constructing jurisdictional advantagemaryannfeldman.web.unc.edu/files/2011/11/...1238 m. feldman, r....

Research Policy 34 (2005) 1235–1249

Constructing jurisdictional advantage

Maryann Feldman∗, Roger Martin

Rotman School of Management, University of Toronto, Toronto, Canada

Received 20 October 2004; received in revised form 18 January 2005; accepted 24 March 2005Available online 2 August 2005

Abstract

This paper aims to advance economic development theory through the concept of jurisdictional advantage; demonstrating howplaces might strategically position themselves to gain economic advantage; then considering how this place-specific advantagemight be constructed. We choose the term jurisdiction to define the set of actors that have a common interest in a spatiallybound community. Jurisdictions are entities with a legitimate political ability to influence social and economic outcomes withintheir boundaries. Borrowing from the literature on corporate strategy, the uniqueness of local capabilities becomes a source ofadvantage for jurisdictions. We consider how to measure and construct jurisdictional advantage.© 2005 Elsevier B.V. All rights reserved.

Keywords: Innovation; Economic geography; Economic development policy

1

asrctsAafo

andrms,ions,bouticork

anstereco-ncy

ughting

owshis-ted

0

. Introduction

In the face of increasingly convincing evidencebout the returns to agglomeration economies, and thetrategic importance of location, pragmatic questionsemain about how places may use this information toreate economic growth. How can a given place in a par-icular situation with a limited set of resources pursue atrategy that will provide sustained economic results?ddressing this question requires prescriptive theorynd conceptual frameworks. Moreover, perspectives

or improving a region’s economic growth tend to focusn the role of government, under-emphasizing the role

∗ Corresponding author. Tel.: +1 416 946 5511.E-mail address: [email protected] (M. Feldman).

of other actors that have significant local interestsplay momentous roles. Actors such as business fiarts and civic organizations, professional associatuniversities and individual citizens are concerned aquality of life and want to influence the economprospects of the communities where they live and w(Becherer, 2000).

The relevant policy question is how to applyappreciation of the benefits of location and cludynamics to implement strategies to promotenomic growth. This question takes on greater urgegiven structural changes in the world economy broon by competitive pressures from newly emerglow cost counties. While corporate outsourcing allfirms to lower production costs technologically sopticated firms compete on the basis of differentia

048-7333/$ – see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.respol.2005.03.015

1236 M. Feldman, R. Martin / Research Policy 34 (2005) 1235–1249

performance and innovation. And, it is in this case thatcertain locations confer an advantage that increases theproductivity of investments made in innovative capac-ity by providing a platform to leverage key resourcesand relationships. Of course, this advantage is typicallythe result of specific unique characteristics that arebuilt up over time to form a coherent place specificactivity set that is not easily transferred or replicatedand forms the basis for sustainable advantage for bothfirms and industries (Feldman and Martin, 2005).

This paper aims to advance economic develop-ment theory on three fronts: introducing the conceptof jurisdiction to the literature on geography and clus-ters; demonstrating how places acting as jurisdictionsmight strategically position themselves to gain eco-nomic advantage; then considering how this place-specific jurisdictional advantage might be constructed.We choose the term jurisdiction to define the set ofactors that have a common interest in a spatially boundcommunity rather than using the more passive termof locational advantage. Jurisdictions, at a variety ofspatial scales, are entities that have a legitimate politi-cal ability to influence outcomes within their borders.Section3 of the paper further defines our concept ofjurisdiction, and considers the economic developmentobjective for which jurisdictions might realisticallystrive, with a discussion of measurable outcomes. Sec-tion5draws on corporate strategy literature to considerhow communities may position themselves in order tocreate opportunities for economic growth. We arguet y beb uldb s ofa h-i ry’sl stab-l fica .T phyo ingj

2

vel ca-t

well-known rather than a historical artefact primarilybecause of the recent economic success of Silicon Val-ley and Route 128 (Saxenian, 1994; Kenney and vonBurg, 1999). Even as the Internet provides real timeconnectivity, there is little dispute that global economicactivity is agglomerating: industries are concentrat-ing spatially in tightly defined geographic areas pre-cisely because proximity and access matters. Empiricalresearch increasingly provides evidence of the benefitsthat accrue to location (e.g.Aharonson et al., 2004;Almeida and Kogut, 1999; Baptista and Swann, 1998;Beaudry, 2001; Beaudry and Breschi, 2003; Glaeseret al., 1992; Henderson, 1994; Sorenson et al., 2004;Rosenthal and Strange, 2003; Swann and Prevezer,1996). As a result, a consensus is developing that com-pact geographic areas, typically centered in cities, aremore important than countries or sub-national regionswhen considering economic growth and prosperity(Krugman, 1998; Fujita and Thisse, 1996).

An important distinction is between the geographicconcentrations of production and the location of inno-vation. Whereas the geographic concentrations of pro-duction is often due to the location of natural resources,ease of transportation or historical inertia, the locationof innovation is due to knowledge externalities andsubject to increasing returns (Audretsch and Feldman,1996). While innovation yields greater productivity andthe increases in wages that jurisdictions seek, jobs asso-ciated with routine production remain geographicallyin place as long as the physical investments are eco-n iatedo comeu ed.

o-m zedk osi-t rketm sig-n utst ostc pro-c eas.L va-t toe ientk thatw -t the

hat every jurisdiction has unique assets that mauilt upon to construct a coherent activity set that woe difficult for others to copy. Considering examplectivity sets from Hollywood and the New York Fas

on district we examine the elements of the industocal advantage and how the industry became eished. Section6 considers how jurisdiction-specidvantage may be constructed and Section7concludeshe next section reviews the literature on the geograf innovation to provide a framework for consider

urisdictional advantage.

. The endogenity of industry—place success

Economists, starting with Marshall in 1890, haong recognized the strategic importance of loion to economic activity (Marshall, 1920). This is

omically viable. Once physical assets are deprecr obsolete, and if the market changes or costs bencompetitive, these locations are easily abandon

Being at the locus of activity for an industry protes innovation. Certain locations supply locali

nowledge externalities or spillovers that provide pive economic value but are beyond the ability of maechanisms to price and efficiently allocate. Theificance of localized knowledge spillovers as inp

o firms’ innovative activities suggest that their mreative and highest value-added activities do noteed in isolation, but depend on access to new idocation mitigates the inherent uncertainty of inno

ive activity: proximity enhances the ability of firmsxchange ideas and be cognizant of important incipnowledge, hence reducing uncertainty for firmsork in new fields (Feldman, 1994). Innovation clus

ers spatially where knowledge externalities reduce

M. Feldman, R. Martin / Research Policy 34 (2005) 1235–1249 1237

costs of scientific discovery and commercialization. Inaddition, firms producing innovations tend to be locatedin areas where there are necessary resources: resourcesthat have accumulated due to a region’s past successwith innovation. In this way, firm success and city eco-nomic growth are endogenous and mutually dependent.The cumulative nature of innovation manifests itselfnot just at firm and industry levels, but also at the geo-graphic level, creating an advantage for firms locatingin areas of concentrated innovative activity. These fac-tors can generate positive feedback loops or virtuouscycles, as clusters attract additional specialized laborand other inputs, as well as the greater exchanges ofideas. What is critical is that these clustered industriestend to be more innovative and have greater productiv-ity which is why we observe wage premium for suchclusters (see for reviewsBaptista, 1998; Audretsch andFeldman, 2005).

The fact that clusters exist and provide economicbenefit does not imply that they can be easily built—thecosts associated with trying may outweigh realizedbenefits. The difficulty is trying to determine whatactivities may aid the formation and building of clus-ters and what activities waste or dissipate resources.While there is little disagreement over the propositionthat clusters are economically advantageous, the notionthat every place should attempt to build clusters is morecontroversial; proactive attempts to build clusters oftenfail or yield results that are different than anticipated(Feldman and Francis, 2004; Orsenigo, 2001). Typi-c g,h ndt hesee ore otechi h asa tatem am-i cuso yO r-i tlecK e,e , thet avel ,2

Incentives that dissipate the natural advantages thataccrue to agglomerations may not only waste resourcesbut may act to the detriment of innovation and tech-nological change nationally or globally, as resourcesare pulled to places where they will be less produc-tive and the benefits of external scale economies arelessened. An examination of the ways in which tech-nologies and industry evolves reveal the importance ofprior actions and the cumulativeness of advance or pathdependencies (Arthur, 1990). The history of regionsand industries become intertwined and it is difficult totransplant industries. Moreover, the path of emergingindustries is difficult to predict and is extremely fluid.Our current understanding of an industry may not beable to anticipate future scientific developments, thetemperament of consumers and their acceptance of aproduct and the directions that technology may take.

When a technology reaches a stage when it can beeasily understood and valued, the established centersmay be described as first movers, already having anadvantage over other locations. As a result, there isa tendency for activities which are ahead to get evenfurther ahead. By the time an industry is well knownenough to be targeted for economic development first-mover jurisdictions have probably already captured thelion’s share of the benefits and are positioned for greateradvantage, making it difficult to catch-up or overtakethem following the same technology. Moreover, forclusters to be successful there is a need not to be cap-tured by the prevailing logic of the industry but to adapta ch-n

3

em-b – isn in-i fiess thea ent ley,l con-c

inga om-m the

ally, cluster building initiatives focus on emerginigh growth industry with great political flourish a

he substantial commitment of public resources. Tfforts are often mimicked by similar jurisdictions. Fxample, in the U.S. there are 40 states that have bi

nitiatives and another nine states that list biotectarget in their state technology plan. The only sissing of the 50 states in the U.S. is Indiana, but ex

nation of the state budget reveals line items that fon building biotech resources (Biotechnology Industrrganization, 2004, p. 28). However, the vast majo

ty of cluster initiatives currently underway have lithance of achieving their articulated goals (Leslie andargon, 1994). In retrospect, the investment of timnergy and capital may not be justified. Of course

hreat of being left behind is great and localities hittle choice but to participate in these races (MacRae002).

new model that reflects a different vision for a teology or an industry.

. Defining jurisdictions

When considering clusters, it is important to remer that the most cited example – Silicon Valleyot a real place in any political, geological or adm

strative sense. Geographically, Silicon Valley detandard classification—it is simply more that thandministrative units of Palo Alto or Menlo Park or ev

he larger unit of Santa Clara County. Silicon Valike other industry clusters, is a more amorphousept that spans multiple administrative units.

When we talk about clusters we are really talkbout spatially defined epistemic communities of con interest. Most generally, we are talking about


construction of a common vision around an industry,technology or set of related interests. History is repletewith examples of co-located firms defining technologi-cal frontiers (c.f.Allen, 1983; Maskell and Malmberg,1999). The geographic concentrations of related enti-ties create synergies that provide unique activity setsthat promote the emergence of new industries: com-bining new knowledge with existing expertise is theessence of innovation. Sociologists such asLatour andWoolgar’s (1986) discussion of scientific discoveries orAbbate’s (1999)history of the Internet documents thecascading decisions that shape human creative work.This suggests that innovation is socially constructedand reflects a consensus vision of what is possible, aseries of complementary activities and imitation andexperimentation. Innovation is a creative activity influ-enced by the expectations, conversations and personali-ties of the individuals involved. Certainly, to the extentthat social interactions are facilitated by co-locationand frequent interaction, innovation becomes a localevent. These epistemic communities self-organize andthere are gains to frequent, face-to-face contact. Whilea specific address may connote prestige, it is proxim-ity, access to concentrations of resources and venues forsocial relationships that matter most in promoting inno-vative activity, productivity growth, and higher wages.

Why then introduce the term jurisdiction? The termconnotes a broader and more inclusive descriptionof the economic assets and joint decision-making ofa spatially defined system that contributes to eco-n ofj gu-l emso beb vastm thej fore reg-u d bya thea letef l ofs

rms,g nali ella s andp nan-

imate actors has the potential to contribute to the devel-opment – or lack thereof – of economically beneficialclusters. A physical asset, such as a great port, cancontribute to the growth of a logistics cluster. Universi-ties or government labs may or may not spin-off ideasthat create new products, new firms and new industries(Bania et al., 1993). Great arts and cultural organiza-tions can attract highly qualified personnel who valueamenities and are critical human capital for certainindustries (Florida and Gates, 2002). Norms supportingentrepreneurship can increase the number of start-upfirms. Conversely, non-compete employment clausesor punitive bankruptcy laws may lessen entrepreneurialactivity (Gilson, 1999; Becker and Hellmann, 2003).All of these are features that should be considered whenanalysing a jurisdiction.

What should be the goals of a jurisdiction? A basicmetric is the prevailing wealth in the geographic area.Wealth is a combination of wages and investments. Thegreater the positive variance in wage levels from thenational or worldwide mean, the greater the jurisdic-tional advantage from which residents benefit, otherthings being equal.1 Investment is an important mea-sure of the wealth of the jurisdiction. For the majorityof the population of the developed world, the singlelargest investment is home equity and the value of localamenities, quality of life and general economic outlookis capitalized in housing prices. Higher local wages anda growing local economy are expected to contributeto appreciation of real estate values and the wealth ofp val-u ich,i ser-v thm

s toi adly

o thism agesa , as ist apita.S agesf ided.F ns ofa nt fort enefitst .S.,h

omic prosperity. Most importantly, the concepturisdiction implies the development of laws, reations, norms and conventions that provide systf governance and innovation. A jurisdiction mayroad—for example, a country, in which case theajority of laws and regulations are the product of

urisdiction. Or, a jurisdiction may also be narrow—xample, a city, in which case some of the laws andlations are self-imposed while others are imposesuper-ordinate body, such as the country. While

nimate actors in the jurisdiction do not have compreedom of action, they certainly have a great deacope for decision-making and implementation.

Jurisdictions include animate actors such as fiovernments/administrative institutions, educatio

nstitutions, cultural institutions, and citizens as ws inanimate features such as natural endowmenthysical infrastructure. Each of these animate and i

roperty owners. Moreover, increases in propertyes yield higher tax revenues for the jurisdiction wh

f used judiciously, increase amenities and publicices. In this way, virtuous cycles of economic groway be created.The logical step is to recognize that strategie

ncrease economic potential should focus on bro

1 However, there are two adjustments that should be made teasure. First, when comparing jurisdictions across countries, wdjusted for purchasing power parity should be the measure

he case when comparing gross domestic product (GDP) per cecond, an additional refinement would be to adjust after-tax w

or major differences in government services and amenities provor example, if after-tax wages are used to compare jurisdictioU.S. city with a Canadian city, there should be an adjustme

he fact that Canadian residents receive greater health care bhrough government expenditure. For individuals living in the Uealth care costs would be paid for from after-tax wages.


defined jurisdictions and the construction of what weterm jurisdictional advantage as a means to promoteeconomic growth and prosperity. This paper takes anext step toward a prescriptive theory of clustering andagglomeration by providing a theory of jurisdictionaladvantage. It takes these steps because the authorsbelieve that the literature will not be able to create auseful prescriptive theory of jurisdictional advantagewithout these building blocks.

4. What is a jurisdictional strategy?

The collective choices that actors make over timethat shape the activity in a place define a jurisdictionalstrategy. This is not meant to imply that there is alwaysclear and conscious coordination of the choices to pro-duce a strategy. Sometimes the choices will be madewith no coordination, the logical outcome of marketforces. Nevertheless, the sum of these choices formsthe basis for what the jurisdiction actually does. Andas with corporations, strategy is choice; strategy is notpublic proclamations or planning processes, it is whatan entity actually does. In this respect, every corpora-tion has a strategy, whether that strategy is articulatedor not. Similarly, every jurisdiction has a strategywhich is defined by what its actors choose to do,whether in coordination or not and whether articulatedor not.

With this background, we define a successful juris-d igha val-u thej m-p tingi co-n d ifw bei ress-i arei uris-d e oft ainh ablet

at-e tureo ints

to the importance of small and compact geographicunits as being a critical element in the performanceof industries. Countries and states provide institutionalframework within which cities work and help definethe opportunity set. However, differences in jurisdic-tional performance appear to derive from the choicesmade by local actors.

5. What is jurisdictional advantage?

The subsequent question is, what is the nature ofa set of jurisdictional strategy choices that producesadvantage evidenced by relatively high and rising wageand property value levels? Here we borrow from cor-porate strategy.

Over the past 30 years, the concepts of corpo-rate strategy and strategic thinking have become wellaccepted by firms. The literature on corporate strategyfinds that observed differences in firm performance aredue to organizations capabilities which are unique andnot easily imitated (Barney, 1991; Dosi et al., 2000;Nelson, 1991; Rumelt, 1984; Wernerfelt, 1984). Whilesimilar resources may be available to all firms, theability to deploy these resources productively is notuniformly distributed and depends on the firms’ capa-bilities (Penrose, 1959). Moreover, these capabilitiesdevelop over time as the result of historically deter-mined endogenous learning processes (Nelson andWinter, 1982). As a result of this path dependency, firmc Ther aree ffec-t sets( m-p hichi

5s

van-t m’sc thataT on ofu erentw ems

ictional strategy as one which produces relatively hnd rising wages for the workers and real estatees for property owners within the boundaries of

urisdiction. If wage levels are higher than other coarable jurisdictions, then the jurisdiction is transla

ts human, physical, and other capital into higher eomic output per worker than its counterparts. Anages are rising, then the jurisdiction is likely to

ncreasing its relative effectiveness rather than regng toward the mean. Similarly, if real estate valuesncreasing, those already owning real estate in the jiction are rewarded with appreciation of the valu

heir property. The longer a jurisdiction can maintigh and rising wages and values, the more valu

he jurisdictional strategy.We are particularly interested in jurisdictional str

gy at the level of the city-region because the literan clustering and agglomeration increasingly po

apabilities are unique and not easily replicated.esult is that specialized resources used by a firmmbedded in an organizational context and their e

ive use is contingent on other complementary asRumelt, 1987). The mechanism through which a coany produces advantage is an activity system w

s unique, not easily replicated and sustainable.

.1. Activity systems: borrowing from corporatetrategy

One school of corporate strategy holds that adage results either from lower cost relative to the firompetition or the production of a set of attributesre uniquely valued by the market (Porter, 1980, 1985).hese advantages are based on the constructinique activity systems, which are defined as a coheb of activities. Taken together, these activity syst


provide an advantage because the individual activitiesand components fit well together and actually reinforceeach other. The fit and reinforcement makes it diffi-cult, if not impossible, for competitors to replicate asuccessful firm’s strategy—a rival firm would have toreplicate every single aspect of the activity system inorder to match the advantaged firm’s strategy. Thus,the essence of strategy is to construct an activity sys-tem that allows the firm to perform differently than thecompetition or to perform different activities than thecompetition.

Consider the example of Southwest Airlines, whichhas been the most successful airline in the U.S. mar-ket over the past 30 years, in terms of the level ofprofitability, stability of earnings and growth in mar-ket share (Porter, 1996). Southwest is noted for flyinga completely standardized fleet of Boeing 737s, flyingto and from secondary airports, having the most fre-quent number of daily departures from each of its loca-tions, and utilizing the Internet rather than travel agentsfor booking. Importantly, its exemplary performanceoutcomes are not the result of any single thing thatSouthwest does. As demonstrated inFig. 1, Southwestachieves advantage by performing a set of activities in

a comprehensive manner that fit together and reinforceeach other to produce a significant and sustainable costadvantage over its competitors. To challenge SouthwestAirline’s success, a potential competitor would have tomatch every single aspect of Southwest’s activity sys-tem.

An activity system can provide a low cost advantageby enabling the firm to produce a product or service thatis roughly equivalent to the competition but at a signif-icantly lower cost. This results in higher profitabilitythan the average competitor and has been the case forSouthwest in the airline industry. Any firm with a costadvantage is able to set prices in the industry and ulti-mately force marginal firms who do not have a similarcost advantage out of the industry.

It is important to note however, that being a low costfirm is not the same as being the firm offering the low-est price. Having the same cost structure as competitorsand deciding to sell at a lower profit margin is not astrategy for long-term advantage. Indeed, it is a recipefor transferring economic value from corporate share-holders to customers. This is simply not a sustainablelong-term strategy. Any competitive firm in the sameindustry can simply cut its own prices and margins to

Airlines
Fig. 1. Southwest : low cost advantage.


Fig. 2. Progressive insurance: differentiation advantage.

compete in the short run. This, ultimately, leads to arace to the bottom in terms of profitability as firms suc-cessively continue to lower prices and squeeze profitmargins.

The second strategy for corporate advantage is prod-uct differentiation—producing a product or service thatis considered to be uniquely valued by a segment of cus-tomers and for which those customers are willing topay a premium price. For example, Progressive Insur-ance, as demonstrated inFig. 2, offers a differentiatedautomobile insurance service to a non-standard seg-ment of drivers (Porter, 1996). It offers quotes thatare better-tailored to the true risks associated with dif-ferent categories of drivers and provides quick andeasy settlement of claims using an extensive fleet ofvan-based adjusters. Like Southwest Airlines, Progres-sive also has a distinctive activity system that featuresmany reinforcing activities, such as its sophisticatedpricing database, a fleet of claims-settling vans, andunique training and compensation structures, whichfit together to produce a service that is highly val-ued by its customers and is produced at a competitivecost.

An effective differentiation strategy also requiresoperational efficiency and a competitive cost structure.

Firms that charge a premium price but do not main-tain an efficient cost structure end up with lower profitmargins and do not have the resources to reinvest insustaining long-term advantage. An effective differ-entiation strategy requires an activity system that isefficient in terms of cost structure while simultaneouslyproducing a differentiated product or service.

Strategy allows firms to define their objectives andmake decisions that focus their energies and invest-ments. Most importantly, strategy allows firms to knowwhat activities are not going to realistically advancetheir goals.

5.2. Jurisdictional activity sets

Following this logic, it seems that places might alsohave similar capabilities that develop over time, areunique and not easily replicated. Jurisdictions may ben-efit from adapting a similar strategic orientation andbuilding an activity system that is unique and valuablein order to produce either a low cost or differentiationadvantage over other jurisdictions. Next, we look atHollywood and the New York garment industry as twoexamples of how jurisdictions have built such activitysystems.


Hollywood is so well known that its name connotesboth an industry and a geographic place. Hollywoodhas been the global center for film production sincethe 1920s and has maintained its dominance throughchanges in production technology and business models(Scott, 2005). In 2001, the southern California enter-tainment cluster, as defined by The Institute for Strat-egy and Competitiveness, employed 178,000 people,or 16% of national employment for this industry. Theaverage wages in the entertainment industry are US$60,000 which was 50% higher than the national aver-age wage for the industry and greater than three timesthe average wage for the region.

Scott (2004, 2005)rejects the conventional expla-nation that the industry located in southern Californiawas due to favorable weather conditions and scenicvistas, but instead offers a view that suggests a coher-ent activity set created the industry’s dominance. Atthe turn of the century, the then-dominant New Yorkbased Motion Picture Patent Trust priced its films bythe foot, a policy that gave producers little incentiveto raise quality or innovate (Scott, 2005, p. 17) Incontrast, Hollywood independent producers concen-trated on distinctive feature films and promoted indi-vidual stars and the development of methodological

procedures to break down the shooting process intodisconnected segments that were reassembled later.This lowered costs and gave rise to the studio sys-tem with its advanced division of labor and sophis-ticated managerial model of production. Hollywoodwas able to differentiate itself from New York withan activity set that included new business models,a cost advantage and the ability to experiment andinnovate.

Fig. 3diagrams the activity set that established Hol-lywood’s advantage, ca. 1928, or about the time whenKoszarski (1990)notes that the term Hollywood wasused in a generic sense to refer to the motion pictureindustry. A supporting infrastructure developed thatreinforced the activity set with skilled crafts such asfilm editing, file laboratories, agents, orchestras, cos-tumes and props. In addition, the open shop labormarket arrangements in Los Angeles, which were inplace until 1935, provided flexibility. The Academyof Motion Pictures Arts and Science was created in1927 as an overarching union with five branches repre-senting producers, writers, directors, actors and techni-cians. The Academy, well known for the Oscar Awards,encourages artistic achievement and diffused bestpractices.

d Cali
Fig. 3. Hollywoo fornia (ca. 1928).


Another example is the New York City fashionindustry, one that has become synonymous with style(Rantisi, 2002a, 2002b). The New York apparel indus-try had a modest beginning on the lower east side ofManhattan, the site of skilled immigrant communitiesin the 1880s. Most large cities had garment districtsat the time and there was nothing particularly uniqueabout New York. By 1920 the industry migrated tothe Garment District in the western part of midtownwhere it is currently located. In these intervening years,the focus shifted to high value-added fashion (Rantisi,2002b, p. 447–448). Currently, the New York clusteraccounts for 40% of U.S. value added for the categoryof women and children’s fashions (Rantisi, 2002b, p.442). Although Los Angeles has greater employmentin this industry category, wages in New York are morethan double the national average for the industry.

Fig. 4 provides the activity set that reinforces theprominence of the fashion industry in New York. Thedevelopment of a range of retailing formats in NewYork – from department stores like Macy’s, Bloom-ingdales and Lord and Taylor’s, as well as specializedboutiques such as Henri Bendel – helped to developvaried markets for the industry’s output. Fashion mag-

azines, notablyWomen’s Wear Daily (now WWD),Harpers Bazaar andVogue helped establish a fashionculture and disseminate industry trends to the nationalmarket. This was supported by the developing advertis-ing industry and expertise in fashion photography. Theinternational Ladies Garment Workers Union helpedpromote safe working conditions and standardizedwages. This ensured high-quality goods, encouragedwomen to upgrade their skills and also enabled workersto become active consumers (Rantisi, 2002b, p. 448).Supporting educational institutions such as the PrattInstitute, the Parson School of Design and the Fash-ion Institute of Technology ensured a steady supply ofskilled labor. In addition, the production system cameto be characterized by a large number of specializedcontractors, job shops and specialty fabric designers.These small, niche firms provide a flexibility and cul-ture of experimentation.

In conclusion, these are two examples of jurisdic-tional advantage. Similarly, the Boston Biotech Clus-ter (Owen-Smith and Powell, 2004) or Silicon Valley(Kenney and Patton, 2005) also have a coherent activ-ity set that connect biotech firms to venture capitalists,specialized business services (such as contract research

wome
Fig. 4. New York City n’s garment production.


organizations), legal services, and universities. How-ever, these examples illustrate the importance of locallabor agreements, experimentation with new businessmodels, the development of a local division of inno-vative labor and other institutional linkages which arespecific and idiosyncratic. Taken together, these activ-ities comprise the local activity set that provides aneconomic advantage to companies, industries and juris-dictions. Moreover, these examples demonstrate thepath dependencies and cumulativeness of the activitysets (Arthur, 1996). Innovation, the ability to introducenovel ideas and consistently improve performance, isa notable characteristic of these locations and a key totheir economic performance.

In the examples examined above, jurisdictions wereable to translate an initial cost advantage into a sus-tained competitive advantage through the constructionof an activity set that yielded an advantage based ondifferentiation. This is to say, the interaction of thefeatures of the activity set, make Hollywood (in film)and New York (in fashion) jurisdictions that provideunique value to a firm in the respective industry, suchthat locating in that jurisdiction is uniquely valuablevis-a-vis alternative jurisdictions. Moreover, the twoindustries examined here are illustrative as they are cre-ative industries where human capital known as talentis important (Caves, 2002). In a global economy whereunskilled labor is inexpensive, transportation and com-munication costs are negligible and raw material matterlittle, talent is becoming an important competitive asset(

e tot ris-d auset igha onl ris-d arda thel lmo icea ficia-r notb do.T y –w oww not as us-

tries where talent and human capital are important.Producing talent requires investments in human cap-ital and incentives that motivate individuals to createand engage their talents. Moreover, talented individ-uals do not work alone. Indeed, talent may only berecognized in a specialized setting where genius maybe appreciated and appropriately rewarded.

Hollywood demonstrates that innovation in businessmodels that accompanied technological breakthroughsin movie-making were critical—a vision of what themotion picture industry might be. While certain indi-viduals stand out as influential, the activity set reflectsthe efforts of many people, acting individually and col-lectively. The New York fashion industry demonstratesthe success of building on rather pedestrian activities.Interestingly, both industries made important transi-tions in the form of advantage they enjoyed. Eachbegan with a low cost advantage—it was less expen-sive for firms to do business in those industries in thosejurisdictions. This was not because of low wages, butrather because of other cost reducing activities. How-ever, both industries eventually developed differenti-ation advantages—with activity systems that providedan environment that firms in the industry could not findelsewhere.

What allowed these jurisdictions to differentiatethemselves is the construction of reinforcing activitysets that both promote and further define the industry.Of course, this advantage is typically the result of spe-cific unique characteristics that are built up over timet t isn asisf us-t ga itu-t acyw herea t bet pira-t ainsa ctedc nesiso

andt d toc tionst mplyr we

Florida and Gates, 2002).The current obsession with outsourcing gives ris

he question of whether low wage regions enjoy juictional advantage. We argue that they do not bec

he manifestation of jurisdictional advantage is hnd rising wages. A jurisdictional strategy based

ow wages would undermine itself on the way to juictional advantage. That is, in order to move towdvantage, wages would have to rise, unraveling

ow wage approach. A low wage policy in the reaf jurisdictional strategy is analogous to a low prpproach in corporate strategy. The targeted beneies of corporate strategy – the shareholders – doenefit from a low price strategy; the customershe targeted beneficiaries of jurisdictional strategorkers in the jurisdiction – do not benefit from a lage strategy. In particular, decreasing wages istrategy for long-term competitive advantage in ind

o form a coherent place specific activity set thaot easily transferred or replicated and forms the b

or sustainable advantage for both firms and indries (Feldman and Martin, 2005). Without engagin

variety of different agents and supporting instions these jurisdictions may not have achieved primithin their industries. The examples presentedre intended to be illustrative. The tendency migh

o dismiss these examples as offering unrealistic asions for average locations. Yet the literature cont

large number of detailed and carefully construase studies about different industries and the gef the jurisdictions in which they reside.

The heuristic of activity systems presented herehe study of coherent activity systems may be useonsider elements that are missing in certain locahat are less successful. Moreover, rather than sieplicating existing successful clusters when


consider these activity systems, it becomes apparentthat activities are reinforcing and that the success ofthe industry and jurisdiction co-evolve. This paper willnow sketch out some ideas on how the animate actorsin a jurisdiction can use the theory of jurisdictionaladvantage to construct jurisdictional advantage.

6. Constructing jurisdictional advantage

Constructing jurisdictional advantage is not easy,just as it is not simple for companies to craft strategiesthat yield long run growth and profitability. However,there are a few things that can be noted. The evi-dence that firm location decisions are not responsiveto jurisdictional tax differentials except at the intra-metropolitan area (Bartik, 1991; Papke, 1991) suggeststhat individual municipalities may gain if they droptheir tax rates or offer special incentives. Of course,this creates artificial competition between administra-tive units that may be part of the same industry clusterand may be viewed as a unified jurisdiction in termsof common economic objectives. These actions, whileyielding short-term benefits, may result in a race to thebottom. Individual municipalities may achieve long-term benefit if they view themselves as subsidiaries ordivisions of a larger geography and cooperate to theirmutual advantage.

Jurisdictional activity system is not the product ofany one class of actors—not firms, not individuals, notg lus-t avea ct-i itsu

sts.P theiro thath andc iver-s entsd ow-e ed bya andr tor isw test rec-o go

so far. Individual firms, for example, will not investmeaningfully in things that payoff primarily for otherfirms or actors in the system. Governments need to fundthose items for which there are high externalities fromthe perspective of any individual firm.

Because no one actor can do everything and eachactor has an interest in experiencing jurisdictionaladvantage, constructing jurisdictional advantage takesthe will of all the actors—a consensus vision and visionof uniqueness. Each party needs to ask what uniquecontribution it can make to the unique activity system.Governments are in the best position to play a coordi-nating role.

Two extreme philosophies anchor the ends of thejurisdictional policy spectrum. At one end of the spec-trum lies aggressive centralized planning that targetsan industry or industries and puts in place a strategy toattract that industry or industries to the jurisdiction,often with tax incentives. There are myriad exam-ples where politicians and civic leaders focus on someemerging, high-growth industry with greater fanfare.This is certainly the case with current example of the49 states with biotech industry initiatives. Even whenefforts are successful at generating start-up companiesit is difficult for a jurisdiction to garner long-term ben-efits if complementary assets are lacking (ConnecticutCenter for a New Economy, 2004).

The opposite extreme is to let market forces deter-mine the allocation of resources, a straight-forwardlaissez faire philosophy. The underlying rationale oft parto llyc asilyr tiesf pre-d t them ar-k

thec rum,t ment.F fulj ris-d entb entb ro-v onV tion

overnments, not universities. Highly competitive cers and therefore highly advantaged jurisdictions hctivity systems in which multiple actors play intera

ng and reinforcing roles to produce and reinforceniqueness.

Each of the actors pursues their own self-intereeople choose to work where they can maximizewn wealth and happiness. Firms invest in thingsave the prospect of increasing their returns. Artsultural organizations pursue self-expression. Unities follow the desires of their scholars. Governmo the things they need to do to get re-elected. Hver, each of these self-interested needs is furtherjurisdictional activity system that produces high

ising wages and real estate values, so each acorking toward their self-interest when it contribu

o jurisdictional advantage. However, it has to begnized that the actions of any one actor will only

his philosophy is that industrial clusters that aref successful cities arise for a variety of historicaontingent or serendipitous factors that are not eeplicated. Firms locate and invest in particular cior reasons that are not well understood, much lessictable and controllable. This view suggests thaost constructive thing a jurisdiction can do is let m

et forces determine its future.However, just as there are logical problems with

entralized, command and control end of the specthere are weaknesses to the laissez faire arguirst, it is difficult to find examples of highly success

urisdictions or even individual clusters in such juictions where there was little evidence of involvemy actors other than firms—in particular involvemy governments acting in coordination with the pision of externality-laden investments. Even Silicalley, often used as the paragon of a jurisdic


driven by free market forces, would not have devel-oped as a highly advantaged jurisdiction with amongthe highest wage levels and real estate values in theworld without massive investment by governments inhigher-education and research focused on the indus-tries that have made Silicon Valley the envy of virtuallyall jurisdictions. Indeed Silicon Valley exemplifies thechallenges in the increasingly knowledge-based econ-omy where market failures lead to under-allocation ofthe very goods that provide advantage without coordi-native activities. After all, this is one of the classic rea-sons for government provision of infrastructure, fund-ing of basic research and promotion of public goodssuch as education. These resources which are associ-ated with market failure take on new importance inthe emerging knowledge-based economy and suggestthat there is a role for collective action and governmentparticipation.

The idea that firms act alone as solo players is aromantic image that has no good exemplars. Laissezfaire proponents tend to point to high tech industrieslike software because they feature cut-throat competi-tion and rapid entry and exit. However, it is one of theindustries most dependent on and linked into the U.S.higher education system. This holds true for pharma-ceuticals, medical devices, aerospace and virtually allhigh-wage industries.

Strategy is choice and closer inspection reveals thateach of these industries has benefited from a jurisdic-tional activity set that coordinates the choices of variousc hav-i turef ia-t gy– rat-e thet ec-t h isu ces-s lsor ina-t d ifi nala osta andg enta di-c e as

a realistic goal, let alone a stated goal, of creating suchan outcome.

As we know from corporate strategy, investmentsmade by businesses to simply replicate what other firmshave already done or improve a firm’s competitive posi-tion from significantly behind to merely lagging aretypically investments that do not earn an acceptablereturn on capital. Similarly, we would not expect gov-ernment investments that are not made to support anadvantaged jurisdictional strategy to have a positive netpresent value. As such the framework of jurisdictionaladvantage provide guidance for the difficult challengeof choosing coordinative and costly interventions thathave a sufficiently high probability of paying off.

What is the role of firms in jurisdictional advantage?A firm can act simply as a taker and exploiter of a juris-diction. However, a firm is better served by being anactive partner in jurisdictional advantage rather than apassive taker. The existence of externalities critical tofirm success suggests that firms can receive benefitsfrom the jurisdictional activity system that are outsideof the market mechanism to price. As soon as it hasmade investments in a jurisdiction, a firm has an incen-tive to encourage other actors in the jurisdiction to buildexternality-laden investments so that the jurisdictionprovides more advantages for itself in the future.

While it may be argued that firms pay more taxesas a result of the higher profits they earn as a resultof externalities, a higher level (not rate) of taxes is asmall price to pay for increased competitive successa rmsm tiveb ralo ore-o tiblea ts inj ov-e maya turet t ofs

taxb r thec . Thee thebe hichi nal

ategories of actors. Given that this is the case,ng a goal – high and rising wages – and a strucor thinking about achieving that goal – differention or low cost – and a tool for guiding the strate

a distinctive activity system – can be useful. Stgy has utility in suggesting that a route betweenwo ends of the spectrum is likely to be most effive. It suggests that a strict laissez faire approacnlikely to provide the externality-laden assets neary for the jurisdiction to succeed. However, it aestricts the overly expansive government coordors. Government intervention can only be justifiet contributes to the development of a jurisdictioctivity system that provides differentiation or low cdvantage to the jurisdiction in attracting, retainingrowing firms. Rigorous analysis of the governmctivities in cluster initiatives would undoubtedly inate that the vast majority of them do not even hav

nd sustainability. In fact, it may be argued that fiay actively cultivate the sources of the agglomeraenefit by investing in local universities and culturganizations, building infrastructure, and so on. Mver, these investments are typically tax deducnd provide a means to make targeted investmen

urisdictions rather than relying on the process of grnment budgeting. That is to say, those firmsctively build the external resources and infrastruc

hat benefit their bottom line, but all in the contextrengthening the jurisdictional activity system.

Industrial recruitment incentives with specialreaks and various other inducements that loweosts of doing business are not low cost strategiesvidence is that this type of strategy is a race toottom in a zero-sum game (Bartik, 1991). There is novidence that it eventually leads to higher wages, w

s the measure of a successful low cost jurisdictio


strategy. In many respects, governments that use taxbreaks or other monetary inducements are identical tofirms that use sales as central tenants of their strategies.Sales are nothing but periodic low price strategies andlow price without low cost is not a sustainable strategy.This is a difficult lesson that Sears and Kmart, who havenow merged in an attempt to survive Walmart’s relent-less market penetration, have learned. Sears and Kmarthave historically followed what are called “hi-lo” mer-chandising approaches. That is, they have a normalprice for their products, which is set at a sufficientlyhigh level to earn a reasonable profit if they sell a tar-geted amount of products at that price. However, todrive volume in the stores, they hold sales by whichthey discount some or all merchandise for periods oftime. Unfortunately for them, customers figure out thisapproach and disproportionately wait for goods to goon sale which simply reduces the overall margin on theportfolio of goods to a level that does not allow suffi-cient investment in upgrading by Sears and Kmart.

Walmart entered the fray against Sears and Kmartwith a different merchandizing strategy, which focusedon cost structure, through leaner overheads, better dis-tribution systems, and narrower product selection. Thiswas known as EDLP – or every day low pricing – mer-chandising strategy by which its everyday low pricemet or beat Sears’ and Kmart’s sale prices. As a resultconsumers could shop at Walmart every day and beconfident they were buying the goods in their basketat a price that was below Sears and Kmart, even ifS heirh om-p laredC

bsi-d ingw od-i nd“ hej ce-n adyi her“ hej t tow umeo taxb axeda t a

subsidy from the current firms in the jurisdiction. Aswith Sears and Kmart, the “hi” sales are extremely vul-nerable when they realize that they are paying higherprices. This both hurts the competitiveness of the exist-ing firms and tends to drive them away, perhaps toanother jurisdiction that will offer them lower prices.As with Sears and Kmart, “hi-lo” is not a robust strat-egy but a dangerous act of desperation that begs for acrushing competitive response.

7. Reflective conclusions

The pursuit of jurisdictional advantage is not with-out its challenges because there are so many factorsthat influence the outcomes. Given that future prosper-ity and quality of life in local communities are at stake,the questions of how this might be done are of morethan academic interest. Previous work on clusters hasemphasized the random nature of geographical loca-tion (Krugman, 1991; Klepper, 2004) suggesting thatclusters arise from serendipitous events. In contrast,we argue that clusters may be constructed, but not inthe way that policy typically proceeds by targeting anindustry that is poised to take off in another location.Instead we argue that policy may be fruitfully employedby building upon unique place-specific assets.

To develop the concept and practice of jurisdictionaladvantage, more work needs to be done on the successmodels of building jurisdictional advantage throught and-a thew es toa eni tingw turew ssesa uris-d orkt ateo

A

us-s arch(

ears or Kmart had the product on sale. With tigher cost structures, Sears and Kmart could not cete and were battered in the process: Kmart dechapter 11.Jurisdictions that offer tax incentives or cash su

ies to attract new firms to the jurisdiction are followhat may be called a modified hi-lo strategy. It is m

fied in that rather than offering alternative “hi” alow” pricing to all customers at rotating times, turisdiction offers at the same time a higher price sario to some “customers” – i.e. firms that are alre

n the jurisdiction – and a lower price scenario to otcustomers”—i.e. firms that it is trying to attract to turisdiction. As with Sears and Kmart, for the gambiork, the jurisdiction desperately needs a large volf high price sales in order to fund the discounts orreaks. This means that existing firms must to be tt a higher rate to fund the “low” pricing—in effec

his middle path between laissez faire and commnd-control. In particular, work needs to be done onay in which governments can constrain themselvctivities that focus on providing the externality-lad

nvestments without discouraging firms from inveshere they are most capable of investing. Also, fuork needs to determine how governments, businend other actors may work together to coordinate jictional strategy. These are the next pieces of w

o be done and we hope that this effort will motivthers to join in this pursuit.

cknowledgements

This paper benefited from comments from discants at the German Institute for Economic ReseDIW Berlin) Regionalization of Innovation Policy


Conference held in June 2004 and from partici-pants in the Athena Alliance at the Woodrow Wil-son International Centers for Scholars. This paperhas benefited from comments and suggestions byAjay Agrawal, Pontus Braunerhjelm, Wendy Dobson,Michael Fritsch, Ken Jarboe, Norma Rantisi, AllenScott, Scott Stern, Will Strange and David Wolf. RichBryden at the Institute for Strategy and Competitive-ness provided assistance with data. The comments andsuggestions of the anonymous referees are acknowl-edged and appreciated.

References

Abbate, J., 1999. Inventing the Internet. The MIT Press, Cambridge,MA.

Aharonson, B.S., Baum, J.A.C., Feldman, M.P., 2004. Industrialclustering and the returns to inventive activity: Canadian biotech-nology firms, 1991–2000. Danish Research Unit on IndustrialDynamics (DRUID), Working Paper 2004-3. ISBN: 87-7873-153-4.

Allen, R.C., 1983. Collective invention. Journal of Economic Behav-ior and Organization 1 (4), 1–24.

Audretsch, D.B., Feldman, M.P., 1996. R&D spillovers and the geog-raphy of innovation and production. American Economic Review86, 630–640.

Audretsch, D.B., Feldman, M.P., 2005. The geography of innovationand spillovers. In: Henderson, V., Thisse, J.F. (Eds.), Handbookof Regional and Urban Economics: Cities and Geography, vol.4. North-Holland/Elsevier Science Publisher, pp. 2713–2739.

Almeida, P., Kogut, B., 1999. Localization of knowledge and theence

A doSci-

A ess.

B the128(4),

B y ofics

ting–51.

B ore?

B dvan-

B co-ent

Beaudry, C., 2001. Entry, growth and patenting in industrial clusters:a study of the aerospace industry in the UK. International Journalof the Economics of Business 8, 405–436.

Beaudry, C., Breschi, S., 2003. Are firms in clusters really moreinnovative? Economics of Innovation and New Technology 12,325–342.

Becherer, H.W., 2000. A more cosmopolitan way of life: why localeconomic development matters. Vital Speeches of the Day 66(15), 473.

Becker, R., Hellmann, T., 2003. The genesis of venture capital:lessons from the German experience. CESIFO Working PaperNo. 883.

Biotechnology Industry Organization, 2004. Laboratories of Inno-vation: State Bioscience Iniatives 2004. Biotechnology IndustryOrganization, Washington, DC.

Caves, R., 2002. Creative Industries: Contracts between Art andCommerce. University of Chicago Press, Chicago.

Connecticut Center for a New Economy, 2004. Incubating Biotech:Yale Prospers, New Haven Waits, March.

Dosi, G., Nelson, R.R., Winter, S.G., 2000. Introduction. In: Dosi,G., et al. (Eds.), The Nature and Dynamics of OrganizationalCapabilities. Oxford University Press, New York, pp. 1–22.

Feldman, M.P., 1994. The Geography of Innovation. Kluwer Aca-demic Publishers, Boston.

Feldman, M.P., Francis, J., 2004. Homegrown solutions: foster-ing cluster formation. Economic Development Quarterly 18,127–137.

Feldman, M.P., Martin, R., 2005. Jurisdictional advantage. In: Jaffe,A.B., et al. (Eds.), Innovation Policy and the Economy, 5. MITPress, Cambridge, pp. 57–85.

Florida, R., Gates, G., 2002. Technology and tolerance. The Brook-ings Review 20, 32–64.

Fujita, M., Thisse, J.-F., 1996. Economics of Agglomeration. In:Proceedings of the 10 Japanese International Economy, vol. 339,p. 341.

G dus-t to.

G 92.(6),

H ent.1 (1),

K sup-omic

K andand

K troitaper.

K f the8.

K ridge,

mobility of engineers in regional networks. Management Sci45, 905–917.

rthur, W.B., 1990. Silicon Valley locational clusters: whenincreasing return imply monopoly? Mathematical Socialences 19, 235–251.

rthur, B., 1996. Increasing returns and the new world of businHarvard Business Review 74, 100–110.

ania, N., Eberts, R.W., Fogarty, M.S., 1993. Universities andstartup of new companies: can we generalize from Routeand Silicon Valley? Review of Economics and Statistics 75761–766.

aptista, R., 1998. Clusters, innovation and growth: a survethe literature. In: Swann, G.M.P., et al. (Eds.), The Dynamof Industrial Clusters: International Comparisons in Compuand Biotechnology. Oxford University Press, Oxford, pp. 13

aptista, R., Swann, P., 1998. Do firms in clusters innovate mResearch Policy 27, 525–540.

arney, J., 1991. Firm resources and sustained competitive atage. Journal of Management 17 (1), 99–120.

artik, T.J., 1991. Who Benefits from State and Local Enomic Development Policies?. Upjohn Institute for EmploymResearch, Kalamazoo, MI.

ilson, R.J., 1999. The legal infrastructure of high technology intrial districts: Silicon Valley, Route 138, and covenants nocompete. New York University Law Review 74 (3), 575–629

laeser, E.L., Kallal, H.D., Scheinkman, J.A., Shleifer, A., 19Growth in cities. The Journal of Political Economy 1001126–1152.

enderson, J.V., 1994. Externalities and industrial developmCityscape: A Journal of Policy Development and Research75–93.

enney, M., Patton, D., 2005. Entrepreneurial geographies,port networks in the three high technology industries. EconGeographies 81, 201–228.

enney, M., von Burg, U., 1999. Technology, entrepreneurshippath dependence: industrial clustering in Silicon ValleyRoute 128. Industrial and Corporate Change 8, 67–103.

lepper, S., 2004. Agglomeration through spinoffs: how debecame the capital of the US automobile industry. Working PCarnegie Mellon University.

oszarski, R., 1990. An Evening’s Entertainment: The Age oSilent Feature Picture. Scribner’s, New York, pp. 1915–192

rugman, P., 1991. Geography and Trade. MIT Press, CambMA.


Krugman, P., 1998. Space: the final frontier. Journal of EconomicPerspectives 12 (2), 161–174.

Latour, B., Woolgar, S., 1986. Laboratory Life: The Construction ofScientific Facts. Princeton University Press, Princeton, NJ.

Leslie, S., Kargon, R., 1994. Electronics and the geography ofinnovation in post-war America. History and Technology 11,217–231.

MacRae, M., 2002. Biotech growth: will your city be a leader or aloser? Pharmaceutical Technology, 106.

Marshall, A., 1920. Principles of Economics. Macmillan, London.Maskell, P., Malmberg, A., 1999. Localised learning and indus-

trial competitiveness. Cambridge Journal of Economics 23 (2),167–185.

Nelson, R.R., 1991. Why do firms differ, and how does it matter?Strategic Management Journal 12, 61–74.

Nelson, R.R., Winter, S., 1982. An Evolutionary Theory of EconomicChange. Belknap Press, Cambridge, MA.

Orsenigo, L., 2001. The (failed) development of a biotechnologycluster: the case of Lombardy. Small Business Economics 17(1/2), 77–92.

Owen-Smith, J., Powell, W.W., 2004. Accounting for emergenceand novelty in Boston and Bay Area biotechnology. In: DraftPrepared for 2004 Genesis of Clusters Meeting, Waxholm,Sweden.

Papke, L.E., 1991. Interstate business tax differentials and new firmlocation: evidence from panel data. Journal of Public Economics45, 47–68.

Penrose, E.T., 1959. The Theory of the Growth of the Firm. JohnWiley, New York.

Porter, M.E., 1980. Competitive Strategy: Techniques for AnalyzingIndustries and Competitors. Free Press, New York.

Porter, M.E., 1985. Competitive Strategy: Creating and SustainingSuperior Performance. Free Press, New York.

Porter, M.E., 1996. What is strategy? Harvard Business Review 74(6), 61–78.

Rantisi, N.M., 2002a. The local innovation system as a source ofvariety: openness and adaptability in New York City’s garmentdistrict. Regional Studies 36, 587–602.

Rantisi, N.M., 2002b. The competitive foundations of learning andinnovation: the case of women’s garment production in New YorkCity’s garment district. Economic Geography 78, 441–462.

Rosenthal, S.S., Strange, W.C., 2003. Geography, industrial orga-nization, and agglomeration. The Review of Economics andStatistics 85 (2), 377–393.

Rumelt, R.P., 1984. Toward a strategic theory of the firm. In: Lamb,R.B. (Ed.), Competitive Strategic Management. Prentice-Hall,Englewood Cliffs, NJ, pp. 556–570.

Rumelt, R.P., 1987. Theory, strategy, and entrepreneurship. In: Teece,D. (Ed.), The Competitive Challenge. Ballinger, Cambridge, MA,pp. 137–158.

Saxenian, A., 1994. Regional Advantage. Harvard University Press,Cambridge, MA.

Scott, A.J., 2004. Origins and growth of the Hollywood motion-picture industry: the first three decades. In: Paper Prepared for theConference on Cluster Formation and Development, Waxholm,Sweden, July 1–3, 2004.

Scott, A.J., 2005. On Hollywood: The Place, the Industry. PrincetonUniversity Press, Princeton, NJ.

Sorenson, O., Rivkin, J.W., Fleming, L., 2004. Complexity, networksand knowledge flow. Working Paper, Anderson School of Man-agement, UCLA and Harvard Business School.

Swann, P., Prevezer, M., 1996. A comparison of the dynamics ofindustrial clustering in computing and biotechnology. ResearchPolicy 25, 1139–1157.

Wernerfelt, B., 1984. The resource-based view of the firm. StrategicManagement Journal 5, 171–180.

constructing jurisdictional advantagemaryannfeldman.web.unc.edu/files/2011/11/...1238 m. feldman, r....

Documents