Applying the natural disasters vulnerability ... - Wiley Online Library

doi:10.1111/disa.12069

Applying the natural disasters vulnerability evaluation model to the March 2011 north-east Japan earthquake and tsunami Mario Arturo Ruiz Estrada Faculty of Economics and Administration, University of Malaya, Kuala Lumpur, Malaysia, Su Fei Yap Faculty of Economics and Administration, University of Malaya, and Donghyun Park Principal Economist, Asian Development Bank, Manila, the Philippines

Natural hazards have a potentially large impact on economic growth, but measuring their economic impact is subject to a great deal of uncertainty. The central objective of this paper is to demonstrate a model—the natural disasters vulnerability evaluation (NDVE) model—that can be used to evaluate the impact of natural hazards on gross national product growth. The model is based on five basic indicators—natural hazards growth rates (αi), the national natural hazards vulnerability rate (ΩT), the natural disaster devastation magnitude rate (Π), the economic desgrowth rate (i.e. shrinkage of the economy) (δ), and the NHV surface. In addition, we apply the NDVE model to the north-east Japan earthquake and tsunami of March 2011 to evaluate its impact on the Japanese economy. Keywords: earthquake, economic desgrowth, gross national product growth, Japan, natural disaster, NDVE model, tsunami

Introduction Initially, this paper aims to examine the differences that exist between natural hazards and natural disasters. According to Okuyama and Chang (2004), natural hazards can be considered physical events (the causes of disasters) and natural disasters are the final effects of natural hazards. Hence, any natural hazard (Sorkin, 1982) can have a potentially large effect on economic growth, but measuring the economic impact of natural disasters is subject to a great deal of uncertainty. This is because they impose both direct and indirect costs that change and evolve over time (Greenberg, Lahr, and Mantell, 2007). Natural hazards adversely affect economic activity in the short run in a number of ways. For example, the north-east Japan earthquake and tsunami of March 2011 severely curtailed manufacturing output by destroying power stations (Rose et al., 1997), production facilities, and transportation and other infrastructure (Rose and Benavides, 1998). Beyond the very short term, however, the negative economic impact of natural hazards tends to fade. For example, in the Kobe earthquake of January 1995 (Okuyama, 2003) government reconstruction spending spearheaded a robust recovery in private investment and consumption. As a result, macroeconomic indicators recovered very quickly after an initial drop (Skidmore and Toya, 2002; Noy, 2009). Disasters, 2014, 38(S2): S206−S229. © 2014 The Author(s). Disasters © Overseas Development Institute, 2014 Published by John Wiley & Sons Ltd, 9600 Garsington Road, Oxford, OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA

9,856 13.8 334 87.4 160,892 102.7 41,637 Source: EM-DAT (n.d.).

423,227 Total

156.3

2,034 0.7 0 9.4 147,457 5.8 7,701 183,979 South-east Asia

12.8

2,964 1.5 20 2.9 6,856 19.7 18,668 73,221 South Asia

6.9

381 0.0 0 0.1 270 0.1 56 59 Pacific

0.0

769 10.8 134 73.4 5,582 66.6 9,302 91,003 East Asia

130.9

3,708 0.8 180 1.6 727 10.5 5,910 5.6 74,965 Central and West Asia

Deaths Damage (USD billion) Damage (USD billion) Damage (USD billion) Damage (USD billion)

Deaths

Floods Earthquakes

Deaths

Table 1. Major natural disasters in developing Asia, 2000–2011

Deaths

Storms

Deaths

Droughts

Epidemics

Applying the natural disasters vulnerability evaluation model

Given the potentially large effects of natural hazards on economic growth, it is important for policymakers to have reasonably accurate estimates of these effects (Scanlon, 1988). However, such estimates are difficult to calculate, given the high uncertainty surrounding the measurement of these effects. The motivation for this paper comes from the large numbers of natural hazards that seem to be inflicting damage on the world economy with growing frequency. Developing countries in particular are more vulnerable to natural hazards due to their weaker infrastructure and lack of anticipatory measures (Cuaresma, Hlouskova, and Obersteiner, 2008). Developing Asia in particular accounted for 61 per cent of global fatalities and 9 per cent of all persons affected globally by natural hazards between 1971 and 2011. Table 1 shows the fatalities and estimated damage from various types of natural hazards in developing Asia between 2000 and 2010. The estimated damage implies a sizable negative economic impact on the region. The central objective of this paper is to demonstrate a model—the natural disasters vulnerability evaluation (NDVE) model—that can be used to evaluate the impact of natural hazards on gross national product (GNP) growth. The model is based on five basic indicators: natural hazards growth rates (α i), the national natural hazards vulnerability rate (ΩT), the natural disaster devastation magnitude rate (Π), the economic desgrowth rate (i.e. shrinkage of the economy) (δ), and the NHV surface. Furthermore, the model is also based on elements from an alternative mathematical approach analysis framework from a multidimensional perspective. We look at different

S207

S208

Mario Arturo Ruiz Estrada, Su Fei Yap, and Donghyun Park

types of natural hazards that occurred around the world between 1971 and 2011. To illustrate the NDVE model, we use it to assess the economic impact of the earthquake and tsunami that devastated north-east Japan in March 2011 (Shibusawa, 2011). For comparative purposes we also apply the model to an earlier earthquake in Japan that affected the Kobe region in January 1995. We hope that the NDVE model will contribute towards a more systematic and accurate measurement of the economic impact of natural hazards.

Economic modelling in the evaluation of natural disasters Classic economic modelling in the evaluation of natural disasters Firstly, this paper studies the origins of the economics of natural disasters (Tol and Leek, 1999). Dacy and Kunreuther’s (1969) book entitled The Economics of Natural Disasters: Implications for Federal Policy gives us the first basic economic theoretical framework and policies framework dealing with the analysis of the effects of natural disasters. While this book’s treatment of the issues of policies and implications shows some deficiencies, nonetheless its economic modelling remains the cornerstone of the study of the economics of natural disasters. We believe that the major contribution of this book is the analysis of a long-term recovery model that refers to the need for a reconstruction process that returns the community to its pre-disaster economic level. It is important to mention that the long-term recovery model is based on the use of Solow-Swan’s neoclassical growth model. From a mathematical modelling point of view, the long-term recovery model has a series of basic equations that include the negative effects of natural hazards at different levels from a macroeconomic perspective. It also uses the long-term recovery model, which applies basic arithmetic operations and calculus—in the latter case using partial differentiation (e.g. first derivatives). The graphical modelling is based on the application of basic geometry using the two-dimensional coordinate system. In our opinion, building a model of this magnitude in 1969 was a significant achievement. If we observe the technological limitations (e.g. the scarcity of computers with large capacity and high speed, and the non-existence of econometrics and statistical software) and the limitations of the database, which was confined to simple observations and interviews, it is clear that this model is major contribution. Modern economic modelling in the evaluation of natural disasters Since the 1980s the economics of natural disasters has experienced a deep transformation (in form and content) and faster research expansion using sophisticated analytical tools to evaluate the effects of natural disasters such as the implementation of more modern statistical, mathematical and econometric modelling through the use of advanced software (modern econometrics software programs) and hardware


(high-speed computers with large memory storage capacity). Interesting studies in this area are those of Albala-Bertrand (1993), Okuyama and Chang (2004), Hallegate and Przyluski (2010), Kunreuther and Rose (2004), Loayza et al. (2009), Okuyama (2007), Rose and Liao (2005), Sorkin (1982), and Skidmore and Toya (2002). Some of these studies use basic ideas from Dacy and Kunreuther (1969). Additionally, we can observe that the focus of these studies is on real physical infrastructure damage that directly affected consumption and production. According to Okuyama (2007), the most common model employed to study the economics of natural disasters is the input-output model. Okuyama (2004) observes that this model can only show the basic interdependency that exists among different sectors, and leaves out explicit resources constraints, import substitution, and price changes behaviour (Okuyama, 2007). Therefore, many economists have specialised in the study of the economics of disasters. Subsequently, they prefer to use the computable general equilibrium (CGE) model rather than the input-output model, because the CGE model is more flexible in capturing more variables in the process of economic modelling (Yamano, Kajitani, and Shumuta, 2007). However, the CGE model also presented certain limitations in that it is based on the use of a large database compared to the inputoutput model. Another theoretical framework that is widely used in the study of the economics of natural disaster is the social accounting matrix (SAM) (Cole, 2004). The SAM is designed to study different levels of socioeconomic agents and factors simultaneously. It employs a group of coefficients that estimate the impact of natural disasters by evaluating the feasibility of different possible public policies designed to manage natural disasters (reconstruction) under different magnitudes (Okuyama, 2005; 2007). Finally, the econometric models used to analyse natural disasters show some deficiencies in their incorporation of non-economic variables and technical indicators into the analysis of the effects of natural hazards as a whole. Therefore, we need to integrate into the study of the economics of natural disasters a new dynamicity and complexity through innovative mathematical and graphical approaches to better understand the behaviour of natural disasters and reconstruction management. The idea in designing the NDVE model is to innovatively access the impacts and consequences of a natural hazard. The model tries to evaluate higher order effects of uncertainty after a natural hazard that need to be incorporated into the analysis of the economic impacts of a natural disaster. Here we use the CGE model. Our main objective is to account for this uncertainty and behavioural changes from a multidimensional perspective (mathematical and graphically) within the framework of a dynamic imbalanced state (Ruiz Estrada and Yap, 2012) and the Omnia Mobilis assumption (Ruiz Estrada, 2011). The idea is to move on from classical economic modelling such as linear and non-linear models (e.g. the input-output model, the CGE model, the SAM, etc.) to new economic mathematical modelling and mapping of natural hazards (ex-ante—before the natural hazard—and ex-post—after the natural hazard) by using high-resolution mapping software.

S209

S210


The natural disasters vulnerability evaluation model The NDVE model assumes that any country is vulnerable to natural hazards at any time and anywhere. Additionally, each natural hazard has its own level of potential damage and impact on the GNP of any country. Hence, our world is in a constantly dynamic imbalanced state. This means that at any time and anywhere the possibility exists of a natural hazard occurring and that it can generate different magnitudes of natural disasters. When we speak of a natural hazard, we are referring to any event beyond human control that can generate massive destruction at any time, anywhere, without any advance warning. The quantification and monitoring of natural hazards are inherently difficult, and we cannot evaluate and predict them with any degree of accuracy, but we can compute series of natural hazards within a fixed period of time (per year or per decade) (Wilson, 1982). In addition, the NDVE model is useful for demonstrating how the GNP growth rate is directly connected to the presences of natural hazards. In the context of the NDVE model, we would like to propose five new indicators: natural hazards growth rates (αi), the national natural hazards vulnerability rate (ΩT), the natural disaster devastation magnitude rate (Π), the economic desgrowth rate (δ) (see below), and the NHV surface. These five indicators simultaneously show the various levels of vulnerability and devastation arising from different natural hazards. They are determined by the collection of historical data on various natural hazards that have impacted any country in terms of which such hazards are defined according to certain intervals of time and the magnitude of destruction in terms of the loss of material resources (infrastructure) and non-material resources (human lives) (Rose, 1981). According to our model, the analysis of any natural disasters from an economic point of view must simultaneously take into account the reduction in production (national output) (Pelling, Özerdem, and Barakat, 2002) and human capital mobility (labour) (Rose and Liao, 2005). In this part of our model we introduce a new concept called the economic desgrowth rate (δ) (Ruiz Estrada, 2010), which is defined as a shrinkage of or reduction in economic growth due to any natural hazard. The main function of δ is to determine the ultimate impact of any natural hazard on GNP growth rate behaviour over a particular period of time. The basic data used by the NDVE model is based on the occurrence of 16 possible natural hazard events: earthquakes, tsunamis, floods, volcanic eruptions, typhoons, fire pollution, snow avalanches, landslides, blizzards, cyclones, tornadoes, epidemics, droughts, hailstorms, sandstorms, and hurricanes. The national natural hazards vulnerability rate (ΩT) According to the NDVE model, we assume a continuous but irregular oscillation in and out of various natural hazard events. We do so by applying the natural hazards growth rate (αi), which is equal to the total sum of the same type of natural hazard event in the present year (Σλo) minus the total sum of the same type of natural hazard event in the past ten years (Σλn-1) divided by the total sum of the same type of natural hazard event in the past ten years (Σλn-1) (see Formula 1).


αi = (Σλo – Σλn-1)/Σλn-1

(1)

This means that our world will be in a permanent dynamic imbalanced state under high risk of experiencing a natural hazard event at any time. The NDVE model allows for different magnitudes of destruction. Therefore, we have different natural hazard events growth rates (α i), as described in Formula 2. Therefore, we assume that the national natural hazards vulnerability rate (ΩT) is directly connected to time (Tj). At the same time, Tj is affected directly by different natural hazards growth rates (αi). In our case, ‘j’ is a specific period of time and ‘i’ represents the particular type of natural hazard according to our classification of 16 different types of natural hazards. Hence, ΩT includes a total of 16 possible natural hazard events, as follows: earthquakes (α1), tsunamis (α2 ), floods (α3), volcanic eruptions (α4 ), typhoons (α5), fire pollution (α 6 ), snow avalanches (α7 ), landslides (α 8 ), blizzards (α 9 ), cyclones (α10 ), tornadoes (α11), epidemics (α12 ), droughts (α13), hailstorms (α14 ), sandstorms (α15), and hurricanes (α16 ). Each natural hazard has its magnitude of intensity according to its geographical position and related environmental problems. We assume that if any natural hazard event follows another at a great geographical distance from the first then it cannot be predicted with accuracy, as in Formula 4. Hence, the national natural hazards vulnerability rate (ΩT) is equal to the total sum of all αi divided by the total of natural hazards in the analysis (itotal) (see Formula 3). ΩT = (Σαi)/itotal Є[0 < Σαi < 1] itotal = 16

(2)

ΩTe = Ln[(αi)Tj – (αi)Tj-1]/(αi)Tj]  ΩTe ≠ 0

(3)

ΩTp = Ln[(αimax)Tj] – [(αimin)Tj)] 0 > αimax ≤ 1 or 0 ≥ αimin < 1

(4)

ΩTe ‡ ΩTp

(5)

Formulas 3 and 4 show the effective national natural hazards vulnerability rate (ΩTe) and the potential national natural hazards vulnerability rate (ΩTp). ΩTp is based on a comparison of the past and present natural hazards events growth rates. We assume that the present national natural hazards vulnerability rate (ΩT) cannot be equal to zero (see Formula 3). However, ΩTp is based on the use of a maximal and minimal natural hazard events growth rate in a determinate period of time (Tj) (see Formula 4). Additionally, we need to assume that for the purposes of ΩTp a random database exists that makes it possible for the NDVE model to analyse unexpected results from different natural disaster events that cannot be predicted and monitored with traditional methods of linear and non-linear mathematical modelling. The effective natural hazard events growth rate is identified in Formula 3. Finally, the potential natural hazard event growth rate cannot be equal to the effective natural hazard events growth rate in either the short run or long run (see Formula 5). This is because we assume at the very outset that our world is in a dynamic imbalanced state. Thus the ΩT calculation can be observed in Table 3 for different countries by using different α i and a single ΩT. The evaluation of the national natural hazards vulnerability rate (ΩT) is applied to three different levels of vulnerability (see Formula 6).

S211

S212


Table 2. Calculation of the natural disaster devastation magnitude rate (Π) and the economic desgrowth rate (δ) ΩT

ΩT – 1

√ΩT – 1

Ln √ΩT – 1

ψL

Φk

Π

δ

0.00

1

1

0.0000

0.000

0.00

0.000

0.000

0.01

0.99

0.99

-0.0050

0.001

0.01

0.000

0.000

0.02

0.98

0.99

-0.0101

0.002

0.02

0.000

0.000

0.03

0.97

0.98

-0.0152

0.003

0.03

0.000

0.000

0.04

0.96

0.98

-0.0204

0.004

0.04

0.000

0.000

0.05

0.95

0.97

-0.0256

0.005

0.05

0.000

0.000

0.06

0.94

0.97

-0.0309

0.006

0.06

0.000

0.000

0.07

0.93

0.96

-0.0363

0.007

0.07

0.000

0.000

0.08

0.92

0.96

-0.0417

0.008

0.08

0.001

0.000

0.09

0.91

0.95

-0.0472

0.009

0.09

0.001

0.000

0.10

0.90

0.95

-0.0527

0.010

0.10

0.001

0.000

0.11

0.89

0.94

-0.0583

0.011

0.11

0.001

0.000

0.12

0.88

0.94

-0.0639

0.012

0.12

0.001

0.000

0.13

0.87

0.93

-0.0696

0.013

0.13

0.002

0.000

0.14

0.86

0.93

-0.0754

0.014

0.14

0.002

0.000

0.15

0.85

0.92

-0.0813

0.015

0.15

0.002

0.000

0.16

0.84

0.92

-0.0872

0.016

0.16

0.003

0.000

0.17

0.83

0.91

-0.0932

0.017

0.17

0.003

0.000

0.18

0.82

0.91

-0.0992

0.018

0.18

0.003

0.000

0.19

0.81

0.90

-0.1054

0.019

0.19

0.004

0.000

0.20

0.80

0.89

-0.1116

0.020

0.20

0.004

0.000

0.21

0.79

0.89

-0.1179

0.021

0.21

0.004

-0.001

0.22

0.78

0.88

-0.1242

0.022

0.22

0.005

-0.001

0.23

0.77

0.88

-0.1307

0.023

0.23

0.005

-0.001

0.24

0.76

0.87

-0.1372

0.024

0.24

0.006

-0.001

0.25

0.75

0.87

-0.1438

0.025

0.25

0.006

-0.001

0.26

0.74

0.86

-0.1506

0.026

0.26

0.007

-0.001

0.27

0.73

0.85

-0.1574

0.027

0.27

0.007

-0.001

0.28

0.72

0.85

-0.1643

0.028

0.28

0.008

-0.001

0.29

0.71

0.84

-0.1712

0.029

0.29

0.008

-0.001

0.30

0.70

0.84

-0.1783

0.030

0.30

0.009

-0.002

0.31

0.69

0.83

-0.1855

0.031

0.31

0.010

-0.002

0.32

0.68

0.82

-0.1928

0.032

0.32

0.010

-0.002

0.33

0.67

0.82

-0.2002

0.033

0.33

0.011

-0.002


ΩT

ΩT – 1

√ΩT – 1

Ln √ΩT – 1

ψL

Φk

Π

δ

0.34

0.66

0.81

-0.2078

0.034

0.34

0.012

-0.002

0.35

0.65

0.81

-0.2154

0.035

0.04

0.001

0.000

0.36

0.64

0.80

-0.2231

0.036

0.36

0.013

-0.003

0.37

0.63

0.79

-0.2310

0.037

0.37

0.014

-0.003

0.38

0.62

0.79

-0.2390

0.038

0.38

0.014

-0.003

0.39

0.61

0.78

-0.2471

0.039

0.39

0.015

-0.004

0.40

0.60

0.77

-0.2554

0.040

0.40

0.016

-0.004

0.41

0.59

0.77

-0.2638

0.041

0.41

0.017

-0.004

0.42

0.58

0.76

-0.2724

0.042

0.42

0.018

-0.005

0.43

0.57

0.75

-0.2811

0.043

0.43

0.018

-0.005

0.44

0.56

0.75

-0.2899

0.044

0.44

0.019

-0.006

0.45

0.55

0.74

-0.2989

0.045

0.45

0.020

-0.006

0.46

0.54

0.73

-0.3081

0.046

0.46

0.021

-0.007

0.47

0.53

0.73

-0.3174

0.047

0.47

0.022

-0.007

0.48

0.52

0.72

-0.3270

0.048

0.48

0.023

-0.008

0.49

0.51

0.71

-0.3367

0.049

0.49

0.024

-0.008

0.50

0.50

0.71

-0.3466

0.050

0.50

0.025

-0.009

0.51

0.49

0.70

-0.3567

0.051

0.51

0.026

-0.009

0.52

0.48

0.69

-0.3670

0.052

0.52

0.027

-0.010

0.53

0.47

0.69

-0.3775

0.053

0.53

0.028

-0.011

0.54

0.46

0.68

-0.3883

0.054

0.54

0.029

-0.011

0.55

0.45

0.67

-0.3993

0.055

0.55

0.030

-0.012

0.56

0.44

0.66

-0.4105

0.056

0.56

0.031

-0.013

0.57

0.43

0.66

-0.4220

0.057

0.57

0.032

-0.014

0.58

0.42

0.65

-0.4338

0.058

0.58

0.034

-0.015

0.59

0.41

0.64

-0.4458

0.059

0.59

0.035

-0.016

0.60

0.40

0.63

-0.4581

0.06

0.60

0.036

-0.016

0.61

0.39

0.62

-0.4708

0.061

0.61

0.037

-0.018

0.62

0.38

0.62

-0.4838

0.62

0.62

0.384

-0.186

0.63

0.37

0.61

-0.4971

0.063

0.63

0.040

-0.020

0.64

0.36

0.60

-0.5108

0.064

0.64

0.041

-0.021

0.65

0.35

0.59

-0.5249

0.065

0.65

0.042

-0.022

0.66

0.34

0.58

-0.5394

0.066

0.66

0.044

-0.023

S213

S214


ΩT

ΩT – 1

√ΩT – 1

Ln √ΩT – 1

ψL

Φk

Π

δ

0.67

0.33

0.57

-0.5543

0.067

0.67

0.045

-0.025

0.68

0.32

0.57

-0.5697

0.068

0.68

0.046

-0.026

0.69

0.31

0.56

-0.5856

0.069

0.69

0.048

-0.028

0.70

0.30

0.55

-0.6020

0.070

0.700

0.049

-0.029

0.71

0.29

0.54

-0.6189

0.071

0.71

0.050

-0.031

0.72

0.28

0.53

-0.6365

0.072

0.72

0.052

-0.033

0.73

0.27

0.52

-0.6547

0.073

0.73

0.053

-0.035

0.74

0.26

0.51

-0.6735

0.074

0.74

0.055

-0.037

0.75

0.25

0.50

-0.6931

0.075

0.75

0.056

-0.039

0.76

0.24

0.49

-0.7136

0.076

0.76

0.058

-0.041

0.77

0.23

0.48

-0.7348

0.077

0.77

0.059

-0.044

0.78

0.22

0.47

-0.7571

0.078

0.78

0.061

-0.046

0.79

0.21

0.46

-0.7803

0.079

0.79

0.062

-0.049

0.80

0.20

0.45

-0.8047

0.080

0.80

0.064

-0.052

0.81

0.19

0.44

-0.8304

0.081

0.81

0.066

-0.054

0.82

0.18

0.42

-0.8574

0.082

0.82

0.067

-0.058

0.83

0.17

0.41

-0.8860

0.083

0.83

0.069

-0.061

0.84

0.16

0.40

-0.9163

0.084

0.84

0.071

-0.065

0.85

0.15

0.39

-0.9486

0.085

0.85

0.072

-0.069

0.86

0.14

0.37

-0.9831

0.086

0.86

0.074

-0.073

0.87

0.13

0.36

-1.0201

0.087

0.87

0.076

-0.077

0.88

0.12

0.35

-1.0601

0.088

0.88

0.077

-0.082

0.89

0.11

0.33

-1.1036

0.089

0.89

0.079

-0.087

0.90

0.10

0.32

-1.1513

0.090

0.90

0.081

-0.093

0.91

0.09

0.30

-1.2040

0.091

0.91

0.083

-0.100

0.92

0.08

0.28

-1.2629

0.092

0.92

0.085

-0.107

0.93

0.07

0.26

-1.3296

0.093

0.93

0.086

-0.115

0.94

0.06

0.24

-1.4067

0.094

0.94

0.088

-0.124

0.95

0.05

0.22

-1.4979

0.095

0.95

0.090

-0.135

0.96

0.04

0.20

-1.6094

0.096

0.96

0.092

-0.148

0.97

0.03

0.17

-1.7533

0.097

0.97

0.094

-0.165

0.98

0.02

0.14

-1.9560

0.098

0.98

0.096

-0.188

0.99

0.01

0.10

-2.3026

0.099

0.99

0.098

-0.226

Source: authors.


Level 1: High vulnerability (dark grey cells in Table 2): 1 – 0.75 Level 2: Average vulnerability (light grey cells in Table 2): 0.74 – 0.34 Level 3: Low vulnerability (white cells in Table 2): 0.33 – 0.0 (6) However, in Figure 1 it is possible to observe diminishing returns between the economic desgrowth rate (δ) and the national natural hazards vulnerability rate (ΩT). There are three possible scenarios for this relationship between δ and ΩT. In the first scenario, if ΩT is very high, then δ will be high. In the second scenario, if ΩT is very low, then δ will be low (see Figure 1). Finally, we assume that ΩT can intercept δ, because we are using the ‘dynamic imbalanced state’ (DIS), which never stays static, but constantly changes. Hence, we suggest the application of the Omnia Mobilis assumption to keep the DIS in the long run. It changes according to changes in ΩT. Natural disaster devastation magnitude rate (Π) We use two main variables to calculate the natural disaster devastation magnitude rate (Π). The first main variable is capital devastation (Φk), which we compute by dividing the area of infrastructure destroyed by the natural hazard (km 2 ) by the total infrastructure area (km 2 ) in the same geographical space. The second main variable is human capital devastation (ΨL), which we compute by dividing the number of people killed by or missing due to a natural disaster by the total population in the Figure 1. Relationship between the national natural hazards vulnerability rate (ΩT) and the economic desgrowth rate (δ) ΩT

δ

Source: authors; see Table 2.

S215

S216


same geographical space. After calculating both main variables, we can then multiply the results to get our natural disaster devastation magnitude rate (Π), which is equal to the product of the capital devastation rate (Φk) and the human capital devastation (ΨL) (see Formula 7). Finally, we generate the natural logarithm. Π = ƒ(Φk ,ΨL) = Ln [(Φk) x (ΨL)]

(7)

We decided to apply the product rule of differentiation in Formula 7 to obtain the first derivative test to find the relative maximum and minimum of the capital devastation (Φk) and capital devastation rates (Φk) (see Formulas 8, 9, and 10). ∂ƒ/∂(Φk) = Φ’(k)ΨL/Φ(k) ΨL ∂ƒ/∂(ΨL) = Ψ’(L)Φ(k)/Ψ(L)Φ(k)

∂Π = Φ’(k) Ψ(L) + Φ(k) Ψ’(L)

(8) (9) (10)

Figure 2. How the national natural hazards vulnerability rate (ΩT) can affect the natural disaster devastation magnitude rate (Π)

ΩT

Π Source: authors; see Table 2.


Moreover, we can also observe that the natural disaster devastation magnitude rate (Π) is directly proportional to the national natural hazards vulnerability rate (ΩT). Refer to Table 2 and Figure 2. Economic desgrowth rate (δ) We define economic desgrowth (δ) (Ruiz Estrada, 2010) as a macroeconomic indicator that shows the final impact of a natural hazard on GNP. We can say that the final GNP post-natural hazard effect is a function of the natural disaster devastation magnitude rate (Π) (see Formula 11). At the same time, Π is directly dependent on the national natural hazards vulnerability rate (ΩT) (see Formula 11) according to Figures 1 and 2. In Formula 12 we calculate the preliminary GNP post-natural hazard effect (Q’). Hence, Q’ is a function of Π. Π = ƒ(ΩT)

(11)

Q’ = ƒ(Π)

(12)

Therefore, the economic desgrowth rate (δ) depends on these two functions in our model, as shown in Formula 13 (i.e. a function of a function). Therefore, δ can only obtain values between 0 and -∞ . . . δ = ƒ(Π(ΩT))

(13)

In the last instance, the final GNP preliminary hazard effect (Q’) directly depends on the changes between the natural disaster devastation magnitude past rate (∂Π o) and the natural disaster devastation magnitude present rate (Π o+1) (see Formula 14). Q’ = ∂Πo/∂Πo+1

(14)

Finally, the economic desgrowth rate (δ) is equal to the preliminary GNP postnatural hazard effect (Q’) minus the final GNP pre-natural hazard effect (Qo) (see Formula 15). δ = Q’ – Qo

(15)

Figures 1 and 2 show that a strong relationship exists between the economic desgrowth rate (δ), on the one hand, and Π and ΩT, on the other. The empirical results show that if Π and ΩT are higher, then δ shows the same behaviour. Our experiment is based on the uses of different rates from 0.00 to 0.99 in the case of ΩT. The finals results calculated for δ show that when the Π and ΩT are high, the effect on δ is magnified. Hence, δ is directly proportional to Π and ΩT in the long run (see Table 2). Finally, we assume that δ, Π, and ΩT are moving significantly together (see Formulas 16 and 17). According to our model, δ always starts from zero and retains negative values. ↑δ = ƒ↑Π (↑Ω ) (16) T ↓δ = ƒ↓Π (↓ΩT) (17)

S217

S218


The natural hazards vulnerability surface (NHV surface) The construction of the NHV surface is based on the natural hazard growth rate (α i) results and the mega-surface coordinate space (see Formula 18 and Figure 3). The NHV surface is a four-by-four matrix that contains the individual results of all 16 variables (taken from Table 3). However, the 16 variables are plotted in a fourby-four array with the vertical value on the NHV surface. The idea is to produce a surface for a quick pictorial representation of the overall propensities for any one country. The underlying idea here is to use the results of 16 variables in α i to build a symmetrical surface. When the NHV coordinate system (η) has strictly the same number of rows as the number of columns, then the natural hazards growth rates can always be perfectly symmetrical.

η=

α1 α5 α9 α13 α2 α6 α10 α14 α3 α7 α11 α15 α4 α8 α12 α16

(18)

The final analysis of the NHV surface depends on any changes that this surface can experience in a fixed period of time.1

Applying the NDVE model: Japan Applying the NDVE model to the Japanese economy will give us a much better idea of how the model works. Before we do so, it is useful to have a look at general data Figure 3. The mega-surface coordinate space

Source: authors.


about Japan such as the contribution of each region to the country’s final GNP and the geographical distribution of Japanese industry. In terms of the geographical distribution of Japanese GNP, we find that Hokkaido contributes around 15 per cent of Japan’s GNP, while the Honshu region contributes 43 per cent, the highest share. The region with the second highest contribution to Japan’s GNP is the Shikoku region with 27 per cent. Therefore, the major contributors to Japanese GNP are the Honshu and Shikoku regions, which collectively account for 70 per cent of Japanese output. Finally, the Kyushu region contributes 15 per cent to Japanese output (see Figure 4). Honshu and Shikoku also account for about 70 per cent of Japanese industrial output, with the remaining output divided among the other regions (see Figure 5). Figure 4. Contribution of each Japanese region to GNP

Natural hazards growth rates (αi) In this section we first examine the natural disaster vulnerability propensity rates for countries around the world and then take a closer look at Japan’s natural disaster vulnerability propensity rate.

Global natural hazards growth rates (αi )

Source: METI (2011a; 2011b; 2011c); JETRO (2011a; 2011b; 2011c; 2011d).

Figure 5. Concentration of industries in Japan

Source: METI (2011a; 2011b; 2011c); JETRO (2011a; 2011b; 2011c; 2011d).

Table 3 shows the natural hazards growth rates in 59 countries around the world. These countries show a wide range of probability of natural disaster event based on their historical data. We use three different shades of grey to classify countries according to their αi: the dark grey shading represents high vulnerability, medium grey represents medium vulnerability and light grey represents low vulnerability. We can observe in Table 3 that the ten countries with the highest risk of natural disasters are China, Japan, the USA, Indonesia, the Philippines, Australia, South Korea, Taiwan, Chile, and Guatemala. Therefore, Japan is among the top ten countries with the highest natural hazard growth rates (α i); specifically, second highest, according to Table 3. On the other hand, countries such as Mongolia, Hungary, South Africa, Denmark, Belgium, and Luxemburg have

S219

Country

China

Japan

USA

Indonesia

Philippines

Australia

South Korea

Taiwan

Chile

Guatemala

El Salvador

Honduras

Nicaragua

Costa Rica

Panama

Mexico

Russia

Vietnam

Caribbean

India

Cambodia

No.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

0.70

0.65

0.95

0.65

0.85

0.95

0.95

0.95

0.95

0.95

0.95

0.95

0.85

0.80

0.80

0.35

0.99

0.95

0.65

1.00

0.95

α1

0.85

0.55

0.85

0.90

0.15

0.20

0.25

0.25

0.25

0.25

0.25

0.25

0.75

0.75

0.75

0.35

0.98

0.95

0.25

1.00

0.55

α2

1.00

1.00

0.45

1.00

0.75

0.70

0.55

0.55

0.55

0.55

0.55

0.55

0.65

0.55

0.70

1.00

0.85

0.85

0.85

0.35

0.99

α3

0.00

0.00

0.00

0.00

0.20

0.75

1.00

1.00

1.00

1.00

1.00

1.00

0.55

0.35

0.00

0.00

0.95

1.00

0.25

0.75

0.05

α4

0.95

0.85

0.99

0.95

0.10

0.55

0.35

0.35

0.35

0.35

0.35

0.35

0.35

0.95

0.85

0.50

0.95

0.95

0.75

0.85

0.95

α5

0.00

0.75

0.00

0.00

0.85

0.35

0.85

0.85

0.85

0.85

0.85

0.85

0.35

0.00

0.00

1.00

0.00

0.65

0.65

0.00

0.65

α6

0.00

0.00

0.00

0.00

0.95

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.85

0.00

0.25

0.00

0.00

0.00

0.50

0.65

0.55

α7

0.65

0.10

0.20

0.55

0.35

0.45

0.90

0.90

0.90

0.90

0.90

0.90

0.75

0.55

0.35

0.00

0.85

0.85

0.20

0.55

0.65

α8

0.00

0.00

0.00

0.00

0.95

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.3

0.00

0.25

0.00

0.00

0.00

0.7

0.35

0.30

α9

0.00

0.00

0.95

0.00

0.00

0.20

0.25

0.25

0.25

0.25

0.25

0.25

0.25

0.55

0.55

0.55

0.75

0.25

0.65

0.65

0.20

α10

0.00

0.00

0.00

0.00

0.00

0.20

0.00

0.00

0.00

0.00

0.00

0.00

0.20

0.65

0.85

0.65

0.00

0.00

0.70

0.45

0.35

α11

0.98

0.85

0.55

0.95

0.75

0.75

0.85

0.85

0.85

0.85

0.85

0.85

0.25

0.45

0.65

0.90

0.65

0.65

0.50

0.85

0.95

α12

0.00

0.45

0.00

0.00

0.10

0.45

0.00

0.00

0.00

0.00

0.00

0.35

0.30

0.00

0.00

0.65

0.00

0.00

0.35

0.00

0.65

α13

Table 3. Natural hazards growth rates (αi) and national natural hazards vulnerability rate (ΩT), 1971–2011

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.3

0.00

0.00

0.00

0.00

0.00

0.4

0.10

0.30

α14

0.00

0.15

0.00

0.00

0.05

0.05

0.00

0.00

0.00

0.00

0.00

0.00

0.20

0.35

0.35

0.65

0.00

0.00

0.25

0.85

1.00

α15

0.75

0.55

0.99

0.95

0.00

0.55

0.55

0.55

0.55

0.55

0.55

0.55

0.10

0.99

0.89

0.75

0.98

0.95

0.65

0.90

0.85

α16

0.368

0.369

0.371

0.372

0.378

0.384

0.406

0.406

0.406

0.406

0.406

0.428

0.431

0.434

0.453

0.459

0.497

0.503

0.513

0.581

0.621

ΩT

S220 Mario Arturo Ruiz Estrada, Su Fei Yap, and Donghyun Park

Country

Thailand

Bangladesh

Colombia

Egypt

Venezuela

Singapore

Ecuador

Italy

Malaysia

Pakistan

Brazil

New Zealand

Greece

Iceland

Holland

Morocco

Bolivia

Cyprus

Canada

Kenya

Sweden

Norway

No.

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

0.00

0.00

0.55

0.10

0.25

0.95

0.90

0.15

0.65

0.95

0.95

0.35

0.95

0.03

0.98

0.95

0.03

0.95

0.85

0.95

0.10

0.35

α1

0.15

0.15

0.00

0.10

0.85

0.00

0.85

0.85

0.20

0.90

0.35

0.85

0.05

0.75

0.20

0.10

0.95

0.20

0.15

0.10

0.90

0.85

α2

0.05

0.05

0.10

0.10

0.45

0.55

0.25

1.00

0.01

0.25

0.35

0.85

0.85

0.75

0.95

1.00

0.95

1.00

0.90

1.00

1.00

0.85

α3

0.00

0.00

0.15

0.00

0.00

0.35

0.00

0.00

0.35

0.05

0.55

0.00

0.00

0.00

0.95

0.00

0.00

0.00

0.00

0.00

0.00

0.00

α4

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.55

0.00

0.20

0.25

0.00

0.00

0.85

0.00

0.35

0.90

0.65

0.00

0.55

0.99

0.90

α5

0.00

0.00

0.15

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.35

0.55

0.00

0.00

0.15

0.00

0.25

0.10

0.20

0.00

0.00

α6

0.45

0.45

0.00

0.65

0.00

0.00

0.00

0.00

0.75

0.00

0.20

0.00

0.10

0.00

0.65

0.00

0.00

0.00

0.00

0.00

0.00

0.00

α7

0.00

0.00

0.00

0.35

0.50

0.65

0.25

0.00

0.00

0.55

0.00

0.00

0.65

0.35

0.25

0.90

0.00

0.75

0.25

0.85

0.55

0.65

α8

0.55

0.55

0.00

0.6

0.00

0.00

0.00

0.00

0.95

0.00

0.00

0.00

0.00

0.00

0.15

0.00

0.00

0.00

0.00

0.00

0.00

0.00

α9

0.30

0.30

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.35

0.25

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

α10

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

α11

0.05

0.05

0.90

0.25

0.45

0.55

0.35

0.85

0.03

0.85

0.30

0.65

0.55

0.85

0.30

0.55

0.95

0.45

0.85

0.65

0.90

0.98

α12

0.00

0.00

0.65

0.00

0.00

0.15

0.60

0.00

0.00

0.00

0.00

0.25

0.15

0.00

0.00

0.00

0.00

0.00

0.65

0.00

0.03

0.00

α13

0.75

0.75

0.00

0.75

0.00

0.10

0.00

0.00

0.85

0.00

0.00

0.00

0.45

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

α14

0.00

0.00

0.15

0.05

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.05

0.05

0.00

0.00

0.00

1.00

0.00

0.00

0.00

α15

0.00

0.00

0.00

0.00

0.55

0.00

0.10

0.25

0.00

0.35

0.85

0.65

0.00

0.75

0.15

0.65

0.90

0.45

0.00

0.55

0.85

0.90

α16

0.144

0.144

0.166

0.181

0.191

0.206

0.206

0.228

0.237

0.256

0.259

0.263

0.269

0.274

0.289

0.291

0.293

0.294

0.297

0.303

0.333

0.343

ΩT

Applying the natural disasters vulnerability evaluation model S221

Israel

Spain

Laos

Paraguay

Argentina

France

Uruguay

England

Germany

Czech Rep.

Mongolia

Hungary

South Africa

Denmark

Belgium

Luxemburg

44

45

46

43

48

49

50

51

52

53

54

55

53

57

58

59

34

0.00

0.10

0.00

0.00

0.00

0.00

0.10

0.10

0.25

0.55

0.05

0.35

0.95

0.25

0.75

0.85

α1

22.4

0.00

0.05

0.15

0.25

0.00

0.00

0.00

0.05

0.35

0.35

0.10

0.10

0.00

0.00

0.10

0.05

α2

31

0.00

0.15

0.05

0.10

0.15

0.00

0.15

0.05

0.25

0.25

0.10

0.35

0.45

0.55

0.35

0.01

α3

13.35

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.10

0.00

0.00

0.00

0.00

α4

19.83

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.05

α5

12.1

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

α6

8.55

0.25

0.00

0.00

0.00

0.20

0.00

0.25

0.35

0.00

0.00

0.35

0.15

0.00

0.00

0.00

0.00

α7

20.8

0.05

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.15

0.00

0.65

0.00

0.00

α8

6.5

0.00

0.00

0.00

0.00

0.25

0.00

0.00

0.25

0.00

0.00

0.35

0.10

0.00

0.00

0.00

0.00

α9

9.15

0.00

0.00

0.00

0.35

0.00

0.00

0.00

0.00

0.25

0.15

0.00

0.00

0.05

0.00

0.10

0.00

α10

4.25

0.00

0.00

0.00

0.00

0.00

0.10

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.10

0.00

α11

33

0.05

0.15

0.05

0.15

0.05

0.03

0.45

0.20

0.35

0.25

0.55

0.25

0.35

0.55

0.40

0.60

α12

6.83

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.05

0.10

0.00

0.35

0.55

α13

6.2

0.00

0.03

0.35

0.00

0.25

0.00

0.25

0.25

0.05

0.00

0.25

0.10

0.00

0.00

0.00

0.00

α14

6.25

0.00

0.00

0.00

0.00

0.00

1.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.05

0.00

0.00

α15

22.4

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.15

0.00

0.00

0.00

0.10

α16

Source: EM-DAT (n.d.).

Note: We applied probabilities in terms of the record of all natural disasters events mentioned in this table.

High level of risk: 1. Earthquakes; 2. Floods; 3. Epidemics

α1 = earthquakes α 2 = tsunamis α 3 = floods α 4 = volcanic eruptions α 5 = typhoons α 6 = fire pollution α7 = snow avalanches α 8 = landslides α 9 = blizzards α10 = cyclones α11 = tornadoes α12 = epidemics α13 = droughts α14 = hailstorms α15 = Nuclear pollution α16 = hurricanes

αi = natural hazards growth rates

TOTAL

Country

No.

0.271

0.022

0.030

0.038

0.053

0.056

0.071

0.075

0.078

0.094

0.097

0.109

0.116

0.119

0.128

0.134

0.138

ΩT

S222 Mario Arturo Ruiz Estrada, Su Fei Yap, and Donghyun Park


the lowest αi. This means that according to historical data, they face lower risk of natural disaster than the other countries in our sample.

Japan’s natural hazards vulnerability rate (ΩT): max. and min. In the case of Japan, we find large differences between the maximum and minimum of the natural hazards vulnerability rate (ΩT). According to the historical data of natural disasters, Hokkaido has the lowest vulnerability, with a ΩTmin of only 0.15 and a ΩTmax of 0.25. In the rest of Japan the natural disaster vulnerability propensity rates are higher. More specifically, the vulnerability rate ranges from 0.45 to 0.95 in Honshu, from 0.35 to 0.95 in the Shikoku region, and from 0.25 to 0.85 in the Kyushu region (see Figure 6).

Natural disaster devastation magnitude rate (Π) In addition, we would like to compare the natural disaster devastation magnitude rate (Π) between the Kobe earthquake of 1995 and the north-east Japan earthquake of 2011. According to our results the devastation resulting from the 1995 Kobe earthquake was quite limited at -6.68. But according to our computations the devastation caused by the 2011 north-east Japan earthquake and tsunami was much larger at -12.22. In Figure 7 we can observe Figure 6. Japan’s natural hazards vulnerability rate (ΩT) by region

Source: EM-DAT (n.d.); FDMA (2011).

S223

S224


more clearly from a graphical perspective that the north-east Japan earthquake and tsunami (right-hand cone) caused devastation several times greater than the Kobe earthquake (left-hand cone). Table 4. Natural disaster devastation magnitude rates of the Kobe earthquake (Π1995 ) and the 2011 earthquake and tsunami (Π 2011) Natural disaster devastation magnitude rate of Kobe earthquake (Π1995) Φk 150

552

0.271739

ΨL 5500

1200000

0.004583333

0.00124547

(Π)

-6.68

Natural disaster devastation magnitude rate of 2011 earthquake and tsunami (Π2011) Φk 35000

227962

0.15353436

ΨL 16000

50000000

0.00032

4.9131E-05

-12.22

Source: authors; see Table 2.

Figure 7. Natural disaster devastation magnitude rates (Π) of the 1995 Kobe earthquake (left-hand cone) and the 2011 earthquake and tsunami (right-hand cone)

Note: final result from the NDVE model. Source: authors; see Table 2.


Economic desgrowth rate (δ) Finally, to measure the impact of earthquakes and tsunamis on economic growth, we use the new concept of ‘economic desgrowth’ (δ) introduced by Ruiz Estrada (2010). Using this concept, we try to discover possible leakages that can adversely affect GNP performance. Basically, this new concept assumes that in the process of GNP formation, leakages may arise due to various factors, in our case natural disasters. According to our estimates, the economic desgrowth caused by the Kobe earthquake had an impact of -1.51 on Japan’s economic desgrowth. Our estimates indicate that economic desgrowth caused by the north-east Japan earthquake and tsunami of 2011 was much larger, at -4.6 in 2011. Therefore, the economic desgrowth difference between the Kobe earthquake and the north-east Japan earthquake and tsunami of 2011 is -3.1 according to our final result in Table 5.

Concluding observations and policy implications Natural disasters can have a significant negative impact on economic performance, but measuring this impact with any degree of certainty is inherently challenging. In this paper we propose a new model for evaluating the impact of natural disasters on economic performance. The natural disaster vulnerability evaluation (NDVE) model is based on five indicators: (i) natural hazards growth rates (α i); (ii) the national natural hazards vulnerability rate (ΩT); (iii) the natural disaster devastation magnitude rate (Π); (iv) the economic desgrowth rate (δ); and (v) the NHV surface. The underlying intuition is that the economic impact of natural disasters depends on a country’s vulnerability to natural disasters and the devastation caused by natural disasters, which jointly determines the leakage from economic growth and hence the impact on growth. We hope that our model will contribute to a better and deeper understanding of how to measure the economic impact of natural disasters. A more useful measurement of impact is conducive to the drawing up of appropriate policies, both for dealing with the effects of natural disasters and for anticipatory policy measures that seek to lessen the impact of natural disasters before they occur. For example, underestimating the impact may lead to the government allocating too few resources for addressing the impact of disasters—e.g. public investment in physical infrastructure and income support for households most affected by the disaster. On the other hand, overestimating the impact may cause the allocation of too many resources, raising the risk of inefficiency and waste. By the same token, determining the appropriate level of anticipatory investments to limit the impact of future disasters would benefit from an accurate ex-ante assessment of their impact. The NDVE model can also help in determining the appropriate mix of disaster management and prevention policies. For example, the model may allow policymakers to better estimate and compare the impact of different types of natural disasters. The application of our model to two natural disasters in Japan—the Kobe earthquake of January 1995 and the north-east Japan earthquake and tsunami of March

S225

S226


Table 5. Japan’s GNP growth rate, 1971–2011 1

1971

4.4

2

1972

8.4

3

1973

8.0

4

1974

-1.2

5

1975

3.1

6

1976

4.0

7

1977

4.4

8

1978

5.3

9

1979

5.5

10

1980

3.2

11

1981

4.2

12

1982

3.4

13

1983

3.1

14

1984

3.1

15

1985

4.5

16

1986

2.8

17

1987

4.1

18

1988

7.1

19

1989

5.4

20

1990

5.6

21

1991

3.3

22

1992

0.8

23

1993

0.2

24

1994

0.9

25

1995

0.7

26

1996

1.7

27

1997

1.6

28

1998

-2.0

29

1999

-0.1

30

2000

2.9

31

2001

0.2

32

2002

0.3

33

2003

1.4

34

2004

2.7

35

2005

1.9

36

2006

2.0

37

2007

2.4

38

2008

-1.2

39

2009

-5.2

40

2010

1.2

41

2011

-2.8

δ = -4.6

ΩT = 0.99

Variables: δ = GNP desgrowth rate ΩT = national natural hazards vulnerability rate Π = natural disaster devastation magnitude rate Source: IMF (2011).

δ = -1.51

ΩT = 0.95

Π = -6.68

Π = -12.22


2011—indicates that the 2011 disaster will have a bigger impact on the Japanese economy than the Kobe earthquake. Nevertheless, the immediate implication for Japanese policymakers is that they need to support growth with stronger measures than they implemented in 1995. In particular, they need to provide more fiscal resources for reconstruction efforts to rebuild the region’s devastated physical infrastructure, which in turn will lay the foundation for the recovery of the region’s productive activities, in particular manufacturing. In addition to rebuilding infrastructure, the government should provide income support for residents whose homes and livelihoods have been destroyed by disasters. While Japan’s high public debt level constrains the government’s fiscal space, concerted fiscal support is nevertheless vital for north-east Japan’s recovery. At a broader level, our results confirm that natural disasters can have a significant economic impact even in advanced countries with good infrastructure and high levels of preparedness. The inescapable policy implication for developing countries, which tend to suffer the bulk of natural disasters, is that investing in anticipatory measures may yield sizable benefits in the medium and long term, even though they can be costly in the short run. Anticipatory measures can reduce the extent of damage, loss of life, and disruption to economic activity. Such measures include: (1) good design of and adherence to rigorous building codes, earthquake and storm proofing of buildings, floodplain and drainage systems, hillside stabilisation, and other measures related to the natural and manmade environments; (2) early warning system for floods, storms, epidemics, typhoons, tsunamis, and others; and (3) emergency response plans such as evacuation systems, emergency response drills, equipment readiness, and the storage of essential supplies like medicine and water. Given the high opportunity costs of using fiscal resources to mitigate the effects of natural disasters in developing countries, the NDVE model’s more accurate measurement of the economic impact of natural disasters is all the more valuable. Better measurement allows for the more efficient and better-targeted use of fiscal resources. One interesting direction for future research is to examine the importance of effective communication in mitigating the adverse impact of natural disasters. It is widely believed that more effective communication by the Japanese government to the general public, for example about the magnitude and nature of the damage, could have limited the damage from the 2011 tsunami. The failure of authorities to quickly and reliably inform the public led to widespread concerns and fear, which further dented consumer and business confidence. Therefore, more and better information is likely to reduce the impact of natural disasters, and looking at the role of information would contribute to a more accurate measurement of the impact of a disaster.

Correspondence Dr Mario Ruiz Estrada, Faculty of Economics and Administration, University of Malaya, Kuala Lumpur 50603, Malaysia. E-mail: [email protected]

S227

S228


Endnotes 1

Initially it was intended to provide examples of NHV surfaces for Japan, the USA, China, Luxemburg, Guatemala, and South Korea to illustrate the points made here. However, due to technical problems during design this was not possible. The author will be happy to provide such examples on request; see his e-mail address under ‘Correspondence’.

References Albala-Bertrand, J.M. (1993) The Political Economy of Large Natural Disasters with Special Reference to Developing Countries. Clarendon Press, Oxford. Cole, Sam (2004) ‘Geohazards in social systems: an insurance matrix approach’. In Y. Okuyama and S.E. Chang (eds.) Modeling the Spatial and Economic Effects of Disasters. Springer, New York, NY. Cuaresma, J.C., J. Hlouskova, and M. Obersteiner (2008) ‘Natural disasters as creative destruction? Evidence from developing countries’. Economic Inquiry. 46(2). pp. 214–226. Dacy, D. and H. Kunreuther (1969) The Economics of Natural Disasters: Implications for Federal Policy. Free Press, New York, NY. EM-DAT (International Disasters Database) (n.d.) Database. Centre for Research on the Epidemiology of Disasters. http://www.emdat.be/database (accessed April 2011). FDMA (Fire and Disasters Management Agency) (2011) General Reports. http://www.fdma.go.jp/en/ (accessed April 2011) Greenberg, M.-R., M. Lahr, and N. Mantell (2007) ‘Understanding the economic costs and benefits of catastrophes and their aftermath: a review and suggestions for the U.S. Federal Government’. Risk Analysis. 27(1). 83–96. Hallegate, A. and V. Przyluski (2010) ‘The economics of natural disasters: concepts and methods’. Policy Research Working Paper 5507. World Bank, Washington, DC. IMF (International Monetary Fund) (2011) Database. http://www.econstats.com/weo/v002.htm (accessed April 2011). JETRO ( Japan External Trade and Organisation) (2011a) White Book from 1999 to 2010. http://www. jetro.go.jp/en/reports/white_paper/ (accessed April 2011). JETRO (2011b) Japan’s International Trade in Goods Monthly. http://www.jetro.go.jp/en/reports/ statistics/ (accessed April 2011). JETRO (2011c) International Trade and Investment. http://www.jetro.go.jp/en/reports/white_paper/ (accessed April 2011). JETRO (2011d) Survey and Reports on International Trade. http://www.jetro.go.jp/en/reports/survey/ trade/ (accessed April 2011). Kunreuther, H. and A. Rose (eds.) (2004) The Economics of Natural Hazards, Vols. I & II. Edward Elgar, Northampton, MA. Loayza, N., E. Olaberria, J. Rigolini, and L. Christiaensen (2009) ‘Natural disasters and growth: going beyond the averages’. World Bank Policy Research Working Paper 4980. World Bank, Washington, DC. METI (Ministry of the Economy, Trade and Industry) (2011a) ‘Analysis of all industrial activities in Japan’. http://www.meti.go.jp/english/statistics/bunseki/index.html (accessed April 2011). METI (2011b) ‘Inter-regional input-output table by region’. http://www.meti.go.jp/english/statistics/ tyo/tiikiio/index.html (accessed April 2011). METI (2011c) ‘Current survey of production’. http://www.meti.go.jp/english/statistics/tyo/seidou/ index.html (accessed April 2011). Noy, I. (2009) ‘The macroeconomic consequences of disasters’. Journal of Development Economics. 88. pp. 221–231.


Okuyama, Y. (2003) ‘Long-run effects of a disaster: measuring the economic impact of the Kobe earthquake’. Mimeo. CUEE, Tokyo Institute of Technology, Tokyo. Okuyama, Y. (2004) ‘Modeling spatial economic impacts of an earthquake: input-output approaches’. Disaster Prevention and Management. 13(4). pp. 297–306. Okuyama, Y. (2005) Critical Review of Methodologies on Disaster Impact. Niigata, Graduate School of International Relations, International University of Japan. Okuyama, Y. (2007) ‘Economic modeling for disasters impact analysis’. Economic Systems Research. 19(2). pp. 115–124. Okuyama, Y. and S.E. Chang (eds.) (2004) Modeling Spatial and Economic Impacts of Disasters. Springer, New York, NY. Pelling, M., A. Özerdem, and S. Barakat (2002) ‘The macro-economic impact of disasters’. Progress in Development Studies. 2. pp. 283–305. Rose, A. (1981) ‘Utility lifelines and economic activity in the context of earthquakes’. In J. Isenberg (ed.) Social and Economic Impacts of Earthquake on Utility Lifelines. American Society of Civil Engineers, New York, NY. pp. 107–120. Rose, A. and J. Benavides (1998) ‘Regional economic impacts’. In M. Shinozuka, A. Rose, and R.T. Eguchi (eds.) Engineering and Socioeconomic Impacts of Earthquakes. Multidisciplinary Center for Earthquake Engineering Research, Buffalo, NY. pp. 95–123. Rose, A., J. Benavides, S.E. Chang, P. Szczesniak, and D. Lim (1997) ‘The regional economic impact of an earthquake: direct and indirect effects of electricity lifeline disruptions’. Journal of Regional Science. 37. pp. 437–458. Rose, A. and S.Y. Liao (2005) ‘Modeling regional economic resilience to disasters: a computable general equilibrium analysis of water service disruptions’. Journal of Regional Science. 45. pp. 75–112. Ruiz Estrada, M.A. (2010) ‘The economic desgrowth’. SSRN-Working Papers. http://papers.ssrn. com/sol3/papers.cfm?abstract_id=1857277. Ruiz Estrada, M.A. (2011) ‘Policy modeling: definition, classification and evaluation’. Journal of Policy Modeling. 33(3). pp. 523–536. Ruiz Estrada, M.A. and S.F. Yap (2012) ‘Policy modeling: definition, classification and evaluation’. Journal of Policy Modeling. 34(6). Scanlon, J. (1988) ‘Winners and losers: some thoughts about the political economy of disasters’. International Journal of Mass Emergencies and Disasters. 6(1). pp. 47–63. Shibusawa, H. (2011) ‘Evaluating the economic impacts of a disaster: a CGE application to the Tokai region of Japan’. Mimeo. Toyohashi University of Technology. Skidmore, M. and H. Toya (2002) ‘Do natural disasters promote long-run growth?’ Economic Inquiry. 40(4). pp. 664–687. Sorkin, A.L. (1982) Economic Aspects of Natural Hazards. Lexington Books, Lexington, MA. Tol, R. and F. Leek (1999) ‘Economic analysis of natural disasters’. In T. Downing, A. Olsthoorn, and R. Tol (eds.) Climate Change and Risk. Routledge, London. pp. 308–327. Wilson, R. (1982) Earthquake Vulnerability Analysis for Economic Impact Assessment. Information Resources Management Office, Federal Emergency Management Agency, Washington, DC. Yamano, N., Y. Kajitani, and Y. Shumuta (2007) ‘Modeling the regional economic loss of natural disasters: the search for economic hotspots’. Economic Systems Research. 19(2). pp. 163–181.

S229