Data/GDP_data/README.md

GDP data for 231 nations and territories

This folder contains economic data for 231 nations and territories for the period 1960–2023. Real GDP per capita and real GDP growth rates are complete for the period 1990–2023, and extend even further back in many cases. The dataset contains the following variables:

• Nominal GDP
• Real GDP, in 2015 USD
• Nominal GDP/capita
• Real GDP/capita, in 2015 USD
• GDP real annual growth (in %)
• GDP deflator
• Exchange rate

Main sources

Data is mostly taken from four sources:

• The World Bank
• UNCTAD
• The IMF World Economic Outlook Database
• The Maddison Project

Since we need complete data on real GDP/capita and real growth rates for the years 1990–2023, we fill gaps by extrapolating available data back using the real growth rate, where available. Where this is impossible, we use national statistics and estimates from reliable sources to infer gaps; where no data is available at all, we extrapolate using neighbouring or similar countries.

World Bank

The World Bank provides data on nominal and real GDP and GDP per capita, as well as growth rates, the GDP deflator, and the exchange rate.

IMF

The IMF provides GDP/capita in current USD, current PPP, and 2021 PPP, as well as GDP at current prices and GDP % change in their World Economic Outlook dataset.

Maddison Project

The Maddison Project provides historical GDP per capita data in const. international 2011 USD.

UNCTAD

UNCTAD provides real GDP per capita in 2015 USD.

UN DESA

UN DESA provides some nominal GDP per capita figures here

Additional Sources

The above datasets allow us to calculate real growth rates for many missing countries and territories, and extrapolate values backwards or forwards to estimate GDP/capita. However, many territories still lack individual values, which we estimate from national statistics (where available), or infer by extrapolating back based on data (e.g. deflators) from neighbouring countries.

Caribbean Netherlands

UN DESA supplies data for the Former Antilles in nominal terms from 1970–2009. These broadly correspond with the WB figures for Curacao, the largest of the islands (Aruba became an independent constituent in 1986). Statistics Netherlands gives data for the BES islands from 2012 on, also in nominal terms. We convert and fill the gaps using the deflator for neighbouring Aruba (which uses base year 2013, and is also a constituent part of the Netherlands).

French overseas territories

Data from Statista. We extrapolate back to 1989 and forward to 2023 using average growth rates.

Small Pacific Islands

Source: Pacific Community. This gives nominal GDP from 2025–2023, as well as country-specific inflation rates until 2022, which we use as a proxy for the GDP deflator. Additionally, we infer missing values using average growth rates from neighbouring countries:

• Niue, Wallis-Futuna, Tokelau: deflated using the inflation rate as a proxy
• American Samoa, Niue, Wallis-Futuna: extrapolated back using the average growth rates of New Caledonia, Cook Islands, and Fiji
• Tokelau: extrpolated back using the growth rate of Fiji

Guam, Northern Marianas, U.S. Virgin Islands.

These are all U.S. overseas territories. We extrapolate Guam and the Marianas back from Puerto Rico, and the U.S. Virgin Islands using an average of Puerto Rico, Antigua, and St. Kitts and Nevis.

Serbia and Montenegro

We can extrapolate back using the growth rates given by the Maddison Project.

South Sudan

The World Bank data for South Sudan is clearly erroneous. Since we have stocks from South Sudan from before its founding (2011), we can project the GDP of South Sudan back using GDP growth of Sudan and adjusting to match the 2009 value.

East Timor

The growth rate for 1990 is missing; this report by the Australian government suggests an average 10% economic growth from 1987-1997, due to heavy investment in construction. Given the data we already have, this means GDP growth from 1987–1990 must have averaged 11.7%.

Western Sahara

We use GDP/capita from Morocco, since ESH is under occupation.

Saint Pierre-Miquelon

We use GDP/capita figures from Canada.

Eritrea

We extrapolate past 1990 using data from Ethiopia.

Falklands, Isle of Man, Gibraltar, St Helena

• Isle of Man: Government figures estimate real GDP growth at 2.9% in 2022, and 1.5% in 2023.
• Falklands: Some data is available from the Falklands government and historical archives. These datapoints match the population-scaled GDP of the Isle of Man very well, so we will the gaps using data from the Isle of Man.
• Gibraltar Government figures give GDP in nominal terms. We use the deflator for GBR. Missing values (pre-2009) are extrapolated back using Isle of Man data.
• St Helena: Nominal St Helena data is available from the government website. We extrapolate back using data from Brazil.

Faroe Islands

We use the Danish deflator to extrapolate back.