Gross Violations of Duverger’s Law in India

DOI10.1177/2321023013509150
Published date01 December 2013
Date01 December 2013
Subject MatterArticles
Military-Madrasa-Mullah Complex 179
India Quarterly, 66, 2 (2010): 133–149
A Global Threat 179
Article
Gross Violations of Duverger’s
Law in India1
Peter Mayer
Abstract
In his famous study of the impact of electoral regimes on party systems, Duverger suggested that two-
party systems are always found to be associated with single-member constituencies where candidates
are elected by a plurality of votes (Duverger, 1964).
Many studies of what Rae called the political consequences of electoral laws have tended to confirm
this well-known electoral ‘law’. A few studies (Betz, 2006; Chhibber and Murali, 2006; Diwakar, 2007,
2010), however, have drawn attention to India’s exceptionalism in terms of party aggregation.
This article explores a relatively minor—and thus less-examined but nonetheless puzzling—aspect of
Indian exceptionalism: the spectacular failure overall of candidates (and of minor parties) to conform to
Duverger’s law. Since 1957, the first national election after States Reorganisation, the average number of
candidates standing in Lok Sabha constituencies in all states has tended to rise. In 2004, for example, in all
but four states there were more than 10 candidates per constituency in over 90 per cent of constituencies.
In this article I explore why neither the mechanical nor the psychological processes commonly
understood to be so compelling elsewhere exert significant influence in India. This suggests that there
are hitherto unrecognized presuppositions to Duverger’s law.
Keywords
India, elections, Duverger’s law, candidates
Introduction
In his famous study of the impact of electoral regimes on party systems, Duverger suggested that
two-party systems are always found to be associated with single-member constituencies where candi-
dates are elected by a plurality of votes (Duverger, 1964). He suggested proportional electoral systems,
by contrast, tend to produce multi-party systems.
Duverger identified two processes which seem to be responsible for this association between electoral
regimes and party structures. One process, which he termed the ‘mechanical factor’, involves the work-
ing of the electoral laws and rules which convert votes into seats. In single-member electorates (districts
characterized in later literature as having a ‘magnitude of one’) votes for third and minor parties are, in
effect, wasted. Parties which have no realistic chance of winning in a seat will not enter a candidate there
(Duverger, 1964, pp. 224, 226). The second, more reflexive process (the ‘psychological factor’) describes
Peter Mayer, Politics Department, University of Adelaide, Adelaide, S.A. 5005, Australia.
E-mail: peter.mayer@adelaide.edu.au
Studies in Indian Politics
1(2) 179–201
© 2013 Lokniti, Centre for the
Study of Developing Societies
SAGE Publications
Los Angeles, London,
New Delhi, Singapore,
Washington DC
DOI: 10.1177/2321023013509150
http://inp.sagepub.com
180 Peter Mayer
Studies in Indian Politics, 1, 2 (2013): 179–201
the adaptation of voters to that fact. Voters, whose favoured party has little chance of winning, will elect
to vote ‘strategically’ for the least disliked of the two major parties contesting the seat (Duverger, 1964,
p. 226). ‘The simple-majority single-ballot system appears then to be capable of maintaining an estab-
lished dualism [two-party system] in spite of schisms in old parties and the birth of new parties’ (Duverger,
1964, p. 226). Many studies of what Rae called the political consequences of electoral laws have tended
to confirm this well-known electoral ‘law’.
India has a single-member district, plurality electoral system for elections to the lower houses of
national and state legislatures. This article explores a relatively minor—and thus less-examined but
nonetheless puzzling—aspect of Indian exceptionalism: the spectacular failure overall of candidates
(and of minor parties) in elections to the Lok Sabha (the lower house of India’s parliament) to conform
to Duverger’s law.
Since all candidates in India are elected in the same system of single-member constituencies and by a
plurality of ballots, differences in the ‘mechanical effects’ of party systems which are so important in
the general comparative political science literature on the effective number of parties are not relevant
here (Benoit, 2001, 2007; Lijphart, 1990; Rae, 1967). Nevertheless, though, since contests should be
relatively predictable, over time, we would ordinarily predict that mechanical effects would force fewer
and fewer parties to stand in electorates where they cannot succeed. Equally we would expect that ‘psy-
chological’ pressures on voters not to waste their votes on candidates who cannot win should tend to
produce a convergence to a two-party system with fewer and few multi-party contests in constituencies.
Equally, as a corollary, we would predict that, over time, the effects—whether mechanical or psychological—
of the single-member, plurality vote system in India would also result in diminishing percentages of
candidates who are unable to judge their chances of success and who would thus forfeit their deposits.
But, in fact, the opposite is the historical experience in India. Since 1957, the first national election
after States Reorganisation, the average number of candidates standing in Lok Sabha constituencies in
all states has tended to rise. In 2004, for example, in all but four states there were more than 10 candi-
dates per constituency in over 90 per cent of constituencies. The election in Chennai South in 2009, to
cite an extreme recent example, was contested by 43 candidates. And the percentage of candidates losing
their deposits has consequently also risen to high levels. In 2004, for example, over 50 per cent of all
candidates in all states lost their deposits. In Chennai South in 2009, 41 of 43 candidates (95 per cent)
forfeited their deposits.
Only a few studies, however, have drawn attention to India’s exceptionalism in terms of party aggre-
gation. Betz (2006) has examined the institutionalization of parties. Chhibber and Murali (2006) have
explored interesting exceptions to Duverger’s law at the state level. Diwakar (2007, 2010) has extended
the analysis to the electoral district level, finding that there is multi-party competition in many seats.
In this article I explore why it is that neither the mechanical nor the psychological processes com-
monly understood to be so compelling elsewhere exert significant influence in India. This suggests that
there are hitherto unrecognized presuppositions to the working of Duverger’s law.
Candidates
The Number of Candidates per Constituency
In Table 1 we can see just how impotent mechanical and psychological pressures have been in India.
In 1957 the average number of candidates in constituencies was five or fewer. This remained largely the

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT