MNE Headquarters Disaggregation: The Formation Antecedents of Regional Management Centers

• Author: Andreas P. J. Schotter,Brian C. Pinkham,Maximilian Stallkamp
• Date: 01 December 2017
• Published date: 01 December 2017
• DOI: http://doi.org/10.1111/joms.12285

MNE Headquarters Disaggregation: The Formation

Antecedents of Regional Management Centers

Andreas P. J. Schotter, Maximilian Stallkamp and

Brian C. Pinkham

Western University Canada

ABSTRACT This research examines region-bound headquarters disaggregation in multinational

enterprises (MNEs). We link the formation of regional management centres – both dedicated

regional headquarters (RHQs) and regional management mandates (RMMs) granted to

operating subsidiaries – to the complexity argument underlying organizational information

processing theory. We demonstrate how different dimensions of complexity associated with the

number and dispersion of an MNE’s subsidiary network in a focal region affect whether, and

in which form, region-bound headquarters disaggregation takes place. Additionally, we

consider boundary conditions affecting RMC formation based on within-region experience,

global MNE footprint, and between-region effects. Empirically, we utilize a large global

dataset of Japanese MNE foreign investments between 1992 and 2014, which allows us to

perform event history analyses.

Keywords: event history analysis, information processing theory, MNE headquarters

disaggregation, regionalization, regional headquarters formation

INTRODUCTION

Despite the proliferation of research on multinational enterprise (MNE) regionalization

(Arregle et al., 2013; Ghemawat, 2003; Kim and Aguilera, 2015; Rugman and Verbeke,

2004), few studies have examined the structural units that support regional MNE strat-

egies. One important way in which MNEs adapt their structure to better support the

regional component of their strategies is the disaggregation of select headquarters (HQ)

functions to the regional level through the formation of regional management centers

(RMCs) (Ambos and Schlegelmilch, 2010; Enright, 2005a, 2005b; Hoenen et al., 2014;

Verbeke and Asmussen, 2016).

Address for reprints: Andreas P. J. Schotter, Ivey Business School, Western University Canada, 1255

Western Road, London, ON, Canada N6G 0N1 (aschotter@ivey.ca).

V

C2017 John Wiley & Sons Ltd and Society for the Advancement of Management Studies

Journal of Management Studies 54:8 December 2017

doi: 10.1111/joms.12285

Recent research has identified two different forms of RMCs: Dedicated regional

headquarters (RHQs) and regional management mandates (RMMs) delegated to oper-

ating subsidiaries within a region (Alfoldi et al., 2012; Chakravarty et al., 2017; Piekkari

et al., 2010; Verbeke and Asmussen, 2016). Alfoldi et al. (2012) argued that ‘there are

strong reasons why RMMs should be treated as conceptually different from dedicated

RHQs’ (p. 277). They suggest that since RMMs draw on the existing resources of oper-

ating subsidiaries, they perform select region-specific HQ functions more flexibly and at

lower cost than dedicated RHQs. However, RMMs could experience control and coor-

dination challenges, due to their limited and sometimes underspecified decision-making

authority within the MNE network (Alfoldi et al., 2012). Chakravarty et al. (2017) found

significant differences between RHQs and RMMs, notably with respect to size, regional

mandates, and parent-firm characteristics. While these studies help us understand con-

ceptual and empirical differences in existing RMMs and RHQs, theory development

related to the antecedents of RMM and RHQ formation has not been forthcoming.

We extend the literature on RMCs by considering under which conditions MNEs disag-

gregate HQs to the regional level, and which form of disaggregation they choose, by draw-

ing on organizational information processing theory (IPT) (Galbraith, 1973, 1974, 1977;

Thompson, 1967; Tushman and Nadler, 1978). IPT suggests that organizations should

create a structural fit between their information processing needs and their information

processing capabilities in order to overcome information complexity (Premkumar et al.,

2005; Wolf and Egelhoff, 2002). Organizations achieve this by creating self-contained tasks

to streamline information processing needs, or creating vertical or lateral information sys-

tems, which increase information processing capacity (Galbraith, 1973).

Building on research in international business, we argue that RMCs provide MNEs

with information processing capacity and fit at the regional level. Thus, RMCs represent

a structural response to information processing needs and to internal complexity result-

ing from MNEs’ international activities (Egelhoff et al., 2013; Roth, 1995; Tihanyi and

Thomas, 2005; Wolf and Egelhoff, 2002). Specifically, they are a response to the inter-

nal complexity arising from an MNE’s subsidiary network in a geographic region. We

consider two observable proxies of internal complexity: footprint-based complexity

(Devinney et al., 2013) and dispersion-based complexity (Ahmadjian, 2016; Asmussen,

2009). Footprint-based complexity refers to the number of individual subsidiaries within

a focal region. Dispersion-based complexity refers to the dispersion of subsidiaries across

different countries within a region (Boyacigiller, 1990; Roth and Kostova, 2003). We

propose that the two dimensions of complexity affect RMC formation differently, and

that RHQs and RMMs differ in their information processing characteristics. Thus,

whether an MNE establishes an RHQ or an RMM represents a differentiated structural

response to the specific information processing demands it faces in a region.

Empirically, we utilize a global dataset of Japanese MNE foreign direct investment

(FDI) between 1992 and 2014. This allows us to go beyond the analysis of cross-

sectional associations between strategy and structure (Egelhoff, 1982; Egelhoff et al.,

2013; Wolf and Egelhoff, 2002) by performing event history analyses of RHQ and

RMM formation. We investigate the regional expansion of MNEs from the time of their

first FDI in a given region, and link HQ disaggregation decisions to the growing com-

plexity associated with MNEs’ regional activities. In addition, our data allow us to

1145Regional MNE Headquarters Disaggregation

V

C2017 John Wiley & Sons Ltd and Society for the Advancement of Management Studies