The British Election Study is releasing data on the occupational position of respondents of the 2015 face-to-face post-election survey. Respondents’ open ended respondents were coded to standardized occupational codes, which are then coded into social class positions.
We manually coded all British Election Study face-to-face respondents’ occupations into the SOC2010 scheme based on the open-ended responses to the questions: “What [does/did] the firm/organisation you worked for mainly make or do (at the place where you work/worked)?”, “What [is/was] your (main) job?” and “What [do/did] you mainly do in your job?”. NS-SEC classes are then calculated using the respondent’s SOC2010 code and further questions about the respondent’s workplace. We conducted further cross-checks to ensure that the final NS-SEC assignments matched the respondents’ original responses.
The data can be downloaded here.
The class data allows us to analyze the relationship between a voter’s social class and their vote choice in 2015. Prior to the 1990s, class was a major driver of vote choice in Britain, with Labour appealing mainly to the working class and the Conservatives gaining votes mainly from the middle class.
However, class became less important following Tony Blair’s repositioning of the Labour party as a “catch-all” party that tried to appeal to voters from all backgrounds rather than appealing mainly to those in the working class.
The figure below shows the proportion of each class that voted for each party in 2015. The results show that there is certainly a much weaker relationship between social class and voting for Labour or the Conservatives. The Conservatives beat Labour across most classes, with Labour only managing to lead in the routine, lower supervisory and never worked categories, which are far from a winning coalition in modern Britain. This is not to say that class voting has disappeared entirely: the Conservatives still perform best in the higher managerial/professional classes which have always been their strongest base of support. Nonetheless, it is a real change in British politics that the Conservatives can be competitive among groups such as semi-routine workers, who would previously have voted primarily for Labour.
Although the working class votes for Labour seem to have fallen, it is not necessarily the case that they are gone for good. Labour does still have a working class lead in people who identify with a party. However, many of their identifiers either didn’t vote or defected to another party in 2015. We will have to wait for future elections to see if they are gone for good.
Our results also shed further light on the much discussed relationship between UKIP voting and social class. Our previous research showed that Labour had gradually lost votes to UKIP although often via other parties. This seems to be reflected in UKIP’s base of support: the routine and semi-routine classes are the most likely to vote for UKIP. In fact, UKIP now has the most disproportionately working class base of any party. The face-to-face data shows stronger support for UKIP among the working class than our previous analyses using the BES internet panel.
Nonetheless because of the large scale shifts in occupational structure in Britain, only 34% of the UKIP vote comes from the traditional working class (routine and semi-routine NS-SEC classes). In fact nearly 20% of UKIP’s support comes from the lower professional class, more than from any other single source.
Class voting in Britain has not gone away in 2015. However, it has transformed into something very different than existed before the 1990s. Labour’s shift to become a catch-all party allowed them to become the party of government for a decade and attract votes from an unprecedentedly diverse coalition. Of course the downside to trying to catch everything is that sometimes you don’t catch very much at all.