For the complete documentation index, see llms.txt.Skip to main content
District magnitude (M) refers to how many seats are filled in a given electoral district. A district that elects one member has a magnitude of 1 (single-member district, or SMD); a district that elects ten members has a magnitude of 10. Despite sounding like a technical detail, district magnitude is one of the most consequential variables in electoral system design, often considered more determinative of outcomes than the formal seat-allocation formula.
The political scientist Rein Taagepera and others have shown that, in proportional representation (PR) systems, higher district magnitude produces more proportional results and lowers the effective threshold for small parties to win representation. Where M = 1, as in the United Kingdom's House of Commons or U.S. congressional elections, the system mechanically favors larger parties and tends to produce two-party competition — a tendency Maurice Duverger described in his 1951 work Les Partis politiques. As M rises, the effective threshold falls roughly in proportion, allowing smaller parties to clear the bar.
Examples of how magnitude varies:
District magnitude also interacts with gerrymandering (mostly relevant in low-M systems), malapportionment, and legal thresholds. Reformers debating proportional representation, mixed-member systems, or ranked-choice variants almost always frame the choice around what M should be, because changing magnitude is often a more powerful lever than changing the counting rule.
Generally yes, but the seat-allocation formula and any legal threshold also matter. A high-M system with a 5% national threshold (like Germany's) can still exclude small parties.
M = 1. Since the Uniform Congressional District Act of 1967, all U.S. House seats are elected from single-member districts.
Empirical work by Taagepera, Shugart, and Lijphart shows magnitude has a larger effect on proportionality and party-system size than the choice between D'Hondt, Sainte-Laguë, or other allocation methods.