geoelections Correspond
Voter turnout worldwide since 1945
The following study is limited to countries with more than 150,000 inhabitants (in 2024), practicing multiparty politics, at least universal male suffrage
(except for restrictions indicated in the fifth paragraph), and ensuring a minimum of freedoms to present oneself to the voters.
Single-party regimes, or elections boycotted by the entire opposition, or seriously disrupted by the activism of any camp are not included.
The first round of presidential and legislative elections since 1945 is taken into account, by country or overseas territory, according to their five-year average.
The parliamentary elections of future independent countries are taken into account from the moment they acquired a very large internal autonomy;
these elections generally took place with a view to or even as preparatory to independence. French overseas territories, departmentalized in 1946, are taken into account from 1945,
the other territories are taken into account after 1958. Generally speaking, the colonies are not, because they generally did not benefit from universal suffrage.
To calculate participation, the potential electoral population is used and not the number of registered voters, because of certain American states which do not require this procedure,
but also because of the lack of rigor in the maintenance of electoral lists in too many countries, particularly in those which have recently converted to the representative system.
The voting population is defined, according to the rules in force at the time of the election, by citizenship (required almost everywhere; only Chile,
Ecuador and Uruguay grant the right to vote to a significant proportion of foreigners), sex, age, literacy (a common condition in Latin America until the 1960s), and finally race in Canada until the 1958 election.
The other conditions (particularly property) were generally abolished before 1945.
For the maps, the maximum class limit, 76%, will seem modest. However, it allows to include countries other than those where voting is obligatory and accompanied by an effective sanction
(Argentina, Bolivia, Brazil, Chile, Ecuador, Peru, Uruguay and Australia), as well as those with autocratic tendencies (such as those from the former USSR),
all countries where participation often exceeds 80% or even 90%, which is practically never the case elsewhere.
Finally, the relevant data and maps and graphs are up to date for the elections held in November 2025.
The succession of maps shows first the pioneering role of Western Europe and America, of Australasia as a wagon, plus Japan and Thailand, just hooked up to the train of the free representative system.
Secondly, we will note the late arrival at democratic maturity of Africa, which took place at the same time as the Soviet world broke up and opened up to it itself.
Finally, we can guess a strong contrast between the first mentioned, with high and stable civic-mindedness, and the second mentioned, with an erratic electoral trajectory.
The sliding evolution curves will illustrate this better — each peak of the curve represents a moving average in most cases over five years, or over four years as specified in each legend.
Or in a simplified version, by large areas of the world…
See the detailed comment in french (same maps and graphs).
Statistical sources:
– France: personal electoral figures, and INSEE for the population;
– rest of the world, for electoral statistics: International Institute for Democracy and Electoral Assistance (IDEA), widely amended by various online sources, preferably official
(the majority of countries have put this type of statistic on the Web, generally under the aegis of their Ministry of the Interior), secondarily specialized works,
for example the 6-volume compendium coordinated by Dieter Nohlen; the IDEA indeed provides figures that are sometimes fanciful, for example on France,
where the number of voters and registered voters can be underestimated by several million, or where overseas territories are sometimes taken into account and sometimes ignored, without any apparent logic;
it also sometimes takes the votes cast for the voters, which underestimates electoral participation by that much; as for the United States, for mysterious reasons,
the IDEA overestimates the number of votes by up to 11 million: the very comprehensive site uselectionatlas.org was used for the presidential figures,
and the publication of the House of Representatives for the latter, entitled Statistics of the presidential and congressional election,
supplemented by the results of the primaries in the constituencies where the absence of competition on the day of the national election results in an absence of votes and consequently of figures;
– rest of the world, for population: United Nations - Population Division - Department of Economic and Social Affairs - World Population Prospects 2022 - File POP/01-1:
Total population (both sexes combined) by single age, region, subregion and country, annually for 1950-2100 (thousands), Estimates, 1950 - 2021.
For the earlier period and for the supplements necessary for the definition of the electoral population, in particular the illiteracy rate in countries having used this restriction: national censuses.
