Proportional Representation Simulator: D'Hondt, Sainte-Lague & Seat Allocation

simulator intermediate ~10 min
Loading simulation...
ENP ≈ 3.8 effective parties in a 50-seat legislature

With 5 parties, 50 seats, D'Hondt method, and 5% threshold, the effective number of parties (Laakso-Taagepera) is approximately 3.8 — smaller parties lose some proportionality to the threshold and D'Hondt's large-party bias.

Formula

D'Hondt quotient: votes_i / (seats_allocated_i + 1)
Sainte-Lague quotient: votes_i / (2 * seats_allocated_i + 1)
Laakso-Taagepera ENP = 1 / sum(p_i^2) where p_i = seat_share of party i
Gallagher index = sqrt(0.5 * sum((v_i - s_i)^2))

From Votes to Seats

The central challenge of representative democracy is converting millions of individual votes into a finite number of legislative seats. Proportional representation (PR) systems attempt to make seat shares match vote shares as closely as possible. But the mathematics of rounding discrete seats from continuous vote percentages creates unavoidable distortions — and different rounding methods create different kinds of bias.

Three Allocation Methods

This simulation implements three major PR allocation methods. D'Hondt (used in Spain, Belgium, Japan) divides votes by successive integers and awards seats to the highest quotients — it slightly favors larger parties. Sainte-Lague (used in Norway, Sweden, New Zealand) divides by odd numbers (1, 3, 5...) and is more favorable to smaller parties. Largest Remainder (used in Italy, Hong Kong) gives each party its integer quota of seats, then distributes remaining seats to parties with the largest fractional remainders.

The Threshold Effect

Electoral thresholds act as gatekeepers. A 5% threshold (common in Germany, Poland) means any party below 5% gets zero seats — their voters are effectively disenfranchised. Raise the threshold slider and watch small parties disappear from the legislature, their seats redistributed to larger parties. The 'wasted votes' metric shows how many citizens' preferences are excluded from representation entirely.

Measuring Fairness

The Gallagher disproportionality index quantifies how far the result deviates from perfect proportionality. The effective number of parties (Laakso-Taagepera) measures how many parties truly matter in the legislature. Compare these metrics across methods and threshold levels to understand the tradeoffs inherent in any electoral system design. No system is neutral — each embeds choices about fragmentation versus representation, stability versus diversity.

FAQ

What is proportional representation?

Proportional representation (PR) allocates legislative seats in proportion to each party's vote share. Unlike winner-take-all (FPTP) systems where the plurality winner takes the single district seat, PR ensures that 30% of votes translates to roughly 30% of seats.

How does the D'Hondt method work?

D'Hondt divides each party's total votes by 1, 2, 3, ... N, creating a table of quotients. Seats are assigned to the highest quotients one by one. This method slightly favors larger parties, which is why it is used in systems wanting to discourage fragmentation (e.g., Spain, Belgium, Japan).

What is the electoral threshold?

An electoral threshold is the minimum percentage of votes a party must receive to qualify for seats. Common thresholds range from 1% (Netherlands) to 10% (Turkey). Higher thresholds reduce fragmentation but disenfranchise voters of small parties.

What is the Gallagher disproportionality index?

The Gallagher index measures how disproportional an election result is: sqrt(0.5 * sum of squared differences between vote shares and seat shares). A value of 0 is perfectly proportional; values above 5 indicate significant disproportionality.

Sources

Embed

<iframe src="https://homo-deus.com/lab/voting-democracy/proportional-representation/embed" width="100%" height="400" frameborder="0"></iframe>
View source on GitHub