FAND vs CWoN — comparative analysis

How FAND's national balance sheets diverge from World Bank Changing Wealth of Nations.

The narrated Appendix E walkthrough video is still in production (source deck not yet rendered through HeyGen). The comparative charts below are ready.

John pioneered the Changing Wealth of Nations (CWoN) at the World Bank as part of MEP 1995. FAND is the continuation — what MEP 1995 would look like if you could rebuild it with modern data, modern compute, and the lessons of three decades watching what worked and what didn’t.

Where the two methods disagree

Side-by-side comparison of the CWoN and FAND deflation approaches — CWoN strips money out with component-specific chained deflators and treats human capital as the residual that closes the identity. FAND deflates every component by a single Atlas GNI ratio and lets only the value (V) series travel through the Who-What-Where-When hierarchy — so relative prices between components stay visible instead of being absorbed into an implicit claim.

Comprehensive Wealth, ten countries, 1950–2025

Ten small-multiple line charts of Comprehensive Wealth per capita, FAND vs CWoN, log scale, 1950-2025, for the United States, Japan, Germany, United Kingdom, France, China, India, Indonesia, Brazil, and Nigeria — The starkest divergence in the whole comparison. FAND (solid) and CWoN (dashed) trace similar shapes but different levels almost everywhere CWoN has coverage — FAND's top-down capitalized-GNI method and CWoN's bottom-up PA+NC+HC-with-no-residual method are answering closely related but not identical questions.

Natural Capital, ten countries

Ten small-multiple line charts of Natural Capital per capita, FAND vs CWoN, log scale, 1950-2025, for the same ten countries — FAND preserves market commodity prices for fossil fuels, minerals, fisheries, agricultural land, and forest timber/cover; CWoN's deflators smooth much of that price signal away — visible as the gap between the solid and dashed lines widening in resource-exporting economies like Nigeria and Brazil.

What’s different — the short version

FAND’s balance-sheet identity is closed. CW = PA + NC + RW; RW = HR + SI, verified to $0.000000. The identity holds at every scale.
FAND measures substantive rights (HR + SR) explicitly. CWoN treats human capital as residual; FAND derives it from a clean OLS production function (ln(RW_pc) = α + β₁·MYS + β₂·min(LE,65) + β₃·t, with N=10,031 country-years, R²=0.497).
Cross-border is per-side, not net. CBA (gross foreign assets) and CBL (gross foreign liabilities) are independent stocks. No NIIP netting — the gross flows reveal exposure that nets to nothing.
The IS-Branch Principle. Institutional infrastructure that conventional methodologies assume but don’t measure (rule of law, trust in courts, institutional capacity). FAND measures it inside SR.

The balance sheet equation — the FAND identity these charts are computed from.
How we grade what we publish — the badge system behind this page’s Experimental rating.
Global Leaves crosstab — the subnational density × wealth finding this comparison sits alongside.

What’s here at launch

The Appendix E narrated video (pending — source deck not yet rendered through HeyGen)
The comparative analysis charts (FAND vs CWoN for selected countries)
Pointer to the full Appendix E working paper in /deep-dive/