Divisions within the British Parliamentary Labour Party under Keir Starmer: Results of a Cluster Analysis

This paper offers a methodologically innovative two-stage approach for studying divisions amongst parliamentary representatives. Using the Parliamentary Labour Party (PLP) as our case study, we construct a dataset of all Labour MPs elected in the 2019 general election, along with their nominations in the 2020 Labour Party leadership and deputy leadership elections and their membership of, or affiliation with, various party-aligned organisations. We then conduct a cluster analysis based on this dataset, which reveals the existence of a two-cluster model-comprised of the Mainstream (N = 162) and the Left (N = 33)-and a five-cluster model-in which the Left exists alongside the Tribune Soft Left, the Labour Friends of Palestine and the Middle East Soft Left, the Unaligned Centrists, and the Right. Finally, we test the robustness of our clusters via a canonical correspondence analysis of the language used by MPs on social media (Twitter/X) and their contributions to parliamentary debates (Hansard). We show that the MPs from different clusters do use different languages to one another in both fora. We also find that the main divide within the PLP is between the left of the party and the rest of the party, and that the deputy leader, Angela Rayner, has a broader base of support amongst the PLP than its current leader Keir Starmer.