i. Central statistical agencies ii. Historical reconstructions iii. Estimates iv. Conjectures Virtually all benchmark data after the 1920s are derived indirectly from Central Statistical Agencies and therefore probably better classify as historical reconstructions. The remaining data are estimates

General references

The wage and price series shown in this chapter are taken from three sources: (A) a variety of studies on historical real wages that appeared in academic journals and books; (B) the British Colonial Blue Books (circa 1840-1912); and (C) the October Enquiries of the International Labour Organisation (1924-2008). These data were then converted into subsistence ratios, which indicate how many times the daily wage of a male unskilled construction labourer can buy the daily subsistence basket. This methodology has the advantage of providing an absolute yardstick to compare welfare across countries and time periods and, hence, is conceptually close (but not identical) to purchasing power parities. Finally, in order to fill gaps in the data, interpolations were made (D) on the basis of real wages indices from the (older) literature.

* To start, for much of the 19th century data we draw on economic histories. Much European data came from Allen's pioneering study (2001) on European wages and prices from the late Middle Ages to the First World War. Data for Istanbul came from the study by Ozmucur and Pamuk (2002), which was based on over 6 000 account books from the soup kitchens of pious foundations and the Topkapı Palace. In addition, we took data for Japan from Bassino and Ma (2005), for several Southeast Asian countries from Van der Eng and Bassino (2013), for India from Allen (2007), for China from Allen et al. (2011), for Argentina, Bolivia, Chile, Colombia, Mexico and Peru from Arroyo Abad et al. (2012), for the United States from Allen et al. (2012), and for Indonesia from De Zwart and Van Zanden (2012). These studies each draw on a variety of sources that are too extensive to discuss here.

* The Colonial Blue Books (1840-1912) contain data that were collected by the colonial administrators in the various colonies of the British Empire and sent each year to the Colonial Office in London, in response to questionnaires sent out by the latter. Frankema and Van Waijenburg (2012) worked with these data for nine British African colonies, with the earliest observation dating from 1870. We extended their series, where possible, to 1850, and added estimates for South Africa (De Zwart, 2011). In addition, we added data from several non-African colonies, especially in Oceania, Latin America and the Caribbean. Data from the Blue Books are not ideal, since price data do not always reflect retail prices (but prices for produce) and wages are not always representative for the majority of the population, but these are currently the only figures available for many of Europe's former colonial possessions.

* Since 1924, the International Labour Organisation (ILO) has conducted an annual survey, called the October Inquiry, to obtain data on wages and prices worldwide. Every year the ILO has sent two questionnaires (one relating to wages and hours of work, the other to retail prices) to national statistical agencies, which were to complete the questionnaires with the information already available to them (and thus not to conduct specific surveys in order to supply the data). Hence, while the price data are roughly consistent, the wage information returned for the various countries could differ significantly; while some reported average wages per hour from an establishment survey, others reported legislated minimum or maximum wage rates for certain occupations, and others returned minimum wage rates based on collective agreements, etc. (see Freeman and Oostendorp, 2001). These wage data thus require some form of standardisation before they can be used. Since 1983, this work has been performed by Freeman and Oostendorp (2001), Harsch and Kleinert (2011), and Oostendorp (2012). In addition, the number of countries included expanded from 15 in 1924 to over 50 in the 1950s, after which the number of countries has fluctuated (after 1983 it contains data on over 130 countries!).

* Real wage series for various countries in the 19th and 20th centuries are also available from other (older) literature (e.g. Mitchell, 2007; Williamson 1998; Scholliers and Zamagni, 1995) as well as from some more recent literature on wages and prices (e.g. Van Leeuwen, 2004; 2007). In addition, in a few cases we included special reports, such as on the period 1950-1970 in China for which little data is available (e.g. Survey on cities and counties in Guangxi, 1985; Yulin City Gazetteer, 1993). The lack of standardisation in the methodology makes direct comparisons on the basis of these data impossible. However, in order to deal with gaps in the data, those real wage series are used for interpolation in a few cases.

In 2010, the Netherlands Organisation for Scientific Research (NWO) awarded a subsidy to the Clio Infra project, of which Jan Luiten van Zanden was the main applicant and which is hosted by the International Institute of Social History (IISH). Clio Infra has set up a number of interconnected databases containing worldwide data on social, economic, and institutional indicators for the past five centuries, with special attention to the past 200 years. These indicators allow research into long-term development of worldwide economic growth and inequality.

Global inequality is one of the key problems of the contemporary world. Some countries have (recently) become wealthy, other countries have remained poor. New theoretical developments in economics - such as new institutional economics, new economic geography, and new growth theory - and the rise of global economic and social history require such processes to be studied on a worldwide scale. Clio Infra provides datasets for the most important indicators. Economic and social historians from around the world have been working together in thematic collaboratories, in order to collect and share their knowledge concerning the relevant indicators of economic performance and its causes. The collected data have been standardized, harmonized, and stored for future use. New indicators to study inequality have been developed. The datasets are accessible through the Clio Infra portal which also offers possibilities for visualization of the data. Clio Infra offers the opportunity to greatly enhance our understanding of the origins, causes and character of the process of global inequality.