{"id": 1009185, "name": "GDP per capita", "unit": "international-$ in 2011 prices", "createdAt": "2025-01-29T15:22:47.000Z", "updatedAt": "2026-03-05T16:54:55.000Z", "coverage": "", "timespan": "1870-2021", "datasetId": 6918, "shortUnit": "$", "columnOrder": 0, "shortName": "gdp_per_capita", "catalogPath": "grapher/wpf/2025-01-17/famines_by_regime_gdp_population/famines#gdp_per_capita", "descriptionShort": "Average economic output per person in a country or region per year. This data is adjusted for inflation and differences in living costs between countries.", "descriptionFromProducer": "Real GDP per capita in 2011$\n\nIn two ways, this analysis leads to departures from the original Maddison approach and closer to the multiple benchmark approach as developed by the PWT. There is, to begin with, no doubt that the 2011 PPPs and the related estimates of GDP per capita reflect the relative levels of GDP per capita in the world economy today better than the combination of the 1990 benchmark and growth rates of GDP per capita according to national accounts. This information should be taken into account. At the same time, the underlying rule within the current Maddison Database is that economic growth rates of countries in the dataset should be identical or as close as possible to growth rates according to the national accounts (which is also the case for the pre 1990 period). For the post-1990 period we therefore decided to integrate the 2011 benchmarks by adapting the growth rates of GDP per capita in the period 1990\u20132011 to align the two (1990 and 2011) benchmarks. We estimated the difference between the combination of the 1990 benchmark and the growth rates of GDP (per capita) between 1990 and 2011 according to the national accounts, and annual growth rate from the 1990 benchmark to the 2011 benchmark. This difference is then evenly distributed to the growth rate of GDP per capita between 1990 and 2011; in other words, we added a country specific correction (constant for all years between 1990 and 2011) to the annual national account rate of growth to connect the 1990 benchmark to the 2011 benchmark. Growth after 2011 is, in the current update, exclusively based on the growth rates of GDP per capita according to national accounts.\n\nWe also use the collected set of historical benchmark estimates to fine tune the dataset for the pre-1940 period, but only in those cases where the quality of the benchmark was high and there were multiple benchmarks to support a revision. The most important correction concerns the US/UK comparison. The conventional picture, based on the original 1990 Maddison estimates, indicated that the US overtook the UK as the world leader in the early years of the 20th century. This finding was first criticized by Ward and Devereux (2003), who argued, based on alternative measures of PPP-adjusted benchmarks between 1870 and 1930, that the United States was already leading the United Kingdom in terms of GDP per capita in the 1870s. This conclusion was criticized by Broadberry (2003).\n\nNew evidence, however, suggests a more complex picture: in the 18th century, real incomes in the US (settler colonies only, not including indigenous populations) were probably higher than those in the UK (Lindert & Williamson, 2016a). Until about 1870, growth was both exten- sive (incorporating newly settled territory) and intensive (considering the growth of cities and industry at the east coast), but on balance, the US may\u2014in terms of real income\u2014have lagged behind the UK. After 1870, intensive growth becomes more important, and the US slowly gets the upper hand. This pattern is consistent with direct benchmark comparison of the income of both countries for the period 1907\u20131909 (Woltjer, 2015). This shows that GDP per capita for the United States in those years was 26% higher than in the United Kingdom. We have used Woltjer\u2019s (2015) benchmark to correct the GDP series of the two countries. Projecting this benchmark into the 19th century with the series of GDP per capita of both countries results in the two countries achieving parity in 1880. This is close to Prados de la Escosura\u2019s conjecture based on his short- cut method (Prados de la Escosura, 2000), and even closer to the Lindert and Williamson (2016a) results.\n\nChanging the US/UK ratio on the basis of the new research by Woltjer (2015) raises the question of which country\u2019s GDP estimates should be adapted. In the current PWT approach, the growth of GDP per capita in the United States is the anchor for the entire system. For the 19th century, however, it is more logical to take the United Kingdom as the anchor, because it was the productivity leader, and because most research focused on creating historical benchmarks takes the United Kingdom as reference point. We have therefore adapted the UK series for the period 1908\u20131950 to fit the 1907\u201309 (Woltjer, 2015) benchmark in our view the best available benchmark for this period. The reason is that there are doubts about the accuracy of price changes and deflators for the UK for the period 1908\u20131950, given that it was characterized by two significant waves of inflation (during the two World Wars) and by large swings in relative prices and exchange rates (as documented in the detailed analysis by Stohr (2016) for Switzerland). Future research will have to assess whether this choice is justified.\n\nThis new version of the MPD extends GDP per capita series to 2022 and includes all new historical estimates of GDP per capita over time that have become available since the 2013 update (Bolt & Van Zanden, 2014). As new work on historical national accounts appears regularly, a frequent update to include new work is important, as it provides us with new insights in long-term global development. Furthermore, we have incorporated all available annual estimates for the pre-1820 period instead of estimates per half-century, as was usual in the previous datasets.\n\nA general \u201cwarning\u201d is in place here. For the period before 1900 (and for parts of the world such as Sub-Saharan Africa before 1950), there are no official statistics that fully cover the various components of GDP; and the more one moves back in time, the more a scarcity of basic statistics becomes a problem for scholars trying to chart the development of real income and output. The statistics needed for reconstructing GDP are often produced in parallel to the process of state formation, but even large bureaucratic states such as China or the Ottoman Empire only rarely collected the data that allow us to estimate levels of output and income. Much of the work on pre-industrial economies makes use of the \u201cindirect method,\u201d which links data on real wages and levels of urbanization to estimates of GDP per capita. But a few countries, during the Medieval and Early Modern periods, did collect the (tax) data to estimate GDP in the \u201cproper\u201d way (Tuscany in 1427, Holland in 1514, and England in 1086). These benchmarks, in combination with the many \u201cindirect\u201d estimates, allow us to create a tapestry of estimates which becomes\u2014with the increase of the number of studies\u2014increasingly robust. Where the original Maddison dataset included 158 observations for the pre 1820 period, the current 2023 MPD includes close to 2800 data points for the preindustrial period.\n\nFor the recent period, the most important new work is Harry Wu\u2019s reconstruction of Chinese economic growth since 1950. Inspired by Maddison, Wu\u2019s model produces state of the art estimates of GDP and its components for this important modern economy (Wu, 2014). Given the large role China plays in any reconstruction of global inequality, this is a major addition to the dataset. Moreover, as we will see below, Wu\u2019s revised estimates of annual growth are generally lower than the official estimates. Lower growth rates between 1952 and the present, however, substantially increases the estimates of the absolute level of Chinese GDP in the 1950s (given the fact that the absolute level is determined by a benchmark in 1990 or 2011). This helps to solve a problem that arises in switching from the 1990 to the 2011 benchmark: namely, that when using the official growth estimates, the estimated levels of GDP per capita between 1890 and the early 1950s are substantially below subsistence level, and therefore too low. Including the new series as constructed by Wu (2014) gives a much more plausible long-run series for China.\n\nOften, studies producing very early per capita GDP estimates\u2014particularly work on the early modern period (1500\u20131800)\u2014make use of indirect methods. The \u201cmodel\u201d or framework for making such estimates is based on the relationship between real wages, the demand for foodstuffs, and agricultural output (\u00c1lvarez-Nogal & De La Escosura, 2013; Malanima, 2011 among others). This model has now also been applied to Poland (Malinowski & van Zanden, 2017), Spanish America (Abad & van Zanden, 2016), and France (Ridolfi, 2017; Ridolfi & Nuvolari, 2021). In this update, we have now included annual estimates of GDP per capita in the period before 1800 for these countries.\n\nFor some countries during a period before 1870 or 1800, we only have series of a certain province or similar entity. The British series links to estimates for only England for the period before 1700; the series for the Netherlands links to estimates for only Holland for the period before 1807. The switch from the national to the \u201cpartial\u201d series is clearly indicated in the dataset, and the \u201ccorrection\u201d in terms of GDP per capita is indicated.\n\nFinally, we have extended the national income estimates up to 2022 for all countries in the database. For this we use various sources. The most important is the Total Economy Database (TED) published by the Conference Board, which includes GDP per capita estimates for a large majority of the countries included in the Maddison Project Database. The 2013 MPD update took the same approach (Bolt & van Zanden, 2014). For countries unavailable through TED, we relied on UN national accounts estimates to extend the GDP per capita series. To extend the population estimates up to 2022, we used the TED and the US Census Bureau\u2019s International Database 2022.18 The TED revised their China estimates from 1950 onwards based on Wu (2014). As discussed above, we also included Wu (2014)\u2019s new estimates in this update. Finally, we have extended the series for the former Czechoslovakia, the former Soviet Union, and former Yugoslavia, based on GDP and population data for their successor states.", "descriptionProcessing": "When GDP data was missing for certain years but nearby estimates were available, we used the closest or most relevant data. For example, during Cuba's famine from 1895 to 1898, we used the GDP from 1892. For China\u2019s famine from 1876 to 1879, we used the average GDP from 1870 to 1887. In special cases like Russia and Kazakhstan (1932 to 1934), we used the USSR\u2019s average GDP from 1940 to 1946. For recent gaps, we used 2022 data for countries like the Central African Republic, Ethiopia, and Syria.", "type": "float", "grapherConfigIdETL": "0194b2a8-6337-7076-a673-6a0e81a91ea7", "dataChecksum": "14944508969720054795", "metadataChecksum": "-5212437854408800761", "datasetName": "Deaths from famines by regime, GDP and population growth", "updatePeriodDays": 365, "datasetVersion": "2025-01-17", "nonRedistributable": false, "display": {"name": "GDP per capita", "unit": "international-$ in 2011 prices", "shortUnit": "$", "tolerance": 5, "numDecimalPlaces": 0, "entityAnnotationsMap": "Western offshoots (Maddison): United States, Canada, Australia and New Zealand"}, "schemaVersion": 2, "processingLevel": "minor", "presentation": {"titlePublic": "GDP per capita", "titleVariant": "Long-run data in constant international-$", "attributionShort": "Maddison Project Database", "topicTagsLinks": ["Famines"], "faqs": [{"gdocId": "1gGburArxglFdHXeTLotFW4TOOLoeRq5XW6UfAdKtaAw", "fragmentId": "poverty-international-dollars"}]}, "descriptionKey": ["GDP per capita is a comprehensive measure of people's average income. It helps compare income levels across countries and track how they change over time. It is especially useful for understanding trends in economic growth and living standards.", "GDP per capita is calculated as the value of all final goods and services produced each year in a country (the [gross domestic product](#dod:gdp)), divided by the population. It represents the average economic output per person.", "This indicator shows the large inequality between people in different countries. In the poorest countries, average incomes are below $1,000 per year; in rich countries, they are more than 50 times higher.", "This data comes from the [Maddison Project Database](https://www.rug.nl/ggdc/historicaldevelopment/maddison/releases/maddison-project-database-2023), which provides GDP per capita estimates for the _very long run_. Some country series extend as far back as 1 CE, and regional estimates start in 1820.", "This work builds on the efforts of many researchers who have carefully reconstructed historical data on economic growth and population for individual countries. You can find the full list of sources in [the original dataset](https://dataverse.nl/api/access/datafile/421302).", "This data is expressed in constant international dollars at 2011 prices to adjust for inflation and differences in living costs between countries. Read more in our article, [What are international dollars?](https://ourworldindata.org/international-dollars)", "This dataset combines multiple [purchasing power parity](#dod:purchasing-power-parity) (PPP) benchmarks to ensure historical consistency and comparability over time. 1990 PPPs are used up to 1990, and 2011 PPPs are used from 2011 onward. For the years in between, they adjust the series to smoothly connect the two benchmarks. This approach preserves consistency with the original long-run estimates calculated by Angus Maddison.", "Time series for former countries and territories are calculated forward by estimating values based on their last official borders.", "For more frequently updated estimates since 1990, explore our chart of GDP per capita from the [World Bank](https://ourworldindata.org/grapher/gdp-per-capita-worldbank)."], "dimensions": {"years": {"values": [{"id": 1915}, {"id": 1974}, {"id": 1876}, {"id": 1975}, {"id": 2020}, {"id": 1897}, {"id": 1920}, {"id": 1928}, {"id": 1934}, {"id": 1937}, {"id": 1958}, {"id": 1895}, {"id": 1998}, {"id": 1973}, {"id": 1983}, {"id": 2021}, {"id": 1917}, {"id": 1941}, {"id": 1888}, {"id": 1896}, {"id": 1899}, {"id": 1942}, {"id": 1981}, {"id": 1968}, {"id": 2016}, {"id": 1995}, {"id": 1894}, {"id": 1870}, {"id": 1940}, {"id": 1891}, {"id": 1918}, {"id": 1939}, {"id": 1984}, {"id": 1992}, {"id": 2003}, {"id": 2012}, {"id": 1931}, {"id": 1921}, {"id": 1946}, {"id": 1944}]}, "entities": {"values": [{"id": 372649, "name": "Austria-Hungary 1915-1918", "code": null}, {"id": 370693, "name": "Bangladesh 1974", "code": null}, {"id": 370650, "name": "Brazil 1876-1879", "code": null}, {"id": 370681, "name": "Cambodia 1975-1979", "code": null}, {"id": 370668, "name": "Central African Republic 2020-2023", "code": null}, {"id": 370690, "name": "China 1876-1879", "code": null}, {"id": 370709, "name": "China 1897-1901", "code": null}, {"id": 370675, "name": "China 1920-1921", "code": null}, {"id": 370696, "name": "China 1928-1930", "code": null}, {"id": 371320, "name": "China 1934-1936", "code": null}, {"id": 370713, "name": "China 1937-1945", "code": null}, {"id": 370666, "name": "China 1958-1962", "code": null}, {"id": 370697, "name": "China 1975", "code": null}, {"id": 370669, "name": "Cuba 1895-1898", "code": null}, {"id": 370700, "name": "Democratic Republic of Congo 1998-2007", "code": null}, {"id": 370703, "name": "Ethiopia 1958", "code": null}, {"id": 370655, "name": "Ethiopia 1973", "code": null}, {"id": 370682, "name": "Ethiopia 1983-1985", "code": null}, {"id": 370640, "name": "Ethiopia 2021-2023", "code": null}, {"id": 370694, "name": "Germany 1917-1919", "code": null}, {"id": 371324, "name": "Greece 1941-1944", "code": null}, {"id": 370663, "name": "India 1876-1878", "code": null}, {"id": 370685, "name": "India 1888-1889", "code": null}, {"id": 370710, "name": "India 1896-1897", "code": null}, {"id": 370672, "name": "India 1899-1901", "code": null}, {"id": 370716, "name": "Indonesia 1942-1945", "code": null}, {"id": 371316, "name": "Japan 1941-1945", "code": null}, {"id": 370662, "name": "Mozambique 1981-1985", "code": null}, {"id": 371311, "name": "Nigeria 1968-1970", "code": null}, {"id": 370702, "name": "Nigeria 2016-2019", "code": null}, {"id": 370705, "name": "North Korea 1995-1997", "code": null}, {"id": 370660, "name": "Ottoman Empire 1894-1896", "code": null}, {"id": 370652, "name": "Persia 1870-1872", "code": null}, {"id": 370643, "name": "Persia 1917-1919", "code": null}, {"id": 370683, "name": "Philippines 1899-1902", "code": null}, {"id": 371319, "name": "Poland (ghettos and concentration camps) 1940-1945", "code": null}, {"id": 370673, "name": "Poland 1940-1945", "code": null}, {"id": 371323, "name": "Russia 1891-1895", "code": null}, {"id": 371310, "name": "Russia 1918-1920", "code": null}, {"id": 370651, "name": "Russia 1941-1944", "code": null}, {"id": 370679, "name": "Spain 1939-1942", "code": null}, {"id": 370706, "name": "Sudan 1984-1985", "code": null}, {"id": 370699, "name": "Sudan 1992-1994", "code": null}, {"id": 370654, "name": "Sudan 2003-2005", "code": null}, {"id": 370677, "name": "Syria 2012-2023", "code": null}, {"id": 371314, "name": "USSR (Kazakhstan) 1931-1933", "code": null}, {"id": 371327, "name": "USSR (Southern Russia & Ukraine) 1921-1922", "code": null}, {"id": 371325, "name": "USSR (Southern Russia) 1931-1933", "code": null}, {"id": 371312, "name": "USSR 1939-1945", "code": null}, {"id": 371318, "name": "USSR 1940-1945", "code": null}, {"id": 371326, "name": "USSR 1941-1947", "code": null}, {"id": 370707, "name": "USSR 1946-1947", "code": null}, {"id": 370691, "name": "Uganda 2003-2006", "code": null}, {"id": 371322, "name": "Ukraine 1931-1933", "code": null}, {"id": 370656, "name": "Vietnam 1944-1945", "code": null}, {"id": 370661, "name": "Yemen 2016-2021", "code": null}]}}, "origins": [{"id": 7167, "title": "Maddison Project Database", "description": "The Maddison Project Database provides information on comparative economic growth and income levels over the very long run. The 2023 version of this database covers 169 countries and the period up to 2022. The new estimates are presented and discussed in Bolt and Van Zanden (2024), \"Maddison style estimates of the evolution of the world economy: A new 2023 update\", Journal of Economic Surveys, 1\u201341.", "producer": "Bolt and van Zanden", "citationFull": "- 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\u201341. DOI: 10.1111/joes.12618.\n- The Maddison Project Database builds on the efforts of many researchers who have carefully reconstructed historical data on economic growth and population for individual countries. You can find the full list of sources in [the original dataset](https://dataverse.nl/api/access/datafile/421302).", "attribution": "Bolt and van Zanden \u2013 Maddison Project Database 2023", "attributionShort": "Maddison Project Database", "versionProducer": "2023", "urlMain": "https://www.rug.nl/ggdc/historicaldevelopment/maddison/releases/maddison-project-database-2023", "urlDownload": "https://dataverse.nl/api/access/datafile/421302", "dateAccessed": "2024-04-26", "datePublished": "2024-04-26", "license": {"url": "https://www.rug.nl/ggdc/historicaldevelopment/maddison/releases/maddison-project-database-2023", "name": "CC BY 4.0"}}]}