https://orcid.org/0000-0001-7483-5148
![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
In a recent paper, Piketty argues that the vote for the left in France, the UK and the USA tends increasingly to be associated with a high education level whereas a traditional class- or income-based divide separated left from right individuals in the 1950s and 1960s. The current situation would be characterised by a dominance of ‘elites’ in left and right constituencies: financially rich elites vote for the right (merchant right), high-education elites vote for the left (brahmin left). Using ISSP data for 17 countries, this paper tests the influence of income and education inequalities on political leaning and a variety of policy preferences: the support for redistribution, for investment in public education, for globalisation and immigration. Results show that income levels are still relevant for the left-right divide, but the influence differs across education levels. Our findings also point to a certain convergence of opinion among the Brahmin left and the merchant right, which could lead to a new political divide beyond the left and the right, uniting a bloc bourgeois.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Kitschelt (Citation1994), Oesch (Citation2008a,Citation2008b), Norris and Inglehart (Citation2019).
2 Kriesi et al. (Citation2006), Rodrik (Citation2018).
3 Gennaioli and Tabellini (Citation2018).
4 For instance, Autor et al. (Citation2013) have studied the negative effects on jobs and wages in US industries and regions with higher exposure to Chinese import competition.
5 Rovny and Rovny (Citation2017) find that “outsiders” tend to vote more for “radical” parties, left or right depending on how the insider-outsider divide is operationalised.
6 It is not entirely clear whether the reference is Gary Becker’s “human capital” or Pierre Bourdieu’ “social capital”.
7 Piketty (Citation2019) implicitly develops the idea of an intergenerational loyalty for left parties to explain the emergence of the brahmin left: whereas a majority of voters in the 1950s/1960s had a low education level and voted for left parties, their children and their grandchildren in the 1980s/1990s who benefited from mass education continued to vote for those parties. In this sense, the shift from a traditional to a brahmin left would not be based on a specific political strategy to attract more educated voters.
8 Piketty (Citation2019) points out the role played by increasing educational inequalities: education spending has become increasingly more concentrated among high-income households (and to a lesser degree among middle-income households) and has on average declined over time. Therefore, the marginalisation of the popular classes would contribute to increasing inequalities because the left has become increasingly indifferent to redistributive issues.
9 According to Piketty (Citation2019), each of these groups was represented by a candidate in the French presidential election of 2017. The four candidates obtained comparable scores (20 to 24%) in the first round.
10 Frank (Citation2005, Citation2017) analyses the transformation of the party system in the United States in the direction of an increasing opposition between low-income and low-educated voters (supporting the Republican Party) and high-income and high-educated voters (supporting the Democratic Party).
11 Expectations regarding social mobility may play a role too: middle income highly-educated individuals may expect to climb up the income ladder and may have mixed feelings about redistribution.
12 The following variables used are all reflecting the economic dimension of globalisation and immigration, thereby excluding any cultural dimension.
13 When respondents are able to report/estimate amounts of income as exactly as possible, the income variable is referring to the average monthly gross level (before taxes and all other deductions). If this cannot be done, classes or brackets are used.
14 Our classification refers to the 1997 ISCED Classification.
15 Supplementary tables are available to present the results of our regressions and the estimated marginal effects.
16 As robustness checks, we use an alternative variable on the feeling of (in)justice that rich people can buy better education. The opinions on the equality of opportunity give, however, a slightly different picture (Figure A1 in the appendix). Regression results indicate that the probability to consider that it is unjust that rich can buy a better education increases from the lowest level of education to the medium level for almost all income deciles (except for D5 and D7), but to decrease and remain stable among highest educated individuals for all deciles (except for D6). The question mixes redistribution and education issues, which may explain the difference with the findings of the previous question. It confirms however the existence of a divide regarding education that could split apart the low educated from the other individuals.
17 Here again, we find more nuanced results when explaining the opinions on the equality of opportunity (Figure A2): while the probability to think that it is unjust that rich can buy education is decreased with the education level among right respondents, we find a positive relationship among left respondents.
18 This result is weakly confirmed when one looks at the question on the possibility for rich people to buy a better education (Figure A2).
19 By contrast, we find that the support for large companies clearly increases with the education level (Figure A3), especially for high income deciles, whereas proposition G1 stated that the increase in support should take place at lower income levels. This points to the importance of the education level in the globalisation divide and points to a possible community of interests among educated groups on such matters.
20 The support for large companies increases with income level, more strongly among high-educated individuals (Figure A4), which invalidates proposition G2. This points to a divide between the highly educated and the rest that would tend to unite the bloc bourgeois.
Additional information
Notes on contributors
Bruno Amable
Bruno Amable is Professor of Political Economy at the University of Geneva, on leave from the University Paris I Panthéon – Sorbonne. His research focuses on the political economy of institutional change and the diversity of capitalism.
Thibault Darcillon
Thibault Darcillon is an Associated Professor of Economics at the University of Paris 8 (LED). His research interests are in financial macroeconomics and labour economics adopting a political economy approach, with a focus on finance and inequality in the OECD countries.