Content uploaded by Dieter Franz Kogler
Author content
All content in this area was uploaded by Dieter Franz Kogler
Content may be subject to copyright.
June 9, 2008
ca. 16,000 words
Social Foundations of Regional Innovation and
the Role of University Spin-offs
by
Harald Bathelt
University of Toronto, Department of Political Science and Department of Geography,
Sidney Smith Hall, 100 St. George Street, Toronto ON M5S 3G3, Canada,
E-mail: harald.bathelt@utoronto.ca, URL: http://www.harald-bathelt.com
Dieter F. Kogler
University of Toronto, Department of Geography,
Sidney Smith Hall, 100 St. George Street, Toronto ON M5S 3G3, Canada,
E-mail: dieter.kogler@utoronto.ca
Andrew K. Munro
University of Toronto, Institute for the History and Philosophy of Science and Technology,
Victoria College, 91 Charles Street West, Toronto ON M5S 1K7,
E-mail: andrew.munro@utoronto.ca
- 2 -
Social Foundations of Regional Innovation and the Role of University Spin-
offs
Abstract (ca. 235 words): Drawing from the literature on the role of universities in promoting
technology transfer, this paper will develop a regional conceptualization of spin-off processes,
and apply it to a regional case study. In doing this, a typology of spin-off firms will be explored,
which is based on the following variables: university sponsorship, university involvement in firm
formation, character of knowledge applied, and co-localization of the founders. This enables us
to analyze the wider impact of universities on technology transfer and regional development.
Extending propositions of organizational ecology, we argue that start-up processes and intra-firm
adaptations are not competing against one another for superiority in regional growth or selection
processes. The argument is developed that new and existing firms can complement one another
in a regional context if they succeed in both developing wider regional networks and trans-
regional linkages. Our study will focus on the Kitchener and Guelph metropolitan areas about
100 km west of Toronto, sometimes referred to as Canada’s Technology Triangle (CTT), where a
larger number of firms related to information technology (IT) have been successfully launched
since the 1970s around the activities of the University of Waterloo. This research will investigate
to which degree different groups of university spin-off and start-up firms have established
regional producer–user networks, to which degree they have developed, and depend upon, global
pipelines, and whether this dynamic has produced, or will produce, spillovers to other regional
industries.
Keywords: University spin-offs, University-related start-ups, Technology transfer, Regional
networks, Regional innovation, Waterloo, Canada’s Technology Triangle
JEL Classifications: D83, L24, L26, M13, O31
- 3 -
1. Introduction
The goal of this paper is to explore the social foundations of regional innovation,
and the underlying processes which drive innovation. In particular, we are interested in
the role of spin-off processes vs. intra-firm restructuring in guiding technological change
at the regional level. This will be investigated for the Kitchener and Guelph metropolitan
areas
1
, about 100 km west of Toronto, around which the initiative “Canada’s Technology
Triangle” (CTT) was founded in the late 1980s (Cities of Cambridge, Guelph, Kitchener,
and Waterloo 1988).
2
Since the 1970s, this region has experienced the impetus of spin-
off processes from university research. Particularly around the activities of the
University of Waterloo, numerous firms in the area of information technology (IT), such
as Dalsa, Open Text, Research in Motion (RIM),
3
Sybase, or Waterloo Maple, have been
1
In media studies and reports, these municipalities are often referred to as the Waterloo region, in
part because of the important role of the University of Waterloo as a regional driver of technologies,
provider of high-quality technical skills, and generator of start-up firms.
2
CTT was jointly established by the four cities of Cambridge, Guelph, Kitchener, and Waterloo to
market the region’s technological strengths and reduce inter-municipal competition (Bathelt and Hecht
1990). The idea was that all communities would benefit from the attraction of new firms through
additional jobs and incomes (Black 1988; Chevreau 1988). Although Guelph left this initiative in the
meanwhile to market its strengths individually, the local economies are still linked to one another, draw
from a shared labour market, and depend on similar economic conditions. Therefore, the whole region was
of general interest in this study.
3
RIM is Waterloo’s best known start-up success in the IT field. As founder and co-CEO Mike
Lazaridis emphasized many times, RIM is not a classical spin-off from university research. In fact,
Lazaridis who was a student at the University of Waterloo even dropped out before receiving his Bachelor
of Engineering to form RIM as a consulting firm in 1984, doing contract projects (Colapinto 2007; Wahl
- 4 -
successfully launched, altogether establishing a growing technology basis in the region
(Bathelt and Hecht 1990; Bathelt 1991a; Parker 2001; Bramwell et al. 2004; 2008;
Colapinto 2007).
4
Due to this dynamic development, the region received a wide interest
in academic studies to explore the reasons behind the growth process. This has produced
a regional success story of high-technology growth and university spin-off processes.
The region also received a lot of attention by policy makers because it has been
able to shift its economic focus from traditional industries to new IT-related businesses.
The regional economy achieved above-average performance levels, according to
indicators such as job growth, unemployment rate, or average household income as
indicated in Table 1. Between 2001 and 2006, the Kitchener CMA and the Guelph CA,
for instance, experienced an increase in population and jobs that was significantly higher
than the national and provincial growth rates, and similar to the one in the Toronto CMA.
The unemployment rates were among the lowest in Canadian metropolitan areas, while
average household income was about average in Ontario. In terms of employment
2007). The first contract was for General Motors. Early on, Lazaridis and his co-founder Fregin began
providing wireless point-of-sale systems. The firm developed its own radio transceivers and soon
recognized the value of moving into the area of wireless technology. Here, they later began producing the
line of products that is now known as the “Blackberry”. In terms of the firm’s relationship to the
University of Waterloo, it appears relatively week. Lazaridis repeatedly pointed out that the flow of new
graduates would be the most important source of technology transfer from the local universities to RIM
(Brady 2004; Livingston 2007; Fielding 2008).
4
The University of Guelph has generated fewer spin-offs than the University of Waterloo. In
addition, most of these spin-offs are situated in the agribusiness, animal health, and molecular design
technology fields, which are the university’s research focus (see http://www.uoguelph.ca/research.php
),
date accessed June 2, 2007).
- 5 -
growth, it is notable that both, the Kitchener CMA and the Guelph CA, expanded their
job base considerably in professional scientific and technical services (a 24% and 19%
increase compared to only 14%, 10% and 8.5% growth in Canada, Ontario and the
Toronto CMA, respectively). Job growth was also substantial in a wide range of
producer-related services such as finance and insurance (30% increase in the Kitchener
CMA), administrative and support, and other services (36% and 42% increases), and
educational services (20% and 9% increases). A 20% and 14.5% job increase in
construction in the Kitchener and Guelph metropolitan areas, respectively, also indicates
the dynamic nature of growth in the regional economy.
***********************************
Insert Table 1 about here
***********************************
Around the spin-off and start-up activities of the University of Waterloo, a myth of
dynamic technology growth has been created and actively pushed forward by regional
business organizations, such as Communitech
5
and the CTT initiative. This is also
reflected in a recent economic study of the region by BMO Capital Markets (2008),
which suggests that the region’s strengths will fuel further business creation in the future.
In the years 2006 and 2007 (although not in 2008), Waterloo was furthermore identified
5
Communitech is an industry-led organization that was started up by 40 founding companies in
1997 with the goal to support the region’s technology basis. This followed an earlier initiative by
technology pioneers in the region, known as the Atlas Group. Currently the organization has more than
450 members including large firms, such as RIM and Open Text, but also many small firms. In addition,
investors, service firms, educational institutions and government agencies are members of Communitech
and support the initiative (see http://www.communitech.ca
, date accessed May 5, 2008)
- 6 -
as one of the world’s top seven “intelligent communities” by the Intelligent Community
Forum (2007). Even though this organization claims that this evaluation is related to
indicators such as intellectual property (IP), engagement of local businesses, citizens and
government, as well as collaboration patterns and institutional support, this ranking
should be treated with caution, and primarily be seen as a promotion tool.
At closer investigation, knowledge behind the successes in economic development
and the social foundations of regional innovation appears still fairly limited. Clearly, the
region cannot be viewed as a true regional industry cluster of closely interrelated firms of
a particular value chain, and their supplier and service infrastructure (Bathelt 1991a).
What we find instead is a highly heterogeneous and segmented industry structure with
businesses which are characterized by limited commonalities. The region hosts a
substantial variety of larger and smaller establishments, old and young firms, and
businesses with diverse old and new manufacturing and service background. There is no
simple explanation for the overall success of these different economic segments. Aside
from this, we have to take into consideration that there are also processes of economic
restructuring in traditional sectors which involve social conflicts and processes of
crowding out firms. Presently, media reports, for instance, indicate the beginning of an
economic downswing in the automobile supplier industry, as the examples of Kitchener
Frame Ltd. and Linamar Corp. in Guelph suggest (Van Alphen 2008; Toronto Star 2008).
This could have wider regional effects as the labor market is quite dependent on the
automobile industry (Rutherford and Holmes 2008). The example of NCR’s job cuts in
2007 further suggests that restructuring processes due to global reorganization are not
only related to traditional industries but also affect the IT sector (Toronto Star 2007).
- 7 -
It is unclear how the region will deal with these challenges, and whether it will
continue to have as much economic success as before. In this context, this paper will
particularly explore the role of university spin-off firms in the region. We chose the IT
sector for this study as start-ups in this industry segment are likely to become part of
wider production and research networks, and possibly develop into drivers of regional
innovation within and across different sectors. Drawing from the literature on the role of
universities in promoting technology transfer, this paper develops a typology of
university spin-offs based on several variables: type of university sponsorship, university
involvement in firm formation, type of university knowledge applied, and co-localization
of the founders. This will enable us to analyze the wider impact of universities on
technology transfer and regional development.
This paper will be structured as follows: In section 2, we present our conceptual
framework revolving around ideas from organizational ecology and the role of spin-off
firms. Extending common views in organizational ecology, we argue that start-ups in a
regional context have the potential not only to trigger technological change themselves,
but support and guide restructuring and modernization within established firms through
production and research networks. Section 3 summarizes some of the results of former
research in the region, and section 4 discusses the research approach and methodology
applied. In the empirical part of the paper, the nature of start-up and spin-off processes
(section 5), local and non-local social relationships in innovation, and the role of
institutional support in these processes (section 6) will be investigated. At the end,
section 7 will summarize the main findings and draw some conclusions regarding the
effects of IT-related university start-ups and spin-offs on regional innovation.
- 8 -
2. Organizational Ecology and the Role of University Spin-offs
In the social science literature, spin-off and start-up processes related to universities
have been studied intensively since the 1960s, related to successful examples such as
Silicon Valley (e.g. Cooper 1971; Rogers and Larsen 1983; Saxenian 1985) and the
Route 128 region (e.g. Roberts 1968; Keune and Nathusius 1977; de Jong 1987; Bathelt
and Glückler 2003). In these regions, university spin-off processes were important
drivers of technological change and provided a trigger to regional development. In
Boston’s Route 128 region, for example, 156 technology-oriented firms were identified
in the post-World War II period prior to 1965, which had spun off from the
Massachusetts Institute of Technology and its research laboratories (Roberts 1968). A
similar success story has been reported in the Waterloo region and a connection has been
made between regional technological change and spin-off processes from the University
of Waterloo (Bathelt 1991a; Bramwell and Wolfe 2008). An early study found that at
least 50 of 68 firms, which had been identified as “university spin-offs” by the University
of Waterloo in 1988, were established within the Waterloo region (Bathelt and Hecht
1990). Although it was somewhat unclear how spin-off firms were exactly defined at
that time by the University of Waterloo, these firms seemingly contributed to a dynamic
growth process. In 1988, most university spin-offs were small with about 60% having
ten employees or less; all 39 firms interviewed had created a total of 545 jobs (14 jobs
per firms), most of which in the region. The study estimated that about 20% of all spin-
offs had little business success and did not survive until 1988 (Bathelt and Hecht 1990).
- 9 -
2.1 Regional technological change as a selection process
A conception which powerfully describes and emphasizes the role of young start-
ups and spin-offs in organizational and technological change is that of organizational
ecology (Hannan and Freeman 1977; 1984; 1993). This conception explains changes in
organizational forms as an evolutionary process and claims that this change primarily
results from selection processes among organizations, rather than from adjustments
within organizations. Firms are viewed as not being easily structurally adaptable
(Hannan and Freeman 1977). A reason for this would be that decisions are made at a
consensual basis tending toward suboptimal solutions. Adaptations would also be slow
leading to structural inertia. In addition, it would be difficult to identify the most
efficient adaptations due to uncertainties.
As a consequence, it has been concluded that firms themselves become the object
of selection within their organizational ecologies. The selection process is assumed to
depend not only on economic efficiency of the firms, but also on reliability and
accountability, which result in legitimacy. In other words, this conception suggests that
firms which produce reliable high-quality outputs, and are easily accountable for their
produced goods will be more likely to gain a broad customer base and have more market
success. Hannan and Freeman (1984; 1993) argue that although selection processes
prioritize firms with high reliability and accountability, i.e. well-established and capable
of constantly reproducing their own structure, new firms would be in a favorable position
in periods of organizational and technological change. Although new firms are
characterized by high exit rates, some have a greater robustness and potential, and will
survive, while established firms with outdated structures tend to disappear from the
market. As structural inertia increases with size and age, processes of adaptation in
- 10 -
existing organizations become more difficult over time. Progress in major restructuring
processes would likely be slow, involve high switching costs, threaten reliability, and
reduce legitimacy due to internal tensions (Kieser and Woywode 1999). In the end, it
would be primarily new firms that drive development trajectories. This conception has
been applied in the density-dependence model to demonstrate how firm formation
processes drive regional development, based on technological opportunities and
legitimacy (Baum and Oliver 1992; Staber 1997).
Although this approach is important in pointing out the key roles of new start-ups
in processes of technological change, it has also been criticized for its neglect of (i) the
role of agency, and (ii) the significance of permanent adjustment and higher-level
learning processes in reacting to changes in the economic environment (for a summary,
see Kieser and Woywode 1999; Bathelt and Glückler 2003). Regional firms are hardly
comparable to a biological population with shared genetic code. Empirical realities also
show that many sectors are dominated by large firms which have existed and prospered
over a long time period. These have been able to adapt new structures in ongoing
learning processes, and create their own regional environments (Storper and Walker
1989; Lundvall and Johnson 1994; Gertler 2004). Still, the importance of technology-
based start-ups in processes of technological change cannot be ignored. In this context,
we have to particularly emphasize the importance and potential of technology-based
university spin-offs (e.g. Malecki 1991; Hayter 1997). These new firms have the
potential to become triggers of technological change as they are close to and develop
their ideas for new products and services from basic and applied research, not primarily
driven by returns-to-investment considerations. The potential to develop new
technologies is likely greatest if the universities or research facilities are specialized in
- 11 -
particular science and technology fields (Chiesa and Piccaluga 2000), as in the case of the
University of Waterloo. Aside from economic efficiency, the development of reliability
and accountability is, however, a major obstacle for these spin-off firms. The firms are
typically founded by faculty members or graduate students with little experience in
markets, setting up production, and establishing routines. This may, in turn, reduce their
overall performance and regional impact, and limit their chance of survival and success
(e.g. Stankiewicz 1994).
2.2 A combined model of regional selection and adaptation
In an attempt to combine the high technological potential of new organizations with
the effects of ongoing incremental learning processes in existing organizations, we
develop an argument which reduces the risk of deterministic interpretations of
organizational ecology.
6
In applying a spatial perspective, we argue that technology-
based spin-off processes can have a great potential in becoming drivers of regional
change in two aspects: Similar to the ideas expressed in density-dependence theory (e.g.
Baum and Oliver 1992), the spin-offs themselves can develop into a new basis for
regional economic growth and trigger organizational and technological change in the first
stage. This is, of course, highly dependent upon their legitimacy, and whether they are
able to develop reliable and accountable structures. As this seems to be a challenge for
6
Interestingly, the starting point of Hannan and Freeman (1977) is not that different from our
model in that they assume a combined process of corporate reorganization and firm formation. They are
very critical though of the dominant model corporate adaptation at that time, and thus focus on the role of
new firms in organizational and technological change.
- 12 -
university spin-offs (Stuart et al. 1999; Vohora et al. 2004) we can expect that the overall
effects of technology-based start-ups are initially relatively small. This stage can take a
longer time period as structures develop only incrementally. It might be a typical feature
in start-up processes and lead to limited growth in employment and sales (Perez Perez
and Sanchez 2003). If, however, these new firms become well established in their
regional environment, generate production and research networks and turn into role
models for others in adapting new technologies, they can likely have a strong overall
impact in the region. This second stage is entered if regional interaction, input-output
linkages and knowledge exchange stimulate learning processes within existing
organizations (Bathelt 1991a). Rather than being merely fitter than established firms,
start-ups might induce learning processes in these organizations. Since the established
firms in the region have a strong record of efficient, reliable, and accountable structures,
their adaptation likely has a higher potential to generate more visible and stronger effects
in terms of generating or expanding incomes and jobs. The increased fitness of existing
firms might, in turn, strengthen the legitimacy of new firms, if the latter draw from the
growth process of the former firms, and learn to establish routines and institutions that
strengthen reproductivity. As a consequence, the combined potential of new firms and
reorganized established firms can stimulate broad learning and renewal processes at the
regional level, eventually having an impact on other industries in the regional economy
and stimulating further start-ups (Figure 1).
***********************************
Insert Figure 1 about here
***********************************
- 13 -
This is, however, by no means a deterministic sequence of developmental stages.
We know that the success of start-up firms and regional renewal processes depends on
many factors, two of which deserve particular attention: First, we know from empirical
studies that broad regional effects are more likely to occur if the regional economy draws
from technological complementarities and overlapping knowledge bases which enable
firms to establish regional networks and engage in knowledge exchange (Bunnell and
Coe 2001; Boschma 2005). A corresponding regional ensemble of firms can have the
form of a regional thickening of a particular value chain or of a fully-fledged industry
cluster with a well-developed supplier, service and institutional infrastructure (Porter
2000; Maskell and Malmberg 1999; Bathelt and Boggs 2003). In the case of a cluster,
regional networks can develop and dynamic local knowledge flows, or local buzz, enfold
(Storper and Venables 2004; Bathelt et al. 2004). In contrast, if local firms are not
closely related to – or cognitively distant from – one another in terms of their utilized
technology and knowledge base, possibilities for local networking and growth triggers
will likely remain limited (Nooteboom 2000). Instead, we might observe that growth – if
there is growth at all – has few collective qualities but is caused by individual firm
successes which rely on close bonding with partners outside the regional or even national
economy.
Second, strong connections to global value chains, and access to external markets,
technology partners, and knowledge pockets, are key in generating growth impulses, and
initiate and sustain processes of innovation in a regional ensemble (Owen-Smith and
Powell 2004; Bathelt et al. 2004). Ideally, both conditions enable university spin-offs to
provide ideas and incentives for restructuring, renewal and reorganization processes in
established firms. Through this combination, negative consequences of structural crises
- 14 -
in a region could be overcome, as suggested in the model of regional re-bundling (Bathelt
and Boggs 2003). Feldman et al. (2005) describe a similar sequence in the case of the
biotechnology industry as a process that can lead to the establishment of new clusters
(Feldman 2001). On the contrary, if external cluster linkages are weak or non-existent,
regional growth would remain limited, and the danger of negative technological lock-in
might increase resulting in a relatively unstable regional economy (Bathelt and Glückler
2003).
If both conditions are not fulfilled, strong regional economic growth processes
would be an unlikely outcome. Initially promising university spin-offs would remain
small or disappear from the market. What is important to note is that such ongoing
selection processes most likely will not prioritize small firms at the expense of
established firms. In fact, if anything, spin-offs are likely under more pressure to perform
well, and become reliable, while established firms have more resources to master internal
adaptations.
2.3 Toward a typology of university spin-offs and start-ups
While we are skeptical about the prospects of the development of a new fully-
fledged industry cluster from the long-term effects of university spin-off processes alone,
we think that these new firms have a high potential for creating crucial linkages within
the regional economy to enhance learning and further innovation. Having developed the
above argument, it is still necessary to clarify how to best define and differentiate
university spin-offs. This is not a trivial question, as it seems that a widely accepted
definition does not exist, despite attempts to develop more rigorous typologies of
university spin-offs (Pirnay et al. 2003; Mustar et al. 2006). In parts of the literature,
- 15 -
university spin-offs are defined as activities which develop directly from knowledge
produced in university research. Here, the spin-off process involves a direct transfer of
personnel, and novel scientific or technical concepts to the private sector (Smilor et al.
1990; Pappert et al. 1999). Often this provides a source of revenues for universities. In
most instances the university is at least in part owner of the intellectual property rights to
the new product or process developed by its staff through research activities. Although
variations between and within countries exist (Rasmussen et al. 2006), this is something
that is quite common at most Canadian universities. At the University of Waterloo, in
contrast, a different policy was introduced in the 1960s in that the developer or inventor
of intellectual property in generally remains the owner of it (Bathelt and Hecht 1990;
Bramwell and Wolfe 2008). Even though there might be good reasons to use such a spin-
off conception in other investigations, we found that it was not sufficient for our study
because it covers only a portion of the varying technological and economic impacts a
university has on its regional economy.
Other definitions are less specific and more inclusive in their use of the term.
Sometimes, for instance, every firm is considered a university spin-off that has been
founded by a university graduate. This definition, in turn, is also problematic because it
implies that most firms would be considered as university spin-offs because they were
started by someone with a university degree (see, for an overview, Garvin 1983; Mossig
2000). While this latter definition seems to describe the University of Waterloo’s
original view of spin-off firms, the its Technology Transfer Office (now: Intellectual
Property Management Group) has meanwhile adopted a different perspective very much
- 16 -
in line with our first definition.
7
In this study, we introduce a different typology of
university spin-off and university-related start-up firms. Key to our definition is that
these firms draw first upon knowledge that is produced or circulated at the university,
second on individuals or collectives who met or got together in the context of the
university, and third on business opportunities which result in relation to existing areas of
competence in research and teaching.
Table 2 portrays a classification of spin-offs according to university sponsorship
and university involvement. This distinction allows us to distinguish firms according to
how “close” their start-up process is to the university’s core competencies and research
competencies. Sponsored spin-offs are firms which are established with active support
and approval of the university. Only rarely does this support, however, involve direct
money transfers. Instead, sponsorship typically involves training and support in the
management of the start-up phase. Sometimes, this includes granting preliminary
office/research space and the use of university facilities such as libraries and research
laboratories at a low rate. The support can also involve that the university buys a new
firm’s products instead of established products from existing firms.
8
University spin-offs
7
Definitions of university spin-offs often also include activities which were originally conducted
inside the university and later split off. This is the understanding which is, for instance, applied at the
University of Alberta (see http://www.uofaweb.ualberta.ca/tecedmonton/spinoffdefinition/cfm/
, date
accessed July 17, 2007) and the University of British Columbia (see
http://www.uilo.ubc.ca/pdf/uilospinsrvy_1998.pdf/
, date accessed July 17, 2007).
8
In the case of the University of Waterloo, this was a practice in early spin-off firms such as
Volker-Craig, which was founded in 1973 in the area of computer terminals. The firm had substantial
- 17 -
can, however, also be unsponsored. On the one hand, this might be the case when a
researcher establishes a company against the wish of the university, but in a way that is
legally correct. While this case might be rare in a university environment, a threat exists
that firms could be split off from a university–industry joint venture project – driven by
the private partners – using knowledge that was primarily developed in the university.
We assume that if such unintended knowledge transfers have occurred in the past,
universities are less likely to continue collaborating with private-sector firms in research
projects. On the other hand, new ventures that fall in the category of unsponsored spin-
offs might also be in line with the general expectations put forward by a university’s
technology transfer policies.
***********************************
Insert Table 2 about here
***********************************
In terms of university involvement, we distinguish three cases: (i) spin-offs from
university research based on intellectual property developed at the university, (ii) spin-
offs which result from university–industry joint ventures
9
, (iii) spin-offs resulting from
decentralized individual or collective ideas developed at the university, unrelated to
business success in its early years and employed 185 people in 1980. By 1988, however, only 6 employees
were left in the region, and later the firm closed down (Allaby 1984; Black 1988; Bathelt 1991a).
9
In the case of the University of Waterloo, Open Text is an example of a spin-off firm which grew
out of a joint venture research project with Oxford University Press to develop computer-indexing and
string-search algorithms for structuring a digital version of the Oxford English Dictionary (Open Text
Corporation 2001; Colapinto 2007).
- 18 -
research projects (Table 2). The latter group might involve firms which are started up by
former graduate or undergraduate students after they finish school. These students might
have met while studying in the same or a related field, and developed a business idea
from their joint classroom experience. We refer to these firms as university-related start-
ups, as opposed to university spin-offs from research projects. We believe that it is
important to include these firms because they would be unlikely to exist if the university
had not provided the opportunity for the founders to get together and develop their
business ideas, while benefiting from experiences of other start-ups and spin-offs. It is
very difficult to study this group of spin-off firms, however, because there is no natural
data base from which one could draw.
In terms of reliability and accountability, university-related start-ups and the spin-
offs from pure university research projects often have little legitimacy in their early
stages. They have to demonstrate that they are competent and able to reproduce their
structure to earn legitimacy in the market. This is likely a slow process which takes time.
Firms from these groups are less likely to produce high growth than spin-off firms which
result from university–industry collaboration. The latter firms benefit from existing
industry networks and third-party referrals which are passed on to them. Many ventures
might develop in an evolutionary way over a longer time period before being formally
started up (for a summary, see Hayter 1997). If the respective founders manage to
establish networks with initial customers, suppliers, investors and employees in this pre-
launch period, they will likely start at a higher level of legitimacy when formally entering
the market. As start-up processes of these firms are often not associated with a true
location decision between different regions, legitimacy may, to some degree, be focused
on the regional market driving a relatively strong initial local orientation (e.g. Bathelt and
- 19 -
Glückler 2003). Of course, a lack of legitimacy also makes it more difficult for firms to
acquire start-up finance and finance for further rounds of investments, despite the
existence of venture capital. Founders seem to typically make up for such problems in
that they mobilize resources for seed financing from existing social networks which are
often concentrated in their region, such as local banks, family and friends (e.g. Zook
2002; Kenney and Patton 2005). This partially explains why regional linkages tend to be
stronger in the start-up phase compared to later stages.
Another classification of spin-off/start-up activities related to universities might
help us to better understand the focus of relationships of a new firm. It also helps us to
distinguish start-ups and spin-offs according to the character of university knowledge
applied and the pattern of co-location of the founders (Table 3). The pattern of co-
location in the start-up period refers to the question whether the founders of a company
are associated with the same university (which is also the case if there is only one
founder) or with different universities in different regions. In the first case, firms will
likely draw more heavily from regional resources in the start-up stage while, in the
second case, they might be able to develop wider interregional networks from the very
beginning.
10
Both instances might be indicative of different growth trajectories, one
10
One of our interviews fell in this category. The firm produces natural language search and
database systems with customer service applications (Interview 18). The Waterloo-based division of the
firm focused on providing support to university clients. A CEO was hired and, who originated from
Toronto, and, parallel to this, operations were established in Toronto. Management, sales, development,
and customer support activities remained located at both locations. Through the hiring of this CEO from
Toronto, the firm was able to extend its business to the financial organizations – including several of banks
– headquartered in Toronto.
- 20 -
being based on limited regional, the other one based on wider cross-regional legitimacy.
The second variable draws upon the character of university knowledge used for the spin-
off/start-up process. Here, the distinction is made between (i) generic, broader, less
specific knowledge which can, for instance, be transmitted in a seminar or lecture, and
(ii) specific knowledge related to the competence base of the university and closely tied
to university research. The latter firms are expected to coincide to some degree with the
group of university spin-offs described above, while firms in the former category should
correspond more closely with university-related start-ups. The classifications developed
above not only allow us to distinguish between spin-offs and start-ups according to their
individual growth potential, but in aggregate terms they also enable us to draw
conclusions about regional development paths.
***********************************
Insert Table 3 about here
***********************************
3. Economic Growth and Spin-off Processes in the Kitchener and
Guelph Metropolitan Areas: Results from Previous Research
The Kitchener and Guelph metropolitan areas were traditionally – and still are –
characterized by a strong diversified manufacturing base. In the first half of the 20th
century, the region had well-developed economic strengths in the rubber, textile, leather,
furniture, and food processing industries. Despite the differentiated industry structure,
however, regional supplier linkages never seemed to be very strong. While the rubber
industry was, for instance, originally established as a supplier sector to the shoe industry,
- 21 -
when Schlee and Kaufmann founded the Berlin Rubber Company in 1899, it later shifted
toward other customer groups, especially toward the production of tires (English and
McLaughlin 1983). In the post-World War II period, manufacturing growth was driven
by industries, such as fabricated metals, machinery and electrical products (Bramwell et
al. 2004; 2008; Wahl 2007; Wolfe 2007). Furthermore, the region developed a strong
basis in the automobile supplier/transportation equipment sector (Rutherford and Holmes
2008).
Since the 1970s, numerous university spin-offs were started up in the region. This
was related to the foundation of the University of Waterloo in 1959 as a university with
an engineering focus, allowing members of the university to own patents from university
research. Industrial leaders, such as Ira Needles from BF Goodrich, played an important
role in the design of the university. They shaped the University’s co-operative education
program and its openness toward private sector collaboration
11
(Bathelt 1991a; Wahl
2007; Bramwell and Wolfe 2008). Compared to other Canadian universities in the post-
World War II period, the University of Waterloo not only had a more pronounced focus
on establishing university–industry linkages, but also developed a stronger focus on basic
and applied research. According to data presented by Niosi (2000), the University of
Waterloo was Canada’s largest research university in the late 1960s with 533 researchers,
representing about a quarter of all researchers at Canadian universities. In comparison,
11
As part of this, the University of Waterloo received substantial funding from multinational IT
firms. It received about CAD 40 million in IT-related funds from IBM and DEC in 1980s – a practice, that
has continued until today through RIM’s major investments into the University’s research infrastructure.
- 22 -
for instance, the University of Toronto had only 256 researchers. At that time, the
University of Waterloo became an important driver of a more research-oriented as
opposed to a resource-led national production and innovation system. Until 2007, the
University of Waterloo was incubator to 47 spin-off firms, 24 of which were directly
related to IT industries.
12
The University’s initial advantage, however, decreased over
time. Already in the 1990s, observers began speculating that the overall key to economic
growth and success in the region was primarily due to the co-op program and a constant
flow of highly qualified graduates, who found a job in the region’s growing technology
sectors, rather than a consequence of University research and spin-off processes (Bathelt
1991a).
Aside from start-ups around the University of Waterloo, the region also attracted a
number of multinational IT firms such as Google, Hewlett-Packard, Microsoft and NCR,
which established branches or acquired existing technology firms. Although 87% of the
firms surveyed in information and communication technologies in a Communitech (2006)
report had in-house R&D, most of this seemed focused on incremental development tasks
rather than basic or applied research (Bramwell et al. 2004). In addition, local
technology firms seemingly had not developed extensive input–output linkages in the
regional economy (Bathelt 1991a; Xu 2003; Bramwell et al. 2008), although an earlier
study concluded that regional input linkages were not generally weaker than in other
12
There were also other educational organizations which spurred firm formation in the region.
The University of Guelph, for instance, spurred at least 13 spin-off firms outside the IT sector, and
organizations, such as Conestoga Community College, which claims that more than 200 enterprises were
started by its graduates (Conestoga College 2003), also led to technology-based start-ups.
- 23 -
high-technology regions, such as Boston and Silicon Valley (Oakey 1985; Bathelt
1991b).
Due to ongoing investments, the regional economy successfully transformed its
traditional manufacturing base, while remaining strongly diversified. As shown in Table
4, traditional manufacturing sectors such as textile mills, clothing and leather
manufacturing lost between 50 and 60% of their employees in the Kitchener CMA
between 2001 and 2006; chemical and electrical equipment manufacturing lost another
20% of their employees. This structural change was over-compensated by a 20%
increase of the employment in plastics/rubber products and computer/electronic product
manufacturing. Furthermore, most knowledge-based producer-related services
experienced substantial job growth (see, also Table 1). In the areas of professional,
scientific, technical and educational services, for instance, total employment increased
from about 40,000 to 47,500 from 2001 to 2006. The most spectacular job growth in the
Kitchener CMA in this time period occurred in the scientific research & development
services branch. Here, the number of employees increased by more than 250% from 400
to about 1,550 (Statistics Canada 2001b; 2006b; 2006c).
13
***********************************
13
In its 2006 report, Communitech (2006) identified a total of 284 firms with 13,300 employees in
information and communication technologies, 79% of which were headquartered in the region. Further, 65
firms were identified in the area of science, technology and engineering (5,100 employees), 58 firm in
biotech, life sciences and the environment (1,400 employees) and 58 in advanced manufacturing (9,200
employees). The latter category also included parts of the conventional manufacturing sector.
- 24 -
Insert Table 4 about here
***********************************
4. Research Approach and Methodology
To explore the social foundations of regional innovation, and the underlying
processes which drive innovation, we conducted a largely qualitative study of IT-related
university spin-offs/start-ups, using semi-structured interviews. Since this research was
part of a larger study concerning the social dynamics of economic performance in
Canadian city-regions (e.g. Holbrook and Wolfe 2005), the methodology applied was
structured to be consistent with that of the overall study and its goals.
The total population of university-related start-ups and spin-offs was composed as
follows: As a basis for our research, we used a list of 47 spin-off firms from the
University of Waterloo (provided by the University’s Intellectual Property Management
Group in August 2007) and 14 firms where the University of Guelph played an active
role in the start-up process (provided by the University’s Business Development Office in
September 2007). Further, we had a list of 227 start-up firms established by
PricewaterhouseCoopers (2001a) from a survey in which firms identified themselves as
being related to the regional universities, on a scale from strongly related (“but-for”) to
weakly related (“at least in part”) (PricewaterhouseCoopers 2001b).
14
We viewed the
firms indicated by the universities as true spin-off firms since they were closely related to
14
The PricewaterhouseCoopers list is not unproblematic as we cannot say much about its
reliability. It is, however, the only such list available and was, therefore, a “treasure” for our research.
- 25 -
university research. The other firms identified by PricewaterhouseCoopers (2001a) were
viewed as university-related start-ups. From these lists, we identified firms with an IT
focus as the population from which our final sample was drawn (Table 5).
***********************************
Insert Table 5 about here
***********************************
In a first step, the firm lists were consolidated and double entries and firms in other
sectors than IT removed. In the PricewaterhouseCoopers list alone, 133 firms were not in
the IT sector. The list included firms such as engineering consultants, architects,
specialty equipment outlets, and even a dance studio and a flower shop. The resulting
population consisted of a total of 119 firms: 25 university spin-offs
15
and 94 university-
related start-ups. Before starting our sampling process, we tried to track down all of the
119 firms and collect information like business focus, address, firm size and potential
interview partner.
This involved a time-consuming process: We checked the virtual firm directories
of the region and the cities’ business development offices (see
http://www.techtriangle.com/search/business.cfm, date accessed July 27, 2007), as well
as several other websites (i.e. Canada411, Waterloo Tech Digest, uwRyan.com, Strategis
– Government of Canada), and Google. As a consequence of this search process, the
relevant population for our research declined further. Out of the 119 firms in the original
15
As the list provided by the University of Guelph contained only 1 IT-related spin-off firm, our
survey was focused on University of Waterloo spin-offs/start-ups.
- 26 -
population to be investigated, 16 firms (13%) were not found, and another 36 (30%) were
not located in the region. An additional 8 firms (7%) were acquired by another IT firm
on the list, whereas another 17 establishments (14%) were confirmed closed. After this
process, we arrived at a population of a total of 42 firms.
Out of these 42 IT firms, a random sample of firms was drawn which later had to
be extended because numerous firms were not willing to participate in our research. This
left us with too few firms to generate a sufficient pool to interview. At the end we
contacted 32 out of 42 firms (76%) on our list (Table 5).
16
Of the 32 firms contacted, 14
were unresponsive and did not agree to an interview.
17
Although this is a seemingly
small sample of firms with 18 of originally 288 firms, we should keep in mind that these
firms represent a substantive portion of the IT-related start-up/spin-off activities of the
University of Waterloo. Related to the relevant firm population of 42 firms, a substantial
proportion of 18 (43%) were interviewed. The interviews conducted took between 45
and 75 minutes on average, and were mostly done on a face-to-face basis, with few
exceptions. Altogether, we believe that it is feasible to draw conclusions regarding the
impact of IT-related university spin-offs/start-ups in the region from our sample.
In our interviews, we investigated in which way different groups of spin-off and
start-up firms established regional producer-user networks, to which degree they
16
Although the local business association originally agreed to support the project helping to
approach firms, they seemingly lost interest when we mentioned that we wanted to select the interview
firms ourselves.
17
This rejection rate of 45% might also be an indication that the region has been over-studied in
recent years, associated with many interview requests for local firms.
- 27 -
developed, and depended upon, global pipelines, and whether this dynamic produced
spillovers to other regional industries, due to practices such as inter-sectoral networking,
technology transfer, and job hopping. The questions asked focused on three main areas
of interest: First, we asked with which goals and incentives, and under which conditions
the firms were started up in the region. The second set of questions enquired about
material linkages and knowledge flows, which developed within the region or with
partners in other regions and countries. Third, we were interested to find out whether
local institutional support and economic policies provided incentives to develop local
linkages, and even had cross-industry effects. Key characteristics of the firms included
are shown in Table 6. The results of our interviews are discussed in the subsequent
sections.
***********************************
Insert Table 6 about here
***********************************
5. Myths and Realities of University Start-up and Spin-off Processes
The transformation that has taken place in the Kitchener and Guelph metropolitan
areas from an economy based on traditional manufacturing to one with a substantial
proportion of IT-related businesses is often at least partly attributed to knowledge
transfers and growth triggers based on university spin-off processes. In this section, we
examine some of the start-up processes that the firms experienced. Our goal is to identify
how strongly the start-ups/spin-offs are embedded in the regional economy, and how this
has changed over time.
- 28 -
All 18 firms in our sample had direct ties to the University of Waterloo since some
of their founders were graduates (18 individuals), faculty (seven individuals), or staff
members (five individuals) of the University. However, the actual role the University of
Waterloo played in the foundation of these firms was frequently found to be either
minimal or indirect. This was surprising as the University is usually portrayed as the
central agent in the rise of IT-related developments in the region.
18
In six cases, we did
not find any active university involvement in the start-up process. In contrast, another six
firms were founded by faculty or staff member based on a core technology that was, at
least in part, developed within the University. In four of these instances, the technology
was the direct outcome of a university–contract research project, and the University of
Waterloo received some sort of compensation for the intellectual property rights, either
through the direct transfer of funds or shares in the new venture. Essentially, this
constituted the process where an informal business unit was removed from its university
setting and spun-off into a stand-alone business model.
The firms captured in our population were overwhelmingly software-focused, often
based on specific mathematical algorithms or software solutions. Even the ones that had
a hardware product line still produced software as an integral part of their product
offering. Four firms that had once developed hardware eliminated those products from
their portfolio to focus on what they considered to be their core competence, i.e. software
development. As discussed before, the original list of IT-related university spin-
18
The study of Colapinto (2007) also found fewer university-related start-ups than expected with
less than 20% being closely related to technology transfer from the university.
- 29 -
offs/start-ups contained a number of firms that had been acquired by other firms,
sometimes resulting in the relocation of the original business unit. Typically, the
technology and part of the some senior management left the region, but the majority of
employees were let go or quit to stay in the region. Considering that the
PricewaterhouseCoopers’ list was only established in 2001 indicates that the university
spin-offs/start-ups in the region experienced substantial structural changes in their early
years.
Six of the 18 firms interviewed were acquired by another larger firm, typically
occurring 5 to 10 years after the foundation of the venture. In the case of four firms, the
core of the business unit remained part of the main development center for the firm. In
one notable case, the firm was acquired four times by successively larger server and
database management companies, but the actual business unit remained almost
completely intact over that time, with the same management personnel and the same
basic product line. All of these firms were acquired by non-regional players with the aim
to add a product line to their existing business, and also to develop a physical presence in
the Waterloo region.
The spin-off processes common to the region often did not follow the expected
process of intellectual property transfer from university-based research to a spin-off firm
started by a faculty member.
19
From our interviews, as well as media reports and other
19
Based on a standardized questionnaire, Xu (2003) estimated that 65% of all start-ups were based
on some sort of knowledge transfer related to the University. Although this result appears somewhat
optimistic in the light of the interviews conducted in our study, Xu (2003) also found that the knowledge
transfer linkages decreased over time.
- 30 -
academic studies of the region’s technological development (e.g. Bathelt and Hecht 1990;
Bramwell and Wolfe 2008), it became clear that the University and the community
leaders in the region were very proud of the entrepreneurial output of the university
faculty. In our interviews, and in media reports, they always emphasized the “inventor-
owns” policy of the University of Waterloo attracting commercially oriented faculty. It is
frequently suggested that this was an important cause of the economic growth in the
region. This type of spin-off process, whereby a university-developed research product is
commercialized and the firm’s initial development stage is aided by the university
through either economic support or the use of facilities, occurred, however, in only four
of our 18 cases. This was much less than what we would have expected considering the
general perception of the University of Waterloo’s significant role in the generation of
university-based spin-off companies.
One example of this type of spin-off process is a firm that developed software
which allowed for the manipulation of symbolic data at a time when other programs were
only able to handle mathematical data. The firm was founded by the principal
researchers who created this software, and who were themselves studying symbolic
mathematics, after it became necessary to hire a full-time staff member. Despite the
product being offered free to the academic community, the demand was so high for the
product that an individual was needed to keep up with copying floppy disks and fill
orders. The establishment of the firm was more of a decision to be able to distribute the
software to academics rather than a drive to commercialize the product. Support for the
firm was provided by the University of Waterloo in the form of space and office
resources. The technology was not patented, so the University’s Intellectual Property
Management Group (formerly the Technology Transfer Office) was not involved (it
- 31 -
should be noted that none of the firms in our sample indicated that this office was
involved in their spin-off or start-up process). There was no business development
support provided to this firm.
Further, there were two firms that – although both fell into the category of
university-based research – also received no additional support. In both cases, the
founders were involved in the development of technologies within the University of
Waterloo’s Computer Systems Group (CSG) but left the organization with ownership of
intellectual property rights to the technology which was utilized to establish their own
businesses. In one instance, the founder continued to work at the University as a member
for another ten years while developing his own business. The individual continued this,
until the business was sufficiently established, and generated a revenue stream that was
adequate to cover living expenses. Both of these examples represent our category of
spin-offs from university research, but were unsponsored in that the firms received no
direct business building support from the University of Waterloo.
The most common type of start-up that we found in ten of the 18 firms was the
decentralized and unsponsored firm formation process. Although the exact
circumstances of how the founders met varied from instance to instance, the university
was not involved in sponsoring the development of the core technologies around which
these firms were founded. The tie to the University of Waterloo was typically that the
founders were trained there, and met while in school socially or in class, or while
working in the region after graduation. Again, the PricewaterhouseCoopers database
informed us a priori that these firms claimed some relationship to the University of
Waterloo.
- 32 -
One example of a decentralized start-up process is a small firm that was created by
two University of Waterloo students. The two founders were enrolled in the co-operative
education program at the University of Waterloo and met during a work term. They were
both on-line game enthusiasts who believed that they could improve their game play
experience if they had a voice–communications program to use. As a consequence of
this perceived need, they began product development and eventually created a voice–over
Internet program. The firm was acquired by a multinational software development firm
that took the program and used it as a basis for two major products lines. The intellectual
property developed in this unsponsored decentralized case was solely that of the
principals; they met, conceived, and developed their software off-campus, and received
no support from University services. Each of the respondents to our survey indicated that
the reason they started their firms in the region was that they were already living there.
In those spin-offs, where the University of Waterloo played a fundamental role in the
creation of the core technologies, the relationship with the University typically decreased
over time. It should be remembered, however, that only six out of 18 firms fell into this
sponsored category.
As an exception, two university-sponsored spin-off firms identified remained
actively involved in activities at the University of Waterloo. However at the time of our
interviews, neither of the firms received significant inputs into their innovation process
from University collaborations. The founder of one well-known software firm noted that
“the primary value of these projects is to meet, and get access to, future potential
employees” (Interview 13). Firms which were spun off from the output of university–
industry collaboration (5 out of 18 firms), also saw their relationship with the university
decrease since their establishment. In the case of a larger data and work flow
- 33 -
management software developer, initial product innovations occurred while the
development was still situated within the University. Later relations were classified as
“arms length” by one of the founders, who remained a University of Waterloo professor.
He noted that there were still collaborative research projects underway but that “these
projects have not been decisive for the firm” (Interview 9).
In the latter examples, we see that the University of Waterloo was of fundamental
importance in the establishment of the firm, but that the role diminished and the
University later no longer provided key inputs to the innovation process. Regardless of
the initial relationship with the University, it was surprising to find overwhelmingly weak
relationships between the University of Waterloo and these firms. When the 18 firms
interviewed were asked to identify the most important organizations supporting
innovative processes, only four named the University of Waterloo; however, those four
firms also identified customers as equally important. In fact, all 18 firms interviewed
identified customers as the most important organizations they dealt with for stimulating
innovations in their firms. Rather than the radical innovations that marked the
foundations of many of these firms, particularly based on university-sponsored research,
most of these firms had shifted their research toward more incremental innovations
drawing from ongoing customer relationships. In addition to a decreasing number of
university-related start-ups and spin-offs, second-generation spin-offs, such as B2B
Scene which originated from Open Text, were rare (Colapinto 2007). The establishment
of local and non-local social relations will be explored in more detail in the next section.
- 34 -
6. Local and Non-local Social Relations in Innovation
The University of Waterloo is clearly a significant local source for innovation, but
only played a limited role in the formation of half of the firms interviewed (eight out of
18 firms). We classify these firms as unsponsored, decentralized start-ups that did not
derive their technological competence from university research, nor did they receive
specific support from the University. Further, we observed that a number of firms which
had originally spun off from university-generated research had relatively little input or
stimulus from the University in their later innovation process (see, also, Xu 2003).
As discussed in the conceptual part, if local firms are not closely related to one
another and to other firms in their technological and knowledge bases, then the
possibilities for local networking and knowledge flows between firms appear limited.
Under different circumstances, however, university spin-offs may provide a trigger for
technological change within a region. We proposed two conditions under which this
could take place. On the one hand, spin-off firms can potentially acquire sufficient
legitimacy to become a catalyst for regional economic growth driving organizational and
technological change. On the other hand, we proposed that if spin-offs become more
established within the region and generate production and research networks, they can
stimulate learning processes within established firms. In this section, we will investigate
to which degree the start-up/spin-off firms have established local or non-local linkages
relevant to their innovation process, as a way of determining their possible impact on the
regional economy.
- 35 -
6.1 Producer–User Linkages and Knowledge Flows
To acquire information about the regional impact of university-related spin-
offs/start-ups, we asked our sample firms to describe the nature of producer–user linkages
and other important knowledge flows established, and what impact these relationships
had on their innovation processes. In this section, we will focus primarily on findings
related to supplier and customer relationships. Our primary interest in exploring local
linkages was to identify what sort of intra-regional knowledge flows exist between
university-related spin-offs/start-ups and other firms in the region. In addition, an
attempt to account for knowledge pipelines that may exist into or from other regions was
made.
In terms of supply-side relations, firms responded that reliable suppliers are
important but not critical to their innovation process: 10 out of 16 respondents viewed
them as relatively unimportant, and only two as important (Table 7). Similarly, having
local suppliers was rated as being unimportant in all cases. Furthermore, 15 out of 17
firms indicated that the percentage of supplies acquired locally was 10% or less, with 11
firms indicating that the amounts were negligible (Table 6). Key supplies, where they
existed, were typically not drawn from within the region. Many firms recognized
suppliers more broadly as development partners but not as key figures in the generation
of new ideas for innovation and development in their firms. As one interviewee pointed
out, “[o]ur suppliers provide us with tools … the capabilities of those tools can be
important, but as far as applying new ideas, practically zero” (Interview 8). This is also
confirmed by the findings of Rutherford (1996) and Bramwell et al. (2004; 2008). In an
earlier study, it was shown that regional supplier linkages in high technology industries
were not lower than in other high-technology regions in North America (Bathelt 1991b):
- 36 -
Results from a survey conducted in 1988 indicated that regional supplier linkages were,
on average, almost as strong as in the Boston region. It should be emphasized, however,
that software developers do not draw heavily from suppliers, but rather purchase pre-
packaged and standardized software packages universally available. In this respect,
supplier linkages might not be the best indicator to judge the local innovative practices of
these firms.
The three firms which indicated that suppliers played a significant role in the
development of new ideas and product innovations were – not surprisingly – all hardware
producers. One of these firms was purchased by a multinational company with numerous
R&D facilities located around the world. The interviewee indicated that before the firm
was acquired and still had local ownership, “[v]endors were more frequently located here.
Now that we’re an R&D site, we’re linked more with other global players” (Interview 1).
The director of another firm that was also acquired by a multinational server and database
company commented on supplier relationships saying that “[t]hese are negotiated by our
head office in the US. I have no input into these contracts” (Interview 4). In both cases,
as the management of these firms moved out of the region a shift occurred away from
local to global supplier sources, mediated through their respective corporate networks.
This does not imply that the supplier relationships that would have otherwise existed with
local vendors would have been significant inputs for innovation; the absence of these
opportunities, however, will make it unlikely that such regional producer–user
relationships will be built in the near future.
In comparison to suppliers, the role customers play in the innovation process was
considered far more crucial by our interviewees: 12 of 15 respondents indicated that
customers were important in innovation (Table 7). When asked where the ideas for new
- 37 -
products or services came from, each firm cited customers as one of the key sources for
the generation of new or the improvement of existing products and services. In general,
all respondents rated customers as either very important or critical to their innovation
processes. There was no difference in sectoral focus observed here; all the interviewed
firms viewed their customers as key sources for innovation and new ideas.
The location of customers, similar to suppliers, was deemed not to be relevant by
the firms that were interviewed. Nevertheless, it should be noted that that this is likely a
standard answer to questions about the role of proximity which does not necessarily
reflect the actual nature of social relations. The likelihood that close-by customers
receive more attention in terms of regular interaction practices than distant customers is
certainly higher, especially if we consider the significant role face-to-face interaction
plays in this context, whose frequency is influenced by the actual physical distance
between the two parties involved. Our research does not allow conclusive statements in
this respect. However, examples such as Dalsa’s establishment of a small office in
Hollywood, Open Text’s acquisitions in other countries and RIM’s international
expansion, which have been publicized through the media, indicate that firms were aware
of the need to be present in other markets to acquire important information and develop
closer customer relationships.
In contrast to these examples, all 18 firms answered that having local customers
was not critical to their innovation process, while 14 firms also emphasized that they sell
to a global client base (see, also, Bramwell et al. 2004). The Kitchener/Guelph
metropolitan areas and other southern Ontario markets were not significant to these firms
in terms of the percentage of their overall sales: 14 of 15 firms with regional data
available indicated that their regional sales were 5% or less (Table 6). It should be noted
- 38 -
that most of the firms, particularly the software firms, had ongoing support contracts with
their customers, and thus received more or less regular feedbacks from them. The
president of a small software company that provides logistics management software
explained that one of his major clients was located in Cambridge, Ontario, and a second
one in the U.S. in Mississippi. He would experience no difference in his ability to service
the two customers. The primary methods of communication between these firms were
the telephone and the Internet, and he would spend relatively little time at either of the
clients’ sites. This is an example which indicates that close-by customers, who are rare in
our study to begin with, do not automatically experience preferential treatment.
Although customers were a major source of new ideas for the software-related IT
firms in our sample, our interviews suggested that firms who went about implementing
these ideas often stumbled on technological hurdles. In aiming to solve these technical
problems, they seemingly needed to turn to the Internet and its user and expert
communities to find solutions. As a consequence of these practices, they no longer
required close connections with suppliers, customers, or peer-firms. As one interviewee
powerfully stated, “[o]ne of the best skills you can have is finding things on the Internet.
… Ten years ago, I could do my work without the Internet, now I can’t. We’re
dependent on the Internet for this sort of help” (Interview 2). This included the use of
virtual user and expert communities which provide assistance with and solutions for
specific technical problems, potentially leading to incremental innovation. In our sample,
12 of the firms rated the Internet as an important or key source to find solutions for
problems. In addition, six firms mentioned intra-firm problem-solving as a decisive
instrument in this regard (Table 8).
- 39 -
Aside from the increasing role of Internet user communities in this field (e.g.
Jeppesen and Frederiksen 2006), a possible explanation to this might be related to the
context of the firms in our sample. Most firms had a small hardware component in
production and did not have many local customers or suppliers. Consequently, they did
not benefit from specialized local buzz. Instead, they operated very much like “stand-
alone” firms in the regional economy, and were relatively isolated without strong local
linkages. Some were also too small in terms of resources and employees to establish
frequent and intensive face-to-face contacts with distant customers. Given the software
dominance and the degree of specialization in their work, it was no surprise that the
reliance on the Internet and intra-firm information sources were key to these firms. This
supports the conclusion that regional spin-off/start-up firms around the University of
Waterloo did not channel regional technological triggers by the means of local network
linkages. Based on a quantitative network analysis, Xu (2003) also concluded as one of
his most remarkable findings that the region’s inter-firm network was relatively
unconnected.
There were some exceptions, however, which indicated that different practices also
existed. Interestingly, firms who recognized the importance of being physically close
mentioned this in the context of dealing with or visiting foreign customers. When talking
about the local environment, this was only rarely explicitly raised. In one case, the
founder of a start-up firm had identified business opportunities for another firm in
Kitchener-Waterloo, which later materialized in a contract. He described this as follows:
“I’ll be in the middle of Japan on business and realize there is some business
in that particular area that would be useful to somebody else in town here,
and I would simply offer these leads … it was marketing the whole area, it
- 40 -
was putting deals together that involved several companies, but they were all
companies inside the Kitchener-Waterloo area” (Interview 10).
This statement also points towards the importance of joint business contracts and
collaboration in providing product and process solutions. When engaging in such local
multi-firm contracts, a local agglomeration of complementary firms may be a key asset
with the potential to increase the legitimacy of spin-offs, and in turn generate regional
economic growth. This only existed to some degree though within the Kitchener/Guelph
metropolitan areas as most firms operated in different specialized niche markets.
Although we assume that the local community of IT firms created a considerable
pool of specialized labor, which would enable exchange and problem solving within this
technology sector, this was only occasionally pointed out by our interviewees. One
respondent said: “I know where to go and who to ask [in the local community] if I have a
technical problem.” More often though possibilities for such knowledge exchange
processes seemed limited because of different specializations. Whenever more generic
knowledge was discussed in the interviews, possibilities for local interaction seemed
much stronger. Although several of our interviewees talked positively about the local
labor market and mentioned that meanwhile a “critical mass” had accumulated in the
region over time, they also pointed out increasing shortages of labor and greater
challenges in finding the “right people”, brought on by enhanced competition for the
existing labor pool in the region. Although we might expect that the agglomeration of IT
firms induces frequent face-to-face interaction and accidental encounters with colleagues
or neighbors, relatively little of the local-buzz dynamics emphasized by Storper and
Venables (2004) and Bathelt et al. (2004) were obvious or reported during our research.
German clubs (which are common in the Kitchener-Waterloo part of the region due to its
- 41 -
heritage), neighborhood events, and the advantages of living in a smaller urban fabric
rather than a large metropolis were repeatedly mentioned as vital elements that make up
the character of the local community. We believe that these social events might also play
a vital role within the regional community of IT firms, yet the results of our research do
not allow us to say that they have a strong impact on innovation.
Our research regarding the genealogy of start-up and spin-off firms also provided
extensive evidence of organizational changes as many firms became involved in
acquisition and merger processes (Table 4). They merged with other local IT firms or
were acquired by established foreign firms. In the latter case, they either stayed in the
region as more or less independent subsidiaries or were closed down after a few years
and left.
20
As a corollary of these changes, we assume that local producer–user linkages
have also shifted (see, also Xu 2003). Even if local linkages might have played a
somewhat stronger role originally, several interviewees indicated that corporate networks
had become much stronger. In our framework of organizational ecology, we can view
this as a logical move to gain access to a larger customer basis, or to increase legitimacy.
We suspect, however, that direct customer contact was still significant in acquiring
strategic information about product performance and the future direction of innovation,
even if physical proximity had become less important on a day-to-day basis. At the same
time, we found that the Internet was often used intensively in daily routines, search
20
Rarely, however, it seemed that the founders themselves left the region in these cases. In fact,
there were a few instances where the founders started a new regional business in a different technology.
- 42 -
processes and problem solving, enabling firms without local networks to participate in
broader processes of knowledge generation.
6.2 Institutional Linkages and Support
The role of the University of Waterloo during the start-up processes of the sample
firms has already been explored in section 5, we also investigated how this role changed
since the firms were founded, and what impact other universities and organizations in the
region had on the ongoing innovation processes in the start-up firms. This section will
extend these discussions by investigating the role of institutional linkages and support for
our sample firms.
Collaborations with public R&D laboratories and universities were generally not
very common among the sample population: 13 of the firms interviewed did not engage
in university research opportunities, and 15 did not utilize public R&D laboratories. Not
surprisingly, the firms that had collaborative relationships with public research
laboratories also had relationships with universities. The representative of one regional
R&D facility of a multinational imaging company that made use of university research
partners, but not of public R&D laboratories, commented this as follows:
“I don’t know if there’s anything we’re getting out of it other than PR and
marketing type items. We get to go to events, and the staff at Waterloo are
doing a lot of interesting type research, and we have access to that content for
our staff. So there’s an educational advantage for our staff” (Interview 1).
University research generally involves a mix of scientific discovery, testing, and
prototype development, whereas public R&D laboratories tend to be more focused on
mission-oriented assignments, standard-setting responsibilities and scientific projects
- 43 -
which require capital and human resources beyond those available to private sector
research organizations (Leslie 1993). Based on this multitude of different functions, the
motivation of firms to collaborate with the one or the other organization also depends on
the project at hand, and is driven by the needs of their R&D strategies (Bercovitz and
Feldman 2007). Although we have to consider that diverse industries have different
motivations to collaborate with universities and public R&D laboratories, the above quote
exemplifies that many of the firms interviewed did not draw specific knowledge from
their interaction with the University of Waterloo. Given the context of our research, as
well as regular media reports about the region, this was a bit of a surprise. One
respondent from a manufacturer of measuring devices for precision manufacturing
applications who used both university and public R&D laboratories stated:
“They are the ones who are exploring the frontiers of where our technology
will be used in the medical field five or more years out. It’s not where we
look to generate revenue from, it’s an area where we’re forging deep
relationships to allow us to develop insight and create products that meet
those new needs. The universities are mostly for monitoring rather than
problem solving” (Interview 3).
For this firm, public R&D laboratories played a fore sighting role. Parallel, the
firm also tapped into a strong international network of research partners and established a
European development facility as a center of excellence for electromagnetic applications.
Interestingly, we found that there were likely differences between the firms that
primarily draw upon university research and those that focused on public laboratory
research. The three firms which indicated that they collaborated with public R&D
laboratories were all hardware producers. The innovation and product development
- 44 -
processes of software firms aligned infrequently with the research conducted at
universities, whereas hardware developers seemingly found collaborative projects more
readily.
Communitech, the local business organization, was frequently mentioned by the
interviewees as a contact partner, but only five firms in the sample utilized the service
offerings of this organization actively. Communitech partners with public entities from
all levels of government, business associations, educational institutions, and technology
associations. The organization, in sum, supports through various initiatives the region’s
technology sector (see Wahl 2007). These initiatives include business and technology
forums, seminars, conferences, and peer-to-peer events ranging from general business
operations to specific CEO networking events. Many interviewees pointed out the
importance of Communitech for the regional industry, even though relatively few seemed
to be involved in intensive interaction and none viewed it to be decisive in innovation.
The peer-to-peer events were found to be particularly useful as one of our interviewees
emphasized:
“Software developers in the region use common tools, such as Java and
Dot.Net. The software developer folks here get together with their peers at
the Communitech groups and talk shop. That gets them excited about their
chosen domains and they come back here and apply it to the health care
sector. We saw that with virtualization. We’ve tapped into the local
community, sponsored internal forums based on that, and we have sites all
over the world that are using virtualization technology” (Interview 1).
Another firm was less enthusiastic, rating Communitech “not critical, but of value”
(Interview 6). The peer-to-peer sessions did not seem to have a primary educational
- 45 -
value to the personnel, but helped to create and deepen social networks. One CEO noted:
“They are good at sponsoring meetings of best-practice leaders, and that’s really
important for keeping [our firm] really sharp” (Interview 3). Of the six firms who
indicated that they made use of Communitech’s services, three said that they used them
for programmers to keep up with generic knowledge in the programming field; three
indicated that this was a good resource to keep up-to-date with business practices. Four
firms further suggested that it was valuable because it generated access to trends and
gossip, especially related to human resources in the region. The peer-to-peer programs of
Communitech were particularly valuable to firms for increasing generic knowledge, but
domain-specific or specialized knowledge that is an essential input into innovation was
not shared.
21
In general, the firms in our sample population had weak linkages to local business
organizations. CTT Inc., which in its own definition is a not-for-profit, private–public
economic development organization with the goal to market the region to the world, was
not mentioned in this context. Although firms were aware of this organization, and
valued the work CTT did for the region highly, none of the interviewees deemed it
important for their own business activities. One possible explanation for this might be
that CTT is primarily concerned with attracting innovative businesses and entrepreneurs
to the region, and the development of private and public sector partnerships, which
21
We are aware that additional informal peer groups exist outside of the Communitech
organization between colleagues, but these were not mentioned by our interviewees as being significant for
innovation.
- 46 -
seemed less relevant for the IT firms examined in this study. Local, provincial or federal
public business or economic development organizations were not mentioned at all by the
interviewees, but – when asked about them – they were acknowledged as instruments that
made sure locational attributes such as infrastructure or operational space were in place.
Nevertheless, it was also pointed out that, even if such organizations did not directly
impact the innovative capacity of local firms, they had important functions, such as the
provision of adequate immigration policies that allowed firms to attract and hire
international talent. Thereby, there would indirectly influence the innovative output of
firms in the region.
Although the firms interviewed appeared relatively isolated in terms of inter-
organizational linkages that provided significant input into innovation processes, the
individuals interviewed have seemingly developed close peer relationships over the
course of their university and professionally lives. When asked to rate the role that peer
contacts played in problem solving, all firms indicated that these relationships were not
sources of ideas or knowledge regarding innovation; however, they also mentioned that
they sometimes drew upon such contacts for problem solving activities (Table 8). These
were primarily social relationships and sources of gossip, particularly about staffing
issues and the health of other firms in the region. Not surprisingly, contacts with peers
from former collaboration tended to be trustful and allowed relatively open exchange of
information.
All interviewees identified the skills and knowledge of their employees as the
single most important factor contributing to competitive advantage within their respective
market space. The ability of local firms to attract and retain highly skilled personnel can
thus be viewed as key to their continued success. In addition to a general set of
- 47 -
programming skills, the software-based IT firms interviewed demanded a specific skill
set from their workers. The skill sets of university graduates including some degree of
specialization and/or work experience largely seemed to satisfy the entry-level
requirements of the firms. Usually, once working for a firm, employees were thus
expected to engage in extensive on-the-job learning, and the firms were resolved to
provide this extensive training to new employees.
Innovative inputs through collaborative research with the local universities were
clearly less important than what we expected, but many interviewees mentioned two
other specific facets by which particularly the University of Waterloo provided important
linkages which were instrumental for the past success of innovation in the region: First,
the progressive intellectual property policy of the University of Waterloo was mentioned
several times but only few firms benefited from it when they started up their venture.
Second, the university’s reputable co-operative education program, which is currently the
largest such program in the world (see http://www.cecs.uwaterloo.ca/about/
, date
accessed March 25, 2008), provided a steady influx of talent for the local firms on a
contractual basis (see Bramwell and Wolfe 2008; Wahl 2008). The University of
Waterloo is a pioneer of this form of temporary student placement, and one of its main
advantages, as pointed out by several respondents, was that it distributed student
placement among local firms on a cyclic basis, rather than making them available for a
short period in the summer. Several respondents mentioned that the co-op program made
it easy to carry out short-term projects flexibly that needed additional human resources.
- 48 -
7. Conclusions
In summary, our research offers some interesting insights into the social dynamics
of innovation processes in a second-tier city-region in Canada. Although the Kitchener
and Guelph metropolitan area – particularly the Waterloo region –are frequently
portrayed as a dynamic technology region which draws from university-related start-
up/spin-off processes, knowledge transfers, and corresponding regional networks, our
research draws a different picture of the developments in this region. Beyond simple
cluster logic, our research draws from a conception which combines aspects of
organizational ecology with insights from the literature on organizational learning. The
basic idea this approach suggests is that technological change and regional development,
such as that witnessed in the Kitchener and Guelph metropolitan areas, are most
successful if there is a combined impact of technological inputs of university spin-
offs/start-ups and restructuring of existing industries in the form of regional networks. In
that case, start-up firms gain initial legitimacy to demonstrate their technological
superiority while established firms face incentives to change established routines. Some
former studies and reports about the region seemed to at least partially confirm this
hypothesis, although other studies did not.
In this context, we explored a population of 18 firms (43% of the relevant total) in
detail by means of semi-structured interviews out of the 42 IT-related university spin-
offs/start-ups that were identified in the Kitchener and Guelph metropolitan areas at the
onset of the present study. The empirical results derived from our interviews were
somewhat surprising. They showed that these firms operated in very specific cross-
regional networks along market and technology linkages that adhere to their particular
- 49 -
technological expertise. Local linkages with customers and suppliers and the existence of
regional industry networks, such as those described in conventional cluster approaches
were quite limited in their extent, or absent altogether. Furthermore, research linkages
with the local universities were unexpectedly low.
Although the IT sector might be somewhat specific in terms of its ability to create
international networks, it does not have fundamentally different linkage patterns than
other new technologies. In particular, we expected university start-up/spin-off firms to
display a somewhat stronger regional orientation in their early stages. This was not the
case in our sample however. We found three reasons which help to explain this structure:
First, it seemed that firms in the area of specialized software solutions were able to
establish a broader extra-regional customer base more quickly and easily than firms in
other sectors. Second, the regional firms we found were extremely diversified limiting
the opportunities for local network creation in a mid-sized metropolitan region. Third,
acquisitions by larger entities that took place within the region served to provide access to
wider extra-regional corporate networks, and thus boosted legitimacy for the respective
units. Overall, we found that spin-off and start-up firms created surprisingly little
specialized local buzz in innovation. Most firms were stand-alone units in the regional
economy with strong international customer linkages, particularly to the US. They had
little ongoing research activities with R&D laboratories and the regional universities,
except those that had a hardware-related component to their product offering.
Despite the lack of strong local relationships, the observed IT spin-off/start-up
firms appeared to be clearly embedded in the regional industry and community structure.
The University of Waterloo provided important skill flows to the regional firms in the
form of qualified graduates, but these were generic skill flows that did not directly
- 50 -
strengthen innovative capabilities. In this respect, the University was very important. As
Bill Gates, founder and chairman of Microsoft Corp., emphasized during a visit to the
region, “[m]ost years, we hire more students out of Waterloo than any university in the
world, typically 50 or even more” (CTV.ca 2005). In contrast to these labor market
effects, our study indicated that less sponsored university research is being transferred to
the region by entrepreneurial faculty members than we would have expected. In addition,
existing university spin-offs/start-ups seemingly entered a stage of incremental
innovation, with few strong research and development relationships persisting.
From this we can conclude that it is primarily weak ties and generic untraded
interdependencies that hold together the fabric of the region. The role of local
universities as sources of spin-off/start-up firms or partners in leading-edge research has
largely been over-stated in media reports, which does not mean, of course, that the local
universities are not significant places of knowledge generation. This is, nonetheless, a
story different from that portrayed in the media and, at least partially, transported through
academic publication channels.
Acknowledgements
Parts of this paper were presented at the 2008 annual meeting of the Association of
American Geographers in Boston, April 15-19 and the 2008 annual conference of the
Canadian Innovation Systems Research Network (ISRN) in Montreal, May 1-2. We very
much appreciate the comments and suggestions by many individuals during these
conferences, especially by Jürgen Essletzbichler, Meric Gertler, Jill Grant, Adam
Holbrook, Réjean Landry, Henrik Mattson and David Wolfe.
- 51 -
References
Allaby, I. (1984): The not so hallowed halls of higher learning. Canadian Business 57
(9): 85-89.
Bathelt, H. (1991a): Schlüsseltechnologie-Industrien: Standortverhalten und Einfluß auf
den regionalen Strukturwandel in den USA und in Kanada (Key Technology
Industries: Location Behaviour and Impact on Regional Change in the USA and
Canada). Berlin: Springer.
Bathelt, H. (1991b): Employment changes and input-output linkages in key technology
industries: a comparative analysis. Regional Studies 25: 31-43.
Bathelt, H. and Boggs, J. S. (2003): Towards a reconceptualization of regional
development paths: is Leipzig’s media cluster a continuation of or a rupture with
the past? Economic Geography 79: 265-293.
Bathelt, H. and Glückler, J. (2003): Wirtschaftsgeographie: Ökonomische Beziehungen in
räumlicher Perspektive (Economic Geography: Economic Relations in Spatial
Perspective). 2nd edition, Stuttgart: UTB – Ulmer.
Bathelt, H. and Hecht, A. (1990): Key technology industries in the Waterloo region:
Canada’s Technology Triangle (CTT). Canadian Geographer 34: 225-234.
Bathelt, H., Malmberg, A. and Maskell, P. (2004): Clusters and knowledge: local buzz,
global pipelines and the process of knowledge creation. Progress in Human
Geography 28: 31-56.
Baum, J. A. and Oliver, C. (1992): Institutional embeddedness and the dynamics of
organizational populations. American Sociological Review 57: 540-559.
Bercovitz, J. E. L. and Feldman, M. P. (2007): Fishing upstream: firm innovation strategy
and university research alliances. Research Policy 36: 930-948.
Black, D. (1988): Canada’s Technology Triangle. En Route (August): 30-32, 44, 46 and
52.
BMO Capital Markets (2008): Waterloo Region and Guelph: Short-term Pain, Long-term
Gains. Toronto: Bank of Montreal, Economic Research (URL:
http://www.techtriangle.com/UploadedFiles//2008-02-12_BMO_Waterloo-
Guelph_Econ_Outlook.pdf, date accessed February 25, 2008).
- 52 -
Boschma, R. A. (2005): Proximity and Innovation: a critical assessment. Regional
Studies 39: 61-74.
Brady, D. (2004): RIM’s Lazaridis. Interview with Mike Lazaridis. Business Week
Online (URL:
http://www.businessweek.com/magazine/content/04_16/b3879097.htm
, date
accessed April 27, 2008).
Bramwell, A., Nelles, J. and Wolfe, D. A. (2004): Knowledge, Innovation and Regional
Culture in Waterloo’s ICT Cluster. Paper presented at the Innovation Systems
Research Network (ISRN) National Meeting in Vancouver, 12-15 May (URL:
http://www.utoronto.ca/isrn/documents/KW%20ICT%20cluster%20paper_ISRN.
pdf, date accessed February 5, 2008).
Bramwell, A., Nelles, J. and Wolfe, D. A. (2008): Knowledge, innovation and
institutions: global and local dimensions of the ICT cluster in Waterloo, Canada.
Regional Studies 42: 101-116.
Bramwell, A. and Wolfe, D. A. (2008): Universities and regional economic development:
the entrepreneurial University of Waterloo. Research Policy 37: forthcoming.
Bunnell, T. G. and Coe N. M. (2001): Spaces and scales of innovation. Progress in
Human Geography 25: 569-589.
Chevreau, J. (1988): Caution: future under construction. Challenges (March): 14-17.
Chiesa, V. and Piccaluga, A. (2000): Exploitation and diffusion of public research: the
case of academic spin-off companies in Italy. R&D Management 30: 329-339.
Cities of Cambridge, Waterloo, Kitchener and Guelph (1988): Canada's Technology
Triangle: Economic Profile. Brochure. Kitchener.
Colapinto, C. (2007): A Way to Foster Innovation: A Venture Capital District. From
Silicon Valley and Route 128 to Waterloo Region. Working Paper n. 2007-24.
Dipartimento Di Scienze Economiche Aziendale e Statistiche, Universita Degli
Studi Di Milano (URL: http://www.economia.unimi.it/uploads/wp/Colapinto-
2007_24.pdf, date accessed March 23, 2008).
Communitech (2006): State of the Industry – Technology in the Waterloo Region: Report
2006. Waterloo.
Conestoga College (2003): The Economic and Social Value of Conestoga College.
Brochure. Kitchener.
- 53 -
Cooper, A. C. (1971): Spin-offs and technical entrepreneurship. IEEE Transactions on
Engineering Management 18: 2-6.
CTV.ca (2005): Bill Gates draws a crowd at Waterloo University. URL:
http://www.ctv.ca/servlet/ArticleNews/story/CTVNews/20051013/billgates_water
loo_20051013/20051013?hub=Canada, date accessed April 27, 2008.
de Jong, M. W. (1987): New Economic Activities and Regional Dynamics. Nederlandse
Geografische Studies – No. 38. Amsterdam.
English, J. and McLaughlin, K. (1983): Kitchener: An Illustrated History. Waterloo:
Wilfrid Laurier University.
Feldman, M. (2001): The entrepreneurial event revisited: firm formation in a regional
context. Industrial and Corporate Change 10: 861-891.
Feldman, M., Francis, J. and Bercovitz, J. (2005): Creating a cluster while building a
firm: entrepreneurs and the formation of industrial clusters. Regional Studies 39:
129-141
Fielding, D. (2008): Leaps of faith. Globe and Mail, Report on Business 24 (10): 44-49.
Garvin, D. A. (1983): Spin-offs and the new firm formation process. California
Management Review 25: 3-20.
Gertler, M. S. (2004): Manufacturing Culture: The Institutional Geography of Industrial
Practice. Oxford, New York: Oxford University Press.
Hannan, M. T. and Freeman, J. (1977): The population ecology of organizations.
American Journal of Sociology 82: 929-964.
Hannan, M. T. and Freeman, J. (1984): Structural inertia and organizational change.
American Sociological Review 49: 149-164.
Hannan, M. T. and Freeman, J. (1993): Organizational Ecology. Cambridge (MA),
London: Harvard University Press.
Hayter, R. (1997): The Dynamics of Industrial Location: The Factory, the Firm and the
Production System. Chicester, New York: Wiley.
Holbrook, J. A. and Wolfe, D. A. (2005): The innovation system research network: a
Canadian experiment in knowledge management. Science and Public Policy 2 (1):
109-118.
Intelligent Community Forum (2007): The Top Seven Intelligent Communities of 2007.
New York (URL:
- 54 -
http://www.intelligentcommunity.org/associations/5315/files/TopSeven-ICs-
2007.pdf, date accessed November 25, 2007).
Jeppesen, L. B. and Frederiksen, L. (2006): Why do users contribute to firm-hosted user
communities? The case of computer-controlled music instruments. Organization
Science 17 (1): 45-63.
Kenney, M. and Patton, D. (2005): Entrepreneurial geographies: support networks in
three high-technology industries. Economic Geography 81: 201-228.
Keune, E. J. and Nathusius, K. (1977): Technologische Innovation durch
Unternehmensgründungen: Eine Literaturanalyse zum Route 128 Phänomen
(Technological Innovation through Start-up Firms: A Literature Study of the
Route 128 Phenomenon). BIFOA Forschungsberichte – No. 77/4. Köln: Wison.
Kieser, A. and Woywode, M. (1999): Evolutionstheoretische Ansätze (Evolutionary
approaches). In: Kieser, A. (ed.): Organisationstheorien (Organization theories).
Stuttgart: Kohlhammer, pp. 253-285.
Leslie, S. W. (1993): The Cold War and American Science: The Military–Industrial–
Academic Complex at MIT and Stanford. New York: Columbia University Press.
Livingston, J. (2007): Founders at Work: Stories of Startups’ Early Days. NewYork:
Apress.
Lundvall, B.-Å. and Johnson, B. (1994): The learning economy. Journal of Industry
Studies 1: 23-42.
Malecki, E. J. (1991): Technology and Economic Development: The Dynamics of Local,
Regional, and National Change. Burnt Mill: Longman.
Maskell, P. and Malmberg, A. (1999): The competitiveness of firms and regions:
‘ubiquitification’ and the importance of localized learning. European Urban and
Regional Studies 6: 9-25.
Mossig, I. (2000): Räumliche Konzentration der Verpackungsmaschinenbau-Industrie in
Westdeutschland: Eine Analyse des Gründungsgeschehens (Spatial Concentration
of the Packaging Machinery Industry in West Germany: An Analysis of Firm
Formation Processes). Münster: Lit.
Mustar, P., Renault, M., Colombo, M., Piva, E., Fontes, M., Lockett, A., Wright M.,
Clarysse, B. and Moray, N. (2006): Conceptualising the heterogeneity of
- 55 -
research-based spin-offs: a multi-dimensional taxonomy. Research Policy 35:
289–308.
Niosi, J. (2000): Canada’s National System of Innovation. Montreal, Kingston: McGill-
Queen’s University Press.
Nooteboom, B. (2000): Learning and Innovation in Organizations and Economies.
Oxford: Oxford University Press.
Oakey, R. (1985): High-technology industry and agglomeration economies. In: Hall, P.
and Markusen, A. R. (eds.): Silicon Landscapes. Boston: Allen and Unwin, pp.
94-117.
Open Text Corporation (2001): Ten Years of Innovation 1991-2001. Waterloo.
Owen-Smith, J. and Powell, W. W. (2004): Knowledge networks as channels and
conduits: the effects of spillovers in the Boston biotechnology community.
Organization Science 15: 2-21.
Parker, P. (2001): Local–global partnerships for high-tech development: integrating top-
down and bottom-up models. Economic Development Quarterly 15: 149-167.
Perez Perez, M. and Sanchez, A. M. (2003): The development of university spin-offs:
early dynamics of technology transfer and networking. Technovation 23: 823–
831.
Pirnay, F., Surlemont, B. and Nlemvo, F. (2003): Toward a typology of university spin-
offs. Small Business Economics 21: 355-369.
Porter, M. E. (2000): Locations, clusters, and company strategy. In: Clark, G. L.,
Feldman, M. P. and Gertler, M. S. (eds.): The Oxford Handbook of Economic
Geography. Oxford: Oxford University Press, pp. 253-274.
PricewaterhouseCoopers (2001a): Waterloo Region/Canada’s Technology Triangle 2001:
Techmap. Waterloo.
PricewaterhouseCoopers (2001b): University of Waterloo: Regional Economics Benefits
Study. Waterloo.
Rappert, B., Webster, A. and Charles, D. (1999): Making sense of diversity and
reluctance: academic–industrial relations and intellectual property. Research
Policy 28: 873–890.
Rasmussen, E., Moen, Ø. and Guldbransen, M. (2006): Initiatives to promote
commercialization of university knowledge. Technovation 26: 518–533.
- 56 -
Roberts, E. B. (1968): Entrepreneurship and technology: a basic study of innovators; how
to keep and capitalize on their talents. Research Management 11: 249-266.
Rogers, E. M. and Larsen, J. K. (1984): Silicon Valley Fever: Growth of High-
Technology Culture. New York: Basic Books.
Rutherford, T. D. (1996): The local solution? The Schumpeterian workfare state, labour
market governance and local boards for training in Kitchener, Ontario. Regional
Studies 30: 413-427.
Rutherford, T. and Holmes, J. (2008): Engineering networks: university–industry
networks in southern Ontario automotive industry clusters. Cambridge Journal of
Regions, Economy and Society 1: forthcoming.
Saxenian, A. (1985): The genesis of Silicon Valley. In: Hall, P. and Markusen, A. (eds.):
Silicon Landscapes. Boston: Allen and Unwin, pp. 20-34.
Smilor, R. W., Gibson, D. V. and Dietrich, G. B. (1990): Spin-out companies: technology
start-ups from UT-Austin. Journal of Business Venturing 5: 63–76.
Staber, U. (1997): An ecological perspective on entrepreneurship in industrial districts.
Entrepreneurship and Regional Development 9: 45-64.
Stankiewicz, R. (1994): University firms, spin-off companies from universities. Science
and Public Policy 21: 99-107.
Statistics Canada (2001a): Census of Canada 2001: Profile Series, Cumulative Files.
Catalogue No. 95F0495XCB01001 and 95F0495XCB01004. Ottawa: Statistics
Canada.
Statistics Canada (2001b): Industry – 1997 North American Industry Classification
System, Class of Worker and Sex for Labour Force 15 Years and Over. Catalogue
No. 97F0012XCB01009. Ottawa: Statistics Canada.
Statistics Canada (2006a): Census of Canada 2006: Profile Series, Cumulative Files.
Catalogue No. 94-581-XCB2006001 and 94-581-XCB2006004. Ottawa: Statistics
Canada.
Statistics Canada (2006b): Industry – 2002 North American Industry Classification
System, Class of Worker and Sex for Labour Force 15 Years and Over. Catalogue
No. 97-559-XCB2006009. Ottawa: Statistics Canada.
Statistics Canada (2006c): Canadian Business Patterns, 1998-2005. Catalogue No.
61F0040XCB. Ottawa: Statistics Canada.
- 57 -
Storper, M. and Venables, A. J. (2004): Buzz: face-to-face contact and the urban
economy. Journal of Economic Geography 4: 351-370.
Storper, M. and Walker, R. (1989): The Capitalist Imperative: Territory, Technology,
and Industrial Growth. New York, Oxford: Basil Blackwell.
Stuart, T. E., Huang, H. and Hybels, R. C. (1999): Interorganizational endorsements and
the performance of entrepreneurial ventures. Administrative Science Quarterly 44:
315– 349.
Toronto Star (2007): NCR Corp. to cut 450 jobs at its Waterloo operations. January 12.
Toronto Star (2008): Sales and profits fall. March 6.
Van Alphen, T. (2008): 1,200 Kitchener jobs going. Toronto Star, February 12.
Vohora, A., Wright, M. and Lockett A. (2004): Critical junctures in the development of
university high-tech spinout companies. Research Policy 33: 147-175.
Wahl, A. (2007): Innovation station. Canadian Business, October 8.
Wolfe, D. A. (2007): The Nature and Development of the Waterloo ICT Cluster. Paper
prepared for the Local Economic and Employment Development (LEED)
Programme, Organisation for Economic Co-operation and Development (OECD)
in Paris.
Xu, S. X. (2003): Knowledge Transfer, Inter-Firm Networking and Collective Learning
in High Technology Cluster Evolution: A Network Analysis of Canada’s
Technology Triangle. Master Thesis. Waterloo: University of Waterloo.
Zook, M. A. (2002): Grounded capital: venture financing and the geography of the
Internet industry, 1994-2000. Journal of Economic Geography 2: 151-177.
- 58 -
Figure 1. Regional ecologies of technological change and growth
- 59 -
Table 1. Socio-economic indicators of the Kitchener CMA and Guelph CA in relation to
the Toronto CMA, Ontario and Canada (Sources: Statistics Canada 2001a; 2006a)
Socio-Economic Indicators Canada Ontario Toronto
CMA
1)
Kitchener
CMA
1)
Guelph
CA
2)
Population growth,
2001-2006
5.4% 6.6% 9.2% 8.9% 8.2%
Unemployment rate, 2006 6.6% 6.4% 6.7% 5.6% 5.1%
Average household income,
CAD 2005
69,550 77,970 87,820 78,220 77,920
Employment growth,
2001-2006
8.2% 8.0% 9.4% 10.7% 9.7%
- Construction 21.6% 15.8% 19.7% 19.7% 14.4%
- Manufacturing -7.7% -8.6% -6.2% -3.5% 2.9%
- Finance and insurance 8.4% 8.1% 9.3% 14.3% -0.4%
- Professional, scientific and
technical services
14.3% 9.9% 8.5% 23.9% 19.0%
- Administrative and
support, and other services
19.3% 22.2% 19.9% 36.0% 42.3%
- Educational services 12.7% 16.8% 20.1% 20.3% 8.9%
Notes:
1)
CMA = census metropolitan area;
2)
CA = census agglomeration
- 60 -
Table 2. Typology of spin-offs and start-ups according to university sponsorship and
involvement in firm formation processes
University involvement in spin-off and start-up processes University
sponsorship
University research University–industry
joint ventures
Decentralized idea
development
Sponsored
spin-offs
Intellectual property
(IP) development at
the university
through publicly
funded research
grants; as a part of
standard university
operations
Formal development
agreement between
university and industry;
typically involves
preferential licensing
rights for IP that the
research generates for
the sponsoring firm
Firms started by former
graduates or
undergraduate students
after they finish school;
might have developed a
business idea from their
joint classroom
experience
Unsponsored
spin-offs
Researcher develop
an idea within the
university; pay for
the IP and then leave
the university to
develop it further;
possibly without
support of the
university
Off-site, unsolicited
innovation brought
forward by someone in
the research group and
developed into a
product; possibly an
ancillary development
not central to the study
or research project that
is subsequently
developed
Completely self-
developed firms;
principals meet
informally, off-site or
outside campus
academic facilities but
have social ties with the
university; typically the
university is not aware
of these processes
- 61 -
Table 3. Typology of start-ups according to the character of university knowledge
applied and co-localization of the founders
Co-localization of start-up founders Character of
university
knowledge
applied
Co-localized Not co-localized
Generic,
broad
knowledge
Broad epistemic knowledge,
largely based on the capabilities
and focus of the local incubator
university; limited potential for
innovation
Broad epistemic knowledge
drawing from a wider set of
experiences at different places;
innovation benefits from broader
access to generic knowledge pools
Specific
knowledge
Drawing on specific knowledge in
the university’s competencies,
including tacit knowledge pools
(particularly in dynamic
technology fields)
Drawing from different specific
knowledge pools (e.g., different
research projects/specializations);
large potential for innovation;
access to different specialized
regional knowledge pools
- 62 -
Table 4. Employment and establishments by industry group in the Kitchener and Guelph
metropolitan areas, 2001 – 2006 (Sources: Statistics Canada 2001b; 2006b; 2006c)
NAICS
1
industry groups Employees Establishments
abs
2001
abs
2006
%
2001-06
abs
2001
abs
2006
%
2001-06
11 Agriculture, forestry, fishing
and hunting
2,850 3,135 10.0% 25 217 -3.6%
21 Mining and oil and gas extraction 270 390 44.4% 17 18 5.9%
22 Utilities 1,250 1,665 33.2% 27 26 -3.7%
23 Construction 15,935 18,910 18.7% 1,777 1,908 7.4%
31-33 Manufacturing 76,500 74,915 -2.1% 1,556 1,477 -5.1%
313 Textile mills 900 425 -52.8% 18 11 -38.9%
315 Clothing manufacturing 2,275 775 -65.9% 30 22 -26.7%
316 Leather and allied product
manufacturing
290 120 -58.6% 7 9 28.6%
325 Chemical manufacturing 2,405 1,960 -18.5% 50 44 -12.0%
326 Plastics and rubber products
manufacturing
5,255 6,445 22.6% 86 83 -3.5%
334 Computer and electronic
product manufacturing
4,515 5,460 20.9% 70 66 -5.7%
335 Electrical equipment, appliance
and component manufacturing
4,630 3,700 -20.1% 55 49 -10.9%
41 Wholesale trade 13,605 16,365 20.3% 1,089 1,110 1.9%
44-45 Retail trade 31,595 34,885 10.4% 1,942 2,108 8.5%
48-49 Transportation/warehousing 11,725 12,805 9.2% 530 586 10.6%
51 Information and cultural industries 5,925 6,430 8.5% 169 149 -11.8%
52 Finance and insurance 14,875 16,640 11.9% 589 724 22.9%
53 Real estate and rental and leasing 4,105 5,135 25.1% 609 665 9.2%
54 Professional, scientific and
technical services
16,425 20,150 22.7% 1,755 1,924 9.6%
55 Management of companies and
enterprises
1,125 390 -65.3% 278 188 -32.4%
56 Administrative and support, waste
management/remediation services
9,930 13,602 37.2% 708 803 13.4%
61 Educational services 23,545 27,480 16.7% 176 218 23.9%
62 Health care and social assistance 22,955 26,980 17.5% 1,226 1,353 10.4%
71 Arts, entertainment and recreation 4,710 4,650 -1.3% 228 221 -3.1%
72 Accommodation and food services 16,790 18,820 12.1% 940 946 0.6%
81 Other services (except public
administration)
12,825 13,795 7.6% 1,594 1,611 1.1%
91 Public administration 9,365 10,190 8.8% 20 13 -35.0%
Total 296,305 327,332 10.5% 15,255 16,265 6.6%
Notes:
1
NAICS = North American Industry Classification System
- 63 -
Table 5. University-related spin-off/start-up firms in the Kitchener/Guelph metropolitan
areas
Firm category University-
related start-ups
(number)
University of
Waterloo spin-
offs
(number)
University of
Guelph spin-
offs
(number)
Total
(number)
– A. Sectoral split-up –
- All firms 227 47 14 288 (100.0%)
- Non-IT firms 133 23 13 169 ( 58.7%)
- IT firms 94 24 1 119 ( 41.3%)
– B. IT firm status/trajectory –
- IT firms 94 24 1 119 (100.0%)
- Firms
confirmed
closed
14 3 0 17 ( 14.3%)
- Firms not
found
16 0 0 16 ( 13.4%)
- Firms not in
region
30 6 0 36 ( 30.3%)
- Firms acquired
by other IT
firms
6 2 0 8 ( 6.7%)
- Relevant IT
firm population
28 13 1 42 ( 35.3%)
– C. Sample statistics –
Relevant
population (N)
28 13 1 42 (100.0%)
Contacted firms 24 8 0 32 ( 76.2%)
Rejections 11 3 0 14 ( 33.3%)
Interviews 13 5 0 18 ( 42.9%)
Note: University-related start-ups were identified from the PricewaterhouseCoopers
(2001a) techmap; lists of university spin-offs were provided by the universities.
- 64 -
Table 7. Importance of suppliers and customers in innovation for IT-related University
of Waterloo spin-off/start-up firms, 2007/08
Number of responses with the score Importance of suppliers and consumers in
innovation
1 2 3 4 5
Rating of suppliers 5 4 3 1 2
Rating of customers 0 0 3 4 8
Note: Firms were asked to rate the importance of suppliers and customers in innovation
processes on a scale from “1”= unimportant to “5”= critical.
- 65 -
Table 8. Information sources in product design and problem solving for IT-related
University of Waterloo spin-off/start-up firms, 2007/08
Information sources in problem solving/market
development
Responses of IT-related UW spin-offs/ start-
ups
Number Share
Internet as a source for innovation 12 66.7%
Of these:
Internet is a key source for solutions 9 50.0%
Peers (technology issues) 1 5.6%
Peers (business related) 3 16.7%
Public R&D Labs - -
Consultants 3 16.7%
Internally in the firm 6 33.3%
Of these:
Other global sites within the company 3 16.7%
Universities 4 22.2%
Customers and suppliers 5 27.8%
Note: Firms were asked who they turned to for help to solve a problem related to product
design, production or market development
.
Table 6. Economic indicators of the sampled IT-related University of Waterloo spin-off/start-up firms, 2008-08
Share of sales Inter-
view
numbe
r
Date
estab-
lished
Owner
-ship
Products/
services
Regional
emp-
loyees
Share of
local
supplies
Local
sales
Other
Canadian
sales
US
sales
Other
inter-
national
sales
University
sponsorship in
start-up phase
University
involvement in
start-up phase
Specific form of university
involvement
University
collaborat-
ions at time
of interview
1 1991 Public Software 250 5% n/a n/a n/a n/a Unsponsored Decentralized No direct involvement Yes
2 1977 Public Software 84 <1% 1% 1% 60% 40% Unsponsored Decentralized No direct involvement No
3 1981 Private Instruments 145 20% 1% 13% 37% 48% Sponsored University–
industry joint
venture
Founder staff at UW; IP
from custom contract
Yes
4 1991 Public Software 9 <1% n/a n/a n/a n/a Sponsored Decentralized Initial product was spun-out
from another firm
No
5 1997 Private Software 5 n/a n/a n/a n/a n/a Unsponsored Decentralized Founders met during UW
co-op term
No
6 1984 Private Software 65 <1% 1% 9% 40% 50% Sponsored University–
industry joint
venture
Founder’s undergraduate
project at UW was basis for
firm
No
7 1991 Public Software 12 <1% 2% 13% 60% 25% Unsponsored University
developed
Founder was staff and
bought IP
No
8 1984 Private Software 6 10% 5% 85% 5% 5% Unsponsored University
developed
Founder was staff and
bought IP
No
9 1989 Public Software 400 <1% 0% 3% 45% 52% Sponsored University–
industry joint
venture
Initial project was a UW–
industry project
Yes
10 1976 Private Software 5 <1% 0% 10% 70% 20% Sponsored University
developed
Founder was UW
researcher
No
11 1976 Private Consulting 450 <1% 0% 15% 25% 60% Unsponsored Decentralized Founders met at UW No
12 1988 Private Software 150 <1% 0% 6% 54% 40% Sponsored University
developed
Main product was a UW
project
Yes
13 1980 Public Semi-
conductors
255 10% 1% 3% 30% 66% Sponsored University
developed
Founder was a professor at
UW
Yes
14 1984 Public Comm-
unication
5200 <1% 0% 7% 60% 33% Unsponsored Decentralized No direct involvement Yes
15 1988 Public Engineering 3 100% 98% 1% 0% 1% Unsponsored Decentralized No direct involvement No
16 1998 Public Semi-
conductors
45 <1% 0% 0% 0% 100% Unsponsored Decentralized No direct involvement No
17 2004 Private Software 39 <1% 0% 50% 20% 30% Sponsored University
developed
Main product was a UW
project
No
18 2000 Private Software 20 10% 1% 20% 70% 10% Unsponsored Decentralized No direct involvement; UW
was an early client
No
