Converted from a Word document
The paper presents to the research community a new digital resource for research on the social and economic development of Russia and its regions over the last two centuries (1800-2000): The Electronic Repository of Russian Historical Statistics (https://ristat.org). It makes available, free of charge, standardised data-sets on demographics, factors of production (labour, land, capital) and output (agriculture, industry, services) for five historical cross-sections (1795, 1858, 1897, 1959, 2002). Data cover the territory of the modern-day Russian Federation, and are consistently available with a regional break-down at level of individual provinces (
gubernii pre-1917/
oblasti post-1917).
The Electronic Repository of Russian Historical Statistics was created by a Russian-Dutch team of researchers, led by Gijs Kessler of the International Institute of Social History (IISH, Amsterdam) and Andrei Markevich of the New Economic School (NES, Moscow). It became fully operational in 2020.
The project was intended to serve a twofold aim: first, to mine the wealth of Russian historical statistics and make these data available to the global research community, and, secondly, by making available data with a regional break-down, to provide an impulse to the development of research agendas on Russian history with regional or transnational perspectives. The paper provides a brief outline of the data made available and the methodology adhered to in the data-mining project, and briefly presents the results of three use-cases of the data in research.
For a country at its stage of development, and even in comparison to many developed nations, Russia can boost of extremely good historical statistical data. Regions regularly reported back to the centre and since at least the early nineteenth century these reports were accompanied by detailed statistical appendices. The shift to central planning in the Soviet Union during the twentieth century, furthermore, entailed the spread of even more elaborate record keeping.
Yet, despite this riches and their potential, these data have been used very little by researchers. Archives in Russia have long been closed, and only became generally available in the early 1990s, following the collapse of the Soviet Union. In conjunction with language barriers, this has played a role in preventing them from being picked up by the international scholarly community in disciplines to which the excellent Russian data could speak, like development economics, and global history.
The Electronic Repository of Russian Historical Statistics aims to overcome this problem by making available systematic data on a number of key indicators of social and economic development on a bilingual (Russian & English), easy to use, online platform, accompanied by thorough and extensive documentation, and coded using internationally accepted classifications of economic activity (NACE), occupational structure (PSTI) land-use (FAO), and capital assets (CLEMS). Full translation of the data makes them accessible to non-specialists, the use of international classifications to non-historians like social scientists and economists, and dissemination through an easy-to-use platform is intended to make reliable historical data available to public debates, as well as for educational purposes.
Due to the country's vast size, national level data are woefully inept for systematic analysis of its social and economic development. This is generally acknowledged, but where accessibility of data at the national level is a problem already, this holds true even more at the regional level. Studies on Russia that take a regional perspective face the formidable barrier that even some of the most basic quantitative parameters, like population figures, are difficult to come by and require significant outlays of time and resources to be made ready for use. It is precisely by targeting this barrier that the Electronic Repository aims to facilitate regional and transnational research perspectives. By making available some of the basics, research projects can build on this foundation, and concentrate their efforts on the specific topics or issues they aim to investigate.
The data gathering programme reflects this philosophy. To have a solid base in population data, the five historical cross-sections were chosen in line with the availability of population censuses (1897, 1959 and 2002) and proto-censuses (1795 & 1858). Data were then added on for the other topics of the grid (labour, output, land and capital) according to availability. Where possible we tried to get data for the exact benchmark year, if not possible, for the immediately adjacent years.
Data were sourced from published and unpublished sources, after a survey of available alternatives. The choices made, were laid down in the extensive documentation accompanying the data. Documentation also discusses the data gathering procedures underlying the sources we chose to use for the Repository, as well as the potential biases they might contain.
Data were systematised and standardised to improve usability in as far as this was possible without making assumptions that do not follow directly from the source. In other words, we corrected typos, standardised spelling and unified categories only in cases where there is no reasonable doubt as to the validity of these decisions. Data are therefore not published in 'raw' format, but in unified 'historical categories' from the source.
To enhance comparability of the data across time and space we have added a second layer of interpretation to the data – a 'modern classification' in internationally accepted taxonomies and systems of classification. Using the custom query tool of the site data can be accessed in either historical or modern categories. In order to see both sets of categories in conjunction data can also be accessed in the less user-friendly, but comprehensive data-base format through the file catalogue section of the portal.
Because most topics in the database contain numerous subtopics the overall number of indicators in the database reaches over four thousand. With up to a hundred regions per cross-section this results in a database of over half a million individual observations. The majority of these data comes from archival sources and is therefore made available to research for the very first time.
Particular attention was paid to the development of a user-friendly web-interface, accessible to specialists and non-specialists alike. Users requiring more advanced querying tools can easily download the data in full data-base format through the file catalogue and upload them into their own software tools or querying environment. All data from the Electronic Repository are free to use and re-use, as long as proper reference is provided.
In its most elementary form the Electronic Repository of Russian Historical Statistics can simply be used as an online reference tool. In its more advanced uses data can be used in research to analyse the geospatial variation of selected indicators for particular years or across time. The paper provides examples of three such use-cases.
Andrei Markevich drew on the data in the repository for his reconstruction of 1897 gross regional products for the Russian Empire (Markevich 2019). His analysis of the relative impacts of geographical and institutional factors on economic development highlights substantial heterogeneity within a middle-income country. Both natural advantages – sea access, mineral resources, land abundance – and institutions, in particular the legacy of serfdom, are found to account for the observed variation.
Gijs Kessler and Timur Valetov use data from the Electronic Repository in their study of occupational change in the Russian Empire and the Soviet Union between 1897 and 2002 (Kessler / Valetov forthcoming). It finds that the tertiary sector is the largest in terms of its share of the workforce throughout the period under study, but that there is important regional variation in this, particularly for the early cross-section of 1897.
A first example, finally, of the data from the repository being used for cross-country comparison, is provided in a recent volume on the history of regional industrialisation, in which regional variation in secondary sector employment across Eastern Eurasia is traced in a series of maps, which rely on the repository for the Russian data (Van Leeuwen et al. 2021).