Rank-index method

Part of the Politics and Economics series
Electoral systems
Comparison of electoral systems Social choice theory Mechanism design List (By country)
Single-winner methods Single vote-runoff methods Plurality (FPP) Two-round (US: Jungle primary) Related: Partisan primary Instant runoff (IRV) / Alternative vote (US: Ranked-choice/RCV) Condorcet methods Condorcet-IRV Minimax Kemeny–Young Nanson Schulze Ranked pairs Maximal lottery Positional voting Plurality(el. IRV) Borda count (el. Baldwin) Antiplurality (el. Coombs) Cardinal voting Score voting Approval voting Majority judgment STAR voting Quadratic voting
Proportional representation Party-list Nonpartisan proportional representation List type Closed list Open list Panachage Localized list Apportionment Highest averages Largest remainders National remnant Biproportional Quota-remainder methods Hare STV Schulze STV CPO-STV Quota Borda Cumulative Approval-based committees Thiele's method Phragmen's method Expanding approvals rule Method of equal shares Fractional social choice Direct representation Interactive representation Liquid democracy Fractional approval voting Maximal lottery
Mixed systems By results of combination Mixed-member majoritarian Mixed-member proportional By mechanism of combination Non-compensatory Parallel voting (superposition) Coexistence Conditional Majority bonus (fusion) Compensatory Additional member (seat linkage) system (MMP, AV+) Vote linkage system (Scorporo, MBTV) Dual-member proportional Rural–urban proportional Majority jackpot Supermixed systems By ballot type Single vote Two vote Mixed ballot
Voting system criteria Spoiler-resistance Condorcet criterion Smith independence Spoiler independence Clone independence Strategy-resistance Lesser evil voting Binary independence Maskin monotonicity Later-no-harm Later-no-help Rational response Participation criterion Positive response Reinforcement criterion Reversal symmetry
Social and collective choice Impossibility theorems Arrow's theorem Majority impossibility Moulin's impossibility theorem McKelvey–Schofield chaos theorem Gibbard's theorem Fishburn's example Positive results Median voter theorem Condorcet's jury theorem May's theorem Campbell-Kelly theorem Harsanyi's utilitarian theorem Tournament sets Smith set Landau set Copeland set Split-cycle set Bipartisan set
Politics portal Economics portal
v t e

In apportionment theory, rank-index methods^[1]: Sec.8 are a set of apportionment methods that generalize the divisor method. These have also been called Huntington methods,^[2] since they generalize an idea by Edward Vermilye Huntington.

Like all apportionment methods, the inputs of any rank-index method are:

• A positive integer ${\displaystyle h}$ representing the total number of items to allocate. It is also called the house size.

• A positive integer ${\displaystyle n}$ representing the number of agents to which items should be allocated. For example, these can be federal states or political parties.
• A vector of fractions ${\displaystyle (t_{1},\ldots ,t_{n})}$ with ${\displaystyle \sum _{i=1}^{n}t_{i}=1}$, representing entitlements - ${\displaystyle t_{i}}$ represents the entitlement of agent ${\displaystyle i}$, that is, the fraction of items to which ${\displaystyle i}$ is entitled (out of the total of ${\displaystyle h}$).

Its output is a vector of integers ${\displaystyle a_{1},\ldots ,a_{n}}$ with ${\displaystyle \sum _{i=1}^{n}a_{i}=h}$, called an apportionment of ${\displaystyle h}$, where ${\displaystyle a_{i}}$ is the number of items allocated to agent i.

Every rank-index method is parametrized by a rank-index function ${\displaystyle r(t,a)}$, which is increasing in the entitlement ${\displaystyle t}$ and decreasing in the current allocation ${\displaystyle a}$. The apportionment is computed iteratively as follows:

• Initially, set ${\displaystyle a_{i}}$ to 0 for all parties.
• At each iteration, allocate one item to an agent for whom ${\displaystyle r(t_{i},a_{i})}$ is maximum (break ties arbitrarily).
• Stop after ${\displaystyle h}$ iterations.

Divisor methods are a special case of rank-index methods: a divisor method with divisor function ${\displaystyle d(a)}$ is equivalent to a rank-index method with rank-index function ${\displaystyle r(t,a)=t/d(a)}$.

Every rank-index method can be defined using a min-max inequality: a is an allocation for the rank-index method with function r, if-and-only-if:^{[1]: Thm.8.1}

${\displaystyle \min _{i:a_{i}>0}r(t_{i},a_{i}-1)\geq \max _{i}r(t_{i},a_{i})}$.

Every rank-index method is house-monotone. This means that, when ${\displaystyle h}$ increases, the allocation of each agent weakly increases. This immediately follows from the iterative procedure.

Every rank-index method is uniform. This means that, we take some subset of the agents ${\displaystyle 1,\ldots ,k}$, and apply the same method to their combined allocation, then the result is exactly the vector ${\displaystyle (a_{1},\ldots ,a_{k})}$. In other words: every part of a fair allocation is fair too. This immediately follows from the min-max inequality.

Moreover:

• Every apportionment method that is uniform, symmetric and balanced must be a rank-index method.^{[1]: Thm.8.3}
• Every apportionment method that is uniform, house-monotone and balanced must be a rank-index method.^[2]

A quota-capped divisor method is an apportionment method where we begin by assigning every state its lower quota of seats. Then, we add seats one-by-one to the state with the highest votes-per-seat average, so long as adding an additional seat does not result in the state exceeding its upper quota.^[3] However, quota-capped divisor methods violate the participation criterion (also called population monotonicity)—it is possible for a party to lose a seat as a result of winning more votes.^{[4]: Tbl.A7.2}

Every quota-capped divisor method satisfies house monotonicity. Moreover, quota-capped divisor methods satisfy the quota rule.^{[5]: Thm.7.1}

However, quota-capped divisor methods violate the participation criterion (also called population monotonicity)—it is possible for a party to lose a seat as a result of winning more votes.^{[5]: Tbl.A7.2} This occurs when:

1. Party i gets more votes.
2. Because of the greater divisor, the upper quota of some other party j decreases. Therefore, party j is not eligible to a seat in the current iteration, and some third party receives the seat instead.
3. Then, at the next iteration, party j is again eligible to win a seat and it beats party i.

Moreover, quota-capped versions of other algorithms frequently violate the true quota in the presence of error (e.g. census miscounts). Jefferson's method frequently violates the true quota, even after being quota-capped, while Webster's method and Huntington-Hill perform well even without quota-caps.^[6]