Endorse or Not to Endorse: Understanding the Determinants of Newspapers’ Likelihood of Making Political Recommendations

Published date01 September 2016
Date01 September 2016
DOIhttp://doi.org/10.1111/sjpe.12098
ENDORSE OR NOT TO ENDORSE:
UNDERSTANDING THE
DETERMINANTS OF NEWSPAPERS’
LIKELIHOOD OF MAKING POLITICAL
RECOMMENDATIONS
Fernanda Leite Lopez de Leon*
ABSTRACT
This article investigates the determinants of newspapers’ provision for political
opinion. I empirically examine the role of newspapers’ political preferences and
market competition on newspapers’ decision to make endorsements. Regression
results suggest that market competition turns newspapers more likely to make
endorsements. Results from a simple model show that newspapers’ ideology
determine their endorsements, making partisan papers more likely to make polit-
ical recommendations and endorse challengers than non-partisan newspapers.
II
NTRODUCTION
Political endorsements have an important role in American elections. They
affect voters’ perceptions about candidates (Ladd and Lenz, 2009; Knight and
Chiang, 2011) and candidates’ vote share (Leon, 2013). This article investi-
gates the determinants of newspapers’ provision of political endorsements by
focusing on two main issues: market structure and newspapers’ political pref-
erences. Understanding these relationships is important for policy issues. His-
torically, public efforts, such as the Newspaper Preservation Act, have been
undertaken in response to the decline of newspapers’ circulation to ensure
diversity (Busterna and Picard, 1993). The Newspaper Preservation Act autho-
rized the formation of joint operating agreements (JOAs) among competing
newspapers operating within the same market area to help newspapers reduce
their operating costs. While JOAs allowed newspapers to combine business
operations (e.g. advertising and circulation), they were required to maintain
separate news operations and editorial sections to decide, for example, politi-
cal endorsements. Despite such efforts, it is likely that newspapers’ level of
political activism is a response to market incentives, as described below by
The Times Magazine (2008):
*University of Kent
Scottish Journal of Political Economy, DOI: 10.1111/sjpe.12098, Vol. 63, No. 4, September 2016
©2015 Scottish Economic Society.
357
At the time when newspapers are trying to ensure their survival
by attracting young readers, the idea of endorsements is both
counterproductive and an anachronism... in doing so newspa-
pers are undermining the very basis of their business, which is
impartiality. The Times Magazine (2008)
More broadly, this article contributes to a body of work that investigates
newspapers’ political preferences and how these reflect onto media reports
(Ansolabehere et al., 2006; Gentzkow and Shapiro, 2010; Di Tella and Fran-
ceschelli, 2011; Knight and Chiang, 2011; Larcinese et al., 2011; Puglisi and
Snyder, 2011 2015), and how market structure shapes the media (Chiang,
2007; Kim, 2008; Gentzkow and Shapiro, 2010; Cage, 2013).
As discussed above, newspapers’ political endorsements are relevant in elec-
tions and interesting by themselves. More generally, they are a good case
study to overcome measurement challenges and to understand the relationship
between newspapers’ provision of political opinion and market competition.
Firstly, endorsements represent a clear stand in favour of a candidate. Sec-
ondly, during elections, newspapers face an identical opportunity to take a
stand. Therefore, their choice set is observable, as opposed to news that is
determined by a random occurrence of events and is thus unobservable to
readers until reported. Measuring the correlation between competition and a
newspaper’s reporting practice is also challenging. Newspapers face a different
set of competitors in different geographical areas, while news itself is supposed
to reach all newspaper readers. Political endorsements are tailored messages
for a subset of readers namely, those who live in a particular political juris-
diction. This feature allows one to test whether and how the level of competi-
tion a newspaper faces in an electoral jurisdiction correlates with its
behaviour. As I will further develop in Section III, I expect a relationship
between newspapers’ political endorsements and market competition. Market
structure correlates with newspapers’ characteristics, such as ideological posi-
tioning (Chiang, 2007; Gentzkow and Shapiro, 2010) and politicians’ beha-
viour (Snyder and Stromberg, 2010). As a result this will determine
newspapers’ evaluation of candidates and, hence, their endorsements.
To identify the association between market competition and newspapers’
likelihood of making endorsements, I collected information on demographics
and the newspaper industry’s structure at the county level, and transform
those to the level of newspaper-jurisdiction. I then constructed a variable that
is the fraction of the political jurisdiction in which a newspaper operates as a
monopolist, and I test how this variable correlates with the likelihood of an
endorsement. Variations in the data enabled me to compare endorsement
behaviour across areas where newspapers face a different degree of competi-
tion, holding constant politicians’ behaviour, readership, and newspapers’
intrinsic characteristics. Hence, I was able to circumvent and separate con-
founding effects regarding measurement of the impact of market structures on
newspaper behaviour, such as the selection of politicians that run for election
in different jurisdictions, or characteristics of newspapers that are established
358 FERNANDA LEITE LOPEZ DE LEON
Scottish Journal of Political Economy
©2015 Scottish Economic Society

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT