AI ClaimsRapid review

Angus Deaton · 2013

The Great Escape

Health, Wealth, and the Origins of Inequality

Cover via Open Library

Rough AI truth score

83/100

Deaton's long-run picture of extraordinary but uneven gains in income, health, and longevity is robust. He is also persuasive that progress creates inequalities as innovations diffuse, though his skepticism of foreign aid is stronger than the heterogeneous program-level evidence warrants.

Based on three central claims · high confidence

The three claims

01Supported

Humanity has achieved historically unprecedented gains in income, health, and longevity over the last few centuries.

Reconstructed income, mortality, height, and life-expectancy series show enormous long-run improvements despite measurement limits and reversals. Progress is highly uneven across regions and groups, and aggregate gains do not erase deprivation or distributional conflict.

02Mostly supported

The same innovations and growth processes that improve lives initially create inequality because escape occurs at different times.

Health technologies and economic transformations commonly diffuse unevenly, creating temporary or persistent gaps between early and late beneficiaries. Political power, discrimination, inheritance, geography, and policy also generate inequality, so diffusion timing is one mechanism rather than a general justification.

03Mostly supported

Conventional foreign aid often weakens accountability and is not a dependable engine of long-run development.

Aid can fragment institutions, distort incentives, and support unaccountable governments, and macro growth estimates are uncertain. Vaccination, disease control, humanitarian relief, and well-designed programs also deliver clear gains, making composition and governance more informative than a blanket verdict.

Other claims worth checking
  • Public health knowledge can matter more than income for mortality decline.
  • International migration can spread the gains from development.

What this number means. It is an AI-generated first-pass judgment of three central factual or causal claims—not a rating, exhaustive fact-check, or human peer review. Claim credits are 100% for supported, 75% for mostly supported, 50% for mixed, and 25% for weak, then averaged and rounded. Lower confidence means the score should move more as better evidence arrives.

Method three-central-claims/0.1.0 · checked 2026-09-01 · 3/3 selected claims assessed · method and source audit