Strategic Management Journal

Strat. Mgmt. J., 30: 885894 (2009)

Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/smj.770
Received 15 February 2008; Final revision received 25 February 2009

RESEARCH NOTES AND COMMENTARIES

HORIZONTAL ALLIANCES AS AN ALTERNATIVE TO
AUTONOMOUS PRODUCTION: PRODUCT
EXPANSION MODE CHOICE IN THE WORLDWIDE
AIRCRAFT INDUSTRY 19452000
2

BERNARD GARRETTE,1 * XAVIER CASTANER,

and PIERRE DUSSAUGE1
1

HEC Paris, France

HEC Lausanne, Faculty of Business and Economics, University of Lausanne,
Lausanne, Switzerland
2

This study investigates why firms choose to undertake product expansion through alliances with
competitors rather than on their own. We highlight product heterogeneity as a determinant of
this make or ally choice. We propose that firms turn to horizontal alliances in order to implement
product expansion projects that require greater resources than those available to them. More
precisely, we hypothesize that a firm is more likely to launch a new product through a horizontal
alliance rather than autonomously when the resource requirements of the project are greater,
the resources available to the firm are more limited, there is a mismatch between resource
endowment and requirement, and the firms collaborative competence allows it to better cope with
the interorganizational concerns that collaboration with competitors raises. We find support for
our arguments on a sample of 310 new aircraft developments launched between 1945 and 2000,
either by a single prime contractor or as a horizontal alliance in which prime contractorship is
shared with another industry incumbent. Copyright 2009 John Wiley & Sons, Ltd.

INTRODUCTION
This study investigates why firms choose to undertake product expansion through alliances with
competitors rather than on their own. We argue that
both firm and product heterogeneity influence this
make or ally choice. Drawing on and expanding
Penroses theory of firm growth (Penrose, 1959),
we propose that resource conditions influence not
Keywords: alliance; collaboration; fit; governance; resource; competence; growth

Correspondence to: Bernard Garrette, HEC Paris, 1, rue de la

Liberation, 78351 Jouy-en-Josas Cedex, France.
E-mail: garrette@hec.fr

Copyright 2009 John Wiley & Sons, Ltd.

only the decision to grow, but also the choice of

expansion mode: firms turn to horizontal alliances
to implement projects that require greater resources
than those available to them.
Empirical studies that have examined horizontal alliance formation have primarily drawn on
resource approaches and have emphasized firm
heterogeneity over product heterogeneity (Hamel,
Doz, and Prahalad, 1989; Nohria and GarciaPont, 1991; Eisenhardt and Schoonhoven, 1996;
Doz and Hamel, 1998; Stuart, 1998; Ahuja, 2000;
Chung, Singh, and Lee, 2000; Rothaermel and
Boeker, 2008). In this stream of research, only a
few studies have considered horizontal alliances
as an alternative to autonomous expansion (Shan,

886

B. Garrette, X. Castaner, and P. Dussauge

1990; Mitchell and Singh, 1992). In addition, these

studies have produced mixed results: while Shan
(1990) found that poorly endowed firms are more
likely to form alliances to compensate for their
weaknesses, both Mitchell and Singh (1992) and
Stuart (1998) observed that stronger incumbents
are more likely to cooperate because their resource
endowment makes them more attractive as potential partners. More recent research has combined
these two logics in a need and attractiveness view
of alliances, which argues that firms turn to horizontal alliances to fill a particular resource gap,
but that they must also possess valuable resources
in other areas in order to attract potential partners (Eisenhardt and Schoonhoven, 1996; Ahuja,
2000; Rothaermel and Boeker, 2008). We claim
that the ambiguous findings on the impact of
firm resource endowment on horizontal alliance
formation result from the fact that most previous research has overlooked product heterogeneity
as a driver of alliance formation. This research
stream has implicitly assumed that product expansion creates similar resource requirements across
an entire business domain: all firms undertaking
product expansion within their industry are thus
assumed to face the same resource challenge and
will opt for collaborative or autonomous expansion on the sole basis of their resource endowment. We argue instead that, even within the
same business domain, varying levels of resource
requirements across products, together with differences among firms resource endowments, influence the choice between autonomous expansion
and expansion through horizontal collaboration. In
other words, what drives a firm to collaborate with
competitors is primarily the mismatch between the
resources required by the envisioned product and
the resources available to the firm.
We define autonomous product expansion as a
firms independent decision to launch a new product and to fully bear the risks and rewards associated with such a decision. Product expansion
through horizontal alliances, in contrast, involves
sharing product definition, risks, and returns with
firms competing in the same business domain. In
our definition, both autonomous product expansion
and horizontal alliances can coexist with outsourcing activities to suppliers and subcontractors, or
even with vertical collaboration with such partners.
Copyright 2009 John Wiley & Sons, Ltd.

The empirical setting of our study is the worldwide aircraft industry. We test our model on a sample of 310 new aircraft projects undertaken by aircraft manufacturers from 1945 until 2000, either as
single prime contractors (autonomous expansion)
or joint prime contractors (horizontal alliance). We
find support for all our hypotheses. More precisely, we observe that aircraft manufacturers are
more likely to undertake a new project through
horizontal collaboration rather than autonomously
when the resource requirements of the project are
greater, when the firms resource endowment is
more limited, and when the firm has substantial
collaborative competence. Further, consistent with
our main argument, we find a significant impact of
the interaction between project resource requirements and firm resource endowment on the choice
between collaborative and autonomous production.
This supports our view that the effect of available
firm resources is contingent on product resource
requirements.

THEORY DEVELOPMENT AND

HYPOTHESES
The crux of our argument is that a firms decision to expand autonomously or through a horizontal alliance is determined by the fit between
the resource requirements of the envisioned product and the resources available to the firm. We
argue that product resource heterogeneity explains
why a firm with a given resource endowment
may undertake some new products on its own and
others through horizontal alliances. Conversely,
firm resource heterogeneity explains why, when
undertaking similar products with similar resource
requirements, some firms choose autonomous expansion while others opt for an alliance. We
also acknowledge that the firms ability to reduce
interorganizational costs influences its decision to
form a horizontal alliance or go it alone (Simonin,
1997; Williamson, 1999; Anand and Khanna,
2000). More specifically, we argue that when
deciding to launch a new product, managers assess
the products resource requirements, the firms
resource endowment, and thus the firms ability to
undertake expansion on its own. When the firms
resources do not match the project requirements,
managers can opt for a horizontal alliance as an
alternative. When considering the alliance option,
Strat. Mgmt. J., 30: 885894 (2009)
DOI: 10.1002/smj

Research Notes and Commentaries

managers also take into account the firms collaborative competence, that is, its ability to address
potential collaboration hazards (Williamson, 1975,
1991; Kogut, 1988, 1989; Hamel, 1991; Gulati and
Singh, 1998).
Product heterogeneity and resource
requirements
The decision to undertake product expansion alone
or through a horizontal alliance is influenced by
the scope and magnitude of the resources required
to successfully implement such a move. We claim
that product characteristics determine the scale
and scope of resources1 needed to develop, produce, and commercialize the envisioned product
(Poppo and Zenger, 1998; Mayer and Salomon,
2006). Depending on their functional attributes,
products create different resource requirements
in terms of research, development, manufacturing, and marketing. Firms therefore make governance choices on a project-by-project basis.
Ceteris paribus, firms will be more likely to form
horizontal alliances for more complex projects
requiring greater and broader sets of difficultto-trade resources than for simpler projects that
they are more likely to undertake independently.
Indeed, through alliances, two or more incumbents
can combine or pool their resources to collectively address the resource requirements of the
envisioned product, which would otherwise be out
of reach of any one partner (Doz, 1988; Contractor
and Lorange, 1988; Hennart, 1988; Kogut, 1988).
Hence the following hypothesis:
Hypothesis 1: The greater the resource requirements created by the envisioned product, the
greater the likelihood a firm will undertake
product expansion through a horizontal alliance
rather than autonomously.
Firm heterogeneity and resource endowment
Holding a products desired characteristics constant, firms possessing most of the imperfectly
tradable resources needed to undertake the expansion project are more likely to do so on their
own, and are therefore less likely to collaborate
1

We follow Barney (1991) in considering that resources include

both tangible and intangible assets and the capabilities needed
to exploit these assets.
Copyright 2009 John Wiley & Sons, Ltd.

887

with other industry incumbents (Shan, 1990). The

more firms have, in the past, launched products in
the same domain, the more likely they are to have
built relevant production capabilities by developing and codifying activity-specific know-how and
by investing in assets redeployable to new products
in the same domain (Yelle, 1979; Simon, 1991;
Argote, 1999; Zollo and Winter, 2002; Leiblein
and Miller, 2003).
Further, building on the resource-based view, we
claim that more competent firms have an additional
incentive to opt for autonomous expansion to avoid
knowledge leakage (Rumelt, 1984; Hamel, 1991;
Barney, 1999). Moreover, stocks of resources tend
to deplete over time, and renewing an existing
resource stock is less costly than building one up
because of time compression diseconomies (Dierickx and Cool, 1989). Autonomous product expansion is a privileged way through which firms practice and refine their competences in a learning-bydoing process (Penrose, 1959; Nelson and Winter, 1982). Thus, more experienced firms are more
likely to carry out new product expansion on their
own in order to sustain their competence accumulation path and, thereby, their competitive advantage (Conner and Prahalad, 1996; Leiblein and
Miller, 2003). In contrast, less experienced firms
may be less likely to undertake product expansion
on their own because they lack some of the competences required to launch the envisioned product
and have a more limited competence base to maintain through autonomous production. Horizontal
alliances can allow relatively inexperienced firms
to pool their limited resources in order to pull the
project through. Thus, less competent firms are
likely to gain more than more competent firms
from forming a horizontal alliance (Ahuja, 2000).
In addition, the risk of losing valuable knowledge
is substantially lower for less experienced firms.
Hypothesis 2: The greater a firms production
experience in a product domain, the lower the
likelihood it will undertake product expansion in
this domain through a horizontal alliance rather
than autonomously.
The combined effect of product resource
requirements and firm resource endowment
The two prior hypotheses seem to suggest that,
when choosing product expansion mode, managers
independently consider the resources required by
Strat. Mgmt. J., 30: 885894 (2009)
DOI: 10.1002/smj

888

B. Garrette, X. Castaner, and P. Dussauge

the product and the resources available to the firm.

However, we argue that managers make these decisions based on the match between these two factors. The greater the match between available and
required resources, the more likely they are to
undertake the project on their own. Conversely,
the greater the mismatch, the more managers will
turn to alliances to pull through projects that are
beyond the firms reach. In other words, we argue
that the influence of project resource requirements
for less experienced firms is greater than for betterendowed firms. Similarly, for products with greater
resource requirements, differences in firm production experience become more critical: having little
experience in the relevant product domain amplifies the need for forming a horizontal alliance to
meet product resource requirements.
Hypothesis 3: The greater the resource requirements created by the envisioned product, the
stronger the impact of firm production experience on the likelihood to undertake product
expansion autonomously rather than through a
horizontal alliance.
Firms ability to deal with interorganizational concerns
Firms that engage in alliances incur transaction
costs in searching for partners, negotiating, monitoring, and enforcing the contract (Williamson,
1991; Parkhe, 1993; Zaheer, McEvily, and Perrone, 1998), as well as coordination costs in managing joint operations (Gulati and Singh, 1998;
White and Lui, 2005). Further, depending on its
relative bargaining power and the extent of opportunism among the partners, a given firm might
or might not obtain the expected share of the
overall value created, potentially making collaboration less profitable than autonomous production
for that firm. A firms collaborative competence
can make collaboration both more effective and
more efficient by lowering coordination costs and
reducing the potential rent appropriation hazards
and strategic risks of collaboration (Liebeskind,
1996; Simonin, 1997; Williamson, 1999; Gulati,
1999). Having collaborated in the past enables
firms to assess the trustworthiness of potential
partners, better select with whom to collaborate,
and, thus, reduce monitoring costs (Gulati, 1995).
Further, thanks to their collaborative know-how,
firms are capable of better managing day-to-day
Copyright 2009 John Wiley & Sons, Ltd.

cooperationpooling resources and jointly leveraging operational knowledgethus also incurring

lower coordination costs (Simonin, 1997; Anand
and Khanna, 2000; Kale, Dyer, and Singh, 2002;
Mayer and Argyres, 2004; Hoang and Rothaermel,
2005; Sampson, 2005). Therefore, past alliance
experience leads to collaborative competence,
which creates an incentive to form alliances
(Gulati, 1999), while a lack of such experience
may deter collaboration.
Hypothesis 4: The greater a firms horizontal
alliance experience in a product domain, the
greater the likelihood it will undertake product
expansion in this domain through a horizontal
alliance rather than autonomously.

METHODS
The empirical setting for our study is the worldwide aircraft industry from 1945 to 2000, excluding former Warsaw Pact countries. We focus on
airframe manufacturers, excluding suppliers or
subcontractors, such as engine makers, electronic
equipment providers, and others. This is an appropriate empirical setting given that airframe makers
have been forming horizontal alliances with each
other for many years.
Because of increasing resource requirements and
costs, the aircraft industry has undergone major
domestic consolidation over the years (Anand and
Singh, 1997). However, national security concerns
have limited the potential for international consolidation. This has led airframe manufacturers to turn
extensively to collaboration to jointly produce aircraft. Most of these collaborations associate industry incumbents that undertake a project together by
sharing the prime contractorship. Joint prime contractorship entails jointly defining product features,
sharing investments, risks, and benefits. In such
arrangements, the partners split the development
work among themselves and then each assumes
responsibility for manufacturing (sometimes turning to subcontracting) those elements and modules it has developed (Dussauge and Garrette,
1995). Such joint prime contractorship arrangements, which have accounted for close to 20 percent of all new aircraft developed since World War
II (Janes All the Worlds Aircraft, 19452003), are
horizontal alliances between incumbents, formed
to launch new products together.
Strat. Mgmt. J., 30: 885894 (2009)
DOI: 10.1002/smj

Research Notes and Commentaries

Population and sample
To collect our data, we considered the population
of civil and military airframe projects launched
from 1945 to 2000, which includes four types of
aircraft: fighter aircraft, jet transport aircraft, propeller aircraft, and helicopters. We gathered data
from two sources: Janes All the Worlds Aircraft
annual reports and DMS Forecast databases. The
Jane reports classify aircraft by prime contractor. We thus categorized aircraft projects as either
alliances or autonomous projects based on whether
they were listed under one or more prime contractors in Janes reports. Thus, the unit of analysis
in our study is an airframe manufacturer launching a new aircraft, either on its own or through a
horizontal alliance. We identified 412 firm-product
observations with prototype first flights taking
place between 1945 and 2000. Due to missing data,
only 310 cases (75% of the population) were available for the analysis: 84 collaborative firm-product
observations and 226 single-firm products. These
new product launches involved 79 firms. The proportion of autonomous to collaborative projects in
our sample matches the one observed at the population level.
Variables
Our dependent variable, alliance, is a dummy indicating whether a given firm launching a new aircraft is doing so as the single prime contractor
(alliance = 0) or by sharing the prime contractorship with one or several other industry incumbents
(alliance = 1).
Because product characteristics have been
shown to determine the scale and scope of resources needed to develop, produce, and commercialize a new product (Mayer and Salomon,
2006), we estimated product resource requirements
based on product functional attributes and measured it with the logarithm of the aircrafts maximum speed (in km/h) multiplied by its range (in
km) and takeoff weight (in kg), as reported by
Janes and DMS Forecast. Such indicators have
been frequently used in other studies on the aircraft industry (Frenken and Leydesdorff, 2000).
We captured a firms production experience with
the number of aircraft that the firm had previously
developed as a prime contractor in the same product domain as the focal project (fighter aircraft, jet
transport aircraft, propeller aircraft, or helicopters).
Copyright 2009 John Wiley & Sons, Ltd.

889

We measured firms horizontal alliance experience

in the considered product domain by the number
of past horizontal alliances in the same product
domain prior to the focal prototype first flight.
We controlled for nine factors that might affect
the make or ally choice. First, we used three
dummies (helicopter, fighter, and jet propeller
aircraft being the excluded category) to account
for possible resource requirement differences in
the four previously mentioned product domains
(Poppo and Zenger, 1998). Second, we included
firm size because larger firms might have easier
access to a broad range of resources, such as capital and labor, thanks to their greater reputation and
market presence (Shan, 1990). We assessed firm
size on the basis of the firms revenues (in constant millions of dollars) in the aircraft industry one
year prior to the focal prototype first flight. Third,
we controlled for product launch year by using
the products prototype first flight year (year) in
order to eliminate any trend effect. Fourth, we controlled for the level of technology with the average
development time (in months) of products in the
considered category during the five years before
the year of the focal project launch (level of technology). Greater development cycles require more
resources and are likely to prompt the choice of
alliances rather than autonomous expansion. Next
we controlled for several industry-specific factors.
We traced the military, civilian, or dual nature of
aircraft by creating two dummies; military, which
identifies products developed for military purposes
only, and civil, which identifies products developed
for commercial purposes only, dual-purpose aircraft being the excluded category. We also assessed
government support to a firm launching a new aircraft with the logarithm of the defense budget (in
constant millions of dollars) of the country from
which the firm originates (defense budget) one year
prior to the focal prototype first flight. We also
included two variables to control for the impact
of conflicts on firms make or ally choice. The
Cold War dummy denotes whether a project was
launched during the period of East-West tension
(Cold War = 1) or after the fall of the Berlin Wall
in 1989 (Cold War = 0). The wartime dummy
traces whether a new product was launched in
a country engaged in a more localized conflict.2
2
In our period of observation, such conflicts include the wars
in Korea (19501953), Vietnam (19641973), and the Gulf
(19901991) for U.S. firms; the wars in Indochina (19461954),

Strat. Mgmt. J., 30: 885894 (2009)

DOI: 10.1002/smj

Copyright 2009 John Wiley & Sons, Ltd.

0.09
0.29
0.03
0.35
0.18
0.01
0.01
0.02
0.16
0.65
0.00
0.11
0.01
0.10
0.31
0.30
0.14
0.20
0.02
0.20
0.31
0.20
0.05
0.05
0.61
0.08
0.05
0.04
0.04
0.07
0.15
0.46
0.10
0.17
0.11
0.22
0.06
0.39
0.51
0.34
0.08
0.15
0.12
0.26
0.34
0.18
0.12
0.26
0.45
0.68
0.02
0.04
0.03
0.26
0.33
0.30
0.03
0.07
0.42
0.24
0.12
0.14
0.10
0.04
0.31
0.11
0.00
0.05
0.22
0.38
0.15
0.02
0.03
0.11
0.35
0.03
0.08
0.04
0.20
0.16
0.02
0.01
0.14
0.01
0.01
0.09
0.00
0.08
0.06
0.08
0.10
0.50
0.24
0.13
0.13
0.39
0.31
0.29
0.15
0.15
0.26
0.20
0.05
0.05
0.05
0.08
0.03
0.63
0.26
0.51
0.40
0.08
0.24
0.15
0.17
0.13
0.04
0.18
0.01
0.09
0.09
0.09
0.35
0.08
0.03
0.05
0.04
0.33
0.09
0.06
0.05
0.20
0.28
0.07
0.06
1.00
6.78
16.00
13.36
5.00
1.00
1.00
1.00
7.68
1999
8.84
1.00
1.00
2.01
1.00
1.00
35.00
0.00
1.26
0.00
14.47
0.00
0.00
0.00
0.00
3.64
1949
1.13
0.00
0.00
0.10
0.00
0.00
0.00
0.45
1.10
3.06
3.12
0.82
0.41
0.45
0.43
0.79
12.64
1.27
0.48
0.50
0.61
0.39
0.37
6.84

N = 310

0.27
4.26
3.01
0.18
0.34
0.22
0.27
0.24
5.94
1974
3.03
0.35
0.44
1.12
0.82
0.16
16.52

1
Max
Min
S.D.
Mean

Summary data and correlations

Algeria (19541962), and the Gulf (19901991) for French

firms; the wars in the Falklands (1982) and the Gulf (19901991)
for U.K. firms; the Falklands war (1982) for Argentinean firms;
the three Indo-Pakistan wars (19471948, 1965, 1971) for Indian
and Pakistani firms; the two Arab-Israel conflicts (1967, 1973)
for Israeli firms; and the Gulf war (19901991) for Canadian
and Dutch firms.
3
We used the Huber-White sandwich cluster procedure provided by STATA. The results we obtained when using this cluster
procedure are substantially the same as those obtained when not
using it (i.e., running a simple robust logit): both standard errors
and confidence intervals remain virtually unchanged, which suggests that nonindependence among observations is not an issue
in our study.

Table 1.

Table 1 provides the descriptive statistics and the

correlation matrix for all the variables, and Table 2
displays the logistic regression results.
Inclusion of the explanatory variables (Models 3 and 4) improves model significance compared to the model with only controls (Model 1).
Comparing Models 2 and 3 reveals that the impact
of production experience is significant only when
product resource requirements are introduced in
the equation. This supports our claim that accurately modeling the influence of firm resource
endowment on expansion mode choice requires

FINDINGS

Alliance
Product resource requirements
Production experience
Product res. req. prod. exp.
Horizontal alliance experience
Helicopter
Fighter
Jet
Firm size
Year
Level of technology
Civil
Military
Defense budget
Cold War
War time
Nb of competing aircraft

10

11

12

13

14

15

16

Indeed, in addition to the impact they may have

on defense budgets, local conflicts may lead to less
international cooperation with firms from countries
that are not engaged in the same war. Finally, we
controlled for industry structure in each product
domain with a variable (number of competing aircraft) that captures the number of aircraft offered
in the same category (helicopter, fighter, jet, or
propeller aircraft) at the time of product launch.
While prior literature on alliance formation has
suggested that more fragmented industries offer
greater opportunities for collaboration (Hagedoorn,
1995), thus leading to a greater alliance propensity,
it can also be argued that the market power benefit
of horizontal alliances is greater in more concentrated industries (Jarillo, 1988; Kogut, 1988).
To test our hypotheses, we used a logistic regression model. As some of the sample firms launched
more than one project (310 projects launched by
only 79 different firms), we created a categorical variable (firm) that identified each parent firm
and we adjusted the standard errors of the logistic
regression coefficients by clustering the sample on
this variable.3

0.00

B. Garrette, X. Castaner, and P. Dussauge

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

890

Strat. Mgmt. J., 30: 885894 (2009)

DOI: 10.1002/smj

Research Notes and Commentaries

Table 2.

Logit estimates for expansion mode

(1 = horizontal
alliance)

Model 1
Coef.

Product resource
requirements
Production experience
Product res. req.
prod. experience
Horizontal alliance
experience
Helicopter
1.40
Fighter
0.39
Jet
0.53
Firm size
0.06
Year
0.06
Level of technology
0.14
Civil
0.78
Military
0.97
Defense budget
1.09
Cold War
0.12
War time
0.27
Number of competing
0.00
aircraft
Constant
118.84
N
310
Wald chi-square
60.07
Pseudo R square
0.16
(a)

891

Model 2
(a)

Robust
Std. err

Coef.

Model 3
(a)

Robust
Std. err.

Coef.

Model 4
(a)

Robust
Std. rrr.

1.13

0.25

Coef.

Robust(a)
Std. err.

1.10
0.11
0.08

0.26

0.07

0.07

0.14

0.07

0.76

0.18

0.89

0.18

0.85

0.20

0.67
0.60
0.52
0.26
0.02
0.19
0.35
0.41
0.31
0.47
0.40
0.03

1.29
0.35
0.44
0.15
0.05
0.15
0.78
1.06
0.74
0.07
0.21
0.01

0.60
0.60
0.52
0.26
0.02
0.21
0.36
0.45
0.30
0.47
0.40
0.02

2.89
0.23
0.72
0.47
0.07
0.21
0.80
0.79
0.79
0.12
0.06
0.00

0.77
0.59
0.59
0.27
0.02
0.19
0.40
0.44
0.32
0.57
0.46
0.03

2.74
0.30
0.74
0.50
0.07
0.22
0.81
0.81
0.78
0.11
0.09
0.01

0.77
0.62
0.59
0.28
0.02
0.20
0.41
0.46
0.32
0.58
0.46
0.03

40.57

96.37
310
105.67
0.20

43.44

128.56
310
111.84
0.25

45.34

136.98
310
103.43
0.26

45.86

(12df)

(14df)

(15df)

0.06
0.05

(16df)

Standard errors adjusted for 79 firm clusters

p < 0.01; p < 0.05; p < 0.10

taking into account varying levels of resource

requirements at the product level. The full model
(Model 4) is significant (chi-square = 103.43, p <
0.01) and provides support for our four hypotheses.
The resource requirement and availability framework we put forth appears to be a good predictor of
product expansion mode. Our results show that a
misfit between the resources required by a product
expansion opportunity and the resources available
to the firm induces managers to opt for a horizontal alliance rather than for autonomous expansion.
Thus, both product and firm heterogeneity, as well
as their interaction, seem to be driving the make or
ally choice. Our findings confirm the importance
of considering product-level characteristics and
the resource requirements they create as determinants of governance. We also find that larger firms
tend to favor autonomous expansion over horizontal alliances, which suggests that, regardless
of their product domain capabilities, larger firms
are better able to gather the resources needed to
expand on their own. In addition, firms originating
Copyright 2009 John Wiley & Sons, Ltd.

from countries with larger military expenditures

also tend to opt for autonomous expansion over
horizontal alliances. This confirms that in the
aerospace industry, the government is an influential provider of resources through subsidies and
research programs.

DISCUSSION AND CONCLUSIONS

Our study extends Penroses theory of firm expansion by considering alternative modes of growth.
While Penrose (1959) primarily focused on organic
growth and, to a lesser extent, on acquisitions, we
consider horizontal alliances as a means of product
expansion (Kay, 2002). Whereas Penrose argued
that resource conditions influence the decision to
expand, we show that they also impact expansion
mode. Penrose (1959) claimed that firms expand to
find new uses for their underleveraged resources.
In contrast, we show that, despite resource deficiencies, firms still expand, but tend to do so in
Strat. Mgmt. J., 30: 885894 (2009)
DOI: 10.1002/smj

892

B. Garrette, X. Castaner, and P. Dussauge

collaboration with other incumbents rather than on

their own. In doing so, we highlight the importance
of matching available firm resources with project
resource requirements (Poppo and Zenger, 1998).
While, following Penrose (1959), the resourcebased view has emphasized firm resources as
the prime determinant of strategy (Rumelt, 1984;
Barney, 1999), we stress the important role that
differences in resource needs plays in a firms
choice of one strategy alternative over another.
In recognizing that governance decisions are also
driven by resource needs, in addition to available resources and potential transaction issues, we
respond to Winters (1988) call for integrating
production and exchange perspectives (see also
Hennart, 1988).
We also contribute to the alliance formation
literature by highlighting the influence of product heterogeneity on the make or ally choice.
Prior research on alliance formation has focused
on firms resource and capability endowment and
modeled firms propensity to form alliances on the
basis of the strength and scope of this endowment,
regardless of the resource needs that arise from
the particular features of each expansion project
(Eisenhardt and Schoonhoven, 1996; Stuart, 1998;
Ahuja, 2000). Instead, we model the decision to
expand through alliances or autonomously on a
project-by-project basis, as a function of both product and firm features and their interaction. Our
results demonstrate that a firms resource limitations do not influence its likelihood to collaborate
in a linear way: project resource requirements not
only condition, but also exacerbate the firms likelihood to choose collaboration over autonomous
expansion.
While prior research has yielded somewhat
ambiguous results on the link between firm resources and alliance formation, our findings provide
a better understanding of the resource determinants
of interfirm collaboration. Indeed, earlier research
has alternatively pointed to resource weaknesses
and resource strengths as the main determinants
of alliance formation (Shan, 1990; Mitchell and
Singh, 1992; Stuart, 1998). Approaches emphasizing both firm resource needs and firm attractiveness have provided a first attempt at resolving
such contradictions: complementary strengths and
weaknesses in the partners resource endowments
are seen as the primary drivers of alliance formation (Teece, 1986; Eisenhardt and Schoonhoven,
1996; Ahuja, 2000). This view assumes that a
Copyright 2009 John Wiley & Sons, Ltd.

firms resource needs and strengths relate to resources that are different in nature: one partners
strengths compensate for the others weaknesses
and vice versa. However, our findings suggest that
resource needs may not always relate to different
resource categories, but rather to sufficient quantities in the same resource categories (Hennart,
1988; Doz, 1988; Das and Teng, 2000; Dussauge,
Garrette and Mitchell, 2000). Similarly, attractiveness may not necessarily result from the possession of different resources, but rather from additional quantities of the same resources. Indeed, our
results suggest that firms form horizontal alliances
to pool similar resources with competitors in order
to jointly match the resource requirements of
expansion projects that they would be unable to
address on their own. In contrast, for projects with
more limited resource requirements, the same firms
tend to opt for autonomous expansion.
We acknowledge that our measures present some
limitations. In particular, our measure of firms
available resources could be improved. Future
studies should take into account both the scope and
the depth of the firms resource endowment along
the value chain, identifying the strategic nature of
resources as well as their redeployability to the
new project (Poppo and Zenger, 1998). Along the
same lines, future research should attempt to capture resource requirements more directly by assessing the scope and magnitude of resources needed
to implement the new project.
More fundamentally, a possible limitation of our
research is that we consider a firms make or ally
choice without explicitly considering the extent
to which the firm can outsource the products
components or form vertical partnerships (Masten, 1984; Dyer, 1996). We recognize that these
various governance choices might influence one
another. Indeed, outsourcing or vertical partnerships might make it possible for firms with more
limited resources to launch relatively complex
new products without forming horizontal alliances.
Also, in this research we did not consider licensing
and corporate acquisitions as alternative expansion
modes. However, we believe our framework could
apply to expansion modes other than autonomous
production and alliances.
Finally, we believe our study offers additional avenues for future research. In particular,
it shows that weaker (i.e., less well resourceendowed) firms choose to form horizontal alliances
rather than to undertake expansion projects
Strat. Mgmt. J., 30: 885894 (2009)
DOI: 10.1002/smj

Research Notes and Commentaries

autonomously. This might suggest that horizontal
alliances are a fallback option that firms choose
when they do not possess the necessary resources
to carry out product expansion on their own. This,
in turn, raises the issue of whether such governance choices are effective or not. By demonstrating that a firms resource endowment determines
its decision to produce on its own or to cooperate
with other industry incumbents, our study clearly
calls for considering endogeneity when examining
alliance performance (Shaver, 1998).

ACKNOWLEDGEMENTS
We would like to acknowledge the support of
the CNRS-GREGHEC lab and of the Atos Origin Chair on Growth Strategies and Integration
Management, as well as the research assistance of
Louis Mulotte. We are also grateful to Editor Will
Mitchell and two anonymous reviewers for their
constructive comments and help in improving this
paper. We also thank Africa Arino, Bob Gibbons,
Nandini Rajagopalan, and Jeffrey Reuer for their
valuable comments on previous versions of this
paper.

REFERENCES
Ahuja G. 2000. The duality of collaboration: inducements
and opportunities in the formation of interfirm
linkages. Strategic Management Journal , March
Special Issue 21: 317343.
Anand BN, Khanna T. 2000. Do firms learn to create
value? The case of alliances. Strategic Management
Journal , March Special Issue 21: 295315.
Anand J, Singh H. 1997. Asset redeployment, acquisitions and corporate strategy in declining industries.
Strategic Management Journal , Summer Special Issue
18: 99118.
Argote L. 1999. Organizational Learning. Kluwer:
Amsterdam, The Netherlands.
Barney JB. 1991. Firm resources and sustained competitive advantage. Journal of Management 17(1):
99120.
Barney JB. 1999. How a firms capabilities affect
boundary decisions. Sloan Management Review 40(3):
137145.
Chung S, Singh H, Lee K. 2000. Complementarity, status
similarity and social capital as drivers of alliance
formation. Strategic Management Journal 21(1):
122.
Conner K, Prahalad CK. 1996. A resource-based theory
of the firm: knowledge versus opportunism. Organization Science 7(5): 477501.
Copyright 2009 John Wiley & Sons, Ltd.

893

Contractor FJ, Lorange P. 1988. Why should firms

cooperate? The strategy and economics basis for
cooperative ventures. In Cooperative Strategies in
International Business, Contractor FJ, Lorange P
(eds). Lexington Books: Lexington, MA; 331.
Das T, Teng B. 2000. A resource-based theory of
strategic alliances. Journal of Management 26(1):
3161.
Dierickx I, Cool K. 1989. Asset stock accumulation and
sustainability of competitive advantage. Management
Science 35(12): 15041511.
Doz Y. 1988. Technology partnerships between larger
and smaller firms: some critical issues. In Cooperative
Strategies in International Business, Contractor FJ,
Lorange P (eds). Lexington Books: Lexington, MA;
317338.
Doz Y, Hamel G. 1998. Alliance Advantage. Harvard
Business School Press: Boston, MA.
Dussauge P, Garrette B. 1995. Determinants of success
in international strategic alliances: evidence from the
global aerospace industry. Journal of International
Business Studies 26(3): 505530.
Dussauge P, Garrette B, Mitchell W. 2000. Learning
from competing partners: outcomes and durations of
scale and link alliances in Europe, North America and
Asia. Strategic Management Journal 21(2): 99126.
Dyer JH. 1996. Specialized supplier networks as a
source of competitive advantage: evidence from the
auto industry. Strategic Management Journal 17(4):
271291.
Eisenhardt KM, Schoonhoven CB. 1996. Resource-based
view of strategic alliance formation: strategic and
social effects in entrepreneurial firms. Organization
Science 7(2): 136150.
Frenken K, Leydesdorff L. 2000. Scaling trajectories in
civil aircraft (19131997). Research Policy 29(3):
331348.
Gulati R. 1995. Does familiarity breed trust? The
implications of repeated ties for contractual choices.
Academy of Management Journal 38(1): 85112.
Gulati R. 1999. Network location and learning: the
influence of network resources and firm capabilities
on alliance formation. Strategic Management Journal
20(5): 397420.
Gulati R, Singh H. 1998. The architecture of cooperation:
managing coordination costs and appropriation
concerns in strategic alliances. Administrative Science
Quarterly 43(4): 781814.
Hagedoorn J. 1995. A note on international market leaders and networks of strategic technology partnering.
Strategic Management Journal 16(3): 241250.
Hamel G. 1991. Competition for competence and
inter-partner learning within international strategic
alliances. Strategic Management Journal, Summer
Special Issue 12: 83103.
Hamel G, Doz YL, Prahalad CK. 1989. Collaborate with
your competitorsand win. Harvard Business Review
67(1): 133139.
Hennart J-F. 1988. A transaction costs theory of equity
joint ventures. Strategic Management Journal 9(4):
361374.
Strat. Mgmt. J., 30: 885894 (2009)
DOI: 10.1002/smj

894

B. Garrette, X. Castaner, and P. Dussauge

Hoang H, Rothaermel FT. 2005. The effect of general

and partner-specific alliance experience on joint R&D
project performance. Academy of Management Journal
48(2): 332345.
Janes Publishing, various editors. Annual supplements
19452003. Janes All the Worlds Aircraft. Janes
Publishing Company: London, U.K.
Jarillo JC. 1988. On strategic networks. Strategic
Management Journal 9(1): 3141.
Kale P, Dyer JH, Singh H. 2002. Alliance capability,
stock market response, and long-term alliance success:
the role of the alliance function. Strategic Management
Journal 23(8): 747767.
Kay N. 2002. Hercules and Penrose. In The Growth of
the Firm. The Legacy of Edith Penrose, Pitelis C (ed).
Oxford University Press: Oxford, UK; 81101.
Kogut B. 1988. Joint ventures: theoretical and empirical
perspectives. Strategic Management Journal 9(4):
319332.
Kogut B. 1989. The stability of joint ventures: reciprocity
and competitive rivalry. Journal of Industrial
Economics 38(2): 183198.
Leiblein MJ, Miller DJ. 2003. An empirical examination
of transaction- and firm-level influences on the vertical
boundaries of the firm. Strategic Management Journal
24(9): 839859.
Liebeskind JP. 1996. Knowledge, strategy, and the theory
of the firm. Strategic Management Journal , Winter
Special Issue 17: 93107.
Masten SE. 1984. The organization of production:
evidence from the aerospace industry. Journal of Law
and Economics 27(2): 403417.
Mayer K, Argyres N. 2004. Learning to contract:
evidence from the personal computer industry.
Organization Science 15(4): 394410.
Mayer K, Salomon R. 2006. Capabilities, contractual
hazards, and governance: integrating resource-based
and transaction cost perspectives. Academy of
Management Journal 49(5): 942959.
Mitchell W, Singh K. 1992. Incumbents use of preentry alliances before expansion into new technical
subfields of an industry. Journal of Economic Behavior
& Organization 18(3): 347372.
Nelson RR, Winter SG. 1982. An Evolutionary Theory of
Economic Change. Belknap Press: Cambridge, MA.
Nohria N, Garcia-Pont C. 1991. Global strategic linkages
and industry structure. Strategic Management Journal ,
Summer Special Issue 12: 105124.
Parkhe A. 1993. Strategic alliance structuring: a game
theoretic and transaction cost examination of interfirm
cooperation. Academy of Management Journal 36(4):
794829.
Penrose ET. 1959. The Theory of the Growth of the Firm.
Oxford University Press: Oxford, UK.
Poppo L, Zenger T. 1998. Testing alternative theories of
the firm: transaction cost, knowledge-based, and mea-

Copyright 2009 John Wiley & Sons, Ltd.

surement explanations for make-or-buy decisions in

information services. Strategic Management Journal
19(9): 853877.
Rothaermel FT, Boeker W. 2008. Old technology meets
new technology: complementarities, similarities, and
alliance formation. Strategic Management Journal
29(1): 4777.
Rumelt R. 1984. Towards a strategic theory of the firm.
In Competitive Strategic Management, Lamb R (ed).
Prentice-Hall: Englewood Cliffs, NJ; 556570.
Sampson RC. 2005. Experience effects and collaborative
returns in R&D alliances. Strategic Management
Journal 26(11): 10091031.
Shan W. 1990. An empirical analysis of organizational
strategies by entrepreneurial high-technology firms.
Strategic Management Journa, 11(2): 129139.
Shaver J. 1998. Accounting for endogeneity when
assessing strategy performance: does entry mode
choice affect FDI survival? Management Science
44(4): 571585.
Simon H. 1991. Bounded rationality and organizational
learning. Organization Science 2(1): 121134.
Simonin BL. 1997. The importance of collaborative
know-how: an empirical test of the learning
organization. Academy of Management Journal 40(5):
11501174.
Stuart T. 1998. Network positions and propensities to
collaborate: an investigation of strategic alliance formation in a high-technology industry. Administrative
Science Quarterly 43(3): 668698.
Teece DJ. 1986. Profiting from technological innovation.
Research Policy 15(6): 285305.
White S, Lui SS-Y. 2005. Distinguishing costs of
cooperation and control in alliances. Strategic
Management Journal 26(10): 913932.
Williamson OE. 1975. Markets and Hierarchies. Free
Press: New York.
Williamson OE. 1991. Comparative economic organization: the analysis of discrete structural alternatives.
Administrative Science Quarterly 36(2): 269296.
Williamson OE. 1999. Strategy research: governance
and competence perspectives. Strategic Management
Journal 20(12): 10871108.
Winter S. 1988. On Coase, competence, and the corporation. Journal of Law, Economics & Organization 4(1):
163181.
Yelle LE. 1979. The learning curve: historical review
and comprehensive survey. Decision Sciences 10(2):
302328.
Zaheer A, McEvily B, Perrone V. 1998. Does trust
matter? Exploring the effects of interorganizational
and interpersonal trust on performance. Organization
Science 9(2): 141159.
Zollo M, Winter SG. 2002. Deliberate learning and
the evolution of dynamic capabilities. Organization
Science 13(3): 339351.

Strat. Mgmt. J., 30: 885894 (2009)

DOI: 10.1002/smj