Presentation from the launch of the OECD report "Assessing the Real Cost of Disasters - The Need for Better Evidence". For further information see oe.cd/cost-of-disasters

1. ASSESSING THE REAL COST OF
DISASTERS
Stéphane Jacobzone
Public Governance Directorate, OECD
11th Meeting of High-Level Experts and Leaders Panel on Water
and Disasters (HELP)
Geneva, 3-4 May 2018
The need for better evidence

2. 0
50
100
150
200
250
300
350
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012
AnnualeconomiclossesinUSDbillion
• Past decade: USD 1.5 trillion in economic damages from man-
made disasters (industrial accidents, terrorist attacks) and natural
disasters (primarily storms and floods)
Why we need to better account for the
real costs of disasters
Source: EM-DAT: The OFDA/CRED International Disaster Database,UniversitécatholiquedeLouvain,Brussels,Belgium,www.emdat.be
(accessed 14 November 2013).
Economic losses due to disasters in OECD
and BRIC countries, 1980-2012 (USD Billion)

3. • Increase in economic damages believed to outpace national DRR
investments…
• … though this claim cannot be supported by data as there is hardly
any available, especially on an internationally comparative level
• The development of standardised and comparable accounting
frameworks for DRM expenditure and disaster losses can:
– Support the evaluationand prioritisationof economic benefits of DRR
investments
– Faciliatecross-ountry comparisons
– Provide systematic indicatorson global DRR objectives (Sendai Framework,
SDG’s)
Why we need to better account for the
real costs of disasters
Source: EM-DAT: The OFDA/CRED International Disaster Database,UniversitécatholiquedeLouvain,Brussels,Belgium,www.emdat.be
(accessed 14 November 2013).

4. Australia
Bangladesh
Bolivia
Chile
Costa Rica
Egypt
Estonia
Ethiopia
Fiji Finland
France
Germany
Greece
Haiti
Honduras
India
Indonesia
Iran
Italy
Jamaica
Japan
KenyaMadagascar
Malawi
Mexico
Mozambique
Nepal
Netherlands
New Zealand
Norway
Pakistan
Philippines
Poland
Portugal
Slovenia
Thailand
Turkey
United Kingdom
United States
Venezuela
Yemen
0.5
1
1.5
2
2.5
3
3.5
2.7 3.2 3.7 4.2 4.7
AverageDeathTollperDisaster
1980-2013(log)
Real GDP per Capita, Year 2010 (log)
Significant decreaseinfatality ratesfromdisasterswithincreasing income
1980-2013
OECD Non-OECD
• Resilience against disasters in OECD countries is high , but higher
income countries still experience large economic losses
Disaster losses in OECD countries
Source: Source: EM-DAT: The OFDA/CRED International Disaster Database,www.emdat.be - Universitécatholiquede Louvain - Brussels - Belgium". Data for OECD and
BRIC countries (1980-2012).Figures areshown true to the year of the event. OECD Stat National Accounts GDP per capita in US$,constant prices,reference year 2005

5. • Resilience against disasters in OECD countries is high , but higher
income countries still experience large economic losses
• Policy makers need a good understanding of past losses to face this
challenge and understand better whether their DRR investments
are effective
Disaster losses in OECD countries
Source: EM-DAT: The OFDA/CRED International
Disaster Database,www.emdat.be - Université
catholiquede Louvain - Brussels - Belgium;OECD
(2013), “
Gross domestic product (GDP) MetaData : GDP per
capita,US$, constantprices,reference year 2005”,
National Accounts OECD Statistics Database,
accessed on 14 November 2013,
http://stats.oecd.org/

6. 1. Review national and international efforts recording ex-post
disaster losses:
– Analyse their comparability
– Analyse strengths and weaknesses
→ provide basis for developing better methods, setting new international standard and
providing repository for such information
2. Assess ongoing national efforts and propose draft framework for
assessing public spending for DRM:
– To better understand countries‘ expenditures ex-ante and ex-post of disasters
– Help policy makers understand whether their spending efforts lead to future
reductions in disaster losses
→ goal is to be sufficiently comprehensiveso as to account for most such
expenditure items, while being broad enough to capture similar expenditures
across countries
OECD work on assessing the real costs of
disasters: objectives

7. • A review of national and international
approaches and datasets regarding disaster
losses and damages found:
– Significant progress has been made in accounting for a
number of social loss indicators (such as the fatalities from
a disaster)…
– …but that efforts to calculate economic losses remain
inconsistentand incomparable across countries (only an
estimated 30-40%of disasters are reported with economic loss
figures)
Assessing disaster losses and damages:
country evidence

8. • Losses are challenging to asses as they can be:
– Short, medium, or long-term
– Direct or indirect
– Not only felt locally,but trigger through economic sectors and countries
globally
• Identifying expenditure information across different departments
and sectors equally challenging:
– There are no central repositories for DRM expenditure information
– Multipleagencies and levels of government have DRM expenditures, each
their own way of describing this in budgets and nationalaccounts
– Even more complex if expenditure for DRM is “embedded”
– Requires much effort and judgment to identify spending categories across
sectors and levels of government
Disaster losses vs. expenditures

9. Country Host institution Hazards
Australia No centralized national repository,but various comprehensive databases Natural
Austria
No centralized national repository,but sectoral repositories for different
ministries
Natural
Canada Public Safety Canada Natural & man-made
Colombia Unidad Nacional para la Gestión del Riesgo de Desastres Natural
Costa Rica Ministry of Economic Policyand National Planning (MIDEPLAN) Natural
Finland
No centralized national repository,but sectoral repositories for different
ministries
Natural & man-made
France Observatoire national des risques naturels Natural
Japan
No centralised national repository,but sectoral repositories for different
ministries
Natural
Mexico National Disaster Prevention Centre (CENAPRED) Natural & man-made
Poland Ministry of the Interior and Administration Natural
Slovak Republic Ministry of Interior and Ministry of the Environment Natural
Slovenia Administration for Civil Protection and Disaster Relief,Ministry of Defence Natural
Sweden Swedish Civil Contingencies Natural
Switzerland
No centralized national repository,but sectoral repositories for different
ministries Natural
Turkey Disaster and EmergencyManagementAuthority Natural & man-made
9
Assessing disaster losses and
damages: country evidence
Source: 2016 OECD surv ey ;
Many OECD & partner countries collect disaster loss data …

10. … but data comparability is not a given
10
Assessing disaster losses and
damages: country evidence
Thresholds for
collecting disaster
impact data
Distinction between
publicly and privately
accrued damages
 Disaster loss data collection lacks common language
 Aggregated disaster loss figures hide the distributional impacts of disasters,
hampering effective disaster risk reduction strategies
Are direct and indirect
losses separately
accounted?
Source: 2016 OECD surv ey ; 17
country responses

11. 11
Improving disaster loss and damage
assessments: measuring indirect impacts
Damages to infrastructure and businesses have
important cascading effects
 Critical infrastructure failure = cause of triggering cascade effects of
major disasters
 Disaster related business interruption & supply chain disruptions have
an impact far beyond directly affected areas
 Most countries report only direct damages, although sometimes
estimations on indirect losses are made for large-scale disasters

12. 12
Improving disaster loss and damage assessments:
the value of public private partnerships
Sharing data collected by non-government agencies
 Data on insured losses often more comprehensive, systematically recorded
 Can provide a basis for estimating overall economic losses
 Public-private partnerships as an innovative way to improve information sharing
between public authorities and private organisations
Example: National Observatory of Natural Risks (ONRN)

13. … policy makers rely on an incomplete picture of their
country’s spending on disaster risk management
13
How much is spent on disaster risk
management: what do we know?
Is information on
expenditure collected?
41%
47%
12%
Yes No N.a.
 Available data focuses predominantly on
specific spending categories & central
government spending
 Disaster risk management expenditure not
always earmarked in public accounts/ budgets
 Embedded disaster risk management
expenditure across many government sectors
Source: 2016 OECD surv ey ; 17
country responses

14. Assessing public (and private) expenditure for DRM: the
need for increased comparability
 Reviews that exist are result of specific project to retrieve DRM
expenditure information from national accounts and sectoral
budgets
 Usually one-off efforts (although some include historical data)
 Some focus on distinguishing expenditure along the DRM cycle (e.g.
preparedness vs. response spending), others gather information on
specific hazards
 A harmonised approach is needed
 Objective is to find a way to obtain such information from
governments in a comparative way on a continuous basis

15. Disaster-related contingent
liabilities in publicfinance
frameworks
Assessing the real cost of
disasters
15
OECD work on the cost of disasters

16. Thank you
stephane.jacobzone@oecd.org