IL51: Why Most Recessions Are Completely Misunderstood ft. Tyler Goodspeed

Core claim. Goodspeed's book, built on a US/UK recession chronology extended back to 1700 (132 recessions), argues that the standard "boom-bust" narrative — that expansions accumulate excess and must eventually correct — is wrong. Recessions are instead exogenous shocks that knock an economy off a stable long-run trend, which then recovers to that trend (the "pluck" model), rather than an oscillation above and below trend that self-corrects. He frames it as Peter Pan (expansions don't age into death, but can be killed by misfortune) versus Dorian Gray (hidden rot must eventually be purged).

The evidence chain. Using survival-analysis techniques on the extended chronology, Goodspeed finds expansions are no more likely to end in year ten than year one or two — true both before and after 1945, contradicting even Diebold and Rudebusch's 1990s finding that pre-1945 US expansions aged into death (he attributes the discrepancy to using the improved Davis-Romer chronology). Recessions, by contrast, do die of old age: the single strongest predictor of a recession ending is simply its own duration.

He also tests whether the height, speed, or duration of an expansion (GDP growth, bank credit growth, even skyscraper-height records) predicts the depth or length of the subsequent recession, and finds no statistical relationship on either side of the Atlantic. Separately, he tests whether recessions coincident with financial crises produce slower recoveries — using a binary crisis indicator from Stephen Broadberry's UK dataset and the size of bank-credit contraction as a proxy — and finds no difference in UK recovery speed going back to 1700, a result corroborated for the US since the 1880s by Michael Bordo. He attributes the divergence from Reinhart and Rogoff's cross-country finding to their broader country sample potentially picking up other correlates of slow recoveries; he says he has discussed the discrepancy directly with Rogoff and it remains an open comparison.

On the "creative destruction" defense of recessions — that they cleanse inefficient firms and reallocate capital — Goodspeed finds the opposite: R&D spending is procyclical (falls in recessions), job-to-job quitting and hiring both collapse (workers stop leaving jobs, firms stop poaching), and younger, more dynamic firms are disproportionately killed off because they lack collateral for credit. Sectoral output composition after a recession typically looks similar to what it would have been had growth continued uninterrupted along trend — i.e., little net reallocation actually occurs.

On countercyclical government policy, he tested for statistical trend breaks at points of major state expansion (1913 Federal Reserve founding, 1946 UK Beveridge reforms, 1960s Great Society) and found none — no acceleration toward longer expansions or shallower recessions coincides with these institutional changes. Recession depth and duration have been statistically indistinguishable since 1945 versus before, aside from the 1914-1945 period, which he calls the real historical outlier for volatility. He is careful to note this isn't an argument against relief spending — he supports higher unemployment-insurance replacement rates where income loss is concentrated — just against the idea that countercyclical policy has smoothed the cycle itself.

Mechanism: what actually causes recessions. Goodspeed's data point overwhelmingly to sector-specific shocks in goods with few near-term substitutes, especially energy: oil shocks contributed to US recessions in 1948, 1953, 1957, 1970, 1973, 1980, 1981, 1990, and — he argues, underappreciated — 2001 and 2008. Pre-oil, coal strikes and, further back, adverse harvest shocks (which reduced feed for draft animals) and peat shortages played the same role. He estimates unconditional annual probabilities of roughly 1-in-10 for an energy-related recession (a probability that has been declining), 1-in-100 for a pandemic-related recession, and 1-in-50 for a credit-controls-related recession.

This also explains why the historically less-recession-prone UK differs from the US: nationwide branch banking since 1826 diversified UK bank risk versus thousands of small, undercapitalized US unit banks (barred from interstate or even intrastate branching until the 1980s), and the UK ran without a coal strike from 1926 to 1972, insulating it from oil shocks until 1973 — when the Arab embargo combined with a coal strike and forced Ted Heath to call the