Across Europe, from Italy to Hungary through Finland and Greece, the far right is rising in the polls, governments are shifting right and the left is collapsing. Are we entering a new political era, and can the left ever return to power? In other words, is the rightward drift inevitable?
We examined historical voting data going back to the French Revolution – and the findings, contained in our new book, offer a more optimistic view of what may happen in the coming years. They highlight the potential strength of the left’s electoral base and suggest that tacking to the right on migration to win back the working-class vote is a political dead end.
In a nutshell, we studied the determinants of voting behaviour using data gathered at municipal – “commune” – level, covering every election (legislative and presidential) between 1848 and 2022 and the main referendums held between 1793 and 2005. The chief advantage of examining such localised data is that the 36,000 French communes offer a broad range of electoral profiles: they include very poor and very rich districts as well as a full range of industrial and occupational makeups, with varying proportions of graduates, migrants and so on.
This provides a very detailed understanding of long-term voting patterns. Furthermore, we were able to study the interplay between many different factors – including income, wealth, education and profession, but also the size of town or village and the type of geographical area where people live. This is something that cannot be done reliably with surveys, due to limited sample sizes.
Over the past few years, the view has taken hold that the working classes have entirely abandoned the left. Some even argue that the left in France has become a “bobo” vote, in other words, that its support is now drawn mostly from the better-off bourgeois-bohemian class. That particular perception is one largely invented by the rightwing media and promoted by conservative elites.
We show, however, that first: not only is this working-class switch away from the left not the case, but it has never been the case. When we examined every legislative and presidential election since 1848 (nearly 50 elections), we found that the richest communes have always, and systemically, voted much less for leftwing parties (historically the Communist party and the Socialist party, increasingly today La France Insoumise) than for the right, the centre right and the far right.
Similarly, the poorest communes have generally voted much more for the left, especially in cities. That remains true to the present day. The (often intentional) confusion comes from the fact that commentators tend to associate the working classes only with blue-collar manufacturing workers, forgetting that the average wage of supermarket cashiers, restaurant staff, cleaners, care workers and other service industry employees has fallen below that of manufacturing workers for several decades now.
In other words, the French political landscape can be described as follows: low-income urban voters, who tend to be mainly service industry employees and tenants, vote predominantly for the left, while working-class voters outside the main cities, who are mainly blue-collar workers and homeowners, are more likely to vote for parties of the far right.
Such a sharp division between low-income voters in cities and those in small towns or the countryside has not always existed. But since the 1990s (as similarly observed at the end of the 19th century), political conflict in France has been determined mostly by two factors: the urban-rural divide and socioeconomic status (income, wealth, education, home ownership). The left, in other words, has retained the votes of poorer people in urban areas, but only of poorer people in urban areas.
Perhaps more importantly, we show that the role of what we call “the geosocial class” has never been as important as it is today: the socioeconomic character of a commune together with its size allow us to explain more than 70% of the variance of the vote between municipalities at the last French presidential elections, compared, for example, to “only” 50% in 1981, when François Mitterrand won against Valéry Giscard d’Estaing, and 30% in 1848. Perhaps even more surprisingly, when we add measures of identity and immigration, the relevance remains almost the same, increasing to 72-73%). What does this imply?
First, contrary to what is often asserted without much empirical support, voters – including voters of the far-right RN (Rassemblement National) – do not vote mainly on immigration issues. Socioeconomic issues are the main determinants of voting choices. If blue-collar workers have shifted towards the far right in recent years, it is above all because they have suffered disproportionately from globalised trade and deindustrialisation, and a lack of access to public services. From this point of view, they have felt abandoned by the left in power over the past 40 years in France. Of course, this can teach us lessons about what is happening today in other European countries. Throughout Europe, the left needs to convince voters that it can provide adequate protection against social, fiscal and environmental “dumping” – if necessary through unilateral action.
Our findings do give cause for optimism: indeed, the lack of public services in rural areas, deindustrialisation, unequal access to property and widening inequality are all issues that can be addressed by implementing adequate policies. Identity politics, on the other hand, tends only to lead to increased tensions and conflict within society.
Parties of the left should also be buoyed up by the knowledge that, in doing more for poorer people in small towns and peripheral areas, they could enlarge their future electoral base and return to power. Importantly, we document the fact that rural and urban poor people have much more in common that is often thought, in particular in terms of inadequate access to public services and opportunities, and widening disparities with the richest municipalities.
People may argue that, contrary to what we show (using the proportion of migrants in the communes where voters live), voters do care about immigration when they are asked. Is there a contradiction? We think not, for at least two reasons. First, polls give only a limited historical perspective, which is why we chose to rely on commune-level voting data in our book. As a result, it is hard to claim reliably that people care “more” about immigration today that in the past. Second, it is important to see that, historically, in a country like France, far-right voters have indeed voted on immigration (this was the case in the 1965 presidential elections with the candidacy of Jean-Louis Tixier-Vignancour, and in 1974 with Jean-Marie Le Pen). But the relationship between far-right voting and the share of migrants has changed over time, and this is no longer the case. Hence, voters’ motivations appear to have changed over the past 20 years – an issue that has been overlooked.
Nevertheless, this does not mean that no far-right voters are anti-migrant; some most certainly are, in particular Éric Zemmour’s electorate in France. But this electorate is not a working-class one; it is one of the most “bourgeois” electorates in French history (in terms of either voter income or wealth). Nor does it mean that the immigration question is a simple one, or that the refugee crisis can be easily solved.
Can these conclusions be applied to other countries, or should our optimism be limited to the new left alliance in France? Of course, our methodology needs to be extended to other electoral democracies, and we hope to do so. But we see little reason to expect poor voters in France to behave differently from those in other western democracies, particularly given that they are facing many similar threats, from deindustrialisation to unemployment, the cost of living and climate change. Parties – and the media – across western countries may be giving too much importance to the politics of migration, and in doing so they have lost sight of what matters to voters. We hope our research can help to refocus the debate.
Julia Cagé and Thomas Piketty’s Une histoire du conflit politique. Élections et inégalités sociales en France, 1789-2022 is published this month. An English translation, A History of Political Conflict: Elections and Social Inequalities in France, 1789-2022, will be published next year