Professional Documents
Culture Documents
discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/259517582
CITATIONS
READS
68
1 author:
Paulo N Figueiredo
Fundao Getulio Vargas
66 PUBLICATIONS 815 CITATIONS
SEE PROFILE
All content following this page was uploaded by Paulo N Figueiredo on 29 July 2016.
The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
A R T I C L E I N F O
A B S T R A C T
JEL classication:
M16
O32
Q16
Q18
Keywords:
Innovative capability accumulation
Latecomer rms
Catch-up
Competitive performance
Brazil
* Correspondence address: Brazilian School of Public and Business Administration (EBAPE), Getulio Vargas Foundation (FGV),
Praia de Botafogo, 190, 22253-900 Rio de Janeiro, RJ, Brazil. Tel.: +55 021 3799 5742.
E-mail address: pnf@fgv.br.
0923-4748/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jengtecman.2013.10.008
74
Introduction
This article reports on an empirical investigation of outcomes achieved by rms in developing and
emerging economies (known as latecomers) that are related to their accumulation of innovative
capabilities. Unlike most previous studies, this article explores outcomes in addition to technological
catch-up. The accumulation of innovative capabilities has been a central topic in the study of
latecomer rms since the early 1970s when Cooper (1970) examined the mechanisms by which
international technology transfer inuenced the long-term accumulation of these capabilities in
technology-importing rms from developing economies.
Research on innovative capabilities in latecomer rms moved forward with the work of a Latin
American group led by Jorge Katz that initiated the rst systematic research program on these issues
in the mid-1970s. Drawing on detailed rm-level studies, this group unveiled signicant technological
capabilities that permitted rms to undertake diverse innovative activities across different industries
(for a compilation see Katz, 1987). In so doing, they challenged contemporaneous prevailing
arguments that technological activities in latecomer rms lacked creativity and were based merely on
the use of technology generated in advanced economies. This initiative inuenced the emergence of
other studies in Asia (e.g., Bell et al., 1982; Lall, 1987) that eventually gave rise to a research eld
devoted to understanding the process of technological capability accumulation in latecomer rms and
industries (for analytical overviews, see Bell and Pavitt, 1993; Bell, 2006; Bell and Figueiredo, 2012).
Following the rise of several developing economies in the early 1990s, there have been a steadily
growing number of studies on innovative capability accumulation in latecomer rms, including
studies on its sources, underlying learning mechanisms, and consequences. However, when studies
examine the consequences of accumulating innovation capabilities, they focus primarily on the ways
in which latecomer rms close their innovative capability gaps with their counterparts in advanced
economies or engage in technological catch-up. Consequently, there is a paucity of empirical research
on the types of outcomes, other than technological catch-up, that are achieved by latecomer rms that
are accumulating innovative capabilities. Although prior research suggests that the manner in which
these capabilities are accumulated has positive and/or negative implications for latecomer rms
competitive performance (e.g., Bell et al., 1982; Katz, 1987; Figueiredo, 2002), there is a dearth of
empirical studies over the past ten years that would help extend our understanding of the
consequences of innovative capability accumulation in latecomer rms.
This study is intended to contribute to lling this research gap by exploring certain outcomes of
innovative capability accumulation (beyond technological catch-up) that latecomer rms have
achieved by drawing on innovative capabilities accumulated during their lifetimes. To that end, this
article is based on an inductive multiple-case study of homogenous innovative latecomer rms
utilizing rst-hand and long-term empirical evidence gathered in a recursive eldwork process. This
study also builds on previous related empirical research on innovative capability accumulation in
latecomer rms and combines conceptual insights from the literature on innovation in latecomer
rms and the strategic management literature. Thus, this article contributes to extending our
understanding of the role of innovative capabilities as a source of rms competitive performance. This
topic has drawn the attention of researchers from different research traditions, including strategic
management and innovative capability building in rms from emerging economies.
The remainder of this paper is structured as follows. The section Study background and research
question presents the studys background leading to the articles research question, and the section
75
Accumulation of innovative capabilities and their outcomes in latecomer rms outlines certain
conceptual perspectives on innovative capability accumulation and its outcomes in latecomers. The
section Research setting outlines the research context, followed by the research design and methods
in the section Research design and methods. These ndings are presented in the section Findings
and discussed in the section Discussion.
Study background and research question
Several studies have examined latecomer rms from different industries that have attained
innovative capabilities near or at the international innovation frontier, including producers of
automobiles and semiconductors in South Korea and Taiwan (e.g., Kim, 1997; Sher and Yang, 2005);
glass in Mexico (e.g., Dutrenit, 2000); consumer electronics, telecom and telecom-equipment in South
Korea, Taiwan, Malaysia and China (e.g., Lee and Lim, 2001; Amsden and Tschang, 2003; Hobday et al.,
2004; Choung et al., 2006; Fan, 2006; Arifn, 2010); thin-lm transistor liquid crystal display (TFTLCD) panels in Taiwan (e.g., Zhang et al., 2008); electronics in Mexico (Iammarino et al., 2008);
pharmaceuticals in India (e.g., Kale and Little, 2007); ships in South Korea and Taiwan (e.g., Sohn et al.,
2009); oil and gas in Brazil (e.g., Dantas and Bell, 2009; Silvestre and Dalcol, 2009); pulp and paper in
Brazil (e.g., Figueiredo, 2010); metals, ceramics, composites and polymers in Turkey (e.g., Yoruk,
2011); and rms located in clusters (e.g., Giuliani and Bell, 2005). However, latecomer rms may
accumulate capabilities at the level of the international production frontier but not at the international
innovation frontier, such as in the pulp and paper industry in Indonesia (e.g., Dijk and Bell, 2007).
Intertwined with the examination of innovation capability accumulation in these studies is the issue
of technological catch-up as an immediate outcome of accumulating capabilities.
With respect to factors inuencing the accumulation of innovative capabilities, there have been
several relevant studies on the role of underlying learning mechanisms (for a review see Bell, 2006;
Bell and Figueiredo, 2012). Other studies have sought to examine the role of factors other than learning
in latecomer rms current capabilities (not accumulation process), such as rm-specic factors,
including age and size (e.g., Romijn, 1999), leadership (e.g., Kim, 1997), ownership (e.g., Boehe, 2007),
industry-specic factors (e.g., Jung and Lee, 2010) and economy-wide conditions (e.g., Lall, 1992; Arza,
2005).
However, in relation to investigating the outcomes of innovative capability accumulation beyond
technological catch-up, there has been a paucity of empirical studies, although there are exceptions.
Firm-level studies have found that a rms current capabilities have either positively or negatively
inuenced the achievement of specic performance outcomes, such as energy performance (e.g.,
Piccinini, 1993), productivity growth (e.g., Tremblay, 1994), and patents (Joo and Lee, 2010) but did
not examine innovative capability accumulation. Moving further in relation to these studies and
building on past research (e.g., Bell et al., 1982; Katz, 1987), Figueiredo (2002) found interrm
differences in competitive performance based on a wide range of performance parameters associated
with the manner in which rms accumulated different types and levels of technological (production
and innovative) capabilities. Since the early 2000s, there has been growing research interest in the
effects of innovative capabilities on latecomer rm performance. However, both innovative
capability and performance have been dened and measured in different ways in various studies,
which precludes a systematic cumulative body of evidence and leads to inconclusive results.
For instance, using a denition of technological capability similar to Figueiredos (2002) but
operationalizing this construct on the basis of rms current yields, Jonker et al. (2006) found a positive
correlation between capabilities and economic performance (measured as value added) in paper
machines in West Java. Based on a cross-sectional study of 275 rms in Tanzania, Goedhuys et al.
(2008) found a weak association between technological capabilities (proxied as R&D and other
innovative activities) and labor productivity. Drawing on observations of 15 rms over 15 years in the
global integrated circuit manufacturing industry, Bapuji et al. (2011) found that external knowledge
acquisition and innovative activities (measured as patent grants) did not always lead to positive rm
performance (measured as sales). By contrast, in a sample of 215 rms in the Chinese information
technology (IT) industry, Shan and Jolly (2012) found a positive relationship among innovative
capabilities (as dened in the Oslo Manual and substantiated with manager perceptions) and
76
77
of capabilities to generate and manage change in their technologies and may then engage directly in
innovative activities at the international frontier, which is referred to in this article as technological
catch-up.
There are different routes by which latecomer rms achieve technological catch-up, such as
following the technological paths previously pursued by global leaders (technology following), skipping
stages along those paths (stage-skipping) or even by creating their own paths (path creating; Lee and
Lim, 2001). In the latter route, with which this article is concerned, latecomer rms accumulate
capabilities that enable them to take different directions in technological development from those
previously pursued by global industry leaders. This accumulation of innovative capabilities does not
necessarily stop at pre-determined end-points on current technological trajectories. Instead, such
innovative capabilities may enable rms to develop technologies, products and processes that are
different from those developed by global leaders (Bell and Figueiredo, 2012), which reects the uidity
of the international technological innovation frontier that can be explored by any latecomer
(Figueiredo, 2010).
When latecomer rms attain the capability level of undertaking world-leading innovative
activities, that is, when they catch up technologically with global leaders, their technological behavior
becomes similar to global innovative rms from advanced economies. The latecomers become
concerned with how to use, sustain, and expand their innovative capabilities to re-build and re-create
new and distinctive positions of strategic competitive advantage, perhaps even by changing, or at least
adding to, the areas of technology within which they innovate, which is an issue of concern in the
strategic management literature (e.g., Pavitt, 1991; Leornard-Barton, 1995; Eisenhardt and Martin,
2000; Teece, 2007a,b). Therefore, it is important to draw on this literature to explore this stage of the
accumulated innovative capabilities in latecomer rms. Indeed, the bodies of literature of capability
building and of strategic management seem to converge on a common concern, i.e., the outcomes that
rms achieve from their innovative capabilities.
Therefore, on the one hand, it has been argued in the latecomer literature that the ability of rms to
implement innovative activities and achieve distinctive performances reects the nature and depth of
their technological capabilities (e.g., Dosi, 1988; Lall, 1992; Bell and Pavitt, 1993). This argument has
been supported by empirical insights that show that rm capabilities permit innovative activities
(that are not always R&D-based) to be implemented with differing degrees of novelty and complexity
with important positive operational economic impacts (e.g., Enos, 1962; Hollander, 1965). Firm
innovative capabilities may generate notable improvements in operational performance (e.g., Patel
and Pavitt, 1994; Laestadius, 1998; Piccinini, 1993; Tremblay, 1994; Figueiredo, 2002) and their
absence might negatively impact performance (e.g., Bell et al., 1982).
On the other hand, several studies in the strategic management literature have assumed that
innovative capabilities operate as a source of competitive performance. Following insights from
classical studies (e.g., Penrose, 1959; Chandler, 1962), there has been a steadily growing debate over
the past decades on innovative capabilities as the fundamental source of a rms sustainable
competitive advantage and superior performance, which has been reected in different subsets of
the literature such as the resource-based view (e.g., Peteraf, 1993) and the dynamic capabilities
perspective (e.g., Hitt et al., 2000; Eisenhardt and Martin, 2000; Teece, 2007a; Helfat et al., 2007).
However, previous studies have produced inconclusive, contradictory and inconsistent results
(Zahra and Sapienza, 2006; Adegbesan and Ricart, 2007; Helfat et al., 2007). Among the reasons for
this problem, it has been argued, is the fact that most studies in that branch of the literature have been
dominated by theoretical discussions, relatively weak empirical support (Newbert, 2007;
Protogerou et al., 2011) and by a multiplicity of metrics for both innovative capability and
performance (Coombs and Bierly, 2006). Indeed, it has been argued that there has been an
overemphasis on the idiosyncratic nature of such innovative capabilities and that these capabilities
are not a guarantee of sustainable competitive performance because rms are subject to
unpredictable and uncontrolled inuences from within and without (Eisenhardt and Martin,
2000; Zahra and Sapienza, 2006; Costa et al., 2013)although these statements lack rm-level
empirical substantiation themselves. The above perspectives constitute an important conceptual
basis that helped form the research design of this study, particularly the data collection and
analytical processes outlined below.
78
Research setting
Pulp and paper rms based on forestry refer to industries classied by the International Standard
Industrial Classication of All Economic Activities (ISIC) as silviculture and other forestry activities
(ISIC class 0210) that produce feedstock for forest-based manufacturing, such as pulp and paper (ISIC
class 1701). Pulpmaking requires the separation of cellulose bers from non-cellulose materials and
impurities (e.g., lignin) to create woodpulp. Papermaking involves processes such as pulp rening and
screening, the mixing of additives, sheet forming and drying. The pulp and paper industry is processintensive and normally large-scale (Pavitt, 1984). Forestry is considered part of the pulp and paper
industry because 90% of paper pulp is currently generated from wood, and pulp is increasingly
manufactured in the same country in which the plantations are located; in addition, wood represents
55% of the average total cost of making pulp.
Since the 1990s, it has been recognized that trees that yield more cellulose generate gains across
the entire production chain in the form of savings from tree harvesting and transportation, which
minimizes the expansion of forests and reduces efuent waste (Grattapaglia, 2004). After realizing
that the pulp factory is actually the tree (Grattapaglia and Kirst, 2008), pulp and paper rms have
shifted their efforts from wood volume to wood quality. The objective is to reduce the cubic meters of
wood necessary for the production of one ton of pulp, i.e., to decrease wood-specic consumption
(WSC) (Grattapaglia and Kirst, 2008). Using different types of biotechnological processes, these forests
have become an important source of biomass and function as a platform for new products such as ber
cement, biofuels, biochemicals, bio-plastic, bio-materials, and carbon bers, in addition to services
such as CO2 sequestration (Bracelpa, 2012; www.wbcsd.com). To achieve and sustain a global
competitive position in this industryand to take advantage of these innovation opportunitiesrms
must master innovation capabilities at or near world-leading levels, particularly in planted forestry
research that is focused on developing new genetic material.
In 2011, Brazil ranked as the worlds fourth-largest pulp producer, the worlds largest producer of
hardwood pulp (eucapulp), and the ninth-largest paper producer. Of the pulp and paper produced in
Brazil, 100% is derived from planted forests, which are renewable resources. Brazil has 2.2 million
hectares of fully certied planted area for industrial use. In 2011, the revenue from Brazils pulp and
paper industry approached US$ 17 billion, yielding exports of US$ 7.2 billion and a trade balance of US$
5.1 billion. In 2011, this industry generated 128,000 direct jobs, 575,000 indirect jobs in Brazil and US$
1.75 billion in taxes. From 1970 to 2011, Brazils output of pulp and paper grew by an average of 6.8%
and 5.4% per year, respectively. During the same period, Brazils pulp and paper exports increased by
an average of 13.6% and 18.8% annually, and the value of such exports by an average of 17.3% (pulp)
and 22.7% (paper) annually, respectively. Although there are 220 rms engaged in this industry in
Brazil, six large pulp makers were responsible for 85% of the pulp output in 2010; these rms have
their own forests. The same six rms also represent 55% of the paper output. This high concentration of
output from a small number of integrated rms is the result of the substantial investment involved in
forestry and large-scale manufacturing activities (Bracelpa, 2011).
Research design and methods
This article is based on a broad ve-year study about innovative capability accumulation and its causes
and consequences in pulp and paper rms derived from Brazilian forestry during the 19502010 period.
Consistent with Pettigrew (1990) and given the paucity of empirical work regarding the relationship
between these issues, the decision was made to undertake a qualitative inductive study based on multiple
case studies and long-term evidence from rms in a similar industrial sector. This methodological
approach is appropriate for addressing gaps in the literature and the general research question;
moreover, it facilitates a better understanding of what lies behind a subtle and under-researched
phenomenon, the details and nuances of which might not be captured by other methods, including, in
particular, an aggregated analysis derived from purely quantitative methods (Eisenhardt, 1989; Yin,
2003). This methodological approach was implemented over three stages of eldwork, namely
exploratory, pilot and main stages, which involved an iterative process of data collection and analysis
with constant returns to the literature for conceptual clarication in achieving solid construct and
79
Table 1
Overview of the eld research process.
Research
elements/
stages
Exploratory
Pilot
Main
Post-eldwork
Focus and
purpose
Full operationalization
of constructs and data
validation
Data sources
Industry experts at
business associations
and related non-rm
organizations and
rms
Firm professionals
(e.g., CEOs and
industrial directors
and managers)
Firm professionals
(directors, managers,
engineers, researchers,
technicians, consultants,
human resources and
engineering
departments, R&D units,
labs, shop-oor, and
retired staff)
Non-rm professionals
(e.g., universities,
research institutes)
Firm activities and
events
Firm archival records
Targeted rm
professionals
13 formal interviews
and ve informal
meetings
Consultation with
rms archival records
and public documents
Follow-up
questionnaires
Identication of
specic construct
categories (e.g., 1520
categories for
outcomes)
Reduction of overlaps
and redundancies in
construct categories (to
1012)
Identication of nal
construct categories
Data collection
techniquesa
8 informal interviews
Consultation to
industry literature
Data analysis
internal validity (Eisenhardt, 1989). Table 1 provides an overview of the eld research process. During
that process, this author worked closely with two research assistants. All the activities throughout the
four major phases were far from linear; instead, they were somewhat recursive and intertwined.
Cases selection
The rationale for selecting the cases for this study involved a purposeful choice of rms (Miles and
Huberman, 1994) that (i) provided relevant evidence to substantiate the research question and related
constructs and (ii) were likely to both generate rich information about the issues under study and
enhance the analytical generalizability of the ndings. Therefore, a relatively homogenous sampling
(Patton, 2002) was used that provided powerful examples of the phenomenon under study
(Siggelkow, 2007), as shown in Table 2. Certain details relative to data collection and analysis
processes are provided in the Appendix.
Operationalizing the research constructs
The operationalization of these research constructs required a recursive process, including several
rounds of data collection and analysis with literature consultations, which resulted in the emergence
80
Table 2
The selected cases.
Selected cases
Suzano
Klabin
Rigesa
Aracruz
VCP-Jacare
VCP-Luiz Antonioa
a
Start-up year
1941
1945
1974
1967
1988
1988
Ownership
Brazilian
Brazilian
Foreigner
Brazilian
Brazilian
Brazilian
Business lines
Forestry
Pulp
Paper
H
H
H
H
H
None
H
H
H
H
H
H
H
H
H
None
H
H
of categories related to the research constructs as follows: innovation capabilities (as a scale of
innovation capability levels) and outcomes of innovation capability accumulation (innovation-related
performance, improving operational and environment-related performance and patterns of corporate
growth).
Innovative capability accumulation
In the operationalization of the innovative capability construct developed during the past few
decades in advanced economies, assessing innovation capabilities has been mainly based on
quantitative measures such as R&D expenditures and/or patent grants (Hagedoorn and Cloodt, 2003).
These capabilities may only become useful once rms have built up their innovative capabilities to the
point at which they involve measurable R&D activities or recorded patenting. These capabilities reect
only a fraction of a rms innovative capability and none of it with respect to rms that have only nonR&D-based innovative capabilities (Bell and Pavitt, 1993; Bell and Figueiredo, 2012). The limitation of
relying on one aggregate measure of a rms innovation capability (e.g., R&D expenditures) is that it
neglects a range of mixed technological activities that are necessary to develop and produce particular
products (Patel and Pavitt, 1994; Laestadius et al., 2005) and does not capture the process of
technological transformation that involves a spectrum of activities ranging from incremental to
radical types that can have a signicant performance impact on pulp and paper rms (Laestadius,
1998).
That limitation has been overcome by a comprehensive approach that has been the primary basis of
research in this area since the earliest studies of innovation capabilities of latecomer rms, i.e., using
qualitative assessments at the scale of technological capability levels (Katz, 1987; Bell et al., 1982; Lall,
1992; Bell and Pavitt, 1993, 1995; Bell and Figueiredo, 2012). Such an approach has been
operationalized through a typology of approaches based on revealed capability. Rather than
specically identifying capability levels in terms of particular quantities and qualities of human
resources, skills, knowledge bases, etc., such approaches have identied levels of increasing novelty
and signicance of innovative activity and then inferred that different capability levels underlie
different types of innovative activities (Bell and Figueiredo, 2012). The use of such a typology captures
what rms are able to do in technological terms by using a nuanced perspective of the levels of
capabilities required to undertake innovative activities with different degrees of novelty. Consistent
with the nature of the eld evidence, this paper draws on a modied version of the typology developed
in Lall (1992) and further rened in Bell and Pavitt (1995). The modied version of this typology
identies levels of innovative capability that range from basic to world-leading and are consistent
with the characterization of innovation in terms of degrees of novelty and complexity in technological
activities; thus, these levels are consistent with the Oslo Manual (see OECD, 2005).
Such a typology has been used intensively and successfully in studies with different degrees of
capability-level disaggregation that have covered the histories of capability building over considerable
time periods (e.g., Dutrenit, 2000; Figueiredo, 2002, 2010; Dantas and Bell, 2009) and in a much larger
number of rms over shorter periods (e.g., Hobday et al., 2004; Iammarino et al., 2008; Arifn and
Figueiredo, 2004; Arifn, 2010; Yoruk, 2011; Peerally and Cantwell, 2012). Table 3 presents a
condensed version of the typology used in this study. The rst column shows four levels of innovative
81
Table 3
Taxonomy for assessing rm innovative capabilities (condensed version).
Levels of innovative
capabilities
World leading
Advanced
Intermediate
Basic
Production capabilities
capabilities that extend from basic to world leading, and the second column provides illustrative
examples of these capability levels.
The application of this framework to this study was achieved after approximately six months work
and involved several consultations with experts in the forestry and pulp and paper industries. These
interactive and iterative consultations were used to adapt and validate the taxonomy to the
technological specics of these industries. Although Table 3 condenses the levels of technological
capabilities for forestry, pulp and paper, the original framework that was applied during our eldwork
involved the use of a specic matrix for forestry, pulp and paper. Each matrix identied levels of
technological capabilities for specic technological functions in forestry (silviculture, harvesting,
logistics, and socio-environmental management) and pulp and paper (project management, process and
production organization, process equipment, and product-centered). Project management in the
framework used herein is equivalent to the project execution capabilities in Amsden and Hikino (1994).
Outcomes of innovative capability accumulation
Innovative performance refers to the implementation of creative activities with concrete benets for
rms involving (i) implemented inventive activities, which are measured by the quantity and quality of
patents and (ii) implemented innovative activities and their benets consistent with Enos (1962) and
Hollander (1965). Creative or innovative activities may vary in terms of the degree of innovation in
their technological/market novelty or the extent to which they differ from current technologies,
which allows innovations to range from those that are close to being pure imitations to those that are
fundamentally different from anything currently existing (OECD, 2005). This type of differentiation
has been widely used, particularly for innovation analyses of latecomer rms.
Operational and environment-related performance improvement refers to technical performance
parameters related to the pulp and paper rms derived from forestry. Through the recursive processes
of data collection and analysis and in consultation with previous related studies in the pulp and paper
industry (e.g., Laestadius, 1998; Dalcomuni, 1997; Jonker et al., 2006; Dijk and Bell, 2007), the study
identied two types of country-level forestry parameters, six types of rm-level parameters for
forestry (yield, basic density of wood, density, cut-off age, wood volume per amount of pulp produced,
planting density), 15 types of pulpmaking parameters (e.g., specic consumption of steam, electricity,
water and ber losses, in addition to liquid, solid and air efuents) and six types of parameters for
papermaking (e.g., specic consumption of steam, electricity, water plus industrial efuents), in
addition to production and commercialization costs of short ber pulp.
Patterns of corporate growth. During the data collection and analysis processes, there was a
recurrence of categories such as integration, expansion, merger/acquisition, and diversication. A
consultation of the literature led to classifying these categories under patterns of growth. Prior
studies have pointed to the contribution of innovative capabilities to rm growth (e.g., Penrose, 1959;
Chandler, 1962; Amsden and Hikino, 1994; Yang, 2012). By drawing on Torres-Vargas (2006), rms
growth patterns are measured in terms related to (i) horizontal integration, (ii) vertical integration,
and (ii) diversication, which is further disaggregated into direct diversication by the rm and
[(Fig._1)TD$IG]
82
indirect diversication (spin-offs and spillovers). The latter may take the form of new enterprise
creation through the use of more or less formally organized spin-off mechanisms (Bell and Figueiredo,
2012) with positive implications for industrial development (Nelson and Pack, 1999).
The recursive processes of data collection and analysis with constant consultations with the
literature led to the creation of this studys conceptual framework (Fig. 1). In the context of this study,
the framework identies at least three types of outcomes (component C) that can be achieved by
latecomers from their innovative capability accumulation (component A) beyond technological catchup (component B). This framework recognizes that achieving these outcomes may be affected
indirectly and/or directly by other factors (components D and E). Such other factors remain outside the
scope of this article.
Findings
This section presents the articles empirical ndings. The section Innovation capability
accumulation presents evidence of accumulated innovative capabilities, and the section Outcomes
of innovative capability accumulation explores some of the outcomes that were achieved by the cases
with accumulated innovative capabilities. This study examines these issues over as long a period as
possible to capture a large part of the lifetime of these rms (19502000s).
Innovation capability accumulation
In contrast with prior reports on innovative capability building in latecomer rms, the cases
examined here did not follow the imitation to innovation path, i.e., a trajectory based on accumulating
83
Table 4
Levels of innovation capability accumulated in the researched rms.
Levels of innovation capability
Pulp
Paper
World leading
VCP-Jacare
Aracruz
Klabin
Rigesa
Suzano
VCP-Jacare
VCP-Luiz Antonio
Aracruz
Klabin
Suzano
VCP-Jacare
VCP-Luiz Antonio
Klabin
Suzano
Advanced
Intermediate
Basic
All rms
All rms
All rms
Rigesa
All rms
All rms
progressively higher capabilities from production adaptation to duplicative imitation up to R&Dbased innovation (e.g., Kim, 1997). As described in Figueiredo (2010), because of several constraints,
rms could not simply copy recognized global leaders but were instead forced to develop technologies
more suited to their own somewhat different operations. This development involved the use of
different raw materials (eucapulp) and developing an effective means to do this; these rms had to
innovate in their downstream pulp and papermaking processes because of the innovations that had
developed upstream in forestry. These rms could not simply imitate because they were developing
along a different trajectory; thus, the capability accumulation process of these rms can be
summarized as moving from non-imitation to innovation. In light of the framework presented in
Table 3, Table 4 shows the resulting levels of innovative capability accumulated by these cases for each
business line.
The ndings suggest that these innovative capabilities reect these rms proprietary resources
that underlie their technological leadership. Concurrently, certain features of these capabilities were
common across these rms (e.g., common practices of undertaking research activities, innovative
activities in pulp and paper production process, etc.) that will be discussed below. For example, these
rms sought to deepen their world-leading innovative capabilities by re-organizing their research
activities during the early 2000s. For example, VCP integrated its previously dispersed research
activities into the Centre for Pulp Technological Development, Klabin re-congured its research center
based on a review of routines and procedures, documentation and analytical processes and Aracruz
merged its research on forestry and pulp and paper into a stronger research center. From 2002 to 2008,
along with other rms and universities under the coordination of the Brazilian Agricultural
Corporation (EMBRAPA), these rms engaged in a nation-wide project called Genolyptus (the
Brazilian Network of Eucalyptus Genomics Research). This project characterized the complete
phenotypes required to study the functions of the genes in question and employed a multidisciplinary
approach involving researchers in genetics, biochemistry, molecular biology, breeding, phytopathology, wood technology and industrial process engineering. Their world-leading capabilities
permitted these rms to actively collaborate with partners in advanced economies. For example,
Suzano collaborated with the genome project led by the Joint Genome Institute (JGI) in the US by
donating a germplasm base (designated as BRASUZ1) for the complete genomic sequencing of
eucalyptus (Grattapaglia, 2011).
Outcomes of innovative capability accumulation
Innovative performance
Inventive activities: quantity and quality of patents. Table 5 shows the evolution of patents granted to
these rms. The quantity increased by 40% in the 2000s compared with the 1990s. During the 1990s,
Aracruz scored the highest number of patents in forestry; Suzano received the highest number during
the 2000s. Klabin and Suzano had the highest number of paper patents over the entire period. The
84
Table 5
Evolution of patenting activities in the study rms.
Firms and business
lines
1990s
Quantity
Aracruz
Klabin
Forestry
Pulp
Forestry
Paper
2000s
Qualication
Protection of trees against
insects and improvement in
planting and harvesting
equipment. Method for
preventing or controlling the
occurrence of stains on wood
Treatment, bleaching, rening
and test of pulp
1
12
Irrigation of seedlings
Rening process of pulp,
packaging designs, displays for
packages, devices for stacking
packages and towel racks and
paper towel
Quantity
Qualication
Formulation of compound
applied to pulp production
22
None
Packaging design; pallet of
corrugated cardboard; display
packaging design and production
process of devices based on
corrugated paper. Finish applied
to tissue paper
Rigesa
Paper
24
Suzano
Forestry
Paper
VCP
Totals
Paper
None
45
63
Sources: Brazils National Institute of Industrial Property (INPI) and United States Patent and Trademark Ofce (USPTO).
evidence in Table 5 reects the tangible outcome of the rms different types and levels of innovative
capabilities and a basis for implementing innovative activities.
Implemented innovative activities in forestry and pulp and paper
With respect to forestry, Table 6 contains 24 observations of implemented innovative activities and
their related benets from the 1970s to the 2000s. The worlds rst large-scale paper production based
on eucalyptus pulp represented an important innovative activity that emanated from Suzanos newly
developed innovative capability in the 1960s, which paved the way for the introduction of the socalled new pulp in the international market. The second major disruptive innovation was
implemented by Aracruz (mid-1970s to early-1980s), which reected its research capabilities in the
85
Table 6
Implemented innovative activities and related bets in the forestry cases.
Firms
19701980s
Aracruz
Innovative activities
!Benets
Rigesa
1990s
Aracruz
Suzano
VCP
2000s
Aracruz
86
Table 6 (Continued )
Firms
Innovative activities
!Benets
Klabin
Rigesa
Suzano
VCP
mass production of clonally propagated planting stock. For this innovation, Aracruz was awarded the
prestigious Swedish Marcus Wallenberg Prize in 1984, which recognizes world-leading technological
innovations in forestry. During the 1980s, Aracruz and Suzano developed novel eucalyptus varieties
that were both more productive and more resistant to disease, in addition to being adaptable to
Brazils climate. This innovation yielded higher biomass production per unit of planted area and
signicantly improved the quality of wood used as an industrial raw material and energy input. During
the 1990s and 2000s, Aracruz, Suzano, Klabin, Rigesa and VCP expanded their innovation activities in
forestry, which reected a deepening of their related capabilities.
Table 7 presents 28 observations from implemented innovative activities in these rms related to
pulp and paper. During the 1960s and 1970s, these rms utilized their engineering and production
capabilities to change contemporaneous production processes and process equipment, such as
chemical processes, to produce pulp and paper based on the new raw material. Process innovations
involved the development of modied process technology, which was then installed in a succession of
new plants over three decades. These innovative production-based activities might also have
contributed to an increase in average annual pulp production (19802009) in these case rms of 6.08%
compared with the 4.8% average for other rms in Brazil, whereas the paper production of these rms
grew at an average annual rate of 3.9% compared with 3.6% for other rms in Brazil during the same
period.
Since the 1980s, several innovations in the bleaching process became associated with
environmentally targeted efforts and involved research on lignin biosynthesis and the patenting of
the totally chlorine-free (TCF) pulp process that continued through the 1990s. For example, by
augmenting its research capabilities for forestry with pulp and papermaking research, Aracruz
intensied investigations in lignin biosynthesis and pollution control methods based on natural
micro-organisms. By 1992, Aracruz had adopted the elementally chlorine-free (ECF) and TCF process,
following Canada and Scandinavia. However, Aracruz went further by creating a variant in the TCF
process, which was characterized by a much lower level of absorbable organic halogens (AOX); the
process, known as alpha chlorine-free (ACF), was patented in 1997. One year later, VCP created its own
versions of the TCF process. Because of these innovations, fewer chemical products are now required
to whiten pulp used to make paper.
Operational and environment-related performance improvement
Table 8 shows several country-level performance parameters related to forestry for pulp and paper.
According to the eldwork and technical literature (WRI, 1999; Evans and Turnbull, 2004),
considering the signicant technological relevance and scale of Aracruz, VCP, Suzano, Klabin and
87
Table 7
Implemented innovative activities and related benets in pulp and paper cases.
Firms
Innovative activities
!Benets
1970s1980s
Aracruz Adaptation and use of cell membrane technology for
the hardwood-based pulp making process
1990s
Aracruz
Klabin
Suzano
Effective reduction of particulate matter and sulfurbased component emissions to the recovery boiler
Meeting of international standards and the reduction of
environmental risks at the bleaching stage
Setting of new standards in the industry regarding
environmental suitable production processes
Environmentally friendly papermaking process
Paper exports packed with the customer brand
2000s
Aracruz
88
Table 7 (Continued )
Firms
Innovative activities
!Benets
Klabin
Card barrier
Suzano
Special cardboards, such as anti-thermal and antifreeze, and other distinct characteristics dened by
the demands of customers
Tissue paper for the pharmaceutical industry
VCP
Table 8
Country-level operational performance parameters in forestry for pulp and paper.
Types of tree/
parameters
Hardwood
Rotationa (years)
Yield (m3/ha/year)
Softwood
Rotation (years)
Yield (m3/ha/year)
Brazil
Chile
Indonesia
Finland
Canada
(coastal)
USA
7 (eucalyptus)
1012
(eucalyptus)
25
9 (eucalyptus)
3540 (birch)
n.a.
n.a.
2434
n.a.
n.a.
25 (pinus
radiata)
22
7 (pinus
merkusii)
24
7080 (picea
abies)
4
45 (Douglas
Fir)
7
25 (pinus
elliottii/taeda)
10
44
15 (pinus spp)
38
Rigesa, it is highly likely that those leading parameters achieved by Brazil (Table 8) reect the
accumulation of innovative capabilities and related implemented activities of these companies. The
improvements in forestry performance parameters (Table 9) reect the case rms capabilities for
genetic manipulation and selective breeding. For instance, the rst-generation clonal forestry of
eucalyptus during the 1980s reduced wood-specic consumption (WSC) by 20%. A further 20%
reduction was subsequently achieved based on second-generation clones derived from eucalyptus
hybridization, which led to the rst large-scale commercial planting of stands of selected clones
derived from hardwood cuttings, which in turn resulted in exceptional genetic gains in growth, in
addition to adaptability to tropical conditions and higher pulp yields (Grattapaglia and Kirst, 2008).
The evidence in Tables 10 and 11 indicates signicant improvements in certain process
performance parameters for pulp and paper during the 20002009 period. For example, specic water
consumption of 36.7 m3 per ton of pulp and the mean 20.1 m3 per ton of paper achieved by the case
rms were equivalent to those attained for the Finnish and European Union best available technology
standards (www.environment.). Improvements achieved under other indicators (e.g., reductions in
Parameters
Unit
Forest yield
Basic density
of wood
Density
Cut-off age
Volume of wood
per amount of
pulp produced
Planting density
m3/ha/year
kg/m3
1970
1975
1980
1985
1990
1995
2000
2002
2003
2004
2006
2009
37
473
44
473
47
473
45
488
53
488
52
488
45
485
46
489
45
494
46
496
49
493
49
506
ton/m3
years
m3/ton pulp
trees/ha
0.47
9.5
4.2
1651
0.47
9.0
4.1
1651
0.47
8.5
3.9
1512
0.49
8.0
3.8
1512
0.49
7.5
3.7
1486
0.49
7.4
3.7
1419
0.49
7.2
3.8
1224
0.49
7.2
3.8
1259
0.49
7.1
3.8
1259
0.5
7.1
3.9
1326
0.49
7.0
3.9
1326
0.51
7.0
3.8
1326
Average annual
rate of decrease/
increase (%)
(19702009)
+0.7
+0.1
+0.2
0.78
0.2
0.5
Table 9
Evolution of some technical indicators in forestry (19702009).
89
90
Table 10
Evolution of process performance in pulp making.
Parameters
2000
Specic consumption of
Steam
Steam ton/pulp
weight (ton)
Electricity
KWh/pulp
weight (ton)
Water
m3/pulp
weight (ton)
Fiber losses
2001
2002
2003
2004
2006
2009
5.4
4.9
4.9
4.5
4.4
3.6
640.8
646.4
674.2
639.6
4.92
737
ton/day
730
571
Average annual
rate of reduction/
increase (%)
(20002009)
3.4
2.8
41.3
45.6
42.7
39.6
40.9
40.1
36.7
1.3
13.1
16.7
15.5
11.1
8.8
9.5
10.9
2
Average annual
rate of reduction/
increase
(20002009) (%)
Table 11
Evolution of some process performance in papermaking.a
Parameters
Units
2000
2001
2002
2003
2004
2006
2009
3.1
1.9
2.9
2
2.8
3.5
2.7
3.2
2.6
2
2.5
1.9
2.4
1.9
2.8
0.00
1.9
1.7
1.4
1.4
1.4
1.4
1.4
3.3
627.5
457.5
614.3
465.2
591.9
655
576.1
725.8
572
486.3
554.5
432.8
547
391.9
1.5
1.7
412
473.5
439.2
458.1
447.1
398.2
229
6.3
28
31.6
26.2
32.5
24.6
32.6
20.8
33.5
19.2
23.2
18
19.3
17.1
20.1
5.3
4.9
34.8
33.1
31.8
30.6
28.9
25.8
23.3
4.3
specic steam and electricity consumption and ber loss) might also have exerted an important
impact on cost reduction.
With reference to environment-related indicators in pulp and paper (Tables 12 and 13), the average
industrial efuent output within the pulp mills decreased by 3% annually from 2000 to 2009, whereas
the SO2 emission decreased by an annual average of 3.4%. In absolute terms, both indicators were
below the limits delineated by the Brazilian Environment Authority (Conama) and by European best
available techniques (BAT). Similarly, within the paper mills, the decrease in biochemical oxygen
demand (BOD) varied from 6.8% to 23.1% annually which, in absolute terms, were below the limits
established by Conama. Consequently, the mills environmental impact was reduced, particularly in
terms of diminished liquid efuents. The performance improvements in Tables 1013 reect the rms
innovative activities (Table 7) which are highly likely to be an outcome related to these rms
innovative capabilities.
91
Table 12
Evolution of environment-related performance in pulp making.
Types of efuents
and parameters
Liquid
Industrial efuents outputa
COD (chemical oxygen
demand)a
BOD (biochemical oxygen
demand)a
Total nitrogena
Solid
Lime mud/dregs/gritsb
Total ashesb
Air
SO2 (from chemical
recovery boiler)a
NOx (nitrogen-oxides,
from chemical recovery
boiler)a
Average TRS (Total reduced
sulfur)a
Average SO2a
Average TRS (Lime kiln)a
Units
2000
m3/pulp
weight(ton)
46.7
kg/pulp
weight (ton)
11.9
mg/Nm3
ppm
2003
Average annual
rate of reduction/
increase (20002009)%
Limits of
CONAMAc
2006
2009
42.9
38.7
35.3
3
50100
10
7.6
6.1
7.1
10
1.8
1.3
1.1
0.8
8.6
2.5
0.2
0.1
n.a.
n.a.
20.6
n.a.
33.9
11.7
43.5
15.5
68
29.1
96.2
38.5
12.2
14.1
n.a.
n.a.
8.8
24
7.1
6.4
3.4
100
n.a
239.8
187.56
237.91
0.1
470
1.64
0.82
2.13
2.42
4.4
n.a.
1.69
17.1
2.83
42.9
4.4
16.6
7.06
17.2
17.2
0.03
n.a.
n.a.
Table 13
Evolution of environment-related performance in papermaking.
Types of efuents
and parameters
Units
Average annual
rate of increase/
decrease
20002009 (%)
Limits of
CONAMA
2000
2001
2002
2003
2004
2006
2009
28.4
28.4
25.1
37.4
38.8
38.2
39.7
3.7
50100
90
80
80
46.1
43.4
34.1
31.7
10.9
50100
7.23
6.31
12.7
10.0
10.9
4.9
7.1
0.2
10
14.5
18.0
13.6
13.6
15.9
9.3
9.2
4.9
10
n.a.
n.a.
4.3
3.4
4.4
1.5
n.a.
23.1
3.8
5.0
4.32
5.4
5.6
2.4
2.0
6.8
[(Fig._2)TD$IG]
92
exports) in relation to international pulp and paper competitors are manifest in their ability to
produce high quality bleached eucalyptus kraft pulp (BEKP) for approximately US$ 225 per ton (Fig. 2).
Patterns of corporate growth
As indicated in Table 14, growth patterns based on horizontal and vertical integration prevailed
over the entire 19502000 period, which seems to have been enabled by accumulating innovative
project management capabilities that permitted these rms to design and execute plans with partners
and to coordinate expansion projects. For example, Aracruz developed novel techniques for project
engineering that permitted the rm to expand its berlines in world record time. These capabilities
seem to have paid off by allowing Aracruz and VCP to set up large-scale logistics projects that
positively affected their competitiveness.
Direct rm diversication began to become signicant only during the 2000s. These rms were
created under the ISI regime and their businesses evolved around the pulp and paper industry, with a
low degree of diversication. One exception is VCP, which is part of a large Brazilian business group
that is diversied into somewhat related areas (e.g., chemicals, cement, metals, agro-industry, pulp
and paper and banking). However, during the 2000s, these rms began to draw on their worldleading innovative capability in forestry to diversify into new activities from their innovative
capabilities in that eld, which gave rise to new high tech activities in the Brazilian pulp and paper
industries.
For example, by acquiring FuturaGene (with operations in the US, Israel, China and Southeast Asia),
Suzano was able to rmly engage in the international commercialization of modied genes and
develop trees that require less land, water consumption and fewer fertilizers, that produce less lignin
(and fewer chemicals during the pulping processes) and generate higher carbon sequestration, which
contributes to stronger competitiveness in its forestry and pulp and paper businesses. The creation of
Suzano Renewable Energy may allow Suzano to move into the new forestry segment of planted
energy forests by producing genetically modied trees with short cut-off times and caloric
properties. By drawing on its world-leading forestry innovative capabilities, Klabin intensied its
business in medicinal plants, phytotherapy and phytocosmetics. With respect to indirect rm
diversication (spin-offs and spillovers), the evidence suggests that as these rms accumulated
innovative capabilities, they also seem to have stimulated the emergence of spin-offs and spillovers,
such as the four outstanding examples described in Table 15: Imetame Metalworking, Inor
Consulting and Systems, the wood cluster in the municipality of Telemaco Borba and the forest
partnerships program.
Table 14
Patterns of corporate growth in the researched rms.
Horizontal integration
Diversication
19501980s
1990s
2000s
Aracruz: Fiberline A
Merger/acquisition to increase
production of current products
Directly (diversication
by the rm)
None
Aracruz:
organic fertilizer (with Organoeste,
Brazil)
renewable energy (with Ensyn, USA)
Suzano:
bioenergy (by creating Suzano Energia
Renovavel)
biotechnology (by acquiring FuturaGene)
Klabin
biorenery (embryonic).
toterapics
Indirectly (spin-offs/spillovers)
Vertical integration
Upstream
Downstream
94
Table 15
Examples of spin-offs and spillovers generated by these cases.
Examples of spin-offs/
spillovers
Start-up year
Origin
Description
Imetame Metalworking
1980
Aracruz
2001
Aracruz
Mid-1990s
Klabin
Forest partnerships
program
Early 1980s
Aracruz, Klabin,
Suzano and VCP
Discussion
Building on a research tradition in technological capability accumulation and its performance
implications for latecomer rms (e.g., Bell et al., 1982; Katz, 1987; Figueiredo, 2002), the purpose of
this study was to empirically investigate the types of outcomes that are achieved by latecomer rms
from the accumulation of innovative capabilities, up to the world-leading level, in addition to
technological catch-up. In contrast with most previous studies, this article has proxied innovative
capability by drawing on a comprehensive taxonomy that is based on a scale of capability levels for a
wide range of technological activities. Based on an inductive multiple-case study involving rst-hand
and long-term evidence that was derived from extensive eld investigations of a relatively
95
homogenous set of pulp and paper rms in the Brazilian forestry sector, this study scrutinized a
number of outcomes related to innovative performance and business performance as outcomes
achieved by these rms from the innovation capabilities that they had accumulated over their
lifetime. The recursive eldwork processcombined with insights from the literature on innovation in
latecomer rms and the strategic management literaturefacilitated the emergence of the framework
in Fig. 1, particularly with respect to exploring the relationships between components A, B, and C, the
ndings of which respond to the articles general research question and generate further implications
outlined below.
Discussion of ndings
Innovative capability accumulation
The researched rms accumulated innovative capabilities that eventually turned them into world
leaders in a particular segment of the pulp and paper industry, i.e., the short-ber segment
(eucalyptus). This nding is consistent with previous studies that have reported that latecomer rms
in other industries have attained leading technological positions at the international innovation
frontier (see Study background and research question). However, instead of beginning with the
accumulation of production capability and then moving into the progressive accumulation of
innovative capabilities (from adaptation to R&D-based innovation), as is typically documented in the
literature, these rms accumulated innovative capabilities that permitted them to take a direction of
technological development that differed from those directions previously pursued by global industry
leaders. Their innovative capability accumulation process involved a qualitative discontinuity from the
established technological trajectory at an early stage in the development of their capabilities, which is
rarely documented in the related literature. Additionally, as opposed to most studies of technological
capability accumulation that focus on so-called high-tech industries, this study has examined this
issue in natural resource-related rms, which are barely investigated in the literature, despite their
importance for national economies, although there are a few exceptions (e.g., Dantas and Bell, 2009;
Silvestre and Dalcol, 2009). Additional details are provided elsewhere in the literature for this
capability accumulation process (Figueiredo, 2010); thus, the following section discusses ndings
related to outcomes.
Outcomes of innovative capability accumulation
As the rms accumulated these innovative capabilities, they drew on the resources to change and
or create technologies and components of production systems beyond achieving a technological
catch-up; thus, they achieved concrete benets from the accumulation of these capabilities in terms of
innovation and business performance, which could guarantee their international competitiveness.
Thus, the study found that the following outcomes resulted from the accumulation of innovative
capabilities (i) Innovative performance (implemented inventive and innovative activities), which involved
evidence from 108 accumulated patents (Table 5) and 24 examples of signicantly innovative
activities in forestry and 28 in pulp and paper (Tables 6 and 7) of different types and with varying
degrees of complexity and novelty; (ii) Operational and environment-related performance improvement,
which involved country-level and rm-level performance parameters in forestry (Tables 8 and 9),
several performance parameters for pulpmaking and papermaking (Tables 1013), in addition to
country-level product and commercialization costs (Fig. 2); and (iii) corporate growth patterns, which
involved several examples of these patterns in the form of horizontal integration and upstream and
downstream vertical integration, and direct diversication and indirect diversication (spin-offs
and spillovers) (Tables 14 and 15). This study indicates that these outcomes were achieved by
accumulating a wide range of innovative capability levels (from basic to advanced) for diverse
technological functions (e.g., silviculture, harvesting, project management, process and production
organization, product-centered and related engineering-based capabilities) and various implementations of inventive and innovative activities with differing levels of novelty and complexity.
Specically, with respect to the nature of these capabilities and the outcomes that they generated
beyond technological catch-up, this study has yielded the following ndings.
96
First, the study documented a wide range of innovative activities, several of which were
engineering-based and incremental capabilities that have intermediated the achievement of several
improvements in operational and environment-related performance parameters that are vital for the
international competiveness of these rms. Although this result is not really new because the
importance of these types of innovative activities for rm performance has been examined in previous
research (e.g., Enos, 1962; Hollander, 1965; Bell et al., 1982; Katz, 1987; Figueiredo, 2002), studies in
the strategic management literature (see Accumulation of innovative capabilities and their outcomes
in latecomer rms) and studies focusing on latecomer rms (see Study background and research
question) have addressed innovative capabilities since the 1990s narrowly and mainly as R&D
expenditures, patent grants and/or product innovation.
Second, although these innovative capabilities were strategic for these rms, particularly the
advanced and world-leading levels, there were features of these capabilities that were common across
the rms. This result seems to contradict well-accepted assumptions that such high-level capabilities
are highly idiosyncratic resources. This does not imply the absence of distinctiveness across these
rms. However, such distinctiveness should not be attributed solely to innovative capabilities but also
to the interactions among these capabilities and other factors, which are represented in Fig. 1.
Third, the ndings do not imply that the accumulation of these innovative capabilities is any
guarantee of sustained innovative and business performance because rms may go through severe
difculties despite accumulating innovative capabilities. For example, during the 1970s and 1990s,
the global paper industry experienced serious down cycles that maintained prices at historically low
levels and severely impacted earnings (Lamberg et al., 2006); moreover, during the 1980s, Brazils
economy went through a combination of recession and uncontrolled hyperination. During the early
1990s, there was an abrupt change from the import substitution industrialization (ISI) regime in Brazil
to trade liberalization and an open economy, which swept many rms from the market. These events
had signicant negative effects on the competitive and economic performance of rms such as Klabin,
VCP, Suzano and Aracruz. In 2008, Aracruz experienced a deep nancial crisis involving losses of
US$2.1 billion as a consequence of rapid exchange rate movements following the international
nancial crisis and its hedging policies based on derivatives (Zeidan and Rodrigues, 2013), which put
Aracruz on the verge of bankruptcy despite its highly innovative capabilities. Indeed, because rms
operate in increasingly interconnected and changing environments, their performance is more
susceptible to external inuences (Zahra and Sapienza, 2006), including macro-economic conditions
(Lall, 1992; Arza, 2005), changes in institutional frameworks and rm decisions. Nevertheless, the
accumulation of innovative capabilities permits rms to mitigate the negative impacts of external
factors on competitiveness, and also permits rms to cross certain discontinuities in their
environments and overcome crises (Figueiredo, 2002). Therefore, the ndings show that accumulating
signicant levels of innovative capabilities enables rms to achieve not only technological catch-up
but also signicant outcomes related to innovative and business performance.
Fourth, outcomes such as operational and environment-related performance improvement might
have been achieved by rms acquiring new production systems embodying advanced technologies
by contracting with an array of external consultant designers, process engineers, and project
managers to dene and bring a new set of products, processes, and equipment-related technologies
into operational use on its behalf (see Bell and Figueiredo, 2012), or by delegating these duties to third
parties or even by using governments subsidies to provide inputs. Thus, these rms might have
achieved similar types of outcomes through different means and in different ways. Thus, it is
interesting to employ the notion of equinality (Zahra and Sapienza, 2006) in this context to interpret
these performance improvements. Conversely, it seems unlikely that these rms could have achieved
consistent and continuous improvement over a wide range of operational and environment-related
performance parameters across different business lines over time without accumulating a wide range
of levels and types of innovative capabilities. Thus, in the absence of signicant innovative capabilities,
it would most likely be that these rms would not keep achieving further levels of competitive
operational and environment-related performance in the face of erce competition against highly
innovative competitors in the global market. Additionally, even if the same performance were
achieved, the differences in the underlying capabilities do matter. Therefore, these ndings offer an
empirical substantiation to the position presented in Zahra and Sapienza (2006) about the means
97
(in this case, innovative capabilities) used by rms to achieve competitive performance. Therefore,
even if certain of those outcomes might have been achieved by other means, the accumulation of
innovative capability does matter.
Implications of these ndings for the related literature and theoretical contributions
Although these ndings are consistent with some segments of the literature, this study also moves
further in relation to previous studies and approaches in the literature on building innovative
capabilities in latecomer rms and the strategic management literature that is concerned with the role
of innovative capabilities as a fundamental source of competitive performance in several ways.
First, by capturing a wide variety of qualitative outcomes over time, and based on a comprehensive
framework for different capability levels, this study advances research addressing latecomer rms and
the impact of their current technological capabilities on specic performance indicators (e.g., Piccinini,
1993; Tremblay, 1994; Joo and Lee, 2010) and on studies that have examined the effects of innovative
capabilities (proxied in different ways, from R&D efforts to patenting) on rm performance (proxied
with different metrics) based on large data samples (e.g., Jonker et al., 2006; Goedhuys et al., 2008;
Bapuji et al., 2011; Shan and Jolly, 2012; Chen and Tsou, 2012). Therefore, this study adds important
nuanced qualitative evidence to improve the debate. Although corporate growth is explored here in an
incipient manner, it furthers recent studies that have addressed the impact of innovative capabilities
on corporate growth based on large data samples and perceptual evidence (Yang, 2012), different
types of growth patterns (that may be supported by a wide range of types and levels of innovation
capabilities, in addition to R&D) and most likely even production capabilities. Diversication into
related areas is thus consistent with Torres-Vargas (2006) and not with Amsden and Hikino (1994).
Certain of these growth patternsupstream diversication leading to new business lines or
involuntary spillovers led by individuals who create new businesses for the countrymay play an
important role in changing a countrys industrial structure, which is consistent with Nelson and Pack
(1999).
Second, the ndings are consistent with long-standing arguments that the ability of rms to
implement innovative activities and achieve distinctive performance reect the nature and depth of
their technological capabilities (e.g., Dosi, 1988; Lall, 1992; Bell and Pavitt, 1993) and with a tradition
of empirical research on the performance implications of innovative capabilities (e.g., Bell et al., 1982;
Katz, 1987) and that accumulation of innovative capabilities pays off in terms of achieving wide
performance benets (Figueiredo, 2002). The ndings related to the role of a wide range of innovative
capabilities in attaining these outcomes support Bell and Pavitt (1993, 1995), Patel and Pavitt (1994),
and Laestadius et al. (2005) with respect to the importance of non-R&D capabilities in achieving
innovative performance based on implementing innovative activities with differing degrees of novelty
and complexity (not always R&D-based) and with relevant positive operational economic effects for
rms (e.g., Enos, 1962; Hollander, 1965).
Third, this study provides a response to the call for empirical substantiation of the implications
of innovative capabilities for rm competitive performance (e.g., Helfat et al., 2007; Newbert, 2007;
Protogerou et al., 2011). By providing empirical substantiation for this relationship, this study may
move the debate forward and pave the way for further empirical analyses to generate a cumulative
body of evidence. However, this study suggests that the nature of these capabilities is much wider,
involving a wide range of levels and types related to diverse technological functions and
corresponding to different types and degrees of novelty relating to innovative activities. This
narrative contradicts the prevailing notion of innovative capabilities as merely reecting R&D and
patenting activities. This also calls for more comprehensive types of measurement. Therefore, these
ndings are consistent with Eisenhardt and Martin (2000) and Newbert (2007), but move a bit
further by empirically exploring some of the outcomes achieved by rms based on their
accumulated innovative capabilities. Thus, it might be that rms distinctiveness and capacity to
achieve long-term competitiveness lies beyond these innovative capabilities to involve a
combination of factors that may directly or indirectly inuence the achievement of rm innovative
and business performance, as highlighted in Fig. 1, although that broad perspective was beyond the
scope of this article.
98
99
Acknowledgements
Funding from Brazils National Research Council CNPq (grants 477731/2006-6 and 307404/2007-2)
and the Brazilian Pulp and Paper Association (Bracelpa) is gratefully acknowledged. Earlier drafts of this
article were presented at the 6th Globelics Conference (Mexico City 2008) and the 12th International
Schumpeter Society Conference (Rio de Janeiro, 2008). I am deeply grateful to the rms and their
professionals who participated in the eldwork for this study. Insightful and constructive comments,
criticisms and suggestions from two anonymous reviewers and the editor of JET-M are gratefully
acknowledged. I thank Saulo Gomes for his excellent research assistance. All disclaimers apply.
Appendix A. Data collection and analysis processes
Data collection
This study involved an intense triangulation of data sources and data collection techniques to
achieve robust internal validity and reliability (Jick, 1979; Eisenhardt, 1989; Yin, 2003). Data
collection involved several stages that identied the necessary data and its sources and decisions on
techniques to be used for collection. Because the study examines accumulated rm capabilities and
certain outcomes over time, particular efforts were made to collect data from past years. This
collection was undertaken by scrutinizing the rms technological milestones that were provided by
different interviewees (including retired staff), internal presentations and records, annual reports and
independent news reports. The extensive use of triangulation permitted the collection of evidence
from a range of sources to substantiate the study; interviews were used intensively. Creating the
interview protocol involved a breakdown of the research question and its constructs (innovative
capability and outcomes of innovation capability accumulation) until they were transformed into
plain words for the actual interviews. This operation was performed by consulting previous studies.
Each interview was conducted by two investigators on the basis of a structured but inductive
conversation, which encouraged interviewees to talk openly about the research themes (e.g., Tell us
about the main changes undertaken in your area over the past ve years . . . Who led them? Why?
How? Has your company achieved any benet from these changes? If yes, tell us about them).
Interviews were never recorded. After each interview, the notes were expanded and insights were
written down. At the end of each day, there were de-brieng sessions in which the investigators
discussed the responses, matched their interpretations and identied the emerging categories (parts
of each construct) and plans to reach interviewees for snow-balled interviews were made.
Analysis process
The analysis involved three iterative steps, which reected a tension between the studys objective
(deductive) and the categories and interpretations that emerged from the raw data (inductive). The
rst step occurred during the pilot and main stages. During eld interviews, some construct categories
began to emerge (e.g., outcomes, or types and levels of implemented innovations, operational
performance, growth, etc.) and were preliminarily labeled (coded) to facilitate their identication and
association in eld notes and de-brieng sessions. The second step occurred after the pilot and main
stages and sought to organize an overwhelmingly and messy amount of eld evidence (that was
collected from various sources and through diverse techniques) into a manageable amount of
evidence to be initially treated in formal analyses. This data cleaning involved separating and
organizing different pieces of evidence under relatively common blocks of observations, which were
then organized chronologically into within-case display tables, which permitted the grouping of data
and emerging categories and exploring relationships between them, and then in cross-case display
tables (Miles and Huberman, 1994). A close examination of these tables was used to code the
evidence, identify overlaps and reduce the number of categories.
For example, the initial 8590 categories of operational performance, which resulted from
innovation capabilities, were reduced to 4550 categories. Later, these categories were distilled into
100
101
Costa, L.A., Cool, K., Dierickx, I., 2013. The competitive implications of the development of unique resources. Strategic
Management Journal 34 (4) 445463.
Dalcomuni, S.M., 1997. Dynamic capabilities for cleaner production innovation: thecase of the market pulp export industry in
Brazil. SPRU/Sussex University, Brighton (D.Phil. thesis).
Dantas, E., Bell, M., 2009. Latecomer rms and the emergence and development of knowledge networks: the case of Petrobras in
Brazil. Research Policy 38 (5) 829844.
Dijk, M.V., Bell, M., 2007. Rapid growth with limited learning: industrial policy and Indonesias pulp and paper industry. Oxford
Development Studies 35 (2) 149169.
Dosi, G., 1988. Sources, procedures, and microeconomic effects of innovation. Journal of Economic Literature 26 (3) 11201171.
Dougherty, D., 2002. Grounded theory research methods. In: Baum, J.A.C. (Ed.), The Blackwell Companion to Organisations.
Blackwell, Padstow, UK, pp. 849866.
Dutrenit, G., 2000. Learning and Knowledge Management in the Firm. From Knowledge Accumulation to Strategic Capabilities.
Edward Elgar, Cheltenham, UK.
Eisenhardt, K.M., 1989. Building theories from case study research. Academy of Management Review 14 (4) 532550.
Eisenhardt, K.M., Martin, J.A., 2000. Dynamic capabilities: what are they? Strategic Management Journal 21, 11051121.
Enos, J.L., 1962. Invention and innovation in the rening industry. In: Nelson, R.R. (Ed.), The Rate and Direction of Inventive
Cctivity: Economic and Social Factors. Princeton University Press, Princeton.
Evans, J., Turnbull, J.W., 2004. Plantation Forestry in the Tropics, 3rd ed. Oxford, Oxford University Press.
Fan, P., 2006. Catching up through developing innovation capability: evidence from Chinas telecom equipment industry.
Technovation 26, 359368.
Figueiredo, P.N., 2002. Does technological learning pay off? Implications for inter rm differences in operational performance
improvement. Research Policy 31 (1) 7394.
Figueiredo, P.N., 2010. Discontinuous innovation capability accumulation in latecomer natural resource-processing rms.
Technological Forecasting and Social Change 77 (7) 10901108.
Giuliani, E., Bell, M., 2005. The micro-determinants of meso-level learning and innovation: evidence from a Chilean wine cluster.
Research Policy 34, 4768.
Goedhuys, M., Janz, N., Mohnen, P., 2008. What drives productivity in Tanzania manufacturing rms. European Journal of
Development Research 20 (2) 199208.
Grattapaglia, D., 2004. Integrating genomics into eucalyptus breeding. Genetic and Molecular Research 30 (3) 356368.
Grattapaglia, D., Kirst, M., 2008. Eucalyptus applied genomics: from genes sequences to breeding tool. New Phytology 179 (4)
911929.
Grattapaglia, D. (Ed.), 2011. IUFRO Tree Biotechnology Conference 2011: From Genomes to Integration and Delivery. BMC
Proceedings. Bahia, Brazil, 26 June2 July.
Hagedoorn, J., Cloodt, M., 2003. Measuring innovative performance: is there an advantage in using multiple indicators?
Research Policy 32 (8) 13651379.
Helfat, C.E., Finkelstein, S., Mitchell, W., Peteraf, M.A., Singh, H., Teece, D.J., Winter, S.G., 2007. Dynamic Capabilities:
Understanding Strategic Change in Organization. Blackwell, Malden.
Hitt, M.A., Ireland, R.D., Lee, H., 2000. Technological learning, knowledge management, rm growth and performance: an
introductory essay. Journal of Engineering Technology and Management 17, 231246.
Hobday, M., Rush, H., Bessant, J., 2004. Approaching the innovation frontier in Korea: the transition phase to leadership.
Research Policy 33 (10) 14331457.
Hollander, S., 1965. The Sources of Increased Efciency: A Study of Du Pont Rayon Plants. MIT Press, Cambridge, MA.
Iammarino, S., Padilla, R., von Tunzelmann, N., 2008. Technological capabilities and globallocal interactions: the electronics
industry in two Mexican regions. World Development 36 (10) 19802003.
Jick, T.D., 1979. Mixing qualitative and quantitative methods: triangulation in action. Administrative Science Quarterly 24 (4)
602611.
Jonker, M., Romijn, H., Szirman, A., 2006. Technological effort, technological capabilities and economic performance: a case
study of the paper manufacturing sector in West Java. Technovation 26, 121134.
Joo, S.-H., Lee, K., 2010. Samsungs catch-up with Sony: analysis with patent data. Journal of the Asia Pacic Economy 15 (3)
271287.
Jung, M., Lee, K., 2010. Sectoral systems of innovation and productivity catch-up:determinants of the productivity gap between
Korean and Japanese rms. Industrial and Corporate Change 19 (4) 10371069.
Kale, D., Little, S., 2007. From imitation to innovation: the evolution of R&D capabilities and learning processes in the Indian
pharmaceutical industry. Technology Analysis and Strategic Management 19 (5) 589609.
Katz, J., 1987. Technology Generation in Latin American Manufacturing Industries. Macmillan, London.
Kim, L., 1997. Imitation to Innovation: The Dynamics of Koreas Technological Learning. Harvard Business School Press, Boston,
MA.
Laestadius, S., 1998. The relevance of science and technology indicators: the case of pulp and paper. Research Policy 27, 385395.
Laestadius, S., Pedersen, T.E., Sandven, T., 2005. Towards a new understanding of innovativeness and of innovation based
indicators. Journal for Perspectives on Economic, Political and Social Integration, Special Issue XI (12) 75121.
Lall, S., 1987. Learning to Industrialize: The Acquisition of Technological Capability by India. Macmillan, London.
Lall, S., 1992. Technological capabilities and industrialisation. World Development 20 (2) 165186.
Lamberg, J.-A., Nasi, J., Ojala, J., Sajasalo, P. (Eds.), 2006. The evolution of competitive strategies in global forestry industries.
Comparative perspectives. Springer, Dordrecht.
Lee, K., Lim, C., 2001. Technological regimes, catching-up and leapfrogging: ndings from the Korean industries. Research Policy
30 (3) 459483.
Leornard-Barton, D., 1995. Wellsprings of Knowledge: Building and Sustaining the Sources of Innovation. Harvard Business
School Press, Boston, MA.
Mathews, J.A., 2002. Competitive advantages of the latecomer rm: a resource based account of industrial catch-up strategies.
Asia Pacic Journal of Management 19 (4) 467488.
102
Miles, M.B., Huberman, A.M., 1994. Qualitative Data Analysis: An Expanded Sourcebook, 2nd ed. Sage, Thousand Oaks.
Nelson, R., Pack, H., 1999. The Asia miracle and modern growth theory. The Economic Journal 109 (457) 457478.
Newbert, S.L., 2007. Empirical research on the resource-based view of the rm: an assessment and sggestions for future
research. Strategic Management Journal 28, 121146.
OECD, 2005. Oslo Manual. European Commission/Eurostat, Paris.
Patel, P., Pavitt, K., 1994. The continuing, widespread (and neglected) importance of improvements in mechanical technologies.
Research Policy 23, 533545.
Patton, M., 2002. Qualitative Evaluation and Research Methods, 3rd ed. Sage, London.
Pavitt, K., 1984. Sectorial patterns of technical change: towards a taxonomy and a theory. Research Policy 13 (6) 343373.
Pavitt, K., 1991. Key characteristics of the large innovating rm. British Journal of Management 2 (1) 4150.
Peerally, J.A., Cantwell, J., 2012. Changes in trade policies and the heterogeneity of domestic and multinational rms strategic
response: the effects on rm-level capabilities. World Development 40 (3) 469485.
Penrose, E.T., 1959. The Theory of the Growth of the Firm. Basil Blackwell, Oxford.
Peteraf, M.A., 1993. The cornerstones of competitive advantage. Strategic Management Journal 14 (3) 179191.
Pettigrew, A.M., 1990. Longitudinal eld research on change: theory and practice. Organization Science 1 (3) 267292.
Piccinini, M., 1993. Technical change and energy efciency: a case study in the iron and steel industry in Brazil. SPRU, University
of Sussex (D.Phil. thesis).
Protogerou, A., Caloghirou, Y., Lioukas, S., 2011. Dynamic capabilities and their indirect impact on rm performance. Industrial
and Corporate Change 21 (3) 615647.
Romijn, H., 1999. Acquisition of Technological Capability in Small Firms in Developing Countries. Macmillan, London.
Scott-Kemmis, D., 1988. Learning and the accumulation of technological capacity in Brazilian pulp and paper rms. Working
Paper, 187. World Employment Programme Research (222).
Shan, J., Jolly, D., 2012. Accumulation of technological innovation capability and competitive performance: a quantitive study in
Chinese electronic information industry. International Journal of Innovation and Technology Management 9 (5) .
Sher, P.J., Yang, P.Y., 2005. The effects of innovative capabilities and R&D clustering on rm performance: the evidence of
Taiwans semiconductor industry. Technovation 25, 3343.
Siggelkow, N., 2007. Persuasion with case studies. Academy of Management Journal 5 (1) 2024.
Silvestre, B.S., Dalcol, P.R.T., 2009. Geographical proximity and innovation: evidences from the Campos Basin oil & gas industrial
agglomeration Brasil. Technovation 29, 546561.
Sohn, E., Chang, S.Y., Song, J., 2009. Technological catching-up and latecomer strategy: a case study of the Asian shipbuilding
industry. Seoul Journal of Business 15 (2) 2657.
Teece, D., 2007a. Explicating dynamic cabalities: the nature and microfoundation of (sustainable) enterprise performance.
Strategic Management Journal 28, 13192135.
Teece, D., 2007b. The role of managers, entrepreneurs and the literati in enterprise performance and economic growth.
International Journal of Technological Learning, Innovation and Development 1 (1) 4364.
Torres-Vargas, A., 2006. Product-makert and technological direction in the evolution of large business groups in Mexico. In:
DRUID Summer Conference, Copenhagen.
Tremblay, P.J., 1994. Comparative analysis of technological capability and productivity grotwh in the pulp and paper industry in
industrialised countries. SPRU, Sussex University (D.Phil. thesis).
Yang, J., 2012. Innovation capability and corporate growth: an empirical investigation in China. Journal of Engineering and
Technology Management 29, 3446.
Yin, R., 2003. Case Study Research: Design and Methods. Sage, London.
Yoruk, E., 2011. The inuence of technological capabilities on the knowledge network component of innovation systems:
evidence from advanced materials in Turkey. International Journal Technological Learning, Innovation and Development 4
(4) 330362.
World Resources Institute, 1999. Aracruz Celulose S. A. And Riocell S. A: Efciency and Sustainability on Brazilian Pulp
Plantations. The Macarthur Foundation, Chicago.
Zahra, S.A., Sapienza, H.J., 2006. Entrepreneurship and dynamic capabilities: a review, model and research agenda. Journal of
Management Studies 43 (4) 917955.
Zeidan, R., Rodrigues, B., 2013. The failure of risk management for nonnancial companies in the context of the nancial crisis:
lessons from Aracruz Celulose and hedging with derivatives. Applied Financial Economics 23 (3) 241250.
Zhang, W., Shi, Y., Wu, X., Teng, Y., 2008. Enterpreneurship and innovation organizations, Institutions, systems and regions. In:
DRUID Conference, Copenhagen.
Paulo N. Figueiredo is a full professor of the Brazilian School of Public and Business Administration (EBAPE), Getulio Vargas
Foundation (FGV) in Brazil, where he undertakes teaching and research activities in technological learning and industrial innovation
in emerging economies. He holds a PhD in technology and innovation management from SPRU Science and Technology Policy
Research, University of Sussex, UK. In 1999 he set up and still heads up the Research Program on Technological Learning and
Industrial Innovation Management in Brazil at EBAPE/FGV. Senior Research Associate at the University of Oxford, UK. He is the
founder and editor-in-chief of the Int. J. of Technological Learning, Innovation and Development (IJTLID). His research papers appear in
academic journals including Research Policy; Industrial and Corporate Change; Technovation; International Journal of Technology
Management; Oxford Development Studies; International Journal of Technology and Globalization; Science, Technology and Society; Journal
of Commercial Biotechnology; World Development; Asian Journal of Technology Innovation; Journal of Evolutionary Economics and the
Journal of Management Studies.