Professor Michael E.

Porter, Harvard University

Monitor Group
ontheFRONTIER
Council on Competitiveness

CLUSTERS of INNOVATION:
Regional Foundations of U.S. Competitiveness

SAN DIEGO
Pharmaceuticals / Biotechnology
Communications

WICHITA PITTSBURGH
Plastics Pharmaceuticals / Biotechnology
Aerospace Vehicles Production Technology
and Defense

RESEARCH TRIANGLE
ATLANTA Pharmaceuticals / Biotechnology
Financial Services Communications
Transportation and

CLUSTERS OF INNOVATION INITIATIVE

INNOVATION
PRODUCTIVITY
ECONOMIC PERFORMANCE
ECONOMIC COMPOSITION
BUSINESS ENVIRONMENT
SPECIALIZATION
CLUSTERS
STRATEGY
COLLABORATION
This report may not be reproduced, in whole or in part, in any form beyond copying permitted by sections
107 and 108 of the U.S. copyright law and excerpts by reviewers for the public press, without written
permission from the publishers.

ISBN 1-889866-23-7

To download this report or learn more about the Clusters of Innovation Initiative,
please visit www.compete.org or write to:

Council on Competitiveness
1500 K Street, NW
Suite 850
Washington, DC 20005

Tel: (202) 862-4292

Fax: (202) 682-5150
Email: council@compete.org

Copyright ©October 2001

Council on Competitiveness
Professor Michael E. Porter, Harvard University
Monitor Group
Printed in the United States of America
 CLUSTERS of INNOVATION:
Regional Foundations of U.S. Competitiveness

Professor Michael E. Porter, Harvard University

Monitor Group
ontheFRONTIER
Council on Competitiveness
 CLUSTERS OF INNOVATION INITIATIVE:
REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

C ONTENTS
Foreword by the Chairman of the Council on Competitiveness . . . . . . . iv

Foreword by the Co-Chairs of the Clusters of Innovation Initiative . . . . v

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi

National and Regional Steering Committee Members

of the Clusters of Innovation Initiative . . . . . . . . . . . . . . . . . . . . . . vii

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1 Regional Competitiveness and Innovative Capacity . . . . . . . . . . . . . . 5

2 The Economic Performance of Regions . . . . . . . . . . . . . . . . . . . . . 10

3 The Composition of Regional Economies . . . . . . . . . . . . . . . . . . . . 18

4 The Evolution of Regional Economies . . . . . . . . . . . . . . . . . . . . . . . 28

5 TheandDeterminants of Regional Competitiveness

Innovative Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

6 Clusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

7 The Development of Clusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

8 Creating and Implementing a Regional Economic Strategy . . . . . . . . 71

9 Action Agendas for the Public and Private Sectors . . . . . . . . . . . . . 80

Appendices

1. Assessing Regional Economic Competitiveness and

Innovative Capacity: A How To Guide . . . . . . . . . . . . . . . . . . A1

2. Definition of Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . A9

3. Samples Results of Regional Survey: All Five Regions . . . . . . . . . A13

ABOUT the Clusters of Innovation Initiative Participants . . . . . . . . A29

 Foreword by the Chairman of the

COUNCIL ON COMPETITIVENESS

The Clusters of Innovation Initiative is perhaps the most ambitious project in the nearly
20-year history of the Council on Competitiveness. As cited in the Acknowledgments, many

individuals and organizations played key roles in the project’s success. None, however, gave
more of their time, attention, and expertise than the project’s co-chairs, Duane Ackerman,
chairman and CEO of BellSouth, and Michael Porter of Harvard University, both of whom

are members of the Council's Executive Committee. Duane brought his tremendous leadership
and prestige, and Michael his international reputation as the leading expert on cluster theory
and competitiveness. Michael’s pioneering work on innovation and industry clusters is

embedded in this project. The Council on Competitiveness owes a debt of gratitude to

them both.

Raymond V. Gilmartin

Chairman, Council on Competitiveness

Chairman, President and CEO, Merck & Company, Inc.

 Foreword by the Co-Chairs of the Clusters of Innovation Initiative

Since its founding nearly two decades ago, the Council on Competitiveness has addressed a wide
range of economic issues affecting the nation including trade policy, technology policy, the federal
budget, and workforce skills. Competitiveness has tended to be seen primarily from a federal per -
spective, and national policies and circumstances surely affect the prosperity of our economy.
However, the Clusters of Innovation Initiative was undertaken with the realization that the real work
of raising productivity and innovative capacity usually occurs not in our nation’s capital, but in the
cities and regions where firms are based and competition actually takes place.

Regional economies are the building blocks of U.S. competitiveness. The nation’s ability to
produce high-value products and services depends on the creation and strengthening of regional
clusters of industries that become hubs of innovation. Understanding is growing about how these
clusters enhance productivity and spur innovation by bringing together technology, information,
specialized talent, competing companies, academic institutions, and other organizations. Close
proximity, and the accompanying tight linkages, yield better market insights, more refined
research agendas, larger pools of specialized talent, and faster deployment of new knowledge.

Utilizing a unique database developed at the Institute for Strategy and Competitiveness at the
Harvard Business School, we are now able to systematically measure the relative strength of regional
economies and their clusters and track their economic and innovation performance over time. In
addition, a team consisting of individuals at Monitor Group and its affiliate ontheFRONTIER, the
Council on Competitiveness, and the Institute have conducted surveys, in-depth interviews, and
strategic analyses in order to assess the strengths and challenges of five pilot regions: Atlanta,
Pittsburgh, the Research Triangle in North Carolina, San Diego, and Wichita.

This national report draws heavily upon the five regional studies and synthesizes the implications
for any region that seeks to improve its economic performance. The report examines the composi -
tion and performance of regional economies, how industry clusters develop and innovation arises,
how clusters affect a region's economic future, and how a region can establish a strategy and action
program to drive its economy and clusters forward. The framework employed and the lessons
learned apply to every region of the country.

We wish to acknowledge the support we received from the national steering committee, advisors
in the participating regions, the many individuals who gave their valuable time to be surveyed and
interviewed, and the many project sponsors. All of you have helped us to create a unique knowledge
base and a process for catalyzing action. Your thoughts and insights are embedded in this report, and
will, hopefully, benefit not only the five regions that participated in the study but other parts of the
country as well.
 ACKNOWLEDGMENTS
This report benefits from the leadership of co-chairs Duane Ackerman, BellSouth Corporation;
Professor Michael Porter, Harvard University; as well as a national steering committee. They have
guided a partnership involving Monitor Group and its affiliate, ontheFRONTIER, the Institute for
Strategy and Competitiveness at Harvard Business School, and the Council on Competitiveness.
Professor Porter provided the theoretical and methodological framework for the Initiative and led
the research and writing of this national report.
Jeff Grogan of the Monitor Group served as overall project leader. Kurt Dassel of the Monitor
Group managed the efforts in each of the five regions studied. Kurt Dassel and Pedro Arboleda of
the Monitor Group, with assistance and guidance from Jeff Grogan and Mark Fuller of the Monitor
Group, took the lead in preparing this report. Pedro Arboleda, and Randall Kempner, Kyle Peterson,
and Michael Brennan of OntheFRONTIER, under the guidance of Professor Porter, Jeff Grogan,
and Kurt Dassel, prepared the regional reports from which this report draws. These four individu -
als performed the basic economic and cluster analyses and were the primary contacts with business
and government leaders in each region.
The Institute for Strategy and Competitiveness, led by Professor Porter, conducted the Cluster
Mapping Project, a multi-year research effort that developed the data for benchmarking regional and
cluster performance. Elisabeth de Fontenay, Weifeng Weng, Daniel Vasquez and other staff at the
Institute for Strategy and Competitiveness contributed to the conceptual development of the proj-
ect and the interpretation of economic and cluster data presented in the regional reports and the
national report. These individuals include Christian Ketels, Veronica Ingham and Orjan Solvell.
John Yochelson and Alan Magazine at the Council on Competitiveness provided project coordi-
nation and interfaced with business and government leaders. Michelle Lennihan coordinated the
fieldwork, performed data analysis, and contributed to the regional and national reports. Debra
VanOpstal and Jackie Mathewson provided additional national economic data and analysis, as well as
ongoing review and critique of the research. Judith Phair and Lea Kleinschmidt at the Council on
Competitiveness and Jodie Klein, KleinOnPoint, helped communicate the findings of the regional
and national reports to the media and other groups.
Lily Rappoli, Alyson Lee, and Julie Sherman at the DesignStudio at Monitor Group illustrated,
designed, and created the layout of the regional reports and this report.
Almost 1300 business and government leaders contributed to this project in some way by provid -
ing background information, submitting to interviews, completing surveys, and offering their views.
Regional advisors provided the Initiative valuable information and coordination assistance in the
regions. While this report aims to reflect the consensus view of those interviewed and surveyed, it
cannot do justice to all their contributions. Any errors, omissions or inconsistencies are the respon-
sibility of the report writers and not any one individual or institution.
For additional information on this research, contact Kurt Dassel at Monitor Group (e-mail:
Kurt_Dassel@monitor.com), Christian Ketels at the Institute for Strategy and Competitiveness (e-
mail: Cketels@hbs.edu), or Michelle Lennihan at the Council on Competitiveness (e-mail:
Lennihan@compete.org).

vi CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

NATIONAL AND REGIONAL STEERING Clayton M. Jones, President, Rockwell Collins
COMMITTEE MEMBERS George Kozmetsky, Chairman, IC2 Institute
William E. Mayer, Managing Partner,
Co-Chairs
Development Capital, L.L.C.
F. Duane Ackerman, Chairman & CEO,
Mario Morino, Chairman, Morino Institute
BellSouth Corporation
Gary L. Neale, Chairman, President and CEO, NiSource
Michael Porter, Professor, Harvard University
Alan J. Patricof, Chairman, Patricof & Company
Committee Members Pike Powers, Managing Partner, Fulbright & Jaworski, L.L.P.
David Baltimore, President, California Institute of Technology Luis Proenza, President, University of Akron
Richard Bendis, President, Judith Rodin, President, University of Pennsylvania
Kansas Technology Enterprise Corporation Duane J. Roth, Chairman and CEO,
Molly Corbett Broad, President, University of North Carolina Alliance Pharmaceutical Corporation
G. Wayne Clough, President, Georgia Institute of Technology Steve M. Samek, U.S. Managing Partner,
Jared Cohon, President, Carnegie Mellon University Arthur Andersen L.L.P.
Samir Gibara, Chairman, Goodyear Tire & Rubber Company Julie Meier Wright, President and CEO,
William Johnston, President & C O O , New York Stock Exchange San Diego Regional Economic Development Corporation

REGIONAL STEERING COMMITTEE MEMBERS

Atlanta-Columbus San Diego
F. Duane Ackerman, BellSouth Corporation Robert Dynes, University of California, San Diego
Wane Clough, Georgia Institute of Technology Irwin Jacobs, QUALCOMM, Inc.
Milton Jones, Bank of America - Mid South Region Duane Roth, Alliance Pharmaceutical, Corporation
James Kelly, United Postal Service Julie Meier Wright, San Diego Regional Economic
James McDonald, Scientific-Atlanta, Inc. Development Corporation
William Todd, Encino Technology Ventures, LLC Wichita
Sam Williams, Metro Atlanta Chamber of Commerce Rich Bendis, Kansas Technology Enterprise Corporation
Richard Ussery, TSYS Donald Beggs, Wichita State University
Pittsburgh Michael Biggs, Klenda, Mitchell,
Jared Cohon, Carnegie Mellon University Austerman & Zuercher, L.L.C.
Peter Johnson, Tissue Informatics C.Q. Chandler IV, Intrust Bank
Brian Kelley, The Heinz Endowments Bill Hanna, Koch Industries
Sean McDonald, McKesson HBOL Automated Healthcare Ron Holt, KG&E
Dick Simmons, Allegheny Teledyne Charlie Johnson, Cessna Aircraft Co.
Sunil Wadhwani, ‘iGate Capital Corporation Bob Knight, City of Wichita
Peter Zylstra, Pittsburgh Technology Council Phil Neff, Willis of Kansas
Marilyn Pauly, Bank of America, KS
Research Triangle Region
Bill Phillips, Coleman Co.
Molly Corbett Broad, University of North Carolina
Leroy Rheault, Via Christi
Charles Hayes, Research Triangle Regional Partnership
Paul Tobia, Vulcan Chemical
Robert Ingram, GlaxoSmithKline
Hansel Tookes, Raytheon Aircraft Co.
Harvey Schmitt, Greater Raleigh Chamber of Commerce
Jeff Turner, Boeing Wichita
Jim Roberson, Research Triangle Foundation
Tom G. Winters, Sedgwick County Commissioners
Pam Wall, Greater Triangle Regional Council
Jim Ziegler, Bombardier Learjet
Tom White, Greater Durham Chamber of Commerce

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS vii

viii CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS
EXECUTIVE SUMMARY OF THE CLUSTERS OF INNOVATION NATIONAL REPORT

Introduction

During the 1990s, Americans found a way to do what seemed no longer possible — grow the economy,
create jobs, and increase the standard of living, without driving up inflation. Much of the credit goes to
the nation’s ability to develop and commercialize new technology. The result: one of the most robust peri-
ods of economic expansion and prosperity of the past century.
Today, the nation is experiencing an economic downturn. While fiscal and monetary policies pump
dollars into the economy to boost the level of activity, innovation infuses the economy with growth-incu-
bating new ideas, new products, services, and technologies. National policies and national investment
choices have much to do with the growth and capacity of the American economy. For innovation, howev-
er, the real locus of innovation is at the regional level. The vitality of the U.S. economy then depends on
creating innovation and competitiveness at the regional level.
In healthy regions, competitiveness and innovation are concentrated in clusters, or interrelated indus-
tries, in which the region specializes. The nation’s ability to produce high-value products and services that
support high wage jobs depends on the creation and strengthening of these regional hubs of competitive-
ness and innovation.
The Clusters of Innovation Initiative was launched to help meet this challenge. The Initiative examined
five regions around the country: Atlanta, Pittsburgh, the Research Triangle, San Diego and Wichita. These
regions were selected to provide a diversity of size, geography, economic maturity, and perceived econom-
ic success. The regions were similar enough to allow interesting comparisons, yet diverse enough to
encompass a wide variety of challenges and opportunities in regional economic development.
Data for the study were drawn from a number of sources, but the principal sources of data were the
Cluster Mapping Project of the Institute for Strategy and Competitiveness, the Clusters of Innovation
Initiative Regional Surveys™, and in-depth interviews of business and government leaders in each region.
A summary of the findings and implications is provided below:

Innovation and the Standard of Living

Regional Competitiveness and Innovative Capacity
• The economic goal for regions should be a high and
rising standard of living.
• This depends upon creating a high-quality business environ-
ment that fosters innovation and rising productivity.
• Strong and competitive clusters are a critical component of
a good business environment and are the driving force
behind regional innovation and rising productivity.
• The prosperity of a region depends on the productivity of all
its industries.
• Productivity does not depend on what industries a region
competes in, but on how it competes.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS ix

 • The most important sources of prosperity are created not inherited.
• Any regional economic development effort has to start with an assessment of regional
economic performance.
• Economic performance is best measured on multiple levels to capture prosperity, productivity and
innovative capacity.
• Regional economies are composed of three types of industries: traded, resource-driven and local
industries. While local industries account for the majority of employment in regional economies,
traded industries are the dynamic core of a regional economy.
• The evolution of regional economies is a lengthy process. While inherited factors, geography,
climate, and population are important, other factors such as entrepreneurship, the presence of
research and training institutions, the composition of the regional economy, and public and private
sector actions are important influences.
• All levels of government can influence the business environment and the productivity of clusters.
• While government can help foster a favorable business environment, companies and industries
must ultimately achieve and sustain competitive advantage.
• Formal and informal institutions for collaboration such as regional economic development
organizations and alumni of large influential companies are important contributors to cooperation
in advanced economies.

Findings and Implications

Economic Performance of Regions
• Regions vary greatly in terms of economic performance: Some regions have high average wages,

while others have low average wages; some regions are growing rapidly, while others are shrinking.
• A region’s average wages must be assessed in the context of that region’s cost of living:
Regions that exhibit high growth do not necessarily prosper due to cost of living increases that
negate or diminish gains in average wages.
• Higher levels of innovation output lead
Economic Performance Indicators
to higher levels of prosperity: Above-aver-
age economic performance measures are not
enough to ensure regional prosperity.
Maintaining, much less increasing, a region’s
standard of living requires the steady growth of
productivity, which in turn requires innovation.
• Innovation output varies greatly across
regions: Just as regional economies have dif-
ferent levels of average wages and job creation,
so too do they have very different levels of inno-
vation output.

x CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

• The relative effectiveness of commercialization greatly affects the economic impact of
research: Commercialization of basic research is a difficult but important ingredient for generating
entrepreneurship. Some regions have high levels of R&D investments and numerous specialized
research centers, but still lag in terms of innovation output because knowledge is not effectively or
rapidly transferred to companies.
Implications:
- The need for a distinctive strategy: No single policy or strategy will work for all regions. Each
region must craft a distinctive approach based on its unique assets and relative strengths.
- Growth vs. prosperity: Growth is not the same as prosperity. Growth is only desirable if the
standard of living of citizens rises. High growth per se often leads to a rising cost of living
that erodes prosperity and degrades natural resources and physical infrastructure that support
quality of life.
- From efficiency to innovation: Current economic performance does not assure future per-
formance. Maintaining, much less increasing, a region’s standard of living requires the steady
growth of productivity. Innovation output leads to higher productivity levels, and is critically
important if a region is to reach the upper quartile of high-performing regions.
- Measuring multiple dimensions of performance: Measuring only a few indicators of per-
formance will not give an accurate view of a region’s strengths and weaknesses. A number of
performance measures need to be assessed, including economic performance, innovation output,
and effectiveness of commercialization.

Economic Composition of Regions

• The composition of regional economies differs greatly: The Cluster Mapping Project has

identified 41 types of clusters in the U.S. economy. While any given region will have some
employment in the vast majority of these clusters, regional economies are typically very strong in
only a handful.
• A wider geographic focus often identifies more available assets in a region: Regions tend to
focus on narrow geographic areas when devising economic development strategies. A broader
geographic area is sometimes more appropriate.
• Some regional economies are highly dependent on a few clusters or even companies:
Although all regional economies specialize in a few areas, some have especially narrow breadth.
These economies have a disproportionate share of employment in one cluster, and even in a hand-
ful of companies, which makes them unnecessarily vulnerable.
• “High-Tech” clusters account for a small percentage of jobs and wages in most regional
economies: Several types of clusters are especially innovative: communications equipment,
analytical instruments, biotechnology/pharmaceuticals, and information technology. These
clusters are very productive, and pay high wages, and regions with strength in these clusters
certainly benefit from their presence. However, the overall impact of these clusters on a
regional economy is usually relatively small.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS xi

 • Higher wages in traded clusters tend to pull up wages in local clusters: The way to increase
prosperity of an entire region is to increase average wages in traded industries.
Implications
- Defining the right region: The composition of regional economies can shift significantly
depending upon the geographic area considered. Regions have a tendency to follow political
jurisdictions and omit important surrounding areas and assets. A broader, geographic definition
widens opportunities and brings constituencies together.
- Building a strategy: Successful regions build strategies on their unique assets and strong
clusters, where they have the greatest advantages. Strength then spreads to additional clusters
over time.
- Clusters of clusters: Focus on a few clusters exposes a regional economy to booms and busts.
Regional strategies should encompass a wide range of clusters, and be attentive to clusters that
overlap. Overlapping clusters offer potential synergies in skill, technology, and partnership.
- Widen innovative capacity to many clusters: The majority of traded jobs in any region are in
clusters that are not generally perceived to be “high-tech” (e.g., business services, financial serv -
ices, education and knowledge creation, transportation and logistics, and hospitality and tourism).
In order to meaningfully increase overall regional prosperity, innovative capacity must be built in
many clusters.

Specialization of Economic Areas, Narrow Cluster Definition

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

xii CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Evolution of Regional Economies
• Successful regions leverage their unique mix of assets to build specialized clusters: Successful
regions do not pick winners, but build on their inherited assets (e.g., geography, climate, population,
research centers, companies, governmental organizations), to create specialized economies that both
differ from other regions and offer comparative advantages to local companies.
• Building strong regional economies takes decades: There are many steps in building a regional
economy—developing inherited assets, creating new assets, linking companies to these assets, attract -
ing outside companies—and this process takes time.
• Institutions for collaboration play an important role in building regional economies:
Institutions for collaboration help build regional economies by facilitating the flow of information,
ideas, and resources among firms and supporting institutions.
Implications
- Long time horizon: Meaningful changes in regional development require investments that
generally take decades before significant dividends are reaped. Long time horizons create challenges
for leaders seeking reelection, and make it more difficult to mobilize community support
behind an economic development strategy. These realities highlight the need for an institutional
structure for regional development that goes beyond government, as well as the need for private
sector involvement.
- Building on traditional industries: Inherited assets matter to a region’s economic development.
Established and already emerging clusters offer the greatest prospects for near term growth. Strength-
ening established clusters should be one of the early priorities in regional economic development.
- Investing in unique and specialized assets: Many inputs in regional prosperity often require
substantial investments in specialized assets such as university campuses, research programs,
logistical infrastructure, and the like.

Regional Business Environment

• A strong physical and information infrastructure is a baseline requirement to establish and
sustain a prosperous regional economy: Good quality roads, highways, airports, railroads, water,
and power support the efficient movement of people, goods, and services as well as the quality of life
of citizens.
• A strong K-12 educational system is important for developing local talent and attracting
outside talent: The quality of K-12 education is growing ever more critical because it establishes
the baseline of talent for entry-level jobs and the pool of specialized talent critical to cluster devel-
opment. It also helps in the recruitment of individuals and companies.
• Universities and specialized research centers are the driving force behind innovation in
nearly every region: Although companies and individuals do create a large number of innovations,
universities and research centers institutionalize entrepreneurship and ensure a steady flow of new ideas.
• Specialized talent and training are more important than abundant labor: It is not abundant
low wage labor that attracts innovative companies, but rather highly talented, specialized, and often
expensive labor.
• Government can have a significant influence on the business environment, both positively
and negatively: Government at all levels influences the business environment through policies and
services that influence factor inputs, context for firm rivalry, demand conditions, and related and
supporting industries.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS xiii

 • Poor coordination among local jurisdictions impedes efforts to improve the business envi-
ronment: Regional economies encompass many political jurisdictions. Efficient coordination
among them is important for maintaining and improving physical infrastructure (e.g., road,
airports, water ports, communications systems), creating strong K-12 education, offering a busi-
ness-responsive political environment, and promoting cross-cluster collaboration.

Determinants of Regional Productivity

Implications:
- Challenges of success: Successful regional economies tend to experience rapid growth, which
stresses the physical infrastructure. Foresight and a conscious strategy are needed to maintain and
improve infrastructure in advance of the strains caused by growth.
- Recognize the need for strategic transitions: Over time, regional development strategies
run their course. Success at one strategy creates the challenges that need to be addressed by
the next strategy.
- Institutionalizing innovation: Successful regions do not rely on chance, but rather seek to
institutionalize the innovative process by building strong universities and research centers,
and by attracting research divisions of major companies, to create continuous innovation and
entrepreneurship.
- Moving to commercialization: Commercialization is a vital step in the innovation process.
Some regions have high levels of R&D investments and numerous specialized research centers,
but still lag in terms of innovation output because knowledge is not effectively transferred to
companies. Having many different types of research institutions (e.g., public universities,
private universities, for-profit research centers, non-profit research centers, etc.) appears to
foster commercialization.

xiv CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Clusters
• Proximity fosters productivity and innovation: When members of a cluster are located in
close proximity, they can capture synergies that increase productivity, innovative capacity, and new
business formation.
• Clusters often share common industries: Some industries are in more than one cluster.
Overlaps provide opportunities to use strength in one cluster and build new clusters.
• Clusters with depth and breadth enjoy advantages over narrower clusters: Clusters with
strength across a broad range of subclusters have advantages over more narrow clusters due to
the extensive market, technical, and other specialized information which accumulates within a
regional cluster.
• Cluster strength is often disproportionately concentrated in a few subclusters: Clusters
are composed of many subclusters. Even relatively weak clusters can often have strength in a
few subclusters.
• Cluster-specific institutions for collaboration facilitate the flow of information and
resources throughout the cluster: Diverse groups (e.g., rival firms, related and supporting
industries, universities and research centers, training institutions, government, and so forth)
contribute to cluster strength, but their contribution is not automatic. An organization dedicated
to mobilizing these groups does much to strengthen a cluster.
Implications:
- Subcluster interactions: Even if full clusters are relatively weak in a region, there may be a
constellation of related subclusters that constitute a differentiating advantage.
- Proximity: Firms can be encouraged to locate near each other through zoning, and the provision
of easily accessible infrastructure.
The California Wine Cluster

Source: California Wine Institute, Internet Search, California State Legislature. Based on research
by MBA 1997 students R. Alexander, R. Arney, N. Black, E. Frost, and A. Shivananda

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS xv

 The Development of Clusters
• Clusters can be strengthened by increasing awareness of the cluster among local firms and
organizations: Not only must firms be aware of the presence of a local cluster, they must also get
together and coordinate activities to improve the cluster’s business environment. Acceptance of
new companies is important if the cluster is to grow quickly and reach a critical mass.
• New firm and cluster opportunities arise at the intersection of existing clusters: Economic
development strategies can leverage these opportunities to diversify a regional economy.
• Anchor companies play a disproportionate role in seeding cluster development: Anchor
companies support cluster development by acting as magnets for other major companies; organiz-
ing other companies in the cluster for collective action; supporting projects that improve the local
quality of life; and producing numerous spin-out companies, which strengthen key elements of
the cluster.
• Institutions for collaboration can significantly increase the success rate of start-up com-
panies: Cluster development depends in large part on generating new companies from within a
region. Successful regions almost always have a hospitable environment for start-ups.
Implications
- An explicit cluster development program: Although chance events play a role in the
formation and development of clusters, conscious efforts to raise cluster competitiveness and
innovative capacity can meaningfully influence the trajectory of cluster development.
- Recruiting for clusters: Recruitment strategies at the regional level should target clusters in
which the region has strength, or clusters which overlap with other clusters. This allows the
region to market its unique assets rather than compete on subsidies. In recruiting efforts, regions
should also identify gaps within clusters, and seek to attract companies to fill them.
- Opportunities at the intersection of clusters: Opportunities for growth often arise at the
intersection of clusters where a region has strength.

Creating and Implementing a Regional Strategy

• Regions often encounter a common set of pitfalls: Because no single nationwide policy will be
entirely appropriate for every region, policy setting at the regional level is especially important.
• Regions need to overcome transition points in the development of their economies:
Regional leaders encounter transitional challenges as they develop their economies. Addressing
these challenges should be targets of regional economic development strategies.
• Broad-based collaboration is needed for development strategies to succeed: Successful
regional economies benefit from the contributions of a wide array of organizations. Organizing
for action entails arriving at consensus and creating the capacity for regions to implement devel-
opment strategies.
• A shared economic vision helps elicit broad support and coordinate activities: To achieve
good coordination among many diverse groups, a shared vision of common objectives and
methods is vital.

xvi CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 • Strong leadership is a necessary Creating the Capacity to Act
part of any successful economic
development strategy: Strong leader-
ship committed to regional economic
development is needed to ensure that
companies, knowledge centers,
governments, and collaborative institutions
contribute to their full potential.
• An overarching organization for
economic development helps coordinate
and routinize the process:
A formal organizational structure and process
for working on economic issues helps maintain
a consensus behind an economic strategy through
periods of economic and political change.

Action Agendas for the Public and Private Sectors

FEDERAL GOVERNMENT
• Invest in the foundations of science and technology.

- Increase federal funding of research at universities and other research centers.

- Establish federal overhead recovery rules, and other policies, to encourage investment in
universities’ science and technology infrastructure.
- Provide federal support for specialized training programs in science and engineering.
• Improve the innovation policy context.
- Fortify intellectual property protection.
- Strengthen and enforce anti-trust laws with a greater weight on innovation.
- Reinforce federal tax incentives that encourage business investment in R&D and
industry-university collaboration.
• Allocate federal resources to reinforce cluster development.
- Distribute federal research funding through a system of peer-reviewed competitive grants in a
way that fosters cluster development
- Encourage locally-based federal agencies to communicate and coordinate with local business,
institutions for collaboration, and educational and research centers based around clusters.
• Provide better data for measuring regional economic composition and performance.
- Collect more up-to-date data down to the county level.
- Collect measures of both economic performance and innovation.
• Encourage the development of regional economic development strategies that stress innovation.
• Provide federal matching funds for innovation-focused state and regional economic
development strategies.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS xvii

 STATE GOVERNMENT
• Invest in the foundations of science and technology.
- Recognize the state government’s important role in supporting R&D funding at state universities.
- Establish and maintain high levels of state support for community colleges and specialized
training centers.
- Create a strong university or college presence in all major regions of the state.
• Sponsor state programs that encourage cluster development.
- Build cluster thinking into research parks and incubators.
- Organize state systems of higher education around local clusters.
• Focus business recruitment around strong clusters.
- Coordinate activities with firms, universities, and training centers to recruit anchor companies to
their region.
• Create regional dimension to state economic development strategy.
- Encourage and assist regions to develop economic strategies.
- Cultivate attitudes toward collaboration and sharing of information among firms, universities,
training centers, labor, institutions for collaboration, and government.
• Improve information systems to regularly collect data and measure progress.

REGIONAL AND LOCAL GOVERNMENT

• Strongly support K-12 education, and create strong standards and accountability.
• Upgrade core business infrastructure.
- Transportation infrastructure.
- Communications infrastructure.
- Ensure specialized training programs are a high priority in any economic development strategy.
• Develop a regional strategy that involves all stakeholders.
- Support regional benchmarking initiatives.
- Encourage a common vision and collaboration among firms, universities, and training centers.
- Work with firms, universities, institutions for collaboration, and state government to create an
organizational structure to help implement a regional strategy.
• Encourage cluster development.
- Establish research and industrial parks that encourage innovation-based competition.
- Implement cluster-focused and innovation focused recruitment efforts.

UNIVERSITIES AND RESEARCH INSTITUTES

• Recognize the important role of universities in regional economic development.
- Take the lead on, and participate in, regional and cluster development efforts.
• Create and support technology transfer offices.
- Work with firms and venture capital to streamline the technology transfer process.
- Benchmark the commercialization of university-created intellectual property using measures that
promote efficient dissemination of knowledge.
• Actively participate in cluster development efforts.
• Align university curricula and research to meet the needs of local clusters.

xviii CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 - Create cluster-specific institutions to support collaboration between academia and industry clusters.
- Work with local industry to create areas of excellence within universities that differentiate the
university and complement local industry strengths.
- Integrate research and training efforts with the needs of local industry.
- Participate in the recruitment of companies.
• Support company start-up efforts by professors and students through mentorship, entrepreneurial
education, and financing.

CLUSTER-SPECIFIC INSTITUTIONS FOR COLLABORATION

• Promote cluster awareness.

• Engage in ongoing diagnosis of cluster’s competitive position.

- Compare position relative to other regional clusters.

- Identify constraints, obstacles, and advantages.
• Develop training and management programs.
- Provide programs through institutions for collaboration.
- Coordinate with local institutions to provide programs.
• Actively participate with government in recruitment efforts.
- Communicate with firms in clusters to identify gaps in the cluster and recruit accordingly.
• Widen institutional membership to include all cluster constituents.

FIRMS
• Recognize the importance of location to competitive advantage.

• Take an active role in improving competitive environment.

- Consistently communicate your needs and desires (e.g., for talent, ideas, patents) to local
universities, research institutes, and training centers.
• See their cluster as a competitive asset.
• Contribute actively to cluster development activities.
- Actively participate in cluster activities to identify issues of common concern and opportunities
for mutual gain (e.g., regulatory matters, new buyer needs, innovative supplier capabilities).
- Support recruitment activities of local chambers and other regional economic development
officials to bring in companies that will fill missing niches in the cluster (e.g., suppliers,
services providers, competitors).
- Contribute to programs that support new ventures (e.g., improving access to risk capital,
mentoring programs, and specialized services) in order to build-out cluster.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS xix

INTRODUCTION

Why Innovation Matters

During the 1990s, Americans found a way to do what seemed no longer possible — grow the economy,
create jobs, and increase the standard of living, without driving up inflation. Much of the credit goes to
the nation’s ability to develop and commercialize new technology. The result: one of the most robust
periods of economic expansion and prosperity of the past century.
Today, the nation is experiencing an economic downturn. As business and government leaders wrestle
with this new context, most of the attention has been focused on monetary stimulus through lower
interest rates, and fiscal stimulus through lower tax rates and government spending. These are impor -
tant tools to affect economic growth in the short run. However, neither addresses the fundamental causes
of prosperity. Prosperity depends upon the productivity with which the U.S. economy uses labor and
capital to produce goods and services. Productivity rises because of innovation. Moreover, sustained
economic growth will require continued innovation at all levels of the U.S. economy, especially as we
enter a new era when the workforce will be increasing more slowly.
While fiscal and monetary policies pump dollars into the economy to boost the level of activity,
innovation infuses the economy with growth-incubating new ideas, new products and services, and
new technologies. National policies and national investment choices have much to do with the growth
and capacity of the American economy. For innovation, however, the real locus of innovation is at the
regional level. The vitality of the U.S. economy then depends on creating innovation and competitiveness
at the regional level.

About the Clusters of Innovation Initiative

The Clusters of Innovation Initiative offers a new way of thinking about economies that has begun to
take hold as communities across the nation confront the successes of California’s Silicon Valley, and
Massachusetts’ Route 128, Austin, Texas and other areas. In healthy regions, competitiveness and
innovation are concentrated in clusters, or groups of interrelated firms and industries, in which
regions specialize. The nation’s ability to produce high-value products and services that support
high wage jobs depends on the creation and strengthening of these regional hubs of competitive -
ness and innovation.
The Clusters of Innovation Initiative was launched to help meet this challenge. Under the leadership of
Professor Michael Porter, Harvard University; Duane Ackerman, BellSouth Corporation; and a national
steering committee — and supported by a partnership of Monitor Group and its affiliate,
ontheFRONTIER, the Institute for Strategy and Competitiveness at Harvard Business School, and the
Council on Competitiveness — the Initiative has worked to understand how regional economies develop,
how clusters form and gain or lose competitiveness, and how innovative capacity is built. It offers
recommendations for government, universities, the private sector, and other regional institutions. It
aims to inform key decision makers across the country and provide a methodology for analysis that
any region can utilize.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 1

 The Initiative studied five regions around the country: Atlanta, Pittsburgh, the Research Triangle, San
Diego and Wichita. These regions were selected to provide a diversity of size, geography, economic
maturity, and perceived economic success. The regions were similar enough to allow interesting compar-
isons, yet diverse enough to encompass a wide variety of challenges and opportunities in regional
economic development.
Data for the study were drawn from a number of sources, but the principal sources of data were the
Cluster Mapping Project of the Institute for Strategy and Competitiveness, the Clusters of Innovation
Initiative Regional Surveys, and in-depth interviews of business and government leaders in each region.
The Cluster Mapping Project is perhaps the most detailed data set related to economic composition and
performance ever compiled. Comparing regional economies has historically been difficult because clusters
have not been systematically defined and their incidence charted across all U.S. regions. The Cluster
Mapping Project created a detailed statistical analysis using county-level business data, including detailed
metrics on regional economic performance, and data defining 41 types of clusters (e.g., information tech-
nology, automotive, business services) that are found in regions throughout the U.S. economy. The
Cluster Mapping Data also mapped regional economies by cluster and constituent industry and compared
regions to others on various indicators of economic vitality and future competitiveness. One of the goals
of the Cluster Mapping Project is to disseminate this data widely to practitioners. (To access the data over
the Internet, go to www.isc.hbs.edu.)
Monitor Group, its affiliate ontheFRONTIER, in concert with staff from the Council on
Competitiveness designed and implemented a broad reaching survey — the Clusters of Innovation
Initiative Regional Survey™ — to study the business environment and cluster competitiveness in each
region. Over 1025 business and government leaders were surveyed and 264 in-depth interviews were
conducted to determine the historical growth, recent performance and composition of local economies.
Fifteen clusters in the five regions were studied as well.
While many projects around the United States and elsewhere have studied one particular region or one
particular cluster or groups of clusters, the Clusters of Innovation Initiative is unique in its coverage of five
regions and 15 individual clusters using a common methodology, individually and comparatively. The
Cluster Mapping Data, surveys and interviews provide a unique, outstanding information resource for
these regions and the nation as a whole.

2 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Organization of the Report
This report is organized as follows:
- Section 1 describes the fundamental determinants of regional economic performance and
innovative capacity.
- Section 2 builds upon this foundation to describe how economic performance can be
determined and how U.S. regions compare.
- Section 3 describes the composition of regional economies. It highlights the importance of
industries that trade, and the specialization of regional economies by clusters. It describes how
the mix of clusters and the performance of clusters combine to drive regional economic success.
- Section 4 describes how regions develop and the factors that shape the trajectory of
their development.
- Section 5 describes a framework for analyzing the business environment in a region and how it
shapes innovation and economic competitiveness.
- Section 6 demonstrates the importance of industry clusters to regional and national
competitiveness and innovation output, and how to assess the strength of a cluster.
- Section 7 describes the process by which clusters develop and evolve and what forms of
intervention positively or negatively influence their development.
- Section 8 explains the process by which regions create and implement economic strategies.
- Section 9 outlines a potential action agenda for multiple layers of government, universities,
research centers, companies, and institutions for collaboration.

The Appendices includes a “how-to” guide for assessing economic performance and innovative
capacity; a definition of measurements used; and detailed findings of the Clusters of Innovation
Initiative Regional Survey.™

This report summarizes a multi-year, multi-regional analysis. It aims to stimulate other efforts in
regions across the nation to enhance innovation and, through it, lasting economic competitiveness.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 3

4 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS
 REGIONAL COMPETITIVENESS AND INNOVATIVE CAPACITY

1
A nation’s or region’s standard of living is determined by the productivity of its economy. Productivity
is measured by the value of goods and services produced per unit of the labor and capital. It sets the wages
that can be sustained and the returns earned by investors — the two principal components of a nation’s or
region’s per capita income. (See Exhibit 1.)
Competitiveness then, is defined by the level of productivity. Productivity determines prosperity at all
geographic levels, whether it is a nation, a region (metropolitan area), or an inner city. In this report, our
focus will be on the regional level.
Thinking on regional competitiveness is undergoing a significant transition. In many regions, efforts to
enhance competitiveness were bound on lowering the cost of inputs. The focus was on holding down
wages, reducing taxes and recruiting new companies using financial incentives. However, this model has
been superseded for advanced economies and is ultimately self-defeating. Inputs such as cheap labor and
natural resources are widely available. Prosperity comes from the ability to utilize a region’s inputs more
productively than other locations in producing goods and services. Low wages do not yield fundamental
competitiveness, but they hold down the standard of living. Financial incentives are easily matched by
competing regions, and erode the tax base needed to invest in education and local infrastructure. In the
new model, the only path to sustainable prosperity is to build a regional business environment and corpo-
rate capabilities that support high productivity.
Productivity, contrary to popular usage, is more than just efficiency. It also depends on the value of the
products or services that a region’s firms can produce as measured by the prices they can command. In
advanced economies, productivity growth depends heavily on the ability to create higher value products
and services, as well on as improving the efficiency of processes.
The central challenge in enhancing the prosperity of a region is Exhibit 1: Prosperity and Productivity
to create the conditions for sustained productivity growth.
Productivity does not depend on what industries a region
competes in, but on how it competes. There are no industries
that are inherently the most productive and thus more attractive
in generating prosperity. In shoes, for example, Northern Italy
supports high wages and profits because of the high value that con-
sumers place on its products because of their design, materials,
brand recognition, and distribution channels.
Regions should not attempt to pick “winners,” or try to create
new industries where there are no preexisting advantages to build

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 5

 upon. Instead, the challenge is to upgrade the sophistication and productivity of all the region’s industries.
Not all companies and industries in a region will be equally successful, but success should be determined by
the skills and entrepreneurship of the companies in a field rather than selective intervention by government.
The most important sources of prosperity are created not inherited. Inherited competitive advan-
tages such as natural resources, geographic location, or a supply of labor are becoming less important in
determining prosperity. Globalization has expanded the supply of natural resources, and technology has
created new substitutes for them as well as brought distant locations into the economy. A supply of labor
is no longer an advantage in a world where workers are plentiful.
Prosperity depends not on inherited inputs themselves, but on creating the conditions that allow firms
operating in the region to be highly productive in the use of inputs. A good example is the oil and gas clus-
ter in Houston. Oil and gas is still produced in Texas but Texas accounts for only a small and declining
fraction of world production. However, Houston has become the world’s center of technology and
knowledge creation in oil and gas exploration and production, as well as the leading source of most of the
sophisticated equipment and services required. This supports high wages and a large base of thriving
companies. The most prosperous regions do not export natural resources or even only physical products,
but export intellectual capital in various forms.
The prosperity of a region depends on the productivity of all its industries. The productivity of
a regional economy depends on the average productivity of all its companies and industries, not just those
that sell outside the region. Local industries directly affect a region’s impact on the standard of living
because their productivity has a large influence on the local costs of living. However, local industries also
affect the success of a region’s industries competing with firms based elsewhere. For example, research on
Japan1 has shown that poor productivity of local industries such as transportation, construction, and
wholesaling raised the cost of doing business and thus became a drag on the prosperity of the country
despite the existence of some very productive exporting industries. Regional competitiveness, then,
depends on ensuring that local companies in fields such as utilities, transportation, health care delivery, and
other local services are competitive.

INNOVATION AND THE GROWTH OF PRODUCTIVITY

Productivity today sets current competitiveness, but maintaining, much less increasing, a region’s stan-
dard of on living requires the steady growth of productivity. Especially in advanced, high wage economies
no region can maintain high wages, and hold its own in global markets, by producing standard products
using standard methods because they will be imitated by other regions with lower wages.
In advanced regions, prosperity rests heavily on the capacity for continuous innovation. A high level of
productivity itself is not enough when developing countries and regions are improving their skills, and
can rapidly access modern technology. Advanced regions need to innovate to be able to produce prod -
ucts that lower wage regions cannot yet make, and to maintain the productivity advantage that supports
their higher wages. (See Exhibit 2 on the following page.)
Innovation is more than just scientific discovery. Innovation stretches beyond science and technol-
ogy, and includes all the activities involving the discerning of needs and the transformation of knowledge

1Michael E. Porter, Hirotaka Takeuchi, Mariko Sakakibara, Can Japan Compete, New York: Perseus Books (2000).

6 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

into commercial products, processes, and services. Indeed, some of the most important innovations today
occur in marketing, sales, services offered, and distribution; for example, innovation led to the revolution
in the small-package delivery that occurred in the last 15 years and resulted in U.S. global preeminence in
this industry.
There are no low-tech industries, only low-tech
Exhibit 2:
firms. Today, innovation can drive productivity improve- Innovation and the Standard of Living
ment in virtually every industry. Although industry
producing enabling technologies such as biotechnology,
computers, software, and communications equipment
and services have received much attention, opportunities
to apply advanced technology are present in fields as
disparate as textiles, machinery, and financial services. For
example, the small-package delivery industry was trans-
formed by advanced communication and information
processing technologies which led to unheard of efficiency
and the ability to integrate with customers.
In the modern economy, there are no “low-tech” indus-
tries, only low technology companies that fail to incorporate
new ideas and methods in their products and services.
Innovation can upgrade the sophistication of competition
and future productivity throughout a region’s economy, not
just in a few “high-tech” industries.

WHAT IS A CLUSTER? 2
Clusters are geographically close groups of interconnected companies and associated institutions in
a particular field, linked by common technologies and skills. Clusters take varying forms depending on
their depth and sophistication, but most include end product or service companies; suppliers of special -
ized inputs, components, machinery, and services; financial institutions; and firms in related industries.
Clusters also often include firms in downstream industries, producers of complimentary products; special-
ized infrastructure providers; government and other institutions providing specialized training education,
information, research, and technical support.
Drawing cluster boundaries involves a creative process informed by understanding the most important
linkages across industries and institutions to competition. The strength of these “spillovers” and their impor-
tance to productivity and innovation determine the ultimate boundaries.
Why view economies through the lens of clusters, rather than groupings such as companies, industries,
or sectors? Because clusters align better with the nature of competition and the sources of competitive
advantage. Clusters capture important linkages and spillovers of technology, skills, information, etc., that
cut across firms and industries. Viewing a group of companies and institutions as a cluster highlights
opportunities for coordination and mutual improvement.

2Michael E. Porter, On Competition (Boston: Harvard Business School Press, 1996), pp. 199-205.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 7

 DATA SOURCES
The Clusters of Innovation Project examines five regions: Atlanta, Pittsburgh, the Research Triangle, San
Diego, and Wichita. For the purposes of this study, a region is defined as a metropolitan statistical area
(MSA) using U.S. Department of Commerce boundaries, or, in some cases, the Economic Area (EA). The
five regions were selected to provide a diversity of size, geography, economic maturity, and perceived eco-
nomic success. The regions are similar enough to allow interesting comparisons, yet diverse enough to
encompass a wide variety of challenges and opportunities in regional economic development.
Data for the study were drawn from a number of sources. Performance indicators were assembled from
sources such as the County Business Patterns, Department of Commerce Trade Statistics,
PriceWaterhouseCoopers Money Tree, and the Inc. 500 List.
The principal source of data for this study on the composition and performance of the overall regional
economies and specific clusters was the Cluster Mapping Project of the Institute for Strategy and
Competitiveness, the Clusters of Innovation Initiative Regional Surveys™, and the Clusters of
Innovation Initiative Regional Interviews.

The Cluster Mapping Project

The Cluster Mapping Project is one of the largest, if not the largest, data sets related to regional innovation
analysis ever compiled and might well represent the first time this type of work has been conducted in such a
systematic and detailed way. The Cluster Mapping Project created a detailed statistical analysis of county-level
business data, including detailed metrics on regional economic performance, and data defining 41 types of
clusters (e.g., information technology, automotive, business services) that are found in regions throughout the
U.S. economy. Clusters can be defined using “narrow” or “broad” definitions. We use narrow cluster defi-
nition to refer to the subset of the industries that are most correlated with a given cluster. Broad cluster definition
includes all industries with statistically significant locational correlations. This includes industries with
stronger locational correlations with another cluster. The Cluster Mapping Data also maps regional
economies by cluster and constituent industry.
At the broadest level, we compared regions on various indicators of economic vitality and standard of
living such as employment growth, unemployment, wages, wage growth, cost of living, and exports. To
assess potential future competitiveness, we examined measures of innovative output and entrepreneurship
including patents, establishment formation, venture capital investments, the prevalence of fast growing
companies, and initial public offerings. Wherever possible, we tracked both the level and the growth rate
of each performance indicator. We compared the performance of regional economies to the national econ-
omy as a whole, as well as to other technologically intensive regions.
We also analyzed a regional economy’s composition, because differences in composition lead to differ-
ences in performance and innovation. Comparing regional economies has been difficult because clusters
have not been systematically identified or mapped across all U.S. regions. To address this challenge, we
defined clusters statistically and assembled detailed data by industry and cluster on employment, wages,
establishments, and patenting over time for every region in the United States.3
The Cluster Mapping Project also provides an objective, quantitative way to profile and measure the
strength, evolution, and performance of the most important clusters in a region’s economy, understand the

3By traded, we mean that the location of the firms in these clusters is not driven by the need to be near a specific natural resource, or by pop-

ulation concentration. Instead, these industries are located in a specific area for some reason related to the region’s innovative capacity.

8 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

drivers of the region’s relative wages, employment growth, and formation of new establishments, assess
the region’s patenting performance, and examine the region’s relative position versus other regions over-
all as well as in its leading clusters.

Clusters of Innovation Initiative Regional Surveys

Professionals from Monitor Group, ontheFRONTIER, and the Council on Competitiveness also used
a broad reaching survey—the Clusters of Innovation Initiative Regional Survey™— of 1025 business and
government leaders to study the historical growth, and recent performance and composition of local
economies, and selected clusters in the five regions studied (see Exhibit 3 below).
The survey protocol included questions regarding: (1) demand conditions in a region; (2) the context
for firm strategy and rivalry in a region; (3) quality and availability of related and supporting industries;
(4) government’s impact on the determinants of regional productivity; (5) roles and performance of institu-
tions for collaboration; (6) present strengths and future threats to regional innovation; (7) idea generation
and commercialization at regional institutions; (8) help given to entrepreneurs and established companies by
regional institutions; and (9) other key influences on the regional business environment.

Clusters of Innovation Initiative Regional Interviews

Over 264 follow-on in-depth regional interviews were conducted to enhance data from the surveys.
The interviews allowed us to acquire a better, more in-depth understanding of key issues. The interview
protocol included probing questions in the areas of: firm location/expansion, new venture support, regional
development, network focus in development, innovation specific questions, university relations, govern-
ment, and new business formation.
Samplings of questions asked in the interviews include: What barriers do you see to expansion in this
region? Is there a strong group of local business support and strategic advising services for start-ups? Do
you think the region has been successful over time, and if so why? What sort of networks or network
organizations have helped the region develop? Do you partner in R&D with other companies in your
industry? What are the major
environmental or cultural Exhibit 3: Regions and Clusters Studied
factors are important to or
have an impact on innova-
tion in your region? How
aggressive are the universi-
ties in commercializing
applied research (licensing,
equity investor, incubators)?
How effective is your state
and local government in fos-
tering the development of
innovative firms? How does
new business formation
happen in your region?

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 9

 2 THE ECONOMIC PERFORMANCE OF REGIONS

The starting point for regional economic development is an assessment of regional economic perform-
ance. Performance should be measured on multiple levels to capture not only current prosperity and
productivity but also innovative capacity. The ability to create and commercialize innovations, both in exist-
ing firms and new firms, will have a fundamental influence on productivity and prosperity in the future.
Prosperity, or the standard of living, is most strongly the result of the level and growth rate of average
wages, and the proportion of a region’s citizens that are employed. Standard of living is also influenced by
the level of local living costs. These determine the actual purchasing power derived from income.
Productivity and productivity growth are the foundations of per capita income. At the regional level,
available data and data disclosure limitations make productivity difficult to measure reliably. One indica-
tor associated with high productivity that can be measured is export performance. High and rising exports
are normally a reflection of high productivity. Exports allow a region to expand its most productive beyond
serving only the U.S. market, raising regional productivity.
The foundation of future productivity and productivity growth is innovation. Innovation has an
upstream technological or creative component, and a downstream, or commercialization, component.
Technological innovation can be measured by the level of patenting. Patents are not a perfect measure of
technological innovation, but numerous studies have demonstrated that patenting is the best available
measure and correlated with non-patented innovative activity. 4 The ability of a region to commercialize
new ideas and technologies can be measured by the extent of new company formation and the presence of
high growth firms.

PERFORMANCE MEASURES UTILIZED

Exhibit 4 shows the overall regional performance measures employed in our approach. Overall eco-
nomic performance indicators were employment growth, the unemployment rate, average wages, wage
growth, the cost of living, and exports per worker.
Measures of regional innovation and entrepreneurship included patents per worker, new establishment
formation, venture capital investments, initial public offerings and the prevalence of fast growing companies.
Wherever possible, we tracked both the level and the growth rate of each performance indicator.
We compared the performance of regional economies to the national economy as a whole, as other
technologically intensive or otherwise salient regions.

4Furman, J., Porter, M.E., and Stern, S., The Determinants of National Innovative Capacity, Research Policy, forthcoming

10 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 4: Economic Performance Indicators

FINDINGS

Regions vary greatly in terms

of economic performance
Regional economies differ greatly from each
other, and from the national economy, on a number
of dimensions. An important one is economic per-
formance. Some regions have high average wages,
while others have low average wages; some regions
are growing rapidly, while others are shrinking (see
Exhibit 5 below and Exhibit 6 on following page).
Source: Bureau of Labor Statistics, Bureau of Economic Analysis; International Trade
Administration; U.S. Patent and Trademark Office; PriceWaterhouseCoopers Money Tree;
Hoover’s IPO Central; Inc. Magazine, American Chamber of Commerce Researchers Association.

Exhibit 5: Variation in Average Wages across Economic Areas, 1999

Note: Narrow Cluster Definition

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 11

 Exhibit 6: Variation in Employment Growth across Economic Areas, 1990-1999

Note: Narrow Cluster Definition

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

A region’s average wages must be assessed in the context of that region’s cost of living
We found that regions exhibiting high growth do not necessarily prosper because cost of living increases
can negate or diminish gains in average wages. Even if a region has a relatively high average wage, an equal-
ly high or higher cost of living will detract from that region’s prosperity. Higher average wages then are
not evidence enough of a region’s prosperity.

San Diego Average wages in San Diego are just below those of the nation as a whole, while the cost
of living in San Diego is significantly higher (25%) than the national average. In 1999 housing costs were
estimated to be 50% higher than the national average. In essence, San Diego has a California cost of living
but is competing with national average wages. This will make it more difficult for San Diego to compete
successfully in the war for talent—one of the region’s most important assets—and will continue to be an
economic burden to its citizens for the foreseeable future. The majority of survey respondents in San Diego
indicated that the rising cost of living has consequences for the region (see Exhibit 7 on following page).

12 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 7: Select Survey Results on the Consequences of the
Rising Cost of Living, San Diego

Source: Clusters of Innovation Initiative Regional Survey, San Diego

Atlanta The Atlanta region has been successful at creating many jobs. Average wages paid in the Atlanta
cluster rank 15th among the largest 20 clusters, and have been increasing at more than 7% a year in the
1990s. The average wage in Atlanta in 1999 was $35,382, slightly above the national average of $32,100.
However, the cost of living in Atlanta is an estimated 20% higher than the national average. The cost of liv-
ing in the region, once a major draw to relocating companies, is now equivalent to the U.S. average, and the
middle class, as in many U.S. cities, is struggling to find affordable housing. Atlanta’s gains in average wages
need to be measured in relation to its cost of living increases, especially for a region that is among the top
ten fastest growing metropolitan areas in the nation.

Innovation output varies greatly across regions

Just as regional economies have different levels of average wages and job creation, so too do they have
very different levels of innovation output. The highest per capita patent producing region (Boise City, ID)
produces almost nine times the number patents per person of the median region (Eugene, OR). The top
10 patenting regions account for 51% of the nation’s total patents (see Exhibit 8 on following page).

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 13

 Exhibit 8: Variations in Patenting per Capita across Economic Areas, 1998

Note: Narrow Cluster Definition

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

Higher levels of innovation output lead to higher levels of prosperity

We found that the more innovative regions in our study were also the more prosperous, while the
regions that struggled to innovate also struggled in terms of economic performance. Moreover, the low-
cost regions were not the more prosperous regions. We measured the number of patents issued per 10,000
workers compared to average wages in various Economic Areas across the nation for 1998 and found a
correlation between patenting innovation and high average wages (see Exhibit 9 below).

Exhibit 9: Patents per 10,000 Workers

vs. Average Wages in U.S. Economic Areas, 1998

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

14 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Research Triangle In the late 1950s, the Research Triangle, North Carolina area was among the low -
est in the country in terms of wage and employment levels. Despite having three strong local universities,
educational levels were also quite low because university graduates left the area after receiving their
degrees. One interviewee was advised not to go to college because the training would not enable him to
find work after graduating (he disregarded the advice). Textiles and tobacco dominated the economy, and
Research Triangle was on a trajectory of long-term economic stagnation.
Then a group of business, academic, and government leaders worked together to create and develop
Research Triangle Park. The Park offered companies easy access to physical infrastructure, human
resources, and knowledge assets, and research-oriented companies began to locate in the region. Four
decades later, Research Triangle has a growing, prosperous economy that is clearly the result of the success -
ful commercialization of new ideas. In the 1950s, Research Triangle had an abundance of low-cost labor,
but dim economic prospects. Innovation changed that.

The relative effectiveness of commercialization greatly affects the economic impact of research
A difficult and important ingredient for generating entrepreneurship is the commercialization of tech-
nology. We found that some regions have high levels of R&D investments and numerous specialized
research centers, but still lag in terms of commercialization because knowledge is not effectively or rapidly
transferred to companies. Regions with high levels of R&D, or even patenting, thus do not necessarily
exhibit high levels of commercialization.

Research Triangle Although Research Triangle exhibits high levels of R&D, these investments have
not been matched by high levels of commercialization in the region. The Research Triangle’s leaders have
been successful in attracting research institutions to the region—more than ten major specialized research
centers are located in the Research Triangle. The Research Triangle consistently receives more than six
to seven times the national average of R&D investment per worker as a result of having these specialized
research centers. Yet surveys indicate that institutions for collaboration and research centers in the Research
Triangle are not helping the region’s firms as much as they could. We asked survey respondents at compa-
nies in the five regions how often they used various institutions in the commer cialization process. The
results indicate that companies in the Research Triangle area use these institutions and centers relatively
infrequently. Of the five regions studied, Research Triangle ranks 3rd to 5th in all but one of the influ-
ences on commercialization of technology (see Exhibit 10 on following page).

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 15

 Exhibit 10: Influences on the Commercialization of Technology in the Research Triangle

Note: August 2001, n=116

Source: Clusters of Innovation Initiative Regional Survey, Research Triangle Region

Pittsburgh Some of Pittsburgh’s most valuable assets are its research universities and institutes, with
world-renown centers of research in transplantation, bioscience, computer science, and engineering acting
as anchor firms in the region. Pittsburgh is a powerhouse for university research and development as it
receives more than twice the national average on a per worker basis, but it requires better commercialization
mechanisms. The universities are a tremendous source of innovation, and opportunities exist to capitalize on
relationships, such as the University of Pittsburgh’s affiliation with the Healthcare/Hospital Industry.
However, more of these institutions should ensure that their basic research find its way to commercial appli-
cations. It is apparent from the surveys and interviews that university technology commercialization needs
to be better supported and improved.

16 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

IMPLICATIONS
• The need for a distinctive strategy: No single policy or strategy will work for all regions. Each
region must craft a distinctive approach based on its unique assets and relative strengths.
• Growth vs. prosperity: Growth is not the same as prosperity. Growth is only desirable if the
standard of living of citizens rises. High growth per se often leads to a rising cost of living that
erodes prosperity or a degradation of natural resources and physical infrastructure that sup-
ports quality of life.
• From efficiency to innovation: Current economic performance does not assure future perform-
ance. Maintaining, much less increasing, a region’s standard of living requires the steady growth of
productivity. Innovation output leads to higher productivity levels, and is critically important if
a region is to reach the upper quartile of high-performing regions.
• Measuring multiple dimensions of performance: Measuring only a few indicators of perform-
ance will not give an accurate view of a region’s strengths and weaknesses. A number of performance
measures need to be assessed, including economic performance (e.g., job growth, wages, wage
growth, exports, and export growth); innovation output (e.g., patents, new company formation, fast-
growing firms, venture capital investments, and IPOs); and commercialization.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 17

 3 COMPOSITION OF REGIONAL ECONOMIES

Regional economies are composed of three broad types of firms and industries. Each is important to
a region’s prosperity, but in different ways.
The first type is industries that compete across locations. In the US, this competition often occurs
between domestic regions but may also include foreign locations. Grouped into clusters, this type of
industries is called “traded.”
The second type is industries that are resource-driven.
The third type is industries that compete only within their region. This type of industries is called
“local.” Local industries are intrinsically tied to the traded industries located in their region: they directly
serve the needs of the traded industries as suppliers and service providers, and they indirectly depend on
the success of the traded industries through its influence on final consumer demand.

TYPES OF CLUSTERS
Traded clusters have a disproportionate influence on regional prosperity and economic
growth. Traded industries can, in principle, be located anywhere. But similar traded industries tend to
concentrate in specific locations. Because they grow beyond the size and the needs of the local market, they
can become much more sophisticated and productive. Their high productivity can support high wages that
support the prosperity of their employees but also support the prosperity of others through the consumer
demand they create.
Resource clusters can support high wages but have limited scope in advanced economies.
Resource-driven industries also compete across regions but their location is tied to local resources. Their
performance is much more dependent on the way the industries use technology and innovative processes
then on the direct value of the natural resources they process. For example, despite virtually identical nat-
ural conditions, the pulp & paper industries in Finland with their sophisticated use of technology achieve
much higher productivity then their less advanced competitors in Canada.
Local clusters account for the majority of employment in regional economies. Because
local industries serve only the local market and most are services, they have more limited opportunities
for productivity growth. This means that local industries tend to account for an increasing share of
regional employment.
Traded industries seem to be more dispersed than they really are because firms most establish distribu -
tion centers, sales offices, service facilities, and other supporting functions in almost every region. The
locations where truly competitive firms are based are usually limited in number.

18 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 The traded economy is specialized by cluster. Clusters are geographically proximate groups of
interconnected companies and associated institutions in a particular field, linked by commonalities and
complementarities (see section 6). Regional economies can be profiled on the mix of clusters present. It is
on the level of individual clusters that regional economies specialize.
Specialization in a series of strong clusters with a significant national position enhances
a region’s performance. Clusters that can attain the critical mass gain productivity and innovative bene-
fits. A region which depends heavily on just one or very few clusters for much of employment, however,
can be exposed to shocks and instability both in those clusters and the local industries which depend on
them. For example, the recent downturn in the information technology cluster has been especially detri -
mental to Singapore, a city-state with an economy heavily dominated by this cluster.

FINDINGS

The composition of regional economies differs greatly

Regions differ greatly in terms of the clusters that compose the economy. The Cluster Mapping Project
has identified 41 types of clusters in the U.S. economy. While any given region will have some employ-
ment in the vast majority of these clusters, regional economies are typically very strong in only a handful.
In other words, regional economies are highly specialized (see Exhibit 11 below).

Exhibit 11: Top Clusters in Select Geographic Areas

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 19

 Exhibit 12 below provides a benchmark for assessing the specialization of an economic area in the United
States. Every region has some clusters that are relatively more concentrated than what would be expected
based on the region’s total employment, and these would be located above the horizontal axis. Regions will
generally have some clusters growing faster than the national average for that cluster. These clusters would
be located to the right of the vertical line. The chart below identifies the average percentage of traded
employment, the number of clusters, and the percentage of total wages accounted for by each quadrant.

Exhibit 12: Specialization of Economic Areas, Narrow Cluster Definition

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

Atlanta For more than a century, business, government, and university leaders have consistently posi-
tioned Atlanta as a center for commerce. The composition of the Atlanta economy reflects their efforts.
Roughly 50% of the metro area’s traded employment is in four clusters: transportation and logistics,
distribution services, financial services, and business services. Public and private sector collaboration
has produced a highly differentiated, specialized, and above all successful economy.
Exhibit 13 on the following page shows the Atlanta Metro Area’s employment share and growth in share
of the 41 traded clusters in the United States economy. Atlanta had 1.8% of total national employment.
Clusters above the horizontal axis are relatively concentrated in Atlanta and clusters to the right of the ver-
tical axis have grown over the 1990 to 1999 period. The upper right quadrant represents clusters that have
disproportionate share of national employment in Atlanta and are growing their share.

20 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 13: Differentiation and Specialization of the Atlanta MSA, Narrow Cluster Definition

Note: Data points falling outside the range are placed on the axes together with their national share and change in share respectively.
(y-axis, x-axis)
Source: Cluster Mapping Project at Institute for Strategy and Competitiveness, Harvard Business School

Pittsburgh Pittsburgh was the center of steel production for the entire world and flourished because
of that specialization. The early mills were initially attracted to Pittsburgh by natural factors. Local ore
deposits and the rivers that allowed for cheap transportation provided the basis for the development of the
industry and the region. During the early 1800s, iron-smelting factories took advantage of the region’s hard-
working people, waterway and railroad networks, and abundant coal reserves. Pittsburgh flourished due to
low-cost ore, and proximity to the world’s largest demand, with rail and water transportation advantages.

A wider geographic focus often identifies more available assets in a region

When devising economic development strategies, regions tend to focus on the immediate geographic
area, usually a metropolitan statistical area. In some cases this is appropriate. For example, the Atlanta MSA
is large and encompasses 20 counties. Commute times make it difficult to easily reach a wider geography
in one working day. In other cases, however, a focus on the metro area neglects a wide range of assets that
are readily accessible and part of the regional economy.

Research Triangle The Research Triangle area, for example, has considerable assets in regions just out-
side the metro area. Exhibit 14 on the following page shows the Research Triangle’s Metro Area employment
share and growth in share of the 41 traded clusters in the United States economy. The upper right
quadrant represents clusters that have a disproportionate share of national employment in Research
Triangle and are growing their share.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 21

 The Research Triangle has 34% of its MSA employment in six clusters (communications equipment,
pharmaceuticals / biotechnology, distribution services, business services, power generation, heavy con -
struction services) that have a relatively higher share of national employment and are growing in share of
national employment.

Exhibit 14: Research Triangle Metropolitan Statistical Area, Narrow Cluster Definition

Note: Data points falling outside the range are placed on the axes together with their national share and change in share respectively.
(y-axis, x-axis)
Source: Cluster Mapping Project at Institute for Strategy and Competitiveness, Harvard Business School

Exhibit 15 on the following page is the same graphic, but for the Research Triangle Economic Area. The
region has 43% of its EA employment in 12 clusters that have a relatively higher share of national employ-
ment and are growing in share of national employment. These additional six clusters (power transmission
and distribution, construction materials, prefabricated enclosures, agricultural products, heavy machinery,
and building fixtures, equipment and services) represent an additional opportunity for the Research
Triangle to integrate itself more efficiently into broader economy. Moreover, out of these six clusters, the
Research Triangle’s national employment share is ranked 34 th or better in three of these: power transmis-
sion and equipment (15 th ), prefabricated enclosures (14th ), and construction materials (26th ).

22 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 15: Research Triangle Economic Area, Narrow Cluster Definition

Note: Data points falling outside the range and placed on the axes together with them national share of change in share respectively.
(y-axis, x-axis)
Source: Cluster Mapping Project at Institute for Strategy and Competitiveness, Harvard Business School

Some regional economies are highly dependent on a few clusters and sometimes on a small
number of companies
Although all regional economies specialize in a few areas, some have especially narrow breadth. These
economies have a disproportionate share of employment in one cluster, and even in a handful of compa-
nies. Over reliance on a few industries and clusters is dangerous. Excessive narrowness exposes a regional
economy to having only a handful of clusters with high wages. While no region can have high wages in all
41 types of clusters, some diversification is important.

Wichita Although Wichita has diversified its economy during the last ten years, a disproportionate
number of workers are found in one cluster—aerospace vehicles and defense—and within only a few large
companies. In 1998, the aerospace vehicles and defense cluster alone employed close to 20% of narrow
traded cluster employment in the Wichita Economic Area. Furthermore, the majority of workers in the
aerospace vehicles and defense cluster are found in only two industries within one subcluster.
In addition, the region is heavily reliant on fewer than ten companies (see Exhibit 16 on the following
page). The four aerospace vehicles and defense manufacturers (Boeing, Beech, Cessna, and Lear), the
Coleman Company, and Koch together employed approximately 50% of 1998 narrow traded employment
in the Wichita MSA. The Milken Institute ranked Wichita fourth in terms of metro areas sensitive to
“high-tech recession” because of its concentration in aerospace vehicles and defense. 5

5Ross C. DeVol, “America’s High-Tech Economy: Growth, Development and Risks for Metropolitan Areas,” Milken Institute, July 13th, 1999.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 23

 Exhibit 16: Diversification of the Wichita Metro Area Economy

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

Pittsburgh: Despite having positions in oil and gas and defense industries, Pittsburgh was primarily
a steel town throughout most of the 20th century. Then, in the 1980s, a major downturn struck the steel
industry. Pittsburgh lost approximately 150,000 jobs in the steel industry alone. This was close to 90% of
the steel and steel related jobs, and almost 15% of the total jobs in the region. As a result, Pittsburgh expe-
rienced a steep recession.
Today, Pittsburgh has a much more diversified economy and has grown in more than 25 of its 41 traded
clusters during the past ten years. This has not only made the economy less vulnerable to the up-and-downs
of a single cluster, but has also opened up opportunities at the intersections of multiple clusters in which
Pittsburgh has a relatively strong position (e.g., metal manufacturing, construction materials, power gen-
eration, education and knowledge creation, transportation and logistics, production technology, etc).

“High-Tech” clusters account for a small percentage of jobs and wages in most
regional economies
Several types of clusters are especially innovative: communications equipment, analytical instruments,
biotech/pharmaceuticals, and information technology. These clusters are very productive, and pay high
wages, and regions with strength in these clusters certainly benefit from their presence. However, the
overall impact of these clusters on a regional economy is usually relatively small (see Exhibit 17 on fol-
lowing page). Clusters usually referred to as “high-tech” make up only 8.0% of traded employment, and
2.5% of total U.S. employment.

24 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 17: U.S. Traded Employment by Cluster, 1999

Note: Shaded clusters are normally considered “high tech”

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 25

 Atlanta For example, in Atlanta, information technology added less than 5% of the new jobs over the
1990s. Traditionally strong clusters such as business services, financial services, transportation and logis-
tics, and distribution services added nearly 70% of the new jobs over the decade (see Exhibit 18 below).
Even if Atlanta was twice as successful at creating information technology jobs in the future, it would do
little to increase the prosperity of Atlantans in general.

Exhibit 18: Job Creation by Cluster in the Atlanta Metro Area, 1990-1999, Narrow Cluster Definition

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

Higher wages in traded clusters tend to pull up wages in local clusters

It is difficult to increase the percentage of employment in traded clusters, because local clusters are gen-
erally service providers. As the number of traded jobs increases, so too does the number of local jobs that
service them. The way to increase overall prosperity of a region is to increase average wages in traded
industries (see Exhibit 19 on following page).

26 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 19: Traded Wages vs. Local Wages by Economic Area, 1999

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

IMPLICATIONS
• Defining the right region: The composition of regional economies can shift significantly depend-
ing upon the geographic area considered. Regions have a tendency to follow political jurisdictions
and to omit important surrounding areas and assets. A broader, geographic definition widens oppor-
tunities and brings constituencies together.
• Building a strategy: Successful regions build on their unique assets and strong clusters, where they
have the greatest advantages. Strength then spreads to additional clusters over time.
• Clusters of clusters: Focus on a few clusters exposes a regional economy to the booms and busts,
as Raleigh-Durham is now discovering. In extreme cases, such as Wichita, a regional economy can
be dependent the fortunes of a handful of companies. The case of Pittsburgh shows how a regional
economy overly dependent on one or a handful of industries can be severely affected. Regional strat-
egy should encompass a wide range of cluster, and be attentive to clusters that overlap. Overlapping
clusters offer potential synergies in skill, technology, and partnership.
• Widen innovative capacity to many clusters: The majority of traded jobs in any region are in
clusters that are not generally perceived to be high-tech (e.g., business services, financial services,
education and knowledge creation, transportation and logistics, and hospitality and tourism). In
order to meaningfully increase overall regional prosperity, innovative capacity must be built
in many clusters. The same tools apply as in IT or biotechnology: strong university programs in
cluster related skills and technology; efficient knowledge commercialization from local universities;
effective cluster trade associations; the presence of anchor companies; an environment that supports
start-up companies.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 27

 4 THE EVOLUTION OF REGIONAL ECONOMIES

The evolution of regional economies is a slow process that takes a significant amount of time. Change,
be it intentional or not, takes years to materialize and produce results.
Because the evolution process takes so much time, the influence of inherited endowments such as nat-
ural resources or geographic locations is often still recognizable in the current composition of regional
economies. For example, Pittsburgh’s proximity to oil fields in nearby parts of Pennsylvania and its access
to river transport routes set the region up for developing a strong production technology cluster around
the steel industry.
While inherited factors are important, they do not determine the evolution of a regional economy. The
evolution process has multiple equlibria and which one of them materializes depends on initial conditions
as well as on other influencing factors:
• Entrepreneurship. One important factor can be the entrepreneurial influence of an individual. An

idea becomes a firm, a firm becomes an anchor for spin-offs, and over time a cluster develops.
• Specialized Assets. Another important factor is the presence of a research/training institution, for
example a university. Specialized research and training provide a fertile ground for cluster development.
• Government Policies. A broader influencing factor is public policy. Political choices and prioritiza-
tions, for example the decision to invite the US Navy to port in San Diego, have a significant impact
on regional development.
• Past Trajectory. Finally, the opportunities for economic development are at every point in time
strongly influenced by the current composition of an economy. New activities often emerge out of
existing ones. For example, in Wichita earlier oil & gas activities and the presence of aircraft manu-
facturers created an environment conducive for the emergence of a plastics cluster.

Often the evolution process occurs under its own momentum without any intervention. But given the
existence of path-dependency and multiple outcomes there is also clear evidence that this process can be
affected by leadership and choices made in the private and/or public sector. For example, in Wichita the
city made the decision to build on of the first airfields in the US. Wichita still is one of the leading locations
of the nation’s aerospace cluster. In the Research Triangle, the decision was made to develop a strategy based
on the competencies of the three universities in the region. The Research Triangle now is one of the lead-
ing locations for the nation’s biotechnology and communications clusters.

28 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

FINDINGS

Successful Regions leverage their unique mix of assets to build specialized clusters
Successful regions build on their relative strengths, creating specialized economies that both differ from
other regions and offer comparative advantages to local companies. These relative strengths could be
inherited characteristics (e.g., geography, climate, population), or man-made assets (e.g., research centers,
companies, governmental organizations), that differentiate the region. Companies then draw on theses dif-
ferentiating assets to produce innovative goods and services, and generate and sustain innovative strategies
that make them more competitive.

San Diego The San Diego economy benefits from three sets of specialized assets in the region: the cli-
mate and geography, the military presence, and the constellation of bioscience research centers. The good
weather and beachfront location are obviously inherited assets, but local leaders judiciously used them to
create other specialized assets. In the military’s case, the harbor and the city’s geographic location proved
critical, and the City helped by dredging the Bay. For the research institutes, climate and the availability of
land were central, and again the City accommodated through zoning and land grants. Local government
and business leaders teamed up to lobby and get a University of California campus built in La Jolla. In all
cases, the region’s natural endowments were useful to the companies and also led to a high quality of life
which attracted talented people. Irwin Jacobs, a founder of Linkabit and now Qualcomm, came to San
Diego on sabbatical and chose to stay; Jonas Salk visited the region and decided to set up his research cen-
ter on the Torrey Pines Mesa (see Exhibit 20 on next page).
Virtually every strong and growing cluster in San Diego leveraged these specialized assets. The hospital -
ity and tourism cluster benefits from the climate, beach, and proximity to Mexico. The military pres ence
sparked growth in clusters such as aerospace engines, analytical instruments (primarily navigation equipment
in San Diego), communications equipment, information technology, and power generation (technology used
to drive ships was adapted to moving oil and gas through pipelines). The presence of research institutions
contributed to the development of the above clusters and drove growth in clusters such as education and
knowledge creation, medical devices, biotechnology / pharmaceuticals, and printing and publishing. Other
clusters grew out of the intersection of existing clusters. Transportation and logistics benefited from the naval
presence and the need to accommodate shipping, from the inflow of tourists, and from the flow of goods
going to and from Mexico. Sporting and leather goods, San Diego’s fastest growing cluster, began because
of the presence of military industries and tourism.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 29

 Exhibit 20: The Military, Climate, and Research in San Diego

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

Research Triangle The Research Triangle also used and enhanced the specialized assets they had to
attract large multinationals and research institutions. In the Research Triangle’s case, those assets centered
around educational institutions established by benefactors in the late 1800s and early 1900s, assets which
set the stage for a successful creation of the Research Triangle Park in the mid-1900s. North Carolina State
University in Raleigh (NCSU), the University of North Carolina at Chapel Hill (UNC-Chapel Hill),
and Duke University in Durham formed the pillars of the region’s knowledge-based economy by provid -
ing world-class research facilities as well as a critical mass of scientists, researchers and technicians. Their
research capabilities helped in the development of a large number of clusters in the region. These include
not only biotechnology / pharmaceutical and communications, but also plastics, chemicals, fibers, medical
devices, analytical instruments, and education and knowledge creation.

Building strong regional economies takes decades

There are many steps involved in building a regional economy, and inevitably this takes time. Regions
inherit assets (e.g., geography, universities, workforce pool, successful companies) which public and
private sector leaders can leverage. Companies develop from these assets. The activity attracts other com-
panies to locate in the region. Human talent, as well as related and supporting industries are then attracted.
Finally, this process is repeated across a number of clusters.

30 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Research Triangle For example, it took 20 years to build a large corporate R&D presence in the
Research Triangle, and another 20 to see significant economic consequences flow from it. The Research
Triangle Institute opened in 1958, but the biotechnology / pharmaceutical cluster did not take off until the
late 1980s. In both cases, local leaders had to commit significant resources to assemble a critical mass of facil-
ities and institutions, and then wait many years to witness the economic returns (see Exhibit 21 below).

Exhibit 21: Four Decades of Development in the Research Triangle Region

Source: Clusters of Innovation Initiative Interviews, Research Triangle Region

Atlanta With the selection of Atlanta to be the hub for the Georgia Railroad more than 150 years ago,
the region began its ascent from a transportation hub to an international commercial center. The success
enjoyed today is the result of a history of proactive business and government leadership. In 1885, heavily
concentrated in textiles, the Georgia Institute of Technology was founded by the city and state to train the
new generation of industrial engineers and business leaders. In 1980, due to intense private sector lobby-
ing, Georgia Tech President Joseph Petit and Governor Mike Bushee opened the Atlanta Technology
Development Center—designed to spur technology firm development. In the wake of an unsuccessful bid
for the Microelectronic and Computer Consortium, Governor Joe Frank Harris provided state funds to
open six research centers through the University of Georgia from 1986 to 1990. Again the private sector
played a pivotal role when they encouraged Governor Zell Miller and the state assembly to sponsor the
Georgia Research Alliance, an organization that brings together business, government, and the six
regional research institutions. By the late 1990s, this sustained, long-term effort resulted in Atlanta
being able to claim 11 Fortune 500 companies that generated more than $200 billion in sales. (See
Exhibit 22 on next page.)

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 31

 Exhibit 22: Timeline of Atlanta’s Economic Development

Source: Clusters of Innovation Initiative Interviews, Atlanta

Wichita Wichita’s mature aerospace vehicles and defense cluster is the result of 80 years of purposeful
planning, investing, and coordination from the large manufacturers and the local, state, and federal gov -
ernment. The U.S. government invested hundreds of millions in Wichita’s aircraft companies in terms of
facilities, training, and research in order to scale up to meet the needs of wartime (World War II and the
Korean War) aircraft production. The local government also had the foresight to invest a great deal of funds
to improve the local transportation infrastructure around the aircraft production sites and offer hundreds
of millions in industrial bonds for the development of the sites themselves. The state and federal govern-
ment invested in the National Institute for Aviation Research (NIAR), Wichita’s only research institution.
More investment is needed from state and federal sources for NIAR but it is an extraordinarily important
driver of innovation in the region. The most demonstrable example of local leaders’ long time-horizon was
the well-coordinated push for an expensive K-12 bond vote. Recognizing that the labor pool was one of
the region’s most important but undeveloped assets, the private sector worked closely with the city of
Wichita and the Chamber of Commerce to convince citizens that new investment was needed for the
schools. Also, the region’s economic development leaders are stepping up to create a new workforce
development program in concert with the aircraft manufacturers, although this might not yield results
for years to come.

32 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Institutions for collaboration play an important role in building regional economies
There were many different types of institutions for collaboration present in every region we studied.
These institutions for collaboration include formal organizations and informal networks that facilitate the
flow of information, ideas, and resources among firms and supporting institutions. Examples of collabo -
rative institutions that we found include chambers of commerce, local branches of nationwide industry
associations, region-specific industry associations, and university technology transfer offices.
Survey respondents indicated that these types of institutions of collaboration were critical in these regions
because their presence means that several important functions—promotion of the region, recruitment of
companies interested in relocating to the area, lobbying on behalf of business interests, and ensuring the
legal transfer of intellectual property from universities to companies—are performed. Not all of these
institutions for collaboration were effective and not all regions had effective institutions for collaboration.
In fact, most regions had only a handful of truly effective institutions for collaboration. (See Section 5,
Exhibit 26.)

San Diego In response to private-sector efforts to better integrate with the University of California at
San Diego, University President Richard Atkinson asked Mary Walshok, Dean of Extended Studies and
Public Programs, to develop a program that would facilitate university-business interaction. After coordi -
nating with both university researchers and private sector managers, UCSD CONNECT was established
in 1985. Initial programs included:
• Business Environment Assessment. CONNECT sponsored a study to determine what
business leaders felt were the major gaps hindering their success in San Diego.
• Meet the Entrepreneur and Meet the Researcher Events. Entrepreneurs and scientific
researchers had very little understanding about the issues that each faced, or the way each con-
ducted their operations. There was almost no connection between the two groups, but as initial
events that attracted hundreds of participants showed, there was a lot of interest in learning about
each other.
• Financial Forums. These forums brought leading capital providers to the region and educated
them about San Diego companies in order to encourage investments. The forums also con-
nected entrepreneurs with business support services (law, accounting, and marketing firms) to
help them develop and present more effective business proposals.

Research Triangle The Centennial Campus is North Carolina State University’s (NCSU’s) “tech-
nopolis”, a planned mix of university, corporate and government R&D facilities and business incubators, with
a town center, executive conference center and hotel, housing, and recreational facilities. This 1,334-acre site,
adjacent to NC State’s main campus, separates buildings into R&D neighborhoods with multidisciplinary
themes based on the University’s strengths in advanced research and client-driven training programs:
Advanced Communications Technologies, Biosciences and Biotechnology, Advanced Materials,
Environmental Technologies, and Pre-College Education. The campus, home to well over 100 large and
small companies, government agencies and NCSU units, received high praise from our survey and inter-
view respondents.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 33

 IMPLICATIONS
• Long time horizon: Meaningful changes in regional development require investments that
generally take decades before significant dividends are reaped. Long time horizons create challenges
for leaders seeking reelection, and make it more difficult to mobilize community support
behind an economic development strategy. These realities highlight the need for an institution
structure for regional development that goes beyond government, as well as the need for private
sector involvement.
• Building on traditional industries: Inherited assets matter to a region’s economic develop -
ment. Established and already emerging clusters offer the greatest prospects for near term growth.
Strengthening established clusters should be one of the early priorities in regional economic
development.
• Investing in unique and specialized assets: Many inputs in regional prosperity often require
substantial investments in specialized assets such as university campuses, research programs,
logistical infrastructure, and the like.

34 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 THE REGIONAL BUSINESS ENVIRONMENT

5
The productivity and innovativeness of a regional economy benefit from overall conditions such as
a sound fiscal policy, an effective political decision making process, and sound legal institutions. However,
broad regional attributes such as these are increasingly preconditions, not sources of competitive advantage.
Prosperity in a region is actually created by the microeconomic foundations of competitiveness, rooted
in the sophistication with which individuals, firms, and industries based there compete. This is what gives
rise to productivity. Competitiveness requires ongoing improvement in the quality of corporate manage-
ment and in the sophistication of company strategies and operating practices. However, the sophistication
with which firms compete rests heavily on the quality of the regional business environment in which they
operate. For example, the productivity of companies is affected by such things as the specific skills of employ-
ees they can attract, the efficiency of the local logistics and
transportation system, and the extent to which local reg- Exhibit 23: Determinants of Regional Productivity
ulations impede productivity and innovation or
encourage them.

DETERMINANTS OF REGIONAL
PRODUCTIVITY
The quality of a region’s business
environment is embodied in four
broad areas (see Exhibit 23). Each of
them affects the level of productivity
that can be achieved as well as the rate
of innovation.6

6 See Michael E. Porter, The Competitive Advantage of Nations, New York: The Free Press (1990).

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 35

 Factor conditions: Achieving high levels of productivity depends on the presence of high quality and
specialized pools of human resources, applied technology, infrastructure and even sources of capital that
are tailored to the needs of particular industries. The quality of more generic and basic factors such as high
school graduates or the local transportation system are foundations that every region most have.
Increasingly, competitiveness depends on the presence of advanced and more specialized factors. For
example, both Wichita and the Research Triangle offer comparable basic infrastructure. However, the
Research Triangle region has achieved superior innovation performance due to much better advanced fac-
tor conditions in the form of a highly developed and specialized university research and advanced training
capability at Duke, North Carolina State University, and the University of North Carolina.
Demand conditions: The quality of demand in a region has a strong influence on the process of cre-
ating and improving products and services. Sophisticated customers in the region press firms to improve
and offer insights into existing and future customer needs.
Traditionally, regions and countries focused on the size of their local market. A large local market, it was
believed, would allow local companies to exploit economies of scale and improve competitiveness. When
productivity drives competitiveness and firms can easily access national and international markets, howev-
er, the quality rather than the quantity of local demand becomes important because it is crucial for
innovation. For example, the high state of medical practice in Boston has triggered a stream of innovation
in Boston-based biotechnology and medical device companies.
Context for firm strategy and rivalry: The rules, incentives and pressures governing the competi-
tion in a region have a fundamental influence on productivity. Policies that encourage investment, protect
intellectual property, and open the local market for trade, for example, foster productivity growth and com-
petitiveness. Also exerting a strong influence on productivity is the presence of competing rivals in a region
and the intensity of local industry. Spirited local rivalry among Wichita’s Beach, Cessna, Lear, and Boeing
subsidiaries, for example, drove the emergence of the region as the world’s center for general aviation air-
craft. Local rivalry pressures companies to improve, while attracting suppliers and otherwise improving the
overall business environment.
Related and supporting industries: Local sourcing from capable suppliers based in the region can
enhance productivity and improve the capacity for innovation through allowing quicker and less costly
communication, fostering the flow of ideas, and enhancing flexibility through local outsourcing.
Traditionally, many regional development programs have focused on attracting individual companies and
industries. However, isolated companies cannot be productive without the presence of related and
supporting industries. The presence of related and supporting industries gives rise to new clusters. For
example, the plastics cluster in Wichita developed from the presence of petrochemical producers com -
bined with local aircraft manufacturers who became customers for plastic parts.
These four areas of the regional business environment are self-reinforcing and act as a system. Regional
rivalry, for example, stimulates the development of unique pools of specialized skills and the formation or
attraction of specialized suppliers. Active local rivalry also upgrades regional demand by creating more
demanding customers. Weaknesses in any part of the business environment, then, can erode the compet -
itiveness of regions.
All parts of the business environment affect a region’s productivity and competitiveness. A subset of the over-
all environment has particular importance in determining a region’s capacity for innovation (see Exhibit 24 on
following page).

36 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 24: Innovation and the Microeconomic Business Environment

Government
Government affects competitiveness through its influence on the business environment (see Exhibit 25).
Government at all levels influences (positively or negatively) the business environment and the productivity
of clusters. Government is not monolithic, and its influence occurs through a myriad of distinct departments
and entities. While the Federal government is often seen as having the greatest impact on competitiveness,
policies at the regional and even local level are often equally if not more important. Each level of govern-
ment affects various aspects of the business environment, and the policies of different units of government
can frequently be conflicting.

Exhibit 25: Government’s Impact on Regional Productivity

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 37

 Government plays five distinct types of roles in competitiveness, some of which are
often overlooked:
1) Establish a stable macroeconomic, political, and legal environment.
2) Improving the availability, quality, and efficiency of generalized inputs, infrastructure, and
institutions, such as roads and schools.
3) Setting the overall rules and incentives governing competition such as investment incentives,
antitrust laws, and intellectual property protection rules.
4) Facilitating cluster development and upgrading; and
5) Establishing and participating in an ongoing process for defining regional competitive priorities and
implementing them across constituencies.
Most governments realize the first three roles. Few governments have effectively addressed the fourth
and fifth roles and the second role as far as basic factor conditions are concerned. Advanced factor condi-
tions, incentives, and the institutions and processes of cluster development have become much more
important roles of government.

Government’s proper role is to improve the business environment rather than to intervene
directly in the competitive process. Government should not subsidize individual companies but work
to raise the productivity and innovativeness with which companies can operate. Many U.S. regions, for
example, have traditionally sought to attract industry through tax incentives and driving down the cost of
doing business in terms of payroll taxes, unemployment insurance, utilities and the like. This approach
may be necessary in uncompetitive regions, but it is ultimately self-limiting. Pushing down costs can
reduce the revenue necessary to improve education, infrastructure and services. Improving the productiv -
ity of the region, and boosting its innovative capacity, is more effective in increasing standard of living in
the long run.
The traditional separation between the public and private sector no longer applies. In the old
model, the public sector was to provide the infrastructure while the private sector focused on competition.
In the new model, the level of co-dependence of public and private sector has hugely increased: The pub-
lic sector needs to set policies in close interaction with the private sector while the private sector derives
key sources of its competitive success from outside the firm. The new model also includes a much broader
set of institutions such as universities, regulatory bodies, and trade associations.

Institutions for Collaboration

Exhibit 26: Examples of Institutions for Collaboration
Institutions for collaboration are formal and informal
organizations and networks that (1) facilitate the
exchange of information and technology; and (2) foster
various kinds of local coordination and collaboration that
can improve the business environment in the overall
economy or in a cluster. Institutions for collaboration,
then, create and amplify the arrows and feedback loops
(See Exhibit 26).

38 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Institutions for collaboration take various forms (see Exhibit 26). Some are economy-wide or address broad
sectors, while others are cluster-specific. Institutions for collaboration affect productivity and innovation in a
number of ways. First, they create relationships and enhance the level of trust in these relationships. Second,
they facilitate the organization of collective activity. Third, they encourage the definition of common standards,
rules, and norms that stimulate competition or boost productivity. Finally, they can be mechanisms to devel-
op a common economic or cluster agenda.

Economic Attitudes, Values and Beliefs

A final influence on regional competitiveness and innovative capacity is more intangible. Attitudes, values
and beliefs towards the economy—which are often termed “culture”—bear on the behavior and aspirations
of individuals, firms, and other institutions in a region. Of particular importance is an advanced economy
like the U.S. are beliefs about the importance of entrepreneurship, attitudes towards collaboration, and
civic mindedness.

FINDINGS COMMON ACROSS REGIONS

Regions differ substantially in the relative strengths and weaknesses of the business environment.
Nevertheless, a number of issues are common across regions. Our research found that a number of factors in
the business environment were important in all regions. Strength in these factors is a basic ingredient to the
success of any regional economy. These factors include: a strong physical infrastructure, quality primary and
secondary education, excellent universities and specialized research centers, a deep pool of specialized talent,
a business-responsive government, and reasonably efficient coordination among local political jurisdictions.

A strong physical and information infrastructure is a baseline requirement to establish and

sustain a prosperous regional economy
Roads, highways, airports, railroads, water, and power support the efficient movement of people, goods, and
services as well as the quality of life of citizens. All five regions we studied exhibited survey results indicating
that physical infrastructure was one of the most serious challenges facing regions. A deteriorating physical
infrastructure was cited as the leading cause of a deteriorating quality of life by a majority of survey respon-
dents and interviewees (see Exhibits 27 and 28 on following page).
While communication infrastructure was not cited as a problem in any of the regions studied, there was
solid support for the government to make its improvement a priority. Communication, including Internet
access, not only impacts the information technology sector, but affects the efficiency of new business prac-
tices, like e-commerce and logistics. Atlanta is an excellent example of a region that has enjoyed greater
economic success due to major investments in their communications backbone to support the 1996
Olympics and the media giants CNN and Cox Communications.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 39

 Exhibit 27: Government Priorities — Selected Survey Results and Representative Quotes from Interviews

Source: Clusters of Innovation Initiative Regional Surveys and Interviews

Exhibit 28: Physical Infrastructure — Good vs. Poor Innovation Locations

Source: Clusters of Innovation Initiative Regional Surveys and Interviews

San Diego Sixty five percent of biotechnology/pharmaceutical executives and 56% of communications
executives listed the quality of transportation as the second greatest threat to their business expansion. This
issue consistently emerged in both surveys and interviews as a high priority problem, with individuals citing
the need to increase air transport capacity for both passengers and cargo to support many large and fast
growing clusters, such as transportation and logistics, business services, hospitality and tourism, biotech-
nology/pharmaceutical, communications, and information technology.

40 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Although San Diego International Airport is centrally located, it lacks both frequent and direct flights to
many destinations. Population and economic growth has exceeded the airport capacity and will continue
to do so. A recent study estimated that by 2030, the cumulative opportunity cost of not expanding the
regional facilities was between $29.6 billion and $93.8 billion. In 2030, there would be 34,000 to 56,000
fewer jobs. More than half of the job losses, and more than 75% of the lost gross production would occur
in innovative clusters. For example, a number of biotech firms now anticipate the need for air transport as
their products become ready for commercial distribution. 7
A second weakness of the physical infrastructure, as discussed by interviewees, is ensuring reliable, cost-
competitive access to power and water. Southern California is dry, heavily populated, and growing rapid ly.
Water is a perennial issue in the state and demographic shifts guarantee that it will remain an issue. More
recently, deregulation of the state’s energy industry has led to short supplies, higher energy prices, and
even rolling blackouts throughout the state. Although Northern California has been hit the hardest,
rising utility prices affect the entire state.

Atlanta Throughout its history the geographic location and physical infrastructure of the Atlanta region
have served as a strong economic asset. More than 200 million people, 80% of U.S. consumers, are within
two hours’ flight time from Atlanta, or one day’s trucking by highway. However, Atlanta’s road trans-
portation infrastructure has not been able to keep up with its population growth. While the state has
been actively building new roads throughout the decade, traffic and congestion have increased signifi -
cantly. According to the Texas Transportation Index, the average Atlanta driver experienced delays of 68
hours per year in 1997, fourth worst in the country. In 1990, the average Atlanta driver experienced only
27 hours of delay. While Atlanta does have a rail transit system, MARTA, it presently only serves two
counties in the 17 county metro area, too few to make a significant dent in the congestion. Atlanta busi-
ness leaders say that they sometime have trouble attracting workers from within the metro area because
the traffic is so intense within the area. As one business person said, “we have even lost employees simply
because they moved from the Northern suburbs to downtown.”
The rapid growth in some communities has also outpaced the regional authorities’ ability to provide
appropriate sewer services. According to a real estate executive, the problem is particularly acute in the
Buckhead business district where there are “five commercial buildings ready to come out of the ground
waiting for the capacity to expand.” Parts of Southwest Atlanta that lie well within the city limits are forced
to rely on septic systems.
While water has not been a significant problem to date, the State of Georgia is in a battle with Alabama and
Florida over the flow from the Chattahoochee River. For now, Georgia has captured the quantity it needs.
In contrast, thanks to significant capital investments by the city, state and Delta Airlines, Hartsfield
International Airport has grown to take advantage of the region’s location. In 2000, it was the busiest air -
port in the world for passenger travel and handled more than 650,000 metric tons of cargo. Hartsfield has
a $16 billion regional economic impact annually 8, and 44,800 airport employees, making it the largest
employment center in GA.9 While some Atlanta leaders complain that parking at Hartsfield is difficult,
most recognize that the airport, which supports more than 1,000 flights a day, provides the region signifi -
cant economic benefits.

7 “The Impacts of Constrained Air Transportation Capacity on the San Diego Regional Economy,” Hamilton,
Rabinowitz, and Alshuler, Inc., January 5, 2001. The lower estimate assumed maximum expansion of SDIA, while the
higher estimate assumed no change to existing facilities.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 41

 In addition, Atlanta has an excellent communications infrastructure, thanks to major investments made
to support the Olympics and the major media companies like CNN and Cox Communications. Building
on its reputation as a transportation hub, the region has also become a major Internet connection hub.
Leading Internet Service Providers (ISPs) like firms like Mindspring and BellSouth offer services to take
advantage of the high-speed Internet connectivity. Eighty-seven percent of our survey respondents said
that the communications and Internet infrastructure fully satisfied their business needs.

A strong K-12 educational system is important for developing local talent and attracting
outside talent
Community leaders across all five regions placed a high premium on quality K-12 education and
expressed concern about the quality of public schools in their region. The importance of K-12 education
is growing ever more critical because it establishes both the pipeline of talent for entry-level jobs and the
pool of specialized talent critical to cluster development (see Exhibit 29 below).
Demands on the K-12 system have changed in some fundamental ways over the past decade. First, with
greater national and international mobility of capital and companies, the quality of a region’s K-12 system
has become a key differentiator in attracting business investment. Second, most jobs now require a high-
er level of technical and reasoning skills among the workforce at large, not just those requiring advanced
degrees. Third, every high school graduate needs a working knowledge of computers and software to
succeed in the marketplace as well as in higher education.

Exhibit 29: Government Priorities — Selected Survey Results and Representative Quotes from Interviews

8 “Mayor’s Statement” www.atlanta-airport.com

9 Airport Statistics www.atlanta-airport.com

42 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Universities and specialized research centers are the driving force behind nearly every region
Universities and specialized research centers were instrumental in providing the driving force for inno-
vation in nearly every region we studied (see Exhibit 30 below). In San Diego, the University of California
at San Diego, the Salk Institute, the Scripps Research Institute, the Navy’s SPAWAR, among others were
instrumental in generating new ideas and technologies. In Pittsburgh, the most important sources of inno-
vation were Carnegie Mellon University, the University of Pittsburgh, and Dusquense University. In
Research Triangle, many companies (e.g., IBM, Glaxo-Wellcome) were strong innovators. Nevertheless,
the underlying source of talent and ideas are Duke University, the University of North Carolina at Chapel
Hill, and North Carolina State University. In Atlanta, Emory University, and the Georgia Institute of
Technology were prominent.

Exhibit 30: Specialized Research — Good vs. Poor Innovation Locations

Source: Clusters of Innovation Initiative Regional Surveys

San Diego This region not only makes considerable investments in R&D, it also excels at commer-
cializing the knowledge it generates. Part of the reason is the presence of many different types of research
institutions. The University of California at San Diego, a large public university, contains numerous spe -
cialized research centers. In addition, there are many non-university research centers such as the Scripps,
Salk, and Burnham Institutes. Yet another model is the joint university-industry research institution.
Examples of this include the Center for Wireless Communications and the California Institute for
Telecommunications and Information Technology. Finally, there are also government research centers
such as the Navy’s Space and Naval Warfare Systems Command (SPAWAR).

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 43

 Pittsburgh The region’s research workforce has been depleted since the mid 1960s, when it had 14,500
private sector researchers. The number dropped to 8,000 in the 1970s and has stagnated at that number
through the 1990s, making it difficult to improve innovation levels. In fact, Pittsburgh’s patenting numbers
have been regressing through the 1990s, compared to the national averages. Although local universi ties
increased their researchers to more than 5,500 and have more than double the national per capita expen-
diture on research, difficulties with commercialization limit their impact. There is a great opportunity to
mobilize these resources, and university leaders understand this and have made strong improvements.
Now, they must continue to escalate this effort and bring in more private sector firms.

Exhibit 31 — Specialized Talent and Training: Good versus Poor Innovation Locations
Your region has an ample supply of high quality...

Source: Clusters of Innovation Initiative Regional Surveys

Exhibit 32 — Specialized Talent and Training: Good versus Poor Innovation Locations

Source: Clusters of Innovation Initiative Regional Surveys

44 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Specialized talent and training are more important than abundant, low-wage labor
In interview after interview, executives report that they move operations to particular locations in large
part because of the local talent pool. However, it is not abundant,low-paying labor that attracts innovative
companies, but highly talented, specialized, and often expensive labor. This type of labor tends to pool in
various regions around the country, and hence regions with an abundance of specialized workers offer
advantages to companies that other regions cannot match. There are at least four reasons for this: a mass
of specialized workers appears because specialized training institutions already exist; the numerous compa-
nies in a regional cluster offer specialized training to their workers; outside companies are attracted to the
region because of the labor and their relocation brings more specially trained workers to the region; and peo-
ple from outside the region who happen to get the specialized training are more willing to move to the region
because they know there are more employment opportunities in the region (see Exhibits 31 and 32).

Wichita Aside from flat land and good wind, the Wichita region has fewer natural assets or endow -
ments as compared to other regions studied. What attracted Bill Lear and dozens of suppliers to set up
operations in the region, earned Wichita the title of “Air Capital of the World,” and has recently attracted
the Airbus company to investigate the creation of a possible design facility, is not hospitable flying conditions
but rather the existence of specialized talent. To the benefit of the aerospace vehicles and defense cluster,
workers proficient in building aircraft tend to circulate from one large manufacturer to another and from
manufacturer to small supplier. The region’s thick aircraft labor market has seeded the development of other
manufacturing-oriented clusters in the region, transforming Wichita from an aerospace cluster to a major
manufacturing center. In fact, demand for specialized labor is so important in the region that it has become
the aerospace vehicles and defense cluster’s Achilles heal. Availability and cost of labor topped the list of con-
cerns among those surveyed both within the cluster and throughout the region as a whole. In response to
labor shortage concerns, the cluster is embarking on a state-of-the-art workforce-training program.

San Diego UCSD, San Diego State University, local private universities, and the region’s community
colleges offer a variety of general courses and specialized programs at the undergraduate, graduate, and
continuing education levels. One example is the SDSU Center for Bio/Pharmaceutical and Biodevice
Development, which recently launched its first program, a Master of Science in Regulatory Affairs.
Interviewees report satisfaction with local training and talent.

Research Triangle UNC, NC State University, Duke University, local private universities, and the
region’s community colleges offer a variety of general courses and specialized programs at the under -
graduate, graduate, and continuing education levels. Interviewees report satisfaction with local training
and talent as far as quality, but not quantity. The Research Triangle is home to a comparatively large
number of skilled workers in the communications cluster. According to the Bureau of Labor Statistics, in
1998 there were 4,480 electrical or electronic engineers and more than 2,900 electrical or electronic engi-
neering technicians in the region, well above the national average. In recent years, the Research Triangle
has improved its academic and training infrastructure to support specialized research in communications

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 45

 related fields. For example, the optical networking companies are developing partnerships with educators
to train technicians to support their growth. Nortel premiered its Optical Internet Technology training
program that it is offering in conjunction with Wake Technical Community College. Students who suc-
cessfully complete the 15-week pilot program, held at the Raleigh Technical Education Center, a Nortel
facility, can become candidates for Nortel certification – and employment. Competitor Cisco offers its
own brand of optical network training through its Networking Academy Programs. In recent years Cisco
has enrolled students in classes at community colleges, high schools, homeless shelters, and juvenile
centers, and near military installations for outgoing personnel. And Lucent, with its growing optical net-
working center on NCSU’s Centennial Campus, is taking advantage of the people-power and research
available through this type of partnership.

Government can have a significant influence on the business environment, both positively
or negatively
In general, government’s influence on innovation comes through its impact on factor inputs, context for
firm rivalry, demand conditions, and related and supporting industries. Federal and state governments con-
trol much of the legal, fiscal, and regulatory framework (e.g., fiscal policy, competition policy, protection
of intellectual property, and liability law) that affects demand conditions and the context for firm rivalry.
Taken together, these policies create the incentives (or disincentives) for the high levels of investment
and commercialization that characterize an innovation economy. According to our survey, business and
community leaders in the regions identified three primary areas in which local governments must be—
or become—highly effective (see Exhibit 33 and 34 on next page):
1) Speed up regulatory approvals wherever possible, including zoning and licensing, and simplify
compliance procedures.
2) Support for local entrepreneurial activity, including catalyzing partnerships, stimulating funding
for incubators, and supporting start-up ventures.
3) Catalyze the linkages between government and business communities that create the groundwork
for collaboration.

In highly innovative regions, the private sector plays an active role in identifying challenges and work-
ing collectively with government to address them. In innovation-poor regions, by contrast, the prevailing
business attitude is less collaborative and perception that government simply needs “to get out of our way.”

46 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Exhibit 33: Government Priorities—Selected Survey Results

Source: Clusters of Innovation Initiative Regional Surveys

Exhibit 34: Government Priorities —Selected Survey Results

Source: Clusters of Innovation Initiative Regional Surveys

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 47

 Wichita Local government played an important catalytic role in the establishment of the important
aerospace vehicle and defense cluster and has provided economic development assistance since the 1960s.
The seeds of Wichita’s success were sown around the turn of the last century when citizens passed a
$200,000 bond and helped Wichita beat out other cities to attract a railroad. Real estate speculation, aggres-
sively promoted by the city’s two newspapers, followed the railroad. The city’s economic development
organization, the Board of Trade, encouraged manufacturers to come to Wichita with incentives of land
titles. Wichita also funded a landing field to capitalize on the interest in aviation and recruited Clyde
Cessna and Bill Lear. Beginning in 1962, Wichita floated its first industrial bonds. Local government con -
ducted aggressive firm recruitment in the 1970s, following a downturn in the economy. Since 1979, local
government has offered $3 billion in revenue bonds.
More recently, the state of Kansas is widely credited for setting in motion a number of economic
development programs that followed intensive statewide strategic planning during the mid-1980s.
These strategies were prepared in response to a recession in the aerospace vehicles and defense, oil and
gas, and agriculture clusters and a collective fear that Kansas was losing its competitive edge. As a
result, the Kansas Technology Enterprise Corp. (KTEC) was created to stimulate economic devel-
opment by fostering innovation and development of technology. KTEC offers seed capital as
exemplified through the Wichita Technology Corporation, research funding for such institutions
as the National Institute for Aviation Research, and support for technical assistance. In addition,
Kansas Venture Capital, Inc. was created to attract private venture capital in the state through tax
credits and a matching public fund of $5 million.

Research Triangle In 1958, with the economy of central North Carolina still dependent on maturing
industries such as tobacco and textiles, the state’s government and business leaders, including North
Carolina Governor Luther Hodges, set out to foster economic development through far-sighted invest -
ments in universities, research centers, and infrastructure. Governor Hodges, who earned the reputation
as the “businessman’s governor,” led education reform, devoted significant resources to post-secondary edu-
cation, built vocational schools, and embarked on an persuasive industrial recruitment and incentive program.
Alongside these state-led efforts, the local government responded by providing land, special tax zones,
and other incentives aimed at luring large, high-tech companies and organizations to the area. Challenged
by discontinuities, cluster-generating companies and crusading individuals (e.g., former Governors
Hodges, Sanford and, later, Hunt) forged inter-relationships, promoted the development of supportive
institutions (e.g., the Research Triangle Foundation), and pioneered the growth of the present clusters. As
mentioned by a prominent economic development official in one of the cluster interviews, “The shift from
an agricultural focus took real leadership. As a result, the aspirations and expectations of the citizenry have
changed a lot over the last 20-30 years.”10

10 Clusters of Innovation Initiative Interview, Research Triangle, November 6, 2000

48 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Poor coordination among local jurisdictions often impedes efforts
to improve the business environment
One of the keys to successful regional development is a long-term vision for sustainable growth and shared
understanding and coordination between public and private sector leaders about the strategies and goals. The
multiplicity of local jurisdictions encompassed by a growing local economy often impedes this process.
Historically, the location of county seats was determined by a day’s mule drive. Today, sprawling
regional economies cut across these narrow political boundaries, often complicating the process of
regional planning for “smart” growth. Communities face challenges in workforce development, educa -
tion, transportation, housing, and land zoning that require a more coordinated metro-wide approach.
Yet, local political leadership is often fragmented and disconnected. Local jurisdictions do not routinely
coordinate their policies and planning. In rapidly expanding areas, local political leaders may not even
know each other. At the same, however, interviewees indicated that mayors and other locally elected offi -
cials lack the authority to make sound regional policy that affects their economic prospects and quality of
life (see Exhibit 35 below). As a result, a number of the regions studied are experimenting with new forms
of regional coordination but new regional structures are still in the formative stages.

Exhibit 35: Government Collaboration— Representative Quotes from Interviews

Source: Clusters of Innovation Initiative Interviews

Atlanta Atlanta developed a regional solution to address the critical shortfalls in its transportation sys-
tem. In 1999, the Georgia Regional Transportation Agency was created to address multi-jurisdictional
sprawl in the metro Atlanta region. The new agency was given the power not only to build or veto new
road and transit systems, but to affect development patterns by controlling the way new office parks or res-
idential development ties into the transportation network.

Pittsburgh Allegheny County in Pennsylvania described itself in the mid-nineties as having the most
fragmented government structure of any metropolitan county in the United States with 130 municipalities,
the smallest with a population of 100, and 43 school districts. Looking at its prospects for the 21st cen tury,
recommended the creation of a county-wide executive to coordinate economic development and planning
among the local municipal jurisdictions.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 49

 Research Triangle The Research Triangle region has been the 12th fastest growing MSA in the U.S.
over the last decade. Growth has spurred the increase in the political independence of local jurisdictions
that used go along with more regional economic development plans. The former vision of the region as
a growing center of research excellence (built around Research Triangle Park and the metropolitan area)
has reached its limits in terms of being able to generate consensus behind regional economic development
programs. Collaboration among local leaders and institutions for collaboration in the area is more difficult
today. This has made it hard to push through economic development strategies that benefit all jurisdictions
and not just one.

Overall Regional Strengths and Weaknesses

In addition to issues of concern across all regions, business environments have their own unique
strengths and weaknesses. When we examine a region, we look at a complex set of regional business envi-
ronment data. These additional factors in the business environment differentiate regions. Although
strength in these factors confers an advantage on a region, weakness is not necessarily crippling. Other
strengths can compensate for these weaknesses.

Exhibit 36: Determinants of Regional Competitiveness: Atlanta’s Relative Position

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School, Clusters of Innovation Initiative Regional Surveys and Interviews, Atlanta

50 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 37: Determinants of Regional Competitiveness: Pittsburgh’s Relative Position

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School, Clusters of Innovation Initiative
Regional Surveys and Interviews, Pittsburgh

Atlanta Atlanta has a strong business environment overall. Particular assets that have differentiated
Atlanta, however, include its historically high quality of life, strong regional institutions for collaboration,
the state government, and a history of public-private-university collaboration. Institutions like the Metro
Chamber, and the Technology Alliance of Georgia take an active role in pushing economic development.
The State government has initiated programs such as ICAPP (Intellectual Capital Partnership Program) to
partner with firms and educational institutions to provide focused workforce development; and the
Georgia Research Alliance. (See Exhibit 36 on the previous page.)

Pittsburgh The main assets in Pittsburgh’s business environment are its numerous specialized
research and training institutes and high levels of R&D spending. However, slippage in knowledge com -
mercialization limits the impact these research centers have on the economy. Furthermore, the environment
for start-ups is challenging, due to lack of networking, lack of mentorship opportunities, lack of experienced
entrepreneurial talent, and limited VC. Perhaps the most important challenge in the business environment
involves the level of collaboration across diverse groups in the region. (See Exhibit 37 above.)

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 51

 IMPLICATIONS
• Challenges of success: Successful regional economies tend to experience rapid growth, which
stresses the physical infrastructure. Foresight and a conscious strategy are needed to maintain and
improve infrastructure in advance of the strains caused by growth.
• Recognize the need for strategic transitions: Over time, regional development strategies run
their course. Success at one strategy creates the challenges that need to be addressed by the next strat-
egy. For example, Research Triangle’s strategy of focusing development inward on a research park,
amidst three strong local universities proved a great success. Now, however, it is time to use these
assets to reach outward to surrounding communities to leverage their assets with the knowledge
resources proximate to Research Triangle Park.
• Institutionalizing innovation: Regions often prosper from the efforts of great individual entre-
preneurs. Andrew Carnegie’s innovations in production technology, combined with the specialized
assets of southwestern Pennsylvania, were instrumental in making Pittsburgh the steel capital of the
world. Similarly, the innovations of Stedman, Cessna, and Beach made Wichita the air capital of the
world. This model of regional development, however, leaves success almost entirely to chance.
Successful regions do not rely on chance, but rather seek to institutionalize the innovative process
by building strong universities and research centers, and by attracting research divisions of major
companies, to create continuous innovation and entrepreneurship.
• Moving to commercialization: Commercialization is a vital step in the innovation process.
Some regions have high levels of R&D investments and numerous specialized research centers, but
still lag in terms of innovation output because knowledge is not effectively transferred to companies.
Having many different types of research institutions (e.g., public universities, private universities,
for-profit research centers, non-profit research centers, etc) appears to foster commercialization.
Different types of research centers interact with companies in different ways, with varying approaches
to deploying intellectual property, varying incentive structures for researchers to work with companies,
and varying cultures toward commercialization. This gives companies a range of options for
interacting with research centers. Second, a diversity of local research institutions often gener -
ates competition among these institutions, improving effectiveness in knowledge transfer.

52 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 CLUSTERS

6
Clusters are geographically proximate groups of interconnected companies and associated institutions
in a particular field, linked by commonalities and complementarities. Clusters are normally contained
within a geographic area where ease of communication, logistics, and personal interaction is possible.
Clusters are normally concentrated in regions and sometimes in a single town.
Clusters cut across traditional industry classifications. Clusters take various forms depending on
their state of development. Well-developed clusters, however, normally include end product or service
companies; suppliers of specialized inputs, components, machinery, and specialized services; financial
institutions; and firms in related industries. Clusters also often include firms in downstream or customer
industries; producers of complementary products; specialized infrastructure providers; government, univer-
sities, and other institutions providing specialized training, education, information, research, and technical
support; and standard setting agencies. Finally, many clusters include trade associations and other private
sector collective bodies that support cluster members (see Exhibit 38 below).

Exhibit 38: The California Wine Cluster

Source: California Wine Institute, Internet Search, California State Legislature. Based on research
by MBA 1997 students R. Alexander, R. Arney, N. Black, E. Frost, and A. Shivananda

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 53

 Clusters enhance competitiveness in three ways.11 First, they improve productivity because firms have
ready, efficient access to specialized suppliers, skills, information, training and technical in a demanding com-
petitive environment. Extensive market, technical, and other specialized information accumulate within
a regional cluster. Specialized inputs can be assembled, and relationships are forged among cluster partici-
pants. Firms can access trained people and technology at much lower cost than developing it internally. The
presence of a full range of knowledge, inputs, machinery, and services makes experimentation easier and
promotes greater efficiency and flexibility than vertical integration of relationships with distant suppliers.
Second, clusters foster innovation by increasing the ability to perceive opportunities for new proucts, new
processes, and for meeting new needs. The presence of a full range of local suppliers and research institu-
tions encourages knowledge creation and makes experimentation easier.
Third, clusters facilitate the commercialization of innovation by easing the creation of new firms via startups,
spin-offs and new business lines of established firms. Establishing a new business in a cluster location is
easier than elsewhere because all the needed inputs are locally available there as is cluster awareness and
expertise among capital providers such as banks and venture capitalists. The creation of new firms and
business units reinforces productivity and encourages innovation.
Clusters draw on both general and cluster-specific aspects of the business environment.
Clusters benefit from general national and regional attributes such as intellectual property laws, transporta-
tion infrastructure, and education system. However, the competitiveness of a cluster, and its uniqueness
usually owes much to the specialized circumstances of the location for the particular cluster.

Exhibit 39: CLUSTER-SPECIFIC INSTITUTIONS

Tasks of a Cluster Specific Institution for Collaboration FOR COLLABORATION

Institutions for collaboration at the cluster level

include trade associations, university incubators
and institute advisory boards, and alumni net-
works of schools and companies. Many of these
institutions have traditionally been lobbying or
social organizations, but most refocus their
roles towards enabling cluster activation and
development.
Exhibit 39 gives examples of the roles that var-
ious institutions can perform.

Source: Clusters of Innovation Initiative Regional Surveys, Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School; Interviews

11 For an expanded treatment, see Chapter 7 in Porter, On Competition.

54 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 FINDINGS

Proximity fosters productivity and innovation

Firms and industries located in a cluster enjoy commonalities and complementarities that enhance
productivity, innovative capacity, and new business formation. Capturing these synergies often relies on
“personal relationships, face-to-face communication, and close interaction among networks of individu-
als and institutions.”12 When firms are located in the same city or region, it becomes possible for them to
forge these connections. Innovation is further encouraged when members of a cluster are located in close
proximity to each other. Numerous interviewees in regions such as San Diego and Raleigh-Durham men-
tioned this advantage.
Exhibit 40: Research Triangle Park, North Carolina

Research Triangle Park encompasses 150 organizations

employing approximately 45,000 people within 7,000 acres

Source: Research Triangle Foundation

Research Triangle Leaders from the Research Triangle region’s universities, business, and govern-
ment cooperated to create what has become one of the most successful planned science parks in the world,
the Research Triangle Park. The 7,000-acre “Triangle” is defined by three universities located less than 30
miles apart: North Carolina State University in Raleigh (NCSU), the University of North Carolina at
Chapel Hill (UNC-Chapel Hill), and Duke University in Durham. These three educational institutions
formed the pillars of the region’s knowledge-based economy by providing world-class research facilities as
well as a critical mass of scientists, researchers and technicians. Their research capabilities complement
other important research institutions located inside the Park, including the North Carolina Biotechnology
Center (NCBC), the Microelectronics Center of North Carolina (MCNC), the US Environmental

12 Michael E. Porter, On Competition (Boston: Harvard Business School Press, 1996), pp. 213-214

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 55

 Protection Agency (USEPA), the National Institute of Environmental Health Sciences (NIEHS), and the
Research Triangle Institute. Over time these geographically proximate educational and research institu-
tions have been able to undertake major joint-efforts, such as the Triangle Universities Center for
Advanced Studies Inc., and the shared nuclear laboratory at Duke, research luxuries that no school could
justify on its own. Uncharacteristically close cooperation between these three geographically proximate
institutions helps the region win more federal research dollars than would otherwise be the case. But the
“if we will build it, they will come” story is only half of the reason for the Research Triangle Park’s success.
There is an equally important private sector presence that helps explain the region’s ability to compete. The
growing presence of research-oriented companies has promoted, often consciously and deliber ately,
the development of needed supporting and institutions for collaboration (see Exhibit 40 on previous page).

San Diego Jonas Salk came from Pittsburgh on a visit to San Diego, and decided the city would be
a good location for a research institute. San Diego encouraged him by zoning the Torrey Pines Mesa as
a research site. Salk built his Institute on the Mesa, and subsequently numerous research centers located
there as well, including the University of California, the Burnham Institute, and the Sidney Kimmel
Cancer Center. Interviewees comment on how the close proximity encourages the flow of information
and ideas throughout the institutions on the Mesa, a real advantage given the worsening traffic problem
in San Diego.

Clusters share common industries

Some industries are in more than one cluster. Exhibit 41 shows the 43 clusters in the U.S. economy, and
the extent to which they have overlapping clusters. These overlaps provide opportunities to use strength
in one cluster to build new (overlapping) clusters.

Cluster strength is often dispro-

Exhibit 41: Cluster Overlap in the U.S. Economy; Number portionately concentrated in a
of Clusters with Common Industries few subclusters
Clusters are typically composed of
many subclusters, and even relatively
weak clusters often have strength in
a few subclusters. Cluster develop-
ment strategies should attempt to
fill out the cluster by building off
of strong subclusters. Moreover,
strong subclusters from different
clusters (e.g., farm machinery from
the heavy machinery clusters and
motor vehicles from the automotive
cluster) may share commonalities
and complementarities. Regional
development strategies should link
Note: Two clusters are defined to overlap if they share at least 20% of their industries by number in both firms and institutions in these
directions. Numbers in brackets indicate the number of clusters overlap exists with.
Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School subclusters.

56 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Wichita Wichita is widely known as an international center of aircraft production. Locals recognize its
strong positions in the oil and gas cluster, due to Koch Industries, and the plastics cluster, due to the Coleman
Company. But Wichita also has relatively high rankings in share of national employment in many manufac -
turing related subclusters. These include construction machinery, farm machinery, mining machinery, and
equipment and parts (heavy machinery cluster); appliance, specialized pumps, and motorized vehicles (motor
driven products cluster); mobile homes, trucks and trailers, and elevators (prefabricated enclosures cluster);
and process equipment and subsystems, production machinery, and transportation equipment (production
technology cluster)—see Exhibit 42 below.

Cluster with depth and breadth normally Exhibit 42: Select Subcluster Cluster Rankings by Share
enjoy advantages over narrower clusters of National Employment, Wichita Economic Area, 1998
Clusters with strength across a broad range
of subclusters tend to have advantages over nar-
rower clusters. Extensive market, technical, and
other specialized information accumulate within
a regional cluster. Specialized inputs can be assem-
bled, and relationships are forged among cluster
participants. Firms can access trained people and
technology at much lower cost than developing
it internally. The pres ence of a full range of
knowledge, inputs, machinery, and services
makes experimentation easier and promotes
greater efficiency and flexibility than vertical inte-
gration of relationships with distant suppliers (see
Exhibit 43 below).

Exhibit 43:
The Advantages of Cluster Breadth and Depth Source: Cluster Mapping Project, Institute for Strategy and Competitiveness,
Harvard Business School

Specialized suppliers of your business’s materials,

machinery, and services are...

Source: Clusters of Innovation Initiative Regional Survey

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 57

 Boston and Pittsburgh Massachusetts adopted pro-active policies and tax incentives to attract bio-
manufacturing. This helped the Boston area biotechnology / pharmaceuticals cluster develop strength
across all subclusters. In contrast, the Pittsburgh biotechnology / pharmaceutical cluster is much younger
than Boston’s, and inevitably not as strong. Companies in the region have not yet occupied all niches along
the value-chain, but are instead somewhat stronger in research, instruments and equipment, and chemi-
cals (see Exhibits 44, 45, and 46).

Exhibit 44: Composition and Competitive Position of the

Boston Biotechnology / Pharmaceuticals Cluster, Economic Area, 1999

Source: Clusters of Innovation Initiative, Regional Survey and Interviews ; Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

58 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 45: Composition and Competitive Position of the
Pittsburgh Biotechnology / Pharmaceuticals Cluster, Economic Area, 1999

Source: Clusters of Innovation Initiative, Regional Survey

and Interviews ; Cluster Mapping Project, Institute for
Strategy and Competitiveness, Harvard Business School

Exhibit 46: Pharmaceuticals / Biotechnology Cluster Establishments and Patents

of the 15 Largest Economic Areas, Narrow Cluster Definition

Source: Clusters of Innovation Initiative, Regional Survey; Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School; Interviews

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 59

 Cluster-specific institutions for collaboration facilitate the flow of information and resources
throughout the cluster
One type of institution for collaboration that is often missing in a region are cluster specific institu -
tions. The task of such an institution is to link all members within a given cluster to all assets of use to
that cluster. Diverse groups (e.g., rival firms, related and supporting industries, universities and research
centers, training institutions, government, and so forth) contribute to cluster strength, but their contri-
bution is not automatic. An organization dedicated to mobilizing these groups does much to strengthen
a cluster (see Exhibits 47 and 48).

Pittsburgh: The Commonwealth of Pennsylvania provided funding for the Pittsburgh Digital
Greenhouse, an economic development initiative started in June of 1999 to build an industry cluster around
the application of System On Chip (SOC) technology in the digital multimedia and digital networking
markets. With the support of Pennsylvania’s universities, private foundations, regional development
organizations, state and local government, and private industry the Greenhouse is focusing on economic
expansion and job creation in the 21st century. The Greenhouse is creating jobs by attracting new compa-
nies to the region, helping local members grow, and fostering start-ups. Its private sector participants
include Benchmark Photinics, Bridge Semiconductor, Cadence, Casio, CDT (Cable Design
Technologies), Cisco Systems, Compunetix, inc., IMD (Intelligent Micro Designs), and Laural Networks.
Each is a leader in designing, developing, or utilizing SOC technology in digital multimedia or digital net-
working applications. These companies direct Greenhouse research, provide input to its education and
training programs, and collaborate with other Greenhouse members and partners.

Exhibit 47: Cluster-Specific Institutions for Collaboration Facilitate Networking

Source: Clusters of Innovation Initiative Regional Survey

60 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Massachusetts: The Massachusetts Biotechnology Council (MBC), formed in 1985 out of a softball
league consisting of teams from local biotechnology firms, serves as an important source of cluster collabo -
ration in the region. The MBC provides a forum for biotechnology companies to discuss common concerns
and potential solutions, and links the industry to related and supporting industries (e.g. lab equipment com-
panies), pharmaceutical companies, government agencies, and academic training centers.
The four major activities of the MBC are: (1) providing networking opportunities for various constituents
of the cluster (e.g. organizing large conferences), (2) negotiating purchasing agreements with scientific sup-
pliers for MBC members, (3) educating the public and encouraging workforce training programs, and
(4) lobbying government. The MBC’s board consists of representatives from both biotechnology industries
such as Genzyme, Biogen, and Millennium, as well as others such as the MIT Office of Technology
Transfer and Massachusetts General Hospital. This mix of representation ensures that cluster, not just
industry, issues are addressed.

Exhibit 48: Representative Quotes from Interviews, Cluster-Specific Institutions for Collaboration

Source: Clusters of Innovation Initiative Interviews

IMPLICATIONS
• Harnessing proximity: Zoning and industrial parks, and other tools can consciously encourage
proximity of firms in clusters. The city of San Diego zoned the Torrey Pines Mesa for research, and
donated land to the Salk Institute, and these efforts contributed materially to the development of the
medical devices and biotech/pharmaceuticals clusters.
• Building on subcluster strengths: In existing and emerging clusters there are often strong sub-
clusters. Identifying them, and focusing development strategies on them enable regions to build-off
their strength.
• Cluster overlap: Some industries are in more than one cluster. These overlaps provide opportu-
nities to use strength in one cluster to build new (overlapping) clusters. In Wichita, for example,
there is an opportunity to combine assets across several clusters to build Wichita’s position as a
center of advanced manufacturing.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 61

 7 THE DEVELOPMENT OF CLUSTERS

The development of a cluster is inevitably a long process stretching over a decade or more. A good num-
ber of existing clusters trace their roots back for a century. These roots can often be traced to strengths of
the business environment that are present in a location due to historical circumstances. One prominent
motivation for the formation of early companies is the availability of pools of factors, such as specialized skills,
university research expertise, an efficient physical location, or particularly good or appropriate infrastructure.
Many Massachusetts clusters, for example, had their beginnings in research done at MIT or Harvard.
Clusters may also arise from unusual, sophisticated, or stringent local demand. Prior existence of sup-
plier industries, related industries, or entire related clusters provides another seed for new clusters. The
golf equipment cluster near San Diego, California, for example, has its roots in the southern California
aerospace cluster. This cluster created a pool of available suppliers for castings and advanced materials, and
engineers with experience in these technologies.
New clusters may also arise out of the formation of one or two innovative companies that stimulate the
formation and growth of many others. Medtronic played this role in helping to create the Minneapolis
medical device cluster. Similarly, MCI and AOL have triggered the growth of the telecommunications
cluster in the Washington, D.C., metropolitan area.
Chance events are often important to the birth of a cluster. The early formation of companies in a loca-
tion often reflects acts of entrepreneurship not completely explainable by preexisting local circumstances.
Such companies, in other words, could have sprouted at any one of a number of comparable locations. The
role of chance, however, is often less then it seems. What looks like chance may be as much the result of
preexisting local circumstances. Moreover, even when chance provides a central explanation for a devel-
opment, it is almost never the sole explanation. Location not only raises the odds that chance events will
occur, but also the odds that chance events will lead to competitive firms and industries. Chance alone
rarely explains why a cluster takes root or its subsequent growth and development.

CLUSTER GROWTH
While the birth of clusters has many causes, the development or lack of development of clusters is more
predictable. Though there is no guarantee that a cluster will develop, once the process gets started it is like
a chain reaction in which the lines of causality quickly become blurred. The process depends heavily on
the efficacy of the diamond’s arrows or feedback loops, on how well, for example, local educational, reg -
ulatory, and other institutions respond to the cluster’s needs or how rapidly capable suppliers respond to

62 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

the cluster opportunity. Three particular areas deserve special attention: intensity of local competition, the
location’s overall environment for new business formation, and the efficacy of formal and informal mech-
anisms for bringing cluster participants together. Healthy rivalry is an essential driver of rapid improvement
and entrepreneurship. The entrepreneurial climate is important because the creation of new firms and
institutions is so integral to cluster development. Finally, organizational and relationship building mecha-
nisms are necessary because cluster advantages rely heavily on linkages and connections among individuals
and groups.
In a healthy cluster, the initial critical mass of firms triggers a selfreinforcing process in which special -
ized suppliers emerge; information accumulates; local institutions develop specialized training, research,
infrastructure and appropriate regulations; and cluster visibility and prestige grows. Perceiving a market
opportunity and facing falling entry barriers, entrepreneurs create new companies. Spin-offs from existing
companies develop, and new suppliers emerge. Recognition of the cluster’s existence constitutes a mile -
stone. As more institutions and firms recognize the cluster’s importance, a growing number of specialized
products and services become available and specialized expertise responsive to the cluster arises among
local financial services providers, construction firms, and the like. Informal and formal organizations and
modes of communication involving cluster participants develop. As the cluster grows, it develops greater
influence not only over what other firms do but also over public and private institutions and government
policies. Policies that have deferred cluster upgrading are often modified.
Cluster development often becomes particularly vibrant at the intersection of clusters. Here, insights,
skills, and technologies from different fields merge, sparking new businesses. The presence of multiple
intersecting clusters further lowers barriers to entry, because potential entrants and spin-offs come from
several directions. Diversity of learning stimulates innovation.

CLUSTER DECLINE
Clusters can maintain vibrancy as competitive locations for centuries, and most successful clusters
prosper at least for decades. Just as the development of a cluster is not assured, however, neither is its con-
tinued ability to compete.
The causes of cluster atrophy and decline can also usually be found in the business environment. They
can be grouped into two broad categories: Those deriving from the location itself, and those arising from
developments and discontinuities in the external environment. Internal sources of decline stem from inter-
nal rigidities that diminish productivity and innovation. The onset of restrictive union rules or regulatory
inflexibility can slow down productivity improvement. Overconsolidation, mutual understandings, cartels,
or other barriers to competition can undermine local rivalry. Institutions such as schools and universities
can suffer from their own rigidities and fail to upgrade and change. Such rigidities in clusters tend to arise
in locations in which government is prone to suspend or intervene in competition. When internal rigidities
arise, the rate of improvement and innovation in a cluster falters. Increases in the cost of doing business
begin to outrun the ability to upgrade.
External threats to cluster success arise in several areas. Technological discontinuities are perhaps the
most significant, because they can neutralize many cluster advantages simultaneously. Market information,
employee skills, scientific and technical expertise, and supplier bases may be rendered inappropriate. Unless
the requisite new technologies and skills are available from other local institutions or can be rap idly

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 63

 developed, competitive advantage will shift to another location. The shift of golf equipment manufactur-
ing from New England to California provides a good example. The New England cluster was based on
steel shafts, steel headed irons, and drivers with heads made of woods. When the notion of making golf
clubs with advanced materials was pioneered, east coast producers had difficulty competing. Some east
coast firms joined the California cluster; others died or declined.
A shift in buyer needs, creating a divergence between local needs and needs elsewhere, constitutes
another external threat to cluster productivity and innovation. Firms in a variety of U.S. clusters, for exam-
ple, suffered when energy efficiency grew in importance in most parts of the world while the United States
maintained low energy prices, retarding innovation. As this example illustrates, however, the threat posed
by external developments often relates to local choices and policies.

CLUSTER UPGRADING
In some clusters, the process of cluster upgrading leads to reductions in employment or revenue. This
should not be confused with cluster decline. Economic success in a region usually leads to rising local
wages and can also increase the costs of doing business. A natural part of successful economic development
is the migration of less skilled and less productive activities to other locations. Over time, some clusters
in a region will become more specialized, and shift to more advanced segments. The textile cluster in
Massachusetts, for example, has evolved towards segments requiring advanced technology such as special-
ized fabrics and fibers. Such a process of cluster upgrading should lead to higher wages even if total
employment declines. There is no reason for any region to abandon any traditional cluster.
Cluster development can be enhanced by conscious private and public action. Efforts to upgrade the
overall regional business environment must occur in parallel. Without a competitive general business envi-
ronment, regional clusters are disadvantaged.
Initiatives to develop an existing or nascent cluster must begin with cluster awareness. Companies and
local institutions must recognize that a cluster exists, and see themselves as part of it. Second, the need for
and benefits of cluster upgrading must be communicated. This often requires a preliminary study to identify
common problems and opportunities and highlight areas for joint efforts. Third, a cluster action agenda
needs to be developed by a working group consisting of cluster companies, suppliers, financial institutions,
government, and educational and research institutions. Fourth, a more permanent institutional structure needs
to be created to carry over the agenda and monitor progress.

FINDINGS

Clusters can be strengthened by increasing awareness of the cluster among local firms
and organizations
Though clusters take decades to develop, this process can be facilitated provided firms and organizations
in the cluster are aware of, and contribute to, the cluster. Not only must firms be aware of the presence
of a local cluster, they must also get together and coordinate activities to improve the cluster’s business
environment. Acceptance of new companies is important if the cluster is to grow quickly and reach a crit -
ical mass (see Exhibit 49 on following page).

64 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Exhibit 49: Cluster Awareness, Select Survey Results

Source: Clusters of Innovation Initiative Regional Surveys

New firm and cluster opportunities arise at the intersection of existing clusters
In thinking about how to diversify a regional economy, it is important to recognize that new clusters
often grow out of the intersection of existing clusters. In San Diego, for example, one of the fastest grow -
ing clusters in the 1990s was sporting and leather goods, in large part due to the success of local golf club
manufacturers (e.g., Callaway). This cluster grew out of the preexisting hospitality and tourism cluster and
the aerospace vehicles and defense cluster. These golf club manufacturers pioneered the use of lightweight
materials originally used in defense applications.

Wichita In Wichita, there are numerous heavy manufacturing subclusters with relatively high rank -
ings in share of national employment: aerospace vehicles and defense (aircraft and parts industries), heavy
machinery (construction machinery and farm machinery industries), motor driven products (refrigeration
and heating in Wichita industries), power generation, and chemical products clusters. Moreover, metal
manufacturing (broadly distributed over the clusters) and production tech (broadly distributed across the
cluster) are also growing fairly well. There is an opportunity to leverage assets across clusters to build
Wichita into a center of advanced manufacturing (see Exhibit 50 on following page).
These clusters all deal in complex equipment and complex assembly, and as a consequence, they have
similar needs. These include skills of workers, process technology, information needed, materials expert-
ise, and service providers. Companies engaged in heavy manufacturing need to recognize that they are not
only part of a cluster, but also part of a larger heavy manufacturing base. In concert with government, uni-
versities, and institutions for collaboration, these companies need to identify common needs and work
together to strengthen them.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 65

 Exhibit 50: Opportunities at the Intersection of Wichita’s Manufacturing Clusters and Subclusters

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

Research Triangle In Research Triangle as well, there appears to be a number of under-exploited,

crosscutting cluster opportunities. Examples include biotechnology and information technology,
telecommunications and medicine, or biotechnology and agribusiness. Substantial efforts at developing more
collaborative relationships between clusters should also be explored. In the pharmaceutical / biotechnology
cluster, more emphasis should be placed in a couple of areas in particular: bioinformatics, medical software,
universities, analytical instruments, chemicals, and plastics.

Anchor companies play a disproportionate role in seeding cluster development

Anchor companies are firms (e.g., QUALCOMM in San Diego, Boeing in Wichita, and IBM in
Raleigh-Durham) that play a large role in building clusters by performing several functions. First, they act
as magnets for other major companies—both rival and allied companies—that will move to a region simply
to be near the anchor company. Second, anchor companies tend to organize other companies in a cluster
for collective action, such as lobbying government for infrastructure improvements, or collaborating with
universities to build specialized research and training facilities. Third, anchor companies, and in particular
major companies with headquarters in a region, tend to be more involved in supporting community
projects that improve the local quality of life (see Exhibit 51 on following page).
Finally, a particularly important function of anchor companies is that they produce numerous spin-out
companies, which strengthen key elements of the cluster. The anchor firm is generally the buyer of the

66 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

goods and services of its spin-outs, and local demand in the cluster is improved. By producing numerous
spin-outs, the anchor increases the presence of local related and supporting industries. In some cases, spin-
outs directly compete with the parent anchor company, and this results in a more vibrant context for firm
strategy and rivalry. Finally, due to the personal “alumni” relationships among individuals in the anchor and
the spin-outs, a strong web of informal networks tends to form, and cluster collaboration improves.

Exhibit 51: Representative Quotes from Interviews, Anchor Firms

Source: Clusters of Innovation Initiative Regional Interviews

San Diego Anchored by QUALCOMM, San Diego has become a world leading center in wireless
telephony. In 1989, QUALCOMM, a Linkabit spin-off led by Irvin Jacobs and Viterbi, developed an
important new technology for cellular communications—code division multiple access (CDMA). The
new technology led to a highly successful IPO. Qualcomm put San Diego on the international communi-
cations map, motivating other regional entrepreneurs and attracting capital from outside the region. In
recent years, major international companies such as Ericsson and Motorola have set up research and devel-
opment operations in the region, and scores of start-up firms have emerged to exploit new developments
in wireless technology (see Exhibit 52 on following page).
San Diego is now the national center for wireless communications. Nokia has made San Diego its
center for CDMA research and plans to double its size to 1200 in the county by 2002. Ericcson and
Motorola both recently entered San Diego, and both have plans to hire more engineers and focus
research on wireless technology. The region is now well established as having a major national communi-
cations cluster with a particularly strong presence in wireless and internet communications technology.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 67

 Exhibit 52: Spin-outs in the San Diego Communications Cluster

Source: CONNECT, University of California — San Diego (originally created by Martha Dennis, Linkabit).

Wichita Wichita’s key clusters are composed of a few large and powerful anchor firms. In the aerospace
vehicles and defense cluster, for example, world-class companies such as Boeing, Cessna, Bombardier, and
Raytheon Aircraft have been the cluster’s engines and employ the vast majority of aerospace-related workers
in the region. Today, Boeing Wichita specializes in subassemblies, producing 75% of the 737 and major com-
ponents for the 747, 757, 767, and 777 jet aircraft and a number of significant military aircraft. Boeing
Wichita’s impact on the Wichita economy is dramatic. Boeing Wichita accounted for 20% of earnings gener-
ated in the Wichita MSA and 21% of employment in the Wichita MSA in 1998. Boeing Wichita employs
approximately 16,800 workers.
The presence of Boeing and other major aerospace manufacturers in the region has led to an increase of
parts suppliers and sophisticated regional demand for the cluster. Wichita’s suppliers have expertise in
machining, tooling, and metal work. Out of the 120 firms listed under the aircraft parts and equipment
industry category in the Wichita Area Chamber of Commerce’s Directory of Major Employers, 1999/2000,
71% were machining/metal work/tool shops. Similarly, anchor companies spin off other companies that
tend to feed the parent. For example, former workers from the aerospace vehicles and defense manufac-
turers have started machine shops that now supply these firms. The skills sets dominant within the anchor
firms largely determine the character of the spin-off and greatly influence the make-up of the cluster.

68 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Research Triangle Anchored by IBM, Nortel Networks, and Cisco Systems, the Research Triangle
has become a world-leading center in telecommunications. In recent years, major international companies
such as Cisco have set up research and development operations in the region, and scores of start-up firms
have emerged to exploit new developments in wireless technology. Electronics and communication expert-
ise grew up in the region as major multinationals like IBM began to establish manufacturing operations in
the Research Triangle focusing on information technology. IBM has 37 different IBM organizations work-
ing at Research Triangle Park (RTP). Full-time employment here ranged from 10,000 to 11,000 during
IBM’s dark days of 1991 to 1994, but spiked to 13,000 in 1995 as operations shifted from Boca Raton, Fla.
Production employees now account for only 7% of IBM’s full-time staff. But 45% of those workers are
software or hardware engineers. Sixty-two percent of IBM’s local employees were not working for the
company here three years ago. IBM’s RTP facility is now one of its biggest in the world.

Institutions for collaboration can significantly increase the

success rate of start-up companies
There are different types of institutions for collaboration that provide start-up support. Some are formal
and informal organizations and groups that facilitate the exchange of information and technology and fos -
ter various kinds of coordination and collaboration that can improve the business environment in a cluster
or in the overall economy.
The institutions for collaboration that provide start-up support for a specific cluster can prove to be
highly effective mechanisms that create and amplify the arrows and feedback loops in the diamond. The
quality of these institutions has a significant influence on competitiveness. Examples of these types of
organizations include Digital Greenhouse in Pittsburgh and UCSD CONNECT in San Diego.
A second type of start-up-centric institution for collaboration involves one that is geared to all types of
clusters. These types of institutions for collaboration do not distinguish among different types of clusters,
but rather attempt to provide generalized start-up support applicable to many divergent types of industries.
Examples of this type of star-up support include the Council for Entrepreneurial Development (CED) in
the Research Triangle.

San Diego In response to private sector efforts to better integrate with the University of California at
San Diego, University President Richard Atkinson asked Mary Walshok, Dean of Extended Studies and
Public Programs, to develop a program that would facilitate university-business interaction. After coordi -
nating with both university researchers and private sector managers, UCSD CONNECT was established
in 1985. Initial programs included Meet the Entrepreneur and Meet the Researcher Events, which helped
entrepreneurs and scientific researchers gain a better understanding of the issues that each faced. This type
of university-industry institution for collaboration helped connect the two groups and generated substan-
tial interest by hosting several events that attracted hundreds of participants from start-ups, established
companies, and academia.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 69

 Research Triangle The Council for Entrepreneurial Development (CED) was founded in 1984 to
stimulate the creation and growth of high impact companies in the greater Research Triangle area. CED
achieves its mission by providing programs and services in four major areas: education, capital formation,
mentoring, and communications. CED is a private, non-profit organization with more than 4000 active
members representing 1180 companies and is the largest entrepreneurial support organization in the US.
CED is an example of an organization that is instrumental in helping local entrepreneurs, investors, serv -
ice professionals, academics, researchers and public policy makers combine their efforts.

IMPLICATIONS
• An explicit cluster development program: Although chance events play a role in the formation
and development of clusters, conscious efforts to raise cluster competitiveness and innovative
capacity can meaningfully influence the trajectory of cluster development. Useful activities include
forming and supporting cluster-specific institutions for collaboration, creating effective links with
cluster relevant centers of excellence in regional universities, recruiting companies to fill out missing
niches in the cluster, and the like.
• Recruiting for clusters: Recruitment strategies at the regional level should target clusters in which
the region has strength, or clusters which overlap with other clusters. This allows the region to
market its unique assets rather than compete on subsidies. In recruiting efforts, regions should also
identify gaps within clusters, and seek to attract companies to fill them.
• Opportunities at the intersection of clusters: Opportunities for growth often arise at the inter-
section of clusters where a region has strengths. In San Diego, for example, one of the fastest grow ing
clusters in the 1990s was sporting and leather goods, in large part due to the success of local golf club
manufacturers (e.g., Callaway). This cluster grew out of the preexisting hospitality and tourism
cluster and the aerospace vehicles and defense cluster. These golf club manufacturers pioneered the
use of lightweight materials originally used in defense applications.

70 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 CREATING AND IMPLEMENTING
A REGIONAL STRATEGY
8
The development of regional economies can and is affected by purposeful action. Economic develop-
ment is a process which can proceed in different directions and can achieve different degrees of success
depending on decisions by many different actors. Government at all levels, the private sector, and many
other institutions have a role in economic development. If their efforts can be encouraged and aligned,
regional development is more rapid.
Regions must develop a shared understanding of the sources of competitiveness and the role
of clusters. Productivity and innovation—not low wages, low taxes, or a devalued currency—are the def-
inition of competitiveness. Participants need to understand the influences on productivity and the role and
importance of clusters in productivity enhancement. Early and ongoing communication and discussion
educate participants about competitiveness and help to shift mindsets.
An overall strategy for regional development is needed. The regional economic strategy will
include both efforts that improve that general business environment in the region, and an agenda around
specific clusters. The strategy also needs to embraces all clusters in a region. Setting priorities is not only
bad economics; it disenfranchises large parts of the private sector. Successful regional economic strategies
include traditional clusters, such as agriculture and tourism, and even declining clusters. They include
emerging clusters as well as established ones. To avoid misguided attempts at creating clusters that have no
assets on which to build, emerging clusters should have a demonstrable local foundation and a base of
firms that have met a market test.
The private sector must take a leading role in regional economic development, not leave it to
government. Private-sector leadership is often critical for success. Active government participation in
a privately led effort, rather than an initiative controlled by government, will have a better chance of success.
Companies can usually better identify the obstacles and constraints in their path (as well as the oppor tuni-
ties) than can government. Letting the private sector lead also reduces the initiative’s political con-
tent, while taking advantage of the private sector’s often superior implementation ability.
Regions need an institutional structure for economic development that encompasses all the
constituencies. The institutional structure needs to address both cross-cutting and cluster specific issues.
Keeping them separate elements within an integrated structure improves the specificity of the proposals
developed. Regional economic strategy efforts should be as nonpartisan as possible and remain independ-
ent of any party or administration’s political agenda. Legislators and the executive branch, the opposition
parties, and those in power must all be involved. Ideally, the cluster initiative will take place through an
entity independent of government; otherwise, promising efforts may be dropped when a new government
takes office.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 71

 With these elements in place, regions can set out to create a regional strategy, build consensus around
out, and finally implement its action agenda to achieve better economic performance.

FINDINGS

Regions often encounter a common set of pitfalls

Because regions are so unique in their composition, and varied in their performance, no single nation-
wide policy will be entirely appropriate for every region. The decisions that really affect economic and
innovation performance are made in regions, not Washington D.C. Many regions have formulated and exe-
cuted very good development strategies, and we have been fortunate enough to study and learn from some
of them. But, in more than ten years of work in this field, we have also seen numerous poor strategies.
What is more, all participants in a regional economic development strategy contribute to this list of
common pitfalls.
Firms are often not proactive enough in communicating their needs to other organizations that
could be of help. They also often weaken their business environment by pursuing strategies that drive
clusters members out of business and discourage rivals from locating nearby.
In addition to traditional issues of excessive red-tape and taxation, governments hurt economic
development by implementing regulations that discourage investments in local companies, and by
neglecting to invest in research and training institutions.
Economic development organizations frequently focus on an overly narrow geographic area, place too
much emphasis on building “high-tech” clusters, ignore traditional strengths of the “old-economy,” and
target big companies for recruitment. Finally, common pitfalls for research and training centers are an
under emphasis on the commercialization of knowledge, and too little cross-disciplinary collaboration
(see Exhibit 53 below).

Exhibit 53: Common Pitfalls

Source: Cluster Mapping Project, Institute for Strategy and Competitiveness, Harvard Business School

72 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Regions need to overcome transition points in the development of their economies
Regional leaders tend to encounter several sets of challenges as they develop their economies: under-
standing the new paradigm, creating needed assets, creating the capacity for action, overcoming bottlenecks
and seizing opportunities, and revising the shared vision. These challenges become the targets of regional
economic development strategies. The first challenge is informing leaders in business, government, univer-
sity, and economic development organizations of the new understandings of economic development.
Competitiveness is not preordained by a region’s static assets. Rather, regions can choose to compete by
leveraging differentiating assets, investing in specialized training and research, and fostering a business
environment that supports innovation. This understanding of competitiveness will enable leaders to iden-
tify new assets that will be needed to develop the regional economy and build strong clusters. The third
step is then creating leadership and institutional capabilities to fully leverage existing assets and create new
assets as needed.
Once regional leaders understand the new model of regional development, have developed appropri-
ate assets, and create the capacity to use their assets, the economy will grow. At this point, the main
barrier to prosperity will be the bottlenecks that develop as a consequence for growth. Finally, the orig -
inal vision will run its course, and leaders will need to develop a new vision in order to mobilize
companies, government agencies, universities and research centers, and institutions for collaboration
behind a common strategic agenda.
The regions we studied faced a range of different transitions. Leaders in Wichita were absorbing new
understandings and communicating it to others in the region. In the process, they were identifying vari-
ous assets that would have to be created (e.g., specialized training programs), strengthened (e.g., local
research centers), and leveraged (e.g., diversity of heavy manufacturing clusters).
Pittsburgh had many important assets already in place, and the leaders and community understood
regional economic development. The challenge was clearly to create the capacity for firms, universities,
institutions for collaboration, and government to utilize these assets to the fullest extent possible. The San
Diego economy was performing well, and using its assets, but rapid growth created bottlenecks (e.g., over-
use of physical infrastructure, rising cost of living), and valuable opportunities were not being seized. New
strategies were needed to reach full potential. The Research Triangle’s original strategy of focusing on
assets in and around the Park has been very successful, but is also reaching its limits. In addition to deal-
ing with bottlenecks around commercialization, leaders need to rally a wider community around a new
vision, and create institutions for maintaining collaboration. The growth of Atlanta has been remarkable
over the last decade, but this is creating challenges that their leader-centric model of the past cannot meet.
Many groups will have to work together to build regularized institutions capable of solving ongoing prob-
lems of traffic, education, and public-private collaboration. A revised vision emphasizing prosperity, not
growth, will be helpful in accomplishing this task.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 73

 Exhibit 54: Creating the Capacity to Act Broad-based collaboration is needed for
development strategies to succeed
Successful regional economies benefit from
the contributions of many organizations:
companies generate innovations, jobs and
wages; universities produce research and
trained graduates; governments provide basic
education, support workforce training, and
build and maintain the physical infrastructure;
institutions for collaboration provide information,
lobby on behalf of business, help achieve consensus
firms, and create networking opportunities. These con-
tributions, however, are not automatic. One reason is that
companies and organizations pursuing their own parochial
interests may act contrary to the needs of a regional economy.
Another is that no one individual or organization necessarily
exists which has a mandate to lead a regional economic develop-
ment agenda. Finally, the multitude of organizations that need to con-
tribute to regional development may not collaborate well. Organizing for action entails overcoming these
three obstacles and creating the capacity for regions to implement development strategies (see Exhibit 54).

San Diego Collaboration among diverse groups has been critical to San Diego’s economic develop-
ment. In 1908, business and government leaders lobbied President Theodore Roosevelt to build a naval
headquarters in San Diego, the city dredged the bay for military ships, and Roosevelt complied. Later U.S.
Representatives from the region lobbied for military installations and succeeded in bringing a Navy
Training Center, Camp Pendleton (U.S. Marine Corps), and ultimately the Naval Air Station on North
Island, which was attracted to the area by the near ideal conditions for year-round flying. Once again, the
city “accommodated” by making the land for the airfield available. Richard Atkinson, the current president
of the University of California, stated that “one of the most important events” in the economic his tory of
San Diego was the “campaign of Science Applications International Corporation (SAIC) and General
Atomics to attract a University of California campus” to the area. Defense and research companies also
sponsored the extension of the university, once established. For example, General Atomics “sold” the uni-
versity on the International Thermonuclear Experimental Reactor (ITER) and “the university, in turn,
sold the mayor and the governor.” In another example, General Atomics catalyzed the creation of the
Super Computer Center at UCSD. On the other hand, when groups failed to cooperate they lost nation -
al competitions to host computer and superconductor research centers.

74 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Wichita Wichita’s mature aerospace vehicles and defense cluster is the result of 80 years of purposeful
planning, investing, and coordination from the large manufacturers and the local, state, and federal gov-
ernment. Flat land, good winds, and excellent year-round flying weather were initially important for the
cluster, but did little to distinguish Wichita from many Midwestern cities. The presence of oil was also
helpful because local oil magnates provided the local capital used to finance the initial aircraft entrepreneurs.
City leaders took advantage of these circumstances by building an airfield well before anyone real ized how
important commercial air travel would become and, later, by attracting Cessna and Lear. More recently,
local leaders have orchestrated a well-coordinated push for an expensive K-12 bond vote. Recognizing that
the labor pool was one of the region’s most important but undeveloped assets, the private sector worked
closely with the city of Wichita and the Chamber of Commerce to convince citizens that new investment
was needed for the schools. Likewise, the region’s economic development leaders are stepping up to cre-
ate a new workforce development program in concert with the aircraft manufacturers although this might
not yield results for years to come.

A shared economic vision helps elicit broad support and coordinate activities
Numerous companies, government agencies, knowledge centers, and collaborative institutions must
coordinate with each other in order to fully contribute to economic development. They must communi-
cate their respective needs to each other, prioritize collectively, commit resources to solving these needs,
and work together to streamline the delivery of resources. Yet, these organizations do not recognize one
master; there is no CEO of regional economic development.
In order to achieve good coordination among many diverse groups, a shared vision of common objec-
tives and methods is vital. This does not mean all participants agree on a detailed master plan. It does mean
that influential individuals and groups broadly agree on a basic issue such as:
• The main challenges and opportunities facing a region;
• That a broad array of groups (e.g., companies, clusters, knowledge centers, government agencies,
and institutions for collaborative) be included in solving these issues;
• That new and valuable ideas come from interaction with different sectors of an economy;
• That an organization’s (e.g., companies, universities, etc.) long-term interests are tied to the health
of the local economy; and
• That collaboration among diverse groups will be needed to compete in today’s economy.

Wichita Our suggested New Directions for Wichita offer the outlines of a potential shared vision. The
main challenges for Wichita include energizing the region behind a more offensive outlook that recognizes
its differential advantages, and shifting focus to increasing innovation output. To do this, a broader array
of groups in the community must participate in economic development. In particular, university-based
research centers need to be strengthened, and firms in a variety of clusters must increase their level of
interaction. Finally, to gain collaboration among these diverse, sometimes contentious groups, everyone
must take a long-term perspective, and recognize that working with rivals and contributing to regional and
cluster initiatives will build a better business environment, and in turn benefit the individual contributors
(see Exhibit 55 on following page).

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 75

 Exhibit 55: Wichita’s New Directions

Atlanta The combination of a laissez faire business environment, economic and policy entrepreneurism, per-
sonalized leadership, and state-supported economic development are critical facets of the success Atlanta has
enjoyed. These traits have seeded incredible economic growth. However, these same conditions also make solv-
ing the challenges of regional growth very difficult. The Atlanta region is not well configured to solve its present
problems. Neither the Atlanta regional culture nor its institutions support the complex, interrelated efforts nec-
essary to combat its educational, environmental, and transit issues. The region does not have effective regional
government structures to deal with cross-jurisdiction problems. Nor does it seem to generate leaders who have
the mindset to address the particulars of the solutions. To solve these issues will require a thousand little things,
not one or two big things. Atlanta is a Project Town seriously in need of Process (see Exhibit 56 below).

Exhibit 56: Atlanta’s New Directions

76 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

Strong leadership is a necessary part of any successful economic development strategy
Strong leadership committed to regional economic development is needed to ensure that companies,
knowledge centers, governments, and collaborative institutions contribute to their full potential (see
Exhibit 57 below). The reason is that the primary mission of each of these organizations is not to promote
the economic health of the region. Companies are primarily concerned about their own survival and suc -
cess. Universities focus on basic research and educating their students. Governments are interested in
a range of issues (e.g., crime, education, health care, and the economy), and political jurisdictions rarely
coincide with the boundaries of a regional economy. Institutions for collaboration often have a narrow
mandate such as promoting the region, licensing technology developed in a university, or working on
behalf of a single cluster.

Exhibit 57: Leadership from Government and Institutions for Collaboration

To be sure, all of these organizations inevitably contribute to a regional economy in the course of their
normal operations. But that contribution may not be all it could be. Local companies often weaken their
own cluster by driving out competition, and trying to monopolize information and resource flows among
input providers, suppliers, and customers. Universities may disdain commercialization of basic research,
or make the process too cumbersome to be worthwhile. Governments can overtax and poorly regulate
industry. Institutions for collaboration may fail to perform key functions like building cross-cluster net-
works and facilitating the flow of ideas and innovations from knowledge centers to industry.
In the successful regions we studied, strong leaders from the private sector, universities, government,
and institutions for collaboration made sure their organizations contributed to the regional economy (see
Exhibit 58 on the following page). They embraced the notion that “what is good for the community is
good for my organization.” These leaders made development a priority: they identified weaknesses,
leverageable assets, and models of success from other regions; they appointed individuals and created
and supported organizations to carry out desired initiatives; they benchmarked progress; they clearly
communicated to other members in their organization that regional economic development was impor-
tant and that contributions to it would be rewarded.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 77

 Exhibit 58: Leadership from the Private Sector and Research Centers in Atlanta

An overarching organization for economic development helps coordinate and

routinize the process
Building a consensus behind a basic agenda and a true commitment to the above attitudes is a difficult
and ongoing process. It must be sustained throughout changes in political leadership, and during difficult
economic times. Many regions achieve a shared vision primarily due to a crisis that galvanizes the com-
munity and spurs action. Another way to achieve it is to formalize a process for working on these issues.

78 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 Massachusetts An example from our research is the Massachusetts’s Governor’s Council on Economic
Growth and Technology. The Governor’s Council was founded in early 1991 to provide expert, objective
input on issues and policies relevant to the Massachusetts economy, to report periodically on the status of
industry in the Commonwealth and to make recommendations for policy action and assist in implementa -
tion. Composed of the state’s leading business executives, academics and government representatives, the
Council exemplifies the philosophy of chief executive involvement and broad-based collaboration in
support of private sector-led development. The Council was organized into task forces on tax policy and
capital formation, technology policy, and marketing, as well as other committees around important core,
and emerging, industry clusters (see Exhibit 59 below).

Exhibit 59: Massachusetts Governor’s Council

on Economic Growth & Technology

Source: Commonwealth of Massachusetts Governor’s Council on Economic Growth and Technology; Monitor Group

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 79

 9 ACTION AGENDAS FOR THE
PUBLIC AND PRIVATE SECTORS

Although many groups and individuals affect regional economies, the most influential organizations are
government (federal, state, and local), universities and research institutes, institutions for collaboration,
and firms. Each type of organization has an important and distinct role to play in developing regional
economies and clusters. This chapter draws on the material presented in earlier chapters to identify appro-
priate action agendas for each organization. We have distilled the core action implications for each group
and present them here in brief, rather than reiterate earlier concepts, findings, and implications.

FEDERAL GOVERNMENT
• Invest in the foundations of science and technology. (see pp. 43-46)
- Increase federal funding of research at universities and other research centers.
- Establish federal overhead recovery rules, and other policies, to encourage investment in univer -
sities’ science and technology infrastructure.
- Provide federal support for specialized training programs in science and engineering.
• Improve the innovation policy context. (see pp. 46-48)
- Fortify intellectual property law protection
- Strengthen and enforce anti-trust laws with a greater weight on innovation.
- Reinforce federal tax incentives that encourage business investment in R&D and university-
industry collaboration.
• Allocate federal resources to reinforce cluster development. (see pp. 46-48)
- Distribute federal research funding through a system of peer-reviewed competitive grants in a
way that fosters cluster development.
- Encourage locally-based federal agencies to communicate and coordinate with local business,
institutions for collaboration, and educational and research centers based around clusters.
• Provide better data for measuring regional economic composition and performance. (see pp. A1-A9)
- Collect more up-to-date data down to the county level.
- Collect measures of both economic performance and innovation.
• Encourage the development of regional economic development strategies that stress innovation.
• Provide federal matching funds for innovation-focused state and regional economic develop -
ment strategies.

80 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

STATE GOVERNMENT
• Invest in the foundations of science and technology. (see pp. 43-46)
- Recognize the state government’s important role in supporting R&D funding at state universities.
- Establish and maintain high levels of state support for community colleges and specialized
training centers.
- Create a strong university or college presence in all major regions of the state.
• Sponsor state programs that encourage cluster development. (see pp. 55-56)
- Build cluster thinking into research parks and incubators.
- Organize state systems of higher education around local clusters.
• Focus business recruitment around strong clusters. (see pp. 46-49)
- Coordinate activities with firms, universities, and training centers to recruit anchor companies to
their region.
• Create regional dimension to state economic development strategy. (see pp. 46-48, 77-79)
- Encourage and assist regions to develop economic strategies.
- Cultivate attitudes toward collaboration and sharing of information among firms, universities,
training centers, labor, institutions for collaboration, and government.
• Improve information systems to regularly collect data and measure progress. (see pp. A1-A9)

REGIONAL AND LOCAL GOVERNMENT

• Strongly support K-12 education, and create strong standards and accountability. (see p. 42)
• Upgrade core business infrastructure. (see pp. 39-42)
- Transportation infrastructure
- Communications infrastructure.
- Ensure specialized training programs are a high priority in any economic development strategy.
• Develop a regional strategy that involves all shareholders. (see pp. 77-79)
- Support regional benchmarking initiatives.
- Encourage a common vision and collaboration among firms, universities, and training centers.
- Work with firms, universities, institutions for collaboration, and state government to create an
organizational structure to help implement a regional strategy.
• Encourage cluster development. (see pp. 55-56)
- Establish research and industrial parks that encourage innovation-based competition.
- Implement cluster-focused recruitment efforts.

UNIVERSITIES AND RESEARCH INSTITUTES

• Recognize the important role of universities in regional economic development. (see pp. 43-46)
- Take the lead on, and participate in, regional and cluster development efforts.
• Create and support technology transfer offices (see pp. 15-16)
- Work with firms and venture capital to streamline the technology transfer process
- Benchmark the commercialization of university-created intellectual property using measures
that promote efficient dissemination of knowledge.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS 81

 • Actively participate in cluster development efforts. (see p. 29)
• Align university curricula and research to meet the needs of local clusters. (see p. 33)
- Create cluster-specific institutions to support collaboration between academia and industry clusters.
- Work with local industry to create areas of excellence within universities that differentiate the
university and complement local industry strengths.
- Integrate research and training efforts with the needs of local industry.
- Participate in the recruitment of companies
• Support company start-up efforts by professors and students through mentorship, entrepreneurial
education, and financing. (see pp. 69-70)

CLUSTER-SPECIFIC INSTITUTIONS FOR COLLABORATION

• Promote cluster awareness. (see p. 64)
• Engage in ongoing diagnosis of cluster’s competitive position. (see pp. 57-59)
- Compare position relative to other regional clusters.
- Identify constraints, obstacles, and advantages
• Develop training and management programs. (see pp. 60-61)
- Provide programs through the Institutions for Collaboration.
- Coordinate with local institutions to provide programs.
• Actively participate with government in recruitment efforts. (see pp. 46-49)
- Communicate with firms in clusters to identify gaps in the cluster and recruit accordingly.
• Widen institutional membership to include all cluster constituents. (see pp. 69-70)

FIRMS
• Recognize importance of location to comparative advantage.
• Take active role in improving competitive environment.
- Consistently communicate your needs and desires (e.g., for talent, ideas, patents) to local uni-
versities, research institutes, and training centers.
• See their cluster as a competitive asset.
• Contribute actively to cluster development activities. (see p. 64)
- Actively participate in cluster activities to identify issues of common concern and opportunities
for mutual gain (e.g., regulatory matters, new buyer needs, innovative supplier capabilities).
- Support recruitment activities of local chambers and other regional economic development
officials to bring in companies that will fill missing niches in the firm’s cluster (e.g., suppliers,
services providers, competitors).
- Contribute to programs that support new ventures (e.g., improving access to risk capital,
mentoring programs, and specialized services) in order to build-out cluster.

82 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

appendix 1
ASSESSING REGIONAL ECONOMIC COMPETITIVENESS AND
INNOVATIVE CAPACITY: A HOW TO GUIDE

Competition is a game of relative gains. Some regions outperform others because they have a superior
mix of assets and liabilities. Improving regional and cluster competitiveness entails identifying areas of rel-
ative strength and weakness, and building strategies that play to strengths and improve weaknesses. This
guide explains how to identify strengths and weaknesses, and how to develop strategies to exploit them.
The underlying challenge in assessing a region’s competitiveness is accurate benchmarking. How do
you compare diverse regions around the country in order to measure their relative position in terms of
economic performance, innovation output, and the capacity to innovate? Different organizations will
assess different geographical areas, define clusters differently, measure different performance criteria, draw
on different data sources, and use different base years to calculate growth rates. The result is an “apples to
oranges” comparison and an inaccurate assessment of the relative strengths and weaknesses of the regions.
We use a single analytical framework, apply a common methodology, and draw on a number of data
sources in order to make “apples to apples” comparisons of regions.

ASSESSING THE COMPETITIVENESS OF REGIONAL ECONOMIES

Economic Performance and Innovation Output

The first step in assessing a regional economy is to benchmark its economic performance and innova-
tion output. Good economic performance is the ultimate objective, and innovation output is an indicator
of future performance. It is important to benchmark these areas in order to determine basic development
objectives (e.g., raise wages, increase innovation, maintain current performance, etc).
The study examined regional economic performance on two levels. At the broadest level, we compared
the region to other regions on various indicators of economic vitality and standard of living such as
employment growth, unemployment, wages, wage growth, cost of living, and exports. To assess potential
future competitiveness, we examined measures of innovative output and entrepreneurship including
patents, establishment formation, venture capital investments, the prevalence of fast growing companies,
and initial public offerings. Wherever possible, we tracked both the level and the growth rate of each per-
formance indicator (see exhibit on next page).
We compared the performance of regional economies to the national economy as a whole, as well as to
other technologically intensive regions.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-1

 Economic Performance Indicators

Source: Bureau of Labor Statistics; Bureau of Economic Analysis; International Trade Administration; U.S. Patent and Trademark Office; Price
Waterhouse Cooper Money Tree; Hoover’s IPO Central; Inc. Magazine, American Chamber of Commerce Researchers Association

Composition of the Regional Economy

The next step is to benchmark a regional economy’s composition, because differences in composition
lead to differences in performance and innovation. More specifically, regions with a higher percentage of
workers in traded industries, and with a larger portfolio of relatively strong and growing clusters, tend to
be more prosperous. Our analysis of composition also enables us to identify those regional clusters that
create the most jobs and contribute most to the regional payroll.
Especially in advanced nations such as the United States, regional economies are specialized, with each
region strong in a different mix of industry clusters. Comparing regional economies has been difficult
because clusters have not been systematically identified or mapped across all U.S. regions. To address this
challenge, Professor Porter and his team at Harvard Business School have defined clusters statistically and
assembled detailed data by industry and cluster on employment, wages, establishments, and patenting
over time for every region in the United States. (See the boxed insert for a summary of the Cluster
Mapping Project)A1
The Cluster Mapping Project provides an objective, quantitative way to profile regional economies,
compare them over time, and measure the strength, evolution, and performance of the region’s
clusters. The cluster mapping data is used to identify the most important clusters in the region’s econo-
my, understand the drivers of the region’s relative wages, employment growth, and formation of new
establishments, assess the region’s patenting performance, and examine the region’s relative position
versus other regions overall as well as in its leading clusters.

A1 By traded, we mean that the location of the firms in these clusters is not driven by the need to be near a specific natural
resource, or by population concentration. Instead, these industries are located in a specific area for some reason related
to the region’s innovative capacity.

A-2 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

CLUSTER MAPPING PROJECT METHODOLOGY
• The purpose of the Cluster Mapping Project is to assemble a detailed picture of the location and per-
formance of industries in the United States, with a special focus on the linkages or externalities
across industries that give rise to clusters.
• The raw data for the project are County Business Patterns data (excluding agriculture and govern-
ment) on employment, establishments, and wages by four-digit Standard Industrial Classification
(SIC) Code by U.S. county. In addition, U.S. patent data by location of inventor are allocated to
industries and clusters using a concordance of technology classifications with SIC Codes.
- Confidentiality limitations mean that actual data are not disclosed for every county and economic
area in every industry. Various techniques are used to compensate for missing data.
• Economies are analyzed at various geographic levels, including states, Economic Areas, Metropolitan
Statistical Areas (MSAs), and counties.
• All the industries in the economy are separated into “traded” and “local” based on the degree of
industry locational dispersion across geographic areas. Local industries are those present in most if
not all geographic areas, are evenly distributed, and hence primarily sell locally. Traded industries
are those that are concentrated in a subset of geographic areas and sell to other regions and nations.
• Among traded industries, clusters are identified using the correlation of industry employment across
geographic areas. The principle is that industries normally located together are those that are linked
by some external economies. These industries, then, constitute a cluster.
• Clusters are defined initially using state-level data (n=50). The robustness of cluster composition
is verified using Economic Areas as the geographical unit.
• Clusters are constructed using two approaches, which are then reconciled:
- Select a prominent “core” industry in a field or part of the economy. Calculate the locational
correlations of all other industries with the core. Those industries with statistically significant
correlations with the core define the extent of the cluster.
- Calculate locational correlations between all pairs of industries in a general field and potentially
related fields. Those sets of industries with statistically significant and substantial intercorrela-
tions among each other define the cluster.
• In both cases some industries may have spurious correlations to a cluster because of the co-location
of several strong clusters in the same geographical area. Spurious correlation is eliminated using
Input-Output tables, industry definitions, and industry knowledge.A2
• Note that a given industry can be part of more than one cluster. This sometimes reflects overly broad
industry definitions. However, it is also the case that there are multiple forms of externalities, and
some industries are suppliers or customers of many other industries. Thus, overlapping clusters are
expected and their overlaps are important economically (see Exhibit 39, Cluster Overlap in the U.S.
Economy; Number of Clusters with Common Industries, Section 6, page 66).

A2
The 1992 Input-Output Accounts measure the share of economic value traded between industries.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-3

 • The process of statistically defining cluster boundaries resulted in 41 traded clusters in the U.S. econ-
omy. These are shown in the figure below, grouped into broad categories.
• Clusters can be defined using “narrow” or “broad” definitions. We use narrow cluster definition to refer to
the subset of the industries that are most correlated with a given cluster. Analysis using narrow cluster def-
initions eliminates cluster overlaps. An industry is a narrow industry for only one cluster.
• Broad cluster definition includes all industries with statistically significant locational correlations. This
includes industries with stronger locational correlations with another cluster. Analysis using broadclus-
ter definition includes the overlap among clusters. This overlap is important to understanding cluster
competitiveness, but leads to double counting of employment, which leads to difficulties of interpre-
tation for some analyses.
• Subclusters are subsets of cluster industries that are the most strongly correlated with each other
relative to the rest of the cluster. There are subsets of industries where linkages are particularly strong.
We define the subcluster statistically for each cluster. Separate subclusters are defined for narrow and
broad cluster definitions. There are 244 subclusters in the 41 traded clusters defined using narrow clus-
ter definition. An additional 245 subclusters arise among industries outside the narrow cluster that fall
into the broad cluster definition.
• We also grouped the 241 local industries into clusters primarily using industry knowledge. There are
16 local clusters ranging from local health services and local utilities to local retail clothing and acces-
sories. We did not analyze local clusters extensively in this project, but focused on cross-regional
competition. Local clusters are crucial for examining the patterns of location with metropolitan areas.

A-4 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

CLUSTERS OF TRADED INDUSTRIES

The Assessment of the Business and Innovation Environment

The third step in a regional assessment is to analyze the business environment. A healthy business envi-
ronment will lead to growth of employment in traded industries, and to strong and growing clusters, the
drivers of prosperity.
The quality of the overall business and innovation environment includes both common characteristics
that affect the entire economy and the particular circumstances in important regional clusters. We
examine overall competitiveness, but place special emphasis on the region’s innovation environment. The

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-5

 Business Environment and Cluster Indicators

figure below illustrates some of the dimensions of the overall business environment analyzed in the
research.

Data for benchmarking regional economies in terms of basic and specialized factor inputs is readily
available from secondary materials available from such sources as the Department of Education, the
Bureau of Labor Statistics, PriceWaterhouseCoopers, Hoover’s, and others. To obtain benchmarking data
on other elements of the business environment, we survey executives from business, government, aca-
demia, and economic development organizations using a common set of questions. We then follow up
the survey with in-depth interviews with senior executives from a region. The combination of a survey
with a large sample size and 40-60 interviews enables us to accurately benchmark the business environ-
ment of a region against rival regions.

A-6 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

ASSESSING THE COMPETITIVENESS OF REGIONAL CLUSTERS
The steps for assessing regional clusters are similar to those used for assessing a regional economy. We
benchmark a regional cluster to other regional clusters in the categories of economic performance and
innovation output, composition, and business environment.

Economic Performance and Innovation Output

We look at both economic performance and innovation output measures for each regional cluster, and
benchmark a regional cluster against national averages and rival clusters in other regions. See the exhibit
below for cluster specific measures. The Cluster Mapping Project provides data on employment growth,
average wages, wage growth, patents, patents per worker, and establishment formation. Secondary sources
provide data for other measures.

The Composition of Regional Clusters

An accurate breakdown of cluster composition not only helps explain the economic performance and
innovation output of the cluster, but also identifies leverageable strengths and areas of weakness. Clusters
are composed of subclusters, which are composed of industries. The Cluster Mapping Project Data ranks
both regional subclusters and regional industries to precisely identify areas of relative strength and weak-
ness. Strength across a range of subclusters is desirable, as is strength in narrow industries (i.e., the
Composition and Competitive Position of the Research Triangle Communications Cluster

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-7

 subset of the industries that are most correlated with a given cluster). See below for an example of the
breakdown of a regional cluster.

The Cluster Business and Innovation Environment

Finally, we assess a regional cluster’s business and innovation environment, and benchmark it against
other regions and clusters. A healthy environment stimulates cluster growth, helping to broaden and
deepen cluster composition. A good business environment also fosters innovation, helping to increase
the productivity and competitiveness of a regional cluster, and the firms in it. As with the regional econ-
omy, we use a combination of surveys and interviews to gather cross-cluster comparable data.

A-8 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

appendix 2 DEFINITION OF MEASUREMENTS

OUTPUT MEASURES
Measure Definition Calculation Source
Employment Number of persons Sum of employment in all counties constitut- County Business
employed per ing the Metropolitan Statistical Area (MSA) Pattern Data on 4-digit
MSA/cluster Standard Industrial
Classification (SIC)
industries per county

Wages Payroll of Total payroll dived by total employment per County Business
region/cluster per region/cluster; calculated as employment Pattern Data on 4-digit
employed in weighted average of wages per county (for SIC industries per
MSA/cluster region) or industry (for cluster) county

Productivity Value of shipment First, NAICS-based shipment data is trans- Census Bureau
per employee in formed to SIC codes using the bridging Shipment Data;
MSA/cluster methodology provided by the 1997 Economic County Business
Census. The weights of each NAICS code Pattern Data on 4-digit
assigned to a SIC industry are based on the pro- SIC industries per
portions of total sales/ receipts/shipments each county
NAICS accounts for that SIC code. However,
this transformation does not generate data for all
industries defined in the SIC code. Also, some
data is suppressed to avoid disclosing individual
company data.

Second, the value of total industry shipments is

divided by total industry employment.

Of the 720 SIC industries listed, information is

available for 110 industries.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-9

 OUTPUT MEASURES (Continued)
Measure Definition Calculation Source

Exports Value of manufactur- Direct use of data. U.S. Department of

ing and non- Commerce’s
manufacturing International Trade
commodity exports Administration data on
per industry and MSA the two-digit SIC level

INNOVATION MEASURES
Measure Definition Calculation Source
Patents Number of patents Direct use of data for MSAs. Commerce Department
registered per data on patents per
For clusters, we need to distribute the aggre-
MSA/cluster MSA
gate number of regional patents to individual
industries.

Venture Capital Value of Venture Direct use of data PriceWaterhouse-

Investments Capital Investment Cooper’s Money Tree
per MSA/cluster Database

Fast Growth Number of compa- Direct use of data Inc. Magazine Top 500
Firms nies on Inc. 500 list list of high-growth
Inc. Magazine lists companies by sales growth.
and/or Gazelle-type companies
company per MSA “Gazelle”-firms are defined by employment
Cognetics “Gazelle”
growth above 100% over four years
companies’ list

Initial Public Number of IPOs Direct use of data Hoover’s IPO

Offerings per MSA Central.com

A-10 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

COMMON BUSINESS ENVIRONMENT MEASURES
Measure Definition Calculation Source
Basic research Federal funds for Direct use of data National Science
research universities Foundation
per MSA WebCASPAR Database
System

Skills of Number of Direct use of data: US Bureau of Labor

workforce employees per skill Statistics, Occupational
Number of scientists / engineers, technicians
and MSA Employment Statistics
in scientific and engineering fields, managers
and professionals, and science and technology
graduates in the regional workforce

Education Expenditure and Direct use of data: California Department

performance per of Education, National
High school graduation rates, student/teach
student and MSA Center for Education
ratios, average expenditures per student, and
Statistics
SAT scores

Physical Transportation Direct use of data Texas Transportation

infrastructure System, Institute Annual
Communications Mobility Report,
System, Utilities Clusters of Innovation
Initiative Regional
Survey Data, Secondary
Sources

Supply of Risk Size of local venture Direct use of data: Alternative Assets
Capital capital industry
Number of local venture capital firms, and
total funds management by local venture
capital firms

Quality of Life Direct use of data: Clusters of Innovation

Initiative Regional
Cost of housing, and level of traffic
Survey Data, Secondary
congestion
Sources

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-11

 CLUSTER-SPECIFIC BUSINESS ENVIRONMENT MEASURES
Measure Calculation Source
Specialized Direct use of average questionnaire response: Clusters of Innovation
research centers Initiative Regional Survey,
How available are local research centers to use by
and interviews
private firms, and how frequently do they transfer
technology and knowledge to the private sector?

Specialized talent Direct use of average questionnaire response: Clusters of Innovation

base Initiative Regional Survey,
Is there a sufficient number of qualified scientists,
and interviews
researchers, technicians, and business managers to
sustain and grow companies in the region?

Specialized training Direct use of average questionnaire response: Clusters of Innovation

Initiative Regional Survey,
Do local institutions supply a sufficient number of
and interviews
qualified scientists, researchers, technicians, and
business managers, and will this improve or worsen
in the future?

Sophistication of Direct use of average questionnaire response: Clusters of Innovation

demand Initiative Regional Survey,
Are local customers sophisticated in their demand
and interviews
for new and better products, and do companies
receive regular feedback from these customers?

Intensity of rivalry Direct use of average questionnaire response: Clusters of Innovation

Initiative Regional Survey,
How many local rivals are there in your cluster, and
and interviews
would you characterize competition as more intense
or more mild?

Degree of Direct use of average questionnaire response: Clusters of Innovation

cooperation Initiative Regional Survey,
Do firms share knowledge with each other, and do
and interviews
they consistently contribute to cluster-wide projects
and initiatives?

Related and Direct use of average questionnaire response: Clusters of Innovation

supporting Initiative Regional Survey,
What is the quality of local suppliers and supporting
and interviews
industries, how frequently do firms source from out-
side the region, and how much feedback to related
industries give on improving products and processes?

A-12 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

 appendix 3 SAMPLE RESULTS OF CLUSTERS OF INNOVATION
INITIATIVE REGIONAL SURVEY: ALL 5 REGIONS

QUESTION #38a: Which five elements of the business environment currently have the greatest
positive impact on your business’ success?

Positive impact #1 Average of

all Respondents San Diego Pittsburgh Atlanta Wichita RDU

Cost of doing business

(cost of real estate, wages and salary, etc.) 36.97% 7.83% 52.27% 50.71% 37.65% 27.54%

Quality of transportation 12.75% 5.22% 19.70% 10.71% 15.29% 13.04%

Specialized facilities for research 16.82% 41.74% 8.33% 7.86% 5.88% 23.19%

Qualified scientists and engineers 12.01% 25.22% 5.30% 7.14% 4.71% 21.74%

Transfer of knowledge from research

institutions 4.07% 6.09% 2.27% 3.57% 5.88% 2.90%

Communications infrastructure 8.13% 9.57% 3.79% 11.43% 10.59% 4.35%

Available pool of skilled workforce 3.14% 1.74% 3.03% 0.71% 7.06% 5.80%

Quality of K-12 education 0.74% 0.00% 0.00% 2.86% 0.00% 0.00%

Sourcing of employees from advanced

educational programs 0.37% 0.00% 0.76% 0.00% 1.18% 0.00%

Access to capital 1.66% 0.87% 3.03% 1.43% 1.18% 1.45%

Demanding regional customers that provide

feedback 1.11% 0.87% 0.76% 0.71% 3.53% 0.00%

Specialized needs of regional customers 0.92% 0.00% 0.00% 1.43% 3.53% 0.00%

State and regional environmental safety

regulations 0.18% 0.00% 0.00% 0.00% 1.18% 0.00%

State and regional tax and incentives for

investment in R&D 0.18% 0.87% 0.00% 0.00% 0.00% 0.00%

Governments overall responsiveness to the

needs of business 0.18% 0.00% 0.00% 0.00% 1.18% 0.00%

Level of competition in your industry 0.18% 0.00% 0.00% 0.71% 0.00% 0.00%

Quality and in-region location of

your suppliers 0.18% 0.00% 0.00% 0.71% 0.00% 0.00%

Participation with regional institutions

in R&D efforts 0.37% 0.00% 0.76% 0.00% 1.18% 0.00%

Weighted Total Number of Respondents 541 115 132 140 85 69

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-13

 RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
QUESTION #38a: Which five elements of the business environment currently have the
greatest positive impact on your business’ success?

Positive impact #2 Average of

all Respondents San Diego Pittsburgh Atlanta Wichita RDU
Cost of doing business
(cost of real estate, wages and salary, etc.) 0.95% 0.00% 0.81% 2.86% 0.00% 0.00%

Quality of transportation 9.09% 2.68% 0.81% 15.00% 7.23% 4.35%

Specialized facilities for research 6.25% 2.68% 9.68% 4.29% 3.61% 13.04%

Qualified scientists and engineers 17.61% 31.25% 16.13% 9.29% 7.23% 27.54%

Transfer of knowledge from research

institutions 8.90% 21.43% 8.06% 5.71% 1.20% 5.80%

Communications infrastructure 19.32% 11.61% 19.35% 27.86% 21.69% 11.59%

Available pool of skilled workforce 9.85% 8.93% 10.48% 5.71% 15.66% 11.59%

Quality of K-12 education 3.41% 2.68% 4.03% 1.43% 7.23% 2.90%

Sourcing of employees from advanced

educational programs 6.25% 8.93% 4.84% 9.29% 1.20% 4.35%

Access to capital 5.11% 2.68% 4.03% 5.71% 9.64% 4.35%

Demanding regional customers that

provide feedback 2.27% 0.89% 1.61% 1.43% 8.43% 0.00%

Specialized needs of regional customers 2.46% 0.89% 2.42% 2.86% 3.61% 2.90%

State local regulations for production

processes and product 0.57% 0.00% 0.00% 0.00% 1.20% 2.90%

State and regional environmental

safety regulations 0.38% 0.89% 0.00% 0.00% 0.00% 1.45%

State and regional tax and incentives

for investment in R&D 0.95% 0.00% 0.00% 0.71% 3.61% 1.45%

Predictability of government policies 1.14% 0.00% 0.81% 1.43% 2.41% 1.45%

Governments overall responsiveness to

the needs of business 0.76% 0.00% 0.81% 1.43% 0.00% 1.45%

Level of competition in your industry 2.46% 1.79% 2.42% 2.86% 3.61% 1.45%

Quality and in-region location of your suppliers 1.33% 1.79% 0.81% 1.43% 2.41% 0.00%

Assistance from regional suppliers for

new product and proce 0.19% 0.00% 0.81% 0.00% 0.00% 0.00%

Relationships between firms and

organizations in your cluster 0.38% 0.89% 0.00% 0.71% 0.00% 0.00%

Participation with regional institutions

in R&D efforts 0.19% 0.00% 0.81% 0.00% 0.00% 0.00%

Weighted Total Number of Respondents 528 112 124 140 83 69

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

A-14 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
QUESTION #38a: Which five elements of the business environment currently have the
greatest positive impact on your business’ success?

Positive impact #3 Average of

all Respondents San Diego Pittsburgh Atlanta Wichita RDU
Cost of doing business
(cost of real estate, wages and salary, etc.) 2.77% 0.00% 4.20% 5.22% 2.53% 0.00%

Specialized facilities for research 0.40% 0.00% 0.00% 0.75% 0.00% 1.47%

Qualified scientists and engineers 3.95% 1.89% 4.20% 3.73% 1.27% 10.29%

Transfer of knowledge from research institutions 8.30% 22.64% 3.36% 3.73% 1.27% 11.76%

Communications infrastructure 12.06% 15.09% 7.56% 19.40% 6.33% 7.35%

Available pool of skilled workforce 11.46% 13.21% 16.81% 9.70% 7.59% 7.35%

Quality of K-12 education 7.31% 3.77% 13.45% 4.48% 7.59% 7.35%

Sourcing of employees from advanced

educational programs 11.66% 16.98% 6.72% 14.18% 5.06% 14.71%

Access to capital 8.30% 7.55% 8.40% 10.45% 11.39% 1.47%

Demanding regional customers that

provide feedback 6.32% 7.55% 5.88% 5.97% 5.06% 7.35%

Specialized needs of regional customers 6.13% 1.89% 8.40% 5.22% 13.92% 1.47%

State local regulations for production

processes and product 1.98% 2.83% 2.52% 0.75% 3.80% 0.00%

State and regional environmental

safety regulations 1.98% 0.00% 3.36% 0.00% 1.27% 7.35%

State and regional tax and incentives

for investment in R&D 1.78% 0.94% 0.84% 0.00% 6.33% 2.94%

Predictability of government policies 1.38% 0.00% 0.84% 1.49% 1.27% 4.41%

Governments overall responsiveness to

the needs of business 2.57% 0.00% 0.84% 5.97% 2.53% 2.94%

Level of competition in your industry 3.95% 1.89% 2.52% 4.48% 10.13% 1.47%

Quality and in-region location of your suppliers 3.56% 1.89% 3.36% 1.49% 6.33% 7.35%

Assistance from regional suppliers for

new product and proce 1.38% 1.89% 2.52% 0.75% 1.27% 0.00%

Relationships between firms and

organizations in your cluste 1.98% 0.00% 2.52% 1.49% 5.06% 1.47%

Participation with regional institutions

in R&D efforts 0.20% 0.00% 0.00% 0.00% 0.00% 1.47%

Weighted Total Number of Respondents 506 106 119 134 79 68

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-15

 RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
QUESTION #38a: Which five elements of the business environment currently have the
greatest positive impact on your business’ success?

Positive impact #4 Average of Research

all Respondents San Diego Pittsburgh Atlanta Wichita Triangle
Cost of doing business
(cost of real estate, wages and salary, etc.) 2.74% 0.00% 4.50% 3.31% 5.26% 0.00%

Qualified scientists and engineers 0.21% 0.00% 0.00% 0.83% 0.00% 0.00%

Transfer of knowledge from research institutions 1.26% 0.99% 1.80% 0.83% 1.32% 1.52%

Communications infrastructure 2.32% 5.94% 1.80% 2.48% 0.00% 0.00%

Available pool of skilled workforce 5.26% 12.87% 0.90% 5.79% 0.00% 6.06%

Quality of K-12 education 2.74% 2.97% 4.50% 0.83% 2.63% 3.03%

Sourcing of employees from advance

educational programs 10.32% 17.82% 8.11% 6.61% 3.95% 16.67%

Access to capital 11.16% 12.87% 10.81% 16.53% 2.63% 9.09%

Demanding regional customers that

provide feedback 3.16% 1.98% 2.70% 4.96% 3.95% 1.52%

Specialized needs of regional customers 5.68% 3.96% 7.21% 7.44% 5.26% 3.03%

State local regulations for production

processes and product 0.63% 0.00% 0.90% 0.00% 1.32% 1.52%

State and regional environmental

safety regulations 2.53% 0.99% 1.80% 2.48% 3.95% 4.55%

State and regional tax and incentives for

investment in R&D 4.42% 5.94% 6.31% 1.65% 5.26% 3.03%

Predictability of government policies 2.32% 0.99% 1.80% 3.31% 2.63% 3.03%

Governments overall responsiveness to the

needs of business 4.21% 1.98% 2.70% 5.79% 7.89% 3.03%

Level of competition in your industry 9.05% 7.92% 10.81% 11.57% 6.58% 6.06%

Quality and in-region location of your suppliers 8.42% 4.95% 15.32% 4.96% 14.47% 1.52%

Assistance from regional suppliers for new

product and proce 4.84% 1.98% 4.50% 2.48% 10.53% 7.58%

Relationships between firms and

organizations in your cluste 12.84% 14.85% 6.31% 14.88% 14.47% 15.15%

Participation with regional institutions in

R&D efforts 3.79% 0.99% 5.41% 3.31% 1.32% 9.09%

Weighted Total Number of Respondents 475 101 111 121 76 66

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

A-16 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
QUESTION #38a: Which five elements of the business environment currently have the
greatest positive impact on your business’ success?

Positive impact #5 Average of Research

all Respondents San Diego Pittsburgh Atlanta Wichita Triangle
Cost of doing business
(cost of real estate, wages and salary, etc.) 29.98% 0.00% 46.08% 51.33% 37.68% 0.00%

Quality of transportation 0.23% 0.00% 0.00% 0.00% 1.45% 0.00%

Specialized facilities for research 0.23% 0.00% 0.00% 0.00% 1.45% 0.00%

Communications infrastructure 0.46% 1.09% 0.00% 0.00% 1.45% 0.00%

Available pool of skilled workforce 0.92% 1.09% 1.96% 0.88% 0.00% 0.00%

Quality of K-12 education 0.23% 0.00% 0.00% 0.00% 0.00% 1.64%

Sourcing of employees from advanced

educational programs 2.52% 7.61% 1.96% 0.88% 1.45% 0.00%

Access to capital 2.97% 9.78% 0.98% 0.88% 0.00% 3.28%

Demanding regional customers that

provide feedback 0.46% 0.00% 0.98% 0.88% 0.00% 0.00%

Specialized needs of regional customers 0.46% 1.09% 0.00% 0.88% 0.00% 0.00%

State local regulations for production

processes and product 0.23% 0.00% 0.98% 0.00% 0.00% 0.00%

State and regional environmental

safety regulations 0.46% 0.00% 0.00% 0.88% 0.00% 1.64%

State and regional tax and incentives for

investment in R&D 2.29% 5.43% 1.96% 1.77% 1.45% 0.00%

Predictability of government policies 0.92% 1.09% 0.00% 0.88% 2.90% 0.00%

Governments overall responsiveness to

the needs of business 2.52% 3.26% 1.96% 1.77% 5.80% 0.00%

Level of competition in your industry 4.58% 6.52% 3.92% 4.42% 5.80% 1.64%

Quality and in-region location of your suppliers 3.43% 5.43% 4.90% 0.88% 2.90% 3.28%

Assistance from regional suppliers for new

product and proce 2.52% 3.26% 3.92% 0.88% 2.90% 1.64%

Relationships between firms and organizations

in your cluste 10.07% 22.83% 2.94% 7.96% 13.04% 3.28%

Participation with regional institutions

in R&D efforts 34.32% 31.52% 27.45% 24.78% 20.29% 83.61%

Weighted Total Number of Respondents 437 92 102 113 69 61

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-17

 RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
QUESTION #38b: Which factors of the business environment do you consider to be the
greatest future threats to your business if not addressed?

Future Threat #1 Average of Research

all Respondents San Diego Pittsburgh Atlanta Wichita Triangle
Cost of doing business
(cost of real estate, wages and salar 53.58% 87.18% 38.17% 38.13% 51.16% 59.72%

Quality of transportation 23.85% 6.84% 17.56% 38.85% 27.91% 29.17%

Specialized facilities for research 1.83% 0.00% 5.34% 0.72% 0.00% 2.78%

Qualified scientists and engineers 5.69% 3.42% 9.16% 6.47% 4.65% 2.78%

Transfer of knowledge from research institutions 0.55% 0.00% 1.53% 0.72% 0.00% 0.00%

Communications infrastructure 1.10% 0.00% 3.05% 0.72% 1.16% 0.00%

Available pool of skilled workforce 7.71% 0.85% 10.69% 10.07% 12.79% 2.78%

Quality of K-12 education 1.28% 0.00% 2.29% 2.16% 0.00% 1.39%

Access to capital 1.83% 0.85% 5.34% 0.00% 1.16% 1.39%

Demanding regional customers that

provide feedback 0.18% 0.00% 0.00% 0.00% 1.16% 0.00%

State local regulations for production

processes and product 1.10% 0.85% 3.82% 0.00% 0.00% 0.00%

State and regional environmental

safety regulations 0.37% 0.00% 1.53% 0.00% 0.00% 0.00%

Predictability of government policies 0.18% 0.00% 0.76% 0.00% 0.00% 0.00%

Governments overall responsiveness

to the needs of business 0.18% 0.00% 0.76% 0.00% 0.00% 0.00%

Level of competition in your industry 0.37% 0.00% 0.00% 1.44% 0.00% 0.00%

Weighted Total Number of Respondents 545 117 131 139 86 72

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

A-18 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
QUESTION #38b: Which factors of the business environment do you consider to be the
greatest future threats to your business if not addressed?

Future Threat #2 Average of Research

all Respondents San Diego Pittsburgh Atlanta Wichita Triangle
Cost of doing business
(cost of real estate, wages and salar 0.93% 0.00% 0.78% 0.00% 3.53% 1.41%

Quality of transportation 25.23% 52.14% 6.20% 24.09% 11.76% 33.80%

Specialized facilities for research 1.67% 1.71% 0.78% 0.73% 1.18% 5.63%

Qualified scientists and engineers 12.43% 17.09% 14.73% 10.22% 9.41% 8.45%

Transfer of knowledge from research institutions 2.41% 2.56% 3.10% 1.46% 1.18% 4.23%

Communications infrastructure 4.08% 5.13% 4.65% 2.92% 3.53% 4.23%

Available pool of skilled workforce 24.49% 8.55% 23.26% 35.77% 35.29% 18.31%

Quality of K-12 education 10.02% 3.42% 10.08% 13.87% 11.76% 11.27%

Sourcing of employees from advanced

educational programs 0.93% 1.71% 1.55% 0.00% 1.18% 0.00%

Access to capital 5.57% 3.42% 7.75% 4.38% 7.06% 5.63%

Demanding regional customers that

provide feedback 1.11% 0.00% 3.10% 0.73% 1.18% 0.00%

Specialized needs of regional customers 0.74% 0.00% 0.78% 0.73% 2.35% 0.00%

State local regulations for production

processes and product 4.27% 3.42% 7.75% 1.46% 4.71% 4.23%

State and regional environmental

safety regulations 2.23% 0.85% 6.98% 0.00% 1.18% 1.41%

State and regional tax and incentives

for investment in R&D 1.48% 0.00% 4.65% 0.00% 2.35% 0.00%

Predictability of government policies 0.93% 0.00% 1.55% 1.46% 0.00% 1.41%

Governments overall responsiveness

to the needs of business 0.19% 0.00% 0.78% 0.00% 0.00% 0.00%

Level of competition in your industry 0.93% 0.00% 1.55% 0.73% 2.35% 0.00%

Relationships between firms and

organizations in your cluste 0.19% 0.00% 0.00% 0.73% 0.00% 0.00%

Weighted Total Number of Respondents 539 117 129 137 85 71

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-19

 RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
QUESTION #38b: Which factors of the business environment do you consider to be the
greatest future threats to your business if not addressed?

Future Threat #3 Average of Research

all Respondents San Diego Pittsburgh Atlanta Wichita Triangle
Cost of doing business
(cost of real estate, wages and salar 0.19% 0.00% 0.00% 0.78% 0.00% 0.00%

Specialized facilities for research 0.58% 0.86% 0.00% 0.00% 1.20% 1.52%

Qualified scientists and engineers 5.78% 12.93% 2.40% 3.10% 1.20% 10.61%

Transfer of knowledge from research institutions 1.54% 2.59% 2.40% 0.78% 0.00% 1.52%

Communications infrastructure 1.73% 1.72% 2.40% 2.33% 1.20% 0.00%

Available pool of skilled workforce 22.93% 31.90% 16.00% 24.81% 19.28% 21.21%

Quality of K-12 education 16.96% 12.07% 6.40% 24.81% 20.48% 25.76%

Sourcing of employees from advanced

educational programs 5.97% 4.31% 9.60% 5.43% 6.02% 3.03%

Access to capital 9.83% 10.34% 15.20% 9.30% 2.41% 9.09%

Demanding regional customers that

provide feedback 1.16% 2.59% 0.00% 0.78% 2.41% 0.00%

Specialized needs of regional customers 2.70% 0.86% 3.20% 4.65% 2.41% 1.52%

State local regulations for production

processes and product 6.17% 8.62% 5.60% 3.88% 8.43% 4.55%

State and regional environmental safety regulations 7.90% 5.17% 8.80% 6.98% 14.46%
4.55%

State and regional tax and incentives for

investment in R&D 3.85% 3.45% 7.20% 2.33% 2.41% 3.03%

Predictability of government policies 4.43% 0.86% 8.00% 2.33% 6.02% 6.06%

Governments overall responsiveness to

the needs of business 3.66% 0.00% 7.20% 3.88% 4.82% 1.52%

Level of competition in your industry 2.31% 0.00% 3.20% 1.55% 3.61% 4.55%

Quality and in-region location of your suppliers 1.54% 0.86% 0.80% 2.33% 2.41% 1.52%

Assistance from regional suppliers for new

product and proce 0.19% 0.00% 0.80% 0.00% 0.00% 0.00%

Relationships between firms and organizations

in your cluster 0.39% 0.86% 0.80% 0.00% 0.00% 0.00%

Participation with regional institutions

in R&D efforts 0.19% 0.00% 0.00% 0.00% 1.20% 0.00%

Weighted Total Number of Respondents 519 116 125 129 83 66

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

A-20 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
QUESTION #38b: Which factors of the business environment do you consider to be the
greatest future threats to your business if not addressed?

Future Threat #4 Average of Research

all Respondents San Diego Pittsburgh Atlanta Wichita Triangle
Cost of doing business
(cost of real estate, wages and salar 0.82% 0.92% 0.00% 1.67% 1.25% 0.00%

Specialized facilities for research 0.21% 0.92% 0.00% 0.00% 0.00% 0.00%

Transfer of knowledge from research institutions 0.21% 0.00% 0.85% 0.00% 0.00% 0.00%

Communications infrastructure 0.41% 1.83% 0.00% 0.00% 0.00% 0.00%

Available pool of skilled workforce 6.17% 10.09% 5.98% 5.00% 2.50% 6.67%

Quality of K-12 education 7.82% 8.26% 2.56% 15.00% 5.00% 6.67%

Sourcing of employees from advanced

educational programs 5.35% 8.26% 4.27% 5.83% 5.00% 1.67%

Access to capital 9.47% 13.76% 6.84% 7.50% 6.25% 15.00%

Demanding regional customers that

provide feedback 1.23% 0.00% 0.85% 3.33% 1.25% 0.00%

Specialized needs of regional customers 1.65% 0.92% 0.00% 3.33% 1.25% 3.33%

State local regulations for production

processes and product 5.14% 8.26% 3.42% 2.50% 5.00% 8.33%

State and regional environmental

safety regulations 8.23% 11.01% 10.26% 5.83% 6.25% 6.67%

State and regional tax and incentives for

investment in R&D 9.67% 8.26% 9.40% 8.33% 11.25% 13.33%

Predictability of government policies 13.17% 13.76% 14.53% 10.83% 13.75% 13.33%

Governments overall responsiveness to

the needs of business 12.96% 6.42% 23.93% 10.83% 15.00% 5.00%

Level of competition in your industry 11.11% 4.59% 7.69% 14.17% 17.50% 15.00%

Quality and in-region location of your suppliers 2.26% 1.83% 3.42% 0.83% 3.75% 1.67%

Assistance from regional suppliers for

new product and proce 1.65% 0.92% 2.56% 2.50% 1.25% 0.00%

Relationships between firms and

organizations in your cluste 2.06% 0.00% 1.71% 2.50% 3.75% 3.33%

Participation with regional institutions

in R&D efforts 0.41% 0.00% 1.71% 0.00% 0.00% 0.00%

Weighted Total Number of Respondents 486 109 117 120 80 60

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-21

 RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
QUESTION #38b: Which factors of the business environment do you consider to be the
greatest future threats to your business if not addressed?

Future Threat #5 Average of Research

all Respondents San Diego Pittsburgh Atlanta Wichita Triangle
Cost of doing business
(cost of real estate, wages and salar 7.69% 0.00% 10.00% 15.32% 9.46% 0.00%

Available pool of skilled workforce 0.22% 0.97% 0.00% 0.00% 0.00% 0.00%

Quality of K-12 education 1.54% 1.94% 0.91% 1.80% 0.00% 3.51%

Sourcing of employees from advanced

educational programs 1.54% 1.94% 1.82% 1.80% 1.35% 0.00%

Access to capital 3.52% 9.71% 1.82% 2.70% 0.00% 1.75%

Specialized needs of regional customers 0.22% 0.00% 0.91% 0.00% 0.00% 0.00%

State local regulations for production

processes and product 0.66% 0.00% 0.91% 0.90% 0.00% 1.75%

State and regional environmental

safety regulations 4.62% 6.80% 3.64% 5.41% 2.70% 3.51%

State and regional tax and incentives

for investment in R&D 3.08% 4.85% 1.82% 2.70% 2.70% 3.51%

Predictability of government policies 7.47% 6.80% 4.55% 11.71% 6.76% 7.02%

Governments overall responsiveness to

the needs of business 19.12% 30.10% 20.00% 15.32% 13.51% 12.28%

Level of competition in your industry 22.64% 27.18% 21.82% 16.22% 25.68% 24.56%

Quality and in-region location of your suppliers 5.71% 0.97% 7.27% 5.41% 8.11% 8.77%

Assistance from regional suppliers for

new product and proce 2.42% 0.00% 3.64% 0.90% 4.05% 5.26%

Relationships between firms and organizations

in your cluste 8.35% 5.83% 9.09% 6.31% 14.86% 7.02%

Participation with regional institutions

in R&D efforts 4.62% 2.91% 8.18% 5.41% 2.70% 1.75%

Weighted Total Number of Respondents 455 103 110 111 74 57

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

A-22 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY
Below is a list of actions governments at various levels can take that may help increase innovation
in your region. How important is each, as a priority or continuing priority for government over
the next five years?

Promote world-class primary and secondary education

Average of Research
all Respondents San Diego Pittsburgh Atlanta Wichita Triangle
1 (not at all important) 0.93% 0.60% 2.16% 1.01% 0.00% 0.88%

2 1.86% 1.81% 2.16% 0.51% 2.94% 2.63%

3 10.62% 7.83% 15.11% 5.56% 19.85% 7.02%

4 32.01% 34.34% 35.25% 27.27% 36.03% 28.07%

5 (critically important) 54.58% 55.42% 45.32% 65.66% 41.18% 61.40%

Weighted Total Number of Respondents 753 166 139 198 136 114

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Promote specialized education and training programs to upgrade

1 (not at all important) 0.67% 1.21% 1.45% 0.51% 0.00% 0.00%

2 5.87% 8.48% 7.97% 4.59% 4.41% 3.51%

3 16.82% 24.24% 13.04% 12.76% 16.18% 18.42%

4 45.53% 42.42% 50.00% 42.86% 51.47% 42.11%

5 (critically important) 31.11% 23.64% 27.54% 39.29% 27.94% 35.96%

Weighted Total 749 165 138 196 136 114

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Implement tax reform to encourage investment in innovation

1 (not at all important) 0.74% 0.60% 0.00% 0.51% 1.47% 1.77%

2 6.67% 5.99% 2.53% 6.63% 7.35% 14.16%

3 28.64% 27.54% 24.75% 31.63% 33.09% 26.55%

4 39.01% 41.92% 38.38% 42.35% 30.88% 39.82%

5 (critically important) 24.81% 23.95% 33.84% 18.88% 27.21% 17.70%

Weighted Total Number of Respondents 810 167 198 196 136 113

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-23

 RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY

Average of Research
all Respondents San Diego Pittsburgh Atlanta Wichita Triangle

Speed up regulatory approval processes in line with product

1 (not at all important) 5.03% 1.82% 5.73% 7.18% 6.77% 2.70%

2 15.08% 13.94% 13.54% 14.87% 18.05% 16.22%

3 32.04% 21.21% 34.90% 36.92% 35.34% 30.63%

4 32.16% 36.97% 29.69% 27.69% 30.08% 39.64%

5 (critically important) 15.58% 26.06% 16.15% 12.82% 9.77% 10.81%

Weighted Total Number of Respondents 796 165 192 195 133 111

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Simplify compliance procedures for government regulations

1 (not at all important) 1.98% 0.60% 3.05% 3.05% 2.21% 0.00%

2 10.37% 5.99% 10.66% 8.63% 14.71% 14.16%

3 31.48% 27.54% 27.92% 35.03% 36.76% 30.97%

4 36.91% 43.11% 36.04% 37.06% 29.41% 38.05%

5 (critically important) 19.01% 22.75% 21.32% 16.24% 16.91% 16.81%

Weighted Total Number of Respondents 810 167 197 197 136 113

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Reform liability laws to stimulate and reward next generation

1 (not at all important) 3.73% 2.44% 5.64% 4.57% 2.22% 2.65%

2 16.42% 15.85% 13.33% 16.75% 16.30% 22.12%

3 34.33% 31.10% 34.36% 32.99% 37.78% 37.17%

4 30.35% 26.83% 32.31% 34.52% 31.11% 23.89%

5 (critically important) 15.17% 23.78% 14.36% 11.17% 12.59% 14.16%

Weighted Total Number of Respondents 804 164 195 197 135 113

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

A-24 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY

Average of Research
all Respondents San Diego Pittsburgh Atlanta Wichita Triangle

Promote antitrust legislation to encourage competition

1 (not at all important) 18.20% 19.02% 20.62% 15.74% 16.91% 18.75%

2 34.04% 35.58% 35.05% 29.44% 31.62% 41.07%

3 30.67% 24.54% 27.84% 35.53% 38.24% 26.79%

4 13.59% 15.95% 13.40% 16.24% 10.29% 9.82%

5 (critically important) 3.49% 4.91% 3.09% 3.05% 2.94% 3.57%

Weighted Total Number of Respondents 802 163 194 197 136 112

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Support the particular needs of start-up companies

1 (not at all important) 4.20% 4.19% 5.53% 3.59% 5.15% 1.77%

2 12.72% 12.57% 13.57% 8.72% 16.91% 13.27%

3 32.10% 29.34% 25.63% 36.92% 36.76% 33.63%

4 30.00% 35.33% 29.65% 32.82% 25.00% 23.89%

5 (critically important) 20.99% 18.56% 25.63% 17.95% 16.18% 27.43%

Weighted Total Number of Respondents 810 167 199 195 136 113

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Strengthen and modernize intellectual property protections

1 (not at all important) 3.74% 3.66% 3.59% 3.57% 5.19% 2.68%

2 14.46% 7.93% 14.87% 14.80% 22.96% 12.50%

3 33.29% 25.00% 32.82% 37.24% 40.74% 30.36%

4 34.04% 45.12% 35.38% 31.12% 20.74% 36.61%

5 (critically important) 14.46% 18.29% 13.33% 13.27% 10.37% 17.86%

Weighted Total Number of Respondents 802 164 195 196 135 112

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-25

 RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY

Average of Research
all Respondents San Diego Pittsburgh Atlanta Wichita Triangle

Provide services to assist and promote regional exports

1 (not at all important) 8.31% 11.45% 7.07% 9.28% 5.19% 7.96%

2 24.19% 28.31% 22.73% 26.80% 22.22% 18.58%

3 36.72% 33.13% 33.84% 38.66% 34.81% 46.02%

4 24.94% 20.48% 27.78% 22.16% 31.11% 23.89%

5 (critically important) 5.71% 6.63% 8.59% 2.58% 6.67% 3.54%

Weighted Total Number of Respondents 806 166 198 194 135 113

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Improve information and communications infrastructure

1 (not at all important) 3.35% 3.07% 4.06% 3.06% 4.44% 1.75%

2 12.80% 15.95% 13.20% 5.61% 14.81% 17.54%

3 32.55% 37.42% 30.96% 29.08% 37.78% 28.07%

4 34.53% 28.83% 37.06% 37.24% 31.85% 36.84%

5 (critically important) 16.77% 14.72% 14.72% 25.00% 11.11% 15.79%

Weighted Total Number of Respondents 805 163 197 196 135 114

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Assist in attracting suppliers and service providers

1 (not at all important) 10.25% 16.67% 11.28% 10.20% 5.15% 5.41%

2 25.50% 29.01% 24.62% 21.43% 20.59% 35.14%

3 32.50% 29.63% 29.74% 35.20% 38.97% 28.83%

4 25.88% 20.99% 27.18% 27.55% 30.15% 22.52%

5 (critically important) 5.75% 3.70% 7.18% 5.61% 4.41% 8.11%

Weighted Total Number of Respondents 800 162 195 196 136 111

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

A-26 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY

Average of Research
all Respondents San Diego Pittsburgh Atlanta Wichita Triangle

Promote universal computer literacy

1 (not at all important) 3.44% 0.00% 4.62% 2.04% 3.70% 3.54%

2 13.62% 0.00% 16.92% 12.24% 14.81% 8.85%

3 32.55% 0.00% 33.85% 27.55% 36.30% 34.51%

4 33.80% 0.00% 30.26% 35.71% 31.85% 38.94%

5 (critically important) 16.59% 0.00% 14.36% 22.45% 13.33% 14.16%

Weighted Total Number of Respondents 639 0 195 196 135 113

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Government support for funding of specialized research institutions

1 (not at all important) 7.58% 3.64% 7.69% 6.60% 13.33% 7.96%

2 24.22% 20.61% 29.23% 20.81% 28.15% 22.12%

3 33.66% 35.15% 32.31% 34.01% 35.56% 30.97%

4 23.73% 26.06% 21.03% 28.93% 17.78% 23.01%

5 (critically important) 10.68% 14.55% 9.23% 9.64% 5.19% 15.93%

Weighted Total Number of Respondents 805 165 195 197 135 113

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Catalyze partnerships among government, industry and universities

1 (not at all important) 6.59% 6.13% 7.18% 7.11% 8.15% 3.51%

2 19.40% 20.25% 15.38% 18.27% 26.67% 18.42%

3 33.58% 38.04% 32.31% 32.49% 32.59% 32.46%

4 27.99% 24.54% 30.77% 28.93% 25.93% 28.95%

5 (critically important) 12.44% 11.04% 14.36% 13.20% 6.67% 16.67%

Weighted Total Number of Respondents 804 163 195 197 135 114

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-27

 RESULTS OF CLUSTERS OF INNOVATION INITIATIVE REGIONAL SURVEY

Average of Research
all Respondents San Diego Pittsburgh Atlanta Wichita Triangle

Improve transportation and other physical infrastructure

1 (not at all important) 1.37% 0.62% 2.04% 1.02% 2.96% 0.00%

2 4.60% 4.32% 8.67% 2.03% 5.19% 1.75%

3 19.40% 16.67% 22.45% 14.21% 31.11% 13.16%

4 32.96% 34.57% 36.22% 31.98% 28.89% 31.58%

5 (critically important) 41.67% 43.83% 30.61% 50.76% 31.85% 53.51%

Weighted Total Number of Respondents 804 162 196 197 135 114

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

Increase funding for university-based research

1 (not at all important) 7.71% 3.64% 10.26% 8.12% 11.94% 3.54%

2 20.27% 16.97% 23.59% 14.72% 27.61% 20.35%

3 32.84% 32.73% 32.31% 33.50% 35.07% 30.09%

4 25.62% 23.03% 23.59% 29.95% 19.40% 32.74%

5 (critically important) 13.56% 23.64% 10.26% 13.71% 5.97% 13.27%

Weighted Total Number of Respondents 804 165 195 197 134 113

Row Percentage 100.00% 20.34% 24.46% 24.46% 16.71% 14.04%

A-28 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS

ABOUT
THE CLUSTERS OF INNOVATION INITIATIVE PARTICIPANTS

MICHAEL E. PORTER
Michael E. Porter is the Bishop William Lawrence University Professor at Harvard University and
a leading authority on competitive strategy and international competitiveness. He co-chairs the
Clusters of Innovation Initiative at the Council on Competitiveness and is a member of the Council’s
executive committee.
The author of 16 books and over 75 articles, Professor Porter’s ideas have guided economic policy
throughout the world. Professor Porter has led competitiveness initiatives in nations and states such as
Canada, India, New Zealand, and Connecticut; guides regional projects in Central America and the
Middle East; and is co-chairman of the Global Competitiveness Report. In 1994, Professor Porter founded
the Initiative for a Competitive Inner City, a non-profit private sector initiative formed to catalyze business
development in distressed inner cities across the United States. The holder of eight honorary doctorates,
Professor Porter has won numerous awards for his books, articles, public service, and influence on
several fields.

COUNCIL on COMPETITIVENESS
The Council is a nonprofit, 501(c)(3) organization whose members are corporate chief executives,
university presidents, and labor leaders dedicated to setting an action agenda to drive U.S. economic
competitiveness and leadership in world markets. The Council helps shape the national debate on com-
petitiveness by concentrating on a few critical issues including technological innovation, workforce
development, and the benchmarking of U.S. economic performance against other countries.
The Council’s work is guided by a 30 member executive committee. Chief executives of 40 of the
country’s most prominent nonprofit research organizations, professional societies and trade associations
contribute their expertise as national affiliates of the Council.

CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS A-29

 MONITOR GROUP
Monitor Group is a family of competitive service firms linked by shared ownership, management
philosophy, and inter-related assets. Each entity in the Group is dedicated to providing products and
services which fundamentally enhance the competitiveness of our clients. Our aspiration is to operate as
an “intelligent switch” in a closely-linked global network of expertise and experience, not merely as a
narrowly defined consulting firm, a research company or a merchant bank. We are dedicated to creating
innovative, winning, action-oriented solutions by deploying our human, knowledge, and social assets in
unique combinations dictated by each client’s unique circumstances — consulting interventions, capital
infusions, deal structuring, management development programs, customized software, cutting-edge
market research, and so on as appropriate.

Monitor Group is organized into three major operating units:

• Monitor Action Group, which consults to top management to help resolve their most important

and intractable competitive problems;

• The Monitor Merchant Banking Group, which marries capital investment with advisory services

to enhance company competitiveness;

• The Intelligent Products Group, which provides customized data and software products to

support competitive decision making.

ontheFRONTIER, a Monitor Group company, has extensive experience in competitiveness assessment

and cluster development projects throughout the United States and the world. Our private and public
sector client base spans over twenty countries in North and South America, Europe, Asia, Africa and the
Middle East. In addition, we have collaborated extensively with development agencies such as the World
Bank Group and the United States Agency for International Development (USAID) on microeconomic
development issues.

ontheFRONTIER’s work focuses on improving business competitiveness through building winning

strategies, fostering cooperation among clusters of firms, and facilitating productive dialogue between
private and public sector leaders to promote innovation. Our vast network of partners forms the basis
of our collaborative effort to diffuse a new web-based set of offerings. We are working with financial
institutions, industry associations, multilateral agencies, and others to diffuse web-based business
strategy tools and insights to businesspeople around the world. For more information, please visit
www.ontheFRONTIER.com.

A-30 CLUSTERS OF INNOVATION: REGIONAL FOUNDATIONS OF U.S. COMPETITIVENESS