Meta: when and how big?

Thesis

Before answering why the Great Divergence happened, we have to agree on when it happened and how big it was. The data necessary to do this — reconstructed GDP per capita, real wages, life expectancy, agricultural productivity — are thin for pre-1800 Asia and contested even where available. Two rival empirical narratives now anchor the field:

1. The California-school narrative. As late as 1750, NW Europe and the most advanced regions of Asia (the Yangzi delta, Mughal Bengal, Tokugawa Osaka) were at rough parity across the measurable metrics. The divergence is late (18th–19th century), regional (NW Europe vs. specific Asian regions rather than Europe-vs-Asia), and substantially contingent on specific European advantages (coal, ghost acres, state-competition-driven commerce). Associated with Pomeranz, Wong, Frank.

2. The Broadberry / revisionist narrative. Revised long-run GDP reconstructions — incorporating urbanization, sectoral output reconstructions, and careful Asian-source work — show NW Europe pulling ahead of China by 1500 or earlier, and substantially ahead by 1700. The divergence is earlier, deeper, and structural — it cannot be explained by late-18th-century contingencies because it was already substantial before those contingencies kicked in. Associated with Broadberry, Guan, Li, Gupta, Karaman.

The two narratives are empirically incompatible at the strong form, and the choice between them profoundly shapes every causal argument about the divergence. If parity in 1750 was real, causal stories have to explain a late and contingent event. If parity was a measurement artifact and the true divergence goes back to 1500 or earlier, causal stories have to explain a centuries-long structural phenomenon that was already visible in the medieval economy.

This position is therefore a meta-debate: not about why the divergence happened but about what we’re explaining.

Lead proponents

• California school: Kenneth Pomeranz, Bin Wong, Andre Gunder Frank, R. Bin Wong, Roy Bin Wong, Li Bozhong, Jack Goldstone.
• Broadberry school / revisionists: Stephen Broadberry, Hanhui Guan, David Li, Bishnupriya Gupta, Kivanç Karaman, Şevket Pamuk. Also parts of the work of Jan Luiten van Zanden on early-modern European growth.

Key arguments (California school)

1. Appropriate comparators should be regional, not national. Comparing “Europe” to “China” is too coarse. The economically meaningful comparators are the most advanced regions of each: the Yangzi delta, Mughal Bengal, and Tokugawa Osaka vs. England, Flanders, Netherlands. At this granularity, parity in 1750 is defensible.

2. Real-wage measures suggest parity. Pomeranz’s reading of wage and price data in the Yangzi delta and in NW Europe suggests comparable subsistence-basket wages through much of the 18th century.

3. Non-wage welfare metrics support parity. Life expectancy, height data, housing quality, commercial-market access — multiple indicators converge on a picture of rough equality in advanced regions through the early 18th century.

4. Pre-1500 “Europe ahead” claims are retrospective. The notion that Europe was “pulling ahead” by 1500 reads post-1800 European victory back into prior centuries. Before the Atlantic turn, Europe was one of several regional civilizations, not obviously ahead of Song China, Abbasid Baghdad, Ming China, or Mughal India. The “Europe ahead since 1500” claim requires selective reading of sparse pre-modern data.

Key arguments (Broadberry / revisionist)

1. Systematic GDP reconstructions. The British-GDP-1270–1870 series (Broadberry et al. 2015), the Chinese-GDP series (Broadberry, Guan, Li 2018), the Indian-GDP series (Broadberry, Custodis, Gupta 2015), and Ottoman and Dutch parallel series all use consistent methodology. The picture they produce: NW European per-capita GDP crosses Asian levels by 1300–1500 and pulls away steadily from there. By 1700, NW Europe is substantially ahead even of the Yangzi delta.

2. Urbanization rates support the early-divergence story. Urbanization correlates robustly with economic sophistication; NW Europe’s urbanization rate exceeded China’s from the 14th century onward and the gap widened.

3. Pomeranz’s parity-supporting data don’t hold up under scrutiny. Allen’s efficiency-units wage-adjusted data suggest Chinese wages were substantially below English wages even in the Yangzi. Life-expectancy comparisons are fragile — Chinese reconstructions rely on a small number of source populations and later work has found shorter Chinese life expectancies.

4. The California-school claim confuses “high development” with “high population density.” The Yangzi delta was densely populated and commercially integrated, which is impressive, but high population density is not the same as high per-capita output. Broadberry shows that per-capita measures favor Europe even where aggregate or per-acre measures favor Asia.

Key evidence (both sides)

• The Pomeranz 2000 empirical case — wage, life-expectancy, agricultural-yield, and commercial-institution evidence for Yangzi-Europe parity.
• The Broadberry-led GDP reconstructions — explicit per-capita time series for Britain, China, India, Italy, Spain, Holland, Ottoman Empire over centuries.
• Allen’s efficiency-units wage work — showing Pomeranz’s wage-parity claim may evaporate when adjusted for skill content and human capital per worker.
• Urbanization rates — consistently higher in NW Europe from ~1400 onward.
• Silver-income and specie-flow data — showing European silver outflow to Asia (in exchange for Asian goods) consistent with Asian commercial vitality but also consistent with an early Broadberry-style divergence in per-capita production.
• Chinese primary sources (Li Bozhong and others) — documenting the 18th-century Yangzi economy in detail, used by the California school but sometimes reinterpreted by the revisionists.

Where the two narratives agree

• China’s Yangzi delta was genuinely impressive. Neither side thinks 18th-century China was backward.
• The 19th century is unambiguously divergent. By 1900, the per-capita gap between Britain and China is ~10:1; no side contests this.
• The Pomeranz comparative framework was the right framework. Even revisionists agree that comparing “Yangzi delta to England” is more appropriate than “all of China to all of Europe.”
• Coal and ghost acres mattered. Both sides agree these were significant for the 18th–19th-century British acceleration. The disagreement is whether they caused the divergence or merely amplified a divergence already underway.

What’s actually at stake

The empirical question is not just historiographic. Different timings of the divergence imply different causal candidates:

• If the California-school late-divergence story is right: the primary explananda are 18th-century contingencies — coal geography, New World access, the fiscal-military state’s emergence, the specific wage/energy ratio in Britain. Medieval and early-modern institutional, cultural, or epistemic factors are supporting conditions, not primary causes.
• If the Broadberry revisionist story is right: the primary explananda are medieval and early-modern structural factors — the medieval institutional inheritance, the Catholic Marriage Program, the Scientific Revolution’s epistemic culture, NW European state competition, the European Marriage Pattern’s demographic regime. 18th-century contingencies are the final act, not the whole play.

Most contemporary syntheses end up between the two positions: conceding that the divergence was earlier than Pomeranz allowed, but later than the strongest Broadberry readings suggest; and that both medieval structural factors and 18th-century contingencies were load-bearing. The interesting intellectual question is whether that synthesis is stable or is a stopping-place pending more evidence.

Why this meta-debate matters for the other positions

• The California-school coal-and-ghost-acres position stands or falls with the parity claim.
• The institutional-advantage, WEIRD, useful-knowledge, and demographic positions all depend on some form of “structural medieval-to-early-modern divergence” and benefit from the Broadberry reading.
• The state-competition and maritime-commercial-revolution positions work under either reading, though their weight shifts depending on how much of the causal work is already done by 1500.
• The empire/coerced-extraction and geography positions cut across the meta-debate: geography is present in both readings, and empire’s load-bearing weight is large in both.

Status

Contested, active. The Broadberry project has produced substantial new evidence in the 2010s–2020s. The California school has responded (and some of its proponents have moderated their positions in light of new data) but retained its core commitments. A fair summary of where the field currently stands: the strongest version of the California-school parity claim (full 1750 equivalence) has been eroded; the strongest version of the long-Broadberry claim (Europe ahead from 1300) is not fully consensual either; something in between is where the evidence points, and the boundaries are active research territory.