Radical innovations from a dynamic capabilities point of view
A Hybrid-Modeling Approach to Generate Pseudo-E mpirical Data
Christian Weitert and Peter M. Milling
Industrieseminar, Mannheim University
Schloss, D-68131 Mannheim, Germany
phone: +49 621 181 1585
cweitert@ is.bwl.uni-mannheim.de
Keywords dynamic capabilities, radical innovation, agent-based modeling, innovation and compe-
tition, pseudo empirical data
Abstract:
Sustained competitive advantage is a major issue in the field of management research. A growing
number of scholars utilize the Dynamic Capabilities View as a reason for over average perfor-
mance and adaptability of a firm, especially to radical innovations that threaten a firm’s survival.
Due to the abstract character of the concept, the nature and impact of dynamic capabilities is still
vague and empirical evidence is rare. This paper presents a formal simulation model that builds on
previous work on the accumulation of dynamic capabilities to explore the micro foundations of the
concept. To generate pseudo empirical data, a mixed agent based and system dynamics modeling
approach is developed. The preliminary results suggest that further development of the method
promises to be fruitful to understand the micro foundations of dynamic capabilities and their role
for companies which face radical innovations.
Introduction
Neo classical microeconomic theory proposes above zero rents can not be achieved in a competi-
tive market, especially not in the long term (Jacobson 1992). Still companies like General Electrics
or Hewlett Packard managed to sustain not only positive but above average profits over long pe-
tiods of time. Building upon the work of other strategy schools as the market or the resource based
view, a lot of recent management research addresses to the dynamic capabilities framework in or-
der to explain sustainable competitive advantage. While the market based view seeks for an expla-
nation in the company’s competitive environment, concerning industry structure and a firm’s stra-
1
tegic market position as well as the build up of market entry barriers (Porter 1980), the resource
based view stresses the importance of the company’s internal capabilities and resources that allow a
company to perform better than its competitors (Barney 1991; Henderson and Cockburn 1994;
Wenrnerfelt 1984)
Both perspectives fail to explain sufficiently how those barriers can be held up over long
periods of time and in dynamic contexts. Critics address the limits of both perspectives concerning
this static nature that does not account for change, especially in dynamic environments as growing
markets characterized by rapidly increasing returns (Eisenhardt and Martin 2000). In this context
the implications of such a static approach can be very misleading (Merten 1991; Sterman et al.
2007). The dynamic capabilities view, though still based on the resource perspective, accounts for
the mentioned limitations. Still valid in dynamic environments as high-technology industries, the
perspective states that the competitive performance of a company is based on its ability to adapt to
a quickly changing surrounding by reallocation and recombination of its resources — rather than just
on the resources themselves (Teece, Pisano, and Shuen 1997).
Closely related to the questions raised by some companies’ sustained success is the one
asked by Christensen and his concept of disruptive innovations: why do great companies struggle
or even fail when facing rapid technological change? (Christensen 1997). A company like Kodak
was very successful over some time and managed to survive strong competition — mainly based on
a strategy of innovation. Y et a technological innovation, in the case of Kodak digital photography,
led to a severe crisis the company still seeks to recover from. This raises the question what distin-
guishes firms like Kodak, that were able to yield above normal returms on the mid range from the
one sustaining their competitive advantage on the long run as General Electrics. According to the
dynamic capabilities perspective the latter companies possess higher level capabilities enabling
them to modify their operational capabilities and build new ones faster in the face of a changing
competitive landscape (Eisenhardt and Martin 2000; Teece, Pisano, and Shuen 1997).
Literature review and hypothesis development
Companies that have survived an industry shake-out process have adapted to the market setting
through a trial and error process, analogue to the organizational learning described by March
(1994). The history of trial and error is closely linked to the exploitation of an existing technology.
The very question in this context is, why do these firms that have survived harsh competition, ac-
complished to adapt to a dynamic environment, created dynamic capabilities (Stamboulis, Ada-
mides, and Malakis 2002) and emerged as dominant competitors loose their position or even fail at
a high rate when facing another phase of discontinuous technological change (Barnett and Pontikes
2
2006; Spencer, Kirchhoff, and White 2008). There are numerous examples of market leaders like
Kodak that, though dominant with classic lithographic photography, failed to play a similar role
within the following market.
Henderson and Clark (1990) identify a difference in the organizational capabilities required
by a company facing incremental or radical innovations as a cause for failure of former leaders. A
radical innovation destroys the usefulness of an established company’s specific knowledge and
organizational structure. The characteristic of initial inferiority or unrealized potential of a discon-
tinuous innovation makes it especially hard for successful companies to recognize its destructive
power and to react in time. Like a competency trap in organizational learning, processes and rou-
tines remain unchanged because of their success, thereby hindering an exploration of new alterna-
tives (March 1991, 1994). In the context of initially inferior but eventually superior technologies
this has severe consequences for the corporation.
Barnett and Pontikes (2006) find empirical evidence for their hypothesis that a history of
survived fierce competition in one context makes moving into a new context even more hazardous
in data from the U.S. computer history. In concurrence with these results a simulation model based
on the work of Tushman and Romanelli (1985) shows that the structure enabling a companies
competence and success in one context, causes its inertia and reduced capability to adapt to a radi-
cally changed environment (Sastry 1997). Rahmanadad (2007) finds that one reason for this is the
prevailing myopic allocation of investment and resources to operational capabilities instead of
building up dynamic capabilities. This effect is amplified by strong competition and the delayed
impact on performance of investment in higher order capabilities. Looking at these insights from a
dynamic capabilities point of view, a first hypothesis is proposed:
Hypothesis 1: Adapting to a specific context efficiently leads to gradual erosion of an
organization ’s dynamic capability over time, impeding adaptation to a radically new
context.
Following the above findings, a firm’s history has a dominant influence on its future. Or-
ganizational ecologists therefore argue that strong path dependency limits the strategic options for
the actual management, if there are any, to the very early stages of a firm’s history (Carroll and
Hannan 2000). Considering this, dynamic capabilities can not be build up but only born during the
foundation of a company (Winter 2003). Augier and Teece (2008) do not reject this completely, but
argue for a definition of strategy as evolution with design. A company’s path therefore is not com-
pletely bound to prior decisions and investments and can be influenced by its management. This is
supported by examples of companies such as IBM, which were able to respond to a crisis caused by
a fundamental change of their environment — in this case the transition of the computer market
from mainframes to personal computers (Sterman 2000) — through a new orientation, hardly root-
ing in the creation of the organization. Rigid mental models of experienced and formerly successful
3
managers are responsible for a lack of adaptation rather than a general impossibility of actively
managed change (Weil 2004).
Facing the challenge of a radical innovation, researchers recently have identified the need for
an ambidexterity of a firm’s organization, enabling for excellence in explorative as well as in ex-
ploitative tasks (Assink 2006 ; O’Reilly and Tushman 2007). Inertia is useful to complete the day-
to-day tasks of an organization efficiently but a firm needs the ability to overcome this inertia when
adaptation is needed. Feldman and Pentland (2003) challenge the traditional understanding of or-
ganizational routines as creating inertia in organizations, arguing that organizational routines can
be a source of change as well as stability. This is in line with Winter’s (Winter 2003) definition of
dynamic capabilities as ‘high level routines’ that govern the change of ordinary capabilities. One
example of a company overcoming its inertia for a limited period of time, without seriously damag-
ing the basis of daily tasks, is the ‘Innovation Jam’ IBM has established to make use of their em-
ployees ideas and creativity. For a few days all employees are encouraged to discuss on strategic
topics, highly relevant for the company’s future (Bjelland and Wood 2008). While the ‘Innovation
Jam’ origins from a dynamic capability it is not necessarily one in itself. It reflects an increase of
IBMs organizational capability, a change of the organizational structure enabling employees to be
more creative outside of their day-to-day business, yet without reducing the efficiency of their rou-
tines. Based upon this argumentation a second hypothesis is proposed:
Hypothesis 2: Dynamic capabilities enable an organization to overcome its inertia,
when necessary.
Research Design
Due to the conceptional nature of the perspective it has so far proven a tough task to empirically
determine the nature of dynamic capabilities or examine their impact on firm performance. Never-
theless the framework enables researchers as well as managers to look at the ability of an organiza-
tion to adapt and to stay alert in the face of a changing environment. In addition the generic charac-
ter allows applying the concept globally, not limited to specific examples. This seems reasonable,
since strategically significant capabilities of an organization are by definition hard to describe or
measure (Barney 1991; Nelson 1991), otherwise they could not be an enduring source of competi-
tive advantage (Henderson and Cockbum 1994).
The hypotheses postulated in this paper share the abstract character of the approach since it
is one aim of the further research to clarify some theoretical aspects in the understanding of dynam-
ic capabilities. Therefore a system dynamics model — that accounts for conceptional factors and
allows examining their dynamic interaction — helps to understand and discuss the theoretical cha-
racter of dynamic capabilities as well as their impact on firm performance. As every formal model
4
is a set of hypotheses, the structure of a system dynamics model is a collection of structural hypo-
thesis from which dynamic behavior emerges. Even though single hypotheses might seem reasona-
ble and fairly simple, their interplay and connection is complex enough to make prediction of be-
havior a tough task. The simulation of a formal model helps in the understanding of its structure
and above that makes the hypotheses implied in it testable and available for validation. A bove that,
simulation experiment enable for an examination of the impact of dynamic capabilities hardly
available in real world setting — due to the abstract character of the framework. To test the hypo-
theses, we built a generic simulation model, whose structure is based on previous work on the na-
ture and the accumulation of dynamic capabilities, innovation literature and several system dynam-
ics studies. The simulation contains several heterogeneous firms competing in a high technology
market, allowing for entry of new competitors and exit of incumbents.
Presentation of Model Structure
A model addressed to a similar question able to generate this behavior and explain the generic dy-
namics of such a market is presented by Weil and Utterback (Weil and Utterback 2005). Their ap-
proach is an aggregated view on the market, which focuses on the macrostructure of dynamic in-
dustries. The presented model mixes an agent-based and a system dynamics approach to explore
the micro-foundations of dynamic capabilities. In cases where empirical data is hard to obtain or
the research objective is as abstract as dynamic capabilities, agent-based modeling has proven use-
ful in order to create pseudo-empirical data (Axelrod 2006; Weitert 2007).
Market
De d Financial S
oe | Ree | Operations
Capability Product developement. Production
accumulation
Figure 1: Structural Overview of the Model
5
The internal decisions and processes of each firm are modeled in continuous manner as a stock-
flow diagram, while each firms acts as an individual agent on the market. Figure 1 illustrates the
modularized structure of the Model.
The model is divided into 5 main modules:
the market place, where demand is allocated due to relative performance of the firm, and
where all agents compete for the market potential
a financial controlling monitor where each firm’s operative result, its survival and invest-
ment is determined
the capability sector which controls each company’s accumulation of dynamic and opera-
tional capabilities
the product development module, representing the R&D activities of each agent
a production sector, describing the value creation processes of each firm
A
Ne
*
polity ini customers
price sensitivity | Market
<unit price> Potential e
transition.
rice effect
P tecmoogy = | SY) eT
technology a8 ee
ALPHA
<fims inthe Bot {nl
market> attractivity
Prod Introduction
Switch
<average prod Market Share
pee nr ders
4
Figure 2: The market module
Market Sector
Since the demand allocation process is not under research — the main task of this sector is to make
the performance impact of different capability levels among agents comparable — its incorporation
6
in the model is rather simple. The individual market share is a function of each firm’s relative
product attractiveness, which result from its price compared to the average market price and its
relative technological performance. The transition of the market potential in between product gen-
erations is portrayed as a simple bass diffusion process (Thun, Grofler, and Milling 2000). Figure 2
shows the stock and flow structure of the simplified market process.
Financial Controlling
This part of the model calculates the operating figures of each company and governs market entry
and exit due to bankruptcy. The entry rate is exogenous at this stage of the modeling process,
though entry based on market growth and intensity of competition — as proposed by Weil and Ut-
terback (2005) — will be considered for future versions of the model. Upon market entry each firm
receives an initial start-up capital to start its operations. From this point on the company’s capital is
determined by the operating result, which is a function of its turnover from sales reduced through
cost of operations and investment in R&D and capability accumulation. The unit price is defined
through a mark-up on the individual’s variable unit costs. Additional to the mere calculation of
these management ratios the market exit of a firm is dependent on its financial resources. If the
stock of Capital decreases to zero, the agent is considered as insolvent. Figure 3 illustrates the
structure of this part of the model.
fimns in the market.
financial survival
INI CAP AVAIL
available -———— Capital
ele financing <6 invest>
( a
unitcosts <capability invest>
fixed cost
effect of oc on
unit costs
<OC>
Figure 3: Stock and flow structure of the financial model sector
Capability accumulation
As mentioned above, dynamic capabilities are often described as higher level capabilities enabling
a firm to modify its operational capabilities and build new ones faster in the face of a changing
competitive landscape (Eisenhardt and Martin 2000; Teece, Pisano, and Shuen 1997). Winter
(2003) adds that “[...] dynamic capabilities govern the rate of change of ordinary capabilities.”
Therefore the efficiency of the accumulation process of organizational capabilities in the model is
influenced by the level of each firm’s dynamic capabilities. Since generation of organizational ca-
pabilities are said to result from the repeated execution of specific routines and learning as well as
financial investment it seems reasonable to model them as stocks (Chirico and Salvato 2008; Rah-
mandad and Repenning 2008). Processed R&D development tasks — product development is con-
sidered as an example for a dynamic capability (Winter 2003) — are one input factor for the routine
generation and accumulation of DC. In contrast operational capabilities evolve from the firm’s day-
to-day business and are therefore increased by the company’s experience with the production
process.
onhel O
wo tion>
cap om invest a, i
oc whe
va invest Vl
oc wa | J
efficiency mA pe 0c
DC efficiency
muliplikator
<DC>
INIOC
INIDC
inflow DC atrophy DC
De nest DC routine
me generation
ff invest
<developement>
Capability
capability invest 2
fixed ——h Z fraction Depreciation Time
ae)
Figure 4: The capability accumulation process
As illustrated by Figure 4 the capability accumulation process is not only dependent on the organi-
zational routine generation, but the direct investment of resources. For simplification reasons only
the allocation of financial resources is considered. The dependence of capability generation on
endogenous resources acquired through the firm’s operations incorporates the positive feedback of
the accumulation process in the model.
Production and product development
The organizational capabilities affect both product development and operational performance. The
two operational modules are displayed by Figure 5 and 6. While the product
<tumover>
7
‘r&d invest fraction
r&d efficiency
>)
‘r&d invest,
multiplicator
r&d effort
Remaining
Potential Tech
| ——+» developement
INITIAL
Ne
POTENTIAL
<DC>
Level of
Tech
experience with
tech
/
Figure 5: The stock and flow formulation of the R&D process
development is a basic structure to generate the s-shaped growth of technology exploitation
processes (Foster 1986), the production process is modeled as a simple co-flow to generate the
product performance attribute that determines the relative product attractiveness.
<OC> QC Capacity
Multiplyer
<Onlers> capacity
Products in \ a)
production |_Stock sales
financial survival>
ofTech> average prod perf
product
raise in [performance Sua? of
performance performance
NS /
Figure 6: Stock and flow diagram of the production process
Preliminary Results and further Development
To generate pseudo empirical data, all firm specific variables are subscripted to allow the structural
reproduction of the internal processes for all companies. Exogenous input parameters like the in-
vestment fraction in dynamic capabilities, market entry time, and technology transition time were
altered with random numbers to generate a large artificial sample (10000 simulation runs with a
maximum number of 34 firms in the market). Considering firm performance investment in dynamic
capabilities has a significant positive effect. The optimal investment fraction depends on the inten-
sity of competition with product one and the point of time when new firms enter the market with
the following product generation. Over all simulated scenarios an average investment fraction of
0.24 of total invested financial resources to dynamic capabilities seems optimal.
Despite these preliminary results, the quality and validity of the data output is so far not
sufficient to comprehensively test the hypotheses postulated above. Therefore further development
of the model structure and statistical analysis is necessary. Nevertheless the hybrid modeling ap-
proach promises to be fruitful for the further understanding of the micro foundations of the concept
of dynamic capabilities.
10
Bibliography
Assink, Mamix. 2006 Inhibitors of Disruptive Innovation Capability: a Conceptual Model. Euro-
pean Journal of Innovation Management 9 (2):215 - 233.
Augier, Mie, and David J. Teece. 2008. Strategy as Evolution with Design: The Foundations of
Dynamic Capabilities and the Role of Managers in the Economic System. Organization
Studies 29 (8-9):1187-1208.
Axelrod, Robert. 2006. Advancing the Art of Simulation in the Social Sciences. In Handbook of
Research on Nature Inspired Computing for Economy and Management, edited by J.-P.
Rennard. Hersey, PA: Idea Group.
Barnett, William P., and Elizabeth G. Pontikes. 2006. The Red Queen, Success Bias, and Organiza-
tional Inertia. Stanford Graduate School of Business: Stanford Business School.
Barney, Jay. 1991. Firm Resources and Sustained Competitive Advantage. J ournal of Management
17 (1):99.
Bjelland, Osvald M., and Robert Chapman Wood. 2008. An Inside View of IBM’s ‘Innovation
Jam’. MIT Sloan Management Review 49 (1):32-40.
Carroll, Glenn, and Michael Hannan. 2000. The Demography of Corporations and Industries 1ed.
Vol. 1. Princeton, NJ: Princeton University Press.
Chirico, Francesco, and Carlo Salvato. 2008. Knowledge Integration and Dynamic Organizational
Adaptation in Family Firms. Family Business Review 21 (2):169-181.
Christensen, Clayton M. 1997. The Innovator’s Dilenma: When New Technology Causes Great
Firms to Fail. Boston, MA.: Harvard Business School Press.
Eisenhardt, Kathleen M., and Jeffrey A. Martin. 2000. Dynamic Capabilities: What are they? .
Strategic Management J ournal 21 (10/11):1105.
Feldman, Martha S., and Brian T. Pentland. 2003. Reconceptualizing Organizational Routines as a
Source of Flexibility and Change. Administrative Science Quarterly 48 (1):94-118.
Foster, Richard. 1986. Innovation - The Attacker ‘s Advantage. New Y ork: Summit Books.
Henderson, Rebecca, and Iain Cockburn. 1994. Measuring Competence? Exploring Firm Effects in
Pharmaceutical Research. Strategic Management Journal 15:63-84.
Henderson, Rebecca M., and Kim B. Clark. 1990. Architectural Innovation: The Reconfiguration
of Existing Product Technologies and the Failure of Established Firms. Administrative
Science Quarterly 35 (1):9-30.
Jacobson, Robert. 1992. The “Austrian” School of Strategy. Academy of Management Review 17
(4):782-807.
March, James G. 1991. Exploration and exploitation in organizational leaning. Organization
Science 2 (1):71-87.
—. 1994. A primer on decision making: How decisions happen. New Y ork.
Merten, Peter P. 1991. Loop-Based Strategic Decision Support Systems. Strategic Management
Journal 12 (5):371-386.
Nelson, Richard R. 1991. Why do Firms differ and how does it matter? Strategic Management
Journal 12:61-74.
O’Reilly, Charles A. III, and Michael L. Tushman. 2007. Ambidexterity as a Dynamic Capability:
Resolving the Innovator’s Dilemma. Harvard Business School.
11
Porter, Michael E. 1980. Competitive strategy. 1 ed. Vol. 1. New Y ork: Free Press.
Rahmandad, Hazhir. 2007. Why myopic policies persist? Impact of growth opportunities and com-
petition. 25th International Conference of the System Dynamics Society, at Boston.
Rahmandad, Hazhir, and Nelson Repenning. 2008. The Dynamics of Capability Development and
Erosion. Virginia Tech.
Sastry, M. Anjali. 1997. Problems and Paradoxes in a Model of Punctuated Organizational Change.
Administrative Science Quarterly 42 (2):237-275.
Spencer, Aron S., Bruce A. Kirchhoff, and Craig White. 2008. Entrepreneurship, Innovation, and
Wealth Distribution: The Essence of Creative Destruction. International Small Business
Journal 26 (1):9-26.
Stamboulis, Yeoryios A., Emmanuel D. Adamides, and Thomas E. Malakis. 2002. Modeling the
Product-Process R&D Dynamics. Proceedings of the 20th International Conference of the
System Dynamics Society, July 28 - August 1, at Palermo, Italy.
Sterman, John D. 2000. Business Dynamics : Systems Thinking and Modeling for a Complex
World. Boston: Irwin/McGraw-Hill.
Sterman, John D., Rebecca Henderson, Eric D. Beinhocker, and Lee I. Newman. 2007. Getting Big
Too Fast: Strategic Dynamics with Increasing Returns and Bounded Rationality. Manage-
ment Science 53 (4):683-696.
Teece, David J., Gary Pisano, and Amy Shuen. 1997. Dynamic Capabilities and Strategic Man-
agement. Strategic Management Journal 18 (7):509-533.
Thun, Jorn-Henrik, Andreas Grofler, and Peter M. Milling. 2000. The Diffusion of Goods Consi-
dering Network Externalities:
A System Dynamics-Based Approach. Proceedings of the 18th International Conference of the
System Dynamics Society, at Bergen.
Tushman, Michael L., and Elaine Romanelli. 1985. Organizational evolution: A metamorphosis
model of convergence and reorientation. In Research in organizational behavior, 7, edited
by L. L. Cummings and B. M. Staw. Greenwich, CT: JAI Press.
Weil, Henry B. 2004. Disrupting Mature Markets with Innovative Technology. 22nd International
Conference of the System Dynamics Society, July 25-29, at Oxford, England.
Weil, Henry Birdseye, and James M. Utterback. 2005. The Dynamics of Innovative Industries.
Proceedings of the 23rd International Conference of the System Dynamics Society, at Bos-
ton.
Weitert, Christian. 2007. Analysis of Investor Behavior in an Artificial Stock Market. Proceedings
of the 25th Intemational Conference of the System Dynamics Society, at Boston.
Wererfelt, Birger. 1984. A Resource-based View of the Firm. Strategic Management Journal 5
(2):171-180.
Winter, Sidney G. 2003. Understanding Dynamic Capabilities. Strategic Management Journal 24
(10):991-995.
12
