Lab 395¹ Con 215² Lib 12 Others 18³
¹ Includes 2 Independent Labour
² Includes 9 Ulster Unionists, 13 Liberal National, 2 National, 2 Ind Cons, and 1 Ind Ulster Unionist
³ Includes 1 Common Wealth, 2 Communist, 10 Independents, and 2 Irish Nationalists
There is a breakdown of seats and percentage vote by region
The total electorate (i.e. registered voters) was 33,204,517. This comprised the following components:-
The question of turnout is slightly complicated by the existence of 15 two-member seats and four STV elections. Normally, in single-member seats, turnout is interpreted as the proportion of the registered electorate that casts valid votes, and is seen as a measure of the propensity to vote. In those 15 seats where each elector had two votes, it is not clear from the published figures how many people actually voted, and so the figure given for turnout in these constituencies is actually the total number of valid votes as a proportion of the total possible number of votes (twice the registered electorate).
Regional breakdowns were:-
|Total electors §||27,148,612||851,198||3,396,323||1,808,384||33,204,517|
|Votes cast (university)||73,472||2,651||32,789||6,935||115,847|
|Votes cast (non-university)||20,536,849||719,769||2,389,990||1,329,671||24,976,279|
|Total votes cast||20,610,321||722,420||2,422,779||1,336,606||25,092,126|
In the four STV elections, the first preference votes were counted in the total of votes cast. The electorate in unopposed seats was excluded from the maximum vote count. The overall turnout in non-university seats was 72.76%
The average electorate size (registered electors in non-university seats) was 53,816.97. In seats that elected a Conservative the average was 57,682.68 (not including associates), Labour 51,796.24, and Liberal 41,196.55. Three seats were unopposed (Armagh, Liverpool Scotland, and Rhondda West).
18 'parties' stood in the election; see the full list of parties for details. The three major parties between them accounted for 91.60% of the votes cast, or 97.33% if the parties' associates are included. Labour achieved its 61.41% of seats on the basis of support from 34.64% of the electorate. The list shows the percentage of votes obtained by each of the parties. 170 candidates in single-member constituencies (plus 5 in university seats) lost their deposits (though see the note below ¶).
A total of 1,658 candidates stood in the election. The average majority in non-university seats was 7,903.89. In seats won by the Conservatives the average was 5,854.49; in Labour seats it was 9,110.64; and in Liberal seats it was 1,880.27. 37 candidates (36 Labour, 1 Conservative) obtained majorities that were more than 50% of the votes cast, and 9 candidates had majorities that were more than 50% of the electorate. There is a list of constituencies in order of percentage majority, together with separate lists for the three main parties in order of marginality. It is also interesting to note that in 67.36% of the constituencies the winner had more than 50% of the votes cast, and there is a list of constituencies in order of the winner's percentage of votes cast. One issue that often interests people is what proportion of the electorate a winning candidate achieves. This is also included in this list. Turnout ranged from 39.25% in Down (two-member seat) to 87.83% in Fermanagh & Tyrone (two-member seat). Constituencies are listed in order of turnout. A summary of turnout from 1945 onwards may be found here.
Copies of party election manifestos are available for the Conservative, Labour, and Liberal parties
In six constituencies, Armagh, Barrow in Furness, Bury St Edmunds, Eye, Gainsborough, and Winchester, the total electorate given was not the sum of the Civilian, Business, and Service voters in both The Times newspaper version and in the The Times Guide to the House of Commons. In the case of Armagh, the Service vote was missing altogether and the figure I have given is found by subtraction of the Civilan and Business votes from the total electorate. In the case of Barrow, the discrepancy was exactly 3,000 which suggests a simple misprint in the Civilian vote. In this case I reduced the figure from 49,415 to 46,415 on the ground that the total electorate was more likely to be the correct figure and the Service vote given was in line with what would be expected. With the others there seemed no sensible way of determining the correct figure, and so some small discrepancies remain. These problems surprised me because I had always assumed that the the figures given in Guide were checked and corrected versions of those published in the newspaper itself. This appears not always to be the case. In the case of Eye, the figures in the newspaper added up but those in the Guide didn't. Where there were differences in the actual voting figures or party labels, I have generally taken the figures in the Guide to be correct in the fond hope that they are not just misprints. It seems, however, that The Times' reputation for being a 'journal of record' should not be taken too far. It's just the best there is, in the absence of the definitive historical records in each constituency. If you know any of the data to be incorrect, don't hesitate to email me.
I have had considerable difficulty with the calculations on lost deposits. I should like to thank Isobel White of the Parliament and Constitution Centre of the House of Commons Library, and David Boothroyd for their help in clarifying the rule on lost deposits. The Representation of the People Act 1918 specified that in order to avoid losing their deposits:-
The problem I have had is that my calculation on the number of deposits lost is at variance with that published in The Times Guide to the House of Commons 1945, and in other sources. The Times reported that 168 candidates lost their deposits, including 5 in university seats. However, this is not based on an independent calculation from the constituency results but represents what was reported to them by the political parties (Times Guide p.138). David Butler in The Electoral System in Britain since 1918 gives the total number as 163 (p. 169), but it is not obvious what his source is. I believe this is also the figure given by McCallum & Readman in The British General Election of 1945, though I haven't been able to check that. It is possible that they both got their figures from The Times and excluded university seats.
NB: Craig Aston has since pointed out to me that FWS Craig, whose information I did not have access to when I first looked at this issue, gives 182 in his British Electoral Facts, 1885-1975.
My calculations are that 170 non-university candidates in single-member seats, plus 5 university candidates, secured less than 12.5% of the vosts cast and thus should have lost their £150 deposits. The main difference lies in the number of Liberal lost deposits. The Times says that 64 Liberals lost their deposits, while my calculations suggest that this figure should be 72. My calculations are only as good as the original data, which come from The Times and The Times Guide, and I have produced a table listing each of the candidates, in the non-university seats, that appears to have received less than 12.5% of the vote. I'm not sure whether the RPA 1918 specified rules about the calculation and rounding. There were two cases (Chertsey and Lewes) where a candidate received just less than one eighth, but if percentages were used it could have been rounded up to 12.5, and there were several other cases where the percentage was around 12.3 - 12.5. I am also not clear whether the calculations were made by the Treasury, who administered the refunding of deposits, or by the individual returning officers. If it was the latter, it is possible that the calculations were not done entirely consistently. But I'm clutching at straws here. Presumably only the Treasury records have the correct figures on deposits refunded (and therefore also on those lost). Without figures on the number of valid ballot papers in the two-member seats, which I haven't found yet, it is not possible to say how many deposits were lost in these seats, though if everyone had cast two ballots the figure should have been 7, making a grand total of 182.
It is interesting to note that two candidates, Lindsay - in English Combined Universities (2-member, STV) - and Kerr - in Scottish Combined Universities (3-member, STV) - managed to win a seat while at the same time receiving less than one eighth of the appropriate total of first preference votes. They are obviously not included in the total of lost deposits.
Craig Aston has pointed out to me that "any candidate who received less than 6.25% of the vote definitely lost his deposit, example as follows:
In a 2-member seat, 50,000 ballot papers are cast. Candidate X receiving 6,249 votes loses his deposit. If every voter cast both votes, 100,000 votes were cast, and candidate X gets 6.249% of the vote.
If, however, 40,000 voters cast 2 votes and 10,000 are plumpers, 90,000 votes are cast. Candidate X therefore gets 6.94333% of the vote yet loses his deposit. The more plumpers, the higher the threshold for saving the deposit, but it is always equal to or above 6.25%.
Therefore, a candidate receiving less than 6.25% definitely loses his deposit. In 1945 this certainly happened for:
1) Blackburn (two Liberals)
2) Oldham (the lowest polling Liberal only)
3) Preston (the Liberal and the Communist)
4) Sunderland (the Communist)
5) Dundee (the SNP candidate)
And indeed quite apart from all this FWS Craig (Results 1918-1949) gives the breakdowns of the 1945 results for all these except Oldham which bears this out.
So the number of lost deposits in 1945 was 182: 170 single-member, 7 two-member, 5 Universities." This is also the conclusion I reached.