THE REGIONAL DYNAMICS OF INNOVATION:

A comparative case study of oil and gas industry

development in Stavanger and Aberdeen*

Sachi Hatakenaka, Petter Westnes,

Martin Gjelsvik, and Richard K. Lester

MIT-IPC-06-008

November 2006

*
The authors are grateful to the Research Council of Norway and IRIS (International Research Institute of
Stavanger) for their generous support of this research. We also wish to acknowledge the valuable research
assistance provided by Wei Gao of the MIT Industrial Performance Center.
The views expressed herein are the authors’ responsibility and do not necessarily reflect
those of the MIT Industrial Performance Center or the Massachusetts Institute of
Technology.
 Paper presented at the SPRU 40th Anniversary Conference on

The Future of Science, Technology and Innovation Policy: Linking Research and Practice
University of Sussex, Brighton, United Kingdom
September 11-13, 2006

The Regional Dynamics of Innovation:

A comparative case study of oil and gas industry development in

Stavanger and Aberdeen*

Sachi Hatakenaka1, Petter Westnes2, Martin Gjelsvik3, Richard K. Lester4

1
Industrial Performance Center, MIT, [email protected]
2
International Research Institute of Stavanger, [email protected]
3
International Research Institute of Stavanger, [email protected]
4
Industrial Performance Center, MIT, [email protected]

Abstract
This paper addresses a central question confronting politicians, business leaders, and
economic planners throughout the world: How can local economic communities survive
and prosper in the rapidly changing global economy? The paper reports on a
comparative case study of two key regions in the North Sea oil and gas province: the
Stavanger region on the southwest coast of Norway and the Aberdeen region in
northeast Scotland. These two regions proved an ideal setting for a matched pair
comparison, as the circumstances under which they developed into oil capitals are
strikingly similar. Yet the development of local technological and industrial capabilities
followed different paths in the two locations. On the other hand, these differences do
not appear to have led to significantly different levels of international competitiveness.
Although Stavanger and Aberdeen are characterized by very different local innovation
systems, the available evidence suggests that outcomes have been similar along
significant dimensions of industry performance.

Keywords: Regional innovation systems, oil and gas, comparative case study

* The authors are grateful to the Research Council of Norway and IRIS (International
Research Institute of Stavanger) for their generous support of this research. We also wish to
acknowledge the valuable research assistance provided by Wei Gao of the MIT Industrial
Performance Center.

1
Introduction
This paper addresses a central question confronting politicians, business leaders,
and economic planners throughout the world: How can local economic communities
survive and prosper in the rapidly changing global economy? An increasingly common
response to this question in recent years has been to focus on ways to strengthen local
capabilities for innovation. By ‘capabilities for innovation’, we mean the ability to
conceive, develop, and/or produce new products and services, to deploy new
production processes, and to improve on those that already exist.
Local and regional innovation systems have attracted much attention from policy
makers as well as academics in recent years, with several strands of literature developing
(Breschi & Lissoni, 2001; Cumbers & McKinnon, 2004; Doloreux & Parto, 2005;
Iammarino, 2005). The concept of ‘national innovation systems’ was introduced by
scholars of economic and technological change in recognition that a wider set of
institutions than the firms directly involved in bringing new products to market are
participants in the innovation process (Freeman, 1987; Lundvall, 1992; Nelson, 1993).
Porter introduced the notion of ‘clusters’ as key to industrial competitiveness, further
emphasizing the importance of firm interactions with supply chains and with public
science (Porter, 1990; 1998). The concept of ‘clusters’ caught the imagination of policy
makers around the globe and helped popularize the idea that ‘regions’ matter for
competitiveness. A closer examination of successful industrial districts such as Silicon
Valley, Third Italy and Baden-Wurttemberg led to the idea that local interactions played
a special role in such regions (Piore & Sabel, 1984; Saxenian, 1994) and the ‘region’ was
soon proposed as an alternative level of analysis (Acs & Varga, 2002; Braczyk, Cooke,
& Heidenrich, 1998; Cooke, 1992). Economists joined the debate with their renewed
interest in economic geography and the role of geographic specialization (Krugman,
1991), and ‘knowledge spillovers’ were proposed as key mechanisms for agglomeration
dynamics (Audretsch & Feldman, 1996; Jaffee, 1989; Jaffee, Trachtenberg, &
Henderson, 1993).
More recently, there has been a growing understanding that the diverse nature of
regions may preclude ‘one-size-fit-all’ solutions or general ‘best practices’ for regions,
and the need for policy recommendations to be differentiated is increasingly recognized
(Cooke & Schienstock, 2000; Tödtling & Trippl, 2005). Various typologies of regional
innovation systems have been proposed based on reviews of multiple case studies
(Asheim & Isaksen, 2002; Cooke, Heidenreich, & Braczyk, 2004) and the presence of
different configurations of competition, collaboration and cooperation has been
recognized (Polenske, 2004). But the lack of comparability across regions meant that
most of these studies merely juxtaposed different systems, which prevented claims
from being made about relative economic performance. In the end, the underlying
assumptions about ‘successful’ regional innovation systems have remained surprisingly
uniform in the literature, powerfully shaped by the image of Silicon Valley (Saxenian,
1994). Thus, in spite of the wide recognition that regions are pursuing their own
developmental paths, with only limited ability to change course (Heidenreich, 2004), the
goal for all regions is to develop the same kind of regional innovation systems based on
strong local interactions and the culture of cooperation and competition, as suggested
by a handful of successful cases (Cooke & Morgan, 1998). This is perhaps not
surprising given the analytical history of the regional innovation system framework, in
which ‘regions’ were proposed as a critical level of analysis precisely because of the
importance of local interactions (Cooke, Uranga, & Etxebarria, 1997; Doloreux, 2002).

2
 If regional innovation systems are different systems, each following its own
evolutionary path, how do these systems evolve, how do they differ from each other,
and what are the implications for economic and innovation performance at the regional
level? While these issues have not been much explored in the past, the literature does
provide helpful clues as to the circumstances under which different innovation systems
are most likely to manifest themselves.
First, there have been discussions about how regions may connect to global
industry in different ways (Acs et al., 2002), which in turn may demand different
configurations of local capabilities and institutions. Regions may connect to
multinational corporations in different ways as the latter make decisions about their
locations (Cantwell & Iammarino, 2003). Silicon Valley’s dynamism was enhanced by
its linkages to other ‘hubs’ such as in Taiwan (Saxenian & Jinn-Yuh, 2001), and indeed a
broader review of New Silicon Valleys showed that connection to sizable markets, in
this case in the US, was a significant factor in their emergence as innovative regions
(Bresnahan, Gambardella, & Saxenian, 2001). If it is possible for regions to develop
different ‘innovation systems’ because of differently configured links with global
industry, it may be worth exploring the influence of such links on the development of
these systems.
Second, the conditions of success for any region change over time as circumstances
both internal and external to the region change. National contexts for innovation
change over time (Freeman, 2002). The factors important for initiating agglomeration
dynamics may be different from those that keep the process going (Bresnahan et al.,
2001). Once agglomeration sets in, a different concern about ‘lock-in’ may arise and
demand different goals and strategies (Isaksen & Remøe, 2001; Tödtling et al., 2005).
Regions may thus experience different demands for their capabilities over time.
Consequently it may be worth studying the histories of regions as they grow and mature
(Iammarino, 2005).
Finally, since the conditions for successful innovation mechanisms/systems vary
across sectors and technologies (Mowery & Nelson, 1999; Nelson, 2000), it may be
important to control for these effects by exploring inter-regional variations within the
same industrial sector.
In this paper we report on a comparative analysis of the development of the oil and
gas industry in Stavanger, Norway and Aberdeen, Scotland, two regions which
emerged as ‘oil capitals’ in the North Sea in the past 40 years (Hatakenaka, Westnes,
Gjelsvik, & Lester, 2006). This study is part of a larger research project based at the
MIT Industrial Performance Center on the dynamics of innovation system
development at the local level, the Local Innovation Systems Project (Lester, 2005).
The oil and gas industry is among the most technology intensive and highly globalized
of industries. A comparison of Stavanger and Aberdeen affords a valuable analytical
opportunity as the circumstances under which the two regions developed into oil
capitals are strikingly similar. The two regions developed over the same period,
interacted initially with the same group of global oil companies, and faced similar
market conditions and geological and technological challenges. Both regions had to
struggle with the issue of how to build local capabilities in the context of an already
well-established global industry, and both were able to develop industrial complexes of
similar size. Oil production in the Aberdeen region has peaked, and in Stavanger will
shortly peak, and both regions are today confronting the challenge of how to sustain
local innovative capabilities as depletion of the natural resource continues.

3
 But the development of local technological and industrial capabilities in the two
regions has followed very different paths. In Norway, the national, regional and local
authorities made concerted efforts to develop local capabilities in the oil and gas
industry, and to concentrate industry-related institutions in Stavanger. In contrast, the
industry in Aberdeen grew despite a lack of consistent support from the national and
local authorities.
These differences in the institutional and policy environment, though considerable,
do not appear to have led to significant differences in the international competitiveness
of the two local industries, at least so far. Stavanger and Aberdeen are characterized by
very different local innovation systems, but both appear to have enjoyed similar
successes in the race to internationalize and export their expertise to other oil
provinces.
In short, these two regions provide an ideal matched pair for analyzing the origins
and consequences of different approaches to building local systems of innovation.

Method
Our approach was to conduct a comparative case analysis using multiple sources of
data. We relied primarily on semi-structured interviews, supplemented by industry
statistics, and the secondary literature on these two regions and the oil industry more
generally. It is widely recognized that carefully constructed comparative case studies are
valuable for the study of complex socio-economic systems (Asheim, 2002; Doloreux,
2002), particularly in unravelling causal links and underlying mechanisms (Markusen,
1999).
The comparative approach is inspired by the method of structured comparison of
two similar cases proposed originally by Mill (1843; Skocpol & Somers, 1980). The main
benefit of comparison in this case, however, is not so much to enhance the
generalizeability of findings, but to inject rigor into the qualitative understanding of each
of the cases, particularly by drawing contrasts between them. This is a particularly
powerful approach in well-matched cases, where the circumstantial similarities help
highlight the differences in terms of paths taken as well as paths not taken.
We conducted a total of 71 interviews as a key method to explore firm and
institutional dynamics. Interviewees were identified in a cascading manner starting with
key informants in local governments/industry organizations, and were carefully selected
to ensure coverage of different types of stakeholders including different types of firms;
different levels of government; universities; and public/private research institutions. 31
interviews were conducted in Stavanger with 29 key informants -- 14 representing
industry, 7 from research institutions, 4 from universities, and 3 representing industry-
related organizations. In Aberdeen, 40 interviews were conducted with 40 key
informants -- 17 representing industry, 8 from government, 9 from three universities and
6 from other industry-related organizations.
To complement the qualitative data obtained from the interviews as well as archival
sources, we also conducted studies of patenting and publications in the two regions. We
searched the US patent database for all oil and gas industry-related patents cumulatively
issuing through June 2005 which met the criterion that at least one of the inventors was
located either in the Aberdeen region or in the Stavanger region. We further identified all
oil and gas industry-related patents issuing prior to June 2005 for which at least one of

4
the assignees was located in one or other of the two regions.1 We also searched two
major petroleum-focused publication databases -- the e-library of the Society of
Petroleum Engineers (SPE) and Petroleum Abstracts -- for publications by authors
employed by research and education institutions in the two regions.
Half of our team came from RF-Rogaland Research2, a research institute in Stavanger
that is closely associated with the oil and gas industry. These team members’ industry
knowledge and proximity to key players in the region was particularly helpful for this
research. At the same time, we made conscious efforts to ensure neutrality of our
analysis (Markusen, 1999). Non-Stavanger-based team members played a role as
‘outsiders’ to avoid any ‘insider bias,’ and a total of 13 (6 in Stavanger and 7 in Aberdeen)
interviews were conducted jointly to ensure comparability of interviews. In addition, the
analytical results were tested against three key informants from each region to avoid any
biases in interpretation (Cumbers et al., 2004; Markusen, 1999).

Overview of the two cases: Stavanger and Aberdeen

Stavanger and Aberdeen developed to become significant oil industry centers, with a
similar set of industrial players and similar levels of industry employment (about 37,000
in Rogaland County, of which Stavanger is the largest municipality, and about 39,000 in
Aberdeenshire.) In each region oil and gas is by far the largest industry, and the local
industry in each region also accounts for an important share of national employment in
the oil sector. This is especially true of the Stavanger region, where nearly 50% of
Norway’s total employment in the oil and gas industry is located, but in the U.K. too,
more than 20% of total oil and gas sector employment is found in the Aberdeen region.
The overall levels of oil and gas production were similar in the two regions until
recently, although UK production peaked in 1999 for oil and in 2002 for gas, whereas
Norwegian production is expected to peak in 2006 for oil and in 2010 for gas.
Exports of oil-related services are an increasingly important performance metric for
the two regions, as local oilfields reach maturity. Here, Scotland leads Norway, with 6
billion USD in exports in 2003 compared with 4.9 billion USD from Norway, but this
margin is small. Given the fact that the majority of oil-related firms in Scotland and
Norway are localised in Aberdeen and Stavanger, it is reasonable to assume that these
national export numbers also reflect exports from the two regions.
Neither locality has managed to diversify significantly out of the oil and gas industry
to date, though there are on-going efforts, for instance, in renewable energy, supported
by respective authorities as well as industry and universities.

Different development pathways

At the outset of the North Sea oil era, both Norway and the UK confronted the
problem that they had virtually no local capabilities in the oil and gas industry. Even
though the UK was at an advantage, given the broad experience of BP and Shell, the
extraction and production of oil required a range of supply industry functions, none of
1
We defined the greater Aberdeen and greater Stavanger areas by reasonable commuting
time. In each case, the area extends roughly 30 miles from the center of the city.
2
Rogaland Research changed its name to International Research Institute of Stavanger (IRIS)
on January 1 2006.

5
which existed in either country. The first step for both countries, and both localities, had
to be to invite foreign companies – most notably the Americans.
The paths subsequently taken by the two countries were significantly different,
however. Government policies differed in three important ways: in managing the speed
of depletion (by deciding what to license); in the emphasis on domestic capacity building;
and in localization decisions.
From the earliest days, the Norwegians saw oil as a national asset to be managed
carefully. In contrast, the British government moved quickly to adopt a fast depletion
policy, prompting a larger number of foreign companies to move in (Cameron, 1986;
Cook, Lesley, & Surrey, 1983). This difference in approach was at least partly dictated by
differences in the two countries’ macroeconomic circumstances (Noreng, 1980). The
British government was preoccupied with a crippling balance of payments crisis, and
therefore needed a rapid scale-up of oil production. Norway had close to full
employment and generally healthy macroeconomic conditions. Indeed, there were real
concerns that if the development of the oil industry was left to market forces, the
relatively small Norwegian economy might be overwhelmed; it was sensible for them to
move slowly if only to avoid inflation. Environmental concerns and co-habitation with
fisheries were other issues.
Domestic capability building was a clear policy priority for Norway from early on.
This was reflected in the establishment of a national oil company, Statoil, and in
specifying licensing conditions, which often required technology transfer from foreign
companies to domestic organizations. The government was systematically evaluating and
rewarding foreign oil companies who were contributing to domestic capacity building.
Concessionary procedures were used as an instrument to force the international
companies to engage in technology transfer and local content development.
Efforts to promote local industrial development in the UK started later, changed
over time and did not go as far. On the one hand, the UK oil industry long predated the
North Sea oil discoveries and featured major international players like BP and Shell. On
the other hand, the fast depletion policy also required technical inputs that only the
American oil service companies were equipped to provide at that time. The goal of local
capability building was a secondary consideration. The national government later
pursued a ‘Buy British’ policy through the Offshore Supplies Office (OSO), and also
established a national company, the British National Oil Corporation (BNOC).
However, the UK government was also undergoing a radical change in its economic
policies; and the oil sector was seen as too transient and too small to warrant close and
consistent attention from policymakers. BNOC was dismantled shortly thereafter, and
OSO treated foreign-owned companies in the UK as ‘domestic.’
While the regional authorities played critical roles in establishing local education and
research capabilities relevant to the oil industry in Stavanger, no comparable efforts were
made in Aberdeen, which already had a well established university and a polytechnic.
Local authorities in both locations worked hard to prepare the infrastructure
necessary to attract foreign companies; but at the level of the central government there
were significant differences in location policy. In Norway, the decision was taken to
locate both Statoil and the Norwegian Petroleum Directorate (NPD) in Stavanger. In
contrast, Glasgow, which had virtually no oil industry, but which was then confronting
serious problems of unemployment and economic stagnation, was selected as the

6
location for both BNOC and OSO, even though Aberdeen was already attracting oil
related industry.

Different Innovation Systems

The different pathways taken in the development of the two regions have
contributed to the emergence of quite different local innovation systems.
For the oil and gas industry in general, collaboration and interaction between three
types of companies are critical for innovation: oil exploration and production companies
(also known in the industry as ‘operators’), who have the rights to explore oilfields and
without whose participation new technologies cannot be tested in the oilfields; integrated
service providers, large global companies capable of providing most exploration and
production-related services to oil companies; and small specialized suppliers/service
companies, which are often the pioneers in developing new technologies. While all three
types of companies are present in both regions, the mix is subtly different in each case,
leading to different patterns of interactions between them. Stavanger is generally
characterized by high levels of both industry and industry-government coordination and
collaboration, while the prevailing ethos in Aberdeen appears to be market coordination
and competition.
In Stavanger, Statoil, which has grown to become an internationally recognized oil
company known for its technology orientation, has played major roles in orchestrating
collaboration as a demanding user, as a project sponsor and as a provider of information
and expertise. Since 1991, Statoil has operated a program to develop and support
innovative supply companies, providing opportunities for the development of local
firms. However, the regional impact of this initiative is contested. In Aberdeen, although
leading operators like BP and Shell were active in earlier years, these activities have been
scaled back since the 1990s and today there is no leading corporate patron of local
innovation to match Statoil’s role in Stavanger.
While there are complaints in Aberdeen about the lack of collaborative opportunities
for innovation (Crabtree, Bower, & Keogh, 1997), Stavanger companies continue to
report high levels of collaboration. In a survey conducted in 2003, 55% of SMEs in the
Norwegian oil and gas industry found that collaborating with national oil companies was
of great importance to their own technological development, while 37% reported that
collaborations with foreign oil companies were similarly important (Kristiansen et al
2004). In a separate survey conducted in Aberdeen in 2000, only 30% of SMEs in the oil
and gas industry were collaborating with operators at all (Cumbers, Mackinnon, &
Chapman, 2003).
On the other hand, Aberdeen is characterized by a larger number of companies, with
an estimated total of 900-1000, compared with 500-600 in Stavanger (Aberdeen City
Council & Aberdeenshire Council, 2001; BI, 2003; Cumbers, 2000; Frontline
Management Consultants & Ron Botham Associates, 2000; Jakobsen, Vikesland, &
Moen, 2000; Leknes & Steineke, 2001; MacKinnon, Chapman, & Cumbers, 2001; Trends

7
Business Research, 2001)3. Aberdeen hosts a more diverse group of operators, with
nearly three times as many independent operators as Stavanger. Most of these companies
do not have internal technological capabilities, and so they are often willing partners in
trying new ways of doing business, and thus provide different opportunities for
innovation to suppliers. These differences are not necessarily permanent – it is possible
that they merely reflect differences in the maturity of oilfields, which could
narrow/disappear over time. In this regard, the number of independent operators in
Stavanger has risen rapidly in the past couple of years.
The second main difference concerns the nature of innovation. Stavanger has
developed a reputation for technology-driven innovations, while Aberdeen is known for
its operational innovations. Such reputations are consistent with industry benchmarking
data. Aberdeen leads Stavanger in terms of cost competitiveness. The Norwegian
industry is known for high costs, which in turn are partly associated with generally more
severe regulatory requirements (Kon-Kraft, 2004). On the other hand, industry
benchmarking also suggests that Norway leads the UK in the use of new technology, and
Norway is seen as ‘a test-bed for new technology.’ (Duncan, 2001). The high cost level
and a benevolent tax structure are major drivers for exploiting new technology.
Such a difference in orientation is consistent with the way Stavanger and Aberdeen
have attracted the four integrated oil and gas industry service providers. Stavanger has
become the North Sea headquarters of Schlumberger, which has a reputation as the most
scientifically oriented of the four and which has established a significant research
capability in seismic and reservoir monitoring there. Aberdeen has attracted
Weatherford, which is known for its operational orientation, and which has concentrated
its research and training activities in more operationally relevant fields. Baker Hughes,
which also set up its North Sea headquarters in Aberdeen, is focused on operational
rather than technology-related activities there. The fourth integrated service company,
Halliburton, is equally present in both regions.
Similarly, differences in the characteristics of the most visible small innovative
companies are also consistent with this difference in technological orientation. In
Stavanger, the most salient of the small innovative companies arose from local technical
research capabilities. Roxar ASA is recognised as an industry leader in specialist software
and grew directly out of the R&D unit of another local company. Hitec, another
Stavanger-based company, is known for its early innovations in computer-controlled
remote drilling, and its technology arose directly from joint industry projects undertaken
by the regional research institute RF-Rogaland Research.
In contrast, the most visible small firms in Aberdeen were founded by individuals
who had previously worked in global oil industry companies, gained significant
operational experience internationally, and developed many of their ideas directly from
such experience. Andergauge is one such example. The original idea for its innovative
drilling equipment was developed by its founder, based on his extensive international
3
The estimates emerge as means from a wide range of attempts to identify and map the
petroleum related industry in Scotland and Norway. The estimates vary depending on how
the companies are counted, which criteria are used to define the cluster, and which
geographical parts of the two regions are included in the counting. The estimates reflect total
number of companies including companies within the cluster’s core, such as oil companies,
engineering and construction, and supplier companies who depends heavily on business
from the oil and gas sector. These include not only drilling and service companies, but also
companies within maritime transport, catering, ICT, and commercial service.

8
operational experience in drilling, gained while he was employed in several US
companies. Similarly, the innovative ideas of PES, the pioneer in smart well technology,
came from its founder who had developed the concepts based on his extensive
international field experience in Schlumberger and Shell.
None of the educational or research institutions in either region have had explicit
policies to support ‘spin-offs’ until recently. When RF adopted more proactive policies in
the latter part of the 1990s, it was able to quickly establish a portfolio of 20 companies,
and discovered that a number of companies had been formed from the results of its past
research. In contrast, there is only one on-going case of a technology-based university
spin-off in Aberdeen, with other previous examples of university-related companies
mainly in non-technical fields.
As another indicator of inventive activity, we compared patenting behaviours in the
two regions. We found that the total number of US patents in oil and gas-related fields
with at least one Aberdeen-based inventor is more than twice the number associated
with Stavanger-based inventors (see Table 1.) This is an intriguing result, given the
reputation of Stavanger as a technologically driven location. Possible explanations
include differences between the regions in the degree of connectivity to the US oil and
gas industry, as well as more general differences in the propensity to seek patent
protection.

Table 1: U.S. oil and gas industry-related patents with a connection to Stavanger and
Aberdeen (cumulative through June 2005)

ABERDEEN STAVANGER

No. of Patents No. of Patents

With at least one Aberdeen assignee 177 With at least one Stavanger assignee 251

With at least one Aberdeen inventor 756 With at least one Stavanger inventor 307

Total Aberdeen-Related Patents 780 Total Stavanger-Related Patents 412

(Patents with both Aberdeen assignee (153) (Patents with both Stavanger assignee (146)
and Aberdeen inventor) and Stavanger inventor)

Source: US patent database

Table 2 suggests that one key contributor to the regional difference in patenting
patterns is the greater role of US firms in Aberdeen than in Stavanger. 366 patents, or
nearly half the total number of patents with Aberdeen-based inventors, were assigned to
US companies, whereas only 86 (28% ) of the patents with Stavanger-based inventors
were assigned to US companies. It may also be significant that the three American
integrated service companies with a presence in both locations received a considerably
larger number of patents on the basis of Aberdeen-related inventions than from
Stavanger.

9
Table 2: Location of patent assignees for Aberdeen-based and Stavanger-based
inventions (through June ‘05)

ALL PATENTS WITH AT LEAST ONE ALL PATENTS WITH AT LEAST ONE
ABERDEEN-BASED INVENTOR STAVANGER-BASED INVENTOR

Assignee located in: Assignee located in:

Aberdeen 153 (20.2%) Stavanger 146 (47.5%)

Elsewhere in the UK 217.5 (28.8%) Elsewhere in Norway 60.5 (19.7%)

United States 366.5 (48.5%) United States 86 (28%)

Other countries 19 (2.5%) Other countries 14.5 (4.7%)

Total 756 (100%) Total 307 (100%)

Source: US patent database

Another possibility is that these results reflect underlying differences in the level of
inventive activity in the two regions. A third possibility is that there are locational
differences in the propensity to patent. The latter explanation was also suggested by
comments from our interviewees. Regional differences in the propensity to patent may in
turn reflect differences in corporate ‘knowledge’ strategies (Cohen, Nelson, Welsh, &
John, 2000), and they may also reflect broader national differences in patenting behavior.
There is some evidence to support this last possibility. In 2001, for example, 3965 US
patents were granted to inventors located in the UK and 295 patents were granted to
Norwegian inventors (NSF 2004). Adjusting these figures to account for the difference
in the overall scale of inventive activity in the two countries, using reported R&D
expenditures as a proxy, UK inventors in all fields received an average of 170 U.S.
patents per billion dollars of R&D expenditures, while their Norwegian counterparts
received 96 patents per billion dollars of R&D expenditures.

Roles of universities and public research institutions

Differences between the two innovation systems, both in terms of how they evolved
and how they operate today, are exemplified by the different roles played by local
universities and research institutions. The institutions in Stavanger developed relevant
technological capabilities based on implicit and explicit coordination and collaboration
with government and industry. In Aberdeen, the universities developed ties to industry
not as a result of institutional efforts to respond to industrial needs, but rather through
the actions of individual academics in a diverse range of disciplines.
Both the university in Stavanger (UiS) and Rogaland Research (RF) were created
more or less coincident with the founding of the oil industry in the region. Established in
1969 as a regional college, UiS always saw its role as serving the educational needs of
local industry, and developed key capabilities in relevant fields such as petroleum
engineering. Working closely with the oil and gas industry to provide key educational
programs was a natural part of its mission. RF was created in 1973 by the regional

10
authorities, originally as the research arm of the college. But it soon developed into an
independent research institute with capacities to undertake applied research and testing
for the oil and gas industry. Taking part in collaborative research with industry was a
raison d’etre and an assumption since its foundation, and was aided by government
policies which often forced foreign companies to conduct research in Norway.
In contrast, Aberdeen’s two universities – the University of Aberdeen and Robert
Gordon University (RGU) -- were both well established institutions when the first North
Sea oil discoveries were made, and they were not as responsive to the technological
needs of the nascent industry. The University of Aberdeen, a 500-year old academic
institution, opted out of working with the oil and gas industry, which it saw as a
transitory presence in the region. It was only through the efforts of individual academics
from a variety of disciplines, including economics, geography, geology and zoology, that
relationships were formed with the industry. RGU, a former polytechnic, moved quickly
to meet the industry’s training needs, and continues to offer petroleum-related graduate
educational programs -- though mainly through the use of its location and the
importation of external expertise rather than by developing an internal capability.
This is not to say that no UK institution developed technical capabilities to meet the
needs of the oil and gas industry. Heriot-Watt (HW) University, located in Edinburgh,
about two hours by train from Aberdeen, developed a national and later international
reputation for petroleum engineering, but largely through its own initiatives. Once HW
had taken the lead in developing industry-relevant technological capabilities in petroleum
engineering, competition made it harder for either of the Aberdeen institutions to
develop similar fields of specialization. Imperial College also already had an established
reputation in petroleum related fields at the time of the first North Sea oil discoveries.
RF and UiS can also be contrasted with what in some respects are their institutional
twins in Trondheim, the Norwegian University of Science and Technology (NTNU) and
the independent research organization SINTEF, which were both well established by the
time the first North Sea oil discoveries were made. In this case, however, RF and UiS
moved more quickly to develop applied technological capabilities in petroleum
engineering, partly to differentiate themselves from existing institutions such as NTNU
and SINTEF.
One key difference between Stavanger and Aberdeen in the role of universities and
public research institutions is the extent to which these institutions operated to provide
‘public space’ (Lester & Piore, 2004) for their industrial partners. This difference is best
illustrated by the different ways in which the two regions developed and used similar
experimental facilities for drilling. In Stavanger, the drilling facility was developed at RF
and played a critical role in helping develop the region’s capabilities in drilling,
particularly in leveraging joint industry projects with multiple industrial partners. In
Aberdeen, a similar drilling facility was developed at a site that was not connected to
researchers, and played a role principally as a testing site for individual companies.
The differences in these roles are also suggested by the publication record in two
industry-related bibliographies: the papers contained in the E-library of the Society of
Petroleum Engineers (SPE); and Petroleum Abstracts (PA), a database administered by
the University of Tulsa.
The number of papers published in the Society of Petroleum Engineers (SPE)
database is a quantitative measure of application-oriented research relevant to industry
(see Table 3.) SPE papers generally do not report on fundamental scientific research, and

11
their coverage tends to center on exploration and production-related petroleum
engineering and technology and does not extend to related fields such as geosciences or
petroleum economics; nevertheless, the SPE database constitutes one of the most
important bodies of codified knowledge for the industry.
Petroleum Abstracts (PA) provides a more comprehensive measure of research
contributions covering a broader range of fields relevant to exploration and production,
including geosciences, social sciences and economics. In addition, while the SPE
database focuses on applied technology, the PA database covers the whole spectrum of
publications from basic science to applied technology.
Table 3 shows that the two institutions in Stavanger published 308 papers in SPE
between 1990 and 2004, many more than the 70 produced by the two Aberdeen
institutions, but roughly the same as Heriot Watt. The pattern of authorship also varies.
The majority (around 2/3) of the papers from University of Stavanger and RF were
written in collaboration with others. For the University of Aberdeen and Robert Gordon
University it is the other way around: around 60% of the papers were written solely by
their own faculty.

Table 3: SPE-papers from key research institutions in Norway and the UK, 1990-2004 (% of total papers
in parentheses)

Norway UK

Stavanger Aberdeen

UiS RF SINTEF NTNU UofA RGU Heriot Imperial

Watt College

Written solely by in-house 38 74 34 18 24 16 166 92

researchers (38%) (36%) (35%) (24%) (55%) (62%) (54%) (59%)
Written in collaboration with 63 133 63 58 20 10 144 65
others (62%) (64%) (65%) (64%) (45%) (38%) (46%) (41%)
Total 101 207 97 76 44 26 310 157
Regional total 308 173 70 - -

Source: estimated from SPE database

More generally, researchers at the Norwegian institutions are more ‘collaborative’

than their British counterparts. This difference is robust even when collaboration
between twin institutions like RF and UiS is netted out (only 5% of the Stavanger papers
were written by authors from these two institutions collaborating solely with each other.)
In general, the patterns of co-authorship suggest that Norwegian researchers may be
more collaborative across organizational boundaries than their UK counterparts.
The two Stavanger institutions can also be compared with the two institutions in
Trondheim. Here, we might have expected to find that SINTEF, a public research
institute, together with NTNU would have a far greater number of publications than RF
and UiS, given their longer history and larger size (SINTEF has about 290 employees in
their Petroleum and Energy Division, compared with about 90 in RF Petroleum). While

12
the pattern of collaboration is similar in Trondheim, with about two thirds of the papers
co-authored with authors from other institutions, the total number of publications is less
than 60% of that of the Stavanger institutions.
Table 4 shows that the contribution from the University of Aberdeen is much greater
when the wider range of subjects covered by PA is included. The contribution of Heriot
Watt is larger, as before, but on this broader measure the contribution of Imperial
College is larger still, and by a substantial margin. Among the Norwegian institutions,
SINTEF and NTNU are together more than twice as prolific as RF and UiS.

Table 4: Number of Publications in Petroleum Abstracts (1965-2005)

Institution Publications in Petroleum Abstracts

UiS 248

RF 574

NTNU 1020

SINTEF 810

Aberdeen 586

RGU 48

Heriot Watt 1073

Imperial College 1473

Source: Petroleum Abstracts

These data confirm our qualitative findings that the Stavanger institutions specifically
developed application-oriented research capabilities in petroleum engineering fields, on a
scale comparable to Heriot Watt and on a larger scale than that of the University of
Aberdeen (or indeed SINTEF/NTNU or Imperial College). However, once a broader
array of academic subjects is included, the University of Aberdeen is at least as prolific in
producing industry-relevant research, and the contributions from other institutions such
as NTNU/SINTEF, Heriot Watt and Imperial College are correspondingly larger. A
striking difference was found in the level of collaboration across institutional boundaries;
Norwegian institutions were consistently more collaborative with industry than their UK
counterparts – suggesting that they may have been more effective in providing ‘public
space’ to their industrial partners.

Discussions and conclusions

Stavanger and Aberdeen have clearly taken different paths in exploring and
exploiting the opportunities arising from the North Sea oil and gas resources to develop
local capabilities. In Stavanger these capabilities developed as a result of collaboration
and coordination, orchestrated and supported powerfully by the national and local
governments. In contrast, in Aberdeen, the past developments are better characterized
as market-based, where the most powerful role was played by competition.

13
 The Stavanger story looks in many ways like a textbook case of how to build local
innovative capabilities; the Norwegians set ‘Norwegianization’ as a policy goal and went
about its implementation in a systematic and consistent manner. In contrast to the UK
experience, the Norwegians were perhaps fortunate that key contextual factors such as
macroeconomic conditions were aligned with their policy goals.
Whatever the cause, the result has been that the two regions have developed
different innovation systems, with different characteristics and different strengths.
Interestingly, these differences in development strategies and innovation systems do not
appear to have led to significantly different levels of international competitiveness.
Rather, the two regions are competitive in different ways. If Aberdeen appears to have
some advantages in terms of operational costs, Stavanger shows its competitiveness in its
ability to introduce and use new technologies. In the race to internationalize and export
their expertise to other oil provinces, both localities appear similarly successful thus far
and have each seen rapid increases in the level of exports in recent years, though
Aberdeen leads in overall export volume. Neither region shows significant signs of
diversification into other industries. All in all, our measures of ‘industrial
competitiveness’ do not suggest that either region is the clear winner – an interesting and
even surprising result, given the significant differences in the underlying local innovation
systems.
There are two possible reasons for such a finding. The first is that our measures are
simply inadequate to capture performance differences that may in fact exist. The second
and, we suspect, more likely explanation is that differences in local innovation systems
and practices may be associated with similar performance outcomes over a sustained
period. If true, this would add significantly to current understandings of innovation
systems. Both possible explanations call for continued close study of the dynamics of
local innovation systems.

References
Aberdeen City Council & Aberdeenshire Council. 2001. Oil and Gas Prospects - 2001 Update, Report
published in August 2001.
Acs, Z. J. & Varga, A. 2002. Introduction to the special issue on regional innovation systems. International
Regional Science Review, 25(1): 3-7.
Asheim, B. 2002. Nordic SME-Forum-Regional Systems of Innovation: Nordisk Industrifond, Oslo.
Asheim, B. T. & Isaksen, A. 2002. Regional innovation systems: The integration of local "sticky" and
global "ubiquitous" knowledge. Journal of Technology Transfer, 27(1).
Audretsch, D. B. & Feldman, M. P. 1996. R&D Spillovers and geography of innovation and production.
American Economic Review, 86(3): 630-640.
BI. 2003. Kartlegging av olje- og gassnæringens betydning for norske regioner [Mapping of the oil and gas
industry's implication for Norwegian regions], CD-rom published by BI - The Norwegian School of
Management, Senter for verdiskapning.
Braczyk, H.-J., Cooke, P., & Heidenrich, H. (Eds.). 1998. Regional Innovation Systems. London: UCL
Press.
Breschi, S. & Lissoni, F. 2001. Knowledge Spillovers and Local Innovation Sytems: A critical survey.
Industrial and Corporate Change, 10(4).

14
Bresnahan, T., Gambardella, A., & Saxenian, A. 2001. ‘Old Economy’ Inputs for ‘New Economy’
Outcomes: Cluster Formation in the New Silicon Valleys. Industrial and Corporate Change, 10(4): 835-
860.
Cameron, P. 1986. The Oil Supplies Industry: A Comparative Study of Legislative Restrictions: Report
written for FT Business Information.
Cantwell, J. & Iammarino, S. 2003. Multinational Corporations and European Regional Systems of
Innovation (Studies in Global Competition): Routledge.
Cohen, W., Nelson, M., Welsh, R. R., & John, P. 2000. Protecting their intellectual assets: appropriability
conditions and why U.S. manufacturing firms patent (or not), NBER Paper 7552, Feb. 2000.
Cook, Lesley, & Surrey. 1983. Government Policy for Offshore Supplies Industry: Britain compared with
Norway and France: SPRU Occasional Paper Series No. 31.
Cooke, P. 1992. Regional Innovation Systems - Competitive Regulation in the New Europe. Geoforum,
23(3): 365-382.
Cooke, P., Uranga, M. G., & Etxebarria, G. 1997. Regional innovation systems: Institutional and
organisational dimensions. Research Policy, 26(4-5): 475-491.
Cooke, P. & Morgan, K. 1998. The Associational Economy: Firms, Regions, and Innovation: Oxford
University Press.
Cooke, P. & Schienstock, G. 2000. Structural competitiveness and learning regions. Enterprise and
Innovation Management Studies, 1(3): 265-280.
Cooke, P., Heidenreich, M., & Braczyk, H.-J. (Eds.). 2004. Regional Innovation Systems: the Role of
Governance in a globalized world: Routledge.
Crabtree, E., Bower, J., & Keogh, W. 1997. Conflict or Collaboration: The Changing Nature of Inter-firm
Relationships in the UK Oil and Gas Industry. Technology Analysis & Strategic Management, 9(2).
Cumbers, A. 2000. Globalization, Local Economic Development and the Branch Plant Region: The Case
of the Aberdeen Oil Complex. Regional Studies, 34(4): 371-382.
Cumbers, A., Mackinnon, D., & Chapman, K. 2003. Innovation, collaboration, and learning in regional
clusters: a study of SMEs in the Aberdeen oil complex. Environment and Planning A, 35(9): 1689-1706.
Cumbers, A. & McKinnon, D. 2004. Introduction: clusters in urban and regional development. Urban
Studies, 41(5/6): 959-969.
Doloreux, D. 2002. What should we know about regional systems of innovation? Technology in Society,
24: 243-263.
Doloreux, D. & Parto, S. 2005. Regional innovation systems: current discourse and unsolved issues.
Technology in Society, 27(133-153).
Duncan. 2001. Benchmarking the Norwegian Offshore Industry, Presentation held by Mr. Duncan from
Independent Project Analysis at the Norwegian Strategy Conference in Stavanger 12.12.2001.
Freeman, C. 1987. Technology Policy and Economic Performance: Lessons from Japan. London: Pinter.
Freeman, C. 2002. Continental, national and sub-national innovation systems - complementarity and
economic growth. Research Policy, 31: 191-211.
Frontline Management Consultants & Ron Botham Associates. 2000. Marine engineering, technologies
and service cluster analysis, Report to Scottish Enterprise.
Hatakenaka, S., Westnes, P., Gjelsvik, M., & Lester, R. K. 2006. From "Black Gold" to "Human Gold" - A
Comparative study of the Oil and Gas Industry in Stavanger and Aberdeen, MIT Industrial Performance
Center Working Paper (forthcoming)
Heidenreich, M. 2004. Conclusion: the dilemmas of regional innovation systems. In P. Cooke & M.
Heidenreich & H.-J. Braczyk (Eds.), Regional Innovation Systems: the role of governance in a globalized
world: Routledge.
Iammarino, S. 2005. An Evolutionary Integrated View of Regional Systems of Innovation: Concepts,
Measures and Historical Perspectives. European Planning Studies 13(4).
Isaksen, A. & Remøe, S. O. 2001. New Appraches to Innovation Policy: Some Norwegian Examples.
European Planning Studies, 9(3): 285-302.

15
Jaffee, A. 1989. Real Effects of Academic Research. The American Economic Review, 79(5): 957-970.
Jaffee, A., Trachtenberg, M., & Henderson, R. 1993. Geographic Localization of Knowledge Spillovers as
Evidenced by Patent Citations. Quarterly Journal of Economics, 63: 577-598.
Jakobsen, E. W., Vikesland, M., & Moen, E. 2000. Den norske olje- og gassklyngen - internasjonalisering,
omstilling eller forvitring? [The Norwegian oil and gas cluster - Internationalisation, change or
disintegration], Research report nr. 9/2000, Norwegian School of Management (BI), Sandvika (in
Norwegian).
Kon-Kraft. 2004. Norsk Petroleumsvirksomhet ved et veiskille - Kartlegging av kostnadsbildet på norsk
sokkel [Norwegian petroleum activity at a crossroad], Report published in March 2004 (in Norwegian).
Krugman, P. 1991. Geography and Trade. Cambridge, MA: MIT Press.
Leknes, E. & Steineke, J. M. 2001. Den petro-maritime næringsklyngen i Sør-Rogaland [The petro-
maritime industry cluster in Rogaland], RF-report 2001/146, RF - Rogaland Research, Stavanger (in
Norwegian only).
Lester, R. & Piore, M. J. 2004. Innovation: The Missing Dimension. Boston MA: Harvard University Press.
Lester, R. K. 2005. Universities, Innovation, and the Competitiveness of Local Economies: MIT Industrial
Performance Center Working Paper 05-010, 13 December 2005 (available at
http://web.mit.edu/lis/papers/LIS05-010.pdf.).
Lundvall, B.-Å. (Ed.). 1992. National Innovation Systems: Towards a theory of innovation and interactive
learning. London Pinter.
MacKinnon, D., Chapman, K., & Cumbers, A. 2001. Networks Learning and Embeddedness amongst
SMEs in the Aberdeen Oil Complex, ESRC Research Project Working Paper 3, University of Aberdeen,
May 2001.
Markusen, A. 1999. Fuzzy concepts, scanty evidence, policy distance: the case for rigour and policy
relevance in critical regional studies. Regional Studies, 33: 869-884.
Mill, J. S. 1843. [1976] A System of Logic” On the Logic of the Moral Sciences Indianapolis: Bobbs-
Merrill.
Mowery, D. & Nelson, R. 1999. Sources of Industrial Leadership. Cambridge: Cambridge University Press.
Nelson, R. R. (Ed.). 1993. National Innovation System: A Comparative Analysis: Oxford University Press.
Nelson, R. R. 2000. National Innovation Systems. In Z. Acs (Ed.), Regional Innovation, Knowledge and
Global Change. London and New York: Pinter.
Noreng, O. 1980. The Oil Industry and Government Strategy in the North Sea. London: Croom Helm.
Piore, M. & Sabel, C. 1984. The second industrial divide. New York Basic Books.
Polenske, K. 2004. Competition, Collaboration and Cooperation: An Uneasy Triangle in Networks of
Firms and Regions. Regional Studies, 38(9): 1029-1043.
Porter, M. 1990. The Competitive Advantage of Nations. New York: Basic Books.
Porter, M. 1998. Clusters and the new economics of competition. Harvard Business Review, 76(6).
Saxenian, A. & Jinn-Yuh, H. 2001. The Silicon Valley-Hsinchu Connection: Technical Communities and
Industrial Upgrading. Industrial Corporate Change, 10: 893-943.
Saxenian, A. L. 1994. Regional Advantage: Culture and Competition in Silicon Valley and Route 128
Cambridge, MA: Harvard University Press.
Skocpol, T. & Somers, M. 1980. The Uses od Comparative History in Macrosocial Inquiry. Comparative
Studies in Society and History, 22(2): 174-197.
Trends Business Research. 2001. Business Clusters in the UK - a first assessment, Report for DTI in
February 2001.
Tödtling, F. & Trippl, M. 2005. One size fits all? Towards a differentiated regional innovation policy
approach. Research Policy, 34(8): 1203-1219.

16