Order Code RL33206
CRS Report for Congress
Received through the CRS Web
Vulnerability of Concentrated Critical
Infrastructure: Background and Policy Options
December 21, 2005
Paul W. Parfomak
Specialist in Science and Technology
Resources, Science, and Industry Division
Congressional Research Service ˜ The Library of Congress

Vulnerability of Concentrated Critical Infrastructure:
Background and Policy Options
Summary
“Critical infrastructure” consists of systems and assets so vital to the United
States that their incapacity would harm the nation’s physical security, economic
security, or public health. Critical infrastructure is often geographically concentrated,
so it may be distinctly vulnerable to events like natural disasters, epidemics, and
certain kinds of terrorist attacks. Disruption of concentrated infrastructure could
have greatly disproportionate effects, with costs potentially running into billions of
dollars and spreading far beyond the immediate area of disturbance. Hurricanes
Katrina and Rita demonstrated this kind of geographic vulnerability by disrupting a
substantial part of the U.S. energy and chemical sectors in 2005.
As the nation’s response to recent hurricanes and other disasters continues, and
as its homeland security activities evolve, Congress is examining federal policies
affecting the geographic concentration and vulnerability of critical infrastructure. For
example, the Energy Policy Act of 2005 (P.L. 109-58) facilitates construction of new
liquefied natural gas import terminals in diverse ports by granting the Federal Energy
Regulatory Commission exclusive siting approval authority (Section 311). The
Gasoline for America’s Security Act of 2005 (H.R. 3893) would similarly facilitate
construction of geographically distributed oil refineries.
Geographic concentrations of U.S. critical infrastructure have developed for
multiple reasons—typically some combination of market influences including
resource location, agglomeration economies, scale economies, community
preferences, and capital efficiency. Congress and federal agencies also have adopted
a wide range of policies affecting the capacity and location of critical infrastructure,
including prescriptive siting, economic incentives, environmental regulation, and
economic regulation. Some federal policies have been developed specifically to
address perceived threats to critical infrastructure. These influences often have been
in place for decades, gradually driving critical infrastructure to its geographic
configuration today.
Some analysts may argue that little government intervention is necessary to
alleviate geographic vulnerabilities of critical infrastructure because the private sector
will adjust its practices out of its own financial interest. However, if Congress
concludes that federal intervention is needed, it may employ a number of policy
options to encourage geographic dispersion (including eliminating policies that
encourage concentration ), ensure survivability, or ensure that effective infrastructure
recovery capabilities are in place to mitigate impacts of concentrated infrastructure
disruption. Addressing geographic vulnerabilities may call for a combination of
options. Congress may also consider whether other legislative proposals with the
potential to affect critical infrastructure development—directly or indirectly—are
likely to relieve or exacerbate geographic vulnerability. The economic efficiency of
public critical infrastructure and the efficient use of federal funds for infrastructure
development may also be important considerations.
This report will not be updated.

Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Scope and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Geographic Infrastructure Concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
What is Critical Infrastructure? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
What is Geographic Concentration? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Geographic Hazards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Meteorological Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Earthquakes and Tsunamis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Infectious Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Terrorism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Frequency of Major Geographic Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Market Influences on Geographic Concentration . . . . . . . . . . . . . . . . . . . . . 9
Resource Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Agglomeration Economies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Scale Economies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Community Preferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Capital Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Federal Policies and Infrastructure Concentration . . . . . . . . . . . . . . . . . . . . . . . . 12
Prescriptive Siting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Economic Incentives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Environmental Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Economic Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Policy Options to Reduce Infrastructure Vulnerability . . . . . . . . . . . . . . . . . . . . 16
Eliminating Policies Encouraging Concentration . . . . . . . . . . . . . . . . . . . . 17
Encouraging Geographic Dispersion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Ensuring Infrastructure Survivability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Ensuring Infrastructure Recovery Capabilities . . . . . . . . . . . . . . . . . . . . . . 19
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
List of Figures
Figure 1. Oil and Gas Pipelines in the Continental United States . . . . . . . . . . . . 9
List of Tables
Table 1. Selected U.S. Disasters Since 1990 with Costs Exceeding $1 Billion . . 8

Vulnerability of Concentrated Critical
Infrastructure: Background and Policy
Options
Introduction
Critical infrastructure is so vital to the United States that its incapacity would
harm the nation’s physical security, economic security, or public health. The federal
government has a key role in helping protect the nation’s critical infrastructure from
all types of hazards through programs of mitigation, preparedness, response, and
recovery. Accordingly, Congress has a strong interest in the vulnerability of critical
infrastructure to natural hazards, accidents, or terrorism. Since September 11, 2001,
legislators, government agencies, and industry increasingly have been focused on the
sources of infrastructure vulnerability and potential measures to address those
vulnerabilities through operational changes and capital investment.
When infrastructure is physically concentrated in a limited geographic area it
may be particularly vulnerable to geographic hazards such as natural disasters,
epidemics, and certain kinds of terrorist attacks. Whereas a typical geographic
disruption is often expected to affect infrastructure in proportion to the size of an
affected region, a disruption of concentrated infrastructure could have greatly
disproportionate—and national—effects. A catastrophic ice storm in metropolitan
Chicago, for example, would undoubtedly create local emergencies, but could also
temporarily disrupt rail transportation and associated commerce throughout the
country because Chicago is a major railway hub. Extended closure of the port of
Long Beach, the largest port in the nation, would greatly harm California’s economy,
but could also disrupt vital supply chains for a number of national industries.1 The
social and economic impacts of geographic disasters are often difficult to quantify,
but their costs can quickly run into the billions and can spread far beyond the area of
the event itself. Hurricanes Katrina and Rita demonstrated this kind of geographic
impact by disrupting a substantial part of the national U.S. energy and chemical
sectors, both heavily concentrated in the Gulf of Mexico.
As the nation’s responses to recent natural disasters continue, and as its
homeland security activities evolve, Congress is examining federal policies related
to the geographic concentration and vulnerability of critical infrastructure. For
example, the Energy Policy Act of 2005 (P.L. 109-58) facilitates the construction of
new liquefied natural gas import terminals in diverse ports by granting the Federal
Energy Regulatory Commission exclusive siting approval authority (Section 311).
1 Hall, P.V. “‘We’d Have to Sink the Ships’: Impact Studies and the 2002 West Coast Port
Lockout.” Economic Development Quarterly, vol. 18, no. 4. Nov. 2004, pp. 354-367.

CRS-2
The Gasoline for America’s Security Act of 2005 (H.R. 3893) would similarly
facilitate the construction of geographically distributed oil refineries.
This report provides an overview of geographic concentration and related
vulnerability among critical infrastructures in the United States. The report illustrates
the nature of such geographic concentration and how it may expose infrastructures
to catastrophic failure due to geographic hazards. It identifies several long-term
forces which have contributed to infrastructure concentration. These forces include
resource location, agglomeration economies, scale economies, community
preferences, and capital efficiency. It reviews several ways in which the federal
government has also influenced critical infrastructure, such as prescriptive siting,
economic incentives, environmental regulation, and economic regulation. The report
concludes with options to address geographic vulnerability in the context of current
federal infrastructure policy.
Scope and Limitations
This report focuses on “nationally” critical infrastructure and related federal
policies. While many of the infrastructure and policy issues addressed in this report
may also apply at the state and local levels, the report discusses them only in the
context of federal activities. This report also discusses a number of specific
geographic hazards to critical infrastructure in the context of a broader federal policy
discussion. The report does not attempt to quantify the likelihood of any particular
hazard occurring in any particular location, or the degree of vulnerability of any
particular infrastructure concentration to geographic hazards. Such projections are
available elsewhere2 and are beyond the scope of this analysis.
Geographic Infrastructure Concentration
What is Critical Infrastructure?
Twenty years ago, “infrastructure” was defined primarily with respect to the
adequacy of the nation’s public works. In the mid-1990’s, however, the growing
threat of international terrorism led policy makers to reconsider the definition of
“infrastructure” in the context of homeland security. Successive federal government
reports, laws and executive orders have refined, and generally expanded, the number
of infrastructure sectors and the types of assets considered to be “critical” for
purposes of homeland security. The USA PATRIOT Act of 2001(P.L. 107-56
Section 1016e) contains the federal government’s most recent definition of “critical
infrastructure.” According to the act, “critical infrastructure” is
systems and assets, whether physical or virtual, so vital to the United States that
the incapacity or destruction of such systems and assets would have a debilitating
impact on security, national economic security, national public health or safety,
or any combination of those matters (Section 1016e).
2 See, for example: Risk Management Solutions, Inc., “Catastrophic Risk in the United
States,” Map, 2005, at [http://www.rms.com/Publications/Maps.asp#].

CRS-3
This definition was adopted, by reference, in the Homeland Security Act of 2002
(P.L. 107-296, Section 2.4) establishing the Department of Homeland Security
(DHS). The Bush Administration’s National Strategy for the Physical Protection of
Critical Infrastructures and Key Assets
(NSPP),3 contains a detailed list of critical
infrastructures and assets of national importance, as follows:
! Information technology
! Drinking water/water treatment
! Telecommunications
! Energy
! Chemicals
! Banking and finance
! Transportation
! National monuments and icons
! Emergency services
! Defense industrial base
! Postal and shipping services
! Key industry/technology sites
! Agriculture and food
! Large gathering sites
! Public health and healthcare
As the list suggests and the NPP acknowledges explicitly, “the lion’s share of
our critical infrastructures and key assets are owned and operated by the private
sector.”4 The list may continue to evolve as economic changes or geopolitical
developments influence homeland security policy.
What is Geographic Concentration?
This report defines “geographic concentration” of critical infrastructure as the
physical location of critical assets in sufficient proximity to each other that they are
vulnerable to disruption by the same, or successive, regional events. To be of
national significance, the collection of concentrated assets may account for a
significant fraction of the nation’s total infrastructure capacity in a given sector or
subsector. Alternatively, the collection of regional assets could make up an
infrastructure hub, accounting for a nationally significant fraction of commodity or
service flows through that infrastructure sector or subsector. The threshold above
which such assets could be considered “nationally” concentrated would depend upon
the type of impact resulting from a prolonged disruption. From strictly a market
perspective, for example, some policy makers have suggested that a change in energy
infrastructure capacity of as little as 10% to 15% could have an exaggerated effect
on related market prices.5 The corporate merger guidelines used by the United States,
Canada, and the European Union variously assume that a company must have a 25%
to 35% market share to exercise market power, and so uncompetitively influence
market prices or supplies.6 Although the loss of critical infrastructure would have
3 Office of the President, The National Strategy for the Physical Protection of Critical
Infrastructure and Key Assets
, Feb. 2003, 83 pp.
4 Ibid.
5 Hon. Joe Barton, Remarks at the House Energy and Commerce Committee Hearing on
Recovering from Katrina, Sept. 7, 2005.
6 Facey, B.A. and H. Huser, “A Comparison of Horizontal Merger Guidelines in Canada, the
European Union, and the United States,” Antitrust, fall 2004, pp. 43-50.

CRS-4
effects beyond market price, other possible metrics of concentration (e.g.,
environmental) offer little additional clarity on concentration thresholds.
Many of the critical infrastructure sectors identified in the NSPP exhibit some
degree of geographic concentration, as illustrated by the following examples.
! Chemicals (chlorine)—Over 38% of U.S. chlorine production is
located in coastal Louisiana.7
! Transportation (marine cargo)—Nearly 32% of U.S. waterborne
container shipments pass through the ports of Long Beach and Los
Angeles in southern California.8
! Transportation (rail)—Over 36% of U.S. freight railcars pass
through Illinois, primarily around Chicago. Nearly 25% of freight
railcars pass through Missouri, primarily around St. Louis.9
! Agriculture and food (livestock)—Approximately 27% of U.S. hog
inventories are located in Iowa. Another 16% of hog inventories are
located in the eastern counties of North Carolina.10
! Public health and health care (pharmaceuticals)—Approximately
25% of U.S. pharmaceuticals are manufactured in Puerto Rico,
primarily in the San Juan metropolitan area.11
! Energy (refining)—Over 43% of total U.S. oil refining capacity is
clustered along the Texas and Louisiana coasts.12
! Banking and finance (securities market)—Nearly 46% of U.S.
securities are traded on the floors of the New York and American
7 U.S. Census Bureau, Alkalies and Chlorine Manufacturing: 2002, Economic Census,
Manufacturing Industry Series, EC02-31I-325181 (RV), Dec. 2004, Table 2, Table 6b.
8 Army Corps of Engineers, Waterborne Commerce Statistics Center (WCSC). “U.S.
Waterborne Container Traffic by Port/Waterway in 2003.” Dec. 1, 2004. See
[http://www.iwr.usace.army.mil/ndc/wcsc/by_portname03.htm.].
9 Association of American Railroads (AAR). “Rail Carloads Carried by State: 2003.” Oct.
26, 2005. See [http://www.aar.org/PubCommon/Documents/AboutTheIndustry/
RRState_Rankings.pdf].
10 U.S. Department of Agriculture, National Agricultural Statistics Service (NASS),
Quarterly Hogs and Pigs, Sept. 30, 2005, p. 5.
11 U.S. Census Bureau, Pharmaceutical Preparation Manufacturing: 2002, Economic
Census, Manufacturing Industry Series. EC02-31I-325412 (RV), Dec. 2004. Table 2; Puerto
Rico Manufacturing
. 2002 Economic Census of Island Areas. IA02-00I-PRM (RV). Oct.
2005. Table 1.
12 Energy Information Administration (EIA). Petroleum Supply Annual 2004, Volume 1.
DOE/EIA-0340(04)/1. June, 2005. Table 36.

CRS-5
Stock Exchanges in lower Manhattan.13 Approximately 22% of U.S.
security industry employees are located in New York City.14
! Defense industrial base (shipyards)—Over 31% of U.S. naval
shipbuilding and repair capacity is in and around Norfolk, VA.15
In addition to single infrastructure concentrations, some regions of the United
States contain concentrations of multiple critical infrastructures. As indicated in the
examples above, coastal Louisiana has concentrations of both refining and chemical
production capacity. In addition to a concentration of financial services, the
metropolitan New York and New Jersey area contains a concentration of U.S. port
capacity (12% of container shipping) and airport capacity (8% of airline passengers),
among other critical infrastructure.16
Geographic Hazards
Where critical infrastructure is geographically concentrated, it may be distinctly
vulnerable to a range of geographic hazards, including natural or unnatural events.
These events could have varying potential for infrastructure disruption depending
upon the type of event, its location, and the infrastructure sectors present in that
location. What such events have in common is their geographic scale. Among the
geographic events posing the greatest hazard to U.S. critical infrastructure
concentrations are the following.
Meteorological Events. Major meteorological events, such as hurricanes,
tropical storms, floods, and ice storms, have the potential to physically disrupt critical
infrastructures or displace related critical workers in large geographic areas. For
example, the damaging effects of hurricanes Katrina and Rita (and associated
flooding) on energy and chemicals infrastructure in the Gulf of Mexico have been
widely reported. In 1998, a major ice storm in Quebec, Canada and the northeastern
United States caused widespread, persistent power and communications blackouts,
disrupted other power-dependent services, and prevented critical workers from
traveling to their jobs.17
13 New York Stock Exchange. Annual Report 2004; American Stock Exchange. Annual
Report 2004.

14 Securities Industry Association (SIA). “Securities Industry Employment.” Sept. 2005. p. 9.
See [http://www.sia.com/research/pdf/NYMonthly.pdf].
15 Colton Company. “Employment in the Major Shipbuilders.” Oct. 26, 2005.
[http://www.coltoncompany.com/shipbldg/statistics/jobsbyyard.htm] Capacity estimate
based on 2003 shipyard relative employment data.
16 WCSC. Dec. 1, 2004; U.S. Department of Transportation, Bureau of Transportation
Statistics (BTS). Airport Activity Statistics (AAS) 2000. BTS01-05. 2002. Table 3.
17 Environmental Index. “The Storm of the Century in Canada, January 1998.” 2000.
[http://www.scor.com/astrehelp/en/Qualif/env/NATF02CA.htm]. Some 700,000 Canadians
were without power for over two weeks; U.S. Army Cold Regions Research and
Engineering Laboratory. “An Evaluation of the Severity of the January 1998 Ice Storm in
(continued...)

CRS-6
Earthquakes and Tsunamis. Earthquakes have the potential to damage
concentrations of critical infrastructure in seismically active regions of the United
States, including the west coast, Alaska, and the central Mississippi Valley. The
1994 earthquake in Northridge, CA, is a recent example of such seismic activity in
a region with concentrated critical infrastructure. The Northridge earthquake had
limited impacts on the region’s major ports, airports, and energy infrastructure, but
it did cause significant damage to bridges and highways vital for commercial trucking
and public transportation.18 A 1995 earthquake in Kobe, Japan was far more
destructive to Japanese critical infrastructure. In addition to highway damage, the
earthquake heavily damaged the port of Kobe, Japan’s largest container shipping
port, as well as chemical manufacturers, steel manufacturers, railroads, and utilities
in the area. Repairs to the port took almost a year to complete.19
Coastal infrastructure concentrations are also potentially vulnerable to
disruption by tsunamis. The infrastructure damage to Sri Lanka, India, Indonesia,
and other Asian nations from the 2004 tsunami in the Indian Ocean is still being
evaluated, but was extensive. Experts have testified before Congress that the United
States is also potentially vulnerable to a major tsunami.20 Depending upon its
magnitude, such an event could disrupt ports and other critical transportation
infrastructure.21 According to California’s Seismic Safety Commission, for example,
a major tsunami in southern California could close the ports of Long Beach and Los
Angeles for two months and cause $60 billion in economic losses.22
Infectious Disease. Epidemics and pandemics of infectious diseases such
as Severe Acute Respiratory Syndrome (SARS) and avian influenza (bird flu) have
the potential to disrupt critical infrastructure by infecting critical workers or
restricting their movement. The Bush Administration’s National Strategy for
Pandemic Influenza
states that “while a pandemic will not damage power lines,
banks or computer networks, it will ultimately threaten all critical infrastructure by
removing essential personnel from the workplace for weeks or months.”23 An
outbreak of infectious disease may sicken critical workers or force them into
quarantine. It may also restrict their access to critical facilities where the disease may
be present. As one federal government report states, during such an event
17 (...continued)
Northern New England: Report for FEMA Region 1.” April 1998.
18 M.G. Boarnet, “Business Losses, Transportation Damage, and the Northridge
Earthquake.” Journal of Transportation and Statistics, vol. 1, no. 2, May 1998.
19 Risk Management Solutions, Inc. 1995 Kobe Earthquake 10-year Retrospective. Newark,
CA. Jan. 2005. p. 5.
20 C. Groat, Director, United States Geological Survey (USGS). Testimony before the House
Science Committee hearing, Tsunamis: Is the United States Prepared?, Jan. 26, 2005.
21 J. Borrero, S. Cho, J.E. Moore II, H.W. Richardson, and C. Synolakis, “Could it Happen
Here?,” Civil Engineering. Apr. 2005. pp. 54-65.
22 California Seismic Safety Commission. The Tsunami Threat to California. CSSC 05-03.
Dec. 2005. p. 6.
23 Office of the President, Homeland Security Council. National Strategy for Pandemic
Influenza
. Nov. 1, 2005. p. 2.

CRS-7
“operations become disrupted, exposed people and facilities undergo extensive
testing ... and buildings and equipment require decontamination.”24
The 2003 SARS outbreak in Toronto demonstrated the vulnerability of critical
health and transportation infrastructure in Canada to such an infectious disease.
More recently, the World Health Organization, the U.S. Department of Health and
Human Services, and other health organizations have expressed concern about the
growing likelihood of a bird flu pandemic with more serious potential consequences
than SARS.25 In the event of a bird flu or similar outbreak in a particular geography,
some analysts have predicted up to 40% absenteeism among workers during the peak
weeks of a regional outbreak.26
Concentrations of livestock may be similarly vulnerable to infectious disease,
with the potential to catastrophically affect the nation’s food supply. As one expert
has testified before Congress, “animal diseases can be quickly spread to affect large
numbers of herds over wide geographic areas. This reflects the intensive and
concentrated nature of modern farming practices in the US.”27 Foot and mouth
disease (FMD), in particular, has the potential to infect regionally concentrated stocks
of hogs, cattle, and sheep should they be exposed. A 2002 General Accounting
Office found that an FMD outbreak could cost the U.S. economy up to $24 billion
dollars and could have “significant social impacts, such as enormous psychological
damage, especially on families and localities directly affected by the outbreak.”28
Terrorism. Certain types of terrorist attacks could be of sufficient scale to pose
a geographic threat to critical infrastructure. Nuclear bombs, radiological weapons
(“dirty” bombs), or electromagnetic pulse (EMP) devices could damage or render
inaccessible concentrated critical assets. Cyber-attacks on regional computer systems
also have the potential to damage or disrupt computer networks’ ability to control
critical infrastructure.29 Biological attacks could have impacts similar to those of
epidemics, although they could be more specifically targeted at particular regions.30
24 A. Brecher, U.S. Department of Transportation (DOT), Volpe National Transportation
Systems Center. “Cleanup and Recovery of Passenger Transportation Facilities after a
Bio-attack.” Workshop resource paper. Mar. 30, 2004.
25 For further background see CRS Report RL33145, Pandemic Influenza: Domestic
Preparedness Efforts
, by Sarah A. Lister; World Health Organization (WHO). Avian
Influenza: Assessing the Pandemic Threat
. WHO/CDS/2005.29. Jan. 2005.
26 U.S. Department of Health and Human Services. “Pandemic Influenza Planning.” Internet
pag. Dec. 5. 2005. See [http://pandemicflu.gov/plan/pandplan.html].
27 P. Chalk, RAND Corp. “The Bio-Terrorist Threat to Agricultural Livestock and Produce.”
Testimony before the Senate Government Affairs Committee. Nov. 19, 2003.
28 General Accounting Office. Foot and Mouth Disease. GAO-02-808. July 2002. pp20-21.
29 Weiss, J. “CyberWar.” Frontline. Public Broadcasting System. Television interview. Apr.
24, 2003.
30 For further discussion see Senate Judiciary Committee, Subcommittee on Terrorism,
Technology and Homeland Security hearing, Lessons Learned from Hurricane Katrina in
Regard to Emergency Preparedness for a Terrorist Attack
, Oct. 26, 2005.

CRS-8
Frequency of Major Geographic Events
Taken individually, the types of disasters discussed above occur only rarely in
a specific location. Taken collectively, however, such events occur often enough to
warrant dedicated policy attention. As Table 1 shows, reviewing only the past 15
years, major disasters have occurred in North America almost annually. Not all of
these events have impacted regions of concentrated critical infrastructure, nor have
they all significantly affected such infrastructure where it has been present.
Nonetheless, the cost estimates for these events indicate their disruptive power.
Table 1. Selected U.S. Disasters Since 1990 with Costs
Exceeding $1 Billion
Costs
Year
Event
Location
($ billions)
2005 Hurricanes (Category 4, 3)
Central Gulf of Mexico
70.0+
2004 Hurricanes (Category 2, 3, 4) Florida, Alabama
45.0
2003 Hurricane (Category 3)
Mid-Atlantic
3.4
2003 Epidemic (SARS)
Ontario
0.8*
2001 Terror attacks
New York, Virginia, Pennsylvania
38.0
2001 Tropical storm
Texas, Southeast U.S.
5.0
1998 Ice storm
Quebec, Northeast U.S.
1.4**
1997 Flood/Tornados
Ohio/Mississippi Valleys
1.0
1996 Hurricane (Category 3)
North Carolina
3.2
1995 Flood / Tornados / Hail
South Central U.S.
5.5
1994 Ice storm
Southeast U.S.
3.0
1994 Earthquake
California
26.0+
1993 Flood
Midwest
21.0
1992 Hurricane (Category 5)
Florida, Louisiana
26.5
Sources: National Hurricane Center. “The Thirty Costliest Mainland United States Tropical Cyclones
1900-2004.” Dec. 2005. [http://www.aoml.noaa.gov/hrd/tcfaq/costliesttable.html]; National Climatic
Data Center. “Billion Dollar U.S. Weather Disasters.” 2004. [http://lwf.ncdc.noaa.gov/oa/reports/
billionz.html#chron]; Dixon, L. and Stern, R.K., RAND Corp. “Compensation for Losses from the
9/11 Attacks.”Nov. 8, 2004; Holtz-Eakin, D., Congressional Budget Office (CBO). “Macroeconomic
and Budgetary Effects of Hurricanes Katrina and Rita.” Testimony before the House Budget
Committee. Oct. 6, 2005; Eyesenbach, G. “SARS and Population Health Technology.” Journal of
Medical Internet Research
, vol. 5, no. 2, 2005; J.L. Witt, “Comments by FEMA Director James Lee
Witt on the Third Anniversary of the Northridge, California Earthquake.” Jan. 17, 1997.
[http://www.fema.gov/library/wittspch5.shtm].
* Includes only direct costs for extra protective gear, clinics, isolation rooms, and lost wages for
quarantined health-care workers. Cost likely exceeds $1 billion with business losses included.
**Excludes Canadian costs.


CRS-9
Market Influences on Geographic Concentration
Although attention to the geographic concentration of U.S. critical infrastructure
has increased in the wake of recent terrorist attacks and natural disasters, such
geographic concentrations are not new. They have developed for multiple reasons
—typically some combination of market influences including resource location,
agglomeration economies, scale economies, community preferences, and capital
efficiency. These influences often have been in place for decades, gradually driving
critical infrastructure development to its geographic configuration today.
Resource Location. The location of certain critical infrastructures is driven
by the location of related natural resources or, in some cases, natural terrain. Such
influences are particularly apparent in energy, agriculture, and transportation. United
States oil and natural gas basins, for example, are located in particular regions of the
country, including the Gulf of Mexico, the Rockies, and Appalachia. These locations
are generally far from Northeastern urban centers which are the primary locations of
oil and gas demand. Consequently, large oil and gas pipelines tend to be
concentrated between these widely separated resource regions and the Northeast, as
shown in Figure 1. Likewise, production of phosphoric acid, a key component of
agricultural fertilizer, is concentrated in Florida, which has the nation’s largest
deposits of phosphoric rock.31 Agricultural production is also driven by geography,
since particular crops require particular climates, weather conditions, and types of
soil. Terrain may also be a driver of infrastructure concentration. There are
relatively few natural harbors suitable as deepwater ports in the western United States
compared to the eastern part of the country. Consequently, the ports that exist, such
as Long Beach, have become very heavily utilized. In these cases, and others, the
concentration of a natural resource drives the concentration of infrastructure
exploiting that resource.
Figure 1. Oil and Gas Pipelines in the Continental United States
Source: Energy Information Administration.
31 Environmental Protection Agency (EPA). “About Phosphogypsum.” Nov. 30, 2004. See
[http://www.epa.gov/radiation/neshaps/subpartr/more.htm].

CRS-10
Agglomeration Economies. Where resource location and terrain are not
constraints, concentrations of critical infrastructure may emerge due to economic
factors collectively referred to as “agglomeration economies.”
Broadly speaking, it is clear that [industry] concentrations form and survive
because of some form of agglomeration economies, in which spatial
concentration itself creates the favorable economic environment that supports
further or continued concentration.32

For critical infrastructure, such agglomeration economies may include the availability
of specialized knowledge, the availability of skilled workers, access to production
inputs, and access to large markets for the goods and services produced.33 The
concentration of semiconductor manufacturing in Silicon Valley illustrates such
economies. Silicon Valley emerged near major research institutions (e.g., Lawrence
Livermore National Laboratory), with the ready availability of highly-skilled
graduates from leading research universities (e.g., U.C. Berkeley, Stanford
University), and with access to both product suppliers, computer manufacturers (e.g.,
Apple Computer), and software companies. Agglomeration economies have also
been demonstrated in food manufacturing, and may similarly influence other critical
infrastructure sectors such as financial services, chemicals manufacturing, and
telecommunications.34
Scale Economies. Critical infrastructure may become geographically
concentrated in pursuit of “scale economies.”

Scale economies are found in industries where unit costs fall as the scale of
operations increases. This phenomenon was first studied in pipeline industries
... when it was observed that the amount of material required to make a pipe of
a given diameter increased only two-thirds as quickly as the carrying capacity of
the pipe. This observation led to larger pipes having lower unit costs.35
In addition to pipelines, researchers have identified scale economies across many
critical infrastructure sectors.36 The size of new chemical plants, for example,
increased by a factor of five between the late 1950s and early 1980s, in part due to
scale economies.37 Some analysts likewise suggest that the concentration of shipping
32 M. Fujita, P. Krugman, and A. Venables, The Spatial Economy. MIT Press. 1999. p. 3.
(Hereafter cited as Fajita, et al., The Spatial Economy.)
33 Fujita, et al., The Spatial Economy, p. 5.
34 J.P. Cohen, and C.J.M. Paul, “Agglomeration Economies and Industry Location
Decisions: The Impacts of Vertical and Horizontal Spillovers.” Working Paper 01-010.
University of California, Davis, U.S. Department of Agricultural and Resource Economics.
Oct. 2001.
35 P.W. Bauer, “Are We in a Productivity Boom? Evidence from Multifactor Productivity
Growth.” Economic Commentary. Federal Reserve Bank of Cleveland. Oct. 15, 1999.
36 L.R. Christensen, and W.H. Greene, “Economies of Scale in U.S. Electric Power
Generation.” Journal of Political Economy, vol. 84, no. 4, 1976, p. 655.
37 M.B. Lieberman, “Market Growth, Economies of Scale, and Plant Size in the Chemical
(continued...)

CRS-11
container traffic among several U.S. “megaports” is partly due to economies of scale
in warehousing and terminal operations.38 Because scale economies tend to drive an
increase in size of individual facilities, they may also geographically concentrate
regional infrastructure capacity where multiple facilities are located in the same
region.
Community Preferences. Community preferences have sometimes led to
concentrations of critical infrastructure by preventing or inviting the development of
new facilities in particular new locations. Such preferences have affected, for
example, ongoing efforts by energy developers to site new liquefied natural gas
(LNG) import terminals. Since 2000, developers have proposed the construction of
over 70 new LNG terminals in U.S. ports or U.S. waters, many near major natural gas
markets in California and the Northeast. But most near-to-market terminal proposals
have struggled for approval due to community concerns about LNG safety, effects
on local commerce, and other potential negative impacts. Due primarily to local
community opposition, LNG developers have withdrawn terminal proposals in
Alabama, California, Maine, Massachusetts, North Carolina, and Florida. Other
terminal proposals in Rhode Island, New York, and New Jersey are facing stiff
community opposition. In some cases state and local agencies also have been at odds
with federal agencies over LNG terminal siting jurisdiction. Communities in only
a few states, notably Louisiana and Texas, have encouraged the siting of new LNG
facilities. As a result, 9 out of 10 of the new LNG terminals approved by federal
agencies are located in the Gulf of Mexico, where natural gas infrastructure is already
heavily concentrated.39 Similar siting preferences have faced other types of critical
infrastructure and industries, including electric power, telecommunications, and
transportation.40
Capital Efficiency. Capital efficiency seeks to maximize financial returns on
capital investment. Since most U.S. critical infrastructure is in the private sector,
capital efficiency has long influenced how and where private companies have
invested in infrastructure capacity.41 Attention to capital efficiency sharply increased
in the 1990s, however, as financial markets grew dissatisfied with other measures of
37 (...continued)
Processing Industries,” Journal of Industrial Economics, vol. 36, no. 2, 1987, pp. 175-191.
38 National Research Council. Making the Nation Safer: The Role of Science and Technology
in Countering Terrorism
. National Academies Press. 2002. p. 216.
39 For further discussion of LNG siting issues see CRS Report RL32073, Liquefied Natural
Gas (LNG) Infrastructure Security: Issues for Congress
, by Paul W. Parfomak; and CRS
Report RL32205, Liquefied Natural Gas (LNG) Import Terminals: Siting, Safety and
Regulation
, by Paul W. Parfomak and Aaron M. Flynn.
40 D. Laws, and L. Susskind, “Changing Perspectives on the Facility Siting Processes.”
Maine Policy Review. Dec. 1991. pp. 29-44.
41 General Accounting Office (GAO). Challenges for Critical Infrastructure Protection.
GAO-03-233. Feb. 28, 2003. p. 1. This report states that over 80% of critical infrastructure
is private.

CRS-12
company performance such as simple revenue growth.42 This attention, in turn, led
a range of capital-intensive infrastructure companies such as electric utilities,
telecommunications providers, pipelines, and other industrial companies to sharply
reduce annual capital requirements from historical levels.43 Economic deregulation
of the energy, telecommunications, and transportation industries, among others,
accelerated this trend.
In some cases companies reduced capital requirements by cutting “excess”
infrastructure capacity—reducing reserve capacity in capital equipment or reducing
inventories of production supplies (“just-in-time” inventory). For example, power
generation capacity margins for the U.S. electric utility industry as a whole fell by
almost 40% between 1992 and 2000.44 There has been a similar reduction in excess
capacity among other critical infrastructures, many of which now operate near or at
capacity. Oil refineries, for example, have seen capacity utilization rise from below
77% in 1985 to 93% in 2004.45 While such a reduction in reserve capacity has not,
itself, led to geographic infrastructure concentration, it has greatly increased the
sensitivity of infrastructures to the disruption of concentrated capacity.
Federal Policies and Infrastructure Concentration
Although market forces have been the primary influence on critical
infrastructure development, especially in the private sector, Congress and federal
agencies historically have, from time to time, adopted policies intended to affect the
capacity and location of critical infrastructure in the national interest. Although these
policies often have been motivated by the desire to promote specific social objectives
(e.g., economic development, environmental protection, infrastructure reliability)
they have sometimes also encouraged or discouraged the geographic concentration
of critical infrastructures. Examples of these policies follow.
Prescriptive Siting
The federal government has prescriptively sited, constructed, and operated
federally-owned or operated critical infrastructure. Such infrastructure includes
transportation facilities, military bases, postal facilities, federal energy facilities, and
national laboratories. In some cases, such prescriptive siting has led to geographic
concentration of critical infrastructure. For example, the federal government sited
and constructed the Panama Canal in the early 1900s, encouraging a concentration
of military and marine shipping through the new waterway which persists today. (The
42 See, for example, S.R. Rajan, “Turning Capital to Wealth: A Ranking of U.S. Utilities.”
Public Utilities Fortnightly. Dec. 1999.
43 M. Singer, and K. Turnipseed, “Curing Capital Addiction.” McKinsey Quarterly. 1993.
no. 4. pp. 69-77.
44 T. Karier, “Keeping the Lights On: A Banking Industry Model to Avoid Shortages.”
Public Utilities Fortnightly. July 1, 2002.
45 Energy Information Administration. “U.S. Percent Utilization of Refinery Operable
Capacity.” Dec. 7, 2005. Table at [http://tonto.eia.doe.gov/dnav/pet/hist/mopueus2a.htm].

CRS-13
canal carried 48% of U.S. grain exports in 2003.)46 In the Pacific Northwest, dams
constructed or operated by federal agencies on the Columbia River system now
account for 22% of U.S. hydroelectric generation capacity.47 In other cases,
prescriptive federal siting has dispersed critical infrastructure. In the early 1940s, for
example, the federal government financed and sited a major steel plant in Utah, far
from existing U.S. steel plants and steel markets. A key reason for siting the plant
(viewed as critical for shipbuilding during World War II) in Utah was “as a
precaution against steel shortages in the West in case of a Pacific coast invasion or
closure of the Panama Canal.”48 A more current example is the U.S. Postal Service,
which routes no more than 1.3% of all mail through any single processing and
distribution center.49
Although the federal government prescriptively sites its own infrastructure, it
is difficult to find examples of federal prescriptive siting of private sector
infrastructure. One way the government has done so, however, is through its control
of federal lands and other federal assets necessary for infrastructure development.
For example, the Trans-Alaska Pipeline Authorization Act of 1973 (P.L. 93-153)
directed the Secretary of the Interior to authorize a right-of-way for construction of
the Trans-Alaska Pipelines System (TAPS) through federal lands in Alaska. The
construction of TAPS physically diversified U.S. oil supplies, although it initiated a
new geographic concentration of critical infrastructure in Alaska. TAPS transports
nearly 20% of United States domestic oil production.50 More recently, the Energy
Policy Act of 2005 (P.L. 109-58) directs federal agencies to designate “energy
corridors” on federal lands in 11 western states for the siting of new oil, gas, and
hydrogen pipelines and electricity transmission facilities (Section 368). While both
the TAPS and energy corridor projects involve privately-owned facilities, the
location of those facilities is established prescriptively by the federal government.
Economic Incentives
Economic incentives are another policy mechanism employed by the federal
government to direct private sector infrastructure siting. Such incentives are intended
to encourage private developers to build infrastructure that might otherwise not be
built or to build infrastructure in a location favored by government. While most
46 U.S. Army Corps of Engineers. Waterborne Commerce of the United States: Calendar
Year 2003.
Part 5—National Summaries. 2005. Table 2-1; Panama Canal Authority.
“Commodity Movement through the Panama Canal over Principal Trade Routes.” 2005.
Tables 10-11. See [http://www.pancanal.com/eng/maritime/statisti.html].
47 Energy Information Administration . Electric Power Annual 2003. DOE/EIA-0348(2003).
Dec. 2004. Table 2.2; Federal Columbia River Power System. Federal Columbia River
Power System.
Aug. 2003. p. 1. See
[http://www.bpa.gov/power/pg/fcrps_brochure_17x11.pdf].
48 A.K. Powell, “Geneva Steel Plant.” Utah History Encyclopedia. University of Utah Press.
1994.
49 U.S. Postal Service. Personal communication and non-public data. Dec. 7, 2005.
50 Alyeska Pipeline Service Co. Internet page. Anchorage, AK. Apr. 2005. See
[http://www.alyeska-pipe.com/about.html].

CRS-14
federal incentive programs are not geographically targeted, some are intended to
affect infrastructure development in particular geographic areas. The construction
of the transcontinental railroad and telegraph is an historic example of such a policy.
Under the Pacific Railway Act of 1862, the Congress provided private companies
with 30-year bonds, federal land grants, and other incentives to construct a rail and
telegraph line along a specified route from Nebraska to the Pacific coast. The federal
government continues to offer such financial incentives for critical infrastructure
projects today. For example, in 2004 Congress passed the Alaska Natural Gas
Pipeline Act (P.L. 108-324, Div. C) offering an $18 billion loan guarantee,
accelerated depreciation, and investment tax credits to private developers for the
construction of a new natural gas pipeline similar to the existing Trans-Alaska oil
pipeline. In both the railroad and the Alaska gas pipeline cases, Congress has viewed
the new infrastructure as critical for expanding and diversifying (geographically) the
nation’s critical assets.
Environmental Regulation
Federal environmental laws, such as the Coastal Zone Management Act and the
Clean Air Act, also have influenced the geographic development of critical
infrastructure. The Coastal Zone Management Act of 1972 (CZMA, P.L. 92-583)
was enacted to enable states to establish coordinated coastal zone management
programs balancing environmental protection with coastal development. State
coastal management plans implemented under the CZMA may affect the geographic
concentration of infrastructure by encouraging or discouraging the siting of coastal
infrastructure. Research has shown, for example, that one third of states with coastal
management plans under CZMA appear to have adopted policies seeking to confine
the physical expansion of ports to areas already committed to port and industrial
uses.51 Energy industry representatives have argued that state plans under CZMA
have also been used to block the development of new energy infrastructure in many
parts of the country.52
The Clean Air Act of 1970 (CAA, P.L. 91-604) created a national program to
mitigate the harmful effects of air pollution by regulating pollution sources. Among
other provisions, the CAA requires that new facilities emitting certain air pollutants
install best available control technology as determined by the Environmental
Protection Agency. New facilities being sited in counties not in attainment of federal
air quality standards may have more stringent—and potentially more costly—
emissions control requirements than facilities sited in counties that are in attainment
of those standards, depending upon a state determination of the lowest emission rate
available and the need to acquire emissions offsets. By affecting facility costs in this
way, some analysts argue that the CAA encourages the concentration of infrastructure
in geographic “pollution havens,” or, alternatively, encourages the dispersion of
facilities away from existing infrastructure in polluted regions. One empirical study
in New York, for example, suggests that air quality regulations have significantly
51 M.J. Hershman, “Seaport Development and Coastal Management Programs: A National
Overview.” Coastal Management, vol. 27, 1999, pp. 271-290.
52 Argonne National Laboratory. Environmental Policy and Regulatory Constraints to
Natural Gas Production
. ANL/EAD/04-1. Dec. 2004. p. 25.

CRS-15
affected the destination choices of relocating manufacturing plants.53 Because facility
siting decisions are complex, however, other empirical studies of CAA effects on
siting concentration, specifically, have been less conclusive.54
While the environmental regulations under the CZMA and CAA apply generally
to many regions in the United States, some federal environmental policies have been
directed at more specific geographic areas. Congressional moratoria on oil and
natural gas development in specific parts of the outer continental shelf (due to
concerns about local economic and environmental impacts) are one example of such
federal policy.55 Energy industry analysts have argued that the moratoria have
resulted in oil and gas infrastructure concentration in the central and western Gulf of
Mexico, where such development is permitted. This example notwithstanding, the
federal government does not appear to impose regional-level (as opposed to facility-
level) environmental restrictions frequently.56
Economic Regulation
Economic regulation of critical infrastructure, or the lack thereof, by the federal
government may also influence infrastructure concentration. Under federal price
regulation, the U.S. airlines industry offered primarily direct, point-to-point service.
After economic deregulation in 1978, the airlines began offering far more indirect
flights, routing air traffic through concentrated “hub” airports—a largely
unanticipated consequence of inter-carrier competition.57 Federal deregulation of
banking has led to a consolidation of the banking sector, with ever-larger banks
concentrating critical operations in centralized administration facilities to capture
economies of scale.58
Limited federal regulation does not necessarily lead to infrastructure
concentration, however, especially if state or local agencies have regulatory authority.
Under the Federal Power Act of 1935 (FPA), for example, retail electricity sales and
generation investments of investor-owned electric utilities are regulated by the states.
State regulators have historically required utilities to meet state electric generation
needs by constructing in-state plants, or by jointly constructing plants with
neighboring utilities. Consequently, privately-owned electric power plants have been
53 J.A. List, et al. “Effects of Air Quality Regulation on the Destination Choice of Relocating
Plants.” Oxford Economic Papers, vol. 55, no. 4, 2003, pp. 657-678.
54 T. Jeppesen, et al. “Environmental Regulations and New Plant Location Decisions:
Evidence from a Meta-Analysis.” Journal of Regional Science, vol. 42, no. 1, 2002, pp. 19-
49.
55 For further discussion see CRS Report RL31521, Outer Continental Shelf Oil and Gas:
Energy Security and Other Major Issues
, by Marc Humphries.
56 Environmental Law Institute, Institutional Controls in Use. Washington, D.C. Sept. 1995.
p. 14.
57 G. Gowrisankaran, “Competition and Regulation in the Airline Industry.” FRBSF
Economic Letter
. No 2002-01. Federal Reserve Bank of San Francisco. Jan. 18, 2002.
58 S.J. Pilloff, Bank Merger Activity in the United States, 1994—2003. Board of Governors
of the Federal Reserve System. Staff Study 176. May 2004.

CRS-16
geographically dispersed among the 50 states. The largest shares of U.S. generating
capacity in individual states are 10% in Texas, and 6% in California.59 By contrast,
the federally-owned Tennessee Valley Authority (TVA) had plans in the 1960s to
construct 17 nuclear power reactors at seven sites.60 Although TVA only completed
five reactors due to changes in the energy market and nuclear safety regulation, its
original plans would have created a nationally significant concentration of nuclear
generating capacity within TVA’s territory. It is interesting to note that the recent
restructuring of the electric utility industry, which exempts new generation plants
from state economic regulation, appears to be encouraging the geographic
concentration of new generating plants near certain transmission corridors because
plant developers are no longer constrained by state regulators in their site selection.61
Policy Options to Reduce Infrastructure
Vulnerability
Since helping to reduce the overall vulnerability of critical infrastructure is an
objective of the federal government, it is useful to outline what options, if any, may
be considered to reduce vulnerabilities and potential national consequences arising
specifically from the geographic concentration of such infrastructure.
Some analysts may argue that little government intervention in infrastructure
concentration is necessary because the private sector will appropriately adjust its
infrastructure practices out of its own financial interest. Catastrophic insurance
premiums, for example, or internal corporate risk management programs, may
influence corporate practices in a way that reduces vulnerabilities and associated risk
to future profits by reducing the geographic vulnerability of private infrastructure.
As the National Strategy for the Physical Protection of Critical Infrastructures and
Key Assets
states,
Customarily, private sector firms prudently engage in risk management planning
and invest in security as a necessary function of business operations and
customer confidence.... Consequently, private sector owners and operators
should reassess and adjust their planning, assurance, and investment programs
to better accommodate the increased risk....62
Holders of such a view would assert that the socially optimal geographic distribution
of critical infrastructure, balancing economic efficiency with geographic risk, is best
left to the market forces outlined earlier in this report.
59 Energy Information Administration. Electric Power Annual 2003. DOE/EIA-0348(2003).
Dec. 2004. Table 2.1.
60 McCullough, Jr, G.L., Chairman, Tennessee Valley Authority. Testimony before the
Senate Energy and Natural Resources Committee, Energy Subcommittee. Mar. 4, 2004.
61 For further information on utility restructuring see CRS Report RL32728, Electric Utility
Regulatory Reform: Issues for the 109th Congress
, by Amy Abel.
62 Office of the President. The National Strategy for the Physical Protection of Critical
Infrastructure and Key Assets
, Feb. 2003. p. X.

CRS-17
Other analysts argue that the private sector does not properly account for the full
social costs of critical infrastructure failure, or that individual companies cannot
independently and significantly influence geographic concentration in a critical
sector.63 Holders of this view would see a definite and active role for the federal
government in alleviating geographic vulnerability of critical infrastructure in
addition to the market-driven measures taken by the private sector on its own. If
Congress concludes that federal intervention is appropriate, it has several broad
policy options for doing so.
Eliminating Policies Encouraging Concentration
One way Congress may alleviate geographic concentration and associated
vulnerability is to eliminate existing policies that encourage such concentration. As
the previous discussion has shown, some federal policies may increase concentration
prescriptively. Others, especially certain economic and environmental policies, may
implicitly or unintentionally encourage geographic concentration. Without such
government influence, market forces may drive developers to less geographically
concentrated locations for future infrastructure projects. The challenge to this
approach of alleviating geographic concentration is that it may conflict with other
objectives of federal legislation. In the case of the economic deregulation, for
example, geographic concentration often provides the consumer cost reductions and
service improvements that deregulation was intended to achieve. In the case of
environmental laws, concentration is often viewed as a desirable means of preserving
undisturbed natural areas from destructive development. Would the CZMA be able
to fulfill its fundamental balance of environmental protection and economic
development if states were not free to concentrate infrastructure where they choose
to? The resolution of such policy questions would require a careful and complex
reconsideration of long-standing policy objectives in light of evolving concerns about
critical infrastructure risk.
Encouraging Geographic Dispersion
Another remedy for geographic vulnerability is to encourage the geographic
dispersion of concentrated assets where such dispersion is possible. As discussed in
this report, the federal government may implement a range of targeted policies,
including prescriptive siting, economic incentives, and regulation, to help bring about
infrastructure dispersion. Such dispersion could involve the development of new
infrastructure capacity or the shifting of critical goods and services among existing
infrastructure. Some transportation analysts, for example, have proposed shipping
containers through Mexican ports and then on to the United States by rail as a means
of reducing cargo traffic in Southern California’s ports.64 Shifting concentrations of
critical supplies and services to alternative infrastructure already in place (and not
itself concentrated) may be one way to alleviate geographical vulnerabilities
relatively quickly. If the alternative infrastructure lies outside the United States,
63 See, for example, P.R. Orszag, The Brookings Institution, “Homeland Security and the
Private Sector.” Testimony before the National Commission on Terrorist Attacks Upon the
United States. Nov. 19, 2003.
64 R.C. Mireles, “A Cure for West Coast Congestion.” Logistics Today. Jan. 2005.

CRS-18
however, such a strategy may create new vulnerabilities since it might no longer be
under U.S. protection or administration.
While encouraging infrastructure dispersion through federal policy may be
helpful, doing so may be challenging. It may be difficult, for example, to identify
and prioritize geographic infrastructure concentrations amenable to such dispersion.
Predicting the long-term effects of such polices on market economics, especially the
effects on market competition, may also be uncertain. If incentives are involved,
dispersion policies may also be costly to the federal government, potentially drawing
resources away from other federal programs. Furthermore, since infrastructure
development is mostly in private and regional government (state and local) hands,
ensuring that regional infrastructure projects are consistent with federal objectives
may also be a problem. As the Congressional Budget Office has stated,
The federal government’s most important role in infrastructure provision is as a
source of finance.... Thus, for infrastructure to be managed in a way that furthers
national objectives, federal agencies must offer incentives for local managers to
align their choices with the welfare and equity goals of federal programs.
Choices for infrastructure systems that aim at such broad objectives must
similarly be based on wide searches among new investments, rehabilitation
efforts, or operational changes. They must also be derived from consistent
evaluations of the long-term effects of these possible choices on the efficiency
of activities using the infrastructure.65
Infrastructure owners and regional governments are also likely to have vested
interests in existing concentrations of critical infrastructure and may oppose
dispersion on competitive or other economic grounds. State and local governments
may also have concerns about federalism, particularly where federal policies
affecting infrastructure dispersion may supercede local infrastructure priorities.
Notwithstanding the challenges of promoting infrastructure dispersion, Congress
appears to be pursuing such policies with respect to the siting of new energy
infrastructure, including LNG import terminals, oil refineries, electric transmission
lines, and an Alaska gas pipeline. Even where such federal policies may be
implemented successfully, however, it may still take years or decades to achieve
dispersion objectives because critical infrastructure often develops slowly. For
example, industry experts project that it would take five to seven years, absent
community opposition, to construct a new U.S. oil refinery.66 It would take at least
10 years to build the Alaska natural gas pipeline.67
65 Congressional Budget Office (CBO). Federal Policies for Infrastructure Management.
June 1986. p xi.
66 O’Conner, T., ICF Consulting. Testimony before the House Government Reform
Committee, Subcommittee on Energy and Resources. “Petroleum Refineries: Will Record
Profits Spur Investment in New Capacity?” Oct. 19, 2005.
67 Energy Information Administration. Annual Energy Outlook 2005. DOE/EIA-0383(2005).
Jan. 2005. p. 95.

CRS-19
Ensuring Infrastructure Survivability
For geographic concentrations of critical infrastructure that are difficult to
diversify, or that may take a long time to diversify, Congress may wish to ensure their
near-term ability to function, or “survivability,” during and after a major geographic
disaster. Particularly where resource location provides few geographic alternatives
(as in the case of ports) reducing vulnerability through infrastructure protection may
be effective. Such an approach would broadly align with the President’s existing
strategy for protecting critical infrastructure from terrorist attack as stated in the
NSPP, although it would incorporate explicitly geographic vulnerabilities. In the
context of the NSPP or other emergency management programs, geographic
vulnerability could be viewed as a distinctive type of infrastructure vulnerability and
therefore considered in federally mandated risk assessments.
Increasing standards for design, construction, and operation, and retrofitting
existing infrastructure to higher standards may also enhance infrastructure
survivability. While there are numerous industry and government building standards
for protection from earthquakes, hurricanes, and floods in regions where such hazards
exist, such standards may not account for the critical nature of certain types of assets.
In particular, the degree of general survivability these standards impose on critical
assets may not appropriately reflect the economic and social costs that might arise
should such an asset fail. Federal authority to change such standards may be limited,
however, if they fall primarily under state or local jurisdiction.
A principal challenge to alleviating geographic infrastructure vulnerabilities is
incorporating the geographic dimension appropriately into the broader infrastructure
risk management and decision-making process. One key question is whether
survivability measures in place to protect against a facility-specific event would be
effective against a regional event. Backup supply networks, redundant control
centers, and other systems intended to “harden” infrastructure may themselves be
subject to disruption from a geographic hazard. Although history does provide some
guidance as to the likelihood of disruptive natural events and their potential effects,
quantifying such geographic vulnerabilities in a way that allows comparison to a
broader set of vulnerabilities may be analytically complex. Predicting the likelihood
of future terrorist attacks, for which history provides little guidance, adds
considerable uncertainty. Attempting to allocate limited public resources for critical
infrastructure survivability based on geographic considerations may also be
challenging. In particular, it could complicate ongoing policy debates about using
quantitative, risk-based formulas to distribute federal support for critical
infrastructure protection.68 Increasing the private costs of infrastructure through new
construction standards to improve survivability could also be controversial.
Ensuring Infrastructure Recovery Capabilities
In addition to policies promoting geographic dispersion and survivability,
Congress may consider infrastructure recovery as a means of mitigating the impacts
68 For further discussion see CRS Report RL33050, Risk-Based Funding in Homeland
Security Grant Legislation: Issues for the 109th Congress
, by Shawn Reese.

CRS-20
of geographic hazards on concentrated critical infrastructure. The federal
government, through the Federal Emergency Management Agency (FEMA) and other
agencies, provides a range of emergency aid programs for communities affected by
disasters such as hurricanes, earthquakes, or terrorist attacks. Among other
assistance, these federal programs can provide grants, loans, loan guarantees, food,
and shelter to disaster victims. They may also provide long-term infrastructure
assistance, such as repair of public utilities, to affected communities.69
While the evolving objectives of federal emergency assistance programs are a
topic of ongoing debate in Congress, the programs are intended to assist primarily in
the recovery of an immediate disaster area.70 However, if a natural disaster, terrorist
attack, or other regional incident disrupts critical infrastructure, it may have serious
social or economic consequences far beyond the area where the disaster occurs. The
loss of concentrated natural gas supplies in the Gulf of Mexico after hurricanes
Katrina and Rita, for example, sharply increased U.S. energy prices and threatened
to create significant shortages of fuel for home heating and electric power generation
in New England.71 These natural gas shortages have prompted Congressional
proposals to increase federal aid for low income households nationwide facing high
natural gas bills this winter.72 In light of the far-reaching impacts like these,
Congress may wish to incorporate into existing federal infrastructure recovery plans
and aid programs measures that account for the distinctive vulnerabilities of
concentrated critical infrastructure. Measures related to the restoration or alternative
provision of critical infrastructure services away from the immediate area of a
geographic incident may warrant particular attention.
Conclusions
Geographic concentrations of critical infrastructure exist across a number
infrastructure sectors. Although such concentrations often provide substantial
economic and social benefits, they may also be distinctly vulnerable to catastrophic
geographic disruption. Any public policy addressing critical infrastructure
concentration must try to balance these benefits and potential costs. Both government
and industry have taken steps to try to protect critical infrastructure from natural
disasters, epidemics, and terrorist attacks. Nonetheless, questions remain as to
whether these steps appropriately address such geographic vulnerabilities. If
Congress concludes that more federal intervention is needed to alleviate
vulnerabilities due to geographic concentration, it may employ a number of policy
options to encourage geographic dispersion (including eliminating policies that
encourage concentration ), ensure survivability, or ensure that effective infrastructure
69 See CRS Report RL31734, Federal Disaster Recovery Programs: Brief Summaries, by
Mary Jordan.
70 For an overview of current policy debates regarding U.S. emergency preparedness, see
P.M. Prah, “Is the U.S. Ready for Another Major Disaster?,” CQ Researcher, vol. 15,
no. 41, Nov. 18, 2005.
71 R. Adams, “A Hard Winter, A Thin Natural Gas Pipeline,” CQ Weekly, Oct. 7, 2005.
72 Congressional Record - Senate. “LIHEAP” Nov. 18, 2005. p. S13339.

CRS-21
recovery capabilities are in place to mitigate impacts of concentrated infrastructure
disruption. Because geographic hazards exist today, and geographic dispersion
would likely take decades to achieve, addressing geographic vulnerabilities may call
for a combination of options.
In addition to these issues, Congress may assess how geographic infrastructure
vulnerability and survivability fit together in the nation’s overall infrastructure
policies. As Congress evaluates diverse proposals with the potential to affect critical
infrastructure development—directly or indirectly—Congress may consider whether
such proposals are likely to relieve or exacerbate geographic vulnerability. The
economic or social benefits of adding capacity (e.g. refinery, airport, shipping) to an
existing concentration of critical infrastructure, or developing additional
infrastructure in a new location, may be outweighed by the increased geographic risk
implicit in such an expansion. Fiscal implications, especially related to the economic
efficiency of public critical infrastructure and the efficient use of federal funds for
infrastructure projects, may also be an important consideration. Reviewing how such
infrastructure priorities fit together could be an oversight challenge for Congress