Review on Energy Resilience

Economic resilience to transportation failure: a computable general equilibrium analysis

Zhenhua Chen1 • Adam Rose2

Published online: 22 September 2017 � Springer Science+Business Media, LLC 2017

Abstract This study develops and applies a multimodal computable general equilibrium (CGE) framework to investigate the role of resilience in the economic consequences of

transportation system failures. Vulnerability and economic resilience of different modes of

transportation infrastructure, including air, road, rail, water and local transit, are assessed

using a CGE model that incorporates various resilience tactics including modal substitu-

tion, trip conservation, excess capacity, relocation/rerouting, and service recapture. The

linkages between accessibility, vulnerability, and resilience are analyzed. The model is

applied to the transportation system failures in the aftermath of Hurricane Katrina to

illustrate its capabilities. The analytical framework, however, has broader applications and

can provide insights for resource allocations to enhance emergent responses to unexpected

events and to improve resilient design of transportation infrastructure systems.

Keywords Transportation � System failure � Economic resilience � Accessibility � Computable general equilibrium (CGE) modeling � Hurricane Katrina

Introduction

Transportation infrastructure plays a critical role in facilitating economic growth and

development. However, with the increasing number of unexpected events, including both

natural disasters such as earthquakes and hurricanes, and man-made threats like terrorism

attacks, critical infrastructure systems are facing unprecedented risks and have become

much more vulnerable. Negative economic consequences due to a transportation

& Zhenhua Chen [email protected]

1 City and Regional Planning, Knowlton School of Architecture, The Ohio State University, Columbus, OH, USA

2 The National Center for Risk and Economic Analysis of Terrorism Events (CREATE), Sol Price School of Public Policy, University of Southern California, Los Angeles, CA, USA

123

Transportation (2018) 45:1009–1027 https://doi.org/10.1007/s11116-017-9819-6

infrastructure failure stemming from various extreme events can be large. Conversely, the

negative consequences may be reduced due to the inherent and adaptive resilient mech-

anisms of transportation systems, such as modal substitution, conservation, excess

capacity, and rerouting. Given the coexistence of the negative consequences and various

resilience effects, the extent to which a transportation system failure may ultimately affect

the economy remains unclear. The issue is still not fully understood for the following three

reasons.

First, although there is an increasing interest among scholars in assessing the economic

consequence of transportation system failure empirically, few analyses consider the

economy-wide ramifications. Even fewer have utilized the computable general equilibrium

modeling approach. Some studies, for example Wright and Hogan (2008) analyzed the

physical impact of a natural disaster, such as sea level rise on transportation infrastructures,

using digital elevation models, whereas other studies, such as Koetse and Rietveld (2009)

only qualitatively discussed the economic consequences of transportation system failure

due to natural disasters.

Second, a plethora of studies measured the economic resilience of transportation

infrastructure based on hypothetical threat scenarios using various simulation approaches,

whereas only a few studies were conducted using real-world data. This is not surprising

given that information regarding transportation system failure is normally very limited.

Rose (2015) indicated that studies based on hypothetical or simulated data are likely to

provide more optimistic estimation of resilience than those based on actual disaster data

because the former has not incorporated various obstacles to the implementation of resi-

lience tactics and strategies. Among these exceptions, Cox et al. (2011) examined the

resilience of the transit system in the aftermath of the London subway and bus disruption

due to a terrorist attack in 2005. The post-event travel demand for various transit systems

including bus, rail, taxi, motorcycle, cycle and walking were estimated using time series

analysis. The study found that the resilience of the transportation system was about 77.4%,

primarily due to the effect of modal substitution.

Third, most economic resilience studies of transportation failure were investigated

from a unimodal perspective with a focus on each specific mode, such as airport, road,

seaport, or public transit. For instance, Snelder et al. (2012) analyzed vulnerability of

road networks using a robustness analysis. Rose et al. (2014) and Chen et al. (2017)

evaluated the economic consequences of aviation system disruption due to a terrorist

attack, in which resilience was measured as mode substitution for travel related

expenditure using a computable general equilibrium (CGE) model. Cox et al. (2011)

quantified resilience for the London transit bombings of 2005. Although it analyzed

other modes in terms of their ability to offer resilience for subway and bus disruption,

this study has a limited focus on subways and buses that were attacked. These studies

are pioneers in providing in-depth assessments with considerations of specific resilience

factors related to the mode of focus. However, they are still limited in that the system-

wide resilience, such as trip conservation, substitution among various modes of trans-

portation, excess capacity, relocation/rerouting, and service recapture, were ignored. In

addition, the lack of a modal comparative assessment may constrain the understanding

of the economic vulnerability and resilience of the transportation system as a whole,

which may consequently bias decision-making for planning and investment related to

critical transportation infrastructure protection.

1010 Transportation (2018) 45:1009–1027

123

Our study investigates the economic consequence of transportation system failure from

a multi-modal perspective using a CGE model. 1

For the first time, vulnerability and

economic resilience of the different modes of transportation, including air, road, rail, water

and local transit are assessed and compared in a real word scenario based on the aftermath

of Hurricane Katrina. The research is developed by extending the CGE framework of Rose

et al. (2009, 2014). The model enables us to conduct a comprehensive assessment of

vulnerability and resilience of different modes of transportation through a detailed pro-

duction function nesting structure. The results provide insights to understand the interac-

tions among accessibility, vulnerability, and resilience of critical infrastructures, which

will be useful for resource allocations to enhance emergent responses to unexpected events

and to improve resilient design of transportation infrastructure system.

The rest of the paper is organized as follows. Section 2 discusses the theoretical

motivation through a literature review. Section 3 introduces economic resilience with

respect to transportation systems. Section 4 discusses the methodology and data, whereas

Sect. 5 introduces the analytical framework and interprets the results. Section 6 discusses

the linkages between accessibility, vulnerability, and resilience. Section 7 summarizes the

analysis.

Literature review

Literature on economic resilience of transportation systems is burgeoning in recent years

given the increasing attention to critical infrastructure protection in the U.S. and among

other countries. Several studies assessed the issue based on scenarios of major disruptions

using either econometric analysis (Schlake et al. 2011) or CGE analysis (Xie et al. 2014;

Rose et al. 2014; Shi et al. 2015). Foster (1993) characterized robustness, which is a major

aspect of reliability and risk management, as a system’s ability to accommodate variable

and unexpected conditions without catastrophic failure, or ‘‘the capacity to absorb shocks

gracefully.’’ Overall, there are two views of Resilience. Rose (2009) and Cox et al. (2011)

confine resilience to actions taken after the disaster strikes to maintain system function and

to recover quickly. On the other hand, Bruneau et al. (2003) view a resilient system as one

that exhibits reduced probability of failure and negative consequences, as well as reduced

time to recover, which thereby includes pre-disaster actions such as mitigation. Thus, their

framework includes robustness, redundancy, resourcefulness and rapidity as features of

resilience.

In terms of the measurement of resilience in transportation, various approaches have

been developed. For instance, Ta et al. (2009) suggested that the system performance

resilience of transportation can be measured in terms of reliability, travel time, speed, and

vehicle counts. Tamvakis and Xenidis (2012) discussed the parameters affecting trans-

portation resilience and proposed a framework for measuring resilience from an engi-

neering perspective. Leu et al. (2010) estimated the resilience for three types of ground

transportation systems in Melbourne, including train, tram and street car, using network

analysis. The resilience of different transportation system was computed based on the

interdependency and interactions of various transportation networks. Similarly, Dorbritz

1 ‘‘Failure’’ often means total shutdown or complete cessation. The more general case is partial shutdown—

disruption. We consider the influence of Hurricane Katrina on transportation infrastructures in Louisiana as a failure because the systems were completed shutdown for days. See additional evidence provided by Smyrlis (2005).

Transportation (2018) 45:1009–1027 1011

123

(2011) investigated the topological and operational disaster resilience of railway and

tramway networks in Switzerland. Similarly, Reed et al. (2009), Chen and Miller-Hooks

(2012) and Miller-Hooks et al. (2012) assessed resilience of freight transport systems as an

aspect of network’s inherent ability. Resilience was measured as a system performance

optimization problem with considerations of topological and operational attributes. Hence,

one should note that resilience in all these analyses is essentially a system performance

measure, in which resilient ability is measured in technical network performance from an

engineering perspective. The primary goal is to enhance the pre-event capability of a

transportation system to withstand a disaster (or robustness).

Economic resilience, on the other hand, measures the ability of a system to recover from

a disruption in terms of post-event actions and their effectiveness. The focus is on the most

effective resilient strategies from both the demand and the supply sides to reduce the

economic losses in terms of Gross Domestic Product (GDP) and employment after a

disaster/disruption. In a transportation system, economic resilience can be implemented by

several tactics such as modal substitution, conservation of service supply, excess capacity,

rerouting, and etc. These various tactics can be applied by many decision makers within a

transportation system, whether suppliers, such as mass transit providers and public

transportation authorities, or customers, such as businesses and individuals (Cox et al.

2011). Economic resilience differs from system performance resilience in that the former

evaluates the effectiveness of various resilience tactics to reduce economic losses, whereas

the latter primarily addresses mitigation strategies for the infrastructure system through

optimized solutions to improve network topology and system design. Furthermore, eco-

nomic resilience is less concerned with system output than it is with what the services

contribute to the economy. For example, if the pre-disaster level of production can be

obtained by reduced transportation services, via telecommuting or walking, then economic

resilience is achieved (Rose 2009; Cox et al. 2011).

In addition, transportation resilience and vulnerability have also been examined

extensively from the perspective of regional science, with focuses on the roles of con-

nectivity and accessibility in relation to transportation security (Reggiani 2013). For

instance, Östh et al. (2014) analyzed economic resilience and accessibility from a spatial

perspective and found a positive and significant correlation between resilience and

accessibility. Reggiani et al. (2015) indicated that transportation resilience is more difficult

to measure than transportation vulnerability because the latter is more concrete in real-

world networks with links, areas and morphological features. They suggested that one

possible measurement is to identify resilience/vulnerability characteristics using various

topological network metrics such as clustering, degree centrality and correlated networks.

In fact, such a measurement and analysis is termed ‘‘spatial economic resilience’’ by

Modica and Reggiani (2014), and the concept has both linkages to the notions of stability

(engineering resilience) and the idea of adaptivity (ecological and economic resilience).

Our study differs from the previous studies in the following aspects. First, our assess-

ment of resilience is evaluated from a post-disaster perspective, rather than a pre-disaster

perspective. Second, our assessment focuses on economic resilience rather than system

performance resilience. Third, the economic resilience of a multimodal transportation

system is evaluated in the context of the aftermath of Hurricane Katrina, which is a real-

world event rather than a hypothetical scenario.

1012 Transportation (2018) 45:1009–1027

123

Economic resilience

Basic considerations

Resilience has become a popular concept, but its use has been confused by the large variety

of definitions that have recently arisen. In this paper we focus on one aspect—Economic

Resilience. Following Rose (2007, 2009), this concept is derived from more fundamental

resilience definitions:

• Static resilience is the ability to maintain the functioning of a system, while the economic counterpart is utilizing remaining resources as efficiently as possible in order

to maintain function. This reflects the core of the economic problem on how to best use

scarce resources.

• Dynamic resilience refers to the ability and speed of recovery, where the economic counterpart pertains to investing wisely in repair and reconstruction. This is a dynamic

consideration from the standpoint of economics because it involves a time trade-off—

diverting resources for investment represents setting aside current consumption to

enhance productivity at future times.

Note that the definitions are couched in terms of functionality, typically measured in

economics as the flow of goods and services, such as GDP or broader measures of human

well-being, as opposed to property damage. It is not the property (capital stock) that

directly contributes to economic welfare but rather the flows that emanate from these

stocks for either businesses or households, and, for this paper, transportation services. Two

things should be kept in mind. First, while property damage takes place at a point in time,

the reduced flow, often referred to business interruption (BI), just begins at the time of the

disaster but continues until the system has recovered or has attained a ‘‘new normal.’’

Second, the recovery process, and hence the application of resilience, depends heavily on

the behavior of economic decision-makers and on public policy. Since resilient ability

implies a level of attainment will be achieved, the definitions of economic resilience are

contextual, suggesting that the level of function has to be compared to the level that would

have existed had the ability been absent. This means a reference point must be established

for the economic consequence analysis of resilience. In the case of static economic resi-

lience it refers to the worst case outcome. In the case of dynamic resilience it refers to a

normal recovery path. Further discussion of this oft-neglected point is provided below.

Following Rose (2009) resilience pertains to the economy at three levels:

• Microeconomic (individual business or household). • Mesoeconomic (individual industry or market). • Macroeconomic (combination of all economic entities, including their interactions).

At the micro-level, on the business supplier side, static economic resilience includes

redundant systems, improved delivery logistics, and planning exercises. Several options

also exist on the business customer side. An important distinction is between inherent and

adaptive resilience (see Rose and Liao 2005). The former refers to aspects of resilience

already built into the system, such as the availability of inventories, excess capacity,

substitutability between inputs (travel modes), and contingent contractual arrangements

accessing suppliers of goods from outside the affected area (imports). Resilience capacity

can be enhanced through these means and is then accessed after the disaster. Adaptive

resilience arises out of improvisation under stress, such as conservation of ordinary travel

Transportation (2018) 45:1009–1027 1013

123

services by walking to work, changes in the way goods and services are produced, and new

contracting arrangements that match customers who have lost their suppliers with suppliers

who have lost their customers. 2

At the mesoeconomic level, resilience can bolster an industry or market and include, for

instance, industry pooling of resources and information and innovative pricing mecha-

nisms. What is often less appreciated is the inherent resilience of market prices that act as

the ‘‘invisible hand’’ to guide resources to their best allocation in the aftermath of a disaster

(see, e.g., Horwich 1995). Some pricing mechanisms have been established expressly to

deal with such a situation, as in the case of a non-interruptible service price premium that

enables customers to estimate the value of a continuous supply of electricity and to pay in

advance for receiving priority service during an outage. The price mechanism is a rela-

tively costless guide to redirect goods and services. Price increases, to the extent that they

do not reflect ‘‘gouging,’’ serve a useful purpose of reflecting highest value use, even in the

broader social setting.

At the macroeconomic level, resilience is very much influenced by interdependencies

between sectors. Consequently, macroeconomic resilience is not only a function of resi-

lience measures implemented by single businesses, but it is also determined by the actions

taken by all individual companies and markets including their interactions. Examples of

resilience options at the macro-level would be economic diversity to buffer impacts on

individual sectors or geographic proximity to economies not affected by a disaster in order

to facilitate access to goods or aid. Others include fiscal (e.g., infrastructure spending to

boost the affected economy) and monetary policy (e.g., keeping interest rates low to

stimulate private sector reinvestment).

An operational metric

Following Rose (2007, 2009), we provide an admittedly crude but operational metric of

resilience. Direct Static Economic Resilience (DSER) refers to the level of the individual

firm or industry (micro and meso levels) and corresponds to what economists refer to as

‘‘partial equilibrium’’ analysis, or the operation of a business or household entity itself.

Total Static Economic Resilience (TSER) refers to the economy as a whole (macro level)

and would ideally incorporate what is referred to as ‘‘general equilibrium’’ effects, which

include all of the price and quantity interactions in the economy, as in this paper), and

potentially more broadly to macro-aggregate considerations and the ramifications of fiscal,

monetary and security policy.

An operational measure of DSER is the extent to which the estimated direct output

reduction deviates from the likely maximum potential reduction given an external shock,

such as the curtailment of some or all of a critical input. In essence, DSER is the percentage

avoidance of the maximum economic disruption that a particular shock could bring about.

In this paper, we apply this metric to the direct resilience of transportation systems and the

economy in terms of their effect on GDP. This is in contrast to a narrower application to

transportation resilience in terms of the number of trips by Cox et al. (2011).

Studies have found resilience to be especially effective—being able to reduce business

interruption losses by 70% or more (see, e.g., Rose et al. 2007; Kajitani and Tatano 2009;

Rose et al. 2009; Rose and Wei 2013). However, nearly all of these studies have simulated

resilience rather than having estimated it for an actual disaster. Hence, they are likely to be

2 See Rose (2009) on how these various tactics relate to the basic concept of the economic production

function.

1014 Transportation (2018) 45:1009–1027

123

somewhat optimistic, as they have not incorporated various obstacles encountered during

the implementation of resilience tactics and strategies.

Modeling economic resilience

Table 1 summarizes some major aspects of modeling resilience in transportation systems.

Five major types of resilience and the extent to which they do or do not apply to 9

transportation modes with separations of freight and passenger services are illustrated and

compared. The first refers to modal substitution and lists the possibilities and limitations.

Conservation is defined in this context in relation to telecommuting, bicycling and walking,

Table 1 Resilience tactics and measurements in transportation systems

Model Substitution Conservation Excess Capacity

a Relocation/ rerouting

Recapture e

Modeling procedure

Substitution elasticities

c Productivity

parameters Loosen

constraint d

Loosen constraint

d Recapture factor

f

Air (freight) All freight modes

n.a. Yes Yes All commodities except perishables

Air (passenger)

All passenger modes except light rail

Telecommuting Yes Yes Yes

Rail (freight) All freight modes except light rail

n.a. Limited b

Limited b

All commodities except perishables

Heavy rail (passenger)

All passenger modes except light rail

Telecommuting Limited b

Limited b

All commodities except perishables

Light rail (passenger)

Highway Telecommuting; bicycling; walking

Limited b

Limited b

Yes, except limited for some sector work trips

Highway (freight)

All freight modes

n.a. Limited b

Yes for urban; otherwise limited

All commodities except perishables

Highway (passenger)

All passenger modes; ship limited

Telecommuting; bicycling

Limited b

Yes for urban; otherwise limited

Yes, except limited for some sector work trips

Ship (freight)

Air; others limited

n.a. Yes Yes All commodities except perishables

Ship (passenger)

Very limited Telecommuting Yes Yes Yes

a Refers to excess equipment, route capacity, or route networks that are all for easy rerouting

b Limited because of fixed routes

c Refers to inherent substitution; adaptive substitution requires an increase in substitution elasticities

d Limited in a subset of modes

e Limited possibilities

f Also decays over time

Transportation (2018) 45:1009–1027 1015

123

because they reduce the number of formal transportation services required. 3

Excess

capacity, which is similar to redundancy, refers to the existence of unused conduits and

transportation equipment. Relocation/rerouting is self-explanatory. Recapture refers to the

ability to make up lost production at a later date by working overtime or extra shifts. 4

Note

that this resilience tactic decays over time as customers become impatient and seek

alternative suppliers.

Table 1 also presents a summary of how various resilience tactics are modeled in the

context of a production function receiving a shock from a disaster. Substitution is modeled

by increasing the elasticity of substitution between transportation modes and between the

transportation aggregate and other inputs. Conservation is modeled by adjusting the pro-

ductivity parameter of the production function. Excess Capacity and Relocation/Rerouting

can readily be incorporated by loosening the constraint on the system, i.e., lessening the

shock. Recapture can be applied by a scalar adjustment factor. See Rose et al. (2007) and

Rose and Wei (2013) for examples of the application of these resilience adjustments.

Methodology and data

Computable general equilibrium (CGE) analysis is the state-of-the-art approach to

macroeconomic simulation modeling. It is a multi-market model of the behavioral

responses of individual producers and consumers to changes in prices, technology, taxes,

and other external shocks, subject to the constraints on capital, labor, natural resources

(Dixon and Rimmer 2002). Essentially, CGE characterizes the economy as a set of

interconnected supply chains. It represents a significant advance over its predecessor,

input–output (I–O) analysis, by maintaining I-O model’s strengths, such as fully

accounting of all inputs, multi-sector details, focusing on interdependencies. It also

overcomes the limitations of linearity, lack of behavioral content, and omission of markets

and prices (Rose 1995). CGE models have been used extensively to model various types of

disasters (see, e.g., Dixon et al. 2011; Sue Wing 2015; Rose 2015), including considera-

tions of resilience (see, e.g., Rose and Liao 2005; Rose et al. 2007).

An enhanced version of the USCGE model is adopted for this assessment. 5

The model

consists of 58 producing sectors, along with multiple institutions: nine household income

groups, three government actors (two federal and one state and local), and external agents

(i.e. foreign producers). One of the most recent applications of the USCGE model was the

estimation of economic consequences of aviation system disruptions in the U.S. (Chen

et al. 2017).

To translate the various direct impacts caused by the transportation system failure to

relevant CGE shock requires identifying appropriate exogenous variables in the USCGE

model for each direct impacts driver. In our methodology, ordinary direct impacts in terms

of infrastructure damages are modeled as capital stock reductions. Labor shocks as a result

3 Bicycling and walking conserve on the number of trips on formal transportation modes (which relate to

transportation service providers, both private and government) that have to be made, which is consistent with the definition of economic resilience. We reserve the term substitution for shifts between formal modes. 4

Excess capacity promotes production recapture, and is also a separate resilience tactic (similar to system redundancy). 5

The model was originally developed by Oladosu and Rose for environmental policy analysis (Rose and Oladosu 2002) and for consequence analysis of terrorism events (Rose et al. 2009).

1016 Transportation (2018) 45:1009–1027

123

of death and injuries are not included in this study because they were not extensive enough

to affect the service ability of the transportation system itself. The change in capital is

measured in a constant elasticity of substitution (CES) form as in Eq. (1):

Qi ¼ aklK rkl�1 rkl

i þ ð1 � aklÞL rkl�1 rkl i

� � rkl rkl�1

ð1Þ

where Qi denotes the output of sector i, K and L denotes capital and labor input in sector i.

akl and rkl represent share parameter and the constant elasticity of substitution. 6

Figure 1

illustrates the nesting structure of the model, which is a revision based on Rose et al.

(2009), disaggregated to reflect substitution among different transportation modes.

The behavioral impacts caused by a direct change of output are modeled through the

change the price of final demand in the form of the factor of productivity parameter

adjustment. The determination of the price of final demand is denoted as:

PDMDfi;i ¼ dfi;i � PDMD0fi;i � X inpt

shi;inpt PDMDinpt;i

PDMD0inpt;i

� �1�ri;fi ! 11�ri;fi ð2Þ

where PDMD is the demand price; fi and inpt are composite factor inputs, with fi repre-

senting the upper level in a nest and inpt representing the lower level in a nest (e.g. KELM

(the capital, energy, labor and materials nest) is fi to the inpt of KEL (capital, energy and

labor nest) and MAT (material nest)); d is the factor of productivity, set to 1 across all nests and with respect to all sectors by default, except where changed at the KELM level of the

nesting structure for the purposes of modeling technology change.

International trade is represented through an Armington substitution function between

imports and domestic production (meaning that imports and good produced in the U.S. are

imperfect substitutes). A constant elasticity of transformation function represents the

substitution between exports and domestic sales. Input and import substitution elasticity

values have been sourced and checked against the literature on these parameters. 7

Household consumption is represented by a Linear Expenditure System of aggregate

commodities. A Social Accounting Matrix of the U.S. national economy in Year 2012 from

IMPLAN (2012) forms the empirical core of the model.

Nevertheless, the USCGE model is subject to the standard limitations of most CGE

models. First, for the most part, it assumes the economy is in equilibrium, though we do

incorporate disequilibrium in the labor market (unemployment equilibrium). Second, the

model is static, so that it does not trace the time-path of impacts, including various

economic cycles associated with employment and investment changes. Third, the model

construction is based on a deterministic approach on the basis of a single base year of data

(in contrast to the superior approach of econometric models, which use time series data and

have goodness of fit measures).

The direct impact drivers of the transportation failure in our assessment are calculated

based on the damage statistics of Hurricane Katrina in 2005 (GAO 2006). In particular, the

U.S. Federal Highway Administration estimated that Hurricane Katrina resulted in about

$2.1 billion in highway damages, most of which locate in Louisiana (GAO 2006). The

6 It refers to the depiction of a hierarchal decision-making process. The nested structure provides the reader

with an illustration with regard to transport in Level 4 and Level 3. 7

An example of the estimation of the Transportation nest substitution elasticities can be found in Avetisyan (2012).

Transportation (2018) 45:1009–1027 1017

123

damages to railroad and airport infrastructure were found to be over $40 million and $141

million, respectively. Public transit systems, especially in New Orleans, suffered a con-

siderable amount of losses as most of the transit fleet, bus garages, operation and main-

tenance facilities were severely damaged. The estimates found that at least $47 million

were necessary to make basic transit services recovered in the affected regions. The

damages to port facilities were also substantial, as several ports along the Gulf Mexico,

including the Port of New Orleans, the Port of Gulfport, the Port of Mobile and the Port of

South Louisiana, were all severely damaged by Hurricane Katrina with an estimate of total

loss at $822 million (GAO 2006).

To measure the economic consequence of transportation system failure due to Hurricane

Katrina, the national share of capital loss for each transportation mode was calculated

using the deflated direct impact numbers, which were estimated based on GAO (2006). All

the numbers are measured in 2012 U.S. dollars to be consistent with the CGE modeling

framework. The basic shock values of capital loss for different transportation mode are

illustrated in Table 2 8 . Specifically, the national share change of capital loss on air, road,

rail, water and other transportation modes including public transit are -0.85, -5.51 -0.25,

-22.94 and -0.19, respectively, all of which are measured in percentage terms.

In addition, the resilience effects related to substitution can be modeled through an

adjustment of the elasticity of substitution for factor inputs. A lower substitution elasticity

suggests a higher cost penalty to the economy when a policy shock is implemented in the

CGE model, which in turn amplifies the impact results. On the other hand, a higher

elasticity value indicates a lower cost when factor inputs are substituted during the pro-

duction process when a negative shock is simulated. Hence, a relatively lower economic

consequence is expected.

Conservation is measured through the adjustment of productivity parameter in the CGE

model. This is based on the authors’ judgement that conservation improves the efficiency

of productivity by 0.1%. Since the modeling effect works opposite to the change of the

Level 1

Level 2

Level 3

Level 4

Level 5

Level 6

L = Labor T = Transport FS = Finance Air = Air Transport K = Capital S = Services CM = Chemicals Road = Road Transport E = Energy M = Materials OM = Other Materials Rail = Rail Transport FUEL= Fossil Fuels M1 = Materials Sub Water = Water Transport ELEC = Electricity Others = Other Transport Services

KELM

KEL M

S M1 T

FS OS CM OM Air Road

KE

ELEC

E

FUEL

K

L

Intermediate Goods

Rail Water Others

Fig. 1 USCGE production nesting structure

8 Again, our analysis focuses on static resilience only, i.e., we evaluate the adaptive ability to utilize

remaining resources available as efficiently as possible to maintain function. Investment related to the implementation of these resilience tactics is beyond the scope of the analysis.

1018 Transportation (2018) 45:1009–1027

123

parameter, an increase of productivity by 0.1% corresponds to the adjustment of the

parameter to 0.999.

Economic consequence and resilience analysis for transportation failure

Basic considerations

We now apply our methodology to analyze transportation failure due to Hurricane Katrina.

As used by Reggiani et al. (2015), vulnerability has a negative connotation, which is in

contrast to resilience. The concept can be interpreted as the other side of the coin with

respect to resilience, i.e., the overall reduction of a system’s quality as a consequence of an

external shock. Based on this definition, the vulnerability of a transportation system in our

analysis is measured based on simulations without considering resilience, whereas resi-

lience is measured by taking various resilience tactics into account. We simulate the impact

of transportation system failure in Louisiana on the U.S. national economy in the aftermath

of this major disaster.

In Table 3, we present basic data for the Louisiana transportation system in relation to

that of the U.S. as a whole in terms of sectoral value added (GDP). 9

The combination of

‘‘Other Transport’’ modes is the largest of the five modes in the state. This consists of both

Public and Private Transit and Ground Passenger Transportation and Other Transportation

and Support Activities. It excludes Pipeline, as it is unlikely to be affected much by the

wind and flood damage that ensued from Katrina. The next largest transportation sector is

‘‘Road’’ (measured as highway freight transportation), followed by Water and then Rail

and Air, for a total GDP value of $5.4 billion (in $2005). The relative prominence of the

sectors differs significantly from that of the U.S. as a whole, and is presented in row one of

the table. The share of Louisiana transportation as a portion of the U.S. for each travel

Table 2 Adaptive resilience adjustments to reduce transportation failure impacts

Simulation Capital shock (%) a

Substitution elasticity

Productivity parameter

Air Road Rail Water Others

Base scenario (no resilience) -0.85 -5.50 -0.25 -22.94 -0.19 r = 0.25 d = 1

Resilience scenarios

Substitution -0.85 -5.50 -0.25 -22.94 -0.19 r = 1 d = 1

Conservation (0.1% productivity increase)

-0.85 -5.50 -0.25 -22.94 -0.19 r = 0.25 d = 0.999

Excess capacity (5% capital shock reduction)

-0.81 -5.22 -0.23 -21.79 -0.18 r = 0.25 d = 1

Relocation/rerouting (10% capital shock reduction)

-0.77 -4.95 -0.22 -20.64 -0.17 r = 0.25 d = 1

Recapture (scalar scaling) -0.72 -4.67 -0.21 -19.50 -0.16 r = 0.25 d = 1

Aggregate transport resilience -0.60 -3.85 -0.17 -16.06 -0.18 r = 1 d = 0.999

a Shock in context of the U.S. economy in relation to direct impacts in Louisiana from Hurricane Katrina

9 Due to data limitations, our assessment measures the national economic impact of transportation system

failure after Hurricane Katrina. One should also note that a regional level impact assessment may be more relevant once such data become available.

Transportation (2018) 45:1009–1027 1019

123

mode is presented in the first row. Not surprisingly, Water transportation has the largest

share. The share values are fundamental, as they are utilized to gauge the impacts of

transportation resilience to Hurricane Katrina in Louisiana on the nation as a whole in the

CGE assessment.

Our resilience scenarios are summarized in Table 2, and follow the various resilience

tactics presented in previous section. It also highlights the various modifications that we

make to our model and analysis to estimate the impacts of resilience. These modifications

are as follows:

• Transportation mode adaptive substitution is modeled by increasing the basic elasticity of substitution value of 0.25 to a value of 1.0.

• Conservation of transportation systems is assumed to be 5% and is modeled by increasing the productivity of each transport mode by 0.1%.

• Excess Capacity is assumed to be 5% and is modeled by reducing the capital stock in each transportation sector by 5%.

• Rerouting of transportation of each mode by 10% is modeled by a capital shock reduction of that amount.

• Recapture of lost production is modeled by a scalar adjustment of the gross output results downward by 22.5%.

The Excess Capacity adjustment is based on U.S. Department of Commerce estimates,

and Production Recapture is based on recapture factors in Federal Emergency Management

Agency (FEMA)’s hazard loss estimation software, HAZUS (FEMA 2015). The trans-

portation elasticity and conservation parameter changes are based on estimates derived in

Rose and Liao (2005). The rerouting of transportation by mode is based on the author’s

judgement that a 10% loss of transportation services can be offset by optimizing trans-

portation rerouting.

To implement the parameter adjustments, it is necessary to adjust the base estimates in

Table 3 to reflect the resilience only in Louisiana to capital stock damage. This is done by

multiplying the appropriate parameter values in Table 3 by the Louisiana-U.S. share for

each travel mode.

Our analysis proceeds on a comparative static basis, as we simulate the disruption and

resilience to that disruption for each of our five transportation sectors in turn. A simul-

taneous analysis of them all is also conducted to reflect interactions and synergies.

Results without resilience

The base case results of our analysis (in the absence of resilience) are presented in Table 4,

and indicate the vulnerability of the transportation system and the entire U.S. economy to

transportation failures in the aftermath of Hurricane Katrina. The table has two partitions,

Table 3 Basic economic data for transportation sectors in the U.S. and Louisiana, 2005

Economic indicator Air Road Rail Water Others All modes

National GDP (billions of dollars) 60.0 116.8 28.1 9.3 113.0 327.2

Louisiana GDP (billions of dollars) 0.4 1.5 0.4 0.8 2.3 5.4

Louisiana share of US GDP for transportation sectors (%)

0.7 1.3 1.4 8.6 2.0 1.7

1020 Transportation (2018) 45:1009–1027

123

where the top portion represents impacts in percentage terms for the value of output, while

the lower partition presents the results in percentage terms for the quantity of output. We

focus here on the quantity impacts, as they represent the most straightforward interpre-

tation, because they do not consider offsetting price changes that obscure the basic dis-

ruption impacts.

Not surprisingly, given the relative prominence of Water transportation of Louisiana in

relation to the U.S. as a whole (recall Table 3), the impacts on the Water transportation

sector are the largest of all sectors in terms of percentage of the value of output (GDP). In

interpreting the results, it is best first to consider diagonal elements in the table. These

consist primarily of the impact of a disruption to that sector on itself, and are thus largest of

any impact from a disruption to that sector. Note, however, that these diagonal figures also

show general equilibrium effects on the sector, but that these are relatively minor. The off-

diagonal elements represent only general equilibrium effects (on sectors not being directly

shocked). The results indicate that some general equilibrium effects can be significant.

Table 4 also presents the aggregate impacts on the U.S. economy in terms of percentage

reduction in the value of GDP. Interestingly, although the Water Transportation sector is

relatively the most prominent and incurs the greatest sectoral loss, the Road transportation

sector imposes the largest loss on the U.S. economy. Note that all the percentage reduc-

tions in U.S. GDP are relatively small, less than one-half of one percent for each mode and

not much more than that in the aggregate. One should also note that the Louisiana economy

represents only about 0.6% of that of the U.S. Of course, these percentage impacts are

much larger if we simply place them in the context of the state alone.

Results with resilience

Table 5 presents the simulation results for resilience for each transportation mode in the

aftermath of Hurricane Katrina in reducing losses to the U.S. economy. Again, we perform

Table 4 Economic consequences for the U.S. economy due to transportation system failure following Hurricane Katrina without resilience (%)

Transport mode Air Road Rail Water Others All modes

Output value

Air -0.192 -0.008 0.002 0.081 -0.003 -0.120

Road -0.001 0.589 0.000 -0.006 -0.001 0.582

Rail 0.009 -0.039 0.050 0.037 -0.001 0.056

Water 0.019 0.004 0.002 -11.421 -0.020 -11.414

Others -0.021 -0.178 -0.001 -0.451 0.022 -0.629

Total transport -0.039 0.144 0.005 -0.546 0.006 -0.430

Output quantity

Air -0.477 0.004 0.001 0.052 -0.004 -0.424

Road 0.000 -2.048 -0.001 0.005 -0.004 -2.048

Rail 0.005 -0.019 -0.096 0.016 -0.001 -0.095

Water 0.014 0.004 0.001 -15.100 -0.022 -15.100

Others -0.015 -0.134 -0.001 -0.328 -0.037 -0.515

Total transport -0.086 -0.773 -0.010 -0.640 -0.015 -1.525

Aggregate impact on GDP (value added) -0.003 -0.046 -0.001 -0.016 -0.001 -0.067

Transportation (2018) 45:1009–1027 1021

123

T a b le

5 S

im u

la ti

o n

o f

im p

a c ts

o n

U .S

. G

D P

fr o m

tr a n

sp o

rt a ti

o n

sy st

e m

fa il

u re

w it

h c o

n si

d e ra

ti o

n s

o f

v a ri

o u s

re si

li e n

c e

ta c ti

c s

fo ll

o w

in g

K a tr

in a

(p e rc

e n t

re d u

c ti

o n

in im

p a c ts

)

T ra

n sp

o rt

m o d e

A ir

R o a d

R a il

W a te

r O

th e rs

A ll

m o d e s

A ir

R o a d

R a il

W a te

r O

th e rs

A ll

m o d e s

S u b st it u ti o n

C o n se rv a ti o n

A ir

0 .0

0 .3

0 .0

0 .0

0 .2

0 .0

0 .4

0 .3

1 .0

0 .8

1 6

.0 0

.7

R o

a d

0 .2

0 .2

- 1

.3 -

0 .5

2 .8

0 .2

0 .2

0 .3

3 .2

0 .4

9 .6

0 .3

R a il

0 .0

0 .8

0 .2

0 .1

2 .9

- 0

.4 0

.5 0

.2 1

.0 0

.8 9

.5 1

.2

W a te

r 0

.0 1

.6 0

.1 0

.0 0

.0 0

.0 0

.4 0

.2 0

.8 0

.6 1

6 .1

0 .7

O th

e rs

0 .1

0 .5

1 .5

0 .0

0 .1

0 .2

0 .3

0 .3

- 1

.4 0

.7 1

1 .7

- 0

.1

A ll

m o d

e s

0 .0

0 .0

0 .0

0 .0

0 .0

0 .0

0 .4

0 .3

1 .2

0 .6

1 0

.9 0

.5

T o

ta l

G D

P 0

.0 0

.0 0

.0 0

.0 0

.0 0 .0

1 .0

0 .5

1 .5

1 .3

2 4

.1 1 .5

E x c e ss

c a p a c it y

R e lo c a ti o n /r e ro u ti n g

A ir

4 .7

5 .5

8 .0

5 .9

5 .3

3 .9

9 .4

1 0

.8 1

2 .0

1 1

.7 1

0 .5

7 .9

R o

a d

4 .7

5 .8

8 .3

5 .7

5 .2

5 .8

9 .4

1 1

.3 1

2 .5

1 1

.3 1

0 .3

1 1

.3

R a il

4 .8

5 .1

8 .1

5 .8

4 .6

8 .1

9 .6

9 .9

1 2

.1 1

1 .5

9 .3

1 2

.8

W a te

r 4

.7 4

.5 7

.5 5

.0 5

.3 5

.0 9

.4 8

.9 1

1 .3

1 0

.1 1

0 .6

1 0

.1

O th

e rs

4 .7

5 .2

8 .2

5 .3

5 .3

5 .3

9 .4

1 0

.2 1

2 .4

1 0

.6 1

0 .5

1 0

.5

A ll

M o

d e s

4 .7

6 .0

8 .1

5 .1

5 .3

4 .7

9 .4

1 1

.7 1

2 .1

1 0

.2 1

0 .6

9 .6

T o

ta l

G D

P 4

.7 5

.4 8

.0 5

.6 5

.3 5 .5

9 .5

1 0

.6 1

2 .0

1 1

.1 1

0 .5

1 0 .7

R e c a p tu re

A g g re g a te

re si li e n c e

A ir

2 2

.5 2

2 .5

2 2

.5 2

2 .5

2 2

.5 2

2 .5

3 4

.6 3

5 .5

3 5

.7 3

6 .5

4 7

.1 3

3 .7

R o

a d

2 2

.5 2

2 .5

2 2

.5 2

2 .5

2 2

.5 2

2 .5

3 4

.6 3

5 .9

3 7

.8 3

5 .5

4 1

.2 3

5 .9

R a il

2 2

.5 2

2 .5

2 2

.5 2

2 .5

2 2

.5 2

2 .5

3 4

.9 3

4 .7

3 5

.9 3

6 .4

4 1

.8 3

6 .7

W a te

r 2

2 .5

2 2

.5 2

2 .5

2 2

.5 2

2 .5

2 2

.5 3

4 .7

3 4

.1 3

4 .9

3 4

.7 4

7 .4

3 4

.7

O th

e rs

2 2

.5 2

2 .5

2 2

.5 2

2 .5

2 2

.5 2

2 .5

3 4

.7 3

5 .0

3 4

.2 3

5 .3

4 4

.9 3

4 .9

A ll

m o d

e s

2 2

.5 2

2 .5

2 2

.5 2

2 .5

2 2

.5 2

2 .5

3 4

.6 3

6 .3

3 6

.0 3

4 .8

4 4

.6 3

4 .2

T o

ta l

G D

P 2

0 .4

2 1

.9 2

1 .9

1 9

.6 2

1 .5

2 1 .3

3 3

.4 3

4 .8

3 5

.6 3

3 .9

5 2

.8 3 4 .9

B o

ld v

a lu

e s

in d

ic a te

th e

e ff

e c t

o f

th e

c o

rr e sp

o n

d in

g re

si li

e n

c e

ta c ti

c o

n m

u ti

n g

th e

G D

P lo

ss m

e a su

re d

in p

e rc

e n

t c h

a n

g e

1022 Transportation (2018) 45:1009–1027

123

comparative static analyses distinguishing the impact of each resilience tactic on each

transportation mode individually, and holding all other basic disaster impacts constant.

Positive numbers in the table represent offsets to losses incurred in the absence of

resilience.

The analysis of adaptive substitution indicates this tactic has a relatively small effect on

muting economic losses, despite the four-fold increase in the substitution elasticity across

transportation modes. The largest effects for this tactic stem from the Other Transportation

sector, as the shock has caused substantial increases in outputs of Road and Rail sectors

Conservation also has a relatively small effect on muting losses, and also has the largest

improvement stemming from the Other Transportation sector. Excess Capacity resilience

does not vary much from the basic 5% level that was simulated for each of the modes,

except Rail Transportation. Rerouting also does not vary much from the 10% rate simu-

lated. These latter two results indicate that general equilibrium effects are not extensive for

these tactics. Production recapture resilience is applied equally to all transportation modes

on the order of 22.5% directly. The calculation is applied to the sectoral total impact

results, which is why there are all equal. However, the impact of recapture on total GDP

varies across transportation modes, because recapture factors differ across non-trans-

portation sectors and because both the shock and other types of resilience differ across

sectors as well.

The aggregate impacts, which are modeled through simultaneous CGE simulation,

indicate that resilience can reduce business interruption losses, as measured by GDP, by

nearly 35%. Interestingly, the results in percentage terms do not differ much across modes,

except, again for the Other Transportation sector. One should note that in the case of

relocation and rerouting, the Rail system is found to have a relative larger effect, which

might be due to the fact that the accessibility and connectivity of the rail system were less

impaired during Hurricane Katrina. Conversely, Air transportation is found to have a

relative smaller effect from relocation and rerouting, which may confirm the rigidity of Air

transportation connectivity in this region.

In addition, the aggregate impacts on other economic sectors due to a transportation

system failure were evaluated with consideration of various economic resilience tactics. As

illustrated in Table 6, the simulation shows that resilience also has positive impacts on

reducing the GDP (value-added) losses of other sectors. The impacts are found to be

relatively larger in manufacturing sectors than agriculture and service sectors.

Accessibility

Although our assessment of transportation system failure focuses on vulnerability and

economic resilience, accessibility of transportation infrastructure is embedded in the

modeling framework in various ways. There are several ways to define transportation

accessibility. The traditional concept of accessibility is generally defined as ‘‘the ease with

which activities can be reached from a certain place and with a certain system of transport’’

(Morris et al. 1979). More recently, the definition has evolved to a much broader scope,

which can also be defined in relation to facilitating the use of transportation by either

physically disabled person or by economically disadvantaged persons (Östh et al. 2015).

Accessibility is considered from additional standpoints in our assessment, which measures

the general availability of transportation following a disaster. In particular, two types of

general system accessibility are considered:

Transportation (2018) 45:1009–1027 1023

123

Table 6 Value loss of U.S. GDP by sector from transportation system failure (with resilience)

Sector Value change

a Percent change

Sector Value change

a Percent change

Beef cattle and farming -15.9 31.8 Private Transit 5.9 b

34.5

Dairy cattle and farming -13.4 31.8 Communications -209.0 35.0

Other livestock -9.0 31.9 Information -65.2 37.3

Poultry and egg production -6.0 31.8 Private Electric Utilities

-80.7 34.8

Fishing and aquaculture -1.2 31.8 Gas utilities -25.1 34.7

Other agriculture -37.0 32.0 Private water utilities (retail)

-0.3 34.4

Coal -5.1 31.6 Sanitary services -0.3 34.4

Crude oil and natural gas -167.4 36.3 Wholesale trade -236.5 34.6

Other mining -10.0 37.2 retail trade -312.6 34.5

Construction 261.6 b

34.1 Real estate -643.0 35.0

Fluid milk manufacturing -7.9 36.8 Finance banking and credit

-341.5 35.0

Non-milk dairy product Mfg -7.1 36.8 Security brokers -213.2 34.8

Animal slaughtering and meat processing

-12.2 36.7 Insurance -278.2 34.8

Poultry processing -10.9 36.8 Owner occupied dwellings

-449.6 35.1

Seafood product -1.6 36.7 Hotel and restaurants -191.5 28.8

Other food processing -68.9 36.7 Personal services -37.7 26.9

Chemicals -193.0 36.8 Veterinary services -5.3 35.1

Petroleum refining -251.2 36.1 Waste manage -96.5 34.7

Other non-durable Mfg -168.7 36.7 Other business services

-884.7 35.0

Primary metals -49.4 36.8 Entertainment -58.2 28.8

Ordnance -1.2 36.6 Education -91.0 28.8

Electronics -54.4 38.2 Medical services -294.0 28.8

Other durable Mfg -370.1 36.9 Other health services -75.0 28.8

Air transport -51.6 33.7 Electric utilities -5.5 28.6

Truck transport 554.3 b

35.9 Local public transport

-1.0 34.3

Water transport -1050.3 34.7 Water utilities (wholesale)

-0.8 28.3

Rail transport 14.8 b

36.7 Other government -65.4 28.3

Other transport -559.0 34.9 State-local government

-25.3 28.3

The results were based on a CGE simulation incorporating simultaneous shocks to all modes of trans- portation and various resilience tactics a Measured in millions of 2012 dollars

b The positive increases in GDP (value-added) for some sectors are likely due to a combination of the

stimulus effects from inherent substitution across sectors

1024 Transportation (2018) 45:1009–1027

123

Regional accessibility This dimension relates to the ability to travel either short or long

distances, and is primarily related to both the usability of transportation networks and the

location of origins and destinations. Hence our analysis of relocation discussed above is

pertinent. The simulation results indicate that Regional Accessibility after Hurricane

Katrina was offset by resilience by 10.7%, which was predominantly due to the effect of

relocation/rerouting. This is understandable, as regional accessibility improves when

business and workers relocate out of affected area. Similarly, rerouting also improves

Regional Accessibility, though it may take a relatively longer time for passengers to travel

and freight to be delivered than before the threat of Hurricane Katrina.

Modal accessibility This pertains more to a substitution among various modes, and

related strategies such as conservation in terms of telecommuting or walking, which reduce

the number of trips needed in the first place. Of course, some aspects of relocation/

rerouting are pertinent here, but they are likely to be much less prominent than in the case

of Regional Accessibility.

This type of accessibility can be modeled by modifying the elasticity of substitution

between different transportation modes. As such it might be considered as a subset of

adaptive resilience. However, a conflict might arise in that the objective of dynamic

resilience is to recover more quickly. The objectives of accessibility relate to other con-

cerns. They are more obviously related to equity, for example, in the cases of the physi-

cally disabled or economically disadvantaged. They can achieve the goals of efficiency and

equity when they involve facilitating the access of low-income workers that help them get

to their place of employment.

Several additional steps can advance this analysis. The first is to distinguish substi-

tutability on a sectoral basis. Some commodities can readily be transported by all modes,

primarily lighter weight items, such as electronic equipment and pharmaceuticals. Other

commodities, especially heavier/lower value ones, such as minerals and agricultural

commodities, are not good candidates for air transportation. Moreover, substitutability is

also limited by geography, especially with regard to water transportation. All of these

aspects can be conceptualized by modifying the basic nesting structure and substitution

elasticities across sectors, though the greater challenge is in the empirical specification.

Conclusion

Our study indicates that economic resilience tactics can be very effective in reducing

disaster losses in transportation sectors of the economy. We have used relatively conser-

vative estimates of potential resilience for each of the tactics. Our economic impact

estimates are far lower than those of several other studies for transportation, other sectors,

and the economy as a whole (Rose et al. 2007; Kajitani and Tatano 2009; Cox et al. 2009).

One reason for the relatively lower impacts and resilience in our study is that our scaled-

down resilience parameters are intended to take into account various real-world obstacles

in the implementation of resilience, while most prior studies simply assumed that the

maximum potential resilience would be achieved.

One should note that the results presented are preliminary in the sense that some

anomalies, such as negative resilience impacts for some modes in the substitution and

conservation cases, were observed, which may be due to the simulation of overwhelming

shocks. Because of limitations of data, we caution the reader to take our results as generally

Transportation (2018) 45:1009–1027 1025

123

indicative, but not necessarily definitive. In addition, the results might differ for other

disasters in other locations. It is clear that real-world information, such as data collected

from surveys, will be valuable to calibrate the model more accurately.

Future research could be pursued in several directions. First, the existing static CGE

model could be upgraded to a dynamic model by enabling capital accumulation and labor

market equilibrium combined with a long-run closure rule, which would thus capture the

dynamic economic resilience of transportation systems. Second, the current single region

model could be expanded into a multiregional model, which would be able to capture the

impacts from the change of regional accessibility in various forms. In addition, the spatial

economic resilience of different modes of transportation could be further assessed using a

Spatial Econometric CGE (SECGE) model with the integration of spatially estimated

elasticities of factor substitution, as suggested by Chen and Haynes (2015a, b).

Acknowledgements This material is based upon work supported by the U.S. Department of Homeland Security under Grant Award Number 2010-ST-061-RE0001-05. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. Department of Homeland Security. The authors are grateful for the excellent assistance from Joshua Banks, and Noah Miller. An early version of this paper was presented at the Eighth NECTAR International Conference held at Ann Arbor, Michigan in 2015. Any errors or omissions are the sole responsibility of the authors.

References

Avetisyan, M.: Essays on the International Trade Impacts of Climate Policy. Unpublished Ph.D. dissertation, Department of Economics, Purdue University, West Lafayette, IN (2012)

Bruneau, M., Chang, S.E., Eguchi, R.T., Lee, G.C., O’Rourke, T.D., Reinhorn, A.M., von Winterfeldt, D.: A framework to quantitatively assess and enhance the seismic resilience of communities. Earthq. Spectra 19(4), 733–752 (2003)

Chen, L., Miller-Hooks, E.: Resilience: an indicator of recovery capability in intermodal freight transport. Transp. Sci. 46(1), 109–123 (2012)

Chen, Z., Haynes, K.: Spatial impact of transportation infrastructure: a spatial econometric CGE approach. In: Nijkamp, P., Rose, A., Kourtit, K. (eds.) Regional science matters—Studies dedicated to Walter Isard, pp. 163–186. Springer, Berlin (2015a)

Chen, Z., Haynes, K.: Multilevel assessment of public transportation infrastructure: a spatial econometric computable general equilibrium approach. Ann. Reg. Sci. 54(3), 663–685 (2015b)

Chen, Z., Rose, A., Prager, F., Chatterjee, S.: Economic consequences of aviation system disruptions: a reduced- form computable general equilibrium analysis. Transp. Res. Part A Policy Pract. 95, 207–226 (2017)

Cox, A., Prager, F., Rose, A.: Transportation security and the role of resilience: a foundation for operational metrics. Transp. Policy 18(1), 307–317 (2011)

Federal Emergency Management Agency: HAZUS. http://www.fema.gov/hazus (2015) IMPLAN: Social accounting matrix for the U.S. National Economy. http://implan.com/ (2012) Kajitani, Y., Tatano, H.: Estimation of lifeline resilience factors based on empirical surveys of Japanese

industries. Earthq. Spectra 25(4), 755–776 (2009) Koetse, M.J., Rietveld, P.: the impact of climate change and weather on transport: an overview of empirical

findings. Transp. Res. Part D Transp. Environ. 14(3), 205–221 (2009) Leu, L., Abbass, H., Curtis, N.: Resilience of ground transportation networks: a case study on Melbourne.

Paper delivered at the 33rd Australasian Transport Research Forum Conference held in Canberra, 29 September–1 October (2010)

Miller-Hooks, E., Zhang, X., Faturechi, R.: Measuring and maximizing resilience of freight transportation networks. Comput. Oper. Res. 39(7), 1633–1643 (2012)

Modica, M., Reggiani, A.: Spatial economic resilience: overview and perspectives. Netw. Spat. Econ. 15(2), 211–233 (2014)

Morris, J.M., Dumble, P.L., Wigan, M.R.: Accessibility indicators for transport planning. Transp. Res. Part A Gen. 13(2), 91–109 (1979)

Östh, J., Reggiani, A., Galiazzo, G.: Spatial economic resilience and accessibility: a joint perspective. Comput. Environ. Urban Syst. 49, 148–159 (2015)

1026 Transportation (2018) 45:1009–1027

123

Reed, D.A., Kapur, K.C., Christie, R.D.: Methodology for assessing the resilience of networked infras- tructure. IEEE Syst. J. 3(2), 174–180 (2009)

Reggiani, A.: Network resilience for transport security: some methodological considerations. Transp. Policy 28, 63–68 (2013)

Reggiani, A., Nijkamp, P., Lanzi, D.: Transport resilience and vulnerability: the role of connectivity. Transp. Res. Part A Policy Pract. 81, 4–15 (2015)

Rose, A.: Economic resilience to natural and man-made disasters: multidisciplinary origins and contextual dimensions. Environ. Hazards 7(4), 383–395 (2007)

Rose, A.: Economic Resilience to Disasters. CARRI Research Report #8. ORNL (2009) Rose, A.: Macroeconomic consequences of terrorist attacks: estimation for the analysis of policies and rules.

In: Mansfield, C., Smith, V.K. (eds.) Benefit transfer for the analysis of DHS policies and rules, pp. 172–201. Edward Elgar, Cheltenham (2015)

Rose, A., Liao, S.: Modeling resilience to disasters: computable general equilibrium analysis of a water service disruption. J. Reg. Sci. 45(1), 75–112 (2005)

Rose, A., Wei, D.: Estimating the economic consequences of a port shutdown: the special role of resilience. Econ. Syst. Res. 25(2), 212–232 (2013)

Rose, A., Oladosu, G., Liao, S.: Business interruption impacts of a terrorist attack on the electric power system of Los Angeles: customer resilience to a total blackout. Risk Anal. 27(3), 513–531 (2007)

Rose, A., Oladosu, G., Lee, B., Asay, G.B.: The economic impacts of the September 11 terrorist attacks: a computable general equilibrium analysis. Peace Econ. Peace Sci. Public Policy 15(2), 1–28 (2009)

Rose, A., Avetisyan, M., Burns, W., Slovic, P., Rosoff, H., Chan, O.: The role of behavioral responses in the total economic consequences of terrorist attacks on U.S. air travel targets. Presented at the Annual Meeting of the Society for Integrated Risk Management, October (2014)

Schlake, B.W., Barkan, C.P., Edwards, J.R.: Train delay and economic impact of in-service failures of railroad rolling stock. Transp. Res. Rec. J. Transp. Res. Board 2261(1), 124–133 (2011)

Shi, Y., Jin, S., Seeland, K.: Modeling business interruption impacts due to disrupted highway network of Shifang by the Wenchuan earthquake. Nat. Hazards 75(2), 1731–1745 (2015)

Snelder, M., Van Zuylen, H.J., Immers, L.H.: A framework for robustness analysis of road networks for short term variations in supply. Transp. Res. Part A Policy Pract. 46(5), 828–842 (2012)

Smyrlis, L.: Transportation infrastructure is decimated by Katrina. http://www.trucknews.com/features/ transportation-infrastructure-is-decimated-by-katrina/ (2005). Accessed 28 Dec 2015

Ta, C., Goodchild, A., Pitera, K.: Structuring a definition of resilience for the freight transportation system. Transp. Res. Rec. J. Transp. Res. Board 2097, 19–25 (2009)

Tamvakis, P., Xenidis, Y.: Resilience in transportation systems. Proc. Soc. Behav. Sci. 48, 3441–3450 (2012) U.S. Government Accounting Office: Hurricane Katrina GAO’s preliminary observations regarding pre-

paredness, response, and recovery: statement of David M. Walker comptroller general of the United States. GAO-06-442T. Washington, DC (2006)

Wright, K.M., Hogan, C.: The potential impacts of global sea level rise on transportation infrastructure. In: ICF International Prepared for the U.S. DOT Center for Climate Change and Environmental Fore- casting (2008)

Xie, W., Li, N., Li, C., Wu, J.D., Hu, A., Hao, X.: Quantifying cascading effects triggered by disrupted transportation due to the Great 2008 Chinese Ice Storm: implications for disaster risk management. Nat. Hazards 70(1), 337–352 (2014)

Zhenhua Chen is an assistant professor in City and Regional Planning at the Knowlton School of Architecture at The Ohio State University. His research interests include infrastructure planning and policy, resilience and big data. He is particularly interested in integrating these fields through regional impact assessment with a primary focus on transportation infrastructure system. The objective of his research is to improve the understanding of resilience, the effectiveness of infrastructure investment on regional economic development, and to provide policy implications that aim to improve the allocative efficiency of public resources to achieve a sustainable development.

Adam Rose is a research professor in the Sol Price School of Public Policy and research affiliate of the Center for Risk and Economic Analysis of Terrorism Events (CREATE), University of Southern California. His major research interest is economic consequences of and resilience to terrorism and natural disasters. His other major research area is energy and climate change mitigation policy. His work is widely published and has received several major awards. He has served as an advisor to local, regional and national governments in several countries and for several international organizations.

Transportation (2018) 45:1009–1027 1027

123

Reproduced with permission of copyright owner. Further reproduction prohibited without permission.