Signed in as:
filler@godaddy.com
1. Background and Source Database
The GDP share series presented in this dataset draw exclusively from the Maddison Project Database (MPD), maintained by the Groningen Growth and Development Centre (GGDC) at the University of Groningen. The project continues the work of economic historian Angus Maddison, who spent decades compiling long-run estimates of GDP per capita and population for countries across the world. Coverage in the 2023 release extends to 169 countries and aggregate regions, spanning 1 AD to 2022.
Two distinct vintages of the MPD are used, applied to non-overlapping time segments. The derivation, methodological differences, and implications of joining them are described below.
2. From GDP per Capita to GDP Shares
The Maddison Project publishes GDP per capita in internationally comparable price units and population at the country level. Aggregate GDP for any unit is obtained by multiplying the two. The share of world GDP held by any unit in year t is:
Share(i, t) = [GDP per capita(i, t) × Population(i, t)] / Σ GDP(j, t)
where the denominator sums across all countries with available observations. All values in this dataset are expressed as decimal proportions (e.g., 0.197 = 19.7% of world GDP).
3. Database Vintages and Coverage
MPD 2016 — Years 1 AD to 2006
Observations through 2006 come from MPD 2016, the legacy release. GDP is expressed in 1990 Geary-Khamis (GK) international dollars, a multilateral price index anchored to 1990 price structures. A defining feature of this vintage is its back-casting and imputation approach: for countries lacking direct historical observations, estimates are constructed by projecting available data using growth rates from adjacent periods or related economies. This ensures temporal continuity but implies that many pre-industrial data points reflect constructed estimates, not primary sources.
Pre-1820 figures rest on particularly thin empirical foundations, Maddison himself noted that only 6 to 14 countries are directly observed in this period. These shares are suitable for broad qualitative characterization of pre-modern economic structure, not for precise quantitative comparison.
MPD 2023 — Years 2008 to 2022
Observations from 2008 onward come from MPD 2023 (Bolt and Van Zanden, Journal of Economic Surveys, 2024), anchored on the 2011 ICP benchmark in GEKS purchasing power parity dollars—now the international statistical standard for multilateral comparisons. This vintage applies a source-fidelity policy: only observations present in primary historical sources are reported, with no back-casting. It also disaggregates the former Soviet Union into successor states, a distinction absent from the 2016 release.
4. Early Period: Linear Interpolation
The original Maddison data provides observations only for years 1, 1000, and 1500 within the pre-modern period. Six additional years were inserted—200, 500, 800, 1100, 1250, and 1400 AD—to give the time axis more balanced visual spacing. Values for these years were computed by simple linear interpolation between adjacent anchor points (years 200/500/800 between anchors 1 and 1000; years 1100/1250/1400 between anchors 1000 and 1500).
These interpolated points carry no independent historical evidence and were constructed solely for presentational continuity. They must not be cited as estimates of GDP shares in those centuries, nor used in any form of quantitative analysis.
5. Regional and Country Groupings
Ten series are reported. Seven are geographic or sub-regional aggregates (Western Europe, Western Offshoots, Eastern Europe and former USSR, Latin America, Asia excluding Japan, Japan, and China); two are derived composites (Developed = Western Europe + Western Offshoots + Japan; Developing = 100% minus Developed); and one is a country-level series (United States, shown separately within Western Offshoots).
The Developed aggregate corresponds to Maddison's traditional Group 1 of early-industrializing economies and is not equivalent to the IMF Advanced Economies classification, which additionally includes South Korea, Taiwan, Singapore, Hong Kong, Israel, and several Eastern European EU members. The Developing residual implicitly includes Middle East and North Africa and Sub-Saharan Africa, which are not shown as standalone series. The China series covers mainland China only; Hong Kong, Macau, and Taiwan are included in Asia (Excluding Japan). The United States series is a subset of Western Offshoots, shown separately.
6. Recommended Citations
For years 1–2006 (MPD 2016):
Bolt, Jutta, Robert Inklaar, Herman de Jong, and Jan Luiten van Zanden (2018). "Rebasing 'Maddison': New Income Comparisons and the Shape of Long-Run Economic Development." Maddison Project Working Paper 10. University of Groningen.
For years 2008–2022 (MPD 2023):
Bolt, Jutta and Jan Luiten van Zanden (2024). "Maddison Style Estimates of the Evolution of the World Economy: A New 2023 Update." Journal of Economic Surveys, 1–41. DOI: 10.1111/joes.12618.
Citations for GDP data
University of California, Los Angeles (UCLA)
North American Integration & Development (NAID) Center
COPYRIGHT © 2026 - ALL RIGHTS RESERVED | 2023-26