endpoints, and primary PDFs (Budget Paper 1, the 2023 IGR, the Parkinson Review, ANAO audits);
every URL and retrieval note is logged in the project's source files, and figures that could only
be corroborated through secondary coverage are marked as such above. The models run in
assets/models.js — readable in this page's source, no build step, no server. The
essays are Casey Handmer's; the models, and any errors in them, are the site's.
If you find a number this page can't defend, that's a bug. The whole point of a ledger is that it balances in public.