Value Creation in Innovation Ecosystems: How the Structure of Technological Interdependence Affects Firm Performance in New Technology Generations

TLDR: It is proposed that the effectiveness of vertical integration as a strategy to manage ecosystem interdependence increases over the course of the technology life cycle.

Abstract: The success of an innovating firm often depends on the efforts of other innovators in its environment. How do the challenges faced by external innovators affect the focal firm's outcomes? To address this question we first characterize the external environment according to the structure of interdependence. We follow the flow of inputs and outputs in the ecosystem to distinguish between upstream components that are bundled by the focal firm, and downstream complements that are bundled by the firm's customers. We argue that the effect of external innovation challenges depends not only on their magnitude, but also on their location in the ecosystem relative to the focal firm - whereas greater innovation challenges in components enhances the benefits that accrue to technology leaders, greater innovation challenges in complements erodes these benefits. We further argue that the effectiveness of vertical integration as a strategy to manage ecosystem interdependence increases over the course of the technology life cycle. We explore these arguments in the context of the global semiconductor lithography industry from its emergence in 1962 to 2005 across nine distinct technology generations. We find strong support for our arguments.

Figures

TABLE 2: Variable descriptions.

TABLE 1: Major technological innovations by the lithography equipment ecosystem to overcome the challenges accompanying new technology generations.

Figure 1: Generic Schema of an Ecosystem.

Figure 2: Basic Schema of the Semiconductor Lithography Process.

TABLE 4: Coefficient estimates from panel regression. Dependent Variable = Ln(Generation Share)

Figure 4: Distribution of challenges among ecosystem elements in nine semiconductor lithography equipment generations.

Full text

(94A0=>4?D:1$099>D7A,94,(94A0=>4?D:1$099>D7A,94,
&.3:7,=7D:88:9>&.3:7,=7D:88:9>
!,9,20809?$,;0=> *3,=?:9,.@7?D%0>0,=.3

),7@0=0,?4:94999:A,?4:9.:>D>?08>:B?30&?=@.?@=0:1),7@0=0,?4:94999:A,?4:9.:>D>?08>:B?30&?=@.?@=0:1
'0.39:7:24.,79?0=/0;09/09.0110.?>4=8$0=1:=8,9.049"0B'0.39:7:24.,79?0=/0;09/09.0110.?>4=8$0=1:=8,9.049"0B
'0.39:7:2D090=,?4:9>'0.39:7:2D090=,?4:9>
%:9/90=
(94A0=>4?D:1$099>D7A,94,
%,3@7 ,;::=
(94A0=>4?D:1$099>D7A,94,
:77:B?34>,9/,//4?4:9,7B:=6>,?3??;>=0;:>4?:=D@;0990/@828?+;,;0=>
$,=?:1?30@>490>>/8494>?=,?4:9!,9,20809?,9/#;0=,?4:9>:88:9>
%0.:8809/0/4?,?4:9%0.:8809/0/4?,?4:9
/90=% ,;::=%),7@0=0,?4:94999:A,?4:9.:>D>?08>:B?30&?=@.?@=0:1
'0.39:7:24.,79?0=/0;09/09.0110.?>4=8$0=1:=8,9.049"0B'0.39:7:2D090=,?4:9>
&?=,?024.
!,9,20809?:@=9,7


3??;/C/:4:=2>85
'34>;,;0=4>;:>?0/,?&.3:7,=7D:88:9>3??;>=0;:>4?:=D@;0990/@828?+;,;0=>
:=8:=0491:=8,?4:9;70,>0.:9?,.?=0;:>4?:=D;:-:C@;0990/@

),7@0=0,?4:94999:A,?4:9.:>D>?08>:B?30&?=@.?@=0:1'0.39:7:24.,7),7@0=0,?4:94999:A,?4:9.:>D>?08>:B?30&?=@.?@=0:1'0.39:7:24.,7
9?0=/0;09/09.0110.?>4=8$0=1:=8,9.049"0B'0.39:7:2D090=,?4:9>9?0=/0;09/09.0110.?>4=8$0=1:=8,9.049"0B'0.39:7:2D090=,?4:9>
->?=,.?->?=,.?
'30>@..0>>:1,9499:A,?492G=8:1?09/0;09/>:9?30011:=?>:1:?30=499:A,?:=>494?>09A4=:9809?
:B/:?30.3,770920>1,.0/-D0C?0=9,7499:A,?:=>,110.??301:.,7G=8>:@?.:80>':,//=0>>?34>
<@0>?4:9B0G=>?.3,=,.?0=4E0?300C?0=9,709A4=:9809?,..:=/492?:?30>?=@.?@=0:149?0=/0;09/09.0*0
1:77:B?30H:B:149;@?>,9/:@?;@?>49?300.:>D>?08?:/4>?492@4>3-0?B009@;>?=0,8.:8;:909?>?3,?
,=0-@9/70/-D?301:.,7G=8,9//:B9>?=0,8.:8;70809?>?3,?,=0-@9/70/-D?30G=8>.@>?:80=>
*03D;:?30>4E0?3,??300110.?>:10C?0=9,7499:A,?4:9.3,770920>/0;09/9:?:97D:9?304=8,294?@/0-@?
,7>::9?304=7:.,?4:949?300.:>D>?08=07,?4A0?:?301:.,7G=8*04/09?41D,60D,>D880?=D?3,?=0>@7?>
1=:8?307:.,?4:9:1.3,770920>=07,?4A0?:,1:.,7G=8F2=0,?0=@;>?=0,8499:A,?4:9.3,770920>49
.:8;:909?>093,9.0?30-090G?>?3,?,..=@0?:?0.39:7:2D70,/0=>B34702=0,?0=/:B9>?=0,8
499:A,?4:9.3,770920>49.:8;70809?>0=:/0?30>0-090G?>*01@=?30=;=:;:>0?3,??300110.?4A090>>:1
A0=?4.,749?02=,?4:9,>,>?=,?02D?:8,9,200.:>D>?0849?0=/0;09/09.049.=0,>0>:A0=?30.:@=>0:1?30
?0.39:7:2D7410.D.70*00C;7:=0?30>0,=2@809?>49?30.:9?0C?:1?3027:-,7>084.:9/@.?:=74?3:2=,;3D
0<@4;809?49/@>?=D1=:84?>080=209.049?:,.=:>>9490/4>?49.??0.39:7:2D2090=,?4:9>*0
G9/>?=:9208;4=4.,7>@;;:=?1:=:@=1=,80B:=6
0DB:=/> 0DB:=/>
?0.39:7:24.,7.3,920G=>?8:A0=,/A,9?,20-@>490>>0.:>D>?08A0=?4.,749?02=,?4:9.:8;70809?>
>084.:9/@.?:=74?3:2=,;3D
4>.4;7490>4>.4;7490>
@>490>>/8494>?=,?4:9!,9,20809?,9/#;0=,?4:9>
'34>5:@=9,7,=?4.704>,A,47,-70,?&.3:7,=7D:88:9>3??;>=0;:>4?:=D@;0990/@828?+;,;0=>

Electronic copy available at: http://ssrn.com/abstract=1353582
We thank Phil Anderson, Ramon Casadesus-Masanell, Charles Galunic, Martin Gargiulo, Javier Gimeno, Rebecca
Henderson, Dan Levinthal, Bill Lazonick, Alessandro Lomi, Marwan Sinaceur, Jasjit Singh, Olav Sorenson, Peter
Zemsky and seminar participants at Dartmouth, Duke, Harvard and NYU for their helpful reactions. We especially
thank VLSI Research and Rebecca Henderson for their generosity in sharing their data with us.
Value Creation in Innovation Ecosystems: How the Structure of Technological
Interdependence Affects Firm Performance in New Technology Generations
Ron Adner
Tuck School of Business, Dartmouth College
Strategy and Management 100 Tuck Hall
Hanover, NH 03755, USA
Tel: 1 603 646 9185
Email: ron.adner@dartmouth.edu
Rahul Kapoor
Wharton School of Business, University of Pennsylvania
3620 Locust Walk, Suite 3025
Philadelphia, PA 19104
Tel: 215-898-6458
Email: kapoorr@wharton.upenn.edu
December, 2008
ABSTRACT
The success of an innovating firm often depends on the efforts of other innovators in its
environment. How do the challenges faced by external innovators affect the focal firm‟s
outcomes? To address this question we first characterize the external environment according to
the structure of interdependence. We follow the flow of inputs and outputs in the ecosystem to
distinguish between upstream components that are bundled by the focal firm, and downstream
complements that are bundled by the firm‟s customers. We argue that the effect of external
innovation challenges depends not only on their magnitude, but also on their location in the
ecosystem relative to the focal firm – whereas greater innovation challenges in components
enhances the benefits that accrue to technology leaders, greater innovation challenges in
complements erodes these benefits. We further argue that the effectiveness of vertical integration
as a strategy to manage ecosystem interdependence increases over the course of the technology
life cycle. We explore these arguments in the context of the global semiconductor lithography
industry from its emergence in 1962 to 2005 across nine distinct technology generations. We
find strong support for our arguments.

Electronic copy available at: http://ssrn.com/abstract=1353582
2
INTRODUCTION
A firm‟s competitive advantage depends on its ability to create more value than its rivals
(Porter, 1996; Brandenburger and Stuart, 1996). Greater value creation, in turn depends on the
firms‟ ability to innovate successfully. To capture the returns from innovation, many firms strive
to be technology leaders in their industry by being first to introduce new innovations to the
market. A given innovation, however, often does not stand alone; rather, it depends on
accompanying changes in the firm‟s environment for its own success (Adner, 2006). These
external changes, which require innovation on the part of other actors, embed the focal firm
within an ecosystem of interdependent innovations.
Consider, for example, Airbus‟s monumental investment in pioneering the super-jumbo
passenger aircraft with its A380 offer. Airbus, as the focal firm, faces significant challenges in
designing and manufacturing the core airframe of the airplane. Beyond this internal challenge, it
also relies on a host of suppliers for subassemblies and components. Some of these suppliers are
themselves confronted with significant innovation challenges to deliver components that meet
Airbus‟s requirements (e.g., engine; navigation system), while others will not need to innovate at
all (e.g., carpeting). Receiving these various components, Airbus faces the additional challenge
of integrating them with the airframe in order to deliver a functioning aircraft to its airline
customers. In order for the aircraft to be used productively by airlines, however, a number of
other actors in the environment, outside of Airbus‟s direct supply chain, confront additional
innovation challenges as well. Complementors such as airports need to invest and develop new
infrastructure to accommodate the oversized aircraft; regulators need to specify new safety
procedures; training simulator manufacturers need to develop new simulators on which aircraft
crews can be trained. The A380 innovation ecosystem is thus comprised not only of Airbus as

3
the core innovator, but also its upstream suppliers, and its downstream buyers and complementors,
all of whom need to resolve their own innovation challenges in order for the focal A380 offer to
create value.
Understanding competitive advantage in such „innovation ecosystems,‟ requires a change
in the way in which the strategy field has traditionally approached the relationship between a firm
and its external partners. Specifically, it requires an approach that extends beyond the focus on
how different actors will bargain over value capture (e.g., Porter, 1980; Brandenburger and Stuart,
1996) to include an explicit consideration of the innovation challenges that different actors will
need to overcome in order for value to be created in the first place. In this paper we present an
approach to analyzing the dynamics of value creation that focuses on the role of innovation
challenges in a firm‟s ecosystem as potential bottlenecks to value creation. We draw key
distinctions between challenges that need to be confronted by the focal firm, its suppliers, and its
complementors, and reveal differences in their impact on the firm‟s ability to create value, as well
as their implication for the firm‟s ability to capture value.
The ecosystem construct, as a way of making interdependencies more explicit, has gained
prominence in both business strategy (Moore, 1996; Iansiti and Levien, 2004; Adner, 2006) and
practice (e.g., Intel, 2004; SAP, 2006). These approaches have focused on understanding
coordination among partners in exchange networks that are characterized by simultaneous
cooperation and competition (Brandenburger and Nalebuff, 1997; Afuah, 2000). Studies in this
vein explore the challenges that arise when incentives across the ecosystem are not aligned
(Casadesus-Masanell and Yoffie, 2005), the role of established relationships with ecosystem
partners in shaping firm‟s motivations to compete for different market segments (Christensen and

Frequently asked questions

Q1. What are the contributions in "Value creation in innovation ecosystems: how the structure of technological interdependence affects firm performance in new technology generations" ?

The authors follow the flow of inputs and outputs in the ecosystem to distinguish between upstream components that are bundled by the focal firm, and downstream complements that are bundled by the firm‟s customers. The authors explore these arguments in the context of the global semiconductor lithography industry from its emergence in 1962 to 2005 across nine distinct technology generations. The authors further argue that the effectiveness of vertical integration as a strategy to manage ecosystem interdependence increases over the course of the technology life cycle. 

Q2. What are the future works mentioned in the paper "Value creation in innovation ecosystems: how the structure of technological interdependence affects firm performance in new technology generations" ?

Future studies could address these limitations by incorporating how actors coordinate, through mechanisms such as alliances, and relational contracts, within the context of innovation ecosystems. 

Q3. What is the main argument for the argument that a leader will make slower progress down the learning?

because lower rates of adoption will reduce the firm ability to gain experience (i.e., lower demand will lead to lower production quantities), the leader will make slower progress down the learning curve during its period of exclusivity. 

Q4. What is the main driver of progress in semiconductor manufacturing?

Improvements in lithography tools have been the main driver of progress in semiconductor manufacturing, enabling the production of higher performance chips at lower marginal cost (Moore, 1995). 

Q5. What is the key performance attribute of a lithography tool?

The key performance attribute of a lithography tool is its resolution, the smallest geometryon which the tool can focus on the wafer surface. 

Q6. What was the role of suppliers and complementors in the development of the IC?

Although suppliers and complementors were always critical to enabling each new technology generation, the magnitude of their challenges varied across generations – whereas in some generations it was possible to simply reuse existing elements, other generations required suppliers and complementors to completely reinvent their offers. 

Q7. What is the effect of a higher complement challenge on technology leaders?

When complements are open, such that they can be used with the offer of any firm, laggards will benefit from any progress that had been made in the market to advance the leader‟s offer; that is, spillovers will be higher and technology leaders will see their advantage eroded. 

Q8. What are the two mechanisms that stand out when interfaces are well specified?

Two mechanisms stand out when interfaces are well specified: first, with regard to the focal firm, there may be less potential for learning because key developments are undertaken by the supplier. 

Q9. What is the significance of the correlation between technology leadership and generation share?

The significant negative correlation between technology leadership (where lower values indicate earlier entry timing) and generation share suggests that early entry is rewarded with higher market share in the semiconductor lithography equipment market. 

Q10. What technologies attracted the interest, resources, and market share?

Amongst these, the two technologies that attracted the most interest, resources, and market share were X-ray and electron-beam (E-beam) lithography. 

Q11. Why does the fixed effects estimation require temporal variance within a panel?

In addition, because the fixed effects estimation requires temporal variance within a panel, it effectively excludes firms that participated in only one technology generation. 

Q12. What is the degree to which component challenges increase the performance advantage of technology leaders?

The degree to which component challenges increase the performance advantagetechnology leaders may be tempered by the extent of modularity between the component and the focal product.