Spatial Political Economy

Political cleavages often have a distinct geography and politico-economic theories often have implications that can most adequately be validated at the high spatial resolution of cities or electoral constituencies. Yet applied work in the field of spatial political economy is complicated by the lack of sufficiently granular geocoded electoral and socio-economic data, which are often reported for differing administrative boundaries, and suitable cross-country operationalizations of political party classifications. The Spatial Political Economy in Europe Database (SPEED) collects (i) georeferenced electoral maps, (ii) harmonized political party names linked to several party classification systems and (iii) estimates of (per capita) GDP for all the constituencies of most European countries in the 17th release of the Constituency-Level Electoral Archive (CLEA), a repository of detailed worldwide constituency-level electoral results.

SPEED combines the information of three ancillary datasets, which are also made available separately below, namely:

Geo-Referenced European Electoral Districts (GREED): geo-referenced maps for a subset of 28 European countries covering all the electoral districts in nearly all the general elections contained in the 17th version of the CLEA Lower Chamber elections archive, considerable extending their historical coverage in CLEA's own GeoReferenced Electoral Districts Datasets (GRED). Constituency names match those of the CLEA, such that they can be easily linked.
European Political Taxonomy (EPT): a database that harmonizes the political party names for a subset of 28 European countries covering all the electoral districts in nearly all the general elections contained in the 17th version of the CLEA Lower Chamber elections archive and links them to several party classification systems.
Gridded European Economic Data (GEED): yearly gridded estimates of population and GDP for a large sample of European countries between 1800 and 2021 at a high spatial resolution of 5-arcminutes. Gridded estimates are harmonized to make them consistent with (sub)national data in several existing authoritative sources.

While the databases below are fully compatible with the 16th release of the CLEA and do not attempt to correct or extend the substantive data, I also provide a more comprehensive 'CLEA corrected' version which corrects a few apparent errors and extends the electoral data with additional electoral results from several primary and secondary sources. More information on these corrections and revisions are provided in the supplementary documentation and the relevant code in the last data section below.

Download SPEED (CLEA compatible)

Download SPEED (CLEA corrected)

Geo-Referenced European Electoral Districts (GREED)

The Geo-Referenced European Electoral Districts (GREED) database contains geo-referenced maps for nearly all the territorial constituencies for the lower chamber elections contained in the 16th release of the Constituency-Level Elections Archive (CLEA) for a subset of 28 European countries. The database currently provides 689 georeferenced maps between 1847 and 2023.

The full database can be downloaded in a single shapefile containing separate layers for each election or as a separate shapefile for each election. Shapefiles are also available separately for each country. For reference, dynamic maps visualizing the available information can be downloaded too (the country-specific maps show the current NUTS3-borders in red). The downloadables are compressed zip files containing the shapefile(s) (.shp) and a separate PDF documentation. A minimal working example of downloading the CLEA data, cleaning them and linking them to the GREED in Stata can be downloaded here.

GREED, single shapefile (Dynamic map, see also below)
GREED, separate shapefiles (Dynamic map, see also below)
Austria (Dynamic map)
Belgium (Dynamic map)
Bulgaria (Dynamic map)
Czech Republic (Dynamic map)
Denmark (Dynamic map)
(West) Germany (Dynamic map)
Estonia (Dynamic map)
Finland (Dynamic map)
France (Dynamic map)
Greece (Dynamic map)
Hungary (Dynamic map)
Iceland (Dynamic map)
Ireland (Dynamic map)
Italy (Dynamic map)
Latvia (Dynamic map)
Lithuania (Dynamic map)
Luxembourg (Dynamic map)
Netherlands (Dynamic map)
Norway (Dynamic map)
Poland (Dynamic map)
Portugal (Dynamic map)
Romania (Dynamic map)
Slovakia (Dynamic map)
Slovenia (Dynamic map)
Sweden (Dynamic map)
Switzerland (Dynamic map)
Spain (Dynamic map)
UK (Dynamic map)

Documentation

Gridded European Economic Data (GEED)

The Gridded European Economic Data (GEED) Database contains gridded estimates for population and GDP for a set of 28 European countries covering the period 1800-2023 at a high spatial resolution of 5-arcminutes. The estimates are harmonized with several authoritative (sub)national sources and are sufficiently finegrained to be aggregated to arbitrary administrative boundaries using standard spatial weighting methods.

The database is a compressed zip file containing the gridded estimates in a .csv file, a separate folder containing the shapefile of the spatial grid and a separate PDF document summarizing the data sources and the methodology. A minimal working example of importing the GEED in Stata to visualize the spatial distribution of per capita GDP in the year 2020 can be downloaded here. I also provide a complementary dataset of constituency weights that describe the share of each grid cell's surface area that belong to the CLEA-constituencies in the GREED. These spatial weights can be used to aggregate the gridded data to the administrative borders of the CLEA-constituencies.

GEED
CLEA constituency weights
Map of the spatial grid (overlaid with current European NUTS3 regions)

Documentation

Below a dynamic image of the available gridded per capita GDP estimates between 1800 and 2023.

European Political Taxonomy (EPT)

The European Political Taxonomy (EPT) harmonizes 6189 political party names in a subset of 28 European countries for the lower chamber elections contained in the 17th release of the Constituency-Level Elections Archive (CLEA) and links them to political party classifications scraped from Wikipedia, Inglehart & Norris (2019) and the PopuList. Further linkages to other well known party classification schemes could be implemented using the harmonized party names in the Party Facts database for the available political parties there.

The database is a compressed zip file containing an Excel file (.xlsx) containing the harmonized party names and the linked party classifications described in a separate PDF documentation. A minimal working example of downloading the CLEA data, cleaning them and linking them to the PTE in Stata can be downloaded here.

Documentation

Correcting and extending the CLEA

While the databases above are fully compatible with the 17th release of the CLEA and do not attempt to correct or extend the substantive data, the program below corrects a few apparent errors and extends it with additional electoral results from several primary and secondary sources. All of the corrections are explained in the documentation below. As these revisions extend the number of constituencies and the number of political parties in the database, below, I also provide extended versions of the Geo-Referenced European Electoral Districts (GREED) and the European Political Taxonomy (EPT) that encompass the newly added constituencies and political parties. Finally, I also provide an extended set of constituency weights that describe the share of each grid cell's surface area that belong to the CLEA-constituencies in the Gridded European Economic Data (GEED). These spatial weights can be used to aggregate the gridded data to the administrative borders of the CLEA-constituencies in this extended database.

Documentation