EXTREMIST PLATFORMS: POLITICAL CONSEQUENCES OF PROFIT‐SEEKING MEDIA

• Author: Jaideep Roy,Kalyan Chatterjee,Siddhartha Bandyopadhyay
• DOI: http://doi.org/10.1111/iere.12453
• Date: 01 August 2020
• Published date: 01 August 2020

INTERNATIONAL ECONOMIC REVIEW

Vol. 61, No. 3, August 2020 DOI: 10.1111/iere.12453

EXTREMIST PLATFORMS: POLITICAL CONSEQUENCES OF PROFIT-SEEKING

MEDIA∗

BYSIDDHARTHA BANDYOPADHYAY,KALYAN CHATTERJEE,AND JAIDEEP ROY1

Department of Economics, University of Birmingham, Birmingham B152TT, U.K.; Department

of Economics, Pennsylvania State University, University Park, PA 16801 U.S.A.; Department of

Economics, University of Bath, Claverton Down, Bath BA27AY,U.K.

We analyze how information about candidate quality affects the choice of electoral platforms made by an

office-motivated political challenger. The incumbent is of known quality and located at the ideal policy of the

voter. The voter cares for both policy and the candidates’ quality and can learn about the challenger’s quality by

buying information. A high-quality challenger then has an incentive to signal her quality by choosing a policy that

induces the voter to buy information. We first study the benchmark case in which the information is supplied

exogenously, and its quality is independent of the challenger’s platform; this yields multiple equilibria and

indeterminacy of equilibrium platforms. By contrast, when the information is supplied by a profit-maximizing

media outlet, its quality depends on the challenger’s platform and we obtain a unique equilibrium platform. In

particular, when the incumbent’s quality is relatively low, the media coverage rises and the challenger’s platform

diverges further from the voter’s ideal policy as the voter’s preference for quality increases.

1. INTRODUCTION

We study how information about the quality of politicians can affect their choice of platforms.

In particular, we analyze whether, and under what conditions, this can cause the rise and electoral

success of politicians who espouse policies that are distinctly positioned to the right or left of the

median voter. While politicians with median platforms are supposed to be more successful by

being closer to a majority of voters, there are examples of politicians (and parties) who have run

for office by taking stands on certain issues that are different from the median platform.2This

has been seen in the success of Narendra Modi in India, of Donald Trump in the United States

and the rise of far right parties across Europe, all occurring at a time of voter dissatisfaction

about politicians in power.3At the same time, there is evidence—for example, from the 2014

∗Manuscript received July 2017; revised February 2019.

1We are grateful to the editors and an anonymous referee for several comments and suggestions that have helped us

improve the presentation of the article. We also thank Sandro Brusco, Matthew Cole, Kaustav Das, Amrita Dhillon,

Jon Eguia, Matthew Gentzkow, Aditya Goenka, Matt Golder, Indridi Indridason, Navin Kartik, Hongyi Li, Simon

Loertscher, Bryan McCannon, S. Nageeb Ali, Colin Rowat, Kunal Sengupta, Francesco Squintani, Samarth Vaidya and

seminar participants at University of Adelaide, University of Bath, King’s College London, University of Birmingham,

UNSW, University of Melbourne, Jadavpur University, Monash University, Calcutta University, the Indian Statistical

Institute (Kolkata), the SAET 2015 conference in Cambridge, the 2016 NICEP Inaugural Conference in Nottingham

and the 2016 META workshop in Cardiff for comments. Chatterjee wishes to thank the Richard B. Fisher Endowment of

the Institute for Advanced Study, Princeton, NJ, for making his membership of the Institute possible for 2014–15 during

which work on this project was undertaken. We thank Sreeparna Saha and Ritvij Roy for helping us with the diagrams.

Please address correspondence to: Jaideep Roy, Department of Economics, University of Bath, Claverton DriveBath

BA2 7AY, U.K.; Phone: +44 (0) 1225 38 5269. E-mail: J.Roy@bath.ac.uk.

2Following Kartik and McAfee (2007) and the subsequent articles in the literature, in the rest of the article, we shall

refer to candidate platforms that do not coincide with the median voter’s as “extremist.” This is to be understood as a

comparative term; the further the platform from the median voter’s, the more extremist it is.

3See Golder (2016) for a review that analyzes the rise of the far right for Europe. This dissatisfaction was

also seen in the United States where only 33% of voters thought the country is in the right direction (see

1173

C

(2020) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

1174 BANDYOPADHYAY,CHATTERJEE,AND ROY

Indian General elections and the 2016 U.S. Presidential elections—of unprecedented media

attention for such candidates.4

There can be many reasons behind media attention for extremist platforms. This article

shows how the presence of a partially revealing signal about candidate quality—for example,

through media coverage— with voter beliefs about quality to determine the politician’s choice

of platform away from the voter’s ideal policy. We study this in a scenario where there is an

incumbent of known quality facing a challenger whose quality is unknown. If the source from

where the information emanates has an incentive to satisfy the voter’s demand for news (e.g., a

profit maximizing media), the challenger will choose a particular policy that moves away from

the voter’s ideal point—thus becoming extremist—as the voter’s preference for a high-quality

challenger increases. To strip out the impact of various competing channels that can also lead to

extremist platforms, we construct a model where neither the media nor the challenger has any

ideological bias and there is only a single voter. Extremist platforms are chosen by candidates

because the demand for information about a relatively unknown challenger increases with both

the distance between the platform from the voter’s ideal policy and with higher quality of

information. Voters face a higher cost of making a wrong choice when electing a candidate

who makes policy commitments that are away from the median. This generates demand for

information that a high-quality challenger exploits, aided by the media outlet responding to its

profit-seeking interests by providing enhanced coverage for such a candidate.

We now explain the mechanism in more detail. The voter in our model cares about two

dimensions, a horizontal dimension which we interpret as policy and a vertical dimension which

we call quality. The voter has a bliss point on the policy dimension while utility is monotonic

on the quality dimension. There is an unknown challenger facing a known incumbent in a

winner take all election. The challenger knows his or her quality while other participants hold

a common prior belief on this quality. The challenger incurs a cost to enter the race against

the incumbent. There is a media outlet that is willing to provide the voter a signal about the

challenger if the voter pays a fee.5The signal reveals the true quality of the challenger with some

probability. This probability is increasing in the level of (costly) investment made by the media

outlet. This induces candidates of high-quality, who want more media coverage, to rationally

take a policy position away from the voter, generating voter demand for information. The voter

of course has a choice of whether or not to buy information from the media outlet. In a situation

where the voter is not particularly happy with the incumbent, he would like to consider voting

for the challenger. However, if the challenger takes a relatively extreme position (compared to

the voter’s bliss point) the cost of choosing the winner incorrectly is greater for the voter. This

increased value for information increases the voter’s demand for news and the media outlet, in

anticipation, invests more in coverage of a more extreme candidate.

The assumption of profit-maximizing behavior of the media outlet is consistent with the

findings by Genztkow and Shapiro (2010) that the slant that a newspaper chooses are on

average close to what it would have chosen if it had “independently maximized its own profits.”

In addition, our observation on media coverage of candidates with platforms away from the

median is consistent with empirical evidence from the United States and some stylized facts

across the world. McCluskey and Kim (2012) examined the coverage of 208 political action

groups in 118 newspapers in the United States. They conclude that “groups that expressed

more polarized opinions on political issues were mentioned in larger newspapers, appeared

http://www.rasmussenreports.com/public-content/politics/mood-of-america/right-direction-or-wrong-track). Even

when they did not win, far left politicians like Bernie Sanders in the United States as well as Jeremy Corbyn in the

United Kingdom have been far more popular electorally than anticipated.

4A recent article by Frank Bruni at the New York Times entitled, “Will the media be Trump’s accomplice again in

2020” highlights this.

5Subscription fees are widespread for better known outlets like the New York Times or The Times. However it is

important to understand that our results remain intact if instead we assume that the voter must incur some cost in

following the media, while media revenues are generated through commercial adverts that respond positively to the

size of viewership.