The Coming Change

Contested By-Elections in Wartime Britain and the General Election of 1945

Examining Mass Observation’s reports on the electorate and the various by-elections in the latter half of World War II show that the Tory-led national unity government was facing continually competitive by-elections intended to have been uncontested. This series of government defeats and close calls against left-wing insurgent, truce-breaking candidates from the Common Wealth Party or Radical Action points towards the eventual Conservative defeat in July 1945. These by-elections become an even stronger indicator of the Tory Government’s fate when considering that they were: fought by newly formed organizations, based on out-of-date electoral registers that excluded many important left-wing constituencies (the young, the displaced, and most members of the armed forces), faced candidates backed by the all-party government, were met by stiff social suppression of the insurgent candidates, and were taking place amidst a war. Despite the fact that Mass Observation’s (M-O) more formal non-predictions of the general election vaguely pointed a swing towards Labour, not a decisive Labour victory, M-O’s reporting and the general course of by-elections during the latter half of the war provides a clear line of evidence that the 1945 general election was going to be a realigning election.[1]

The expectations of the outcome of the 1945 general election were decidedly mixed. In its primary document on the 1945 general election, M-O reports stated that most voters were expecting that “it [the election] would be a very close thing, and that the Conservatives would get in by a very narrow margin.”[2] In fact, the M-O report on the general election records only one report of someone “definitely expecting” a Labour win, a book shop manager who said that “I was told last week that the betting on the London Stock Exchange was 6 to 4 on Labour getting in.”[3]

M-O’s analysis was most likely intentional indecisive bet-hedging. This is best seen in its pre-election report on the electorate where it stated that a clear left-wing swing was apparent but lessening, with a clear rightward shift in the civilian vote and that if that shift continues, “an indecisive result seems likely on a high poll.”[4] This waffling is followed up by an equivocation that “Mass Observation is not attempting any national sampling in this election.”[5] Another potential reason for M-O’s sheepish reporting is that it had never studied a general election before. The last general election was in 1935, almost ten years prior and two years before Mass Observation was even founded. Regardless, M-O’s reporting articulated that there was likely to be at least a somewhat significant leftward shift, but it was written in such a way to give room for a narrow Tory victory; qualitative herding possibly the product of the relatively sparse expectations of a Labour win and Mass Observation’s inexperience with national elections.

However, Mass Observation’s equivocations belie their previous reporting, which continually shows competitive by-elections that indicate the electorate’s changing mood and interest in replacing the Tory government. From February of 1943 to May of 1945, there were 41 by-elections in the United Kingdom, resulting in seven-seat changes.[6] A 17% success rate may seem low, but that ignores several very close losses and that the electoral landscape strongly disfavored these insurgent candidates. That the Common Wealth Party and other groups like Radical Action or the Independent Labour Party were even competitive is an achievement and an indication of the future electoral fortunes of the Conservative majority in Parliament.  

Wartime by-elections were intended to be functionally uncontested under the rules of the wartime electoral pact between the members of Churchill’s War Ministry. That meant that every principal national party—the Conservative Party, the Labour Party, the Liberal Party, the Liberal National Party, and the National Labour Organization—endorsed a single candidate for each by-election.[7] The idea was that parties would not run contenders if a vacancy were to occur in a seat held by another coalition member. However, independents still contested many by-elections, and eventually, the left-wing Common Wealth Party (CWP) was formed in 1942 with the sole intention of contesting Conservative and Liberal National by-elections.[8] Thus, left-wing insurgent candidates took up the gauntlet against a nominal all-party coalition and remained electorally competitive.

The CWP had to face candidates formally backed by a government of national unity, during a war, and had to do so on electoral registers that were years out of date. The war had stopped both general elections and electoral registration. That is why the 37th Parliament elected in 1935 was extended four times to have a 10-year term, double the length of a ‘normal’ parliament.[9] Likewise, the war stopped the registration of voters who had come of age or voters that had moved since 1939. For example, during the Eddisbury by-election in April 1943, M-O noted that about 15% of those interviewed did not have votes because the electoral register for the constituency had not been updated since 1939.[10] Of that group, 22% could not vote because they had become 21 since 1939, and 58% could not vote because they had moved to the constituency since 1939.[11] Thus, those displaced by the war and the young, those more likely to be partial to left-wing or anti-government candidates, especially as the war continued and the “rally ’round the flag” effect was replaced by war fatigue.

Even if the observations of the M-O report did not hold for the entire constituency, and less than 15% of the population was disenfranchised, the number of people disenfranchised would still be significant because of the narrow margins of the by-elections, including Eddisbury.[12] If, however, 15% of the electorate was disenfranchised as the M-O report suggests, that would amount to thousands of potential votes, much more than the 486-vote majority of the CWP candidate in Eddisbury, for example.[13]

Further, these by-elections were contested without the votes of most armed forces personnel stationed in a constituency or who had previously lived in that constituency. Mass Observation’s report on the status of the armed forces vote in July 1944 states that the application for absentee voting had only been available since January 1944.[14] It is unclear that absentee vote also applied to by-elections or just an eventual general election. And, like other citizens, members of the armed forces would not be able to vote in the constituency they were residing in if they had not lived there before 1939. This was seen during the by-election in Bury St Edmunds in February 1944, where M-O reported that service personnel being disappointed at not being able to vote; and if they could vote, they would vote for the truce-breaking Radical Action candidate (who was backed by the CWP).[15] The lack of vote by the forces could further have contributed to the lack of truce-breaking victories as M-O reported consistently that the forces vote was generally for the left.[16] Thus, these closely fought by-elections often had left out even more potential left-wing voters that would be present at the 1945 general election.  

However, there were still more barriers to the insurgent candidates beyond just skewed electorates and a united backing of government candidates; those insurgent candidates often faced severe suppression by local powerbrokers. M-O documented this to a great extent in their report on the Bury St. Edmund by-election where the influence of the Green King brewery was brought to bear against the supporters of the truce-breaking candidate. This included the “case reported of a man said to have been dismissed from the brewery for canvassing for Mrs. Ashby, and an innkeeper who was also canvassing said he expected to have his license take away within a week.” And further allegations that “it was alleged that brewery employees had been locked in the brewery and told what would happen to them if they did not vote the right way.”[17] There was such a continuous and vicious campaign of suppression by the local industrial brewery that people were too frightened to put out window cards and that Mrs. Ashby felt compelled to mention in her stump speech that the voting would be by secret ballot.[18] Local industrialists’ were not even suppressing a radical socialist candidate with such vigor; Mrs. Ashby was just a left-liberal. That a moderate progressive would be subject to such repression shows what barriers even more progressive candidates would face.[19]

Left-wing insurgent candidates faced a daunting task; they had to run against a united government front as freshly founded organizations facing their first election within a few months of their existence, in addition to violating a broadly popular electoral truce.[20] Candidates were fighting in elections with voter rolls at best four years out of date, denying the vote to the young, the displaced, and most of the armed forces, all vital constituencies for left-wing candidates in the era. They further faced active reprisals against those voters who wished to support them. All these disadvantages and they still won elections, and if they did not win, they often significantly narrowed the majorities of the government candidate.

Insurgent candidates did find success. They won in Eddisbury in 1943 (a CWP gain); in 1944 they won in Skipton (a CWP gain), West Derby (an independent Labour gain), and Combined Scottish Universities (an independent gain); and in 1945 they won at Chelmsford (a CWP gain). Of these, the Eddisbury Common Wealth MP would join Labour (but be defeated), the Skipton Common Wealth MP would run for a different constituency (and lose), the West Derbyshire independent would join the Labour Party and retain his seat in 1945, the Scottish University MP would win re-election, and the Chelmsford Common Wealth MP would win as a CWP candidate, but join the Labour Party in 1946.

Even when they did not win, left-wing truce-breakers often saw significant swings in their favor and very close elections. The Common Wealth Party’s second-ever by-election in Midlothian and Peebles Northern in February 1943 saw the government majority cut by 22 points, down to just 869 votes.[21] Likewise, in Watford in February 1943, the Common Wealth candidate cut down the Conservative majority by 23 points, from 13,290 to just 2,0001 votes.[22] At Daventry in April 1943, the Tory majority of 8,167 was reduced to 2,652 by a CWP candidate.[23] Darwen, in December 1943 saw the Tory majority of 1,164 reduced to just 70 votes by a Radical Action candidate.[24] Bury St Edmund, in February 1944 saw a previously unopposed Tory majority cut down to just 2,584.[25] Manchester Rusholme, in July 1944 saw the CWP cut a Tory majority from 10,420 to just 1,760.[26] Bilston, in October 1944, saw an Independent Labour Party candidate cut the Tory majority to just 349.[27]

Of the 39 constituencies in Great Britain (i.e., not Northern Ireland) that held by-elections from 1943 to 1945, 24 were retained by the Conservative Party; of that number, 15 would be won by the Labour Party in the July general election.[28] These constituencies included King’s Lynn, Portsmouth North, Bristol Central, Watford, Buckingham, the Hartlepools, Birmingham Aston, Burton-on-Trent, St Albans, Peterborough, Woolwich West, Acton, Manchester Rusholme, Bilston, and Newport.[29] Further, Labour won the Liberal-held Middlesbrough West, which held an uncontested by-election in May 1945 just before the general.[30] Almost all of the constituencies that were contested saw predictive swings away from the Tories during their by-elections.

Insurgent, truce-breaking candidates fighting close elections on exceedingly unfavorable ground from 1943 to 1945 are an obvious trendline away from the Conservative Party and towards the left. The Common Wealth Party effectively being a stand-in for Labour. M-O’s reporting consistently showed that despite the expectation that races would be uncompetitive routs for the government’s chosen candidates—they were not. That a newborn party could give Conservative candidates a run for their money is not an idle task. The CWP's relative success is a feat that is even more impressive considering that the infant party was running without the votes of working-class voters in the armed forces, young voters, or displaced voters. Not only were truce-breakers running against a nominally united field and a skewed electorate, but they also faced harsh anti-democratic maneuvering by local powerbrokers. The deck was stacked against these truce-breakers, but by-elections almost uniformly remained competitive. Mass Observation's own reporting continually paints a qualitative picture of unexpectedly competitive by-elections despite unfavorable conditions and significant leftward shifts in the general electorate. So, despite not actually predicting anything, Mass-Observation’s work shows that there were cracks in the Conservative’s electoral majority well-before 1945.

[1] “Post-Mortem on Voting at the Election,” Mass-Observation File Report 2282, Septmer 1945, page 1. (Located online at http://www.massobservation.amdigital.co.uk.grinnell.idm.oclc.org.) [2] “A Report on The General Election, June-July 1945,” Mass-Observation File Report 2268, October 1945, page 124. (Located online at http://www.massobservation.amdigital.co.uk.grinnell.idm.oclc.org.) [3] “A Report on The General Election, June-July 1945,” Mass-Observation File Report 2268, October 1945, page 125. (Located online at http://www.massobservation.amdigital.co.uk.grinnell.idm.oclc.org.) [4]“The New Voters and the Old,” Mass-Observation File Report 2261, June 1945, page 3. (Located online at http://www.massobservation.amdigital.co.uk.grinnell.idm.oclc.org.) [5] Ibid. [6] One NI Labour gain from UUP, one, independent labour gain from Conservative, one independent gain from Liberal National, one SNP gain from Labour, three CWP gain (2 from Conservatives, one from Liberal National) [7] “Qn. 3 Political Truce,” Mass-Observation File Report 1962, September 1943, page 2. (Located online at http://www.massobservation.amdigital.co.uk.grinnell.idm.oclc.org.) [8] Steven Fielding, “What Did ‘the People’ Want?: the Meaning of the 1945 General Election,” The Historical Journal 35, no. 3 (1992): pp. 623-639, https://doi.org/10.1017/s0018246x00026005, 627. [9]Prolongation of Parliament Act 1940 (UK) 3 & 4 Geo. 6 c. 53; Prolongation of Parliament Act 1941 (UK) 4 & 5 Geo. 6 c. 48; Prolongation of Parliament Act 1942 (UK) 5 & 6 Geo. 6 c. 37; Prolongation of Parliament Act 1943 (UK) 6 & 7 Geo. 6 c. 46; Prolongation of Parliament Act 1944 (UK) 7 & 8 Geo. 6 c. 45. Each of these acts prolonged parliament for an addition year. [10] “Report on Feelings about the Obsolete Register, at Eddisbury,” Mass-Observation File Report 1659, April 1943, page 2. (Located online at http://www.massobservation.amdigital.co.uk.grinnell.idm.oclc.org.) [11] Ibid, 5. [12] “By-Elections, February-June 1943,” Mass-Observation File Report 1844, June 1943, page 2. (Located online at http://www.massobservation.amdigital.co.uk.grinnell.idm.oclc.org.) [13] “Report on Feelings of Obsolete Register,” Mass-Observation File Report 1659, page 4-5. [14] “The Forces’ Vote,” Mass-Observation File Report 2136, July 1944, page 2. (Located online at http://www.massobservation.amdigital.co.uk.grinnell.idm.oclc.org.) [15] “Bury St. Edmunds By-Election,” Mass-Observation File Report 2035, February 1943, page 13. (Located online at http://www.massobservation.amdigital.co.uk.grinnell.idm.oclc.org.) [16] “The New Voters and the Old,” Mass-Observation File Report 2261, June 1945, page 4. [17] “Bury St. Edmunds By-Election,” Mass-Observation Fire Report 2035, page 12. [18] Ibid. [19] “Bury St. Edmunds By-Election,” Mass-Observation Fire Report 2035, page, page 9. [20] “Qn. 3 Political Truce,” Mass-Observation File Report 1962, page 2. [21] “By-Elections, February-June 1943,” Mass-Observation File Report 1844, June 1943, page 2. [22] Ibid. [23] Ibid. [24] Fred W.S. Craig, British Parliamentary Election Results, 1918-1949 (London, NY: Macmillan, 1974), 394. [25] “By-Elections, February-June 1943,” Mass-Observation File Report 1844, June 1943, page 2. [26] Fred W.S. Craig, British Parliamentary Election Results, 192. [27] Fred W.S. Craig, British Parliamentary Election Results, 280. [28] Ibid. [29] Ibid. [30] Ibid.