CEPII, Recherche et Expertise sur l'economie mondiale


EconMap is the database developed by the CEPII in 2010 to picture the world economy in the long term.

Based on a three-factor production function of labour, capital and energy, plus two forms of technological change, we propose various long-run growth scenarios for 170 countries at the 2100 horizon, relying on the model MaGE (Macroeconometrics of the Global Economy). Our model is fitted with United Nations, International Labour Office and Energy Information Administration projections and econometric estimations of (i) capital accumulation, (ii) saving rates, (iii) relationship between saving rates and investment rates, (iv) education, (v) female participation, and (vi) technological progress (which includes energy and total factor productivity).

The data files cover GDP at constant and variable relative prices, together with the production factors and technological change, for each of the 170 countries, from 1980 to 2100.

Reference document to cite: Fontagné, L., Perego, E., and Santoni, G. (2021), MaGE 3.1: Long-Term Macroeconomic Projections of the World Economy, CEPII Working paper 2021-12.

Fouré, J., Bénassy-Quéré, A., and Fontagné, L. (2013), Modelling the world economy at the 2050 horizon, Economics of Transition 21 (4): 617-654.

Person in charge & contact: Erica Perego & Gianluca Santoni, econmapcepii.fr

Licence: Etalab 2.0


EconMap 3.1

Documentation of the SSP scenarios: SSP_scenarios_2.4.pdf
File Description
EconMap_3.1.zip Results | Central scenario and five SSP scenarios (CSV files)
Raw_Data_EconMap_3_1.zip Raw data
Reproduction_EconMap_3_1.zip Reproduction files (STATA files + Raw data)


Our projections present four charateristics.

First, we account for energy constraints by including its consumption in the production function and by taking account of rents accruing to oil exporting countries.

Second, we estimate a non-unitary relationship between savings and investment, departing from the assumptions of closed economy and full capital mobility.

Third, we model female participation rates consistently with education catch-up.

Finally, we disentangle real gross domestic product (GDP) growth rates from relative price effects through a consistent Balassa-Samuelson effect.

Older versions
Release Data Files
2.4 Historical data and projections 2.4.xls
2.3 Historical data and projections 2.3.xls
2.2 Historical data and projections 2.2.xls
2.1 Historical data and projections 2.1.xls
2.0 Historical data and projections 2.0.xls
1.0 Historical data GDP
Production factors
  Projections GDP
Production factors
Release Date Description
3.1 November 2021 Updated database and estimations. Source code and data available
2.4 January 2016 Introducing SSP scenarios and 2100 time horizon. Updated energy price projections. Introducing smoothing of non-annual projections. Source code available.
2.3 October 2014 20 new countries available. Updated source data. Revised methodology on energy productivity. Source code available.
2.2 September 2013 Updated parametrization (in line with published article) and new scenario capabilities (price of investment, skills of migrants). Source code available.
2.1 September 2012 Updated methodology and GDP data. Source code available.
2.0 February 2012 New TFP model, female labor market participation rate, correction for oil production bias, updated data and new countries. A new working paper is available.
1.1 Limited access New education data and new countries.
1.0 December 2010 First public release.