How Do Social Media Affect Analyst Stock Recommendations? Evidence from S&P 500 Electric Power Companies' Twitter Accounts
| Date | 01 December 2017 |
| DOI | http://doi.org/10.1002/smj.2678 |
| Author | Eun‐Hee Kim,Yoo Na Youm |
| Published date | 01 December 2017 |
Strategic Management Journal
Strat. Mgmt. J.,38: 2599–2622 (2017)
Published online EarlyView 31 July 2017 in WileyOnline Library (wileyonlinelibrary.com) DOI: 10.1002/smj.2678
Received 8 January 2015;Final revisionreceived 22 January 2017
How Do Social Media Affect Analyst Stock
Recommendations? Evidence from S&P 500 Electric
Power Companies’ Twitter Accounts
Eun-Hee Kim1*and Yoo Na Youm2
1Strategy & Statistics Area, Gabelli School of Business, Fordham University, New
Yor k, N ew Yo rk
2Management Department, Quinlan School of Business, Loyola University Chicago,
Chicago, Illinois
Research summary: The importance of rm-stakeholder relationships is gaining increasing
attention. Although a theory of the drivers and consequences of stakeholder pressure has been
developing, it focuses on pressuresfrom organized stakeholders such as shareholders, NGOs, and
activists, and does not incorporate the emerging possibility that individual voices may matter.
By exploring corporate Twitter, which facilitates movement of individual stakeholders such as
customers to a higher stakeholder class by providing them with a greater sense of power and
urgency,we study the circumstances under which customer voices signicantly affect analyst stock
recommendations. We nd that favorable reactions to rm-initiated messages matter, directly or
indirectly,depending on the messages’ growth implications. Customer-initiated negative messages
have a signicant impact only with high volume and formal institutions that support customer
opinions.
Managerial summary: Social media is increasingly used by rms for disclosing information and
engaging stakeholders. Yet,we know little about whether and how social media usage matters. We
show how corporate Twitter usage may inuence analyst stock recommendations. Our interviews
of securities analysts suggest that social media is not institutionalized yet, but increasingly used
as a source of channel checks, especially for vibes, validations, and so on. Our analyses of
corporate Twitter accounts show that both rm-initiated and customer-initiated tweets can have
signicant impact on analyst recommendations under certain conditions. For rm-initiated tweets,
the extent of retweets is an important factor, along with the content of tweets, in particular,growth
implications. Forcustomer-initiated tweets, negative tweets matter, but only with high volume and
regulatory structure that supports customer protection. Copyright © 2017 John Wiley & Sons,
Ltd.
The importance of rm-stakeholder relation-
ships has gained growing interest as stakeholders
become increasingly relevant to rm operations.
Questions examined in the literature include
which stakeholders are important and whether
Keywords: social media; Twitter; rm-stakeholder rela-
tionship; corporate social responsibility (CSR); analyst
stock recommendations
*Correspondence to: Eun-Hee Kim, Gabelli School of Busi-
ness, Fordham University, New York, NY 10023. E-mail:
ehkim@fordham.edu
Copyright © 2017 John Wiley & Sons, Ltd.
rm-stakeholder relationships matter for rm
performance. We seek to shed new light on this
stream of research by exploring the ramications
of increased salience of individual stakeholders in
the social media space. A growing number of rms
began using social media such as Twitter and Face-
book, which has brought about signicant changes
in how rms communicate with their stakeholders.
Notably, it allows rms not only to engage in
more frequent and informal exchanges with their
stakeholders as compared to more traditional means
such as annual reports or shareholder meetings, but
2600 E.-H. Kim and Y. N. Youm
also to receive feedback on an ongoing basis. In
addition, individual voices have begun to matter
more as a result of social media, a development
quite different from that of commonly studied cases
of more organized stakeholders applying pressure,
such as shareholder resolutions, NGO attacks,
media pressure, and so on.
Together, these idiosyncrasies present an
excellent opportunity to study the link between
rm-stakeholder relationships and rm perfor-
mance from different angles than those used by
most previous studies. Prior work has largely
focused on how rms’ actions to address stake-
holder interests relate to rm performance. For
example, an often-used proxy for rm-stakeholder
relationships is the KLD score. This provides
ratings on attributes of corporate social perfor-
mance such as community relations, environmental
sustainability, and customer and employee pro-
tection based on how rms perform on these
dimensions (see Graves & Waddock [1994] for
details). However, this popular measure does not
capture how rms’ actions to address stakeholder
concerns are, in turn, received by stakeholders
and how this two-sided relationship affects rm
performance. This is a critical issue that has
received little attention. Stakeholder engagement
efforts, if not well received, will not bring about
enhanced support from external stakeholders
(Henisz, Dorobantu, & Nartey, 2014). The inter-
active nature of social media allows us to explore
how stakeholders’ reactions to rms’ actions, that
is, rm-initiated messages, relate to rm perfor-
mance. It also allows us to examine whether and
how stakeholder-initiated postings matter for rm
performance. Thus, we are able to extend the prior
literature in ways that take into account additional
aspects of rm-stakeholder relationships. This is
especially important in rethinking rm-stakeholder
relationships in the digital age.
Among various stakeholders, perhaps individ-
ual stakeholders such as customers stand to gain
the most from rms’ use of social media. Individ-
ual customers are legitimate stakeholders in a rm
(Freeman, 1984), but do not necessarily have power
or urgency (Mitchell, Agle, & Wood,1997). Firms’
use of social media enables individual customers
to increase power, and possibly, urgency, vis-à-vis
the rm for the following reasons. The voice of the
individual customers itself is magnied because of
the public nature of the social media space (Treem
& Leonardi, 2012). Also, individual customers are
able to give ravereviews or exert greater pressure on
rms by amplifying their voices and those of others
by actively promoting and propagating messages
and raising awareness, and by connecting with other
stakeholders and garnering support (Coombs, 1998;
Rowley, 1997). This means that individual cus-
tomers can become denite stakeholders (Mitchell
et al., 1997) in the social media space.
In this article, we suggest specic mechanisms
by which customer feedback on rms’ social media
accounts may signicantly affect rm performance.
A recent study shows that social performance mea-
sures such as rm-stakeholder relationships are
often uncertain and ambiguous to general investors,
and security analysts serve as the informational
pathway connecting corporate social performance
to rm stock returns (Luo et al., 2015). Accord-
ingly, we examine how customer feedback may
affect analyst stock recommendations by looking
into the mechanisms of decision making. In mak-
ing recommendation decisions, security analysts
give signicant consideration to growth potential
(Bradshaw, 2004; Jegadeesh etal., 2004). Good
rm-stakeholder relationships can hint at future
growth prospects in various ways, such as by signal-
ing general support for smooth operations and func-
tioning of rms without interruptions (Henisz et al.,
2014), by lowering transaction costs and easing
capital constraints (Cheng, Ioannou, & Serafeim,
2014; Sharfman & Fernando, 2008), or by facilitat-
ing efcient use of resources to create greater value
for customers (Harrison, Bosse, & Phillips, 2010).
We argue that popularity in the social media space
can point to good rm-stakeholder relationships. In
particular, favorable response to rm-initiated mes-
sages indicates well-receivedness of rms’ efforts
to engage stakeholders and give a good impression.
Positive stakeholder response is especially mean-
ingful because social media offers a low-cost plat-
form for rms to voluntarily disclose information,
where rm-initiated messages are subject to selec-
tive disclosure bias and tend to include favorable
information about the rm (Kim & Lyon, 2015).
Thus, stakeholder response plays a role as a screen-
ing device to differentiate levels of receptivity.
Unless rm-initiated messages provide a direct
gauge of growth, however, the implications of favor-
able stakeholder response for future rm growth
and performance may differ across rms. Even for
rms in the same industry, differences may exist
due to differences in the extent of competition,
market share, economic and demographic factors,
Copyright © 2017 John Wiley & Sons, Ltd. Strat. Mgmt. J.,38: 2599–2622 (2017)
DOI: 10.1002/smj
To continue reading
Request your trial