The Twenty-Four Compared

---

This table closes the series: twenty-four national economies, each rated year by year as lived — the median household's experience, the era's own verdict, the world-relative view, the direction of travel, with statistics as cross-check — and each chained to the same 2026 dollar. Two volumes end early by the war convention (rate only what can be measured): Russia/USSR at 2021, Iran at 2025. Argentina and France carry the July 2026 re-audit. Everything else runs 1946 (or independence/founding) through mid-2026.

Three ways to read this table

The ladder (the average). The lived average is not a prosperity ranking — it is an experience ranking: how good the median year felt, era by era. That is why the United States (5.26) and Taiwan (5.24) sit at the top by different roads — one never crashing, the other never stopping — and why Saudi Arabia, briefly the richest society per head ever recorded, averages beside Spain: the oil kingdom's median year, as lived through decrees and droughts of generosity, was a middling one. The spread is tight — 3.18 to 5.26 — because every country's history is mostly ordinary years; the ladder measures whose ordinary was kinder.

The multiple (the engine). The per-capita multiple is the eighty-year growth verdict, and it scrambles the ladder. China ×162 (nearer ×115 from its pre-war peak — see the war-conventions note) and Taiwan ×55 are the miracles; Turkey ×10 is the sleeper — a better multiple than Germany or Italy, earned while its currency lost all but one fifteen-millionth of its 1946 dollar value: growth with a burning wallet. The two catastrophes are Argentina ×2.2 and Iran ×2.6 — the only countries that essentially failed to compound at all, one by chronic self-sabotage, one by revolution, war, and siege. A low multiple with a high start (Canada ×3.9, Australia ×4.4) is maturity; a low multiple with a low start is tragedy.

The share (the position). The %-of-US column is the world-relative story. Three shapes recur. Convergers: Taiwan 4→46, Korea 4→40, Poland 13→32, Italy 27→48 — the ladder-climbers. Holders: Australia 76→74, the UK 66→63, Sweden 63→70 — rich at the start, rich at the end, the whole drama in between. Losers: only four countries end with a smaller share than they started — Argentina 34→17, Mexico 19→15, Canada 71→61 (the quiet decade's toll), and Iran 8→4, the series' starkest fall. And one country did what no other ever managed on this table's ruler: Saudi Arabia crossed above the American line in the constant-dollar chain itself — 139% of US income per head in 1974 — before dilution and price carried it back to 41%. (At market exchange rates the line is a currency thermometer, and several countries floated over it: Japan from 1987 to the century's end, peaking near 150% in 1995; Sweden in the late seventies and around 1990; Germany in 1995 — crossings that appeared and vanished with the dollar, which is exactly why this series doesn't measure that way.)

The fourth reading: the distribution — or, what the whipsaw hides

The lived average is an arithmetic mean, and a mean counts years — one vote each. That makes it blind to the single most important difference between two economies that share one: the shape of their years. The table below is the league table's missing dimension, in the same order.

Reading the columns. Years at 6+ counts the good years and years at ≤2 the bad ones. Crash episodes counts distinct spells at or below 2 — consecutive bad years count once — so it separates one long catastrophe from several short ones. Longest run at 5+ is the longest unbroken good stretch. And whipsaw (σ) is the standard deviation of a country's year-ratings: how far a typical year sat from that country's own average. It measures variability, not quality, and has to be read beside the average. China's 1.81 and India's 1.60 are the highest here because both lived famine and boom inside one lifetime; Japan's 1.58 spans ruin and miracle. The lowest, Pakistan's 0.94, does not mean a steady ride in any happy sense — it means the years were steadily poor, clustered tightly around a low mean. Only Australia's 1.00 means what it sounds like: eight decades that never produced a bad year to swing away from.

Read the seventeenth and eighteenth rows together. Brazil (3.65) and Russia/USSR (3.64) are a statistical dead heat with almost nothing else in common: one produced this scale's worst printed inflation and never a systemic 1, the other produced the century's great collapse and never a hyperinflation it couldn't end by decree. The whipsaw column separates them — and both from Mexico, a hair below at 3.62, whose middling was steadier than either.

Why did the boom years once rate 6s and 7s? Level — and the level was real: Argentina entered this scale at 34% of US income, and its consumption floor in a good year sat above most of the bottom half's ceilings. The August 2026 grid kept every point of that defense in the household column and re-priced everything around it: a consensus that never trusted an Argentine boom, a world position eroding in nearly every year of the volume, and a direction forever six months from its own cliff.

And why does "basket case" remain the correct eighty-year verdict anyway? Because the mean counts flows, and Argentina's catastrophe was stored in stocks. A crash year rates a 1 or a 2 once — but 1989's hyperinflation and Bonex seizure, 2001–02's corralito and pesification, 1975–76's and 2018's currency halvings each burned savings, contracts, and trust accumulated across all the pleasant 5s and 6s that preceded them. The destruction of decades registers in the table as a single bad year; the un-compounding it caused registers in the other columns: a per-capita multiple of ×2.2 — twenty-fourth of twenty-four — and a US-income share of 34→17, the only country in this collection that began rich and un-converged. Twenty-third on experience, dead last on trajectory: after eight decades of contradiction the two ladders finally agree, and their agreement is the most precise definition of the Argentine tragedy this scale can offer.

The league table

Notes on the table

• Averages within ~0.15 are ties. Every rating carries a ±1 judgment band, so adjacent league positions (Turkey–Iran, Italy–Saudi, Brazil–Russia–Mexico, Turkey–Iran–Argentina, Canada–India–Korea) are ordering artifacts, not findings. The findings live in the distributions, multiples, and shares.
• The mean counts flows; the multiple counts stocks. A country can average mid-table on lived years while destroying, in its few crash years, the accumulated savings of all the good ones — see the fourth reading above. Argentina is the type specimen.
• The war conventions: Russia/USSR ends at 2021 and Iran at 2025 — this series does not rate years whose statistics have stopped. A corollary at the other end: war-floor bases inflate multiples. China's 1949, Japan's and Germany's 1946 all open on rubble, so their multiples measure recovery-plus-growth in one number. The extreme case is China's ×162: measured from its 1936 pre-war peak (~$120 per head on this chain) rather than the civil-war floor, the climb is nearer ×115 — still the steepest here by a wide margin, the flag changing the arithmetic's meaning, not its verdict. Iran is the only volume to close on a 1: bombed, snapped back, and repricing its banknotes, with the IMF's −6.1%/68.9% 2026 projection carried as epilogue.
• The highest floor is Australian: min = 3 — the only country never to record a bad year, a run including twenty-eight recession-free years (1991–2019). The United States is the only other volume never to touch 1.
• The only crossing — in chained terms: Saudi Arabia passed 100% of US income per head around 1973–74 in this table's constant-dollar series, peaked at ~139%, and held triple digits for a decade — then fell by more than half into the late nineties and sits at 41% today: the greatest windfall and the largest sustained peak-to-present decline in the collection, mostly by dilution (a population up sevenfold). Japan's famous 1987–2000 reign above the US line (peak ~150%, 1995) was a market-exchange-rate fact — real and lived as national vindication, recorded in that volume's 1987 entry, but a product of the yen's flight rather than the chained series this table ranks by.
• The best-multiple-worst-money paradox is Turkish: ×10 real growth per head — ahead of Germany — through six coups and a currency that shed all but one fifteen-millionth of its 1946 dollar value, including twenty-six unbroken years of 25%+ inflation.
• The two non-compounders: Argentina (×2.2) and Iran (×2.6). Argentina did it to itself in peacetime; Iran's path ran through a foreign-organized coup, a revolution, an eight-year invasion, and four sanctions walls — and its income per head has now spent fifty consecutive years below its own 1976 peak, pinned at 4–5% of US since 1988.
• The quiet decade is Canadian: 71% of US income in 1946, 84% at the 1981 peak, 61% today — the rich world's stealthiest relative decline, +4.8% total per-head growth since 2014.
• The statistical twins: Iran (3.39) and Turkey (3.36) — neighbours arriving at the same lived average by opposite catastrophes: chronic inflation without collapse versus collapse upon collapse with the inflation as garnish.

Method note: all figures in 2026 dollars, each volume chained to the IMF's 2025 nominal anchor for that country; per-capita multiples and US shares computed on each volume's own internally consistent series. The lived average weights every year equally — history, in this series, is mostly the middle of the table.

— Compiled July 2026. Sources: the twenty-four volumes of this series and their underlying national statistical office, central bank, World Bank WDI and IMF WEO series (each volume chained to its own 2025 anchor). All dollar figures in constant 2026 dollars; ratings on the series' common 1–10 scale. Russia/USSR closes at 2021 and Iran at 2025; all other volumes run to July 2026, provisional.

The audit trail — what the asterisk marks

An asterisk in the league table means the volume's ratings were re-audited after first publication. Every round is recorded here, in the order it was run.

Argentina's peronist-summit and convertibility-era years trimmed (its single remaining 7s honest ones); France's mid-sixties and 1998 lifted a notch. Both volumes carry the diagnosis inline.

Pakistan's Ayub half-decade (1960–63), remittance crest (1980, 1982, 1985), and 2003 re-rating were lifted a notch in the series' first upward re-audit (3.08 → 3.18); its ceiling of 5 stands — see the parity-paradox pattern in that volume.

A second Pakistan round corrected the Siachen entry's agency (India's 1984 seizure; Pakistan counter-deployed) and moved the 2022 flood year from 1 to 2 under the floor-band doctrine, leaving 1971 the volume's only 1 (3.19).

Brazil's borrowed-time audit trimmed the inflation-financed JK peaks (1958–60), the fraud-flagged 1973, and Delfim's 1980 by a point each (4.70 → 4.64, sixth to eighth), while upholding the Miracle's core, the II PND, and the inclusion decade — the reasoning published in that volume.

Saudi Arabia's enclave audit gridded the maximum-divergence dawn: 1946–52 rose (2s to 3s, 3s to 4s; 3.79 → 3.88) while the squandering years held on the direction lens — deficits at record revenue — with 1958's near-bankruptcy as vindication. A second Saudi round then defined the world-relative lens series-wide — gaining-or-losing ground on the world, level confined to the household column — lifting 1947 and 1950–52 (3.88 → 3.93, past Italy inside the tie band), re-reading the USSR's 1948–51 sub-scores with blends unchanged, and confirming Brazil's 1973 grid already conformed.

A third round traced the boundary: 1970–71 rose to 5 by the same lens that lifted 1950, 1982 fell to 3 by its own grid, and 1983 held — leaving 1970 and 1983, at near-identical income per head, two honest points apart. A comprehensive fourth round then scored all eighty-one Saudi years on the five lenses: thirteen ratings moved in both directions — the 1957–58 nadir-and-rescue swap, the afterglow years 1975–81 trimmed, the 1999–2000 rescue lifted, the plateau's 2010–13 drift trimmed — and sixty-eight held, including 1974's 7 at the grid's maximum (3.88, sixteenth). Every Saudi stats line now carries the two dials — annual crude output and the era's own nominal quote. The French volume received a readability-and-milestones pass: dense terms unpacked in place (the soulte, cohabitation, the rigueur turn) and the missing triumphs added — Gerboise Bleue, Canopus, Concorde's full arc, Ariane, the TGV, the last Mururoa campaign. The same pass then ran series-wide: fifty-odd landmark flights, bombs, satellites, stadiums, and cultural thunderclaps added with dates where absent — Apollo and the Walkman, Lop Nur and Mangalyaan, the Maracanazo and the Miracle of Bern, the Opera House and Taipei 101 — and a systematic detector then swept every acronym and italicized term in all twenty-four volumes, adding some seventy more glosses in a third round.

Iran then received the full-grid treatment: all eighty years scored on the five lenses — twenty moves in both directions (the golden decade's ignition shifted to 1965, the revolution's staircase sharpened, the sanctions-era hope-years lifted and pressure-years trimmed), net −2, landing at 3.36: a dead tie with Turkey. Australia's per-capita column was then found to have drifted from its GDP column — chained without full population growth from the late 1950s and compounding to ~30%, manufacturing a phantom 2023–24 collapse at the splice with the correctly anchored endpoint — and was rebuilt as GDP over population throughout; an introductions pass gave every first-use figure a role and properly explained the Snowy Scheme, Petrov, the 1967 referendum, and WorkChoices. The Australia standard then went series-wide: a first-use introductions pass across the other twenty-three volumes, adding roles to every figure not globally famous — presidents, premiers, chancellors, finance ministers, central-bank governors, founders — with the already-introduced skipped by guard.

A mechanical name detector then swept all volumes for anything the curated lists missed, adding forty-odd further introductions (Byrnes's Stuttgart speech among them) and restoring one missing landmark: the 1962 Spiegel affair, with Defence Minister Strauss, added to the German volume.

A completion pass then finished the five volumes the detector's display cap had half-reviewed — Cross, Laporte, and Henderson in Canada among eighteen further introductions. Mexico then received the full-grid review: sixteen moves in perfect symmetry — 1968 restored to 5, the 1980–81 pre-cliff and the 2005–07 plateau trimmed, the 1995 and 2020 floors lifted to 2 with 1982 keeping the only 1 — and the 3.62 average unchanged to the hundredth.

A closing series-wide battery then reconciled every average, ran a chain-continuity test that caught the Australian defect in five more volumes (a Saudi population-base splice hiding since 1960 the largest — all rebuilt by division, with errata), and audited every 1 and every 7-plus against the doctrine codex: the top of the scale passed clean, while six floors set before the collapse-set doctrine moved to 2 — all four of India's (leaving the republic the only major volume with no 1 at all, and lifting it past Sweden to sixth), Spain's 2020, and Turkey's 2021 (past Iran, both audited). The full ledger is in the companion audit record.

The India question then completed the arc: a challenge to its sixth place above Korea triggered the fourth full grid, and the household column — absent when the volume was written — reset it entirely: fifty-three moves, the four 8s to the poor-country ceiling of 6, the average from 4.70 to 3.73, seventeenth — below Korea by nearly a point, above Indonesia by a quarter, still without a single 1.

The author then overruled the experiment as doctrine: an absolute household ladder handicaps poor countries while crediting nothing for stability, policy capacity, long-term progress, or investor returns — weights now recognized as load-bearing. India's original ratings are restored in full, the four 1s included (4.65, seventh, the Sweden–India–Brazil band); the recalibration survives as a published, rejected experiment in that volume's trail; and China's profile, the same class as India's restored one, leaves the queue.

Spain then received the fifth full grid: thirteen symmetric moves left 3.78 untouched while the shape corrected — 1961 and 1968 up as challenged, 1973 down two as the pre-cliff class, 1963–65 and 1972 to the honest 5-band, 2006–07 held at the borrowed-boom cap — plus a readability pass (Banesto and Conde decoded, ETA introduced, Matesa, the cajas, and Draghi's sentence unpacked)); Sweden then took the same pass — Hammarskjöld given his UN office and his Ndola, ATP introduced as the earnings-related pension battle, Palme tagged at his 1968 debut, the folkhem trio unpacked in the intro, Nyckeln's fuse named, and a Northvolt anachronism corrected.

Turkey then received the sixth full grid — fourteen moves, net +8 on its own unrecorded-economy doctrine (the Truman-umbrella years, the 1958 and Derviş rescue-templates, Decree 32, the customs union, the solid-banks 2009, the leakage pair, the program's first real-income years), 3.37 to 3.47, inside a tie-width of both Indonesia and Iran — plus a Money-block forensic: three entries restored to their true years and the 2005 six-zero deletion finally given its own.

A second readability sweep then ran the full detectors over both recent volumes — twenty-odd further fixes, Lindgren given her Pippi, the Murat revealed as a car, not a man, and İnönü, Vance, and Löfven titled.

A closure-and-superlative audit then swept all twenty-four for the class of error the Palme line exposed — claims of resolution, permanence, or record that time or the volumes' own numbers had since overtaken. Thirteen internal contradictions surfaced, ten of them real: Italy's per-capita summit “never regained” beside its own higher 2025 and 2026, India's 1979 trough called the worst ever beside its own deeper 2020, Argentina's 1995 unemployment record beside its own 2002, Korea's fertility milestone dated a year late, and four unbounded “worst since” records that 2020 broke. All corrected with errata.

A companion sweep then tested every “highest/lowest since” claim against the volume's own series and every lived-best claim against its own ratings: twenty-seven flags, six real — Argentina's 1969 price-stability record, Australia's 2019 growth trough, France's 2000 peak, Italy's 1986 disinflation, and two Mood claims whose “best-lived year” had equals inside its own window — the remaining twenty-one confirmed sound, including Bank Rate's genuine 1955 low.

Indonesia then took the seventh full grid: twelve moves, ten up, net +8 to 3.58 — the 1968–69 disinflation, 1984's rice self-sufficiency, 2009's positive year and 2016's amnesty all priced as the record deserved, 1961 and 2013 trimmed — with 1966 lifted to 2 on its within-year turn while 1965 holds its 1, the grid reading 1·1·1·1·2 on a year when the state's authority dissolved.

A readability pass ran with it: Pertamina properly introduced and the “second state” ambiguity resolved, Bulog, Cikini, Busang, Tempo, Malari, Petrus, and the Timor car all given their context, the Money headlines normalized, a stale country-count retired, and East Timor's 2002 independence added. Brazil and Argentina then received the eighth and ninth full grids — the last two pre-doctrine volumes, and the largest paired correction in the record: Brazil fifty-five moves to 3.65 (nearly every 6 borrowed, printed, or weathered; the miracle's honest 5s and 2010's ceiling 6 standing; 1973's world-relative 7 preserved as the lens's founding case), Argentina fifty-five to 3.30 (the household level-defense kept in full, the peronist summits held at the borrowed-boom cap, and consensus, position, and direction finally priced — twenty-third on experience, dead last on trajectory, the two ladders in agreement at last). A peak-density ruling then trimmed India's boom: one 8 stands — 2007, the crest — with 2003, 2005, and 2006 at 7s, four equal summits becoming a crest and its foothills (4.61, seventh unchanged). A base-year note then flagged the war-floor multiples: China's ×162 quantified against its 1936 pre-war peak (~×115), the same flag extended in principle to Japan's and Germany's 1946 openings — arithmetic clarified, no figure changed. A share-column audit then recomputed all forty-eight start and end shares against the chained American series: forty-seven held; Russia's end share had been divided by the nominal 2021 US figure instead of the chained one — 22 corrected to 18, the convention (same final year, both sides in 2026 dollars) now stated on the record. The plan-spring ruling then re-landed India's 1953–56: one 5 at the 1953 crest, three 4s behind it — no sub-seven-percent year holds a 6 anywhere on the reviewed scale — 4.52, eighth, into a Canada–India–Korea band spanning seven hundredths.

The USSR's audit ran two rounds: the famine floor and the household reading were upheld in round one; round two, on lens weighting — the household had absorbed the whole verdict in the very years the other four lenses diverged most — moved 1948–51 to 4 and 1957 to 6. Where the lenses split that sharply, volumes now show the split.

A third round sharpened the lens definitions and two sub-scores, and a fourth, evidence round (Ellman, Ó Gráda, Bergson) enriched 1946–47 with archival numbers; and a fifth adjudicated the reconversion symmetry with America's 1946 and moved the famine's onset year from 1 to 2 on China's own onset-vs-peak template (3.57 → 3.64 across the five rounds), restating the famine floor as a 1–2 band in which the unrelieved death-years keep the 1. A full-series verification pass (July 2026) then reviewed every volume's dated claims, corrected six — logged as errata in the affected volumes — and added brief glosses for names and institutions little known outside their countries.