Research Policy 53 (2024) 104949

Contents lists available at ScienceDirect

Research Policy
journal homepage: [Link]/locate/respol

Focusing the ecosystem lens on innovation studies

Carliss Y. Baldwin a, Marcel L.A.M. Bogers b, c, d, Rahul Kapoor e, Joel West f, g, *
a
Harvard Business School, Boston, MA, United States of America
b
Eindhoven University of Technology, Eindhoven, Netherlands
c
University of Copenhagen, Frederiksberg, Denmark
d
University of California, Berkeley, Berkeley, CA, United States of America
e
The Wharton School, University of Pennsylvania, Philadelphia, PA, United States of America
f
San José State University, San José, CA, United States of America
g
Hildegard College, Costa Mesa, CA, United States of America

A R T I C L E I N F O A B S T R A C T

Keywords: For nearly a century, the key role of innovation in economic growth has been acknowledged and studied. Today,
Ecosystems innovations are increasingly understood as being embedded in ecosystems of autonomous actors, whether firms,
Platforms other organizations, or individuals. These actors contribute in complementary ways to create a value proposition
Value creation
that is greater than the sum of the parts, with the integration of their products and processes made possible by
Value capture
modular interfaces between actors. Here we review the emergence of the ecosystem lens within innovation
Complementarity
Ecosystem governance studies in the context of the Special Issue on Innovation Ecosystems and Ecosystem Innovation. After summa­
rizing the history of the special issue, we review the nine articles in the special issue and show how they relate to
defining the actors, joint value creation by the actors, coordinating the actors, value capture by the actors, and
then the large issue of analyzing ecosystems as the unit of analysis. From this, we offer suggestions for future
ecosystem research, including opportunities to combine the ecosystem lens with other lenses used in innovation
studies, and new methods for studying ecosystem phenomena.

1. Introduction and policymakers now increasingly view innovations as embedded in

ecosystems made up of autonomous actors, including individuals, firms
The role of innovation as a key driver of firm performance and and other organizations such as universities and public agencies.
economic growth, as laid out by Schumpeter (1934, 1942) almost a Members of an ecosystem create products and systems whose value is
century ago, is now well established among practitioners, policymakers, greater than the sum of their separate parts. Each member then captures
and scholars. Innovations may involve new products, processes, ser­ a part of the resulting “complementary surplus” — the difference be­
vices, or technologies. Scholars have made major strides in uncovering tween the joint value created by all and the sum of the values they could
different facets of this phenomenon, including upstream generation of create separately. (Ecosystems in the social sciences are made up of
focal inventions, downstream commercialization of inventions into in­ individual actors and various kinds of organizations. They are distinct
novations, the resulting value creation, and, finally, value capture by from “natural” ecosystems made up of organisms and species.)
inventors, innovators, and others. But even as scholars were deciphering In innovation studies, the term “ecosystem” was initially used as a
implications of Schumpeterian creative destruction for firms and in­ metaphor to highlight the fact that firms might cooperate as well as
dustries (e.g., Teece, 2006; Fagerberg and Verspagen, 2009; Cohen, compete — these are often referred to as “business ecosystems” (Moore,
2010; Martin, 2012), the locus of innovation was gradually shifting from 1996; Iansiti and Levien, 2004). In management and economics, the
single firms within traditional industries to groups of firms and in­ term is now widely used to denote a network of autonomous economic
dividuals offering complementary goods and services, spanning multiple actors interacting to create value, including a complementary surplus,
industries (Moore, 1996; Baldwin and Clark, 2000; Gawer and Cusu­ which is distributed across actors. This is the definition we use in this
mano, 2002; Adner, 2006; Kapoor, 2018). article.
As a result of this shift in innovative activity, scholars, practitioners, However, as with any emerging point of view, there is still confusion

* Corresponding author.
E-mail address: [Link]@[Link] (J. West).

[Link]
Received 18 December 2023; Accepted 21 December 2023
Available online 15 January 2024
0048-7333/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ([Link]
nc-nd/4.0/).
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

about what exactly the ecosystem perspective on innovation entails, and Separately, Andy Grove (1996) argued that, in the late 1980s and
how it differs from previous approaches. In this article, we explain how early 1990s, the computer industry shifted from being dominated by a
the ecosystem perspective has provided a new lens within the field of set of vertically integrated companies to a “horizontal” structure where
innovation studies and define its key tenets. We then use the ecosystem different companies created components, which were then combined by
lens to identify common themes found in the nine articles published in system integrators and users into complete systems. Baldwin and Clark
the special issue on “Innovation Ecosystems and Ecosystem Innovation.” (1997, 2000) studied the technical underpinnings of computer systems
We end by discussing opportunities for further research. and argued that, in a complex system, “modularity” would permit design
and production to be divided among many different firms, coordinated
2. Incorporating an ecosystem lens within innovation studies by “design rules.” Modularity was a key property of technologies,
products and organizations, but firms might “know more than they
2.1. A brief history of innovation studies make” (Brusoni and Prencipe, 2001; Brusoni et al., 2001; Kapoor and
Adner, 2012).
Innovation studies originated in the field of industrial organization The three research streams — ecosystems, platforms, and modularity
(IO) beginning in the late 1950s and 1960s. Spurred by an increasing — proceeded separately for more than a decade until practitioners
appreciation of the importance of innovation to economic progress and converged on the concept of platform ecosystems (Kretschmer et al.,
social welfare, IO economists initially focused on the relationship be­ 2022) and modularity was recognized as a precondition for the creation
tween industry characteristics and firm size and the rate and direction of of both platforms and ecosystems (Jacobides et al., 2018). Implicit
technological change (Kamien and Schwartz, 1982; Cohen, 2010). In­ within these research streams is the notion that ecosystems can arise
dustry characteristics included attributes such as market structure, de­ through an evolutionary process of disaggregation from integrated to
mand, technological opportunity, and appropriability conditions. more modular architectures or a process of aggregation in which new or
Scholars in this tradition explored firms’ incentives to invest in R&D and existing components are linked in new ways for joint-value creation
the advantage that might accrue to large firms, a core thesis advanced by (Kapoor, 2018).
Schumpeter (1942) (e.g., Schmookler, 1966; Scherer, 1986). Responding to the new emphasis on distributed organizations and
In the 1970s and 1980s, a new evolutionary economics paradigm complementarities, advice to corporate managers on how to manage
was put forward as an alternative to neoclassical economics. Introduced innovation shifted 180◦ in the early 2000s. Where once companies were
by Nelson and Winter (1982), with important contributions by Freeman advised to protect their intellectual property via close control, patents,
(1974), Dosi (1982), Rosenberg (1982), Pavitt (1999), and others, and secrecy, managers were now encouraged to draw on external
evolutionary economics combined Schumpeter’s view of creative sources of knowledge including suppliers, complementors, users, start-
destruction with the behavioral theory of the firm (Simon, 1947, 1962, ups, universities, and public research institutions. At the same time,
1969; Cyert and March, 1963). This theory emphasized organizational they could benefit by opening their own organizational boundaries and
search and routines, under the assumption that individuals have limited sharing their knowledge. As such, “open innovation” emerged as a new
cognitive capacity and thus exhibit “bounded rationality.” Scholars paradigm for innovation management (Chesbrough, 2003; Chesbrough
adopting the evolutionary paradigm focused on organizational-level et al., 2006; Laursen and Salter, 2006; West et al., 2014; Chesbrough and
processes related to learning and the accumulation of capabilities Bogers, 2014; Dahlander et al., 2021).
within innovating firms, as well as industry-level dynamics related to
firm entry and exit (Utterback and Abernathy, 1975; Abernathy and 2.2. The ecosystem lens in innovation studies
Utterback, 1978; Tushman and Anderson, 1986; Anderson and Tush­
man, 1990, 2001; Henderson and Clark, 1990; Klepper, 1996, 1997; As this historical overview of innovation studies suggests, until
Christensen, 2013). The primary goals were to understand how firms recently, the Schumpeterian phenomenon of creative destruction has
and industries evolve, as well as the implications of evolutionary change been studied through an industry-level lens, a firm-level lens, and an
for innovation management. innovation-level lens. Applying an ecosystem lens changes the focus of
Innovation research in the evolutionary tradition went through a key inquiry from the origins and diffusion of a specific innovation to the
transition in the late 1980s with Teece’s (1986, 2006) account of the “value proposition” seen by users. In particular, it recognizes that
critical complementarities that often influence the adoption and concerted action by a group of actors may be needed for an innovation to
commercialization of innovations. Teece’s core question was: under create value in the eyes of the users. For this reason, in contrast to
what conditions do innovating firms capture value from their in­ scholars applying other lenses, ecosystem researchers are much more
novations? The answer, Teece argued, depended on whether other ac­ explicit about the demand-side of innovation.
tors controlled access to specialized complementary assets (e.g., The assets and knowledge needed to realize a given value proposition
distribution, manufacturing, marketing, sales, etc.) that were needed to often reside in different sectors, both public and private, and multiple
place the innovation in the hands of users. Any actor controlling an industries. For example, complementors often play an important role in
essential and unique complementary asset would be able to claim a ecosystems by creating compatible products and services that can be
share of the surplus value created by an innovation. In contrast, non- bundled with the focal innovation at the point of use (Adner and Kapoor,
essential or non-unique complements could be obtained by the inno­ 2010, 2016; Jacobides et al., 2018; Teece, 2018).
vator (or by users) at market prices (Baldwin, 2024, Ch. 4). Because innovation activities are distributed across many different
In the 1990s and early 2000s, three new streams emerged in the actors, modularity is an important underlying enabler for the func­
innovation literature. First, Moore (1996) and Iansiti and Levien (2004) tioning of ecosystems. Accordingly, the ecosystem lens focuses not only
drew an analogy between natural ecosystems and cooperative and on complementarities in systems of use, but also on the structure of
competitive relations between firms, especially in the computer in­ interdependencies that exist between technical components of the sys­
dustry. They were followed by Adner (2006) and Adner and Kapoor tem (Parnas, 1972, 1972b, 2001; Baldwin, 2024, Ch. 3). These in­
(2010) who considered innovating firms as dependent on an ecosystem terdependencies are caused by the technical architecture in which a
of upstream suppliers and downstream complementors for value crea­ given product or service is embedded (i.e., the linkages between
tion. In a separate line of work focused on the computer industry, different modules and subsystems), as well as the production architec­
Bresnahan and Greenstein (1999) and Gawer and Cusumano (2002) ture of input-output flows (Ganco et al., 2020). For example, innovation
described how platform owners like Intel and Microsoft could use con­ ecosystems, such as the one surrounding the iPhone, create value via the
trol over standards to recruit complementors who would innovate and technical and production architectures that unite the platform owner
supply compatible hardware and software to users. (Apple) to upstream suppliers (component makers), downstream

2
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

complementors (application software developers), and iPhone users. simply the sum of their separate values.
Modular architectures enable actors with very few organizational If a group of actors satisfies all three conditions, they qualify as an
ties to take complementary actions, but they can also be subject to ecosystem within this framework. Among all ecosystems, it is also useful
performance bottlenecks at different locations (Ethiraj, 2007; Adner and to distinguish between platform ecosystems and non-platform ecosys­
Kapoor, 2010; Baldwin, 2018; Hannah and Eisenhardt, 2018; Kapoor, tems. Platform ecosystems are coordinated by one or more central hubs
2018). For this reason, the interdependencies in a modular system must (platforms). Non-platform ecosystems use other means of coordination,
be explicitly recognized and managed through different forms of including bilateral transactions and contracts, multilateral agreements
governance, e.g., via standards, restrictions on access, or negotiation as arranged by “orchestrators,” and temporary linkages organized by
well as transactions and contracts (Staudenmayer et al., 2005; Kapoor “system integrators” (Kretschmer et al., 2022; Jacobides et al., 2024;
and Lee, 2013). In an ecosystem, governance choices are less likely to be Baldwin, 2024, Ch. 3).
bilateral involving two actors and more likely to be multilateral, In an innovation ecosystem, joint value is created through in­
involving a group of actors and facilitated by a focal orchestrator novations whose components are (by definition) both complementary
(Adner, 2017; Uzunca et al., 2022). and modular. The innovative products and/or processes do not have to
be sold in a market for a price, but users must be willing to expend re­
2.3. A framework for ecosystem research sources (money or labor) to obtain them (von Hippel, 1988, 2005, 2019;
Baldwin and Clark, 2006; Baldwin and Von Hippel, 2011). Like all
The ecosystem lens often provides a way of understanding existing ecosystems, innovation ecosystems can be coordinated via platforms,
phenomena (cf. Wurth et al., 2022). At the same time, it has revealed transactions, contracts, multilateral agreements, systems integration, or
new phenomena, including new sources of innovation, new patterns of some combinations of these mechanisms.
interaction among innovators and users, and new approaches to value Closely related but not identical to innovation ecosystems are
creation. The complexity of these findings calls for a framework that can entrepreneurial (or regional) ecosystems and knowledge ecosystems,
be used to structure research and identify related streams of work. A which also focus on creating something new (Table 1). These ecosystems
robust framework should permit researchers to identify organizational create value in the aggregate, while innovation ecosystems create value
forms that are ecosystems as well as those that are not. Furthermore, for users that is delivered through a series of new products or processes.
within the set of organizations that qualify as ecosystems, the framework Entrepreneurial ecosystems create value through productive entrepre­
should facilitate sensible divisions into different research streams. New neurship (Nicotra et al., 2018), while knowledge ecosystems create and
research projects can then be placed in relation to prior work, enabling disseminate new ideas that do not necessarily have commercial value or
cumulative progress in the field. value-in-use (Cobben et al., 2022).
We propose using the following criteria to identify ecosystems in the In many cases, an innovation, entrepreneurial, or knowledge
economy: ecosystem may also have an explicit regional or place-based focus, and
they may, in practice, be overlapping as well. For example, well-known
1. Autonomy: The actors in the ecosystem are autonomous organiza­ regional ecosystems such as Silicon Valley in the US or Brainport
tions and individuals. As such, they are subject to distributed Eindhoven in the Netherlands demonstrate attributes of all three
governance and value capture. ecosystem types. Similarly, an innovation ecosystem may overlap with a
2. Complementarity: The actors contribute in complementary ways to knowledge ecosystem (Miric and Jeppesen, 2023).
a focal value proposition. The joint value created by the whole sys­ Just as natural ecosystems evolve, so do socio-economic ecosystems.
tem is greater than the sum of the values of the separate parts. Over time, in a healthy ecosystem, actors, activities, and architectures
3. Modularity: The products and processes in the ecosystem are will change, as preferences shift and innovations make new value
modules within a larger technical architecture. propositions possible (Kapoor, 2018). Using a metaphor from physics, at
the most basic level, evolution within business ecosystems is shaped by
The dimensions of our framework have been chosen carefully, so that centripetal forces that push economic activities toward integration and
an organization can be separately classified on each dimension. First, centrifugal forces that pull economic activities into separate organiza­
autonomous actors, by definition, have separate decision rights. Also, in tions or out onto the market (Holgersson et al., 2022; Baldwin, 2024, Ch.
a free market economy, every autonomous actor must remain solvent or 3). For example, strong technological complementarities caused by
be reorganized. This in turn implies that each actor (individual or or­ process flow synchronization, non-contractible effort, and/or co-
ganization) must capture enough value to pay for the costs they incur. specialized assets imply centripetal forces that call for unified gover­
However, in general, autonomous actors do not have to contribute nance, the use of direct authority, and managerial hierarchies. In
complementary inputs to a focal value proposition and they do not contrast, product modularization, distributed knowledge, and network
necessarily operate within a modular technical architecture. effects create centrifugal forces that reward autonomous organizations
Complementarity means that the joint value of all the members’ with diverse policies and structures, capable of independent search and
contributions is greater than the value single members (or subsets) can experimentation.
attain separately (Milgrom and Roberts, 1995). Modularity means that In summary, in the social sciences, we define an ecosystem as a set of
the actions creating these complementary contributions are not so interacting autonomous organizations and individuals united around a
tightly connected that withdrawal of one will destroy the value of the focal value proposition. Members of the ecosystem each capture enough
whole (Baldwin and Clark, 2000). of the value created by the group to keep them involved. The value
Complementarity without modularity arises when two or more created by the combination minus the value created by the separate
agents are linked by strong relational ties, thus forming a “thick crossing parts is the complementary surplus of the ecosystem. In an innovation
point” in the underlying task network (Baldwin, 2008). Examples of ecosystem, value is created by innovations — new products and services
“complementary but not modular” relationships can be found in Japa­ linked by an architecture that users (including user-innovators) are
nese supplier networks, such as the Toyota Production System (Womack willing to expend resources to acquire.
et al., 1990; Sako and Helper, 1998; Sako, 2004; Helper and Sako, 2010) Our framework and definition are broadly consistent with earlier
as well as in systems where attempts to modularize are incomplete or definitions found in the management and innovation literatures. These
premature (Staudenmayer et al., 2005; Colfer and Baldwin, 2016). include “the alignment structure of the multilateral set of partners that
Modularity without complementarity arises in conglomerate corpo­ need to interact in order for a focal value proposition to materialize”
rations, holding companies, and financial portfolios: each business unit (Adner, 2017: 40); “a set of actors that contribute to the focal offer’s user
or investment constitutes a separate module, but their joint value is value proposition” (Kapoor, 2018); “an interdependent network of self-

3
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

Table 1
Comparing ecosystem attributes.
Innovation ecosystem Platform ecosystem Entrepreneurial ecosystem Knowledge ecosystem

Source of value Focal innovation Focal platform Productive entrepreneurship Novel knowledge
creation
Typical actors Innovators, Suppliers, Platform owner(s), Entrepreneurs, Funders, Research Universities, research institutes, firms,
Complementors Complementors Organizations, Accelerators government agencies
Primary Interaction Technological and input- Technological and multi-sided Knowledge and resource flows Knowledge flows
output flows markets
Link with Innovation Innovation and Innovation by platform owners, Innovation clusters and regional University-industry knowledge flows
studies technological change complementors ecosystems

interested actors jointly creating value” (Bogers et al., 2019: 2); and “the 3. About the special issue
evolving set of actors, activities, and artifacts, and the institutions and
relations […] that are important for the innovative performance of an Since 2010, there has been an exponential rise in the number of
actor” (Granstrand and Holgersson, 2020: 1). However, our framework academic publications explicitly examining business-related ecosystems
provides an explicit way to determine whether a group of autonomous (see Fig. 1). Such research has included innovation ecosystems (Adner,
actors is or is not an ecosystem, and if so, what type it is. We believe this 2006, 2012; Adner and Kapoor, 2010), platform ecosystems (Parker
structure can reduce the ambiguity and confusion that currently exists in et al., 2017; Kretschmer et al., 2022), entrepreneurial (or regional)
the field. ecosystems (Acs et al., 2017; Stam and Spigel, 2018), and knowledge
An ecosystem lens explicitly looks at economic complementarities and ecosystems (Järvi et al., 2018; Olk and West, 2023).
technological interdependencies among different actors operating Released in early 2020, our call for papers for a conference and this
within a larger system. The system-level approach recognizes a more special issue was limited to innovation and innovative platform eco­
complex set of relationships between technical components and eco­ systems. We received a strong response, in two separate waves of sub­
nomic agents than the traditional product-firm-industry-market view missions. A total of 75 papers were submitted to the June 2020 special
found in classic industrial organization and innovation studies. As Adner issue conference, of which 33 were accepted for presentation; the con­
(2012) has argued, in complex evolving technical systems, “widening ference was structured to provide detailed feedback for each paper. Four
the lens” to focus on ecosystems offers a better view of the underlying months later, 60 papers were submitted for consideration in the special
phenomena and thus can serve as a foundation for future cumulative issue. Of these, nine papers completed a minimum of four rounds of
research. reviews and have been published.
Other research streams in the field of innovation studies have also Table 2 provides an overview of the contents of the special issue,
advanced a “systems” view, by considering the broader set of in­ including the authors and title of each paper, the empirical context (if
stitutions and actors contributing to innovation within a given region any), and the types of ecosystem considered in each article. Under the
(cf. Nelson, 1993; Freeman, 1995; Edquist, 1997), sector or nation publisher’s current rules, the articles in this and other Research Policy
(Pavitt, 1984; Mowery and Nelson, 1999; Malerba, 2002). However, the special issues are organized as a “virtual issue” but no longer appear
focus in these works has been on the processes of knowledge generation simultaneously in paper or on the website. Our article, rather than being
and learning that underpin innovation in a given regional or industrial first as in a physical special issue, was finalized and submitted last, after
context and the resulting implications for policy. In contrast, our focus all the other articles were in final form.
and that of the special issue is on value creation and capture related to As guest editors, we are excited about the published articles’ po­
the delivery of specific innovations to users — new products, processes, tential to have a significant impact on the field of innovation studies.
services, or technologies — and the implications of the success or failure However, we were struck by the number of submitted papers, which,
of different delivery mechanisms for innovation management and despite studying novel and interesting phenomena, could not convinc­
strategy. ingly articulate the contributions of their research to ecosystem research
and the broader field of innovation studies. This supported our original
conjecture that, because ecosystems are a comparatively new

Fig. 1. Ecosystem articles in Scopus-indexed journals, 1993–2022.

Source: A total of 12,794 Scopus-indexed articles from 1993 to 2022 that have a Subject Area containing “busi” and Title, Abstract or Keywords that contain
“ecosystem” or “ecosystems”.

4
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

Table 2 How do the actors jointly create value through innovation and other
Summary of special issue articles. means? (3) How are members coordinated and interdependencies
Authors Title Empirical Type of ecosystem managed? (4) Who captures value and how? and (5) What are the
context consequences of taking ecosystems as the unit of analysis? We discuss
Borner Another pathway to Smart home Innovation the themes in subsections below.
et al. complementarity: How products ecosystem, w/ both
(2023) users and platforms and 4.1. Who are the actors in an ecosystem?
intermediaries identify intermediaries,
and create new coordinated by
combinations in toolkits
All of the articles touched on the central issue of identifying the
innovation ecosystems different types of actors in an ecosystem, as well as the processes and
Cozzolino Ecosystem disruption Healthcare IT Innovation mechanisms that enabled them to contribute to the focal value propo­
and and regulatory startups ecosystem, not sition. For both Kuan and West (2023) and Song et al. (2024), academic
Geiger positioning: Entry platform-based
research played an important role in reducing entry barriers and getting
(2024) strategies of digital
health startup a new ecosystem off the ground. In the former case, the defense agency
orchestrators and DARPA adopted an explicit strategy of fostering the creation of a fabless
complementors semiconductor industry. In the second case, involving small satellites,
Jacobides Externalities and Theory Innovation
new entrants pursued commercial opportunities suggested by the ex­
et al. complementarities in platforms,
(2024) platforms and transaction
periments of an academic pioneer.
ecosystems: From platforms, all Both Reiter et al. (2024) and van Dyck et al. (2024) show how in­
structural solutions to business ecosystems cumbents created platforms precisely to attract new firms and users that
eendogenous failures (not only innovation would complement their core value propositions. Cozzolino and Geiger
ecosystems)
(2024) show how new entrants whose products complemented the of­
Kuan and Interfaces, modularity Fabless Multilateral
West and ecosystem semiconductors innovation ferings of large incumbents were able to partner with those incumbents
(2023) emergence: How ecosystem and industry groups. Pujadas et al. (2024) document how both in­
DARPA modularized orchestrated by cumbents and new firms exploited low entry barriers (but also low
the semiconductor DARPA
switching costs) as hotel booking transactions shifted from one-to-one
ecosystem
Miric and How does competition Free vs. paid Platform-based
reservations to web-enabled aggregators.
Jeppesen influence innovative iPhone apps innovation There is some disagreement in the literature about who should be
(2023) effort within a ecosystem considered “members” or “participants” in an ecosystem. The issue is
platform-based often framed as a distinction between “generic” vs. “specific” comple­
ecosystem? Contrasting
mentarity. Components with generic complementarity can be used
paid and unpaid
contributors outside the focal ecosystem: they are sometimes called “commercial off-
Pujadas The value and Online travel Complex ecosystem the-shelf” or “COTS” components. Components with specific (or special­
et al. structuring role of web purchase sites with multiple ized) complementarity work only within the focal ecosystem.
(2024) APIs in digital platforms Some authors have suggested that, because suppliers of “generic”
innovation ecosystems:
The case of the online
complements do not require explicit coordination and may be unaware
travel ecosystem that the ecosystem even exists, they need not be considered part of the
Reiter et al. Managing multi-tiered European Comparison of five ecosystem. Adner (2017), for example, recommends “start with a value
(2024) innovation ecosystems banking industry platform-based proposition and … identify the set of actors that need to interact in order
innovation
for the proposition to come about.” (Adner, 2017, p. 41, emphasis
ecosystems
Song et al. From early curiosity to Small satellites Multilateral added; see also Jacobides et al., 2018). However, as Song et al. show,
(2024) space wide web: The innovation providers of generic complements can sometimes play essential role in
emergence of the small ecosystem, ecosystem emergence — even if they may play no role in ecosystem
satellite innovation orchestrated by governance. Specifically, new ecosystems can be built piecemeal,
ecosystem “core actors”
making specialized
starting with mostly generic components and adding more specialized
investments components bit by bit.
van Dyck From product to Agricultural Comparison of two At the same time, Song et al. show that there is risk to the ecosystem
et al. platform: How equipment nascent platform- when key actors do not make specialized investments. In the small sat­
(2024) incumbents’ manufacturers based innovation
ellite ecosystem, the initial generic components fit poorly into the eco­
assumptions and ecosystems
choices shape their system’s central value proposition. The key to growth and success was
platform strategy attracting more committed members who invested in specialized compo­
nents and eventually supplanted the generic contributors.
A related question is: should a would-be disruptor be viewed as part
phenomenon in management research, many authors lack a strong of an ecosystem? In their study of six digital startups in the much larger
conceptual foundation on which to ground their research. Thus, while healthcare ecosystem, Cozzolino and Geiger (2024) show that ecosys­
the upsurge in interest in ecosystems is a noteworthy development, tems have ways of resisting disruption by new entrants. However, the
scholars studying ecosystems do not yet share a conceptual framework resistance can be overcome (1) if the entrant is supported by regulatory
that would allow them to position their findings relative to prior work authorities or (2) existing regulations do not apply to the new entrant.
and to explicate their own work’s novelty and importance. The same conclusion can be drawn from Kuan & West: without DARPA’s
consistent support and funding, the diffusion of Electronic Design
4. Themes in the special issue Automation (EDA) and the separation of chip design and fabrication
would have occurred much later, if at all.
In this section, we demonstrate the value of the ecosystem lens to the Finally, many definitions of ecosystems in the social sciences assume
field of innovation studies by identifying the key themes that multiple that the main actors are firms (Moore, 1993, 1996; Adner, 2006, 2012)
articles brought to the fore. The themes appearing in multiple articles or other organizations (Autio and Thomas, 2021). Kuan & West and
were: (1) Who are the actors in the ecosystem and why do they join? (2) Cozzolino & Geiger both show that government agencies — DARPA and

5
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

healthcare regulators respectively — can play a crucial role both in about the value proposition. Similarly, van Dyck et al. contrast the ways
forcing incumbents to change strategies and enabling entry opportu­ in which two established incumbents in the global agricultural equip­
nities for innovative startups. ment industry pursued different strategies for managing complementary
Only a handful of ecosystem definitions explicitly allow for indi­ innovations. Their divergent strategies resulted in very different results
vidual actors (Qiu et al., 2017; Bogers et al., 2019; Altman et al., 2023; for both the quantity and nature of the applications provided by third-
Baldwin, 2024, Ch. 3). However, the importance of individuals is party complementors.
demonstrated in two articles in the special issue. First, in the smart home In addition to showing how joint value is created, several articles in
ecosystem studied by Borner et al. (2023), the initial platform sponsors the special issue discuss the antecedents to value creation in ecosystems.
were manufacturers of domestic appliances (BHS) or lighting systems Cozzolino and Geiger (2024) point to the difficulty that startup firms
(Philips Signify). However, as users began to combine products from face in realigning an ecosystem to a new model of value creation, and
different platforms, a new layer of intermediaries emerged supplying how regulation can enable, impede or have no effect on such realign­
toolkits that made such combinations easy (Franke and von Hippel, ment. Kuan & West show how the DARPA-funded creation of a modular
2003). Users could then create custom adaptations and systems of use. architecture with open interfaces fueled entry and innovation by EDA
The intermediaries then tabulated the specific combinations users firms and “fabless” semiconductor firms.
created, disseminated this information to other users (and the plat­
forms), and counted how often each combination appeared. From the 4.3. How are ecosystem members coordinated?
data, it became apparent that “users are better able to identify which
combinations create value for them (and for others)” (Borner et al., p. Another commonly asked question is: how are ecosystem members
10). Today enthusiasts and hobbyists are among the most active in­ coordinated (or otherwise aligned) to achieve joint value creation?
novators in this ecosystem. Platform ecosystems have historically been governed by a single firm
Hobbyists are also integral to the iPhone’s “jailbreak” platform (Bresnahan and Greenstein, 1999; Jacobides et al., 2018; Baldwin, 2024,
marketplace, the focus of Miric and Jeppesen’s (2023) article. The Ch. 9) or a coalition of firms (O’Mahony and Karp, 2022). In these cases,
jailbreak market exists for the purpose of disseminating applications not the sponsor defines the value creation goals, the rules of interaction, and
allowed in Apple’s larger App Store. Its members are individual software (often) who is allowed to participate. In some cases, however, multiple
developers (not companies) who can elect to be paid or not paid for their platform sponsors may vie for market share. The articles by Reiter et al.
products. In this sample, unpaid developers outnumbered paid de­ and van Dyck et al. in this special issue describe such cases in the Eu­
velopers by around twenty-five to one (10,132 to 440). Unpaid de­ ropean banking and global agricultural equipment industries
velopers may (1) benefit from the intrinsic satisfaction of developing respectively.
and sharing their creations with like-minded people; (2) receive repu­ Jacobides et al. in the special issue distinguish between platform
tational benefits in the form of increased status within the community; ecosystems and “business” ecosystems without platforms. In the former,
or (3) obtain indirect, long-term financial benefits in the form of the platform sponsor(s) govern via standards (“design rules”) and/or
enhanced career opportunities. As in Borner et al., an important prices charged to different “sides.” The sponsor also determines the rules
contribution of this article is to demonstrate that motives unrelated to of access. In ecosystems without platforms, mutually agreed-upon
direct financial gain can be sources of value for individuals and thus collaborative arrangements bring members of the ecosystem together.
drivers in the evolution and growth of an ecosystem. Members may also agree to create shared assets and develop common
processes that allow them to achieve their joint goals (Jacobides et al.,
4.2. How is joint value created? 2024 Exhibit 3).
In related work, Baldwin (2024, Ch. 3) argues that, in the absence of
Several articles in the special issue advance our understanding of platforms, ecosystems can be coordinated via market prices, bilateral
how joint value is created in ecosystems. First, as discussed above, Song contracts, multilateral negotiations (often orchestrated), and systems
et al. highlight the tension between generic and specific components in integration. She further argues that the primary advantage of platforms
the ecosystem. Components with generic complementarities may be the relative to other means of coordination is that, by standardizing design
only viable option when the ecosystem is very new: specific components rules, platforms can scale efficiently. Thus, as the numbers of partici­
create more value for customers, but suppliers defer making specialized pants and activities in an ecosystem grow, platforms are likely to emerge
investments when there is limited demand. The authors document an in order to cost-effectively handle the increased rates of innovation and/
iterative process in which the initial ecosystem members standardized a or volume of trade (Baldwin, 2024, Ch. 7).
technical architecture and created enough value with generic compo­ For firms and industries where ecosystem-based value creation was
nents to attract more actors and more specialized investments. not previously viewed as modus operandi, coordinating and aligning
Pujadas et al. highlight the joint value created by providing, interests requires new competencies and even cultural norms. Reiter
accessing and recombining data via Web APIs in the online travel agency et al. show that, in the ecosystems sponsored by five different banks, the
(OTA) ecosystem. The network of OTAs and related service providers is sponsors used formal governance mechanisms to solve well-defined
both large and decentralized, with low barriers to entry and few domi­ problems, and informal mechanisms for more uncertain or ambiguous
nant hubs. The revenue from a single booking may be shared with ones. Similarly, in their study of rival ecosystem strategies of two agri­
multiple players whose Web APIs are deployed at some point in the end- cultural equipment manufacturers, van Dyck et al. show how these
to-end process. The study finds that, in this data-driven ecosystem, the strategies were constrained by existing partners, particularly dealers.
competitive advantage gained from controlling specific Web APIs is not Incentives and governance are more complicated in cases of over­
very long-lasting. lapping ecosystems. Horizontal overlaps arise when two or more plat­
Borner et al. find that successful smart home innovations were pre­ forms compete for complementors and users, a practice known as multi-
dominantly recombinations of existing products, coming from down­ homing (Eisenmann et al., 2011; Zhu and Iansiti, 2012; Cennamo and
stream homeowners rather than upstream producers. Miric & Jeppesen Santalo, 2013). In the special issue, Pujadas et al. show that, given very
highlight the importance of innovations by unpaid complementors in low specialization and switching costs, analyzing “overlapping” eco­
ecosystems and show how paid and unpaid actors responded differently systems is meaningless: instead, ecosystem members are all part of a
to competition in terms of nature and speed of their innovations. single indivisible, interwoven network.
Using contrasting case studies of five European banks, Reiter et al. Horizontal overlaps are also possible in non-competing ecosystems.
show how the approach by incumbent banks in Europe to orchestrated For example, a fabless semiconductor firm located in Santa Clara is part
innovation ecosystems differed depending on the degree of uncertainty of the global semiconductor innovation ecosystem and a member of the

6
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

Silicon Valley entrepreneurial ecosystem. It might contribute to ecosystems seen in the computer industry.
different knowledge ecosystems, for example, university-sponsored Finally, when considering issues such as governance, value creation,
research consortia focused on state-of-the-art technologies. It will be or incentives, prior research has tended to distinguish between different
the customer of a semiconductor foundry like TSMC (a platform) and a types of actors based on their structural position within the ecosystem —
supplier of downstream computer and device makers. Finally, its chips typically defined as focal firm, supplier or complementor (e.g., Adner
may be a platform in their own right (e.g., Nvidia’s graphic processing and Kapoor, 2010; Dattée et al., 2018). However, two articles in the
chips are a platform for AI applications). How firms and individuals special issue (Reiter et al., Song et al.) illustrate how differences in ac­
balance their contributions and potential goal conflicts in such situations tions and strategy in an ecosystem may not always be driven by struc­
has yet to be studied. tural position of the actors. For example, the incentives and
Hierarchical (or nested) overlaps arise in industry consortia, when contributions to value may be similar between suppliers and com­
firms or individuals join a specific consortium within a larger industry plementors in one customer segment, while dissimilar across a different
ecosystem (Olk and West, 2023) or when a consortium creates a hier­ customer segment. Future research might thus examine interdepen­
archy of projects and allocates governance rights to managers of both dence between actors using more fundamental definitions of comple­
projects and subprojects (O’Mahony and Karp, 2022; Baldwin, 2024, Ch. mentarity, such as by applying Milgrom and Roberts’ (1995)
16). In the special issue, Cozzolino & Geiger, Reiter et al., and van Dyck observation that the ultimate measure of complementarity is when
et al. all show how firms created new ecosystems by attracting members absence of the complementor reduces the value of the offering.
from a larger ecosystem to specific projects or platforms. Reiter et al.
also show how layered hierarchies (tiers of governance) within each 4.4. Who captures value in an ecosystem and how?
ecosystem allowed the platform sponsors to selectively engage their
complementors in the development of their respective shared value Two articles in the special issue examine factors that can limit firms’
propositions. and individuals’ ability to capture value. In their study of data-based
Recently, researchers have examined weaker forms of control often travel ecosystems, Pujadas et al. show that the constant threat of new
termed orchestration (Reypens et al., 2021; Altman et al., 2022; Olk and entry, imitation and disintermediation meant that value capture op­
West, 2023). Cozzolino & Geiger suggest that in a regulated industry, portunities typically dissipated rather quickly, forcing firms to
new entrants will face difficulty in orchestrating a new set of relation­ constantly update their strategies and reconfigure their ties. Meanwhile,
ships unless the proposed change is supported by existing regulatory by contrasting the platform ecosystem strategies of two agricultural
authorities or laws. equipment makers, van Dyck et al. show that the platform sponsor that
Meanwhile, at both a global and local level, well-defined interfaces shared value more generously with complementors attracted 7× as
provide thin crossing points to structure a value-creating collaboration many complements as its rival. This example demonstrates how “taxing”
between two or more firms (Baldwin, 2008; Baldwin, 2024, Ch. 2). an ecosystem can reduce value creation.
Research on interfaces has typically focused on those provided by a Van Dyck et al. notwithstanding, ecosystem research has rarely
platform sponsor within an overall platform architecture (West and documented or measured this inherent tension between value creation
Dedrick, 2000; Baldwin and Woodard, 2009; Gawer, 2021). However, in and value capture, which has long been observed in other streams of
the special issue, van Dyck et al. show how two agricultural equipment innovation research (see, for example, Simcoe, 2006; Laursen and Salter,
companies used different types of interfaces to manage their respective 2014). Future research might examine possible moderators of this
digital innovation platform ecosystems. One company saw its platform tradeoff, such as growth, complexity, and degree of control, the presence
as a venture separate from its existing product line: they created a of competing ecosystems, and the external appropriability regime
modular connectivity device that could link the platform to any piece of (Teece, 1986). Future research could also validate the prediction set
machinery, regardless of brand. They actively recruited users to their forth in Jacobides et al. (Table 2, Column 1) that centralized governance
digital platform who did not use their machinery. Finally, they mone­ by a platform sponsor can avoid or reduce the risk that misallocated (or
tized the platform by charging third-party app suppliers a 30 % uncertain) values will discourage participation in the ecosystem, thus
surcharge. reducing value created and possibly causing innovation failure.
The second company saw its digital platform as a way to enhance its
existing product line. They retrofitted their entire product line to be 4.5. Ecosystems as the unit of analysis
compatible with the digital platform and created an open API. They did
not charge complementors for access to the platform, but allowed them Researchers increasingly consider ecosystems to be a unit of analysis
to choose prices and keep all the revenue from their apps. Not surpris­ distinct from firms, industries, and markets. In this regard, it is useful to
ingly, they attracted many more complementors than the first company, distinguish between articles that look at the impact of ecosystems on
and, as a result, did not have to staff a large internal group of software different actors and those that look at the impact on ecosystems of ac­
engineers. Despite these differences, as of the time of publication, both tions taken by members and other agents. While causality in any
platform strategies appear viable. ecosystem runs both ways, it is still useful to view each direction
A key benefit of stable interfaces is support for decentralized inno­ separately.
vation. Normally, it is assumed that such interfaces will be established as Much prior research has examined the benefits of a successfully
design rules by a central platform sponsor. (Baldwin and Clark, 2000; functioning ecosystem for its members, including sponsors, orchestra­
Gawer and Cusumano, 2002; Baldwin and Woodard, 2009). Two articles tors, complementors and users (Moore, 1993, 1996; Adner, 2006, 2012;
in the special issue describe exceptions to this rule. First, Borner et al. Ceccagnoli et al., 2012; West and Wood, 2013; Cennamo and Santaló,
demonstrate that end users can recombine off-the-shelf products using 2019; Olk and West, 2020). But how does an ecosystem’s success impact
interfaces provided by intermediaries in the form of toolkits. Second, the rest of the economy? Song et al. show how the maturation of the
Pujadas et al. present a radically decentralized model of interface cre­ small satellite ecosystem — with the concomitant improvements in price
ation — the antithesis of centralized platform sponsorship. In the OTA and performance — enabled the creation of entirely new communica­
ecosystem, very low switching and programming costs allowed both tion businesses that were not economically viable given earlier, more
creators and users to generate and recombine a broad range of ever- integrated, and more expensive satellite designs. Pujadas et al. also show
evolving interfaces. The resulting network contains a large number of how new businesses, supplying both platforms and complementary
interconnected platforms, but no dominant hub (Pujadas et al., Fig. 9). It goods and services were built on top of new Web APIs.
is more like a classic transportation or electrical network with many The articles in the special issue pay relatively more attention to the
transfer points than the classic single or overlapping platform impact of actors on ecosystems. In particular, several articles address

7
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

two key questions: (1) where do ecosystems come from? and (2) why do 5.1.1. Business models
they change? A common finding in several articles was that ecosystems New technologies often enable both new business models (Baden-
may be created to jump-start the value creation strategy of a for-profit Fuller and Haefliger, 2013) and new ecosystems (West and Wood,
firm or a regulator. Thus, Kuan & West describe the history of a key 2013). Like ecosystem research, business model research focuses on
interface that enabled vertical specialization within the semiconductor demand-side value creation (Massa et al., 2017), but it is much more
industry. The interface was prototyped by government-funded academic explicit about choices with respect to value capture (Teece, 2018; Adner,
researchers who predicted that the modularization of chip designs 2017; Kapoor, 2018; Snihur et al., 2021). In the special issue, Cozzolino
would allow chip performance to continue to improve at the rate pre­ & Geiger, Kuan & West, and Song et al. all show that new entrants to an
dicted by Moore’s Law. With DARPA’s backing, a tripartite ecosystem ecosystem can capture enough value to severely disrupt incumbents’
made up of EDA toolmakers, fabless chip designers, and semiconductor business models, adding to the emerging theme of how new entrants
foundries arose and became competitive with established integrated disrupt incumbents’ business models within an ecosystem (see also
device manufacturers like Intel (Baldwin, 2024, Ch. 8). Ansari et al., 2016). However, more research is needed to illuminate the
Similarly, three of Cozzolino and Geiger’s six health care startups relationships between the two phenomena, focusing on the dynamics of
received assistance from a regulator or were founded in response to a value creation and capture among multiple actors in an ecosystem.
new regulatory initiative or law. And Song et al. describe how the An iconic example of these complex relationships is the emergence of
University of Surrey built and launched the first small satellite as a proof the personal computer industry. In the 1970s, the birth of the micro­
that the concept was viable. processor enabled a new ecosystem of microcomputers, peripherals and
Four other articles in the special issue reflect the more common accessories. This allowed Apple Computer and then IBM to create new
reason for an innovation ecosystem to emerge: one or more platform platform business models that depended on the ecosystem for comple­
sponsors sees the benefit of tapping external sources of innovation and mentary hardware and software (Freiberger and Swaine, 1984; Chposky
creates a platform with open (and stable) interfaces for that purpose. In and Leonsis, 1988; Cringely, 1992; Baldwin, 2024, Ch. 9). Subsequently,
two cases, Reiter et al. and van Dyck et al., the platform sponsors were when Compaq and Phoenix Technologies reverse engineered IBM PC
incumbents in traditional industries—banking and agricultural equip­ firmware, another business model emerged based on the control of
ment respectively. critical standards that ensured interoperability between the basic com­
In the two other cases, new platforms emerged that responded to the puter, hardware devices, and software applications (Froot, 1992; Morris
demands of individual users. First, as described above, in the smart and Ferguson, 1993; Intel Oral History Panel, 2008).
home ecosystem studied by Borner et al., when users began to combine In the 1990s, the worldwide dominance of the Wintel standards-
products from different platforms, a new layer of intermediaries based platform led to the formation of a “modular production
emerged supplying toolkits that made such combinations easy. Second, network” — a globally distributed ecosystem comprised of contract
in the iPhone jailbreak ecosystem, described by Miric & Jeppesen, when manufacturers, original design manufacturers (ODMs), and lead firms
Apple opened the App Store, most iPhone developers migrated from the (such as Compaq, NEC, Toshiba, and Dell) which controlled access to
initial jailbreak platform to the Store. However, there were some ap­ final customers (Sturgeon, 2002; Sturgeon and Kawakami, 2011; Thun
plications that Apple did not want to encourage, and would not allow on et al., 2022; Baldwin, 2024, Ch. 11). Each of these specialized roles was
its site. The initial jailbreak platform continues to serve this niche associated with a corresponding business model. The rise of the Internet
market. enabled new online transaction platforms that bypassed traditional re­
Our original call for articles suggested another way to study eco­ tailers. Such a pattern of technological innovation, business model
systems as the unit of analysis: it asked for research on ecosystem inno­ innovation and ecosystem adaptation is common, and deserves further
vation, defined as “new models of ecosystem creation and management.” analysis.
While each of the innovation ecosystems studied in this special issue had
unique story of emergence, in most cases, the eventual ecosystem con­ 5.1.2. Open innovation
formed to our framework based on autonomy, complementarity and The ecosystem lens helped extend open innovation from a dyadic to
modularity. However, the online travel ecosystem described by Pujadas multilateral value creation strategy (West, 2014). For its part, the open
et al. is arguably a new form of ecosystem: a radically decentralized, innovation paradigm has been used to explain the basis of value creation
highly interconnected network of firms and users. This form of organi­ in a range of ecosystems (Rohrbeck et al., 2009; Bogers et al., 2017;
zation can be found in other settings, e.g. international communication Randhawa et al., 2021), although recent evidence shows a complex
networks, electricity grids and airlines route networks. Meanwhile, the relationship between open innovation and financial performance
“massive modular” ecosystem that that designs and manufactures mo­ (Schäper et al., 2023). In the future, open innovation research would
bile communication devices reflects an admixture of the decentralized benefit by using the lens of collaborative value creation from ecosystem
model interspersed with a small number of firm-sponsored platform research, while ecosystem research could learn from open innovation’s
ecosystems (Thun et al., 2022). careful examination of the motives of the collaborating partners and the
implications for value capture. Finally, there is an opportunity to study
5. The future of ecosystem research the nexus of open innovation, business models, and ecosystems, given
the common focus on value creation and capture across organizational
We end this article by discussing two broad areas for future research. boundaries (Zott et al., 2011; Chesbrough and Bogers, 2014).
The first concerns convergent perspectives — questions at the inter­
section of innovation ecosystem research and other established lines of 5.1.3. Research consortia, standards-setting organizations, and open source
inquiry. The second suggests other methods that might fruitfully be used projects
to extend ecosystem research in new directions. Research consortia, standards-setting organizations (SSOs), and open
source projects provide parallel examples of how a coalition of firms (or
other organizations) can sponsor an innovation ecosystem (O’Mahony
5.1. Convergent perspectives in innovation studies and Karp, 2022; Olk and West, 2023). In most cases, this sponsorship is
provided by an incorporated (often nonprofit) organization, which has
In this section we suggest how the ecosystem lens might be combined formal rules for membership and governance. This organization can
with other lenses commonly used by innovation researchers to produce a hold title to common property and gain the advantages of asset sepa­
sharper picture of the underlying phenomena. We emphasize insights ration, limited liability, and indefinite life. At the same time, this
provided by the empirical articles. sponsoring organization is lodged within larger ecosystems whose

8
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

structure and membership are more fluid. ecosystem can enable, constrain or accelerate the development of the
Prior research has tended to focus on governance at one but not both other.
of these levels of organization. Future research could recognize the hi­
erarchical relation between the larger ecosystem and these special- 5.1.7. Knowledge ecosystems
purpose organizations, and examine their interaction. For example, Research on knowledge ecosystems has focused on how networks of
when do the norms of the larger ecosystem constrain the policies of the knowledge-intensive ecosystem participants facilitate knowledge crea­
central governing body? And when does the center affect the direction of tion and sharing, often within strategic alliances and research consortia
innovation of the larger ecosystem? These issues have been studied in (van der Borgh et al., 2012; Cobben et al., 2022; Clarysse et al., 2014). A
the context of open source software projects and communities, but not in key research opportunity at the nexus of knowledge and innovation
innovation ecosystems more generally. ecosystems lies in examining how the mechanisms used to foster trust
and cooperation in knowledge ecosystems influence the processes of
5.1.4. Intellectual property (IP) innovation and commercialization in innovation ecosystems. As with
The allocation of IP rights has been crucial for the effective operation entrepreneurial ecosystems, there are possibilities to study the impacts
of many ecosystems, whether open source platforms (West and Gal­ of overlapping memberships with innovation ecosystems, but also
lagher, 2006; O’Mahony and Karp, 2022) or those organized by R&D nesting, joint control and other stronger ties that sometimes link
consortia to guide academic-industry collaboration (Sydow et al., 2012; knowledge and innovation ecosystems (Leten et al., 2013; Olk and West,
Leten et al., 2013; Olk and West, 2020) or those coordinated by standard 2023).
setting organizations (Toh and Miller, 2017). Indeed, the attributes of
the network of collaborating partners have been identified as one of the 5.2. New insights from new methods
elements that determine the balance between IP sharing and protection
(Bogers, 2011). To better understand interdependencies and comple­ Virtually all papers submitted to and accepted for the special issue
mentarities, patent citations have been used to measure the structure of used qualitative, comparative case studies and/or longitudinal methods
knowledge flows in innovation ecosystems (Lee et al., 2016; Jones et al., to conduct their research. Exceptions in the special issue were Miric &
2021). Future research could study the impact of different types of IP Jeppesen, who use large-sample hypothesis testing and Pujadas et al.
regimes or strategies within an innovation ecosystem, whether endog­ who trace the evolution of the OTA ecosystem using network visuali­
enous (demanded or chosen by the members) or exogenous (imposed by zations. The application of qualtitative and historical methods allows for
legal or regulatory authorities). Future research could also explore how a deep exposition and analysis of the underlying phenomena.
firms in an ecosystem leverage their IP and navigate the tension between Although by definition innovation ecosystems create value through
value creation and value capture. new innovations, researchers rarely seek to measure the quality or
quantity of innovation generated. In contrast, there is a long tradition of
5.1.5. Organizational capabilities measuring both quality and quantity at the level of firms (e.g., Garcia
Previous research has emphasized the influence of capabilities upon and Calantone, 2002; Hagedoorn and Cloodt, 2003), technologies
ecosystem leadership, including how they can help universities lead an (Suarez, 2004), and national economies (Smith, 2006; Gault, 2018).
entrepreneurial ecosystem (Heaton et al., 2019), enable intermediaries Future research could similarly seek to contrast the quality or other
to shape service ecosystems (Randhawa et al., 2022), or allow platform aspects of the innovation with the overall value created by an ecosystem.
sponsors to encourage ecosystem members to be more innovative Such measures could be used to explain the sources of value within a
(Gawer and Phillips, 2013; Haki et al., 2022). However, most work on given ecosystem, or to compare different ecosystems.
both ordinary and dynamic capabilities assumes that these skills are Also, while value capture is central to the ecosystem concept, it is
cultivated and remain within companies or business units (Nelson and notoriously difficult to measure (Rietveld et al., 2019). Thus, studies of
Winter, 1982; Chandler, 1977; Teece et al., 1997; Dosi et al., 2000; value capture have tended to use qualitative measures, such as bifur­
Helfat, 2000; Teece, 2009; Helfat and Peteraf, 2015). cated success vs. failure (West and Wood, 2013) or the trend over time
Future research might consider the development of capabilities for — higher, lower, or the same (Schreieck et al., 2021).
firms in an ecosystem (Helfat and Raubitschek, 2018), the diffusion of Further, it is important not only to identify governance structures
complementary capabilities across firms in an ecosystem, and perhaps and mechanisms in ecosystems and trace their evolution, but also to
the transmission of capabilities from one ecosystem to another. For show which structures and mechanisms are common and which are rare.
example, in contrast to U.S. automakers, Toyota is known to actively This, in turn, requires systematic quantitative measurement across large
build capabilities within its suppliers (Sako and Helper, 1998; Sako, sets of actors and a broad range of industries. Quantitative approaches
2004). TSMC’s Open Innovation Platform provides a toolkit for can also be employed to study the nature and rate of innovation in
modeling combinations of designs and services provided by TSMC and ecosystems, and how they contribute to the overall economic progress
third-party companies in its ecosystem, and testing the combinations to and the specific performance of the different actors over time (e.g.,
see they will function properly. The toolkit avoids expensive rework of Adner and Kapoor, 2016; Kapoor and McGrath, 2014).
chip designs, lowering the customers’ time to market (Chesbrough, The most important impediment to such research is the lack of
2020). appropriate datasets. Ecosystems are, by definition, groups of actors,
linked in various formal and informal ways. Detailed data on interfirm
5.1.6. Entrepreneurial and regional ecosystems transactions is very limited: government input-output tables are too
While research on innovation ecosystems focuses on the value aggregated to be of much use in identifying transactions in ecosystems,
proposition to users and the interplay of technological in­ although social network tools might shed light on the nature and impact
terdependencies and economic complementarities among members of of interdependencies between individual or organizational actors.
the ecosystem, research on entrepreneurial ecosystems generally em­ Another promising methodology is to look at software traces, such as
phasizes the conditions within a given geographic region that support web APIs. These are generally susceptible to large-sample hypothesis
the emergence and growth of new ventures and entrepreneurial activ­ testing (Kapoor and Agarwal, 2017; Agarwal and Kapoor, 2022) as well
ities (Locke, 1996; Saxenian, 1996; Chang, 2009; Sturgeon and Kawa­ as network visualization and deconstruction methods (MacCormack
kami, 2011; Blyde, 2014; Stam, 2015; Acs et al., 2017; Wurth et al., et al., 2006, 2012; Baldwin et al., 2014; Pujadas et al., 2024). Other
2022). The overlapping memberships by organizations in innovation potential sources of data for large-sample studies are the statistics
and entrepreneurial ecosystems suggest numerous opportunities to published by various open consortia and platforms, such as the Linux
research how actors, activities or coordination within one type of and Apache Foundations and the GitHub repository.

9
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

The dual concepts of complementarity and joint-value creation are Declaration of competing interest
central to the application of the ecosystem lens. Today, however, it is
difficult to measure the incremental value created by combinations of The authors declare that they have no known competing financial
complementary components. Identifying the causes of complementarity interests or personal relationships that could have appeared to influence
within an ecosystem, and describing how the joint value created changes the work reported in this paper.
over time are thus important avenues of future research (e.g., Agarwal
and Kapoor, 2022). Data availability
Such analysis requires new methods, however. A structural analysis
of the functional components contributing to value in complex technical No data was used for the research described in the article.
systems can be used to deconstruct joint-value creation and connect it to
innovation outcomes (Adner, 2021, Ch. 1; Baldwin, 2024, Ch. 4). “Value Acknowledgements
structure” analysis also provides a way to distinguish between special­
ized and generic complements and thus predict which contributors to First, we wish to thank the reviewers of the papers submitted to the
the system are likely to capture a significant percentage of the com­ special issue, who provided excellent feedback that greatly improved
plementary surplus. To date, however, these methods have been rarely each submission. We also wish to thank Elizabeth Altman, Frank Piller
used. and Erik Stam for feedback on earlier drafts of this article.
This research was partially supported by the Novo Nordisk Foun­
6. Conclusion dation, grant number NNF 16OC0021630.
All authors contributed equally.
Since Schumpeter’s path-breaking work, the field of innovation
studies has moved from an initial focus on industries and firms to an References
increasing focus on ecosystems. However, despite the rising prominence
of ecosystems, there remains significant ambiguity about what consti­ Abernathy, William J., Utterback, James, 1978. Patterns of industrial innovation.
Technol. Rev. 80, 41–47.
tutes an ecosystem and how an “ecosystem lens” can be applied to Acs, Z.J., Stam, E., Audretsch, D.B., O’Connor, A., 2017. The lineages of the
innovation in ways that facilitate cumulative progress in our knowledge. entrepreneurial ecosystem approach. Small Bus. Econ. 49, 1–10.
To address this ambiguity, this article provides three criteria that can Adner, R., 2006. Match your innovation strategy to your innovation ecosystem. Harv.
Bus. Rev. 84 (4), 98.
be used to define ecosystems. First, ecosystems are comprised of Adner, R., 2012. The Wide Lens: A New Strategy for Innovation. Penguin.
autonomous actors (organizations and individuals) subject to distrib­ Adner, R., 2017. Ecosystem as structure: an actionable construct for strategy. J. Manag.
uted governance that capture enough value to justify their participation. 43 (1), 39–58.
Adner, R., 2021. Winning the Right Game: How to Disrupt, Defend, and Deliver in a
Second, the actors contribute in complementary ways to a focal value Changing World. MIT Press.
proposition, so that the joint value of the whole system is greater than Adner, R., Kapoor, R., 2010. Value creation in innovation ecosystems: how the structure
the sum of the values of its separate parts. Finally, the technical com­ of technological interdependence affects firm performance in new technology
generations. Strategic Manag. J. 31 (3), 306–333.
ponents in an ecosystem are modules within a larger technical archi­
Adner, R., Kapoor, R., 2016. Innovation ecosystems and the pace of substitution: re-
tecture. An ecosystem lens on innovation focuses on innovations and examining technology S-curves. Strategic Manag. J. 37 (4), 625–648.
organizations satisfying these criteria. Agarwal, S., Kapoor, R., 2022. Value creation tradeoff in business ecosystems: leveraging
The articles in the special issue highlight the benefits of using an complementarities while managing interdependencies. Organization Sci. 34 (3),
987–1352.
ecosystem lens to study innovation. These benefits include: (1) the Altman, E.J., Nagle, F., Tushman, M.L., 2022. The translucent hand of managed
identification of different types of actors (e.g., users, suppliers, com­ ecosystems: engaging communities for value creation and capture. Acad. Manag.
plementors, distributors, public actors, and intermediaries); (2) de­ Ann. 16 (1), 70–101.
Altman, E.J., Kiron, D., Schwartz, J., Jones, R., 2023. Workforce Ecosystems: Reaching
scriptions of the processes that shape innovation and joint-value Strategic Goals with People, Partners, and Technologies. MIT Press.
creation in both established and emerging industries; and (3) uncover­ Anderson, P., Tushman, M.L., 1990. Technological discontinuities and dominant designs:
ing common features and themes across the different articles. This lens a cyclical model of technological change. Adm. Sci. Q. 35 (4), 604–634.
Anderson, P., Tushman, M.L., 2001. Organizational environments and industry exit: the
can also help us to understand differences between types of ecosystems. effects of uncertainty, munificence and complexity. Ind. Corp. Chang. 10 (3),
For example, we can distinguish between innovation ecosystems coor­ 675–711.
dinated by platforms and those coordinated by other means, as well as Ansari, S., Garud, R., Kumaraswamy, A., 2016. The disruptor’s dilemma: TiVo and the US
television ecosystem. Strateg. Manag. J. 37 (9), 1829–1853.
between ecosystems concerned primarily with innovation vs. those Autio, E., Thomas, L.D., 2021. Researching ecosystems in innovation contexts. Innov.
focused on entrepreneurship, regional development, or the creation of Manag. Rev. 19 (1), 12–25.
new knowledge. Baden-Fuller, C., Haefliger, S., 2013. Business models and technological innovation. Long
Range Plann. 46 (6), 419–426.
Taken as a whole, the articles shed light on how incumbents and
Baldwin, C.Y., 2008. Where do transactions come from? Modularity, transactions, and
entrants can create novel value propositions by orchestrating their the boundaries of firms. Ind. Corp. Chang. 17 (1), 155–195.
respective ecosystems, while acknowledging the challenges associated Baldwin, C.Y., 2018. Bottlenecks, modules, and dynamic architectural capabilities. In:
with governing autonomous actors and maintaining multilateral align­ Teece, D.J., Leih, S. (Eds.), The Oxford Handbook of Dynamic Capabilities. Oxford
University Press.
ment in an ever-changing environment. Beyond offering new insights to Baldwin, C.Y., 2024. Design Rules, Volume 2: How Technology Shapes Organizations.
better understand innovation ecosystems, we also hope that our special MIT Press.
issue will spur better research, practice and policy on ecosystem Baldwin, C.Y., Clark, K.B., 1997. Managing in an age of modularity. Harv. Bus. Rev. 75
(5), 84–93.
innovation. Baldwin, C.Y., Clark, K.B., 2000. Design Rules, Volume 1, The Power of Modularity. MIT
Press, Cambridge, MA.
CRediT authorship contribution statement Baldwin, C.Y., Clark, K.B., 2006. The architecture of participation: does code architecture
mitigate free riding in the open source development model? Manag. Sci. 52 (7),
1116–1127.
Carliss Y. Baldwin: Conceptualization, Writing – review & editing. Baldwin, C.Y., Von Hippel, E., 2011. Modeling a paradigm shift: from producer
Marcel L.A.M. Bogers: Conceptualization, Writing – review & editing. innovation to user and open collaborative innovation. Organization Sci. 22 (6),
1399–1417.
Rahul Kapoor: Conceptualization, Writing – review & editing. Joel Baldwin, C.Y., Woodard, C.J., 2009. The architecture of platforms: a unified view.
West: Conceptualization, Writing – review & editing. Platforms Markets Innov. 19–44.
Baldwin, C.Y., MacCormack, A.D., Rusnak, J., 2014. Hidden structure: using network
methods to map product architecture. Research Policy 43, 1381–1397.

10
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

Blyde, J.S., 2014. Synchronized Factories: Latin America and the Caribbean in the Era of Garcia, R., Calantone, R., 2002. A critical look at technological innovation typology and
Global Value Chains. Springer Nature. innovativeness terminology: a literature review. J. Prod. Innov. Manag. 19 (2),
Bogers, M., 2011. The open innovation paradox: knowledge sharing and protection in 110–132.
R&D collaborations. Eur. J. Innov. Manag. 14 (1), 93–117. Gault, F., 2018. Defining and measuring innovation in all sectors of the economy. Res.
Bogers, M., Zobel, A.K., Afuah, A., Almirall, E., Brunswicker, S., Dahlander, L., Ter Policy 47 (3), 617–622.
Wal, A.L., 2017. The open innovation research landscape: established perspectives Gawer, A., 2021. Digital platforms’ boundaries: the interplay of firm scope, platform
and emerging themes across different levels of analysis. Ind. Innov. 24 (1), 8–40. sides, and digital interfaces. Long Range Plann. 54 (5), 102045.
Bogers, M., Sims, J., West, J., 2019. What Is an Ecosystem? Incorporating 25 Years of Gawer, A., Cusumano, M.A., 2002. Platform Leadership: How Intel, Microsoft, and Cisco
Ecosystem Research. Working Paper. [Link] Drive Industry Innovation. Harvard Business School Press, Boston.
Borner, K., Berends, H., Deken, F., Feldberg, F., 2023. Another pathway to Gawer, A., Phillips, N., 2013. Institutional work as logics shift: the case of Intel’s
complementarity: how users and intermediaries identify and create new transformation to platform leader. Organization Stud. 34 (8), 1035–1071.
combinations in innovation ecosystems. Res. Policy 52 (7), 104788. [Link] Granstrand, O., Holgersson, M., 2020. Innovation ecosystems: a conceptual review and a
10.1016/[Link].2023.104788. new definition. Technovation 90, 102098.
Bresnahan, T.F., Greenstein, S., 1999. Technological competition and the structure of the Grove, Andrew S., 1996. Only the Paranoid Survive. Doubleday, New York.
computer industry. J. Ind. Econ. 47 (1), 1–40. Hagedoorn, J., Cloodt, M., 2003. Measuring innovative performance: is there an
Brusoni, Stefano, Prencipe, Andrea, 2001. Unpacking the black box of modularity: advantage in using multiple indicators? Res. Policy 32 (8), 1365–1379.
technologies, products and organizations. Ind. Corp. Change 10 (1), 179–205. Haki, K., Blaschke, M., Aier, S., Winter, R., Tilson, D., 2022. Dynamic capabilities for
Brusoni, Stefano, Prencipe, Andrea, Pavitt, Keith, 2001. Knowledge specialization, transitioning from product platform ecosystem to innovation platform ecosystem.
organizational coupling and the boundaries of the firm: why do firms know more Eur. J. Inf. Syst. 1–19.
than they make? Adm. Sci. Q. 46 (4), 597–621. Hannah, D.P., Eisenhardt, K.M., 2018. How firms navigate cooperation and competition
Ceccagnoli, M., Forman, C., Huang, P., Wu, D.J., 2012. Cocreation of value in a platform in nascent ecosystems. Strateg. Manag. J. 39 (12), 3163–3192.
ecosystem! The case of enterprise software. MIS Q. 263–290. Heaton, S., Siegel, D.S., Teece, D.J., 2019. Universities and innovation ecosystems: a
Cennamo, C., Santalo, J., 2013. Platform competition: strategic trade-offs in platform dynamic capabilities perspective. Industrial Corp. Change 28 (4), 921–939.
markets. Strategic Manag. J. 34 (11), 1331–1350. Helfat, C.E., 2000. Guest editor’s introduction to the special issue: the evolution of firm
Cennamo, C., Santaló, J., 2019. Generativity tension and value creation in platform capabilities. Strategic Manag. J. 21 (10− 11), 955–959.
ecosystems. Organ. Sci. 30 (3), 617–641. Helfat, C.E., Peteraf, M.A., 2015. Managerial cognitive capabilities and the
Chandler, Alfred D., 1977. The Visible Hand: The Managerial Revolution in American microfoundations of dynamic capabilities. Strategic Manag. J. 36 (6), 831–850.
Business. Harvard University Press, Cambridge, MA. Helfat, C.E., Raubitschek, R.S., 2018. Dynamic and integrative capabilities for profiting
Chang, L.T., 2009. Factory Girls: From Village to City in a Changing China. Random from innovation in digital platform-based ecosystems. Res. Policy 47 (8),
House Digital, Inc. 1391–1399.
Chesbrough, H.W., 2003. Open Innovation: The New Imperative for Creating and Helper, S., Sako, M., 2010. Management innovation in supply chain: appreciating
Profiting from Technology. Harvard Business Press. Chandler in the twenty-first century. Industrial Corp. Change 19 (2), 399–429.
Chesbrough, H., 2020. Open Innovation Results: Going Beyond the Hype and Getting Henderson, R.M., Clark, K.B., 1990. Architectural innovation: the reconfiguration of
Down to Business. Oxford University Press. existing systems and the failure of established firms. Adm. Sci. Q. 35 (1), 9–30.
Chesbrough, H., Bogers, M., 2014. Explicating open innovation: Clarifying an emerging Holgersson, M., Baldwin, C.Y., Chesbrough, H., Bogers, M.L., 2022. The forces of
paradigm for understanding innovation. In: Chesbrough, H., Vanhaverbeke, W., ecosystem evolution. Calif. Manage. Rev. 64 (3), 5–23.
West, J. (Eds.), New Frontiers in Open Innovation. Oxford University Press, pp. 3–28. Iansiti, Marco, Levien, Roy, 2004. The Keystone Advantage: What the New Dynamics of
Chesbrough, H., Vanhaverbeke, W., West, J. (Eds.), 2006. Open Innovation: Researching Business Ecosystems Mean for Strategy, Innovation, and Sustainability. Harvard
a New Paradigm. Oxford University Press. Business School Press, Boston.
Chposky, J., Leonsis, T., 1988. Blue Magic: The People, Power and Politics Behind the Intel Oral History Panel, 2008. Computer History Museum, Mountain View, CA. [Link]
IBM Personal Computer. Facts on File Publications, New York, NY. [Link]/resources/text/oral_history/Intel_386_Business_Strategy/
Christensen, C.M., 2013. The Innovator’s Dilemma: When New Technologies Cause Great [Link]. August 14, (viewed April 18, 2017).
Firms to Fail. Harvard Business Review Press. Jacobides, M.G., Cennamo, C., Gawer, A., 2018. Towards a theory of ecosystems.
Clarysse, B., Wright, M., Bruneel, J., Mahajan, A., 2014. Creating value in ecosystems: Strategic Manag. J. 39 (8), 2255–2276.
crossing the chasm between knowledge and business ecosystems. Res. Policy 43 (7), Jacobides, M.G., Cennamo, C., Gawer, A., 2024. Externalities and complementarities in
1164–1176. platforms and ecosystems: from structural solutions to endogenous failures. Res.
Cobben, D., Ooms, W., Roijakkers, N., Radziwon, A., 2022. Ecosystem types: a systematic Policy 53 (1), 104906. [Link]
review on boundaries and goals. J. Bus. Res. 142, 138–164. Järvi, K., Almpanopoulou, A., Ritala, P., 2018. Organization of knowledge ecosystems:
Cohen, W.M., 2010. Fifty years of empirical studies of innovative activity and prefigurative and partial forms. Res. Policy 47 (8), 1523–1537.
performance. In: Handbook of the Economics of Innovation, 1, pp. 129–213. Jones, S.L., Leiponen, A., Vasudeva, G., 2021. The evolution of cooperation in the face of
Colfer, L.J., Baldwin, C.Y., 2016. The mirroring hypothesis: theory, evidence, and conflict: evidence from the innovation ecosystem for mobile telecom standards
exceptions. Ind. Corp. Change 25 (5), 709–738. development. Strategic Manag. J. 42 (4), 710–740.
Cozzolino, A., Geiger, S., 2024. Ecosystem disruption and regulatory positioning: entry Kamien, M.I., Schwartz, N.L., 1982. Market Structure and Innovation. Cambridge
strategies of digital health startup orchestrators and complementors. Res. Policy 53 University Press.
(2), 104913. [Link] Kapoor, R., 2018. Ecosystems: broadening the locus of value creation. J. Organ. Des. 7
Cringely, Robert X., 1992. Accidental Empires. Addison-Wesley, Reading, MA. (1), 1–16.
Cyert, R.M., March, J.G., 1963. A Behavioral Theory of the Firm. Prentice Hall, Kapoor, R., Adner, R., 2012. What firms make vs. what they know: how firms’ production
Englewood Cliffs, NJ. and knowledge boundaries affect competitive advantage in the face of technological
Dahlander, L., Gann, D.M., Wallin, M.W., 2021. How open is innovation? A retrospective change. Organ. Sci. 23 (5), 1227–1248.
and ideas forward. Res. Policy 50 (4), 104218. Kapoor, R., Agarwal, S., 2017. Sustaining superior performance in business ecosystems:
Dattée, B., Alexy, O., Autio, E., 2018. Maneuvering in poor visibility: how firms play the evidence from application software developers in the iOS and Android smartphone
ecosystem game when uncertainty is high. Acad. Manage. J. 61 (2), 466–498. ecosystems. Organization Sci. 28 (3), 531–551.
Dosi, G., 1982. Technological paradigms and technological trajectories: a suggested Kapoor, R., Lee, J.M., 2013. Coordinating and competing in ecosystems: how
interpretation of the determinants and directions of technical change. Res. Policy 11 organizational forms shape new technology investments. Strategic Manag. J. 34 (3),
(3), 147–162. 274–296.
Dosi, Giovanni, Nelson, Richard R., Winter, Sidney G. (Eds.), 2000. The Nature and Kapoor, R., McGrath, P.J., 2014. Unmasking the interplay between technology evolution
Dynamics if Organizational Capabilities. Oxford University Press, Oxford, UK. and R&D collaboration: evidence from the global semiconductor manufacturing
Edquist, C., 1997. Systems of Innovation: Technologies, Institutions, and Organizations. industry, 1990–2010. Res. Policy 43 (3), 555–569.
Pinter, London. Klepper, S., 1996. Entry, exit, growth, and innovation over the product life cycle. Am.
Eisenmann, T., Parker, G., Van Alstyne, M., 2011. Platform envelopment. Strategic Econ. Rev. 562–583.
Manag. J. 32 (12), 1270–1285. Klepper, S., 1997. Industry life cycles. Ind. Corp. Change 6 (1), 145–182.
Ethiraj, S.K., 2007. Allocation of inventive effort in complex product systems. Strategic Kretschmer, T., Leiponen, A., Schilling, M., Vasudeva, G., 2022. Platform ecosystems as
Manag. J. 28 (6), 563–584. meta-organizations: implications for platform strategies. Strategic Manag. J. 43 (3),
Fagerberg, J., Verspagen, B., 2009. Innovation studies—the emerging structure of a new 405–424.
scientific field. Res. Policy 38 (2), 218–233. Kuan, J., West, J., 2023. Interfaces, modularity and ecosystem emergence: how DARPA
Franke, N., von Hippel, E., 2003. Satisfying heterogeneous user needs via innovation modularized the semiconductor ecosystem. Res. Policy 52 (8), 104789. [Link]
toolkits: the case of Apache security software. Res. Policy 32 (7), 1199–1215. org/10.1016/[Link].2023.104789.
Freeman, C., 1974. The Economics of Industrial Innovation. Penguin, London. Laursen, K., Salter, A., 2006. Open for innovation: the role of openness in explaining
Freeman, C., 1995. The ‘National System of Innovation’ in historical perspective. Camb. innovation performance among UK manufacturing firms. Strategic Manag. J. 27 (2),
J. Econ. 19 (1), 5–24. 131–150.
Freiberger, P., Swaine, M., 1984. Fire in the Valley. Osborne/McGraw-Hill, Berkeley, Laursen, K., Salter, A., 2014. The paradox of openness: appropriability, external search
p. 279. and collaboration. Res. Policy 43 (5), 867–878.
Froot, K.A., 1992. Intel Corporation, 1992. Harvard Business School Publishing. Lee, S., Kim, W., Lee, H., Jeon, J., 2016. Identifying the structure of knowledge networks
Ganco, M., Kapoor, R., Lee, G.K., 2020. From rugged landscapes to rugged ecosystems: in the US mobile ecosystems: patent citation analysis. Tech. Anal. Strat. Manag. 28
structure of interdependencies and firms’ innovative search. Acad. Manage. Rev. 45 (4), 411–434.
(3), 646–674.

11
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

Leten, B., Vanhaverbeke, W., Roijakkers, N., Clerix, A., Van Helleputte, J., 2013. IP Sako, M., 2004. Supplier development at Honda, Nissan and Toyota: comparative case
models to orchestrate innovation ecosystems: IMEC, a public research institute in studies of organizational capability enhancement. Industrial Corp. Change 13 (2),
nano-electronics. Calif. Manage. Rev. 55 (4), 51–64. 281–308.
Locke, R.M., 1996. The composite economy: local politics and industrial change in Sako, M., Helper, S., 1998. Determinants of trust in supplier relations: evidence from the
contemporary Italy. Econ. Soc. 25 (4), 483–510. automotive industry in Japan and the United States. J. Econ. Behav. Organ. 34 (3),
MacCormack, Alan, Rusnak, John, Baldwin, Carliss, 2006. Exploring the structure of 387–417.
complex software designs: an empirical study of open source and proprietary code. Saxenian, A., 1996. Regional Advantage: Culture and Competition in Silicon Valley and
Manag. Sci. 52 (7), 1015–1030. Route 128, With a New Preface by the Author. Harvard University Press.
MacCormack, Alan, Baldwin, Carliss, Rusnak, John, 2012. Exploring the duality between Schäper, T., Jung, C., Foege, J.N., Bogers, M.L.A.M., Fainshmidt, S., Nüesch, S., 2023.
product and organizational architectures: a test of the “mirroring” hypothesis. Res. The S-shaped relationship between open innovation and financial performance: a
Policy 41 (8), 1309–1324. longitudinal perspective using a novel text-based measure. Res. Policy 52 (6),
Malerba, F., 2002. Sectoral systems of innovation and production. Res. Policy 31 (2), 104764.
247–264. Scherer, F.M., 1986. Innovation and Growth: Schumpeterian Perspectives. MIT Press.
Martin, B.R., 2012. The evolution of science policy and innovation studies. Res. Policy 41 Schmookler, J., 1966. Invention and Economic Growth. Harvard University Press.
(7), 1219–1239. Schreieck, M., Wiesche, M., Krcmar, H., 2021. Capabilities for value co-creation and
Massa, L., Tucci, C.L., Afuah, A., 2017. A critical assessment of business model research. value capture in emergent platform ecosystems: A longitudinal case study of SAP’s
Acad. Manag. Ann. 11 (1), 73–104. cloud platform. J. Inf. Technol. 36 (4), 365–390. [Link]
Milgrom, P., Roberts, J., 1995. Complementarities and fit strategy, structure, and 02683962211023780.
organizational change in manufacturing. J. Account. Econ. 19 (2), 179–208. Schumpeter, Joseph A., 1934. The Theory of Economic Development. Harvard University
Miric, M., Jeppesen, L.B., 2023. How does competition influence innovative effort within Press, Cambridge, MA.
a platform-based ecosystem? Contrasting paid and unpaid contributors. Res. Policy Schumpeter, Joseph A., 1942. Capitalism, Socialism, and Democracy. Harper & Brothers,
52 (7), 104790. [Link] New York.
Moore, J.F., 1993. Predators and prey: a new ecology of competition. Harv. Bus. Rev. 71 Simcoe, T., 2006. Open standards and intellectual property rights. In: Open Innovation:
(3), 75–86. Researching a New Paradigm. Oxford University Press.
Moore, J.F., 1996. The Death of Competition: Leadership and Strategy in the Age of Simon, H.A., 1947. Administrative Behavior. Simon and Schuster.
Business Ecosystems. HarperBusiness. Simon, Herbert A., 1962. The Architecture of Complexity. Proceedings of the American
Morris, Charles R., Ferguson, Charles H., 1993. How Architecture Wins Technology Philosophical Society 106: 467–482, reprinted in Simon (1981) The Sciences of the
Wars, March–April. Harvard Business Review, pp. 86–96. Artificial, 2nd ed. MIT Press, Cambridge, MA, pp. 193–229.
Mowery, D.C., Nelson, R.R. (Eds.), 1999. Sources of Industrial Leadership. Cambridge Simon, Herbert A., 1969. The Sciences of the Artificial. MIT Press, Cambridge MA.
University Press, Cambridge. Smith, K., 2006. Measuring innovation. In: Fagerberg, Jan, Mowery, David C. (Eds.),
Nelson, Richard R. (Ed.), 1993. National Innovation Systems: A Comparative Analysis. Oxford Handbook of Innovation (Oxford).
Oxford University Press, New York. Snihur, Y., Zott, C., Amit, R., 2021. Managing the value appropriation dilemma in
Nelson, R.R., Winter, S.G., 1982. An Evolutionary Theory of Economic Change. Harvard business model innovation. Strat. Sci. 6 (1), 22–38. [Link]
University Press. stsc.2020.0113.
Nicotra, M., Romano, M., Del Giudice, M., Schillaci, C.E., 2018. The causal relation Song, Y., Gnyawali, D., Qian, L., 2024. From early curiosity to space wide web: the
between entrepreneurial ecosystem and productive entrepreneurship: a emergence of the small satellite innovation ecosystem. Res. Policy 53 (2), 104932.
measurement framework. J. Technol. Transfer. 43 (3), 640–673. [Link]
Olk, P., West, J., 2020. The relationship of industry structure to open innovation: Stam, E., 2015. Entrepreneurial ecosystems and regional policy: a sympathetic critique.
cooperative value creation in pharmaceutical consortia. R&D Manag. 50 (1), Eur. Plan. Stud. 23 (9), 1759–1769. [Link]
116–135. 09654313.2015.1061484.
Olk, P., West, J., 2023. Distributed governance of a complex ecosystem: how R&D Stam, E., Spigel, B., 2018. Entrepreneurial Ecosystems. In: Blackburn, R., De Clercq, D.,
consortia orchestrate the Alzheimer’s knowledge ecosystem. Calif. Manage. Rev. 65 Heinonen, J. (Eds.), Sage Handbook of Entrepreneurship and Small Business. SAGE,
(2), 93–128. London, pp. 407–422.
O’Mahony, S., Karp, R., 2022. From proprietary to collective governance: how do Staudenmayer, Nancy, Tripsas, Mary, Tucci, Christopher L., 2005. Interfirm modularity
platform participation strategies evolve? Strategic Manag. J. 43 (3), 530–562. and its implications for product development. J. Product Innov. Manag. 22 (4),
Parker, G., Van Alstyne, M., Jiang, X., 2017. Platform ecosystems. MIS Q. 41 (1), 303–321.
255–266. Sturgeon, Timothy J., 2002. Modular production networks: a new American model of
Parnas, David L., 1972. A technique for software module specification with examples. industrial organization. Ind. Corp. Change 11 (3), 451–496.
Commun. ACM 15, 330–336. Sturgeon, T.J., Kawakami, M., 2011. Global value chains in the electronics industry:
Parnas, David L., 1972b. On the Criteria to Be Used in Decomposing Systems into characteristics, crisis, and upgrading opportunities for firms from developing
Modules. In: Communications of the ACM 15: 1053-58. Software Fundamentals: countries. Int. J. Technol. Learn. Innov. Develop. 4 (1–3), 120–147.
Collected Papers of David Parnas. Addison-Wesley, Boston MA reprinted in Hoffman Suarez, F.F., 2004. Battles for technological dominance: an integrative framework. Res.
and Weiss (eds.). Policy 33 (2), 271–286.
Parnas, David L., 2001. In: Hoffman, D.M., Weiss, D.M. (Eds.), Software Fundamentals: Sydow, J., Windeler, A., Schubert, C., Möllering, G., 2012. Organizing R&D consortia for
Collected Papers by David L. Parnas. Addison-Wesley, Boston MA. path creation and extension: the case of semiconductor manufacturing technologies.
Pavitt, K., 1984. Sectoral patterns of technical change: towards a taxonomy and a theory. Organ. Stud. 33 (7), 907–936.
Res. Policy 13 (6), 343–373. Teece, D.J., 1986. Profiting from technological innovation: implications for integration,
Pavitt, K., 1999. Technology, Management and Systems of Innovation. Edward Elgar collaboration, licensing and public policy. Res. Policy 15 (6), 285–305.
Publishing. Teece, D.J., 2006. Reflections on “profiting from innovation”. Res. Policy 35 (8),
Pujadas, R., Valderrama, E., Venters, W., 2024. The value and structuring role of web 1131–1146.
APIs in digital innovation ecosystems: the case of the online travel ecosystem. Res. Teece, D.J., 2009. Dynamic capabilities and strategic management: Organizing for
Policy 53 (2), 104931. [Link] innovation and growth. Oxford University Press.
Qiu, Y., Gopal, A., Hann, I.H., 2017. Logic pluralism in mobile platform ecosystems: a Teece, D.J., 2018. Profiting from innovation in the digital economy: enabling
study of indie app developers on the iOS app store. Inf. Syst. Res. 28 (2), 225–249. technologies, standards, and licensing models in the wireless world. Res. policy 47
Randhawa, K., West, J., Skellern, K., Josserand, E., 2021. Evolving a value chain to an (8), 1367–1387.
open innovation ecosystem: cognitive engagement of stakeholders in customizing Teece, David J., Pisano, Gary, Shuen, Amy, 1997. Dynamic capabilities and strategic
medical implants. Calif. Manage. Rev. 63 (2), 101–134. management. Strategic Manag. J. 18 (7), 509–533.
Randhawa, K., Wilden, R., Akaka, M.A., 2022. Innovation intermediaries as collaborators Thun, E., Taglioni, D., Sturgeon, T., Dallas, M.P., 2022. Massive Modularity:
in shaping service ecosystems: the importance of dynamic capabilities. Ind. Mark. Understanding Industry Organization in the Digital Age—The Case of Mobile Phone
Manag. 103, 183–197. Handsets. No. 10164. The World Bank.
Reiter, A., Stonig, J., Frankenberger, K., 2024. Managing multi-tiered innovation Toh, P.K., Miller, C.D., 2017. Pawn to save a chariot, or drawbridge into the fort? Firms’
ecosystems. Res. Policy 53 (1), 104905. [Link] disclosure during standard setting and complementary technologies within
respol.2023.104905. ecosystems. Strategic Manag. J. 38 (11), 2213–2236.
Reypens, C., Lievens, A., Blazevic, V., 2021. Hybrid orchestration in multi-stakeholder Tushman, M.L., Anderson, P., 1986. Technological discontinuities and organizational
innovation networks: practices of mobilizing multiple, diverse stakeholders across environments. Adm. Sci. Q. 31 (3), 439–465.
organizational boundaries. Organization Stud. 42 (1), 61–83. Utterback, J.M., Abernathy, W.J., 1975. A dynamic model of process and product
Rietveld, J., Schilling, M.A., Bellavitis, C., 2019. Platform strategy: managing ecosystem innovation. Omega 3 (6), 639–656.
value through selective promotion of complements. Organization Sci. 30 (6), Uzunca, B., Sharapov, D., Tee, R., 2022. Governance rigidity, industry evolution, and
1232–1251. value capture in platform ecosystems. Res. Policy 51 (7), 104560.
Rohrbeck, R., Hölzle, K., Gemünden, H.G., 2009. Opening up for competitive van der Borgh, M., Cloodt, M., Romme, A.G.L., 2012. Value creation by knowledge-based
advantage–how Deutsche Telekom creates an open innovation ecosystem. R&D ecosystems: evidence from a field study. R&D Manag. 42 (2), 150–169.
Manag. 39 (4), 420–430. van Dyck, M., Lüttgens, D., Diener, K., Piller, F., Pollok, P., 2024. From product to
Rosenberg, N., 1982. Inside the Black Box: Technology and Economics. Cambridge platform: how incumbents’ assumptions and choices shape their platform strategy.
University Press. Res. Policy 53 (1), 104904. [Link]
von Hippel, E., 1988. The Sources of Innovation (Oxford).
von Hippel, E., 2005. Democratizing Innovation. MIT Press.

12
C.Y. Baldwin et al. Research Policy 53 (2024) 104949

von Hippel, E., 2019. Free Innovation. MIT Press. West, J., Salter, A., Vanhaverbeke, W., Chesbrough, H., 2014. Open innovation: the next
West, J., 2014. Challenges of funding open innovation platforms: lessons from Symbian decade. Res. Policy 43 (5), 805–811.
Ltd. In: Chesbrough, H., Vanhaverbeke, W., West, J. (Eds.), New Frontiers in Open Womack, J.P., Jones, D.T., Roos, D., 1990. The Machine That Changed the World: The
Innovation. Oxford University Press, Oxford, pp. 71–93. Story of Lean Production. Rawson Associates, New York, NY.
West, J., Dedrick, J., 2000. Innovation and control in standards architectures: the rise Wurth, B., Stam, E., Spigel, B., 2022. Toward an entrepreneurial ecosystem research
and fall of Japan’s PC-98. Inform. Syst. Res. 11 (2), 197–216. program. Entrep. Theory Pract. 46 (3), 729–778.
West, J., Gallagher, S., 2006. Challenges of open innovation: the paradox of firm Zhu, F., Iansiti, M., 2012. Entry into platform-based markets. Strategic Manag. J. 33 (1),
investment in open-source software. R&D Manag. 36 (3), 319–331. 88–106.
West, J., Wood, D., 2013. Evolving an open ecosystem: the rise and fall of the Symbian Zott, C., Amit, R., Massa, L., 2011. The business model: recent developments and future
platform. Adv. Strateg. Manag. 30, 27–68. research. J. Manag. 37 (4), 1019–1042.

13