The Impact of Frictions in Routine Execution on Economies of Scope

• Author: Francisco Brahm,Marcos Singer,Jorge Tarzijan
• DOI: http://doi.org/10.1002/smj.2643
• Date: 01 October 2017
• Published date: 01 October 2017

Strategic Management Journal

Strat. Mgmt. J.,38: 2121–2142 (2017)

Published online EarlyView 9 March 2017 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/smj.2643

Received 11 September 2015;Final revision received24 December 2016

The Impact of Frictions in Routine Execution on

Economies of Scope

Francisco Brahm,1,2Jorge Tarzijan,1*and Marcos Singer1

1Pontificia Universidad Católica de Chile, Santiago, Chile

2University of Cambridge, Cambridge, UK

Research summary: Based on a detailed database of a beverages producer-distributor that

expanded its product variety by leveraging its logistic network, we show that product diversica-

tion generates economies of scope and also higher operational costs. The result is an inverted-U

relationship between variety and productivity: When the rm offers few additional categories,

productivity grows, but as the number of categories rises, the costs of executing the operational

routines increase rapidly and productivity falls. The negative effect on productivity increases if

the added product category is moredissimilar to previous ones, and decreases with learning from

operational experience. Our results highlight how frictions at the operational level can limit the

benets of diversication, even in the absence of other sources of diseconomies, such as increased

coordination needs.

Managerial summary: One of the prevalent reasonsfor companies to expand to adjacent product

lines is attaining economies of scope. However, such growth strategy also generates operational

frictions, even if the day-to-day routinesdo not appear to change at all. Product diversity is disrup-

tive for routine execution, as it requires coordination and exception handling, and may ultimately

overcome any efciency obtained fromgrowth. We estimate the relevanceof such operational fric-

tion using data from a beverages distribution network. When product variety is low, additional

categories do generate efciency, but after reaching a given threshold, friction prevails. We nd

that operational friction increaseswhen products are more dissimilar,but is attenuated when work-

ers learn from their own and other’s experience. Copyright © 2017 John Wiley & Sons, Ltd.

Introduction

The origin and rationale for the existence of mul-

tiproduct rms have been a longstanding research

topic in economics and management (e.g., Penrose,

1959; Teece, 1980, 1982). In the past three decades,

the resource-based view (RBV) has provided a the-

oretical framework that is useful for explaining why

rms diversify (Barney, 1986; see the review by

Wan, Hoskisson, Short, & Yiu, 2011). According

Keywords: economies of scope; learning; product variety;

productivity; routines

*Correspondence to: Jorge Tarzijan, Vicuña Machena 4860.

Santiago, Chile. E-mail: jtarzija@uc.cl

Copyright © 2017 John Wiley & Sons, Ltd.

to the RBV, by leveraging valuable and indivis-

ible resources that are difcult to imitate across

product lines, companies could reap synergies and

improve their performance (Barney, 1997; Chatter-

jee & Wernerfelt, 1991). The canonical RBV holds

that the extent of market unrelatedness provides a

limit to diversication: leveraging a resource has a

higher (lower) payoff if the new market is related

(unrelated) to the market of origin (Barney, 1997;

Palich, Cardinal, & Miller, 2000).

The understanding of the relationships among

diversication, resource sharing, and relatedness

has become more nuanced over time. While the

logic of expanding product variety by leverag-

ing key resources has been gradually conrmed

2122 F. Brahm, J. Tarzijan, and M. Singer

and agreed upon (Neffke & Henning, 2013; Wan

et al., 2011), there has been increasing interest in

uncovering the causes and nature of the different

costs of diversication. The extant literature has

identied several sources of diversication costs,

which we broadly classify as “agency costs”— the

costs of increased misalignment in decision making

as the rm’s scope expands—(Gartenberg, 2014;

Pierce, 2012; Shaver & Mezias, 2009), “coordina-

tion costs”— the costs of coordinating interdepen-

dent business/product units— (Zhou, 2011; Zhou

& Wan, 2016), and “adjustment costs”—the var-

ious costs of adjusting and transferring practices

and routines into the new product category (Hashai,

2015; Kor & Leblebici, 2005; Natividad & Rawley,

2015; Rawley, 2010; Zahavi & Lavie, 2013).

Complementing these important advances, we

shed new light on the sources of diversicationcosts

by proposing and documenting an easy-to-miss but

important cost, which we call a “routine execu-

tion cost.” The force generating this type of cost

is the friction that is introduced into the execution

of operational routines by an increase in the vari-

ety of products. Any resource that is being lever-

aged with higher product variety has at least one

operational routine that needs to be performed that,

if not redesigned or adjusted, will face increasing

difculty handling a wider set of products.1These

frictions, as well as the consequent higher routine

execution costs, limit diversication, even if none

of the other costs listed above is present.

We provide three theoretical arguments for why

product variety is detrimental to the performance

of operational routines. First, routines draw heavily

on tacit knowledge stored in procedural memory

(Becker, 2004; Cohen & Bacdayan, 1994; Nelson

& Winter, 1982), which allows workers to execute

repetitive tasks in a semiconscious way, saving

cognitive resources and increasing execution speed.

Product variety introduces novelty, which severely

reduces these benets of roxxfutines (Cohen &

Bacdayan, 1994). Second, the time needed to solve

several of the tasks associated with routines that

1We dene routine as “a repetitive, recognizable pattern of

interdependent actions involving multiple actors” (Feldman &

Pentland, 2003, p. 96). As we discuss below, in many cases,

companies’ do not have incentivesto redesign routines, maintain-

ing them while product variety increases. By operational routine,

we refer to any routine that is geared toward executing or directly

supporting the production and delivery of the rm’s products or

services to clients (e.g., routines taking place in outbound/inbound

logistics, client ordering, and client service).

are common across businesses and sectors— such

as routing, scheduling, packing, picking, and

forecasting— increases exponentially with variety

(Woeginger, 2003). The operations management

literature has documented this negative effect

of product variety in several studies of process

efciency (e.g., Fisher & Ittner, 1999; Wan, Evers,

& Dresner, 2012). Third, routines are composed

of a number of tasks that interact closely (Becker,

2004). By disrupting any individual task, product

variety can negativelyaffect the rest of the tasks that

need to be performed to complete the routine, gen-

erating ripple effects on the execution of the routine

as a whole. We also claim that all these effects are

particularly detrimental when new products are

dissimilar to those in the current portfolio.

As we focus on operational routines, learning

can fruitfully inform our analysis. Routines are

structured from past experiences, but are improved

and rened over time by new experiences (Becker,

2004; Feldman & Pentland, 2003; Levitt & March,

1988). An important subset of the organizational

learning literature (Argote & Miron-Spektor, 2011;

Levitt & March, 1988) has examined operational

units (e.g., stores, teams, production lines), docu-

menting the benecial impacts of one’s cumulative

experiences (experiential learning) and of proximal

units’ experiences (social learning) on productivity,

quality, and costs (Darr, Argote, & Epple, 1995;

Lapré & Nembhard, 2010). Going beyond these

outcomes, we argue that experience will also neu-

tralize or eliminate frictions in routine execution

that emerge from product variety. Experience pro-

vides opportunities to learn from past disruptions to

routines, improving responses when dealing with

novel products.

Our data are derived from the Compañía Cerve-

cerías Unidas (CCU), the largest beverage pro-

ducer in Chile, which expanded from beer into soft

drinks, wines, spirits, and confectionery by leverag-

ing its logistics network, particularly the transporta-

tion and delivery activities.We use detailed monthly

data on approximately 500 trucks for a period of

2 years.

Our setting is ideally suited to tease apart our

theoretical arguments, mainly because of four

conditions. First, we clearly observe the leveraged

resource— trucks— and its productivity. Second,

the load of the truck in terms of product variety

is exogenously generated; uctuations are trig-

gered by customers’ orders. Third, the operational

routines carried out by these trucks do not change

Copyright © 2017 John Wiley & Sons, Ltd. Strat. Mgmt. J.,38: 2121–2142 (2017)

DOI: 10.1002/smj