Electoral Systems in Advanced Democracies
Introduction
The literature on electoral rules and political systems is heavily focused on the effects of electoral laws.
Less attention has been given to the causes of cross-national variation in electoral laws, with exceptions like Rokkan (1970) and Rogowski (1987).
The article aims to identify conditions under which ruling parties choose electoral rules to maximize parliamentary representation and cabinet posts.
Electoral rules encourage strategic behavior among elites and voters.
Voters avoid wasting ballots on hopeless candidates.
Elites concentrate resources on candidates expected to win.
The degree of strategic behavior varies with the electoral rule's entry barrier.
Governments shape electoral rules to their advantage, anticipating the consequences. If the electoral situation and rules benefit the ruling parties, there is no change. When the electoral arena changes, governments consider altering the system: maintaining high entry barriers (plurality rule) if it reinforces dominance or lowering thresholds to increase proportionality if current rules erode parliamentary power.
Electoral Systems Development in the Developed World
Electoral systems remained unchanged during limited suffrage.
With universal suffrage, left-wing voters entered, creating a new electoral arena.
The plurality/majority system survived where:
The new entrant (socialist party) was weak.
An established nonsocialist party retained a dominant position.
Proportional representation was adopted where:
The socialist party was strong.
Nonsocialist parties controlled similar electorate shares.
Policymakers lost interest in modifying the electoral regime once the electoral arena stabilized and the party system froze.
Abrupt changes in electoral laws have been rare, except in nations with unsettled party systems.
Ethnic or religious fragmentation encourages proportional representation under certain conditions.
The idea that states choose proportional representation to maximize social welfare is unconvincing.
A Theory of Electoral System Selection
The Rokkan Hypothesis
Rokkan interprets that PR rules were introduced via pressures from below and above.
The rising working class wanted access to legislatures.
Threatened old parties demanded PR to protect their position against mobilized voters created by universal suffrage (Rokkan 1970, 157).
Lijphart (1992, 208-9) suggests universal suffrage forced ruling elites and challengers to introduce PR to protect their interests.
Rokkan's argument is underspecified; it doesn't detail the conditions under which policymakers feel threatened enough to change the system.
If $\text{PR}$ is adopted to lessen the chances of a socialist party winning an absolute majority, Great Britain (as well as Australia and New Zealand) embraced universal suffrage without shifting to $\text{PR}$ (as did Sweden and Denmark).
If electoral rules changed due to the extension of universal suffrage, France, Greece, and Spain shifted back and forth between plurality/majority rule and $\text{PR}$ systems over the last century.
Analytical Generalization of the Rokkan Hypothesis
To understand shifts to $\text{PR}$, the argument is structured around three steps:
Consequences of Electoral Rules
Electoral rules encourage strategic behavior, forcing coordination on a reduced set of candidates.
Voters avoid candidates expected to do poorly, supporting second-ranked candidates.
Elites avoid wasting resources on hopeless candidates.
Votes and resources flow to stronger candidates, leading to only a few viable candidates competing.
Strategic behavior varies with the electoral rule.
In single-member plurality systems, voters coordinate around two candidates (Cox 1997, 69-79; Duverger 1954, 217).
Strategic voting declines as proportionality increases.
With $\text{PR}$, voters have fewer incentives to abandon preferred candidates.
The number of viable candidates increases with $\text{PR}$.
Calculations of Rulers and the Stability of the Electoral Arena
Ruling parties choose the regime that maximizes their chances of staying in power, anticipating the effects of different electoral regimes.
Governing parties have no incentive to change the electoral system as long as the electoral arena does not change and the electoral rules serve them well.
The ruling parties consider modifying the electoral system to maintain their political advantage as soon as change takes place and the structure of partisan competition starts to unravel.
Electoral systems were stable throughout the 19th century.
Universal suffrage led to a dramatic change in political competition, increasing the electorate and shifting from rural-urban conflict to capital-labor cleavage.
Old parties' elites had a strong incentive to reshape electoral rules due to the massive entry of new voters transformation in the preferences of already enfranchised citizens threatened the electoral strength of the old parties substantially.
Reform of the Electoral System as a Function of the Viability of the Old Party System
The shift to $\text{PR}$ is not an automatic response to a changing electoral arena.
The viability of the ruling parties in the future determines the extent to which the ruling parties embrace $\text{PR}$.
The strength of the new entering parties (the socialist party) and the coordinating capacity of the ruling parties determine the electoral viability of the ruling parties in the future.
A non-$\text{PR}$ system will survive changes in the electoral market if there are no new parties or one of the old parties leads in the nonsocialist camp.
The incentives to embrace $\text{PR}$ become irresistible as soon as a new party draws substantial support and the ruling parties are tied in votes.
Two points of departure: the single-member plurality system and the single-member dual-ballot system.
Single-Member Plurality System
The governing party's decisions were shaped by the specified conditions.
If the old parties face a strong new party, two scenarios may develop.
Conservative and Liberal parties coexist
Voters cannot determine around which party they should coordinate to defeat the socialist party.
Panel A represents this situation in Figure 1.
The electorate is uniformly distributed on a single-policy dimension, from Left (0) to Right (1).
Before universal suffrage, only half the population has the right to vote -- voters with preferences from 0.5 to 1.
There are two parties, Liberal L1 and Conservative C1, symmetrically positioned around the old median voter m^o = 0.75 at, say, 0.65 and 0.85, respectively, and thus they win 50% of the vote each (Figure 1A).
Panel B represents this situation in Figure 1.
After universal suffrage is introduced, the median voter is m'' = 0.5.
The Socialist Party S enters the electoral process announcing a position 0.35 + \epsilon (Figure 1B).
Under a single-member plurality system, this is enough to snatch the district from the Conservatives.
Voters cannot coordinate on either L1 or C1, unable to determine which nonsocialist party has a better chance to defeat S.
Anticipating a crushing victory of S, the old ruling parties introduce PR.
One of the old ruling parties has a dominant position in the electoral arena.
Panel C represents this situation in Figure 1.
With C2 still at 0.85, the Liberal Party L2 positions itself differently (0.75, that is, the old median voter, m^o) from before.
As a result, L_2 has a dominant position under the old system, polling about 55% of the votes.
Even if S emerges adopting a position equal to 0.35, the incentives of L_2 to shift to PR remain low.
L2 rationally expects all former Conservative voters to coordinate around L2 to defeat S, since single-member plurality systems encourage strategic behavior.
Once C2 disappears, L2 can move toward the new median voter (m'') to regain control of the parliament.
The Weak Socialist Party
The old ruling parties have no incentives to abandon the plurality system.
In Figure 2A, L1 and C1 have a roughly similar share of votes.
In Figure 2B, C_2 enjoys a dominant position.
Voters will not desert either party to stop S. In fact, it is S that is most damaged by strategic voting.
Single-Member Dual-Ballot System
The incentives to shift to PR are similar to those under the single-member plurality system.
If the new party is weak, the ruling elite will not shift to PR.
The first-round socialist voters will likely vote for their second-ranked candidate in the second round.
There are no incentives to abandon the dual-ballot arrangement if there is one dominant nonsocialist party and the Socialist Party is strong.
The dominant nonsocialist force can become the only nonsocialist party in parliament without having to risk losing an election.
If the Socialist Party is strong and the parties of the Right are similar in strength, they have strong incentives to shift to PR.
In the first round, both L1 and C1 will get 25% of the vote each, and S will get 50%.
The odds that the old governing party will disappear are one-half.
It takes only a very slightly risk-averse party to opt for PR under these conditions.
The decision to shift to PR is even stronger if the nonsocialist camp is more fragmented.
Figure 3 represents a case in which three nonsocialist parties share the old electorate.
The Monarchist Party M is positioned at 0.95. C and L are positioned at 0.8 and 0.6, respectively.
Universal suffrage halves each old party's share of the vote.
Trade and Proportional Representation
Rogowski (1987, 204) concludes that the more an economically advanced state relies on external trade, the more it will be drawn to the use of PR.
PR induces the formation of a strong party system that integrates and restrains particular interests, insulating the state from protectionist interests.
PR induces political and policy stability in interaction with societal forces.
Electoral Rules as Generators of Political Stability
PR constitutes the most adequate system to govern any society with a high degree of political segmentation (Lijphart 1977).
The earliest moves toward proportional representation came in the ethnically most heterogeneous European countries (Rokkan 1970, 157).
Majority elections could threaten the continued existence of the political system in linguistically and religiously divided societies.
In homogeneous polities, plurality rule can remain in place safely.
The Measurement of Electoral Systems
The Dependent Variable: The Effective Electoral Threshold
Explaining what caused many countries to shift away from plurality or majority rule necessitates an appropriate measure of proportional representation.
Effective Electoral Threshold - the proportion of votes that secures parliamentary representation to any party with a probability of at least 50%.
The effective electoral threshold has two advantages:
Provides an empirically tractable variable, abstracting from the complexity of electoral systems.
Good predictor of the degree to which the electoral law distorts the proportional representation of voters' preferences.
The percentage of votes a party needs to gain representation is a range of possibilities, a function of the strength and fragmentation of the remaining parties.
Single-member plurality system:
With four candidates, the threshold of inclusion is 25%.
Threshold of exclusion is 50% minus one vote.
The thresholds:
Threshold of exclusion: T_{excl} = \frac{V}{M+1}
Threshold of inclusion: T_{incl} = \frac{100}{2M}
V is the total percentage of votes.
M is the number of seats in the district.
The literature calculates the inclusion and exclusion thresholds on the basis of the district magnitude and the presence of a legal threshold.
35% is the effective threshold of single-member plurality and dual-ballot districts.
17% is the effective threshold in a system with four-seat districts and no legal threshold
0.75% is the effective threshold in a 100-seats district
The Historical Evolution of Electoral Regimes
Figures 4 show the average effective threshold for each year from 1875 to 1990.
Plurality and dual-ballot systems were the only ones in use in the developed world in the last quarter of the nineteenth century.
By 1919 all the small European states as well as Germany and Italy had embraced PR.
The average effective threshold had fallen to around 18% by 1919, and has moved downward only slightly, to around 14%, since then.
Figures 5 through 7 present three sets of cases, grouped according to regime evolution and overall stability.
Explaining Variation in Effective Electoral Threshold
Sample
Two samples are built to explain variation in the selection of electoral rules.
The first sample includes the average effective threshold for all those countries that experienced a period of democratic government in the interwar years (22 observations)
The second sample adds those countries in which democracy was restored after 1945 (31 observations).
Explanatory Variables
Two sets of theories are explored in this article.
Ruling parties' calculations on electoral structure:
The proportion of socialist votes, or Strength of Socialism
The Effective Number of Old Parties (nonsocialist), or N
Threat is the interactive term of the two previous variables.
Other possible explanations:
Trade Openness
The size of the country
The presence of minorities, measures through Ethnic and Linguistic Fractionalization
Empirical Results
The effect of Threat
Table 2 simulates the level of the effective threshold, once universal suffrage is introduced, for different combinations of socialist strength and conservative fragmentation.
The effective threshold stays close to 30% for having single-member districts when there are two main nonsocialist parties and the socialist alternative is weak.
The threshold falls by 10 points, with a socialist party capturing just 20% of the vote and four nonsocialist parties.
The German case fits quite well the empirical results
Size, Trade, and Internal Fragmentation
The larger the country, the higher is the electoral threshold.
Size may capture the effect of trade openness.
Size may also a proxy for the way in which ethnic and linguistic fragmentation affects each country and the means elites devise to deal with it
Figure 8 displays the relationship between ethnic and religious fragmentation and geographical size
The variable Fragmentation Area Dummy, where the dummy is 0 for countries larger than 450,000 km², 1 otherwise. In this, colunm 3 has results that are statistically significant
Conclusions
Three historical periods can be distinguished in the evolution of electoral regimes in the developed world over the last century:
During the era of limited suffrage in the nineteenth century, plurality/majority rule was consistently used across all nations.
At the turn of the century, the old consensus around the single-member system broke down.
The 1920s ushered in a new era of remarkable stability in the structure of electoral regimes.
The selection of different electoral rules can be traced to strategic decisions made by the current ruling parties.
As long as the electoral arena is stable, the ruling parties have no incentives to modify any electoral norm.
A sudden transformation of the electoral market is likely to trigger a change in the electoral regime