‘We Didn’t Know What We Were Eating Tomorrow’: How Class Origin Shapes the Political Outlook of Members of the Parliament in Britain

Abstract

Most work in political science on class and political ideology is focussed on politicians’ class destination rather than class origins. Yet that is inconsistent with evidence in the case of the United Kingdom that the conditions of someone’s family upbringing do influence their politics. This article revisits the conceptualisation of class background in the current literature by redirecting attention to the sociological concept of class origin. We draw on in-depth interviews with 24 British Members of the Parliament to unpack how these political elites perceive their class background to have affected their political outlook and behaviour. Our results indicate that ‘class origin’ is more salient in the formation of Members of the Parliament’ political outlook than educational or occupational background. The manifestation of this political outlook is constrained, however, by party discipline. This tension in how British Members of the Parliament relate to their class origins has implications for how we think about the power of descriptive representation in politics.

Introduction

In the build up to the next UK general election, the leader of the Labour Party, Sir Keir Starmer, repeatedly highlighted his working-class origins and the way they inform his politics (Elgot, 2021; McKeon, 2022). ‘When I think about what that new deal for workers needs to look like, I think of my Dad’, he explained to delegates at the 2021 Conference of the Trade Union Congress. ‘He worked on the factory floor all his life. I think about what he would want, what his aspirations were, and how they could have been realised’ (Starmer, 2021). Starmer’s account is an apt illustration of a long-standing theory in the social sciences: that the class origins of political elites have a direct impact on both their ideological outlook and behaviour.

Despite Starmer’s narrative, there is actually surprisingly little evidence connecting a politician’s class origins to their political ideology or behaviour later in life. Most work in political science on class and political ideology is focussed on politicians’ own work experience (their class destinations) rather than the occupations of their parents (their class origins; Alexiadou, 2015; Carnes, 2012; Lupu, 2015; O’Grady, 2019). As Manza and Brooks (2008: 204) argue, a politician’s own ‘occupation provides the most plausible basis for thinking about how specifically class-related political micro processes and influences occur’. Carnes and Lupu (2023b: 184) go even further, suggesting that the primacy of occupational background as the best indicator of class background reflects the fact that there ‘there is little evidence of a link between the social class of their parents and their adult political views’.

Yet there are reasons to be sceptical about this claim. In particular, it is inconsistent with extensive survey evidence in the UK, which suggests that the conditions of someone’s family upbringing and in particular their formative years or ‘primary socialisation’ (De Graaf et al., 1995; Langsæther et al., 2022; Reeves and Friedman, 2024) do influence their politics. Moreover, even if there were no relationship between class origins and ideology among the population in general, it might still be true that class origins affect the political outlook of politicians.

Reflecting on this, in this article, we seek to revisit the conceptualisation of class background in literature exploring the impact of class on the outlook of politicians, and redirect attention to the sociological concept of class origin. Specifically, we draw on in-depth interviews with 24 British MPs, unpacking the ways in which they perceive their class background to have affected their political outlook and behaviour. Here we explore the relative salience, and interplay between, three different conceptualisations of class background – own occupation, own education, and parental occupation – in the way British political elites articulate their political identity. We are interested not just in the way they perceive these elements of their background to have shaped their political outlook, but also how it has affected their behaviour as professional politicians. After all, we know that politicians’ ability to convert their attitudes into concrete actions tends to be heavily constrained by party discipline (e.g. Carnes and Lupu, 2015) and so we are particularly interested in how the impact of class background is mediated by political constraints. Considering that, bar one, all the interviewees were former rather than current or prospective MPs, their answers are likely to reflect what they really think rather than be a case of saying ‘the right things’ to appeal to voters, even more so given the anonymity of those answers.

Our results indicate that parental occupation or ‘class origin’ is much more salient in the formation of political outlook for these politicians than educational or occupational background. More specifically, most MPs deem their education and occupation as playing a consolidating role in concretising their political views but locate the root, genesis, or formation of their overarching political outlook in their upbringing. Moreover, the political direction of this influence is clearly patterned; Conservative and Labour MPs from working-class origins position their backgrounds as a key driver of why they sit to the political left of their more privileged colleagues in each party. This does not mean, however, that the manifestation of this political outlook is unconstrained by party discipline. Indeed, we find that such constraints generally (albeit not always) trump the impact of class origin in determining the policymaking behaviour of political elites.

This tension in how British MPs relate to their class origins has implications for how we think about the power of descriptive representation in politics. If class origins influence self-reported ideology but not behaviour, this poses important questions concerning why many like Starmer foreground their working-class background rhetorically when articulating their political identity, especially in public. One possibility is that when British politicians ground their political identities in terms of their working-class origins, they may be to some extent performing ‘ordinariness’ in an effort to shore up their political legitimacy. These political elites are not alone. Performing ordinariness has become a common feature of contemporary elites in an age of increasing inequalities (Reeves and Friedman, 2024). Thus, particularly in the case of political elites from a working-class background, descriptive representation may not necessarily translate into substantive representation.

Does Class Background of Politicians Affect Their Ideology and Behaviour?

The impact of descriptive representation of class on substantive representation was, for a long time, seen as either non-existent or largely trivial (Matthews, 1985; Norris and Lovenduski, 1995; Putnam, 1976). Hence, while work on the impact of the gender and race of legislators on their policymaking has flourished (e.g. Bird, 2011; Cameron et al., 1996; Hargrave, 2023), the role of class has, until recently, been largely overlooked (Carnes, 2012: 9).

However, over the past decade or so there has been a relative surge in, mostly quantitative, studies trying to establish whether there is a causal relationship between the class background of political elites and their decision-making behaviour, with an overwhelming focus on political elites (for a useful review of the current literature, see Carnes and Lupu, 2023a). A couple of exceptions notwithstanding (e.g. Hahn, 2024), this literature is characterised by a relatively strong consensus: when controlling for other factors, such as party affiliation or constituency, the class background of political elites has a direct effect on their attitudes and/or the policies they promote (Alexiadou, 2015; Bonica, 2020; Borwein, 2022; Carnes, 2012, 2013; Carnes and Lupu, 2015; Grumbach, 2015; Hansen et al., 2019; Hemingway, 2022; Hyytinen et al., 2018; Lupu, 2015; O’Grady, 2019; Witko and Friedman, 2008). In short, those from the upper-class are more likely to favour socio-economic policies that benefit the upper-class, while those with a working-class background are more likely to favour socio-economic policies that benefit the working class.

While the impact of class background on attitudes seems rather straightforward, the relationship between class background and actual behaviour is slightly more complex. On one hand, those from wealthier backgrounds do tend to vote for economic policies that broadly benefit the wealthy in society (Carnes, 2012, 2013; Griffin and Anewalt-Remsburg, 2013; O’Grady, 2019). Occupational trajectories potentially matter here too, as O’Grady’s (2019) work on British Labour MPs shows; MPs who had had political careers prior to entering office were more likely to support cuts to welfare than MPs from a working-class occupational background. As the share of MPs in the UK who have been employed in working-class jobs has steadily declined over the past decades (Quilter-Pinner et al., 2022; see also Carnes and Lupu, 2021), the correlation between class and policymaking may explain why politicians have been slow to respond to rising inequality (Lupu and Pontusson, 2023). Indeed, it may explain why working-class voters have displayed a stronger sense of alienation from political elites (Cowley, 2013) and higher abstention rates in elections (Evans and Tilley, 2017; Heath, 2016).

However, studies examining the impact of legislators’ class on their policymaking behaviour also appear to indicate that this relationship is significantly constrained by factors such as party discipline. In their study on legislators in Argentina, Carnes and Lupu (2015: 14) find that:

When parties have less influence and legislators have more leeway – as they do when legislators introduce bills – class-based differences in legislative attitudes seem to matter. In sharp contrast, when parties have more influence – as they do when legislators cast their votes – class seems to be irrelevant.

Similar findings in relation to parliamentary speeches or questions, where legislators tend to face less stringent constraints on expressing their ideological preferences, have also been recently made in the cases of Germany (Bailer et al., 2022) and Norway (Fiva et al., 2023).

Notably, the most prevalent indicator of class background used in this literature is politician’s own occupational background or experience (Carnes, 2012; Hahn, 2024; Manza and Brooks, 2008; Witko and Friedman, 2008). According to Manza and Brooks (2008: 204):

Occupation provides the most plausible basis for thinking about how specifically class-related political micro processes and influences occur. . . . Workplace settings provide the possibility of talking about politics and forging political identity, and work also provides a springboard for membership in organizations where class politics are engaged: unions, professional associations, business associations, and so forth.

Carnes and Lupu (2015: 4), two of the most innovative researchers in this area, agree, noting that this reflects the fact that ‘among people in similar adult social classes, there also seems to be no relationship between the social classes of their parents and their own political views’.

However, this view is at odds with the long-standing consensus among sociologists that social orientations, including political attitudes, are predominantly shaped during primary socialisation (e.g. Acock and Bengtson, 1978; Ares and Van Ditmars, 2023; Braungart, 1971; Dalhouse and Frideres, 1996; De Graaf et al., 1995; Jennings et al., 2009; Sieben and de Graaf, 2004; but see Langsæther and Evans, 2020). In their recent study on British elites, Reeves and Friedman (2024) show that elites from middle-class origins tend to hold more right-wing views on both economic and cultural issues than elites from a working-class family background. Also, taking a broader look at the British public, Langsæther et al. (2022) find some increase in right-wing economic views among the upwardly mobile, but that increase is significantly smaller than cross-sectional differences between people from different class origins. Overall, they conclude that the ‘association between class and a range of identities, opinions and preferences is therefore more likely to be caused by early life experiences and longer-term socialization than by the immediate material interests associated with jobs’ (Langsæther et al., 2022: 958). Indeed, a recent study by Fasching and Lelkes (2024: 17) reveals the importance of ancestral kinship in ideology formation, finding that ‘pre-industrial family structure, rooted in the local ecology and the cooperation needs of a person’s ancestors, influences contemporary political attitudes’.

Furthermore, even a few studies from the literature on the class background of political elites raise doubts about the pre-eminence of occupational background. Grumbach (2015), for example, looks at both class origins and subsequent class location of Congress members to find that upper-class Democrats of working-class origin are economically more progressive than upper-class Democrats of upper-class origin. In Germany, too, Hayo and Neumeier (2016) show that prime ministers at state level spend significantly more on social security if they had a less privileged upbringing. Interestingly, when looking at the potential influence of social mobility, they find contrasting effects: whereas political elites from lower class origins that experience upwards mobility prior to entering office display a similar spending behaviour to those from an upper-class background, so do those from upper-class origins who experience downwards social mobility. In other words, whether they start from or end up in the upper-class, political elites tend to promote a fiscal policy aligned to the interests of that social group. Similarly, in a broader study covering 74 countries, Han and Han (2021) find that political leaders who experienced economic hardship during their youth increase spending on social welfare spending during their time in office.

Beyond class origins and destinations, there is another dimension to this debate that focusses on education. There is a scarcity of research on whether the education of political elites matters for their ideological and policy preferences. Those who have examined how the political elites’ level of education affects upon their behaviour (Carnes, 2012; Carnes and Lupu, 2015; Erikson and Josefsson, 2019; Göhlmann and Vaubel, 2007) have not looked at the impact on ideology and policy preferences. One exception is the comparative study by Dreher et al. (2009), who find no clear relationship between the education of political leaders and the introduction of measures aimed at the liberalisation of the economy.

What is puzzling about this literature is that it runs against another body of work which suggests that higher education does tend to make people in general more liberal – not only on cultural issues, as the consensus has been for several decades (e.g. Kaiser and Lilly, 1975; Lipset, 1960), but also on class-based, economic issues (e.g. Attewell, 2022; Gelepithis and Giani, 2022; Mendelberg et al., 2017; Scott, 2022). In the case of the UK, Scott (2022: 8) finds that ‘achieving a degree renders an individual significantly more economically right-wing’ (but see Simon, 2022). This effect is even stronger in the case of those coming from an affluent background (Mariani and Hewitt, 2008; Mendelberg et al., 2017) and those attending elite universities (Fisman et al., 2015; Mendelberg et al., 2017). Considering the high share of people that meet all these criteria among contemporary political elites (Elsässer and Schäfer, 2022), particularly so in the UK case (Bukodi et al., 2024; Quilter-Pinner et al., 2022), there is arguably merit in further exploring the impact of education on their outlook and behaviour.

In sum, despite the recent advances in understanding the nexus between political elites’ class background, their ideological outlook, and policymaking behaviour, important questions remain. First and foremost, while the current literature most often operationalises class background as occupational background, there is good reason to believe that the sociological concept of ‘class origin’ might also be pivotal to understanding the political outlook and behaviour of politicians. The rather limited accessibility of data on class origins compared to occupation might be the main reason why the current, overwhelmingly quantitative, literature has not fully addressed this question so far. As Pontusson (2015: 209) pointed out in his comprehensive ‘state of the art’ article, ‘an important item on our collective agenda is to clarify, theoretically and empirically, how education, occupation and income’ (and we would add class origin) ‘relate to each other – how or when, under what conditions, they reinforce each other and add up to class’. This arguably calls for a qualitative approach – virtually absent in the current literature – to shed light on both how the different components of class background (origins, education, occupation) seem to inform the outlook and behaviour of political elites as well as how these components communicate with each other as part of this processes. Equally important, a qualitative approach can help us get a better idea of the factors that mediate the impact of class background on the actual policymaking of political elites, in such a way that goes beyond ‘usual suspects’ such as party discipline. In particular, the interview encounter represents a relatively unconstrained setting in which politicians are able to express their ideologies or attitudes free from the constraints of party discipline.

Data and Methods

The article builds on 24 semi-structured interviews with British political elites (22 former MPs, 1 current MP, and 1 former MEP) conducted between 2022 and 2023. Interviewees were recruited through a survey sent out in June 2022 to entrants to Who’s Who – the leading biographical dictionary of ‘noteworthy and influential’ people in the UK, which includes all MPs. 4074 people responded to the survey, a response rate of approximately 17.4% (which is similar if not higher than other surveys of elites), and 1722 stated that they would be happy to be interviewed for future research. From this we recruited 24 MPs, using the survey data on parental occupation to assemble a sample of eight MPs whose main breadwinner had routine occupations (what we might call ‘working class’) and 16 whose main breadwinner had professional occupations (‘middle class’).

The interviews were conducted mostly online (15) but some face-to-face (9). The period in office covered by the interviewees spans from 1974 to 2024 and the party affiliations were divided as follows: 14 Labour, nine Conservatives, and one Liberal Democrats. While the sample is somewhat skewed in favour of Labour, the findings on Labour MPs are arguably more interesting in the context of the aforementioned decline of Labour MPs from working-class origins in recent decades – a process that was not replicated as much among Conservatives, who never had a sizable share of MPs from this social background.

The interview guide was divided in the following three main sections: the first one explored the participants’ upbringing, education, and pre-politics occupation(s); the second section focused on their time in Parliament, particularly on their policymaking behaviour; the third section asked participants their views on a number of political topics, such as austerity or public ownership. Thus, while we did not ask participants explicitly about the influence of their class background on their political outlook and behaviour, our questions were meant to explore that potential influence and the relative importance of different dimensions of class in that process. The responses inductively revealed the prominence of class origins in shaping participants’ political outlook and, to some extent, policymaking behaviour.

The anonymity of participants was guaranteed from the very start, partly to ensure they felt comfortable in answering the questions. For that purpose, the interviewees are referred to by their assigned pseudonyms and certain biographic details have been left out. At the same time, two exceptions notwithstanding, these participants not having a political career any longer, or any imminent prospect of resuming it, meant that they were even less concerned about what the party, the media, or the voters might think about what they had to say. Hence, their answers may be seen as a relatively accurate reflection of what these people think rather than of pragmatic considerations linked to their political career, such as displaying characteristics that voters might find desirable.

Results

We begin by unpacking the perceived influence of the following three key components of class background – origins, education, occupation – on the political outlook of MPs. We start with class origins and then turn to occupation and education. We then reflect on whether class background affects actual decision-making.

Class Origin and the Root of Political Outlook

When Kier Starmer locates his political ideology in the context of his upbringing, he is not alone. Class origins were absolutely central to the formation of political identities and ideologies among our interviewees, and this was particularly salient among Labour MPs from working-class origins. For example, Clive vividly remembered his working-class mum telling him to ‘vote Labour’, while Howard’s parents, both of which had manual jobs, ‘were both always political . . . they were very active trade unionists’.

This primary socialisation into a political ideology did not only come through parents; however, it was also connected to grandparents and, at times, the wider community in which people were embedded. A crucial moment for Howard was ‘the 1964 General Election. My grandmother took me to vote . . . and I remember her saying to me at the time, “I was 29 before I could vote”. That was a formative experience’. Carolyn, too, while her father had a middle-class occupation, noted how her grandfather’s background informed, years later, her interest as an MP in international development:

My grandfather joined the Labour Party at its creation in about 1900, a trade unionist. He started work in a pit at the age of 10. And I never forget that. . . . when I looked at developing countries, you could see children like my grandfather.

But the role of the extended family was especially salient in the case of Conservative MPs, often because the family network invoked as part of the process of political socialisation was significantly more extensive. As Frank recalled during the face-to-face interview:

on the wall, there is a photograph of a statue of . . . my great-great grandfather, who was the MP for [town in northern England] from 18xx to 18xx. And apparently, I asked what he was doing on the statue, was told he was an MP, and I said, ‘Well, I shall become one of those’. . . . And also, it is said if you’re not a socialist at 20, you have no heart, if you’re not a Tory at 30 you have no head. Well, I’ve said, by that criterion, I’ve been heartless from the age of nine.

Similarly, Roy, another former Conservative MP from middle-class origins, attributed his particular brand of conservatism to one of his grandfathers:

my paternal grandfather in Nottingham, I remember finding an article he’d written criticising the monarchy – not because he was against the actual monarch, but he was against all the hangers-on and the court, and thought that was a waste of public money. So it was quite a radical point of view, which I probably inherited. I was a bit of a radical Conservative.

While parental occupation was dominant in shaping interviewees’ perceptions of their class origins, this class identity was also often influenced strongly by the wider local communities they had been embedded in growing up. Indeed, this more collective source of socialisation had often played a significant role in forging nascent political ideologies. Recalling his childhood, Howard could see the impact his community had on his politics:

The estate was very, very political. . . . I remember the Conservative candidate once tried to come onto the estate and was chased off. I never saw a Conservative poster till I went to university. . . . There was always lots of poverty. There were people in poverty, there were people having to move house very quickly because of debts. There were people going to Australia to make new lives. There were people working in factories, there were trade unionists, there were communists. . . . And the local Labour councillor lived about four doors away from me. . . . Just by chance, today is the day that Jim Callaghan became Prime Minister of the UK in 1976. And the day after that in 1976 I walked over the road and said to the local Labour councillor, ‘Can I join the Labour Party?’.

This does not mean that the process of political socialisation was linear in every case, with kids from working-class backgrounds automatically becoming left-leaning politicians. Some of our interviewees narrated their journey towards the political right as being in spite of growing up in working-class communities. Lawrence’s case is particularly apt here. A Conservative MP who spent his childhood in a working-class community in the north-east of England and described his social background as working class (even though he was rather middle class by parental occupation) said that his parents:

were quite contented in their circumstances. . . . I was very ambitious because I saw my family not being [so], and I didn’t want to spend my days in a council house or whatever. Maybe that was it. I always rather envied the middle-class lifestyle, I wish we were living in our own house, got a nice garden, got all those things. So, I was partially driven by a desire to have the things I can see other people [had].

And yet, Lawrence, still recognised the lingering influence of his early social environment on his outlook, especially when compared to other Conservative MPs:

I just felt that some of my colleagues didn’t quite see that. I mean, you know, Mrs. Thatcher had married a multimillionaire oil family chap, so she’d had nannies for children and all that. So, she had no concept of, I mean, my mother used to go to work at 5:30 in the morning, as a cleaner, so she could be back home before we got up for breakfast. I think if you’re from a very wealthy, privileged background, and you entered parliament because your family’s been there for generations too and all the rest of it, there’s no way you can see things in the same light.

Thus, despite his party allegiance, Lawrence’s background has left a lasting mark on his current outlook:

My father was on the National Committee of the National and Local Government Officers’ Association. So there was that in the background. I’m a great believer in checks-and-balances in politics, and I think that’s what keeps democracy and economies functioning well. . . . I’m not a mad ‘rights of the worker under any circumstances’ exponent, but I think, given the power of corporate entities, organised labour is a very important check-and-balance to that. . . . I think matters have gone out of kilter for those who are in the sort of lowest quartile of economic survival.

In the narratives of the politicians we spoke with, then, the more sociological idea of class origin had very often left a powerful mark on political outlook much later in life. Often this was directly through the influence of parents, but it was also augmented by the impact of grandparents and wider community. This link was not deterministic in the sense that working-class kids become Labour party politicians; but even when their journeys did involve them moving away from the politics of their parents, they still found their outlook shaped in part by their family background.

Consolidating Political Outlooks: The Role of Occupation and Education

Class origins were clearly important but what about MPs’ own occupation and education? Most interviewees we spoke with barely addressed the influence of their class destinations (their occupation) or indeed their education in the formation of their political outlook. They certainly talked at length about their work experiences prior to entering office but they did so in rather descriptive terms, without linking it explicitly to their political views or broader outlook. Some, like Gordon, a former Conservative MP who worked in the corporate sector before being elected in the mid 1980s, was unequivocal in this regard: ‘I don’t think that my business experience taught me anything; well, it didn’t make me change my views, but I think what it did do was give me some real practical experience of seeing businesses at work’. Former Labour MP Howard also emphasised the pre-eminence of his class origins in shaping his outlook:

I might have been a university graduate, I might have had a managerial job, I might have been a minister in the government, but the first 18 years of my life is what has coloured my political and social outlook and how I feel about myself now.

When our interviewees did address that link directly, they typically saw their occupations as slightly modifying a political orientation that had been acquired through the family. We met Carolyn above. She was a former Labour MP who had what might be described as centre-left views. Her father and grandfather exerted the most formative influence on her outlook, but she also noted that her job in the private sector as a lawyer did further solidify certain positions:

I was in a non-unionised workplace. . . . And, actually, having been in business, you can understand some of the pressures that people are under and trying to keep going and commitment to staff and the business and how that can, whichever trial comes towards you, you’ve got to face and get over and get to the next day and get the wages paid. I understood those pressures, so I think that helped me.

Indeed, when asked about her view on bringing key utilities such as energy back into public ownership, Carolyn framed her answer in light of her work experience, thus suggesting that the latter played a role in shaping her view on that topic:

I worked in [an energy company], it’s one of the few remaining companies now called a vertically integrated company. . . . But actually, energy is now produced by a myriad of different companies. The distribution networks are different. And the retail – folks are talking about nationalising the retail, why would you? It’s a basket case!

This consolidating influence of occupation can also be seen in Joseph, a Labour MP:

My dad was involved in politics, but he had to be neutral because of his job. . . . So it came from my dad. And then of course when I got to the shipyard . . . I became a convenor shop steward. . . . At the same time, Jim Conway, who’s the General Secretary of the Amalgamated Engineering Union at the time, I became his driver if he came to Scotland. He was an incredible sort. All he needed to do was take his glasses off and put a bonnet on and walk into the pub next. A thousand folk had been listening to the last one and nobody recognised him in the pub.

In other words, if Joseph got his political leanings from his father, they were further consolidated by important figures that he subsequently came across in his line of work, which suggests that if occupational background does play a role in political socialisation, it does so not merely through the nature of the occupation (e.g. blue vs white-collar) but also thanks to the peer socialisation experienced in the respective occupation.

Education was even less central to these narratives than occupation. When educational settings did play a role, it was again mostly in a consolidating and refining rather than transformative way. Gordon was raised by his middle-class mother and grandmother, who ‘didn’t like The Labour Party . . . both had pretty right-wing views’, but he was also further influenced in that direction by his art teacher, who ‘was very interested in politics, and was very much a small government free market Conservative’.

What education does seem to have done was making people more conscious of their own class origin relative to that of their colleagues, as illustrated by Arthur, a Labour MP of working-class origins:

I remember thinking at one stage that another lad in another class looked like he might have a similar background to me, and we became friends, and he invited me around to his house, and it was like a palace. So he never did come to my house. . . . most of my friends were going to Oxford or Cambridge, and I didn’t want to go to those places because I didn’t want to live there. I didn’t want to be with the people that I went to school with.

Oxford, in particular, which a sizable share of past and present political elites attended (see Kuper, 2022), emerged from some of the interviews as a site where people defined themselves in relation to students from more privileged backgrounds. While coming from a middle-class background, Labour MP Lisa said:

I was conscious . . . and I think generally I just found the ones who were openly Conservative not sympathetic people. . . . So I could just see those people had absolutely no consciousness of other lives that people lead just down the road from us.

The case of Timothy, Conservative MP, is even more revealing, despite his solid middle-class origins:

So I found that in the elections I participated in in the Conservative Association and the Oxford Union, that it was very much a significant block of Old Etonians and other privileged private schools against the grammar schools, slightly less prestigious public schools, like I had been to. To some extent, I was the [relatively deprived area in South London] candidate. Quite a lot of those people looked down their noses at somebody who came from that background. . . . And to some extent, I think that influenced my politics. I felt that I was quite taken aback by what I saw as a privileged attitude and almost a lack of awareness of some of the stresses and strains that everyday people face in life.

The accounts that our politicians offer suggest that occupation and education mattered, but mostly to the extent that they moderated a set of already existing ideas about the world. They were, in other words, consolidating agents of political socialisation, rather than decisive drivers.

From Attitudes to Action

Class origins, then, do not single-handedly determine political outlook later in life but our results suggest they strongly shape the ideologies of British politicians. We cannot assume, however, that class-based political outlooks will always translate into political behaviour. In this vein, we explicitly explored whether and how class background may have informed the political behaviour of our interviewees. Narratively, this usually meant exploring instances where they felt their political outlook conflicted with their actions as a policymaker. What is striking here is that the vast majority of the people that expressed some conflict in that regard were those who came from working-class backgrounds; hence, we focus on that group in this part of the analysis primarily because our middle-class respondents had far less to say about this.

However, there were concrete examples of a political outlook rooted in class origins directly informing decision-making. Howard, a Labour MP of working-class origins from the 2010–2015 legislature, recalled the austerity measures implemented by the Conservative-led coalition government:

I voted against them when I was in Parliament. And the reason is again, because when austerity happens, it’s always the poorest in society who end up being hurt by austerity. . . . I’ve grown up – and it goes back to values again – I’ve grown up where we didn’t know what we were eating tomorrow.

A similar link between a working-class environment and opposition to austerity was made by Matthew, also Labour, who, while coming from middle-class origins himself, emphasised the importance of the local community he grew up in and was still living in during his time in office:

these weren’t just people that you happened to represent, these were your friends and your neighbours that you were living alongside who were experiencing this. And they knew that I was one of them and so it makes that relationship, I think, even deeper. . . . I recall one example of where there was the bedroom tax, one of the more iniquitous cuts the Conservatives brought in. So, visiting a lady less than a mile from me, a disabled lady. And she had the spare room and they were taking £20 a week off her. And that just damaged her quality of life and she was so upset by it. . . . we were able to get the case dealt with but there was that wider political point that you had people from Eton, from Oxbridge, who really just didn’t care.

Indeed, coming from a less privileged background also seems to have cut through in the case of Lawrence, Conservative MP under Thatcher, even if it came at a significant political cost for him:

I revolted over nurses’ pay under Mrs. Thatcher. . . . It had been agreed that judges and senior-Army-officers would get their increase . . . but the nurses not. I just thought that was not right morally, and I voted against the government on that. I paid a heavy price for it. . . . it was made clear to me I wouldn’t be offered any kind of position in Mrs. Thatcher’s government . . . not a person I warmed to: she was everything about the middle-class that I tended to revile against.

More often than not, however, the number of instances in which the political outlooks inherited from their class origins led to rebellions against the party were uncommon. The first and most obvious reason for this is party discipline (Carnes and Lupu, 2015). For instance, Matthew, quoted above for having opposed a specific austerity measure due to his embeddedness in a working-class community, also rationalised his abstention in the vote on other austerity measures by emphasising the importance of party discipline and cohesion:

The two biggest issues I remember, where there was a lot of consultation in the parliamentary Labour Party, there was a vote about benefits, the benefits cap, the Conservatives were bringing in, this twenty-thousand-pound benefit cap. And you know, Labour was either gonna abstain or vote against. Now, we had a discussion, detailed discussion in the Scottish group and also with the whips more generally . . . about whether we were doing the right thing in abstaining. So, there was a discussion over that. But my view was you take a collective decision. There’s nothing worse than people virtue-signalling and saying ‘well, I see what the broad number of my colleagues are gonna do and I have sympathy with that but I’m just going to do something different, I’m going to be a virtue-signalling rebel’. . . . I mean, I think, I was never one of life’s rebels because I just think you’re able to do more in politics, you have to work as a team and do things collectively.

In other words, Matthew’s ties with his working-class community came into play only when they did not clash with the line of his party. When it did, he chose the latter, and doing otherwise would have been, according to him, a form of ‘virtue signalling’. Howard, another Labour MP from working-class origins, quoted above for the defining influence of his ‘first 18 years of life’ over his political outlook, said that:

I always voted with the party line. . . . I wasn’t particularly keen on the Iraq War, but I had to live with that and argue that and I had to vote for that and go back to the constituency and explain why we’d done it at the time. . . . Because it’s a party decision, it’s a party collective.

While Matthew and Howard justify their adherence to the party line in terms of ‘doing things collectively’, others, like Malcom, mentioned they owed it to the party for having been elected in the first place: ‘you’ve got to recognise that you would not have gotten elected if you hadn’t had a ballot paper with Labour against your name. . . . So obviously, you do owe loyalty to the party’. In addition to that, defying the party whip may affect the MP’s career prospects, as illustrated earlier by the case of Lawrence; it was the same for Carolyn, who voted against her government on the Iraq War: ‘I recognised that the consequence was, for me, promotion, which some people had spoken about at that time . . . I think it’s why I didn’t get promotion before 2005’.

And yet party discipline was not the only reason why class origin-informed political outlook did not always translate into action. For some of our respondents from working-class backgrounds, a more decisive factor appeared to be a relative lack of self-confidence they experienced while operating in parliament. Valerie, a Liberal Democrat MP observed:

you know, having watched my parents’ life, I’ve always been aware that you could easily fall back down the scale, whereas I think if you were born middle-class, you’ve not seen or experienced people falling into poverty, then you have this sort of arrogance that you’ll always survive and you’ll always rise.

Labour MP Clive elaborated on that:

you are always more risk-averse, I think, coming from not having wealth as much as anything, not having alternatives. I needed my job to pay the mortgage. I didn’t have money to buy special advisors and support . . . You can’t afford to be a swashbuckling cavalier. You’ve got to be more cautious in the things that you do, hold back, take care. I’m certain. I have regrets. ‘Why didn’t I go that little bit further?’ I know why I didn’t go that little bit further, because I didn’t think I had the power. I can only play the cards I’ve got.

In contrast to that, a Conservative MP of lower-middle class origins, Gordon claimed that ‘what the whips learnt about me was that . . . if I was going to rebel against the government, there was almost no chance they were going to dissuade me from doing so’.

It is true, though, that we observe this self-confidence deficit even with some MPs from more privileged backgrounds but who felt they were not as privileged as some of their colleagues, as in the case of Conservative MP Edward:

I did lack that, the confidence that the private education gives you as well, there’s another thing. My friends, I have many friends from Eton now, they have a greater confidence typically than grammar schoolboys. So, I did very well and I’m very pleased that I did, I’m very happy, I’m not dissatisfied in any sense. . . . I think I would’ve done better if I’d gone to a private school.

That boosting effect of elite education was also acknowledged by someone from middle-class origins and educated at Oxford, like Timothy:

if you’ve been to Oxford, I do think that it gives you a certain self-confidence, a certain self-reassurance, which is a privilege. I have a 14-year-old daughter, and I want her to go to Oxford and I’m going to try my best to get her in.

In the particular case of MPs from working-class origins, that self-confidence deficit may be linked to their worry of potentially jeopardising the gains entailed by the social mobility experienced by becoming part of the political elite. This would fit with the existing evidence that MPs coming into office from a career in politics, and who tend to come already from more privileged class origins, are more likely to defy the party whip (see Heuwieser, 2018). ¹

While class origins informed these MPs political outlooks, there were various constraints on their ability to translate their values into action. There were certainly key moments when their class-based values conflicted with the policy direction of their party, but party discipline often (albeit not always) superseded their sense of what was right. But whether MPs felt able to rebel against party discipline was influenced by class too, with working-class MPs feeling less able to rebel because they also felt more insecure about their position.

Discussion and Concluding Remarks

In this article, we use 24 interviews with largely former MPs to explore how they understand their own political formation. What is crucial for these political elites was class origin rather than pre-parliamentary occupation, with those coming from working-class backgrounds leaning more to the left than others. In this respect, our data cut against existing work, which is almost entirely quantitative and relies on existing datasets that cover the professional and education background of British political elites, but not so much parents’ occupation (but see Bukodi et al., 2024), let alone the influence of grandparents or the wider local community.

Equally important here is that class origin is not merely determining ideology in a linear way – for example, that working-class people are always on the left – but rather it appears to shape their ideological location with respect to other more privileged colleagues. Conservative MPs from working-class backgrounds tend to present themselves as more sympathetic to working-class interests than party colleagues from more privileged backgrounds, while Labour MPs from a middle-class background tend to position themselves on the right of their party. In line with recent research on the process of preference formation among the wider public (Ares and Van Ditmars, 2023), this raises questions about the centrality of occupation as the key indicator of class in the current political science literature on political elites. Indeed, it might well be that this pre-existing outlook informs the choice of occupation, and is then consolidated or moderated by the latter, rather than the other way around. Similarly, our data indicate that education only plays a consolidating role in the process ideology formation, thus giving credence to the minority opinion in the current literature on this topic (see Simon, 2022).

At the same time, the interviews articulated rich and multifaceted conceptions of class origins. Although family background appears to be, by far, the single most influential factor in the process of political socialisation, family background was not merely a matter of father’s occupation. Political socialisation was attributed to the influence of both father and mothers, as well as grandparents, and in that respect, it aligns with recent work on political socialisation (Fasching and Lelkes, 2024; Fitzgerald and Bacovsky, 2024). More broadly, the class position of these politicians did not only stem from the specific economic situation of their families, but also entailed being a member of a community who shared many of those same economic challenges, although that might have seen considerable change as a result of the neoliberal transformations around housing and community cohesion in Britain over the past few decades (e.g. Flint and Robinson, 2022) and further suggests that understanding the impact of class origins on politics requires richer measures than simply father’s occupation.

The relationship between class origins and actual policymaking is, however, more complex. While the interviewees described several episodes that illustrated how the former informed their stance on certain policy issues, generally their behaviour was mostly guided by the party line (Carnes and Lupu, 2015). In other words, some notable exceptions notwithstanding, MPs’ class origins inform their policymaking behaviour as long as it does not directly clash with party discipline. This may suggest that, at least in the case of political systems characterised by strong party discipline, such as the UK, enhancing the descriptive representation of the working class might not necessarily lead to an improvement in the substantive representation of this social group if it cannot also shape the agenda of their party. Thus, the degree of power of political elites may be a distinct factor mediating the influence of class background on political behaviour. The fact that none of our interviewees had the top agenda-setting roles in their parties is one of the limitations of our data and future research could focus on those who do.

The interviews also uncovered an additional potential inhibitor of a working-class background informing parliamentary behaviour: the lower perceived level of self-confidence on behalf of MPs from working-class origins relative to those from more privileged social backgrounds. Given the aforementioned prevalence of the latter in the UK parliament and their lower level of support for pro-working-class policies, it is not far-fetched that this relative deficit of self-confidence among working-class MPs would further dissuade them from pursuing more left-wing policies, particularly so considering the neoliberal turn in British society over the past decades (Davies et al., 2021). Indeed, as some interviewees alluded to, it may be the preservation of the benefits entailed by making it into the political elites that may further prevent socially mobile working-class MPs from acting according to their class-based outlook. Thus, further research could look into how class origins may interact with the current financial interests of political elites – gained not merely prior to but also during their time in office – in (re)shaping their ideological outlook and informing their policy preferences.

One limitation of our analysis is that it has been unable to disentangle the degree to which MPs in general (and perhaps working-class MPs in particular) ground their ideologies in their class origins in response to the ‘symbolic market for ordinariness’ that surrounds elites more broadly. In this symbolic market, elites are seen as better able to understand everyday people, more trustworthy, and even more competent if they come across as ordinary or regular people (Reeves and Friedman, 2024). The motivation to do so is perhaps particularly salient for politicians whose careers rise and fall on their ability to convince citizens they are trustworthy and competent. In other words, class background may be deployed by some political elites in their quest for legitimacy and political capital (which might also explain the discrepancy noted in some cases between objective and subjective class origins). If this is true, it may be that their self-professed (and often class-inflected) ideology is partly explained by this desire to perform ordinariness and should not therefore necessarily be expected to align with their actions while in office. To return to the example of Keir Starmer at the top of the paper (see also Riley-Smith, 2024), some may be inclined to read these and other similar comments in this way in part because Labour have weakened some pro-working-class pledges under his leadership (Rea and Chambre, 2023). Addressing the impact of the symbolic market for ordinariness on the disconnect between attitudes and action would require more sophisticated forms of data collection than we have been able to pursue here; but this remains an important avenue for future research and one which could potentially help us understand whether class-based descriptive representation would in actuality translate into substantive representation.

In sum, this article has built upon the narratives of MPs about their political socialisation to suggest that class origins – taken in a multidimensional way to include family network and community of origin – is particularly salient in how they understand the development of their political outlook. In this respect, our findings run counter to much of the political science literature in this area. Of course, the impact of class origin on their political behaviour as MPs is significantly moderated by party discipline. MPs tend to follow the party line over their own outlook when the two happen to clash and this may well be due to pragmatic considerations about what it takes to be an effective policymaker. For MPs from a working-class background, though, their relatively lower degree of self-confidence and tacit concern about falling back down the social ladder may further contribute to the inhibition of their views. One possible implication of this is that political elites may be inclined to rhetorically mobilise their working-class origins largely in order to gain symbolic capital and that, therefore, increasing the number of MPs of working-class origins may not necessarily alter the degree to which pro-working-class policies are implemented.