Ministry of Land, Infrastructure, Transport and Tourism

21 November 2014

Kenichiro Tachi Director for International Coordination of River Engineering

Water and Disaster Management Bureau,

Ministry of Land, Infrastructure, Transport and Tourism (MLIT)

Strengthening the evidence base for

disaster risk management policy analysis

Why is the progress toward mainstreaming Disaster

Risk Management always sluggish?

1

Key Question

Under such a condition, it is crucial to take evidence-

based approach to justify the effects of Disaster Risk

Management policies.

The effects of investment in Disaster Risk Management are

usually unrecognizable. If structures or facilities for Disaster

Risk Management are constructed and work well, it is hard

for citizens to imagine the damage possibly occurred without

the measures. Justifying investment in Disaster Risk

Management always incurs difficulty.

Such approach helps practitioners on the ground do their

jobs!

1. Why baseline data collection is important?

• Background

• Flood Damage Statistics Survey

• Utilization of the Flood Damage Statistics

2. What would be required to justify Disaster Management

investments?

• Precise estimation of possible disaster damages

• Cost-Benefit Analysis

3. Way forward

2

Contents

1. Why baseline data collection

is important?

3

•Actual damage caused by water-related disasters such as floods, landslides, storm surges, tsunamis, etc. is estimated every year. •The survey provides basic data indispensable for presenting the necessity and the effectiveness of flood control measures.

Damages to General Asset

housings, household commodities, machineries for

businesses, agricultural products

Damages to public

Infrastructure

・Cost for

emergency

rehabilitation, etc

Damages to Assets for Public Services

Transportation, communication, etc

4

1. Why baseline data collection is important?

Flood Damage Statistics Survey (since 1961)

Nishi-Biwajima, Aichi in Shinkawa river 5

Tokai Storm of 2000, September 11-13

Total economic loss

771,492 million JPY

87%

1% 11%

1%

General assets

Agriculture

Public infrastructures

Public Services

Obu City and Kariya City, Aichi September 12,2000

1. Why baseline data collection is important?

Utilization of Flood Damage Statistics :

Estimation of Total Economic Loss

The state of the damage to Toyooka Hospital (parking lot) caused by the flooding of the Maruyama River as a result of Typhoon #23 of 2004.1)

6

Typhoon #23, October 2004

Total economic loss

770,910 million JPY

JMA

1) Breach of Maruyama River - Records of Typhoon #23 and verification , Tajima Shikyoku, Kobe Shimbun, Kobe Shimbun Sogo

Shuppan Center, 2005

Yomiuri Shimbun

67% 1%

31%

1%

General assets

Agriculture

Public infrastructures

Public Services

1. Why baseline data collection is important?

Utilization of Flood Damage Statistics :

Estimation of Total Economic Loss

7 Inundation caused by Ikarashi River levee failure (Sanjo City, Niigata Prefecture)

※：Source; Niigata pref.（March 23, 2005)

Large Flood in Niigata inJuly 2004 (15 people died and approx. 21,000 buildings were damaged in Niigata Pref.)

Utilization of Flood Damage Statistics :

Identification of Effects of flood Control Projects

1. Why baseline data collection is important?

8

Shinano River Ikarashi River

River improvements were implemented based on the July 2004 flood experience. Total cost : about 47.1 billion JPY

Measures in response to 2004 Large Flood in Niigata

Shinano River

(September 2008)

(March 2009)

(August 2005)

(December 2010) (September 2010)

Retarding Basin

Shinano River

Excavation

Embankment

Kariyata River

River Improvement Work &

Retarding Basin

Excavation of flood channel by 1-2 m

Retarding basin

River channel

Levee

Improvement Channel widening

and excavation

1. Why baseline data collection is important?

9

○ In July 2011 the Shinano River Basin experienced a total rainfall of approx. 1,000mm, which was the largest rainfall on record and 1.6 times more than that of July 2004, but both damages to buildings and human casualties were reduced dramatically. The investments were successfully justified in the end!

Total Rainfall 1)

2004.7

Total

rainfall

1,006mm

2011.7

Number of Buildings Damaged 2)

9,778

buildings

2004.7

421 buildings

2011.7

90%

reduction

60 %

increase

Number of fatalities or missing 3)

12

persons

2004.7

1 person

2011.7

90%

reduction Total

rainfall

647mm

1）Kasabori rain gauge station 2）2004.7: 「7.13新潟豪雨 水害記録誌（March 2006 Niigata Prefecture) 2011.7: Produced by Niigata Prefecture based on 「第1回平成23年7月新潟・福島豪雨対策検討委員会」 3）Shinano River Downstream, Ikarashi River, Kariyata River Disaster Rehabilitation Emergency Project Pamphlet (Shinano Karyu River Office, Niigata Prefecture)

1. Why baseline data collection is important?

Inundation in Nishibiwajima Town

The case of Heavy Rain in Tokai (2000.9)

670 billion yen

Total losses

Cost of Prevention Measures

120 billion yen

Effectiveness of prevention

550 billion yen

71.6 billion yen of

investment could

have reduce 550

billion yen of

damages

10

1. Why baseline data collection is important?

Estimated losses with prevention measures

Utilization of Flood Damage Statistics :

Identification of Effects of flood Control Projects

0

5,000

10,000

15,000

20,000

25,000

30,000

1

10

100

1,000

10,000

19

46

19

51

19

56

19

61

19

66

19

71

19

76

19

81

19

86

19

91

19

96

20

01

20

06

Flood Management Cost (Nominal)

GDP (Nominal)

Number of Fatalities

Number of fatalities (people)

Flood manegement cost (Nominal) (hundred of million yen)

GDP(nominal) <billion yen>

<600,000>

<500,000>

<400,000>

<300,000>

<200,000>

<100,000>

<0>

Effect of Flood Management in Japan

*Number of fatalities exclude those who dead by tsunami *GDP : 1980-2009(2000 price), 1946-1979(1990 price)

←

Typ

ho

on

Kat

hle

en

←

Typ

ho

on

Ion

←Ty

ph

oo

n K

itty

←

Typ

ho

on

Jan

e

←

Typ

ho

on

Ru

th

←

Wes

t-Ja

pan

Flo

od

←

Typ

ho

on

To

yam

aru

←Is

ahay

a Fl

oo

d

←Ty

ph

oo

n K

arin

oga

wa

←Ty

ph

oo

n Is

e-B

ay

←

Seco

nd

Typ

ho

on

Mu

roto

←To

kai F

loo

d

←

Typ

ho

on

No

.24

, No

.26

←W

est-

Jap

an W

ater

Dis

aste

r

←

Typ

ho

on

No

.6, N

o.7

, No

.9

←Ty

ph

oo

n N

o.1

7

←

Nag

asak

i Wat

er D

isas

ter

←W

este

rn-S

anin

Wat

er D

isas

ter

←Ty

ph

oo

n N

o.6

, Lan

dsl

ide

in N

agan

o

←

Wes

t-Ja

pan

Wat

er D

isas

ter

←

Typ

ho

on

No

.17

, No

.18

, No

.19

Pyr

ocl

asti

c fl

ow

of

Un

zen

-Fu

gen

-Dak

e ←

Au

gust

Flo

od

in K

yusy

u

←

Deb

ris

Flo

w in

Kag

osh

ima

Har

ihar

a ←

Fuku

shim

a To

chig

i Flo

od

, Ko

chi F

loo

d

←D

ebri

s Fl

ow

in H

iro

shim

a

Ku

mam

oto

Flo

od

Number of Fatalities by Floods , GDP and Budget for Flood Management (2000 price)

Water-related Disasters Statistics in Japan

1. Why baseline data collection is important?

11

Utilization of Flood Damage Statistics :

Identification of Long-term Effects of flood Management Investment

Inundated Area by Floods

0

10000

20000

30000

40000

50000

60000

70000

19

62

19

67

19

72

19

77

19

82

19

87

19

92

19

97

20

02

20

07

Total Inundated Area (10,000ha)

Area of Inundated (Residential & Other Property) (10,000ha)

Inundated area (10,000 ha)

20

10

0

12

1. Why baseline data collection is important?

0

20

40

60

80

100

120

140

0

10000

20000

30000

40000

50000

60000

70000

19

62

19

67

19

72

19

77

19

82

19

87

19

92

19

97

20

02

20

07

Damage to General Assets (10 billion yen)

Area of Inundated (Residential & Other Property) (10,000ha)

Damage to General Assets (2000 Price)

Economic Losses to General Assets

13

1. Why baseline data collection is important?

0

20

40

60

80

100

120

140

0

10000

20000

30000

40000

50000

60000

70000

19

62

19

67

19

72

19

77

19

82

19

87

19

92

19

97

20

02

20

07

Damage to

General Assets

(10 billion yen)

Density of Damage

Cost

(thousand yen / ha)

Total Inundated Area (10,000ha)

Area of Inundated (Residential & Other Property) (10,000ha)

Density of Flood Damage to General Assets

Damage to General Assets (2000 Price)

20 10 0

Inundated area (10,000 ha)

Flood Damage Density (Economic Losses per Area), GDP and Budget for Flood Management (2000 price)

Water-related Disasters Statistics in Japan

1. Why baseline data collection is important?

14

All disaster damage trend in Japan 1990-2009 (2000 price)

0

500

1,000

1,500

house

housing

companies

agri/fishery

housing res

companies res

private total

river

coast/port

sediment

landslide

debris

road/bridge

sewage

park/urban

Year

Billion JPY

1. Why baseline data collection is important?

15

Query and view the survey data

Aggregate, organize, and

screen the survey data

•A general statistics survey based on the article 19 of the Statistics Act (approved by

Minister for Internal Affairs and Communications)

•Carried out by MLIT’s Water and Disaster Management Bureau in collaboration with

prefectures and municipalities.

National Government

(MLIT) Prefectures

Occurrence of

water-related

disaster

General assets Public infrastructures Public Services

Municipalities Municipalities

Prefectures

Prefectures

Public Services

Offices

Submit -Aggregate the survey data

-Calculate the damage

-Compile and publish Flood

Damage Statistics

-Upload the data to the

DB server of Flood

Damage Statistics

Develop the guideline of flood

damage statistics survey

(coverage and procedure of the

survey, survey format, fill-in

guideline, etc.) Submit within 45 days of disaster

occurrence

Access to Flood Damage

Statistics DB National Government

(Regional Offices, etc.)

Submit

Submit

16

1. Why baseline data collection is important?

2.What would be required to

justify Disaster Management

Investments?

17

Along with the basic survey on flood disaster damages,

it is also required to estimate possible future damages by flood

disasters so as to justify investment in “ex-ante Disaster Risk

Management”.

Cost-Benefit analysis is a useful tool that can provide strong

evidence for the relevance of Flood Disaster Risk Management

investments from the economic aspect.

The benefit is presented as the expected value of the reduction of

possible damages that future projects produces. The estimation of

the expected damages should be as accurate as possible.

Therefore, efforts must be done to improve the accuracy of

damage estimation through covering various kinds of damages,

which include indirect damages as well as direct damages..

2. What would be required to justify Disaster Management Investments? -

18

Categories Methods of evaluation

Manual Manual for Economic Evaluation of Flood Control Investment (April 2005, MLIT)

General conditions

Evaluation period The project implementation period plus 50 years after completion of the project.

Lifetimes of assets levee 50 years, dam 80 years

Evaluated damages

Direct damage property damage, agricultural damage, infrastructure damage

Indirect damage business interruption loss, emergency response cost at household levels

Calculating method of total benefit

Damage reduction, difference between the damages calculated for with and without project cases, is multiplied by the annual exceedance probability, and are summed up to calculate the expected average annual damage, which becomes the annual benefit. Annual benefit is accumulated over the evaluation period and the residual value of the asset constructed by the project is added to determine the total benefit.

Calculating method of total cost Project construction cost (including construction cost, land acquisition cost and compensation cost) and maintenance/operation cost

Social discount rate 4%（based on the real rate of interest of national bond）

Quantitative evaluation criteria Benefit Cost Ratio (B/C)

Methodology of Cost-Benefit Analysis in Japan

Methodology of Cost-Benefit Analysis

19

Items Methods of evaluation

Manual “Manual for Flood Damage Indices Analysis” (2013.6, MLIT)

General conditions

Flood magnitudes

In principle damage reduction is not converted into expected annual damage reduction, but instead the magnitude of damage reduction is assessed under a specific size of flood, such as a design flood (single or multiple floods).

Evaluated damages

Direct damage Human damage (inundated population, expected loss of life, maximum number of persons isolated, etc.)

Indirect damage Damage caused by impaired social functions (medical/welfare facilities, disaster management facilities) Cascading impact (traffic disruption, lifeline, economic damage) Others (underground space, cultural facilities, debris waste generated by flood)

Computation of overall non-monetized impact

For each of the damage categories damages are evaluated quantitatively whenever possible, e.g., affected population, etc

Evaluation of Damages with difficulty in quantitative monetary

estimation in the Cost-Benefit Analysis in Japan

20

2. What would be required to justify Disaster Management Investments? -

10. Increase in land prices as a result of improvement in flood protection

Direct damage

Indirect damage

Intensification benefit

3. Business interruption ・ Business interruption damage (household , businesses, public and public interest services )

・ Emergency response costs (household, businesses, public and public interest service)

4. Impairment of social welfare facilities (hospitals, social welfare facilities, disaster management facilities)

5. Economic impact (disruption of traffic , lifeline, cascading economic impact of damages)

6. Other damage (underground space, cultural facilities, waste generated by flood)

7. Psychological damage

8. Risk premium (insecurities due to possibility of damage)

9. Damages that cause permanent changes to the local economic system (Because the region as a whole is

severely damaged by a large-scale flood the economic system cannot recover to its pre-disaster state.)

( : Damages not considered in Cost-Benefit Analysis (difficult to monetize) but are quantitatively assessed

( : Considered in Cost-Benefit Analysis)

1. Damage to asset ・General property damage (houses, residential properties depreciable assets and inventory assets of

businesses, depreciable assets of fishing and farming properties, etc.)

・ Agricultural product damage

・ Infrastructure damage

2. Human damage (human loss, people isolated etc.)

( : Damages not considered in Cost-Benefit Analysis (difficult to monetize) but are quantitatively assessed

( : Considered in Cost-Benefit Analysis)

21

Consideration of damages with difficulty in quantitative monetary

estimation in the Cost-Benefit Analysis in Japan

2. What would be required to justify Disaster Management Investments? -

22

3. Way forward

23

24

3. Way Forward

1. Collection and accumulation of basic disaster damage data is crucial to promote investment in Disaster Risk Management, through the evidence-based approach.

2. OECD’s effort to establish the common framework of evidence data collection would assist nations’ challenges, which are difficult but indispensable.

3. Cost-benefit analysis of investment in Disaster Risk Management provides strong evidence to justify their relevance. Efforts must be done to improve the accuracy of damage estimation through covering various kinds of damages.

4. Analyses of various damage estimation methods of countries and sharing them are meaningful to present a better framework of damage estimation that includes damage categories, such as “indirect damage”, which contribute to the improvement of accuracy.

Thank you for your attention!

25

25