Conflict inside and outside: Social comparisons and attention shifts in multidivisional firms

• DOI: http://doi.org/10.1002/smj.2556
• Published date: 01 July 2017
• Author: Richard A. Bettis,Songcui Hu,Zi‐Lin He,Daniela P. Blettner
• Date: 01 July 2017

Strategic Management Journal

Strat. Mgmt. J.,38: 1435–1454 (2017)

Published online EarlyView 1 November2016 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/smj.2556

Received 2 October 2014;Final revision received11 April 2016

CONFLICT INSIDE AND OUTSIDE: SOCIAL

COMPARISONS AND ATTENTION SHIFTS IN

MULTIDIVISIONAL FIRMS

SONGCUI HU,1*ZI-LIN HE,2DANIELA P. BLETTNER,3and RICHARD A.

BETTIS4

1Eller College of Management, Department of Management and Organizations,

University of Arizona, Tucson, Arizona, U.S.A.

2School of Economics and Management, Department of Management, Tilburg

University, Tilburg, Netherlands

3Beedie School of Business, Strategy / Technology and Operations Management

Department, Simon Fraser University, Burnaby, British Columbia, Canada

4Kenan-Flagler Business School, Strategy and Entrepreneurship Department,

University of North Carolina-Chapel Hill, Chapel Hill, North Carolina, U.S.A.

Research summary: Behavioral Theory highlights the crucial role of social comparisons in

attention allocation in adaptive aspirations. Yet, both the specication of social reference points

and the dynamics of attention allocation have received little scholarly examination. We address

performance feedback from two social reference points relative to divisions in multidivisional

rms: economic reference point and political reference point. Comparing divisional performance

with the two reference points can give consistent or inconsistent feedback, which has important

consequences for the dynamics of attention allocation in adaptive aspirations. We nd consistent

feedback leads to more attention to own experience, while inconsistent feedback results in more

attention to the social referencepoint the focal division underperforms. Results reveal that political

reference point plays an important rolein determining managerial attention allocation.

Managerial summary: This article is based on how goal-based performance of divisions relative

to both their relevant external marketrivals and sister divisions in multidivisional rms inuences

corporate resource allocation. As a result, various combinations of performance against the two

groups of peers drive the reallocation of divisional management attention. We show that specic

attention shifts occur on average as a function of the focal division’s performance relative to the

marketplace performance and that of sister divisions. Copyright © 2016 John Wiley & Sons, Ltd.

INTRODUCTION

A central tenet of the Behavioral Theory of

the Firm (BTOF) is that performance feedback

Keywords: adaptive aspirations; social comparison; polit-

ical coalition; attention allocation; multidivisional rms

*Correspondence to: Songcui Hu, Eller College of Manage-

ment, University of Arizona, 1130 E. Helen St., McClel-

land Hall RM 405MM Tucson, Arizona 85721. E-mail:

songcuih@email.arizona.edu

Copyright © 2016 John Wiley & Sons, Ltd.

directs how organizations learn and adapt (Cyert

and March, 1963). The generalized model of

performance feedback depends on a satisfactory

performance level— referred to as the aspiration

level— against which an organization compares its

actual performance. According to BTOF (Cyert

and March, 1963), organizations adapt their aspi-

rations based on historical comparisons with their

own experience (i.e., prior aspiration and prior

performance) and social comparisons with others’

experience (i.e., others’ performance). That is,

1436 S. Hu et al.

organizations allocate attention to these different

reference points when adapting their aspirations.

While own experience is generally well dened,

there is considerable ambiguity in the specication

of social reference groups, a crucial but largely

neglected source of rm heterogeneity (e.g.,

Massini, Lewin, and Greve, 2005; Washburn and

Bromiley, 2012).

We focus on social comparisons and the related

feedback from social reference points in adaptive

aspirations. Notwithstanding the well recognized

importance of social comparisons in adaptive aspi-

rations, the selection of social reference groups

remains underexplored and lacks theoretical devel-

opment. Extant empirical studies (e.g., Audia and

Greve, 2006; Baum and Dahlin, 2007; Greve, 1998;

Miller and Chen, 2004) have largely assumed an

entire population of external marketplace competi-

tors as the reference group, which we call the eco-

nomic reference group, and the mean or median

performance of this reference group as the social

reference point, which we call the economic refer-

ence point. This simplied specication of a sin-

gle reference group has provided important insights.

However, recent studies (e.g., Massini etal., 2005;

Moliterno, Beck, and Beckman, 2014; Washburn

and Bromiley,2012) have questioned this and called

for a more careful specication of social refer-

ence groups to increase the precision and realism

of performance feedback models. Furthermore, the

rm-level aspiration has been the focus of these

studies whereas rms can have different aspirations

across different levels of an organizational hierar-

chy (March and Simon, 1958), with varying impli-

cations for adaptation of business units or divisions

within the rm (Gaba and Joseph, 2013; Vissa,

Greve, and Chen, 2010).

We extend the study of attention allocation in

adaptive aspirations to divisions of multidivisional

rms by introducing a second social reference

group, consisting of sister divisions that compete

for corporate resources, in addition to the economic

reference group1. With varying degrees of cen-

tralization (or alternatively divisional autonomy),

multidivisional rms feature headquarters that allo-

cate resources among divisions and divisions that

focus on business-level strategy, implementation,

1In this study, we do not differentiate between “divisions” and

“business units” or other similar terminology; similarly, we do

not differentiate between “multidivisional rms” and “diversied

rms.”

and operations (Chandler, 1962; Williamson,1975).

While corporate strategy often dictates that divi-

sions or business units are designed to create syn-

ergies from sharing capabilities and resources such

as manufacturing or distribution (Ansoff, 1965;

Collins and Montgomery, 1997), it is well estab-

lished that competition for corporate resources

among divisions of multidivisional rms exists.

Depending on the particular rm and the degree of

centralization, this internal competition may vary

in intensity. Following March (1962), and Cyert

and March (1963), we model the overall rm as a

conict management system or political coalition

of members with different preference functions2.

Cyert and March (1963) explicitly state that com-

plete goal consistency among the different coalition

members is generally not achievable: “Except at the

level of nonoperational objectives, there is no inter-

nal consensus. The procedures for ‘resolving’ such

conict do not reduce all goals to a common dimen-

sion or even make them obviously internally consis-

tent” (Cyert and March, 1963: 117).

We submit that the management teams of differ-

ent divisions within a multidivisional rm represent

different coalition members with divergent prefer-

ence functions. They often manipulate information,

misdiagnose problems, and devise solutions in

ways that reinforce their power and importance

relative to other members of the coalition (March,

1988; Pfeffer, 1992). They also resort to vari-

ous “inuence activities” to win more favorable

treatment from the corporate ofce, for instance,

in budget allocation, human resources, and new

product introduction (Lechner and Floyd, 2012;

Milgrom and Roberts, 1988). Getting a larger share

of corporate resources means increased prestige for

division managers and more resources to invest in

division projects. This results in a second social

reference group for each division,the political

reference group consisting of sister divisions, and

the associated political reference point representing

the performance of sister divisions.

It is important to incorporate both economic

and political reference points into division-level

adaptive aspirations of multidivisional rms, since

2For those unfamiliar with the concept of a political coalition,

a reasonable example is the Democratic Party in the USA.

As a political coalition, this party consists largely of union

members, the poor, teachers, single women, governmentworkers,

minorities, college professors, and young voters. These groups

have different goals and interests on some important issues, but

cooperate in a manner that benets all to some degree.

Copyright © 2016 John Wiley & Sons, Ltd. Strat. Mgmt. J.,38: 1435–1454 (2017)

DOI: 10.1002/smj