Indian Politics & Policy Volume 3, Number 1, Spring 2020 | Page 16

Indian Politics & Policy
mance. More importantly, these articles
claimed that public opinion (for
instance, the Mood of the Nation study
in 2018) had turned against the party
that had come to power on the promise
of sabka vikas (development for all)
five years earlier. The predicted number
of seats for the BJP ranged between 170
and 270. It appeared that voters would
most likely give the BJP another chance
to govern, albeit with a drastically reduced
majority in 2019.
The BJP’s landslide return to national
power, winning 37 percent of the
vote and 303 seats while confronting
a slowing economy and rising unemployment
following twin policy disasters—the
demonetization of currency
in 2016, and the adoption of Goods and
Services Tax in 2017—raises a broader
question about voting behavior in India.
Does the economy matter to the Indian
voter? Is there a relationship between a
voter’s perception of government economic
performance and vote choice?
In this article, I empirically evaluate a
hypothesis broadly referred to as the
economic vote (or economic voting) in
political science scholarship in the context
of the 2014 and 2019 elections to
the Lok Sabha. The economic vote hypothesis
(more specifically, egotropic
voting) considers the relationship between
a voter’s retrospective evaluation
of their personal or household economic
condition and their vote choice for an
incumbent. Does economic voting find
empirical purchase in India’s parliamentary
elections? I apply a logistic regression
model to individual level data
from the National Election Study (NES)
post-poll surveys (2014 and 2019) in
order to examine this relationship while
controlling for the influence of relevant
factors such as incumbent party attachment
(or partisanship) and the effects
of time, among others.
I find that a voter’s pocketbook
matters to the incumbent vote in an
election. A voter’s positive evaluation of
household economic conditions is positively
correlated with, i.e., increases the
likelihood of, a vote for the incumbent.
Conversely, a voter’s negative evaluation
of household economic conditions
is negatively correlated with, i.e., lowers
the likelihood of, a vote for the incumbent.
Voters react more to a perception
of a negative evaluation of economic
conditions than positive ones. I also
find that the effect of a positive evaluation
was stable over the past two elections
and that of a negative evaluation
increased in magnitude in 2019 relative
to 2014. Finally, I find that attachment
to an incumbent party or coalition has
a strong positive effect on the incumbent
vote, and that the likelihood a voter
chooses an incumbent in 2019 is also
higher. An analysis of the BJP vote in
2019 reveals that socio-demographic
factors such as caste-community identity
also play an important role. All Hindu
caste groups tend to vote for the BJP
relative to other religious communities
and castes. The effects of other economic
and demographic characteristics,
such as class, rurality, and gender, are
smaller or nonexistent.
The structure of this article is as
follows. I first survey the literature on
comparative voting behavior and outline
the theoretical motivation behind
the economic vote and the hypothesis I
12