|
A Brief Introduction to
Electoral Systems* |
||
|
|
||
|
Dr. Cheng-Feng Shih（施正鋒） Professor, Department of Public Administration
and Institute of Public Policy Tamkang University, Tamsui, Taiwan Electoral
systems are not equivalent to electoral laws; rather, they are mechanisms
that translate votes into seats in legislative as well as executive
elections. That electoral systems matter is testifies by the fact that
parties with the same votes may receive disparate percentages of seats under
different electoral systems. In the Third-Wave of democratization,
electoral engineering has been considered crucial for successful political
engineering, if not manipulation. Electoral systems have both micro and
macro political consequences. On the one hand, electoral systems, as
political institutions, while directly acting upon voters’ behavior, would
ultimately provide incentives for and barriers to political parties and
candidates in the process of nomination and of campaign (Figure 1). On
the other hand, electoral systems would produce three systemic effects:
proportionality of the seat in relation to the vote parties receive, the
number of parties, and minority representation. While the former are
termed psychological effects, the later mechanical ones. When electoral
systems and their impacts on the party system are waged against the
constitutional system already adopted, whether they are synchronized will
decide how the political system may operate smoothly (Figure 2).
Political scientists would generally agree that proportional electoral
systems are conducive to multi-party systems, plurality ones to two-party systems,
and majority ones to two-camp systems. As it has been suggested by some
that multi-party systems are incompatible with presidential systems, the
logic goes, either the constitutional system has to shift gear to a
parliamentary one or a non-proportional electoral system has to be devised.
The
structure of electoral systems has three major components: ballot structure,
constituency structure, and translating formula (Figure 3). First of
all, the balloting structure is composed of two dimensions: to whom the vote
goes (candidate or party), and the process of voting, which is in turn
decomposed into the number of votes (equal or smaller than the number of
seats) and the type of vote (categorical or ordinal). Secondly, the
constituency structure has to decide its nature (either at large or
districting), and size (single-member or multi-member district). Last
but not least, we have to choose from among the following translating
formulae: plurality, majority, semi-proportionality, and proportionality. If
we focus on the translating formula, we can come up with two ideal types of
electoral system, plurality/majority and proportional representation, along
with semi-proportionality and some so-called mixed systems (Figure 4).
Within the plurality camp, first-past-the-post is most popular in countries
under Anglo-Saxon influence while various form of block votes are also
adopted. Under the majority branch, two-run systems are simple but
cumbersome if no candidate receives majority of the votes; alternative votes
appear fairer for voters as they may express their preference for the
candidates while casting their votes. For most of Western and Northern
European countries, they seem satisfied with proportional representation in the
form of party-list even though Ireland enjoys the single-transferable vote
that allows voters to cast ordinal votes. The most amazing proportional
representation system is the mixed-member proportionality system (also termed
additional member system), invented by Germany and imitated by New Zealand,
one that guarantees both representatives’ close contact with their
constituencies and proportionality.
Of
course, we should not leave behind the semi-proportional representation
system that has just been abolished in Taiwan earlier this year but is still
employed by some countries in the Middle East, the single-non-transferable
vote. In the 1990s, we witnessed burgeons of some mixed systems, also
know generically as parallel systems. As the name suggests, they are
designed after assorted combinations of proportional representation and
first-past-the-post constituencies. Compared to the mixed-member
proportionality system mentioned earlier, these mixed systems are only
partially proportional depending on the weight of the proportional
representation. It is noted that the electoral system for legislative
elections since 1992 have be syntheses of single-nontransferable-vote and
proportional representation. Starting from the legislative election in
2007, it will be adjusted to a mixture of first-past-the-post and
proportional representation. If the political science wisdom is
correct, it is expected that the fragmentation and thus the centrifugal
tendency of the party system may be alleviated. As a result, a
popularly elected president can be more accountable to the voters even under
divided government, that is, when the opposition controls the legislative
branch. Finally,
if a country is endowed with ethnic diversity, there are some ingenious
efforts at ensuring effective minority representation, ranging from reserved
seats for the indigenous peoples in Taiwan and New Zealand; guaranteed seats
for national minorities in Croatia, Slovenia, and Romania; affirmative
gerrymandering for the Black and the Latinos in the U.S.; lower thresholds
for ethnic parties in Germany and Poland; nominating minority candidates in
safe constituencies, or ranking minorities higher on the party list; and
utilizing alternative votes to incorporate minorities’ perspectives (Figure
5).
* Presented at the
Conference on Electoral Reforms sponsored by the Democratic Pacific Union,
Foreign Service Institute, Taipei, 2005/12/4. |
