Did colonialism enrich Europe or impoverish the rest?

Argue it from the inside, in its own voice, before any qualification. Eric Williams put the strong form on the table in 1944: the slave trade and the plantations did not merely accompany the rise of British industry — they financed it. The commercial capitalism of the eighteenth century built the wealth of Europe by means of slavery and monopoly, and that accumulated wealth turned over into the industrial capitalism of the nineteenth. Beckert’s Empire of Cotton (2014) modernizes the claim into a system rather than a slogan: “war capitalism” — armed merchants, expropriated land, enslaved labor — was the substrate the Manchester mill grew out of, and the Lancashire spinning frame was one node in a transatlantic circulation, not a freestanding feat of British engineering. The institutional-channel depth here — why GDP-share badly measures what war capitalism built — is carried in the slavery walkthrough’s Stage 3. Berg and Hudson (2023) then put a recalibrated number under it: roughly eleven percent of late-eighteenth-century British GDP tied to the slave economy, double the parameter the prior forty-year consensus had settled on — the recalibration arithmetic is walked at that walkthrough’s Stage 3.

The framing reaches past Atlantic slavery. Pomeranz’s ghost-acreage move generalizes it: Britain’s takeoff was land-saving in a way the Yangzi Delta could not replicate, because Britain drew on a coerced-labor-and-resource periphery — Caribbean sugar as caloric input, American cotton as textile feedstock, the silver-driven Indian Ocean circulation — that supplied flows its own soil could not. The parity-until-1750 evidence underneath that claim is the spine of the Great Divergence walkthrough’s Stage 1. Allen’s high-wage account sharpens the mechanism: Britain’s peculiar factor-price configuration — expensive labor, cheap coal — is what induced labor-saving, capital-deepening innovation, and the cotton half of that configuration ran on the slave-cotton complex; the Allen mechanism is walked at Stage 2 there. Inhabited at full strength, the enrichment framing returns one verdict: a real, substantial colonial dividend to European wealth — Williams-massive at the upper bound, Allen-Beckert-Berg-Hudson at the center, the old Engerman benchmark at the floor, but always positive and never trivial.

Argue this framing the same way Stage 1 argued its own — from the inside, at full strength, before any qualification. The colonial period installed extractive institutions where Europeans could not settle, and those institutions did not dissolve at independence; they channel forward into present-day incomes, and the channel is large enough that institutional quality explains a substantial slice of the gap between rich and poor countries. Twenty years of attempts to break the result mostly failed, which is what the 2024 Nobel was recognizing. And the apparatus is informative precisely because it is not uniform: it returns a strong effect in the extractive cases — the Congo under Leopold, the Caribbean sugar islands, the mita zones of Peru, much of British India — and a more modest one in the settler cases, the United States, Canada, Australia, where the institutions imposed resembled the metropole’s own. That variation is not noise to be averaged away; it is the finding. The disruption concentrates exactly where extractive institutions were installed and then kept, because the rent-extraction structure of the colony was the structure post-independence elites inherited and had every reason to maintain. The colonial choice was load-bearing for centuries because it built the political-economic floor that everyone after had to stand on.

The framing runs deeper than institutional persistence alone. Bagchi’s India is not a place that failed to industrialize — it is a place that was deindustrialized, an existing competitive manufacturing base taken apart by tariff asymmetry and fiscal extraction, the receiving end disrupted not by the absence of colonial inputs but by the colonial intervention itself. Davis’s famines are the same mechanism at its most lethal: late-Victorian mass death manufactured by forcing subsistence economies into world grain markets after the institutions that had managed scarcity were already broken — externalities of empire, not acts of God. And Wallerstein and Frank give the whole its name: the periphery was peripheralized, its surplus pulled toward the core by the very terms on which it was admitted to the world economy. Whether that long-run wound is still visible in the cross-country income gap today — and how it stacks against geography, human capital, and the rest of the menu — is the question the rich-and-poor-countries walkthrough takes up at Stage 4. Inhabited at full strength, the impoverishment framing returns its own verdict: a developmental disruption that is real, variable by colony type, and persistent across centuries.

Take the same-phenomenon reading at its strongest first. The colonial system is one system. The cotton-textile-and-finance circulation that Beckert and Berg-Hudson put at the center of the enrichment story is the same circulation Bagchi puts at the center of the impoverishment story — same merchants, same shipping lanes, same flow from plantation extraction to metropolitan investment. Every pound of textile revenue booked in Manchester has a coerced-labor-and-displaced-handloom counterpart at the periphery; the gain and the wound are two readings of a single ledger entry. On this reading the two framings are making one finding from two vantage points, and their verdicts should largely converge. This reading is partly true, and the part that is true is not small.

Now take the different-phenomena reading at equal strength. The two apparatus quantify different things. The enrichment side asks what fraction of British capital formation was attributable to Atlantic colonial profits — a flow in the metropole’s macroeconomy, bounded by the reach of growth accounting. The impoverishment side asks what fraction of today’s cross-country income variation traces to colonial-period institutional choices — a level and a trajectory in the colony’s long-run political economy, bounded by the reach of the settler-mortality instrument. A flow-side estimate and a persistence-side estimate are not interchangeable; you cannot net them, average them, or convert one into the other. On this reading the two framings measure genuinely different aspects of the system, and their verdicts should not be expected to converge. This reading is also partly true.

Both are partly right, and that is the position — not a hedge but a claim about the relationship itself. At the level of what is being measured, the same-phenomenon reading holds: there is one colonial system underneath both framings. At the level of how each apparatus quantifies it, the different-phenomena reading holds: the two instruments cut the system along different axes and return magnitudes that are asymmetrically bounded by which one you picked up. The framings inhabit a single underlying phenomenon and surface different faces of it through different apparatus selections. That is the reframe’s signature move: framing-direction selects apparatus, apparatus-selection bounds what can be seen, and the asymmetric composition of the two verdicts is the thing the flip was built to make visible.