(Page 2 of 21) - Unstable Politics: Fiscal Space and Electoral Volatility in the Indian States authored by Nooruddin, Irfan.

Nooruddin & Chhibber

Introduction

The Indian party system, after decades of relatively stable one-party party system dominance by the
Congress party, has fragmented considerably. Most analysis of Indian electoral politics has focused
on what has transpired at the national level in India where over 30 parties are currently represented
in Parliament and stable governing coalitions (such as the BJP led NDA or the Congress led UPA)
have been composed of at least 10 parties. The national level fragmentation, electoral volatility
and the anti-incumbency sentiments of the electorate have been well documented and commented
upon (Yadav 2000, 2004; Chhibber and Nooruddin 2000; Chhibber and Kollman 2004; Sridharan
2004; Linden 2004).

What about at the state level? Have electoral politics in the Indian states also become more

volatile over time and what explains volatility in the Indian states. What we ﬁnd is that at the
state level electoral volatility has not increased since the 1990s. In fact, electoral volatility in the
states is lesser today than it was in the 1970s though the change is not that substantial. In other
words, while at the national level electoral volatility has indeed gone up since the 1990s that is not
the case at the state level where electoral volatility has been a relatively consistent characteristic
of the states.

How can we explain electoral volatility in the Indian states? Several diﬀerent theoretical frame-

works have been utilized to explain cross-national and inter-temporal variation in electoral volatility.
Scholars studying electoral volatility in Western and Eastern Europe, Latin America, and Africa,
have argued in favor of explanations emphasizing changes in patterns of mobilization (Huntington
1968; Przeworski 1975), variation in electoral laws and party systems (Pedersen 1983; Bartolini
and Mair 1990; Roberts and Wibbels 1999), nature of social cleavages (Ferree 2004; Heath 2005;
Tavits 2005), economic voting (Remmer 1991; Roberts and Wibbels 1999), and the passage of time
(Przeworski 1975; Tavits 2005).

In this paper we develop and test a diﬀerent argument for explaining electoral volatility. We

claim that if governments are to provide services to voters, in exchange for which voters reward
the party in government with their support, this trade, so to speak, of government programs for
citizen’s votes, is only possible if governments possess the revenues to enact policy initiatives and to
ﬁnance public programs. We claim that the availability of these revenues, which we, following Heller
(2005), term ﬁscal space, is crucial to explaining levels of electoral volatility. Where a government’s
budget has ﬁscal space, i.e. “room. . .to provide resources for a desired purpose without jeopardizing
the sustainability of its ﬁnancial position or the stability of the economy” (Heller 2005,32) citizens
reward the incumbent parties with their votes at the next elections. However, when ﬁscal space
is constrained, either for exogenous reasons such as economic crises or endogenous reasons such
as excessive expenditures or low tax revenues, the ability of the incumbent government to provide
such resources is drastically reduced. Citizens therefore have little reason to reward the incumbent
government at the polls, and are ‘available’ to opposition politicians and to alternative appeals.
Vote-switching ensues and the incumbent government is voted out of the oﬃce.

In the ﬁrst part of the paper we describe the variation in electoral volatility across the Indian

states. We then brieﬂy review the principle explanations and ﬁndings from previous studies of
electoral volatility and show why they may not satisfactorily explain the variation in electoral
volatility in the Indian states. We then develop an alternative framework based on concept of ﬁscal

Linden (2004), in a study of incumbency rates for elections to state assemblies ﬁnds that the incumbency

rates have been consistently high and that they have not changed much since the 1970s. This too belies
observations made from national level data that the Indian party system is far more volatile today than it
has been in the past.