Introduction
Examining the
determinants, specifically economic determinants, of elections has been an
important research activity in the discipline of public choice. Around election
times, not just economists, but talking heads on TV channels seek to divine the
outcome of elections on the basis of a smorgasbord of economic and other
determinants. Many of them, possibly without knowing it, are being influenced
by Edward Tufte’s oft-quoted statement: “When you think economics, think
elections; when you think elections, think economics” [2].
The literature in
this area distinguishes between Vote Functions and Popularity Functions. In
estimating popularity functions, one tries to estimate the approval rating of
the government or the political party in power while vote functions seek to
explain the voting choice itself [3]. As an
election approaches, but before the voting actually begins, the agenda is to
gauge the popularity of the incumbent political party on the basis of its
performance. Performance is generally evaluated on the basis of the actual
state of the economy as conveyed by data on macroeconomic variables. Vote
functions are estimated after the event to try and understand voting behavior
and to determine which factors might have been important in the voters’ choice.
Even though popularity functions have been estimated for a long time,
especially in the United States, there is no consensus on which macroeconomic
variables have a predictable impact on the popularity of the government [4]. Nate
Silver makes a similar point [5]. Over time, popularity functions have
progressed from using actual data on macroeconomic variables to voters’
perceptions about the state of economy. It is worth pondering whether the
subjective perceptions about the state of economy match the actual state of the
economy as evidenced by hard data.
Estimating
popularity functions for India is not well-established because sufficiently
long time series data are not yet available. However, since 2011, the Reserve
Bank of India has been carrying out Consumer Confidence Surveys (CCS) [6]. The CCS
of September 2011 described the exercise as follows: “It [CCS] gives an assessment of the consumer sentiments of
the respondents based on their perceptions of the general economic conditions
and own financial situation. The assessments are analysed in two parts, viz.,
current situation as compared to a year ago and expectations for a year ahead”
[7]. However, estimating popularity functions is hampered due to the paucity of
data on the popularity of the government or the ruling party. Lokniti has been
reporting such data in its Mood of the
Nation surveys [8]. However, the time series is not long enough to estimate
popularity functions. More than twenty-five years ago, in one of the earliest
studies in the area, I had tried to estimate popularity of the government using
bye-election results [9]. However, bye-elections do not take place at regular
intervals making it difficult to relate popularity of the government using
bye-elections with macroeconomic data – whether actual data or perceptions of
such data – that have been collected at regular intervals of time.
RBI Consumer Confidence Surveys
The earliest
RBI Consumer Confidence Surveys is for September 2011 [10]. This CCS covered
the cities of Bangalore, Chennai, Hyderabad, Kolkata, Mumbai and New Delhi. 900
respondents were selected from each city yielding a sample of 5,400
respondents. Over the years, the coverage of the CCS has increased and the last
one available for December 2018 covered 13 cities – Ahmedabad, Bengaluru,
Bhopal, Chennai, Delhi, Guwahati, Hyderabad, Jaipur, Kolkata, Lucknow, Mumbai,
Patna and Thiruvananthapuram - and obtained 5,347 responses [11]. The questions
asked of respondents covered the following topics:
1. Perceptions and Expectations on General Economic Situation
2. Perceptions and Expectations on Employment
3. Perceptions and Expectations on Price Level
4. Perceptions and Expectations on Rate of Change in Price Level (Inflation)
5. Perceptions and Expectations on Income
6. Perceptions and Expectations on Spending
7. Perceptions and Expectations on Spending- Essential Items
8.
Perceptions and
Expectations on Spending- Non-Essential Items
Perceptions are
about the current situation and respondents are asked if they believe there has
been an improvement, a worsening or no change as compared to a year ago. The percentage of respondents
who state that there is a worsening is subtracted from the percentage of those
who state there is an improvement while ignoring those who feel there has been
no change. This difference is labelled Net Response (NR), a positive value of which
suggests an overall improvement while a negative value suggests a worsening
situation. Expectations are generally in terms of anticipated changes over the
next one year for which as well the NR is computed [12].
Reading the Tea Leaves…
In this
section, I try to make some sense of the results of the RBI’s CCS. To begin
with let us consider the General Economic Situation.
General Economic
Situation
Figure 1 shows
the NR for the current situation and for expectations.
The
dismal NR for the current economic situation in the last few quarters
of the UPA regime before the 2014 Elections was a clear indication of imminent
change. For over a year after the 2014 Elections, there was a significant improvement
in NR till the economy was hit by the shock of Demonetisation. NR plummeted in the post-demonetisation period reaching negative values.
There is much greater optimism regarding the future economic situation with a
large percentage of respondents expecting an improved economic situation one
year down the line.
The
pessimism about the current economic situation is not as bad as it was just
prior to the elections of 2014 but it should be remembered that the UPA was
facing those elections after having ruled the country for ten years. The NDA,
which has had the largest majority in Parliament in 30 years, is facing a
pessimistic electorate at the end of only its first term in office. Table 1
shows the comparison between UPA and NDA regimes.
Table 1: General Economic
Situation – Average Net Response
General Economic Situation (current)
|
General Economic Situation (future)
|
|
UPA: ONE YEAR PRIOR TO 2014 ELECTIONS
|
-25.12
|
2.62
|
NDA: BEFORE DEMONETISATION
|
7.59
|
36.39
|
NDA: AFTER DEMONETISATION
|
-6.17
|
26.69
|
Table
1 shows that the pessimism at the end of UPA’s rule was much deeper than it is
for the NDA and this was true for both the current and the future economic
situation. Nonetheless it is not a comfortable situation for the NDA and this
is seen in the anxiety, especially of the BJP, to steer the conversation away
from the economy towards issues related to national security and nationalism.
But more on that later.
Employment
In
Figure 2, I look at the CCS component on employment.
The
pessimism with respect to current employment is the worst that it has been seen
since the start of CCS. In fact, the NR for current employment turned
negative in the survey of March 2016 but, surprisingly, it was positive at 6.2
in November 2016, the month in which demonetisation was announced. After
demonetisation, the NR has been persistently negative. This lends
substantial credibility to the estimates of the NSSO that unemployment in India
has reached a 45 year high [13]. Expectations of employment have also been
lowered in the aftermath of demonetisation.
Table
2 presents a comparison of the situation with respect to employment during the
UPA and NDA regimes.
Table 2: Employment
Situation – Average Net Response
Employment Situation (current)
|
Employment Situation (future)
|
|
UPA: ONE YEAR PRIOR TO 2014 ELECTIONS
|
1.58
|
25.24
|
NDA: BEFORE DEMONETISATION
|
3.75
|
38.85
|
NDA: AFTER DEMONETISATION
|
-10.25
|
28.35
|
The
average NR with respect to current employment situation during the
post-demonetisation period is much worse than that seen towards the end of the
UPA regime. Even as far as employment prospects are concerned, there is 10
percentage points drop in the score – from 38.85 to 28.35 – between the two
periods of the NDA rule.
Income
Finally,
in Figure 3, Net Responses for Income are presented.
The
fall in perceptions about current income does not sit well with the claims that
India has been growing at a rapid rate during the NDA rule. Perhaps, there is
some truth in Raghuram Rajan’s doubts about whether India is growing at 7% per
annum at all [14]. NR for future incomes, however, show a rise from
December 2017.
Table
3 presents a comparison of the UPA and NDA regimes for the income situation.
Table 3: Income Situation
– Average Net Response
Income (current)
|
Income (future)
|
|
UPA: ONE YEAR PRIOR TO
2014 ELECTIONS
|
22.24
|
40.88
|
NDA: BEFORE
DEMONETISATION
|
16.73
|
46.59
|
NDA: AFTER
DEMONETISATION
|
2.80
|
42.56
|
The
average NR for current income at 2.80 is the lowest in the three time
periods considered in Table 3.
Conclusions
Questions
may be asked whether the respondents are misreading the economy and whether
their pessimistic perceptions are not a sign of inadequate information as well
inadequate information processing. That may well be true and this becomes
evident from the Net Responses for inflation and price level (not reported
here). The NR for both have been negative right from the first
survey till the latest. While negative NR might have been
appropriate during the UPA regime when inflation was running high, these may be
mistaken for the NDA regime when inflation has come down substantially.
Howsoever mistaken the perceptions might be, political parties have to live
with these: if the economy matters in the voters’ decision making, perceptions
will play a more dominant role than “objective” reality or hard data. As “Kahneman
and Tversky showed that…human beings hardly behave as if they were trained or
intuitive statisticians” [15].
It
is now reasonably well-established in the literature that the state of the
economy is always a pertinent issue in any election campaign [16]. In this
context, the responsibility hypothesis
of Lewis-Beck and Paldam tells us that voters hold the government responsible
for economic events and that we can bring into the picture voter perceptions of
the economy [17]. Negative perceptions about the state of the economy spell
trouble for any incumbent as the UPA found out in the previous elections. The
NDA government also faces negative perceptions about the handling of the
economy especially with respect to employment. The question remains whether
voters are sufficiently disenchanted with the present government to vote it
out. That is a difficulty question to answer for a variety of reasons.
1. We do not know how exactly perceptions
about the economy are transformed into voting
decisions: the paucity of data do not allow us to make any inference.
decisions: the paucity of data do not allow us to make any inference.
2 Voting out the government requires an
alternative who can capture the negative votes
against the government
against the government
3. Non-economic issues gain prominence over
economic issues deflecting attention away
from the negative perceptions
from the negative perceptions
For
the NDA government, issues surrounding voter perceptions are not academic but rather existential
issues. If there is the slightest risk that negate perceptions might harm
its re-elections prospects, it must (and it has been doing so!) bring to prominence non-economic issues
that steer the discussion away from economic issues. Of course, to be on the
safe side, the government has suppressed unpalatable data (e.g. NSSO report on
unemployment) and it has also got the Niti Ayog to massage the GDP data in such
a way that the NDA performance seems better than that of the UPA.
Sometimes
fate intervenes to give the incumbent an advantage: the Pulwama terror attack,
however tragic and ghastly it undoubtedly was, did give the BJP the opportunity
to display its muscular nationalism. For the past several weeks, the campaign
battle has seen the opposition parties seeking to focus attention towards the
economy while the BJP has been dragging the conversation towards nationalism
and security. In Narendra Modi the opposition is facing a master campaigner who
is not above twisting facts so as to
put the opposition on the defensive. Likewise, genuine questions to the government have been mauled out of context to extract political mileage. For example, questions about BJP’s claims
about casualties in the air strike in Balakot [18] were falsely portrayed as
the opposition questioning the Indian armed forces [19]. This is such a far cry
from the strategy followed by Mr. Narendra Modi in 2014 when catch phrases such
as ‘Achhe Din’ and ‘Sab ka saath, sab ka vikas’ rent the
air. As Shekhar Gupta correctly puts it, for the 2019 elections, Mr. Modi is
not seeking votes on the basis of his government’s economic performance but on
the basis of national security [20].
Will
the BJP’s messaging and rhetoric on nationalism be sufficient to divert
attention away from its obvious failings on the economic front? Will the
opposition be able to drag back the discourse towards economic issues? And,
most importantly, how will the mix of economic and non-economic issues play out
in the minds of voters as their hands hover over Electronic Voting Machines in
polling booths across India? We will get
our answer by the middle of May 2019.
ENDNOTES:
1. I
would like to thank Manali Phatak for collecting and organizing the data on
RBI’s Consumer Confidence Surveys.
2. Tufte, E.R.
(1978) Political Control of the Economy,
Princeton University Press, Princeton, NJ, USA.
3. Lewis-Beck, M.S. and Stegmaier, M. (2013) The VP-function revisited: a
survey of the literature on vote and popularity functions after over 40 years, Public Choice, Vol. 157, No. 3/4, pp.
367-385 (https://www.jstor.org/stable/24507619?seq=1#page_scan_tab_contents)
4. Lewis-Beck
and Stegmaier (2013) ibid.
5. Nate
Silver (2012)Measuring the Effect of the Economy on Elections https://fivethirtyeight.blogs.nytimes.com/2012/07/05/measuring-the-effect-of-the-economy-on-elections/
6. Reserve Bank
of India (various years) Consumer Confidence Survey, https://rbi.org.in/Scripts/QuarterlyPublications.aspx?head=Consumer%20Confidence%20Survey
7.
Reserve Bank
of India (2011) Consumer Confidence
Survey : September 2011, https://rbi.org.in/Scripts/PublicationsView.aspx?id=14028
9. Karnik,
A. (1990) “Elections and Government Expenditures: The Indian Evidence”, Journal of Quantitative Economics, Vol.
6, pp. 203-212
10. RBI
(2011) Consumer Confidence Survey : September 2011, https://rbi.org.in/Scripts/PublicationsView.aspx?id=14028
11. RBI
(2019) Consumer Confidence Survey: December 2018, https://rbi.org.in/Scripts/PublicationsView.aspx?id=18776
12. The
methodology of analyzing the CCS is given here: https://rbi.org.in/Scripts/PublicationsView.aspx?id=14028#A2
in Annex II.
13. Sharad
Raghavan, T.C.A. and Jebaraj, P (2019) National Sample Survey Office’s
unemployment data confirms crisis on the ground: experts, https://www.thehindu.com/news/national/national-sample-survey-offices-unemployment-data-confirm-crisis-on-the-ground-experts/article26141744.ece
14. Economic
Times (2019) Raghuram Rajan raises doubts about India growing at 7%, says cloud
over GDP data needs to be cleared, https://economictimes.indiatimes.com/news/economy/indicators/raghuram-rajan-raises-doubts-about-india-growing-at-7-says-cloud-over-gdp-data-needs-to-be-cleared/articleshow/68581012.cms
15. Sunstein,
C.R. and Thaler, I (2016) The Two Friends Who Changed How We Think About How We
Think, The New Yorker, December 7, https://www.newyorker.com/books/page-turner/the-two-friends-who-changed-how-we-think-about-how-we-think
16. Rogers,
J.and Tyszler, M. (2016) Information and Economic Voting. Political Science Research and Methods, Vol. 6 (4), pp.753-770.
Available at: https://s3.amazonaws.com/academia.edu.documents/46751690/InformationAndEconomicVoting.pdf?AWSAccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires=1554110895&Signature=%2FP%2B9Cy2nDIhyzmd36oFPoWC8Uv0%3D&response-content-disposition=inline%3B%20filename%3DInformation_and_Economic_Voting.pdf
17. Lewis-Beck,
M. S. and Paldam, M. (2000) Economic voting: an introduction, Electoral Studies, Vol. 19, pp. 113-121.
Available at: https://s3.amazonaws.com/academia.edu.documents/6179351/s4._economic_voting_an_introduction__lewis-beck_paldam_2000_.pdf?AWSAccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires=1554111666&Signature=bKTa%2FrD6ei8b2PRsFFzcN8WYwqE%3D&response-content-disposition=inline%3B%20filename%3DEconomic_Voting_An_Introduction.pdf
18. Langa,
M. (2019) Air strike killed more than 250 terrorists, says Amit Shah, The Hindu, March 4 2019, https://www.thehindu.com/news/national/other-states/air-strike-killed-more-than-250-terrorists-says-amit-shah/article26430065.ece
19. DD
News (2019) FM Arun Jaitley slams those who question armed forces, http://ddnews.gov.in/national/balakot-airstrike-was-not-revenge-action-fm-arun-jaitley
20. Gupta,
S. (2019) Modi’s 2019 mantra: Forget achhe din, fear terror, Pakistan, Muslim, The Print, https://theprint.in/national-interest/modis-2019-mantra-forget-achhe-din-fear-terror-pakistan-muslim/214495/
