I read in an unrelated but generally reliable book that in one national election in Germany in the early 1930s or late 1920s, the results were in by 3:00 a.m. the next morning.

When you think of it, vote counting is a task that is relatively simple and can easily be parallelized. Adding the individual results is not particularly hard either, even by pen & pencil. Though dividing into percentages and actually allocating parliament seats (if it is a parliamentary election with proportional representation) is a bit less trivial. And communication of local counting results may be a bottleneck.

Is there data on how long it historically took to determine results of general elections (with universal suffrage or universal male suffrage) on a nation-wide level in larger European countries? I.e. results of general presidential or parliamentary elections, in countries at least as large as Belgium or the Netherlands, e.g. not Andorra or Luxemburg. State-wise or national elections in the US would also be interesting.

  • 7
    I think the results of Imperial Elections were available seconds after the prince-electors made their decision, considering there were only 7-9 of them.
    – Semaphore
    Nov 6, 2020 at 9:58
  • 1
    Welcome to History:Stack Exchange. Thank you for your question; please consider revising it to be more in line with our community expectations. Like many other stacks, we expect questions to provide evidence of prior research. That helps us to understand the question, and avoids our repeating work you've already done. Our help center, and other stacks provide additional resources to assist with revisions.
    – MCW
    Nov 6, 2020 at 10:59
  • In case you are asking in reference to a discusson of the exercise in patience that is the US elections ("how come others get it done in twelve hours?"), that is mostly down to various causes inherent in the US system. Too many voters per election office (and people availabe to do the counting), and mail votes being valid by postal stamp (so there are additional votes coming in long after election day) being just two of them. If every election office has to count just a few hundred votes with all the votes being in at, say, 6 PM on election day, then twelve hours aren't that exceptional.
    – DevSolar
    Nov 6, 2020 at 11:00
  • 4
    This describes Lincoln learning he'd won on election day in 1860. The important technology is the telegraph, not any counting method. Counting can be made quite fast if you throw enough people at it. Also, the speed of results is very dependent on the closeness of the election, probably more so than technology, and there's a huge difference between "knowing who won" and "having the exact vote count". Which "result" is meant here? Nov 6, 2020 at 17:46
  • 2
    Not sure if the UK counts as European for the purposes of your question, but I think the answer is 1983 (first election where the overnight count covered enough seats to confirm a parliamentary majority), although enough seats were counted overnight to be able to estimate the result probably since the 1920s/30s.
    – Andrew
    Nov 6, 2020 at 20:11

2 Answers 2


In Germany probably on 4 May 1924.

The final official results of a election for the Bundestag will even today take weeks (the last election was on 24 September 2017, the final official results were published on 12 October 2017). Before that each unit (constituency, state, federal level) will check the protocols of the lower level and decide about problems. But long before that, in the night after election the "vorläufiges amtliches Endergebnis" (provisional official final result - final means all votes are counted) is published (last time 5:25 am, poll closes 6 pm).

The process toward this provisional result is laid down in § 71 Bundeswahlordnung (federal election order). The electoral board (Wahlvorstand) of the electoral district (Wahlbezirk, [1]) begins with counting immediately after they closed the poll (6 pm). When finish they transfer the results as fast as possible (usually by phone or digital) to the municipality (Schnellmeldung - quick report). There they collect all incoming results and transmit it to the election officer of the constituency. Simlar from constituency to state and from state to the federal election officer. On constituency, state and federal level the elction officer publishs the provisional result.

Such an instruction on fast collection of the results was introduced in German federal election law by the decree of 21 December 1920 to alter the Reichswahlordnung (RGBl 1920, 2122). It introduced a § 55a (renamed to § 61 in the republication of the Reichswahlordnung from the same day, RGBl 1920, 2171) outlining a similar procedure as today. For the transmission are phone, telegram or courier suggested. The report to the federal level is prescribed to take place no later than 8 pm the following day (polls close in summer 5 pm, in winter 6 pm, § 48 RWO), if needed with the number of muncipalities missing.

As the previous Reichwahlordnung from 1 Mai 1920 (RGBl 1920, 713), the Reichswahlgesetz for the national assembly from 30 November 1918 and the Reglement for the election to the North-German Reichstag of 31 May 1869 [2] do not include such a provision, but only rules to transfer the protocols to the higher level within days, I suspect such a procedure did not exist before. The causes for the new rule may be identifiable using the protocols of the Reichsrat (the representation of the states on federal level) of 1920 as the Reichsrat has consented the changing decree. As far as I can see these protocols are not online available.

I am not sure if the provisional results of the election to the Reichstag on 4 May 1924 (the first election after the change) really were published 12 hours after the poll closed (5 pm). It may have taken a few hours more. One has to look in the contemporary newspapers for more information. But this was the first federal election in Germany when a mechanism was in force to publish the provisonal results fast.

[1] The German electoral districts have up to 2500 eligible voters (§ 12 BWO) and the elecotral bords have 5 to 9 members (§ 9 II BWahlG). There are separate elecotral bords for mail-in-voting.

[2] I did not check the Reglement of imperial Germany for alterations between 1869 and the Weimar Republic.


K-HB has a nice answer, but I will answer here with some historical data for Germany anyway, because of the relatively good accessibility of sources. I will mainly quote from the paper "Vossische Zeitung", which is available in scanned form from the Staatsbibliothek Berlin here.

In the June 15th, 1893 election, results of all but one constituency were known by Monday, June 19th. The election results from Berlin were already known the morning after the election.

The preliminary(?) official result for the June 16th, 1898 election, were made known before noon on Monday, June 20th. I.e. getting the result took roughly as long as in 1893.

For the 1903 election, there are unfortunately no useful scans in the Staatsbibliothek digital archive.

The results for the 1907*, 1912, and 1924 (May and December) elections were all known by around noon the next day.

The results for the 1920 elections took considerably longer, they were reasonably certain (though not yet official) by early June 9th. Election had been on June 6th.

For the elections of 1928 and 1930, I could not find the time when the official preliminary results were published. According to Joachim Fest, "Hitler", Ullstein 2005, p.420 the results for the 1930 election were ready by 3:00 a.m. the following day. As pointed out in my question, I consider Joachim Fest to be a relatively reliable author, but the topic of the work is not the history of vote counting, so IMHO this timespan is not really set in stone.

So in Germany, vote counting or processing seems to have become considerably faster from 1898 to 1907, and again a lot faster in the late 1920s (assuming that Fest is reliable and is using the same kind of result that the newspaper uses, which for the 1924 elections are "Vorläufiges Gesamtergebnis" / preliminary end result).

(*) the 1907 paper has an analysis article which says "the result of the election is not entirely clear yet" right next to the results. My interpretation is that maybe the analysis article refers to runoff elections, or that the results came in right before the paper was run to the printing press.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.