Download our free export of the 1,300 companies that have raised funds!
World database
France database

No obvious correlation between industrial employment and the RN vote

David

June 19, 2024

For Le Mondelast week, Bastien Bonnefous asked us about the possible link between industrial employment and the RN vote. He wondered why there was a strong RN vote in areas where many industrial investments have been announced (notably in Dunkirk). Our first reaction was to say that many of these projects had not yet seen the light of day. Then we took a closer look at whether there was a link between the RN vote and indicators taken from the Trendeo France database.

We tested the link between the RN score, at departmental level, and our data over three periods (not knowing whether a department is colored according to the state of the economy since 2009, 2020 or 2023). The data tested are quite diverse, and the correlations mentioned below ⤵️


Main observations:

- no very strong link (maximum correlation is 100%) between an economic aggregate and the NR vote: there is no single explanatory factor;
- the shorter the reference period for the aggregate, the stronger the correlation, as if electoral memory were short-term for the aggregates studied;
- job losses are the first explanatory factor we would expect. They are more important than strictly industrial job losses;
- job creations are also correlated with the RN vote: this led us to think that there was probably a size effect. The size of the département, measured by its population, is in fact correlated at 20% with the RN vote;
- more"sophisticated" indicators such as the net balance between job creations and job losses, or their ratio, are very weakly correlated;
- the most surprising result is obviously to find a maximum correlation with job creation in the digital sector (and also, at a lower level, with fund-raising).

Before attempting an explanation, it would be worth going down to commune level for finer correlations than at département level.
It would also be necessary to correlate one flow (job creations or losses over a period, or another aggregate) with another flow (the evolution of the RN vote between two dates).
Further statistical analysis could finally allow us to take several factors into account.

If we were to attempt an interpretation, we'd say that the RN vote in a département where digital employment is on the rise may stem from an electorate that feels left out of a modernization phenomenon (via phenomena such as rising rents?). This feeling of being sidelined would be even stronger than job loss.

We hesitated to publish such a surprising result, with a rather weak statistical link. But we remember reading an article that regretted that negative results were not published more often. Here's an example of a mixed result, which I hope will stimulate further reflection!

.

👉SeeLe Monde article: https: //www.lemonde.fr/economie/article/2024/06/17/la-reindustrialisation-un-rempart-fragile-contre-le-rn_6240865_3234.html