Universal Scaling in Elections
It is a phenomenon observed globally: large cities tend to vote differently than rural areas. In recent US elections and the UK's Brexit referendum, this divide appeared stark. However, our analysis reveals that this is not chaos, but a manifestation of strict Urban Scaling Laws.
Just as infrastructure or patent production scales with city size, political opinion follows a predictable power law. By analyzing data from 1948–2016 (US) and the 2016 EU Referendum (UK), we found that political parties consistently split into two scaling categories.
Superlinear vs. Sublinear
One party always exhibits superlinear scaling (exponent β > 1), meaning their support grows disproportionately faster as the city gets larger. In modern history, this applies to the Democrats (US) and the "Remain" vote (UK).
Conversely, the opposing side—Republicans and "Leave"—follows sublinear scaling (β < 1). Their strength lies in the absence of agglomeration.
We adapted the model of "creative diversity" to explain this. In cities, social diversity acts as an agglomeration force. Voters in dense environments are exposed to a higher number of "complementary factors" (diverse social groups), leading to higher tolerance. This drives the superlinear growth of liberal-leaning votes, often leading to phenomena like the "Immigration Paradox."
A Single Parameter
Remarkably, the entire political landscape can be described by a single parameter. The behavior of the sublinear party is merely a mathematical consequence of the superlinear party's scaling and probability conservation. One side drives the dynamic (urban agglomeration of tolerance); the other fills the remaining space.
← Return to Main Page