Home / Opinion / Do swing voters ensure higher growth?

Given the ongoing assembly elections, the issue of effectiveness of choice made by voters has once again come to the forefront. In a recent working paper (Democracy and Economic Growth: Do Swing Voters Make a Difference? by Prasanna Tantri, Abhishek Bharadwaj and Nagaraju Thota), we examine the effectiveness of electoral choice exercised by swing voters. We find that the representatives supported by swing voters outperform significantly with respect to both constituency-level outcomes as well as measures based on individual effort.

We start by defining and identifying swing voters. The extant literature defines a core voter as the one who is predisposed in favour of a party on partisan or programmatic grounds. Those who are not so predisposed can be termed as swing voters. We identify swing voting based on actual voting records of elections where a citizen simultaneously chooses representatives for state and central legislatures. For example, along with the general election held in 2014, assembly elections were held for states such as Andhra Pradesh and Odisha. We first aggregate the assembly votes received by all candidates belonging to a party or a pre-poll alliance in different assembly constituencies belonging to the same Lok Sabha constituency. We then compare the same to the total votes received by the candidate belonging to the same party or pre-poll alliance in the Lok Sabha election in the same constituency. The difference between the two gives our measure of swing.

We have data pertaining to constituency-level economic outcomes for two parliamentary terms: 2004-09 and 2009-14. In the states of Andhra Pradesh, Odisha and Sikkim, national and state elections were held simultaneously both in 2004 and 2009. Thus our sample is restricted to 128 members of Parliament (MPs) representing the above three states during the two terms specified above. We examine the association between swing and performance using the above specified sample.

We start by comparing the performance of people’s representatives who enjoy very high levels of support among swing voters and those who do not enjoy such support. Please note that a Lok Sabha MP represents a constituency, which is roughly equal to an administrative district. We use growth in gross domestic product (GDP), per capita GDP and per worker GDP at the district level as performance measures. We calculate the growth rate over the entire term of a representative. In other words, the reported growth numbers are not annualized. We find that a 10% higher support among swing voters is associated with 4.88% increase in district GDP growth, 4.76% increase in per capita GDP growth and 4.72% increase in per worker GDP growth over the entire term of five years. These numbers translate into nearly 1% annualized growth in various measures of GDP. The numbers are both economically as well as statistically significant.

Recognizing that the GDP measurement at local levels, especially in emerging economies, is prone to measurement errors, we use growth in bank credit at the district level as a robustness measure for GDP. These numbers are audited by the Reserve Bank of India. The literature on the relationship between finance and growth has shown that the two variables are positively associated. We find that a 10% increase in swing is associated with 15.34% increase in credit growth at the district level during a five-year term.

In order to test the ability of an MP to bargain with the central government and get funds allocated to his/her constituency, we examine the association between swing votes and Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) allocation. We find that a 10% increase in swing voters is associated with a 131.23% higher growth in MGNREGA spending at the constituency level.

In order to address the concern that the performance measures described so far are mostly based on aggregate outcomes and hence may not necessarily fully reflect individual efforts and drive of the incumbent, we use attendance in Parliament as a performance measure. We compare the attendance levels of MPs who have maximum support among swing voters and those that do not enjoy such support. Unlike general outcomes such as growth and deficits, attendance is largely decided by the discretion of the individual MP. We find that a 10% increase in swing votes is associated with a 6.5% increase in attendance.

In sum, we find that the swing voters contribute towards strengthening of the economy as well as democratic institutions and practices. We show that in the presence of swing voters, democracy is likely to lead to higher economic growth as swing voters elect those candidates that promote growth. Swing voting seems to be the result of deliberate and careful choice made by alert citizens.

Our findings provide a key message to both the ruling and opposition parties at the centre. Given the large presence of swing voters, it is incorrect on the part of the ruling party to set targets in assembly elections that are based on the 2014 Lok Sabha election. The growing tribe of swing voters evaluates assembly elections mostly based on state issues. For the opposition parties, it is not totally realistic to fully extrapolate the assembly poll victories in states such as Delhi and Bihar to the 2019 Lok Sabha election. Our paper argues that the swing voters in 2019 are likely to vote based on their assessment of who is likely to administer the country better for the next five years.

Prasanna Tantri is senior associate director at Center for Analytical Finance, Indian School of Business.

Comments are welcome at theirview@livemint.com

Catch all the Business News, Market News, Breaking News Events and Latest News Updates on Live Mint. Download The Mint News App to get Daily Market Updates.
More Less

Recommended For You

Trending Stocks

Get alerts on WhatsApp
Set Preferences My ReadsWatchlistFeedbackRedeem a Gift CardLogout