DISASTER RISK FINANCE - A TOOLKIT
May 2019
GIZ ACRI+ Commissioned Report
Authored by:
Conor Meenan, Risk Management Solutions (RMS)
John Ward, Pengwern Associates
Robert Muir-Wood, Risk Management Solutions (RMS)
3CONTENTS
CONTENTS
EXECUTIVE SUMMARY 05
INTRODUCTION 09
Why we need disaster risk management 10
A toolkit for disaster risk finance: report structure 13
DISASTER RISK FINANCE TOOLKIT 15
1. Risk audit 16
2. Disaster risk management actions 20
3. Dimensions of instrument design 22
3.1. Risk Holder 23
3.2. Purpose 26
3.3. Timing 27
3.4. Risk Level 29
4. Disaster risk finance instruments 30
4.1 Risk Reduction 31
Loans 31
Micro-credit 32
Bonds 33
Grants, subsidies, & tax breaks 34
Crediting 35
Impact bonds 36
4.2. Risk Retention 37
Budget contingency 40
Reserve Funds 41
Contingent loans 42
4.3. Risk Transfer Instruments 43
Micro-insurance 44
Agriculture insurance 45
Takaful & mutual insurance 46
Insurance & reinsurance 47
Catastrophe bonds 48
Risk pools 49
5. Risk management strategy 50
5.1. Complementarity 50
5.2. Risk Layering 53
6. Illustrative urban use case 57
This page intentionally left blank.
EXECUTIVE SUMMARY 5
Executive
Summary
DISASTER RISK FINANCE – A TOOLKIT6
EXECUTIVE SUMMARY
e impacts of climate-related disaster risks are growing.
e Intergovernmental Panel on Climate Change
identies that the frequency and severity of climate-
related hazards are already increasing due to climate
change, and that this will worsen in the future. e
damage that events cause is also growing, as people and
assets continue to concentrate in vulnerable locations and
inadequate measures are taken to reduce the vulnerability
of people and assets to these risks.
ese risks disproportionately aect developing countries.
is is driven both by their greater exposure to risks and
their greater vulnerability once risks materialize. 90 per
cent of those who have been killed by disasters since
the 1990s live in either Africa or Asia, while the direct
economic losses from disasters are 14 times higher in low–
income countries than high–income countries.
ere is an imperative to reduce and better manage these
risks. A key element to achieving this is the development
of disaster risk management plans. ese plans, developed
ahead of a specic disaster arising, can specify what
actions to undertake to reduce risks and also who will do
what, taking account of what evidence, after a disaster.
To be eective, these plans need to be developed in an
inclusive way, with particular focus on the needs of the
poor and vulnerable. ey require the participation of a
large number of stakeholders through processes that can
often be facilitated by development and humanitarian
partners.
However, disaster risk management plans only work when
accompanied by nance. is nance facilitates and
incentivizes activities that reduce risk. It also means that
sucient and reliable resources are available to respond
when remaining risks materialize. Ensuring this nance is
available in a timely fashion after a disaster is crucial for
reducing the human cost of disasters.
Much uncertainty surrounds the dierent nancial
instruments for disaster risk that are available to
governments, municipalities, communities and
households – as well as the development and
humanitarian partners who support them. Dierent
instruments can play dierent roles, providing dierent
amounts of resources to dierent actors at dierent
speeds. is means that dierent instruments are more or
less appropriate to use in dierent circumstances. It also
means that, in most cases, a combination of instruments
will be required to eciently and comprehensively
manage disaster risk.
e purpose of this disaster risk toolkit is to provide
practical guidance on how to choose which disaster risk
nance instruments for which circumstance. e main
audience is policymakers in developing countries who
are responsible for disaster risk management, at national,
regional and local levels. It is also intended to assist
the development and humanitarian community who
support developing country policymakers in disaster
risk management and who, sometimes, either implicitly
or explicitly, also hold some of the risks associated with
disasters in these countries. It is structured as a series of
steps that those actors who hold risk, and the partners
who support them in this role, can follow to better
understand, reduce and manage these risks, and nance
activities accordingly.
EXECUTIVE SUMMARY 7
Step 1: Risk Audit. is involves developing a sound
understanding of the underlying risk in order to shape
the subsequent nancing strategy. is consists of four
phases (i) dening the exposure at risk – both in terms
of people and assets - to understand what needs to be
managed; (ii) identifying what perils and hazards can
impact that exposure, (iii) quantifying the expected
frequency and severity of impact from those perils,
ideally using a probabilistic risk analysis, and; (iv)
setting a resilience target to identify the extent to
which risks will be explicitly managed.
Step 2: Determining disaster risk management actions.
is requires identifying actions that can be taken to cost
eectively reduce the risks that are faced. is will be
determined based on specic circumstances and requires
both sound economic analysis and engaged, participatory
processes. In relation to the remaining risks, a decision
needs to be taken as to which will be retained (the
nancial consequences of the risk materializing are borne
by those who face the risk) and which will be transferred
(the nancial consequences of the risk materializing are
passed to a third party, usually in return for a premium
payment).
Step 3: Understanding the dimensions of the financing
need. Risk reduction, retention and transfer can be
achieved by using a range of nancial instruments.
However, before these instruments can be selected,
a basic situational analysis should be undertaken to
understand the nancial needs associated with these
activities in more detail. is can be structured around
answering four key questions:
– What is the capacity and need of the risk holder?
e risks of disasters fall on a wide range of actors,
from individuals to communities, municipalities
and sovereign governments. ere may also be cases
where the humanitarian and development community
choose to hold risks, in order to reduce the human
suering that events will otherwise cause. Dierent
risk holders will have dierent capacities and nancial
ability to make use of dierent nancial instruments.
– What will the funds be spent on? e ultimate purpose
of disaster risk nance instruments is to fund or
facilitate resource ows towards a diverse range of
activities that make disasters less damaging for people.
is can be further disaggregated between funding
directed towards protecting and managing the impacts
of risk on lives and livelihoods; funding directed at
reducing the damage that events cause on assets and
facilitating the reconstruction of assets,
and the services they provide, after a destructive event;
and funding covering the immediate operational and
humanitarian needs after a disaster strikes.
– When is funding needed? Funding for risk reduction
is required in advance of disaster impact, and can be
independent of any particular event, or based on long
or near-term event forecasts. After an event strikes,
funding needs spike and there is an urgent need
for additional resources, followed by a longer term,
typically larger, but less urgent, need for funding to
support reconstruction. Dierent nancial instruments
are more or less valuable in meeting funding needs at
dierent timescales.
– What level of risk is being addressed? Some risks
manifest themselves on a frequent basis, even annually.
Other risks are much less frequent but, when they do
arise, cause more severe levels of impact. e prole
of risk has an important bearing on which nancial
instruments might be desirable.
Step 4: Selecting disaster risk financing instruments.
is involves understanding the range of nancial
instruments available, their strengths and weaknesses and
applicability to dierent dimensions of nancing needs.
To support risk reduction activities, the key instruments
and incentives that can be considered are loans; micro-
credit; bonds; grants, subsidies and tax breaks; crediting
and impact bonds. e key nancing instruments for
risk retention are budget contingencies, reserve funds
and lines of contingent credit. Risk transfer instruments
include insurance – and its dierent forms including
mutual insurance, Takaful, microinsurance, agriculture
insurances and risk pools – as well as catastrophe bonds.
Many of these instruments have a range of variants
that alter their suitability in dierent circumstances. In
particular, risk retention and risk transfer instruments
where dierent ‘trigger’ mechanisms can be used to
determine whether and how much funding is released
following a disaster. Figure 1 illustrates how these
dierent instruments map on to the dimensions of
nancing need identied in step 3.
FINANCE & INSURANCE TOOLKIT For the Renewable Energy Sector in Barbados8
Figure 1. Taxonomy of DRF instruments
Risk Holder Purpose Timing Risk Level
What is the capacity and need
of the risk holder?
What will funds be spent on? When is funding needed? What level of risk is being
addressed? (return period)
Action Instrument Individual Community Municipality Sovereign Life &
Livelihood
Operational Physical
Assets
Preparation Response Recovery Annual 1-10
year
10-50
year
50+
year
Risk Reduction
Loan
• • • • • • • • • •
Micro-credit
• • • • • • • •
Bonds
• • • • • • • • •
Grants, subsidies,
& tax breaks
• • • • • • • • • • •
Crediting
• • • • • • • • • • •
Impact Bonds
• • • • • • • • • • • • •
Risk Retention
Budget
Contingency
• • • • • •
Reserve Funds
• • • • • • • • • • •
Contingent Loans
• • • • • •
Risk Transfer
Micro-insurance
• • • • • • • • •
Agriculture
Insurance
• • • • • • • •
Takaful & Mutual
Insurance
• • • • • • • • • • • •
Insurance &
Reinsurance
• • • • • • • • • • • •
Catastrophe Bonds
• • • • • • • • •
Risk Pools
• • • • • • • • •
Step 5: Combining disaster risk financing instruments
to create a disaster risk finance strategy. A coherent
disaster risk nancing strategy will involve more than one
instrument. e possibility of combining instruments
opens up a range of further issues that risk holders and
their partners need to consider. Risk reduction activities
reduce the residual risk, and therefore the expected costs
associated with risk retention and risk transfer. Focus is
growing on how to capture this risk reduction in a way
that increases the incentive to reduce risks. As regards risk
retention and risk transfer instruments, a risk-layering
strategy can reduce costs and improve the reliability
of funding. is involves combining risk retention
instruments for high-probability, low impact events with
risk transfer instruments for the lower probability, higher
impact events.
To practically illustrate these steps, the nal section of
the paper presents a hypothetical use case of an urban
environment in South East Asia and shows how these
steps might be followed and the possible implications
that may result.
INTR ODUCTION 9
Introduction
DISASTER RISK FINANCE – A TOOLKIT10
INTRODUCTION
Why we need disaster risk management
Natural systems contain extremes, whether in the
motions of the atmosphere, the concentration of
precipitation, or the accumulation and release of
strain along faults. e gradients of temperature in the
atmosphere can generate vortex storms. e runo from
extreme rainfall can overow river systems. e absence
of rain over many months itself causes drought and can
exacerbate wildre. e continents are being pushed
and pulled by the convective currents within the earth.
Human induced climate change threatens to make
many of these extreme events more likely. e
Intergovernmental Panel on Climate Change
1
identies
that the frequency and severity of climate-related hazards
are already increasing due to climate change, and that
this will worsen in the future. In particular it warns that
we can expect an increased frequency and intensity of
heatwaves; an increased frequency of heavy precipitation
events, resulting in greater risk of ooding at the regional
scale; and an increased frequency and intensity of extreme
sea level events, such as those caused by storm surges.
e impact of these extreme events depends critically
on both the exposure and vulnerability of potentially
aected people and assets. Exposure relates to the
extent to which people, communities and assets are
located in areas that are prone to hazards. For example,
exposure increases when decisions are taken that lead
to people living in ood prone areas (or, alternatively,
when decisions that might prevent people from living in
ood prone areas fail to be taken). Vulnerability relates
to the social, economic and environmental factors which
increase the susceptibility of people, communities or
assets to the impact of a hazard. For example, people who
lack the knowledge or resources to undertake preventative
actions ahead of a disaster arising are more vulnerable
to the impacts of that disaster. Unsurprisingly, the poor
and socially disadvantaged are typically also the most
vulnerable to disasters, lacking access to public services
and with restricted availability or aordability of water,
food and other consumption items.
Both exposure and vulnerability help to explain why the
impact of disasters is far more damaging in developing
countries than in developed ones. According to the
INFORM Index for Risk Management
2
, 9 out of the 10
countries most exposed to natural hazards are developing
countries – while developing countries account for
all of the top 70 positions in the same organization’s
vulnerability index. Correspondingly, 90 per cent of
those who have been killed by disasters between 1990 and
2013 lived in low or middle income countries
3
, while the
direct economic losses from disasters, when expressed as
a percentage of GDP, are 14 times higher in low–income
countries than high–income countries
4
.
Policymakers and humanitarian actors increasingly
recognize the need to respond to these growing risks,
especially in developing countries. As the Box A below
explains, the Sendai Framework
5
and the Warsaw
International Mechanism for Loss and Damage
6
are
multilateral initiatives that reect the urgency that
the international community attaches to reducing and
managing disaster risks while the Agenda for Humanity
7
also places a strong focus on managing disaster risks in
developing countries.
Responding to these risks requires information,
planning and nancial resources, along with an
appropriate enabling environment. ere is little that
can be done to control how hard the wind blows, but it
is possible to assess how much damage it might cause in
which locations. Similarly, it is possible to understand
how the design of the built environment will inuence the
damage caused by wind, ood, re, and ground shaking.
is information allows the development of disaster risk
management plans to better reduce and manage these
risks. ese plans can identify risk-informed actions to
reduce risks – both a long time in advance of a disaster,
and through anticipatory actions taken immediately
before a disaster strikes – and how these actions will
be nanced. ey can also identify what will happen
after a disaster strikes, who will undertake what actions
to respond and recover from an event and where the
associated nancial resources will come from. By making
plans ahead of time, identifying and clarifying roles
and responsibilities (both nancial and otherwise), the
devastating impacts of disasters can be reduced
8
. ese
plans are easier to develop and implement when there is
political consensus on their value – so that they can be
developed through a technocratic, apolitical process –
and when backed by an enabling legal framework.
INTR ODUCTION 11
e success of disaster risk management plans depend
critically on the involvement of all key stakeholders:
policymakers, international actors, humanitarian
agencies, non-governmental actors and community
groups. It is particularly important for plans to be
developed in active consultation with those who are
most vulnerable to disasters – such as disabled, elderly,
women, slum dwellers and indigenous groups. Typically
these groups bear the brunt of any disaster impact but
can be too easily excluded from decisions over what
should be done and where. Only with the full and
active participation of these groups can the devastating
impact of disasters on lives, livelihoods and economic
development potential be reduced and managed
eectively.
e Integrated Climate Risk Management (ICRM)
approach from GIZ’s ACRI+ project provides a
framework for the development and execution of
disaster risk management plans. It emphasizes both the
traditional role of disaster risk management in responding
to growing climate risks, as well as the important role of
risk retention and risk transfer mechanisms. It explains
how the latter could be particularly important as the
adverse eects of climate change pose new forms of risks
that are currently dicult to predict. Figure 2 illustrates
the framework.
Figure 2. Integrated Climate Risk Management (ICRM) Approach.
risk reduction
measures
pre-disaster
financing
emergency
management
relief
post-disaster
financing
rehabilitation
building
back
better
risk
analysis
RESILIENCE
©
G
I
Z
/
M
C
I
I
2
0
1
7
P
r
e
v
e
n
t
i
o
n
R
e
t
e
n
t
i
o
n
&
T
r
a
n
s
f
e
r
P
r
e
p
a
r
e
d
n
e
s
s
R
e
c
o
v
e
r
y
R
e
s
p
o
n
s
e
DISASTER RISK FINANCE – A TOOLKIT12
e development of disaster risk management plans
according to this framework is a substantial exercise –
this Toolkit focuses on the nancial instruments that
can facilitate their implementation. e development
of disaster risk management plans requires consideration
of a wide number of factors including what activities
to undertake and when, and how to ensure active
participation of all key stakeholders. is report does not
seek to discuss all of these issues. Rather, recognizing
the emphasis that the ICRM framework places on risk
retention and transfer, which typically require dedicated
nancial instruments, it has a more focused purpose: to
i While this is partly motivated by the specialised financial instruments associated with risk retention and transfer, it also
considers financial instruments that can be used for all elements of a disaster risk management plan.
examine the nancial instruments that allow the delivery
of disaster risk management plans
1
. Often this is seen as
a technical, somewhat impenetrable, issue. But, it has a
crucial role: the delivery of nance through appropriate
instruments is indispensable for the cost-eective
implementation of any plan. is report aims to provide
a practical disaster risk nance toolkit for policymakers,
humanitarian actors and practitioners to understand the
wide range of nancial instruments that are available;
their characteristics, strengths and weaknesses; and how
they can be combined within a disaster risk management
plan to develop a coherent, cost-eective approach.
Box A. Multilateral initiatives to address disaster risk
The Sendai Framework is a 15-year (from the year of its adoption in 2015), voluntary, non-binding agreement
which recognizes that the State has the primary role to reduce disaster risk but that this responsibility should
be shared with other stakeholders including local government, the private sector and other stakeholders.
It aims for: ‘The substantial reduction of disaster risk and losses in lives, livelihoods and health and in the
economic, physical, social, cultural and environmental assets of persons, businesses, communities and countries’.
This objective is encapsulated in seven targets – relating to, for example, global disaster mortality and direct
disaster loss – to be delivered through four priorities for action. These priority areas are:
• Understanding disaster risk
• Strengthening disaster risk governance
• Public and private investment in disaster risk reduction; and
• Enhancing disaster preparedness for effective response and to Build Back Better
The Warsaw International Mechanism for Loss and Damage associated with Climate Change Impacts
has been mandated with promoting implementation of approaches to address loss and damage associated
with the adverse effects of climate change. It has three main functions:
• To enhance knowledge and understanding of comprehensive risk management approaches to address loss
and damage
• To strengthen dialogue, coordination, coherence and synergies among relevant stakeholders
• To enhance action and support, including finance, technology and capacity-building, to address loss
and damage associated with the adverse effects of climate change
The Agenda for Humanity, arising from the World Humanitarian Summit sets out five major areas to address and
reduce humanitarian need, risk and vulnerability, and 24 key transformations that will help achieve these five
major areas. It places a strong emphasis on managing disaster risk with one of the key transformations being to
anticipate crises, using data and risk analysis to take early action and thereby prevent and mitigate crises.
It also calls for, among other things, international frameworks and regional cooperation to ensure that countries
in disaster-prone regions are prepared to receive and protect those displaced across borders; greater support
for Small Island Developing States to prevent, reduce and address disasters resulting from climate change;
increasing domestic resources for risk management, including by expanding tax coverage, increasing expenditure
efficiency, setting aside emergency reserve funds, dedicating budget lines for risk-reduction activities and taking
out risk insurance; and for developed countries to dedicate at least 1 per cent of official development assistance
(ODA) to disaster risk reduction and preparedness activities by 2020.
INTR ODUCTION 13
A Toolkit for Disaster Risk Finance: Report Structure
e below schematic provides an overview of the
structure of the Toolkit. Disaster risk nancing (DRF)
instruments exist to fund the various costs of managing
disaster risk and set incentives for a behavioral change.
However, instruments dier signicantly in their cost,
how much nance they provide and how quickly they
can mobilise resources.
is implies, critically, that disaster risk nancing
instruments should not be chosen without an
understanding of the underlying disaster risk. is can
be achieved through a risk audit, as explained in section
1. Once the risk is understood, there are a range of
dierent actions that can be undertaken to manage that
risk: the risk can be reduced, the risk can be retained
with resources set aside to manage it, or the risk can be
transferred to others. Section 2 describes these options
in more detail, recognizing that the appropriate mix will
depend on the specic circumstances. Once the actions
have been chosen, they often require a range of dierent
nancial instruments and/or policy mechanisms. But
these nancial instruments and policy mechanisms vary
across a number of important dimensions. Section 3
explains the criteria that can be used to choose between
dierent instruments. Section 4 sets out the dierent
nancial instruments and evaluates them against the
criteria identied in section 3. Section 5 then explains
how instruments do not work in isolation and how
a disaster risk management strategy needs to
combine various instruments, and sets out the key
interdependencies between dierent types of
instruments and the action they facilitate.
Figure 3. A toolkit for disaster risk finance.
1 Risk Audit
Exposure Definition
Quantify risk and define
resilience target to enable
risk-informed action.
Peril Identification
Risk Quantification
Resilience Targeting
2 Disaster Risk Management Actions
Risk Reduction
Design a DRM plan,
consisting of risk reduction,
risk retention, and risk
transfer actions.
Risk Retention
Risk Transfer
3 Dimensions of Instrument Design
Risk Holder
Use situational analysis
to define underlying need
and inform instrument
requirements.
Purpose
Timing
Risk Level
4 Disaster Risk Finance Instruments
Taxonomy
Select appropriate DRF
instruments.
5 Risk Management Strategy
Complementarity
Combine DRF instruments to
create an efficient DRM strategy
using a risk layering approach.
Risk Layering
This page intentionally left blank.
DISASTER RISK FINANCE TOOLKIT: 1. Risk Audit 15
Disaster
Risk Finance
Toolkit
DISASTER RISK FINANCE – A TOOLKIT16
1. RISK AUDIT
A sound understanding of the underlying risk is
fundamental to eective risk management. Risk
managers – those people who implicitly or explicitly bear
the consequences if a risk materialises, and which can
include individuals, governments, and humanitarian
actors – should collectively undertake a risk auditing
process as the rst step towards developing an eective
risk management strategy.
Risk auditing consists of four phases; (i) dene the
exposure at risk to understand what needs to be managed;
(ii) identify what perils and hazards can impact that
exposure, (iii) quantify the expected frequency and
severity of impact from those perils, ideally using a
probabilistic risk analysis, and; (iv) set a resilience target
to identify the extent to which risks will be explicitly
managed.
is risk auditing process provides the foundation to
make eective risk-informed decisions. e phases are
summarised in Figure 4.
Figure 4. Risk auditing process.
Exposure
Definition
Define the exposure to risk in terms of its key characteristics:
• Location
• Vulnerability
• Value
Value can be quantified in a range of ways, for example in terms of number
of people or asset replacement cost, but also in terms of value to society,
or criticality for dependent systems.
Hazard
Identification
Identify the range of possible event types (perils), and the associated hazards.
Peril types may include:
• Shock events: rapid-onset events (e.g. tropical cyclone, flood, earthquake)
• Strain events: slow-onset events (e.g. drought, pandemic)
• Systemic events: events that occur as a result of multiple factors
(e.g. conflict, migration)
Risk
Quantification
Risk analysis is fundamental for developing a targeted risk management strategy.
For a given set of exposure and hazard types - risk models allow a quantified
understanding of the probability and severity of disaster impact to guide
decision-making.
Resilience
Targeting
Some events are so infrequent and severe that it would be prohibitively expensive
to
aim to manage, in advance, the entirety of the impact.
The resilience target describes the threshold between actively managed risk, and
unmanaged ‚residual‘ risk. As residual risk is ultimately retained by the risk holder,
the objective of a risk management strategy is to reduce the residual risk to a
‘tolerable’ level.
The resilience target can be measured in terms of ‘return-period’ impact,
for example a resilience target may be to actively manage risk up to the
1 in 250-year return period impact.
DISASTER RISK FINANCE TOOLKIT: 1. Risk Audit 17
e process of risk auditing should be approached in
an outcome-oriented manner. e data collection and
modeling exercises should therefore aim to provide t-for-
purpose information to support decision making.
is consideration is particularly important in regions
where there is an apparent lack of reliable exposure and
hazard data, and limited catastrophe risk model coverage.
In these cases, simple assumptions can greatly support
risk management, utilising lessons learned in analogous
regions to enhance the risk auditing process.
Furthermore, while risk modelling has relied on extracting
useful insights from large amounts of historical data
for a long time, new ‘big data’ and articial intelligence
techniques opens up the opportunity of utilising more
data sources and processing that information more
quickly and at lower cost
9
.
Importantly, risk management is an iterative process – the
dierence between no risk-information and some simple
risk-information generated using basic assumptions can
be signicant. As a rst step, an order of magnitude
level risk audit, combined with an appreciation of
assumptions and limitations, still allows risk managers
to make substantially more informed decisions. Simple
assumptions might include local estimates of population,
property construction types and values, and historical
or scenario-based impact assessments. ese simpler
analyses can provide good initial insight, and pave the
way for more advanced data collection and risk modeling
exercises.
An illustrative scenario is provided in
→ Box B to show
how risk auditing can be applied in practice.
DISASTER RISK FINANCE – A TOOLKIT18
Box B. Illustrative Risk Audit
Scientific research and observations from previous disaster impacts provide the data necessary to build catastrophe
risk models, which estimate the probability and severity of potential disaster impact. Catastrophe models provide a
framework in which it is possible to quantify and compare the risk from a range of perils, enabling greater insight into
the drivers of risk.
The below table outlines the application of a risk auditing process of definition, identification, quantification, and
targeting, using a state-of-the-art catastrophe risk model to create an illustrative risk analysis.
The modelled risk analysis results for a set of assets are shown in Figure 5 using an ‘exceedance probability’ (EP)
curve.
EXPOSURE DEFINITION
What is at risk?
The analysis covers commercial-type properties in a Southeast Asian country.
The data includes information about:
• The location of people
• The location of assets (including residential property, business and commercial
properties and infrastructure)
• Key determinants of the vulnerability of people – including:
– Gender
– Age
– Proportion affected by disabilities
– Other vulnerable groups
• Key asset characteristics, which inform their vulnerability – including:
• Construction (dominant material used in constructing the building frame/structure)
• Occupancy (typical use of the building)
• Year built (captures building practices/regulation and deterioration)
• Number of stories
• Replacement value – in relation to assets, describes the cost to rebuild,
including both the structure and value of contents.
PERIL IDENTIFICATION
What can cause impact?
The analysis focuses on two climate-related peril (typhoon, and inland flood) and one
seismic peril (earthquake). The secondary hazards associated with these perils include:
• Typhoon: wind, coastal flooding from storm surge, typhoon-induced coastal and
inland flooding
• Inland food: non-typhoon pluvial and fluvial flooding from excess rainfall
• Earthquake: ground shaking
RISK QUANTIFICATION
What is the frequency and
severity of impact?
Catastrophe risk models can quantify the risk of direct damage and loss to assets.
The risk analysis results are presented in an exceedance probability curve (→ Figure 5).
Of course, direct physical damage is only one component of a disaster impact with
loss of lives and livelihoods and downstream impacts also of crucial importance.
Physical damage is, however, often a good indicator for the total potential impact
from all sources, including direct and downstream impacts. ‘Disaster Impact’ is
used to describe all potential impacts.
RESILIENCE TARGETING
What is the risk tolerance
level?
Resilience targeting sets the threshold between the risk which will be actively
managed using a DRM strategy, and the level of ‘residual risk’, which falls beyond
active risk management.
The level of the resilience target depends on the risk tolerance of the risk holder,
and other practical considerations including available budget and regulatory
requirements. An example resilience target is shown at the 200-year return
period impact.
DISASTER RISK FINANCE TOOLKIT: 1. Risk Audit 19
Figure 5. Aggregate Exceedance Probability (AEP) curve for illustrative scenario (source: RMS).
Exceedance Probability Explainer
The exceedance probability curve is an analytical tool used to describe the frequency-severity distribution of
disaster impact. There is typically an inverse relationship between disaster severity and frequency of occurrence,
i.e. the more severe an event, the less frequently it is expected to occur.
• Frequency (x-axis): ‘Return Period’ thresholds are used to describe the frequency of occurrence.
The Return Period (year) is equivalent to
1
⁄
(Exceedance Probability (year
-1
)
.
• Severity (y-axis): ‘Disaster Impact’ is used to describe the total annual aggregate disaster impact. Direct physical
damage and loss is used here as an indicator for total disaster impact (including indirect impacts). Severity is
often measured in financial terms ($ loss), though other metrics can also be used as appropriate (e.g. number of
casualties, storm category, flood extent).
Any point along the exceedance probability curve can be read as “there is a 1 in X-year annual probability of
exceeding a disaster impact of Y”. Note that while the combined exceedance probability curve consists of the risk
from all three perils, is not equivalent to the sum of the independent peril exceedance probability curves. This is
expected due to the methods used to calculate AEPs.
DISASTER RISK FINANCE – A TOOLKIT20
2. DISASTER RISK MANAGEMENT ACTIONS
Once the risks are understood, it is possible to develop a risk
management strategy around three core categories of actions:
(1) risk reduction; (2) risk retention, and; (3) risk transfer.
Figure 6 describes these actions in more detail.
Figure 6. Risk management actions.
Risk
Reduction
Any ex-ante action that reduces the severity of disaster impact. Risk reduction
activities include physical interventions such as building flood defences and
retrofitting property, but also planning activities such as risk-based site selection
for new developments, and evacuation and response plans. It can also include
activities taken immediately before an event impacts such as the distribution of
hygiene kits and water purification tablets, or preparatory actions taken based on
near or long-term forecasts.
The decisions about which risk reduction activities to undertake, in which localities
and to the benefit of which groups should be taken following a combination of
economic feasibility assessments and participatory processes that allow opportunity
for all voices to be heard.
Risk reduction has benefits for all severities of disaster - however the relative size
of the benefit in terms of reduced impact can vary depending on event severity.
Risk
Retention
After an event has occurred, some costs can be financed directly by the risk holder
using funds that are readily available. Risk retention mechanism are a relatively
reliable source of funds, and they are therefore most appropriate to support more
frequent disaster costs, such as those that are expected to occur every 10 years or
less.
In order for funds to flow quickly, the rules concerning how the resources associated
with risk retention mechanisms are allocated should be determined prior to the event,
and, as far as possible, be informed by data. The rules should be determined in an
open, consultative manner.
Risk retention mechanism have longer term cost implications, in that the costs are
held and repaid by the risk holder, potentially for years after an event has occurred.
Risk
Transfer
For lower-frequency higher-severity disasters, it is relatively more uneconomical
to use risk retention mechanisms. Risk transfer mechanisms remove a portion of
disaster risk in return for an annual premium payment. As such, they redistribute the
infrequent and unmanageable total cost of disaster, into an equivalent manageable
annual cost (premium). After an event, if the payment terms of the instrument are
met, funds are paid by the risk transfer provider to the risk holder.
As with risk retention, decisions as to how the resources associated with the use of
risk transfer instruments (after they are triggered) should ideally be taken in advance
(as far as possible) and following an open, participatory consultation process.
DISASTER RISK FINANCE TOOLKIT: 2. Disaster risk management actions 21
Risk reduction is core to disaster risk management, as it
directly reduces the severity of potential disaster impacts,
saving lives and reducing the destruction of homes and
critical infrastructure. However, in reality risk reduction
activities alone are unlikely to be able to reduce residual
risk to meet resilience targets.
Risk retention and risk transfer tools provide additional
options to manage any residual disaster risk. In all
three cases, the decisions as to who should benet from
these dierent actions, and how the actions should be
implemented, need to be taken in a participatory fashion
that provides full representation for those most exposed
and vulnerable to the risks.
ese three actions should be applied in combination
in order to meet dened resilience targets. e specic
combination of actions this requires will be context
specic, and informed by both cost benet analysis
as well as through participatory engagement processes
with local communities, especially the most vulnerable.
→ Section 5 discusses how to combine DRM actions and
DRF instruments eciently and eectively.
ese three types of action are also part of the ACRI+
and International Red Cross and Red Crescent Movement
(ICRM) disaster risk management ‘cycle’. However, this
toolkit separates risk retention and risk transfer whereas
the ACRI+ cycle combines these two elements. In
addition, the framework in this paper distinguishes how
the risk is managed, from the time at which actions are
taken (which is discussed in section
→ 3.2) whereas the
ACRI+ cycle combines these elements. is distinction
between which actions are taken and when they are
taken is powerful when explaining the dierences
between dierent nancial instruments. However, both
frameworks essentially incorporate the same elements.
DISASTER RISK FINANCE – A TOOLKIT22
3. DIMENSIONS OF INSTRUMENT DESIGN
Disaster Risk Financing (DRF) instruments exist to
support the various funding needs associated with disaster
risk management. In practical terms, these instruments
fund or facilitate risk reduction, risk retention, or risk
transfer actions. Dierent instruments are more or less
suited to these dierent actions.
However, DRF instruments also vary according to a range
of other criteria. ese include; (i) the needs and capacity
of the risk-holder (individuals, sovereigns or somewhere
in-between, as well as development and humanitarian
actors); (ii) the ultimate purpose for the funds, (iii) the
required timing of support relative to a disaster; and;
(iv) the level of risk that they help support.
A basic situational analysis can be performed by asking
the following questions.
Figure 7. Instrument design dimensions.
Risk Holder What is the capacity and need of the risk holder?
Purpose What will funds be spent on?
Timing When is funding needed?
Risk Level What level of risk is being addressed?
e answers to these questions can help to inform the
risk holder about which DRF instruments are most
appropriate for the underlying need. ey can also help
articulate the design requirements for individual DRF
instruments. However, the factors which inuence
DRF instrument design are complex and often
interlinked and, as a result, the criteria share some
intersecting themes. e following sections discuss
each of these dimensions in more detail.
DISASTER RISK FINANCE TOOLKIT: 3. Dimensions of Instrument Design 23
3.1. Risk Holder
Disasters impact people and organisations at all scales,
from the farmer to the nance minister.
e needs of the risk holder vary across this range of scales,
as does the nancial and technical capacity to purchase
and maintain DRF instruments as outlined below:
Figure 8. overview of needs and typical technical and financial capacity of risk holders.
Risk Holder Overview
INDIVIDUAL
(personal, household,
smallholder, SME)
At an individual level people are responsible for the wellbeing of themselves and
their families, property including homes and possessions, and their livelihoods. This
might include individual households, smallholders and small and medium-sized
enterprises (SME).
This risk holder has a limited budget, and less need to access sophisticated DRF
instruments.
The types of DRF suitable at an individual level are typically standard consumer
products, including property & life insurance, and loans. Micro-finance has been
developed to address those with limited capacity to pay, especially in developing
countries.
COMMUNITY
(groups of individuals
or businesses,
towns, villages)
The pooling of individual risk and resource increases the range of DRF instruments
that are available to fund DRM at a local level.
Coordinated groups of individuals and businesses, and local authorities have greater
purchasing power and can carry out resilience actions on a greater scale.
The range of responsibilities also increases to include restoration of services,
in order to minimise impacts on population or employees.
Community level DRM initiatives may be supported by external entities, who can
provide greater technical support, more funding, and access to a wider range of DRF
instruments.
MUNICIPALITY
(cities, sub-national
government)
Municipalities are often responsible for supporting large urban populations.
This includes the provision of critical and essential services such as power, water and
waste management, transport, education, emergency, social and healthcare services.
Municipalities can receive income through taxation, and often have independent risk
management capacity, and additional technical and financial support from national
governments.
Municipalities have capacity to purchase a broad range of DRF instruments, across
a range of markets. They can also coordinate and incentivise DRM activities at the
individual and community level, as well as influence national DRM practices.
SOVEREIGN
(state, supra- national
entity, international body)
Sovereign entities are ultimately responsible for the welfare of their populations,
development outcomes, and for near and long-term economic productivity.
The financing needs at a sovereign level are significant, but so are the available
DRM activities and DRF instruments. Sovereign entities can employ budgeting
mechanisms and issue debt, build disaster reserves, and implement risk management
policy and regulation among other activities.
Sovereigns can benefit from international financial, technical and operational support
from supra-national agencies, development banks, as well as international aid.
DISASTER RISK FINANCE – A TOOLKIT24
A discussion of dierent potential risk-holders raises
important questions about the role of humanitarian
actors. is is discussed further in Box C.
Box C. Stakeholders
Humanitarian actors receive funds from public donors and private sources, to enhance, support or substitute
for in-country responses to a population in crisis. They include local and international non-governmental
organizations, UN humanitarian agencies, the International Red Cross and Red Crescent Movement, host
government agencies and authorities, and donor agencies. Humanitarian actors work according to four key
principles:
• HUMANITY: human suffering must be addressed wherever it is found. The purpose of humanitarian action
is to protect life and health and ensure respect for human beings.
• NEUTRALITY: humanitarian actors must not take sides in hostilities or engage in controversies of a political,
racial, religious or ideological nature.
• IMPARTIALITY: humanitarian action must be carried out based on need alone, giving priority to the most urgent
cases of distress and making no distinctions on the basis of nationality, race, gender, religious belief, class or
political opinions.
• INDEPENDENCE: humanitarian action must be autonomous from the political, economic, military or other
objectives that any actor may hold regarding areas where humanitarian action is being implemented.
Historically, the role of humanitarian actors has been to step in following a crisis, when a risk holder has not
been identified, or when the magnitude of the risks overwhelm the ability of a purported risk holder to respond
to the realisation of that risk. In these cases, humanitarian actors provide indispensable services and support
to minimise the human cost of the event.
While this still represents a core role for humanitarian actors, in recent years, there has been a deliberate
attempt to move beyond this role. At least three additional roles can be identified:
• To support national actors to better understand the risks that they face and develop disaster risk management
plans, and associated financing strategies. The Agenda for Humanity
7
encourages humanitarian actors to work
alongside development partners, national governments and other partners with the aim of ‘strengthening local
and national response in risk-prone countries outside of crises’ It recognises that ‘Investment in data and risk
analysis should be increased and action taken early to prevent and mitigate crises.’ This is a key area in which
humanitarian and development actors have sought to work more closely.
• To explicitly become one of the actors within the plans developed ahead of crises – in other words to become
an explicitly identified risk-holder that ex ante commits to provide resources when risks materialise, and/or as
important actors in implementing risk reduction, response and recovery activities. This is broadly similar to the
‘traditional’ role played by these actors, but in a way that is explicitly incorporated within a broader disaster
risk management plan. This has been associated with a shift towards anticipatory finance, as discussed below.
• To encourage greater societal participation in decisions about disaster risk management strategies, recognising
that humanitarian actors can often play a crucial role in ensuring that otherwise marginalised and vulnerable
people can have their needs taken into account
10
.
DISASTER RISK FINANCE TOOLKIT: 3. Dimensions of Instrument Design 25
Red Cross Red Crescent and its role in anticipatory finance
The Red Cross Red Crescent Climate Centre (RCCC) has applied lessons learned from pilot projects to inform
the development of a model of providing humanitarian finance in anticipation of an extreme event
11
.
This involves identifying triggers, Early Action Protocols (EAPs) and an associated financing mechanism.
1
TRIGGERS
Region-specific “impact levels” are identified based on the detailed risk analysis of relevant natural
hazards, impact assessments of past disaster events, and vulnerability data. A trigger model then
determines priority areas where the impact of an extreme weather event is anticipated to be most severe.
Box D in section 4 explores the use of this sort of trigger mechanism, compared to those conventionally
used for disaster risk finance in more detail.
2
EARLY ACTIONS
Once a forecast exceeds the trigger, a pre-agreed set of early actions, specified in an Early Action
Protocol, are undertaken. These actions are aimed at reducing the impact of the predicted event on human
lives, by providing assistance to people at risk and helping them to protect their families and livelihoods.
This can include, for instance, providing veterinary kits, tying down house roofs, providing food and clean
water, as well as transferring cash.
3
FINANCING MECHANISM
A Forecast-based Action Fund automatically allocates funding once a forecast reaches a pre-agreed
danger level to enables the implementation of the Early Action Protocol.
DISASTER RISK FINANCE – A TOOLKIT26
3.2 Purpose
e ultimate purpose of DRF is to fund or facilitate
resource ows towards activities that make disasters less
impactful for people.
is can be achieved by minimising the risks to
populations through reduction in vulnerability and
volume of exposure; reduction frequency and severity
of hazard; strengthening of disaster preparedness and
response plans; and increasing the speed and eectiveness
of recovery, among other activities.
Disaster risk nance provides the funds which enable
these disaster risk management activities. e specic
purpose for the funds has implications for which DRF
instruments are appropriate, and further for the design
of individual instruments (instrument mechanics).
It can be challenging to clearly segment and dene
purpose, given that disaster management costs are diverse
and interconnected. In reality funds from individual DRF
instruments are often used for a mix of activities, and
instruments can be designed to accommodate multiple
purposes.
Nevertheless, the exercise of ‘purpose mapping’ can
help to guide both DRF selection and design processes.
e following three categories are selected to capture
the main purpose groups.
Figure 9. DRF Purpose groups.
Purpose Overview
Life and
Livelihood
Injury, death, and disruption resulting from disaster are the most immediate and pressing
impacts of a disaster. There are immediate impacts for those directly affected, but also for
municipalities and sovereigns who have responsibilities for the wellbeing of their populations.
The costs required to fund life and livelihood impacts are diverse, and relatively challenging
to quantify ahead of an event.
DRF instruments designed to support this purpose should be flexible enough to reflect impact
and needs assessments.
Operations
Disaster management activities have a range of implementation costs, including costs of
personnel and resources required both before and after a disaster. Ensuring that these are met
is crucial both to reducing the impact of a disaster and to ensuring that any negative impacts
from a disaster are quickly dealt with, and helping to avoid detrimental impacts for longer-term
economic and developmental outcomes.
Funding to support operations must be readily available at the point of need. Prior to an event
funding for operations can be directed towards disaster response and contingency planning.
In the time-critical phase leading up to, during, and immediately following a disaster, rapid
access to sufficient levels of funding for operations can significantly mitigate the overall
severity of impact.
Capital liquidity and certainty of payout are key considerations when designing DRF for operational
costs.
Physical
Assets
Physical assets are exposures that can be directly damaged. This damage can have drastic
impacts on the ability of people to meet their basic needs and access essential services such
as water and sanitation, education, or health services. The costs associated with physical
assets include the costs of development, maintenance, repair, replacement of property such
as buildings and infrastructure, property, machinery, and environmental assets.
The costs and risk associated with physical assets are typically most easily quantified.
Catastrophe models are designed to capture direct physical damage, and the downstream
impacts from damage such as business interruption and casualty losses.
DRF to fund physical assets should aim to closely match the total financial needs of the DRM
action, be it the cost of construction or retrofit, or rebuild/ replacement costs following damage.
DISASTER RISK FINANCE TOOLKIT: 3. Dimensions of Instrument Design 27
3.3 Timing
Dierent instruments facilitate access to funds at dierent
speeds, and to varying levels of funding. is means that
they are more or less appropriate for use at dierent times
relative to a disaster event.
is analysis distinguishes between three phases
ii
:
– A preparatory phase where it is not urgent to access
funding immediately but where relatively small
amounts of funding can signicantly reduce the
direct and downstream impacts of a disaster, both
in terms of the lives that will be aected, and the
asset damage that may be realised.
– A response phase where funding needs are urgent
in order to reduce the overall impact of the event,
especially the impact on lives and livelihoods. During
this time critical period it is important that risk
management activities are not dependent on DRF
instruments which take a long time to release funds.
– A recovery phase during which funding needs can
be substantial, especially if there has been signicant
damage to physical assets and infrastructure, but the
urgency of accessing that funding is not so great.
Figure 10 provides a stylised representation of the scale
and timing of these needs. Figure 11 outlines the types
of activities that occur within each phase.
Figure 10. Schematic of Illustrative timing and volumes of funding associated with each phase.
Response
Recovery
Preparation
Disaster Impact
time
DISASTER RISK FINANCE – A TOOLKIT28
Figure 11. example activities associated with preparation, response, and recovery phases.
Timing Activities
Preparation
– Continuous costs of disaster reduction
– Preceding a forecast event impact (using near or long-term forecast data)
·
Evacuation
·
Deploying defences
·
Initiating disaster response plans
Response
– Immediately following disaster impact
·
Search and rescue
·
Humanitarian services
·
Restoration of essential services
Recovery
– Longer-term post-disaster
·
Reconstruction
·
Social support
DISASTER RISK FINANCE TOOLKIT: 3. Dimensions of Instrument Design 29
3.4 Risk Level
e relative cost eectiveness of DRM actions and
DRF instruments vary according to the frequency-severity
prole of the underlying risk.
e following risk level bands are indicative only –
a comprehensive risk audit and expert guidance is
ideally used to provide context-specic guidance
for selecting risk-appropriate DRF solutions.
FIGURE 12. Indicitive risk levels.
Risk Level Overview
Annual
Risk holders who are responsible for large volumes of risk from multiple sources, such
as municipalities and sovereigns, can expect to incur at least some level of disaster
impact on an annual basis.
This type of yearly (‘attritional’) risk can be measured based on previous experience,
and so should be accounted for using established annually recurring DRF instruments.
Budgeting mechanisms and allocated disaster funds are an efficient and effective
means of managing yearly costs.
Risk reduction actions (including maintenance, simple retrofit, and planning, as well
as early actions immediately prior to an event such as preparation of emergency
shelters), can also be very effective in managing attritional disaster impacts.
HIGH-FREQUENCY
LOW-SEVERITY
(1 to 10-year
return period)
For less frequent events which cause impacts in excess of the yearly expected level,
annual budgeting may not be the most cost-effective option for managing risk.
Disasters which occur on a return period of up to 10 years are still relatively frequent.
In isolation, and depending on the country context, the levels of loss they cause might
fall within a ‘manageable’ level relative to the risk holder’s capacity to pay using
ex-post mechanisms. However, the uncertainty associated with disaster occurrence
can easily make potentially manageable losses very unmanageable if events occur
in succession.
MODERATE-FREQUENCY
MODERATE-SEVERITY
(10 to 50-year
return period)
Moderate severity event impacts typically fall beyond a risk holder’s capacity to pay
using available capital reserves. For less-frequent events more sophisticated DRF is
required to manage the potentially significant levels of impact.
Funding may have to be sourced from external providers, including international lenders.
Risk reduction activities must also be more robust to significantly reduce the risk for
more severe impacts.
LOW-FREQUENCY
HIGH-SEVERITY
(50+ year return period)
Low-frequency high-severity events can cause catastrophic impacts which generate
significant funding needs for large risk holders.
This level of impact is likely to far exceed a risk holder’s ability to build sufficient
disaster reserves. Risk transfer offers an effective means of moving risk off the risk
holder’s balance sheet.
Depending on the local context, the international reinsurance and capital markets may
offer the most affordable risk transfer options. The bundling of risk in sovereign-level
risk pools can also be effective.
DISASTER RISK FINANCE – A TOOLKIT30
4. DISASTER RISK FINANCE INSTRUMENTS
is section explores a range of nancial instruments
and policy mechanisms that can be used within a
disaster risk management strategy.
Building on the discussion above, it categorises these
instruments and policy mechanisms into those that
can fund or facilitate risk reduction (in relation to
climate-change related risks, this represents ‘adaptation’
to climate change); those used for risk retention;
and risk transfer instruments.
e taxonomy also characterises appropriate risk
holders, timing, purpose, and risk levels that each DRF
instrument or policy is tailored to support. In doing
this, it recognises that the instruments often have a
range of structural options, which will vary depending
on the specic needs and circumstances of the user.
Dierent options mean that some instruments or policy
mechanisms can be used across a range of scales and
purposes and can be structured to respond to dierent
requirements associated with timing and risk level.
Finally, it also provides examples of how the instruments
have been used in practice, drawing, in particular, on
examples from developing countries.
e taxonomy presented in Figure 13 summarises the
appropriate range of application for each of the DRF
instruments.
Figure 13. Taxonomy of disaster risk finance instruments, categorized by risk management action and design criteria
Risk Holder Risk Level Timing Purpose
What is the capacity and need
of the risk holder?
What level of risk is being
addressed? (return period)
When is funding needed? What will funds
be spent on?
Action Instrument Individual Community Municipality Sovereign Life &
Livelihood
Operational Physical
Assets
Preparation Response Recovery Annual 1-10
year
10-50
year
50+
year
Risk Reduction
Loan
• • • • • • • • • •
Micro-credit
• • • • • • • •
Bonds
• • • • • • • • •
Grants, subsidies,
& tax breaks.
• • • • • • • • • • •
Crediting
• • • • • • • • • • •
Impact Bonds
• • • • • • • • • • • • •
Risk Retention
Budget
Contingency
• • • • • •
Reserve Funds
• • • • • • • • • • •
Contingent Loans
• • • • • •
Risk Transfer
Micro-insurance
• • • • • • • • •
Agriculture
Insurance
• • • • • • • •
Takaful & Mutual
Insurance
• • • • • • • • • • • •
Insurance &
Reinsurance
• • • • • • • • • • • •
Catastrophe Bonds
• • • • • • • • •
Risk Pools
• • • • • • • • •
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 31
4.1. Risk Reduction
is section consists of two components: rst it considers
a range of nancial instruments that are commonly
used to structure the ow of capital into investments
that will reduce the risks that disasters cause; then it
explores a range of policy mechanisms that governments
or development partners can use to make it more
economically attractive to undertake such investments,
using various types of nancial instrument.
Risk Reduction: Financial Instruments
Loans
Individual
to sovereign
Primarily to reduce
risks to physical assets but
can also be used
to reduce risks to
lives and livelihoods
Preparedness
activities plus
recovery
Most effective
at reducing risks
from frequent
(annual or up to
1 in 10-year events)
OVERVIEW
Bank loans are one of the most common instruments for channelling capital into risk-reduction,
and other types of, investments. They can be made by either public or private financial institutions
(FI) and provided to companies, households or other institutions. They are primarily used to finance
investments that reduce risk in preparation of a disaster event but can also be used to finance
reconstruction after a disaster event (where risk reduction is achieved by a commitment to ‘build back
better’). Loans supporting investments that reduce risks are likely to be proportionally more effective
at reducing risks from high-probability, low-severity events; more extreme events are typically so
devastating that risk-reduction investment is less effective
12
.
Regardless of use, the borrower is expected to repay the loan, plus make interest payments on the
balance of the loan that has not been repaid. On some occasions, the FI advancing the loan will
receive the capital to make the loan through a credit-line provided by an International Financial
Institution (IFI). This credit line will provide resources to the FI on more favourable terms than it
could otherwise access, on condition that loans are advanced for a particular purpose.
DESIGN
OPTIONS
The key design characteristics influencing the nature of the loan are the amount advanced; the
duration (tenor) of the loan; the repayment schedule; whether the loan is secured on the asset that
it finances (or other collateral) such that the FI can claim the asset in the event that the borrower
defaults; and the interest rate, and other pricing, charged on the loan. In cases where loans are
supported by IFI credit lines, the IFI may require that the loans offered to the final borrower are
priced on more favourable terms than would otherwise be available in the market.
CHALLENGES
Loans are a very well-known financial instrument used to finance a wide range of capital investments.
As such, the potential challenges in using the instrument are well known. Most importantly, if the
borrower is unable to repay the loan, either because the asset does not perform or otherwise, then
this can cause problems of indebtedness for the borrower and reduces the profitability of the financial
institution, making it more reluctant to lend in the future. Some households and businesses can
also find it difficult to access loans, either because the FI finds it difficult to judge the likelihood of
repayment, or because the distribution channel of the FI does not reach those who would like a loan.
REQUIRE-
MENTS
FIs need to be licensed by, and are subject to supervision from, the national bank authorities in
the countries in which they make loans, influenced by international bodies such as the Bank for
International Settlements‘ Basel Committee on Banking Supervision.
EBRD CLIMADAPT
The European Bank for Reconstruction and Development’s (EBRD) ClimAdapt programme in Tajikistan provides
a good example of how loans, supported by an IFI credit line, can support risk reduction investment
13
. In this
initiative, the EBRD, with the support of various donors, has advanced a $10m credit line to a selection of banks,
who then provide loans to local businesses and households to invest in projects that reduce climate-related risks.
At the time of writing, more than 3500 projects had been supported, with investments in water efficient
technologies, energy efficiency and sustainable land management practices.
DISASTER RISK FINANCE – A TOOLKIT32
MICRO-CREDIT
Individual
and community
Lives and livelihoods,
and small scale
physical assets
Preparedness
activities
plus recovery
Most effective at
reducing risks from
frequent (annual or up
to 1 in 10-year events)
OVERVIEW
Micro-credit involves the provision of relatively low value, frequent repayment loans to individuals,
households, SMEs and communities. The product arose as a reaction to the difficulty that
conventional FIs are unable or unwilling to provide loans to this target customer group. Micro-credit
is typically provided by dedicated micro-finance institutions (MFIs) who are financed by commercial
lenders and for-profit investors, multilateral and bilateral development banks, and donors. Donors
and IFIs may also provide additional support to specific microfinance programs to reduce costs or risks.
A typical characteristic of microfinance is the engagement of the community within the loan appraisal
and monitoring process through, for example, joint liability or peer monitoring. Microfinance also
often specifically targets women. On many occasions, loans are one of a series of financial products
the MFI offers, others include micro-insurance (see discussion on microinsurance below).
MFIs are beginning to consider the use of some of the risk transfer instruments described below,
or alternatively donor support, so that they are in a better position to extend loans quickly after
a climate shock – so called recovery lending. Early results suggest promise
14
.
DESIGN
OPTIONS
There are a number of design elements that influence the microfinance loan. These include whether
the loans must be used for specific activities, the duration (tenor) of the loan, the interest rate
charged and the distribution channel. There is an increasing interest in using mobile banking
solutions to improve access to microcredit by lowering distribution costs.
CHALLENGES
Researchers have extensively analysed the impact of microfinance with conflicting results. Various
studies find no significant impact on poverty or other development indicators; while there are also
concerns about the potential indebtedness of consumers. On the other hand, microfinance (including
microcredit) has been associated with an enhanced ability of poor people to deal with shocks, but
this is not universal
15
.
Microfinance programs specifically targeted at reducing climate risks are in their early stages.
They offer significant potential, although there are challenges in enhancing awareness regarding
the value of risk reduction investments across all stakeholders, finding distribution models that
reach the most climate vulnerable and, when programs are supported by public funds, ensuring
loan repayments.
REQUIRE-
MENTS
Most countries have introduced regulation to license and supervise microfinance institutions,
especially in cases where the MFIs take deposits as well as advance credit.
JAMAICA PPCR AND OTHER EXAMPLES
In Jamaica, the Pilot Program for Climate Resilience (PPCR), working through the Inter-American Development
Bank, has underwritten microfinance loans extended to farmers and small enterprises in the tourism and
agricultural sector
16
. These loans have, among other things, supported farmers in installing dams and grass
and live vegetation barriers.
However, even in cases where micro-credit is not explicitly targeted at investments that reduce climate
risks, they can be an important tool to build livelihoods and assets that enhance broader adaptive capacity
to climate risks
15
.
The investments supported by micro-credit are most likely to be effective at reducing frequent, relatively
low-intensity events.
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 33
BONDS
Municipality and
sovereign (plus
large corporates)
Physical assets
Preparedness
activities
plus recovery
Can be used to fund
more significant
infrastructure projects (all
risk levels)
OVERVIEW
Bonds are issued by national and local governments, and other quasi-public organisations, as well
as large companies, to finance investment. In exchange for the payment of the bond by the purchaser,
the issuer agrees to pay the purchaser interest payments on a set schedule, and repay the principal
at maturity. As such, they are a form of debt instrument. They are attractive to investors as low-
risk securities, depending on the sponsor, that can be easily traded. Due to their expense (see
below), bonds are typically used for financing large scale capital infrastructure, either supporting
preparedness by reducing risks prior to an event, or for less time-sensitive reconstruction of assets.
Bonds can be classified according to who issues the bond (government, municipal, corporate) as
well as according to the use of proceeds from the bond sale. In recent years, there has been a
significant growth in green bonds: bonds that are explicitly issued in order to finance projects that
are environmentally sustainable or support the mitigation of or resilience to climate change. Climate
Bonds Initiative reports that, as of 2018, there were around $1.45 trillion of bonds that claim links
to addressing climate change, although less than 0.1% have an explicit focus on reducing risks to
climate change
17
.
DESIGN OPTIONS
A number of features define the specific characteristics of the bond. These include: size; the use
of proceeds; whether repayment will come from general sources (either corporate cashflow or tax
revenues) or from the specific revenues generated by the financed asset(s); the duration of the bond;
and the interest rate (coupon) that will be paid to investors.
For green bonds, the Green Bond Principles (GBP) provide voluntary process guidelines to issuers for
launching a credible Green Bond. The Principles cover defining criteria for a green project, defining
the processes for selecting green projects, the systems used to trace the green bond proceeds, and
reporting guidelines. The principles also identify that issuers have the option to ask third parties to
certify their green bond, using organisation such as the Climate Bonds Initiative. These organisations
will assess the bonds against pre-agreed criteria, especially related to how the proceeds will be
used. This increases the green credentials of a bond among investors, but also increases transaction
costs
18
.
CHALLENGES
Bonds are expensive to structure, with transaction costs typically of 1% or more of the principal
raised. They take several months to structure. These costs and time increase further if the bond is
certified. This tends to mean that it is only somewhat richer developing countries that issue sovereign
bonds, although the IMF reports that in the 10 years to 2013, Rwanda, Tanzania, Senegal and Cote
d’Ivoire all issued sovereign bonds19 while Nigeria and Fiji have recently issued sovereign green bonds.
REQUIRE-
MENTS
Bond issuance is typically regulated by the capital market authorities in the country where the bond
is issued. The economic aspects of this regulation might, for instance, place nationality restrictions
on who is allowed to issue or purchase bonds within a jurisdiction, whether or not a prospective
bond issuer meets necessary standards; and taxation rules. Prudential regulation focuses on investor
protection and avoiding systemic risks, by identifying principles for, for example, issuance standards
or trading norms.
GROWING GREEN BOND MARKET
Despite constituting a very small proportion of the overall bond market, there are a number of important
examples of institutions issuing bonds to reduce climate risks. For example, the Government of Fiji issued a
$50m green bond which will primarily be used for investments that build resilience against the impacts of
climate change
20
(as well as renewable energy projects) while the City of Cape Town issued a $76m green
bond in July 2017 to refinance a number of assets, including the rehabilitation and protection of coastal
structures
21, 22
.
DISASTER RISK FINANCE – A TOOLKIT34
Risk Reduction: Policy Mechanisms
GRANTS, SUBSIDIES, & TAX-BREAKS
Individual
to sovereign
Physical assets
and lives and
livelihoods
Preparedness
activities plus
recovery and
possibly response
Most effective at reducing
risks from frequent
(annual or up to 1 in
10-year events)
OVERVIEW
A key way to increase the attractiveness of risk reduction activities is through grants to reduce their
capital costs, subsidies to reduce their ongoing operating costs, or tax breaks. These can be used
to support investments that reduce the exposure of both infrastructure and lives and livelihoods
to extreme weather events, and at all scales from individuals through to sovereigns. They are best
suited for preparedness activities or to support asset reconstruction, although quickly arranged
grants may also help with response activities.
DESIGN OPTIONS
Most grants, subsidies or other incentives take a relatively simple form whereby the payment is made
concurrently, or in advance of, when costs are incurred. Large grant payments may be disbursed in
separate tranches and made conditional on evidence that the previous tranche has been used as
intended.
There also is growing interest in making ‘results-based’ grants or incentive payments, whereby
payments are only made when the outputs expected from undertaking a set of activities have been
delivered. This mechanism can help strengthen the incentives of the recipient (they only get the
additional resource if they deliver results) but it can be challenging in contexts where the recipient
faces challenges in accessing upfront finance. A further challenge is identifying metrics that can be
used to demonstrate that the activities have successfully reduced risks.
CHALLENGES
By improving the economics of undertaking risk reduction investments, they can be powerful
mechanisms to encourage such activity. However, activities may become reliant on the incentives
that, over time, can threaten the financial sustainability of the mechanism.
In terms of international climate finance, almost all developing countries have in place the
institutional architecture to engage with relevant multilateral funds. However, there are frequent
criticisms that the requirements to access resources from these funds are too burdensome for many
countries.
REQUIRE-
MENTS
The regulatory requirements for the domestic use of grants, subsidies or tax breaks are relatively
light and will generally already be in place to provide incentives for other activities. Results-based
incentive payments typically require more onerous regulatory regimes, in order to generate assurance
that the result that warrants the incentives has been delivered.
GRANT & SUBSIDIES
Canada has established a 10-year $2 billion Disaster Mitigation and Adaptation Fund that aims to increase
community resilience to natural hazards and extreme weather events
23
. It provides grants of between 25% and
75% of the eligible costs of infrastructure projects costing more than C$20m that serve to reduce risks. For
developing countries, international climate finance is an important source of grants to make risk resilience
investments more attractive to both public and private sectors. For example, the Adaptation Fund provides
grants of up to $10m to country governments for adaptation investments, including those that reduce the risks
from extreme weather events
24
. For example, a $5m grant is helping to enhance resilience and reduce the risk
of flooding in Ulaanbaatar City in Mongolia, primarily through the construction of various community level flood
protection assets
24
. Similarly, the Green Climate Fund (GCF) provides grants to support adaptation investment,
potentially of a larger scale
25
. For example, the GCF will provide a grant of $27.1m to support a $70.3m project
to scale up Georgia‘s Multi-Hazard Early Warning System to provide reliable information on climate-induced
hazards, vulnerability and risks.
In addition, both the funding received by humanitarian and the way that this funding is passed on to support
governments, municipalities, communities and individuals manage and reduce disaster risk is also typically
provided in the form of grants/subsidies. Box D describes the growing trend for some of this support to be
provided in advance of disasters striking, through anticipatory finance mechanisms such as forecast-based
financing.
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 35
CREDITING (MITIGATION BANKING)
Municipality &
community – with
spillovers to individuals
Physical assets
and lives
and livelihoods
Preparedness
activities
plus recovery
Can be used to fund
more significant projects
(all risk levels)
OVERVIEW
This approach incentivises risk reduction investment by allowing the benefits from these projects
to be recognised in a ‘credit’, that can then be sold to (typically) companies. Companies choose to
purchase the credits either for regulatory compliance purposes or corporate social responsibility
reasons. The sale of the credit boosts the revenue from undertaking the investment, making it more
economically attractive. Indeed, in some cases, the credit sales may be the only revenue source for
the risk reduction project.
The investments incentivised by this type of mechanism can help to reduce the damage that disasters
pose to physical infrastructure and to lives and livelihoods. The time taken to set up a crediting
mechanism means that they are most well-suited for preparatory activities while the relative
sophistication of the instrument means that they are most likely to be effective at encouraging
investment by companies in a way that supports the local community, but this can have spillover
benefits at the personal level. As with all risk-reduction activities, they are most likely to be cost-
effective in reducing the risks associated with relatively high-frequency events, though crediting
mechanisms can be incorporated in more significant risk reduction projects.
DESIGN
OPTIONS
Some of the key issues to determine in this mechanism are whether credit purchases will be
voluntary or mandated by regulation, the extent to which credits are just bilaterally exchanged or
whether they can be traded between third parties (the latter potentially allowing for the formation of
a more liquid commodity market but also being likely to introduce additional price volatility) and the
type of investments that are allowed to generate credits.
CHALLENGES
The attraction of crediting mechanisms is that they can create an additional economic incentive for
risk reduction investments without the use of (scarce) public resources. However, to be effective,
there needs to be a sustainable source of demand for the credits. In the case of mitigation banking
(see below), this is achieved through regulatory requirements on developers to make good the
negative biodiversity impact of their developments.
It may be difficult to generate a parallel source of regulatory demand for risk reduction investments,
while CSR demand may not be consistently high.
A further, critical challenge is in quantifying, on a comparable basis, the risk reduction benefits that
a wide range of varying investments deliver.
REQUIRE-
MENTS
The regulatory requirements for this approach are relatively light in cases where any credits are
purchased on a voluntary basis i.e. for CSR purposes. However, if demand for credits stems from a
compliance obligation placed on purchasers by regulation then an associated regulatory architecture
will be needed to ensure that the risk reduction investments, and the associate credits they generate,
are consistent with the objectives of the regulation.
MITIGATION BANKING
One of the most mature examples of this approach is known as ‘mitigation banking’
26
. Developed in the US,
with a focus on the restoration or enhancement of wetland or other aquatic resource areas, purchasing credits
from such projects provides a flexible way for developers to fulfil mandates to compensate for the impact of
other developments. While this mechanism is primarily intended as a mechanism for preventing biodiversity
loss/ achieving net gain, there are many cases where ecosystem restoration can also reduce the damages from
disasters
There are also similar examples in developing country contexts. For example, the African Development Bank
is piloting the concept of an Adaptation Benefits Mechanism
27
. This will create credits (or Adaptation Benefit
Units (ABUs)) that reflect the value of the social, economic and environmental benefits of adaptation activities.
ABUs could then be sold to interested parties who want to demonstrate their commitment to support adaptation
activities in Africa. The pilot, to run between 2019 and 2023, is set to include projects that enhance coastal
protection through afforestation with mangrove trees.
DISASTER RISK FINANCE – A TOOLKIT36
IMPACT BONDS
Can be used to fund
more significant projects
(all risk levels)
Lives and livelihoods,
operations, and
physical assets
Preparedness
activities
plus recovery
Can be used to fund more
significant projects
(all risk levels)
OVERVIEW
Impact bonds encourage risk reduction investment by offering a pay for performance contract
between an ‘outcome based funder’ - typically a government, donor agency or philanthropy - and
private sector investors in relation to a project that has social or development objectives. Under an
impact bond structure, investors will provide capital (either/both debt and equity) to a project with
the outcomes-based funder committing to make repayments to investors depending on the extent to
which independently verified performance targets are met. These targets place a strong incentive
on the overall outcomes expected from the project, rather than just immediate project outputs.
Investors will normally appoint a ‘managing agent’ to implement the project.
The structure could be used to incentivise investments that reduce the risk that disasters pose
to infrastructure, although by boosting the adaptive capacity of individuals and communities e.g.
improving health or education outcomes, the mechanism could also reduce the risks to lives and
livelihoods that disasters cause. The long timescales and substantial transaction costs involved in
structuring impact bonds (see below) mean that they are most appropriate for preparedness activities
and typically at the community, municipal and/or sovereign level. As with all risk-reduction activities,
they are most likely to be effective in reducing the risks associated with relatively high-frequency,
low-impact events.
The structure can be attractive to outcome based funders as they allow the risk of successful
delivery of outcomes to be transferred to investors (if no outcomes are met, less or no money
is paid to investors). They also require that the capital for a project comes from private sources.
At the same time, they are also attractive to private investors as a way of marrying financial
return
iii
while delivering social impact.
DESIGN
OPTIONS
Key design questions include which outcomes to target, how much return investors should earn
if outcomes are delivered and how much they should lose if the outcomes are not delivered.
CHALLENGES
Impact bonds can be complicated to design, often taking 6 months to 3 years to structure
28
.
Moreover a recent report by Lloyds and DFID explores how impact bond could be used to
incentivise risk reduction/resilience investments. The report notes that the structure may be
challenging to adopt for resilience/risk reduction due to difficulties in quantifying outcomes related
to risk reduction and because of questions over who should bear the risk of a disaster striking
during the lifetime of the bond.
REQUIRE-
MENTS
Specific regulation for impact bonds is unlikely to be needed but the ability to structure deals
in the context of existing procurement regulations can sometimes be complicated
29
.
DEVELOPMENT IMPACT BONDS
There are no examples of development impact bonds explicitly targeting risk-reduction investments. However,
humanitarian actors have developed this model in other contexts. For example, so-called Humanitarian Impact
Bond, designed by the International Committee for the Red Cross, involves a selection of governments have
committed to make payments to consortium of investors depending on whether, after 5 years, new physical
rehabilitation centres financed by the investors deliver a level of outcome – in terms of the number of people
receiving mobility devices per physical rehabilitation professional – that is higher than the average in Africa.
If the benchmark is exceeded the investors will receive a return on their investment; if it is below benchmark,
then the investors will lose a certain amount of their initial investment30.
iii A Blavatnik School of Government briefing reports 2 case studies suggesting investor returns of between
15% and 70% for two impact bonds targeting development outcomes
57
.
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 37
4.2. Risk Retention
Risk retention instruments are pre-arranged mechanisms
that provide risk holders access to capital, where funds are
sourced either from their own reserves or external capital
that they are responsible for repaying. e resources
provided through these instruments come from those
aected by the disaster. In other words, those aected
by the disaster are those who retain the responsibility
for covering the costs that arise following the event. e
section explores three main risk retention instruments:
budget contingencies, reserve funds and contingent loans.
In relation to all risk reduction mechanisms, and the
risk transfer instruments discussed in section 4.3, there
are important considerations relating to how resources
are released from the instrument – this consideration is
described as the ‘trigger mechanism’.
Box D. Trigger Mechanisms
For risk retention and transfer – a key design option is the mechanism by which the funds are accessed
and distributed. The ‘trigger mechanism’ determines whether, and the volume of funds that are released from
a DRF instrument for a given event.
Trigger options range in complexity from subjective processes, to pre-defined objective processes that are
based on the measured parameters of an event (parametric triggers).
Parametric-based triggers use observed event parameters as a basis for estimating total disaster impact. In
order to design parametric triggers, catastrophe risk models can be used to quantify the relationship between
event parameters and the associated disaster impact. This understanding is then used to define parametric trigger
thresholds. With careful design, parametric triggers offer a rapid and transparent alternative to subjective or
indemnity-based triggers. Parametric triggers create derivative products, which can cause
payouts that over- or under- estimate the actual need, this challenge is discussed later in
→ Box E.
The main categories of trigger are summarized below:
Trigger
Mechanism
Measurement Description
Subjective Informed
Judgement
Informed judgement is often sufficient to access risk retention mechanisms,
where the capital ‘belongs’ to the decision maker. Totally or partially subjective
triggers are useful for accessing time-critical funds, as it implies no need for
(independent) assessment. Subjectivity can raise issues of transparency. To
deal with this, these triggers should be associated with clear decision-making
processes and a requirement that funds are distributed according to pre-
arranged disaster plans.
Indemnity Reported
Claims
Traditional risk transfer instruments are often triggered based on the reported
level of loss following an event. The majority of insurance and reinsurance
policies, and insurance-lined securities (ILS) trigger on an indemnity basis.
The advantage of indemnity-triggers is that the payout closely matches the
underlying need. A challenge with indemnity-triggers is that they require
regulated claims handling processes, and claims can take a long time to
settle as they are reviewed.
Simple
Parametric
Macro-event
parameters
Where local observation data is limited, remote observations of an event’s
main characteristics can be used to trigger funds. For example, a simple
‘cat-in-a-box’ structure uses hurricane category and track location with
respect to a pre-defined area (‘box’) as the basis for a trigger mechanism.
This simple structure is a useful first step towards developing more
sophisticated trigger mechanisms. And may be the most fit-for-purpose
solution in some contexts.
DISASTER RISK FINANCE – A TOOLKIT38
Pure
Parametric
Local Hazard
Measurement
Where there are robust local observation networks, location-based hazard
measurements (wind speed, flood depth, temperature) can be used in parametric
trigger mechanisms. Local hazard measurements are more highly correlated to
local damage than macro-event parameters, and so can give a more accurate
estimate of total event impact.
Modeled
Loss
Modeled
Footprint
For large spatially distributed sets of exposure, local observation networks
may not provide enough coverage to create an accurate estimate of total
disaster impact. Available observation data (from local observations and
remote sensing) can be used to create a modeled event hazard footprint.
This can then be used in catastrophe models with exposure and vulnerability
datasets to create a ‘modeled loss estimate’ which is used to trigger the
funds.
Innovations in Trigger Design: Forecast-based Finance
Conventionally, trigger mechanisms developed in the private markets have been responsive – they measure
what has happened and make an appropriate payout. However, in a humanitarian context, the potential value
of delivering funds prior to impact has spurred innovation in trigger design.
The need for rapid funding has led rise to an innovative form of DRF called Forecast-based Finance (FbF),
which broadly describes financing instruments which are triggered by forecast data. There are a range of
current initiatives to develop anticipatory trigger mechanisms, which use forecast data and other available
information to anticipate the potential severity of disaster impact – this information is then combined with
pre-arranged response plans, to deliver funds for particular activities prior to the main impact. The attraction
of these mechanisms is that relatively small injections of capital received before an event, if placed in the
hands of local responders, can support preparation and response activities that significantly reduce the
eventual severity of impact.
Forecast-based trigger mechanisms could be incorporated in risk retention or risk transfer instruments,
e.g. a forecast-based trigger could be used to access a reserve fund, or to trigger a payout from a
catastrophe bond. Their potential to reduce the severity of impact also means that they act to reduce risk.
Forecast-based finance is an evolving topic – and early lessons learned will continue to refine approaches.
As described in Box C, the Red Cross Red Crescent Societies have been key proponents of FbF and have
implemented a range of FbF projects
31
. The Start Network Crisis Anticipation Window
32
has similarly used
forecasting to inform disbursement of funds.
Conceptually FbF is a very powerful tool – however, it will not be appropriate in all situations, some
considerations are explored below. Note that some of these considerations also apply to parametric
triggers more broadly.
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 39
Considerations for FbF
Given the innovative nature of forecast-based triggers, care should be taken when applying this trigger mechanism
to DRF instruments. The following table provides guidance on some considerations for FbF.
Uncertainty & Flexibility
There are inherent uncertainties in the approach of using forecasts to trigger funding. Disaster
events are a function of many complex inter-related factors, any one of which can make the
difference between the event unfolding into a minor or major catastrophe. However, given that
relatively small capital injections can have a profound impact on mitigating the overall severity
of disaster, the risks of this approach can be measured against the potential benefit.
Advances in forecasting and risk modelling techniques will continue to reduce the uncertainty
associated with anticipatory triggers. In addition, uncertainty in forecasts can be accommodated
with flexibility in the trigger design and instrument mechanics. For example, it is possible to
design ‘soft’ parametric triggers which make a small initial payout based on early information,
with the option to make a subsequent larger payout when there is more certainty in the event
outcome. Softness can also be designed into the trigger mechanism by allowing for a combination
of both objective and subjective elements. For example, an objective parametric measurement
may ‘flag’ an event, which can then be referred to an expert panel to assess if a payout should
be made.
Delivery
Forecast-based finance is most impactful when it can be spent effectively on the ground, by
local actors who can use the funds to implement pre-arranged response plans. The speed of
forecast-based payouts should therefore be matched by the capacity of the recipient to use the
funding efficiently and effectively. Clear disbarment mechanisms and spending plans are therefore
fundamental to supporting forecast-based payouts in particular. The humanitarian system is well
placed to support forecast-based initiatives. Local, regional, and international networks can act as
the distribution mechanism for funds. Given the uncertainty in the actual costs required to support
action, the combined experience can also guide decisions about how much funding is required,
and how best to allocate and distribute funds to local responders. It should also be noted that
the speed of payout may also be constrained by the DRF instrument itself – for example special
purpose vehicles (SPVs) which hold collateral funds for catastrophe bonds, may only be able to
provide cash payouts days after they are triggered. This is due to the constraints on liquidity of
the underlying funds. This delay can be accommodated provided that the recipient is able to
cover costs based on the promise of repayment.
Shock, Strain, & Systemic Events
The speed of onset and complexity of the peril type are important factors when considering
the applicability of forecast-based trigger mechanisms. For rapid-onset weather events including
typhoons, floods, and convective storms, climate variability means that forecasts may only
provide actionable guidance shortly before impact (hours-days). In addition, forecasts for very
local hazards (e.g. hail, lightning, tornado) can be highly uncertain. For slower onset events
including drought or El Niño events, forecast-based payouts can be made as the event is unfolding
(similarly to the World Bank Pandemic Bond issued in 2017)
33
. An important design consideration
for slow-onset events is to carefully identify payout thresholds. This decision process is strongly
informed by risk modeling – care should be taken to consider model and measurement uncertainty,
as well as forecast skill. To some extent, all disasters are systemic in nature. However, for
complex disasters that result from multiple upstream causes (e.g. mass migration resulting from
climatic and geopolitical factors), it can be challenging to accurately model the complexity in
the system to the extent that it is necessary to develop forecast-based parametric triggers.
For complex systemic disasters, where possible triggers should be deigned to be flexible
(e.g. both objective and subjective) in order to avoid issues when events do occur.
DISASTER RISK FINANCE – A TOOLKIT40
BUDGET CONTINGENCY
Municipal and
sovereign (plus
smaller-scale bodies
if they set their
own budgets)
Operational Response
Most cost effective when
used to respond to high
frequency low-intensity
events e.g. up to
1 in 10-year events
OVERVIEW
A budget contingency is a risk retention mechanism whereby a certain proportion of revenues within
a budget are set aside for dealing with contingencies. These contingencies may be either explicitly
defined but, more commonly, are simply left available to be used for undefined ‘exceptional events’.
The instrument is most typically used by, national or municipal governments but, in principle, could
be used by any organisation or household that face significant risks. In the case of governments,
budget contingencies typically amount to 2-5% of the annual government budget34.
The attraction of budget contingencies is that, compared to other risk retention or risk transfer
mechanism, they are a relatively low cost, flexible instrument for risk holders to manage their risks.
Funds can be accessed almost as soon as they are needed
iv
, and the main cost is the opportunity
cost of the activities that are not supported because the money is being held as a contingency. A
previous analysis estimates the cost of the instrument as just 1–2 times the expected pay-out of the
instrument, making it among the lowest cost instruments explored in that study
34
.
This flexibility, combined with the fact that they are unlikely to be able to provide large sums of
capital (see below), means that they are best placed for dealing with the immediate, response costs
during and following a disaster event, and for high-frequency, low damage events (i.e. events that,
on average, happen every 2 or 3 years).
DESIGN
OPTIONS
The key design options relate to how much funding is placed in the contingency and whether there
are any formal rules determining whether the funding can be accessed or how it can be spent.
CHALLENGES
The flexibility of the instrument is both its biggest advantage, but also its biggest disadvantage.
As the arrangement is voluntary, it can be politically difficult for large sums of money to be
placed in a contingency budget, and it can be politically tempting for governments to use
whatever funding is placed in a contingency for other, non-disaster related reasons.
REQUIRE-
MENTS
There are typically no substantial regulatory barriers to using this instrument among any
organisation that has budget setting powers.
NATIONAL BUDGETS
A number of governments have budget contingencies in place including Japan, Vietnam, Indonesia and Colombia.
For example, in Vietnam, under the State Budget Law of 2002, Central and Local governments are required to
allocate between 2 percent and 5 percent from their total planned budget for capital and recurrent expenditures
to contingency budgets
35
. However, these contingencies are not explicitly linked to disasters. This has led
to situations where the country has experienced a major cyclone hit the country in November, but when the
contingency budget had already been fully exhausted
34
.
iv Although this depends somewhat on the budget and spending rules of a jurisdiction; if these are cumbersome it may take
longer before budget contingency can be released. Ghesquierre and Mahul (2012) suggest that it can take between 0 and
9 months after an event to access resources from a budget contingency.
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 41
RESERVE FUNDS
Municipal and
sovereign (plus
smaller scale
risk holders using
informal reserving)
Life and livelihood,
operational,
and physical
asset costs
Response
and recovery
Most cost effective when
used to respond to high
frequency, low-intensity
events e.g. up to
1 in 10-year events
OVERVIEW
Reserve funds involve the same basic idea as budget contingencies – money that could be spent
now is instead saved to cover future costs after a disaster – but aim to provide more formality
concerning how the money can be accessed and what it can be spent on. Specifically, money is
transferred into a reserve account that sits outside the budget and the transfer of resources to
the fund recognised as a spending line in the budget. In addition, funds are typically not then
transferred back to the budget if unspent in that year. Reserve funds can be set up by national, city
or local governments. Informal reserving mechanisms can also support smaller scale risk holders,
at the individual and community levels.
The attraction of reserve funds is similar to that for contingency funds: they offer a means of quickly
accessing funding for the immediate response costs associated with a disaster event, at relatively
low cost. Previous studies suggest resources can be made available 0-1 month after a disaster
event and that costs are only 1-2 times the expected pay-out
34
. They are therefore typically used for
covering immediate response costs, although the Philippines example below shows how the basic
model can be adapted to cover other costs as well. Moreover, the rules-based nature of the transfer
of resources into the reserve fund and out of the reserve fund to cover costs means that they are a
more predictable source of post disaster financing than, for example, budget contingencies.
DESIGN
OPTIONS
The examples in the box below illustrate some of the most important design issues to consider in
relation to a reserve fund: who should place how much money in the fund (and with what level of
discretion); what determines whether the money can be accessed; and the rules governing what the
money in the reserve fund can be spent on.
CHALLENGES
While reserve funds may provide more predictable funding than budget contingencies, it still remains
politically challenging to allocate substantial funds to a reserve fund. They therefore remain better
suited for providing capital to deal with relatively frequent, low-intensity events.
REQUIRE-
MENTS
There are not normally substantial regulatory challenges associated with setting up a reserve fund,
although in some countries there may be reluctance from the Finance Ministry or Treasury regarding
the extent to which funds can be established that operate beyond its immediate purview.
DISASTER RESERVES
In the Philippines, cities are required under legislation to set up Local Disaster Risk Reduction and Management
Funds (LDRRMF), which are the principal source of funding for all disaster risk events
36
. Cities are required to
allocate at least 5% of their budget to these funds. 30% of the collected resources are allocated to a Quick
Response Fund for past disaster financial liquidity, which is made available upon the declaration of a state of
calamity at a local (city or higher) or national level by the relevant body. The remaining 70% is placed into a
Mitigation Fund for prevention, response and recovery activities. Any unspent balances at the end of the year
transfer to a Special Trust Fund for the sole purpose of funding disaster risk reduction.
Mexico’s disaster fund, FONDEN, provides a further example of a reserve fund operating at a national level: this
receives annual budget appropriations of around $800m per annum and covers 50% of the costs of reconstruction
after disaster events
8
.
While reserve funds are most commonly set up by local or national governments, they can also be set up by
communities.
For example, the FAO has supported the establishment of Community Contingency Funds whereby communities pay
into a fund which then help vulnerable households following an unexpected event such as drought, hurricanes,
floods, earthquakes or other extreme events
37
. Funds can be accessed, typically in the form of low-interest loans,
for households to, for example, purchase supplies for the new agricultural season in the event of crop losses. In
these cases, donors and international organisations might also support the set up or capitalisation of such funds.
DISASTER RISK FINANCE – A TOOLKIT42
CONTINGENT LOANS
Sovereigns Operational costs Response
Most cost effective when
used to respond to high
frequency, low-intensity
events e.g. up to
1 in 10-year events
OVERVIEW
Contingent loans are loans that, in advance of a disaster, it is agreed will be made available on
specified terms following a disaster, if the disaster’s severity meets or exceeds a certain threshold
(trigger). In other words, they are made available contingent on a particular event or level of damage
being incurred. They are typically provided by International Finance Institutions (IFIs) to sovereign
governments. IFIs often only allowing sovereigns to sign up for a contingent loan if they have a
disaster risk management plan.
The main attraction of a contingent loan is that the resources can be accessed quickly following a
disaster. This makes the instrument well suited to dealing with the immediate increases in costs,
and liquidity challenges this can pose, during the response phase of a disaster. A further attraction
is that, especially at rates offered by IFIs, they are a relatively cheap way of accessing capital to
deal with the impact of disaster. Reflecting this, analytical work suggests that they are typically well
suited for ‘medium’ risks, in other words, risks with a relatively low impact but which happen quite
frequently, perhaps once every five years or so.
DESIGN OPTIONS
There are two main variants of contingent loans: soft trigger and hard trigger loans. A soft trigger is
a subjective trigger mechanism whereby the sovereign government can determine whether or not an
event in sufficiently severe to justify the loan being accessed – practically this is achieved by making
the trigger the declaration of a state of emergency. A hard trigger is a parametric trigger, for example,
relating to wind speed of a tropical cyclone. Other design features include how much capital should
be available to each country; the interest rate and other pricing conditions at which the loan will be
made available; and the drawdown period (the period of time over which the loan can be drawn down).
CHALLENGES
There can also be challenges in using contingent loans. An evaluation of the contingent loan offering
by the IDB found that they were not always supported within the organisation as they used up scarce
lending capacity that might never be used. Similarly, potential borrowers were sometime reluctant to
take out contingent loan products because of a fear that this would indicate they were vulnerable to
the impacts of a disaster (especially compared to peer countries). The evaluation found this problem
was exacerbated for products that had standby fees included and in cases where there was some
uncertainty about whether the loan will actually be made available
38
.
REQUIRE-
MENTS
The product is typically provided through a contract between an IFI and a sovereign government.
As such, the regulation that needs to be in place for the product is relatively light. However, the
IFI will typically require that the sovereign has both an adequate macroeconomic policy framework;
and be preparing, or already have, a satisfactory disaster risk management program,
CAT DDO
An established example of a contingent loan product is the World Bank’s Development Policy Loan with a
Catastrophe Deferred Drawdown Option (CAT DDO) product. This product allows countries to borrow up to the
lower of US$250 million or 0.5 percent of GDP (IDA countries39) or US$500 million or 0.25 percent of GDP
(IBRD countries
40
) in the event of a state of emergency being declared by the country. The drawdown period
for the loan is 3 years, renewable up to 4 times.
The interest rate on the loan is the same as for regular IDA/IBRD loans, with no front end fees or renewal
fees (IDA countries)/0.5% front end fee and no renewal fees (IBRD countries). The product is only available
to countries that have, or are preparing, a satisfactory disaster risk management plan, which the World Bank
monitors on a periodic basis.
Between 2008 and 2017, 15 such loans were approved worth US$2.345 billion across countries
41
.
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 43
4.3. Risk Transfer Instruments
In contrast to risk retention instruments, risk transfer
instruments place the obligation for providing (a certain
amount of) capital in the event of a disaster onto third
parties. e capital provider will receive a payment in
exchange for accepting this risk. is section includes an
overview of insurance as the key risk transfer tool, as well
as exploring a number of dierent forms on insurance
examples – focusing on those of greatest relevance to
developing countries – before considering catastrophe
bonds.
A key issue associated with all of these instruments is
that of basis risk, this is discussed in Box E.
Box E. Quantifying Basis Risk
When disaster strikes, it is not unusual for an insurance payout to differ from the policyholder’s expectation. The
possibility of such a discrepancy is referred to as basis risk. Basis risk can be defined simply as the ‘difference
between expectation and outcome’.
Parametric insurance is most commonly associated with basis risk. For example, in the case of a modeled loss
trigger, basis risk will emerge when there is a difference between modeled loss and measured loss after an
event; while for a pure parametric trigger, basis risk refers to the difference between the index loss calculated
from a wind speed measurement and the total actual loss. However, when defined as above, it becomes clear that
basis risk exists within all DRF instruments which contain a trigger mechanism. For example, in indemnity-based
insurance, basis risk could stem from the possibility that a contract fails to pay because of a legal miswording.
The primary drivers of basis risk vary between structures. To quantify basis risk, it is first necessary to identify
the primary sources of uncertainty with respect to each structure. Once identified, basis risk can often be
quantified, and communicated to the purchaser. With the basis risk appropriately understood, the structure can
then be tailored to modify the expectation as appropriate.
A range of methods have been developed to assess basis risk in parametric structures, these can also be applied
in modelled loss and indemnity cover. Catastrophe models provide a useful tool for the assessment of basis risk.
A simple assessment of the correlation between the modeled parametric index and indemnity loss can uncover
if a trigger mechanism tends towards shortfall (no payout when expected) or overpayment (payout when not
expected). Calculation of shortfall and overpayment with respect to a target covered layer can be done using the
following equations – the process of basis risk calculation and trigger refinement is fundamental to the design
of appropriate parametric instruments.
Figure 14. Basis risk plots. left: parametric index against modeled loss. Right: shortfall and overpayment for an
illustrative risk layer (Source: RMS).
DISASTER RISK FINANCE – A TOOLKIT44
v Some definitions of microinsurance also include agriculture insurance for smallholders. However, this is treated separately
in this taxonomy.
MICRO-INSURANCE
Individual
and community
Lives and livelihoods
and physical assets
Response
(especially parametric) and
recovery
(especially indemnity)
Most cost effective when
used to respond to low
frequency, high-intensity
events e.g. beyond
1 in 10 year events
but sometimes used for
more frequent events.
OVERVIEW
Microinsurance is the provision of insurance to transfer risks associated with disasters from poor
and vulnerable households who would otherwise not have access to insurance. Coverage and
premium payments are, by design, low, with insurance payments intended to pay out for losses
of life and property.
v
A review by Climatewise in 2011 identifies 14 Disaster Micro-insurance
schemes in developing countries, one of which was at a proposed stage, and another had been
discontinued.
The speed of pay out and appropriateness for different forms of risk are similar to those for
agricultural insurance as discussed above.
DESIGN
OPTIONS
As the examples below demonstrate, the schemes can be designed with either parametric or
indemnity triggers. There are also choices over the channel to market; as the Swayamkrushi
example below suggests, it is increasingly common to bundle the provision of microinsurance
with microcredit. Other design features across which schemes may vary include which perils
to cover, trigger design (especially for parametric), premia amount and options for payment.
CHALLENGES
The same challenges regarding affordability as discussed for agricultural insurance also
apply to disaster microinsurance.
REQUIRE-
MENTS
The same regulatory issues as discussed for agricultural insurance also apply to disaster
microinsurance.
MICROINSURANCE
‘Mithapukur Sonirvor Mohila Somobay Somity’ has developed a microinsurance product for residents in
the Mithapukur Upazilla District of Bangladesh that also complies with the principles of Takaful insurance
(as discussed above).
Self-help groups make a contribution of 100 taka per year (approximately US$1.15) to manage the scheme.
In addition, individual members each pay 100 taka annually. This entitles them to access pre-defined benefits
in the event of hazards such as death, disability, hospitalisation and business loss, including those caused by
weather-related events. As of 2016, 90% of SHG members (more than 3,300 people) had taken up the scheme,
with the 50 payouts made in that year, and surplus income of 180,000 taka.
The pilot also identified some of the challenges associated with microinsurance, including a lack of
understanding of the product leading to scepticism among potential beneficiaries as to the benefits they
would receive, difficulties associated with pricing due to the lack of weather data, affordability constraints,
and the potential fragility of the scheme to large events that might wipe-out any reserves.
42
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 45
AGRICULTURE INSURANCE
Individuals
Lives
and livelihoods
Response and
recovery e.g.
new seeds
Most cost effective when
used to respond to low
frequency, high-intensity
events e.g. beyond
1 in 10 year events but
sometimes used for more
frequent events.
OVERVIEW
Agricultural insurance is an insurance product specifically designed to transfer risks associated with
agricultural losses caused by weather related hazards. A 2011 Climatewise report identifies
84 agricultural insurance schemes in developing countries
43
.
Agricultural insurance can be an effective tool. For example, one study found that following a drought
in the Horn of Africa, households benefiting from an index-based livestock micro-insurance scheme
were 25 per cent less likely to reduce meals than their uninsured counterparts and 36 per cent less
likely to engage in distress sales of livestock
44
.
Ideally, agriculture insurance would be best placed to transfer risks associated with infrequent, large
events. However, the individuals benefiting from the schemes may not be in a position to substantially
retain risks without resorting to negative coping strategies, implying that insurance may also be used
to transfer the risk of more frequent events.
DESIGN OPTIONS
A key distinction is between products that have an indemnity trigger and those with a parametric
trigger. Parametric triggers have become popular in many developing world contexts as they avoid
costly assessments of losses. Indemnity triggers may be based on either yield or revenue losses.
As with other forms of insurance, the type of trigger determines the speed of payout and hence the
disaster risk financing phase to which they are best suited. Parametric triggers can pay out in less
than 2 months making them suitable to covering the response phase of a disaster; whereas indemnity
schemes may take around 6 months to pay out but may be better for longer term asset acquisition.
Other design features include which hazards are covered and the level of cover provided.
CHALLENGES
A key challenge is whether premia are affordable for those targeted by the scheme as, relative
to the ability to pay of typical customers, scheme set up and operation costs can be high. To address
this challenge, donor and/or public funding may be made available. In addition, or alternatively,
schemes may use innovative approaches for premia payment as explored in box below.
While parametric triggers are better suited to many developing world contexts, trigger design can
be complicated, and basis risk substantial, if granular meteorological information is not available.
REQUIRE-
MENTS
The issues and challenges surrounding regulation for agricultural insurance are broadly the same as
those described more generally for insurance above. Given the prevalence of index based insurance
in developing countries, there can be a particular challenge when regulatory frameworks do not
recognise index based insurance, as has been the case in West Africa. This has held back the
development of the market in this region compared to East or Southern Africa
45
.
‘R4 RURAL RESILIENCE INITIATIVE
The R4 Rural Resilience Initiative
46
, supported by Oxfam and the World Food Programme, as of early 2018, has
reached around 57,000 farms (300,000 people) across Ethiopia, Senegal, Malawi, Zambia and Kenya. It offers
microcredit to support risk reduction, promotes savings so as to allow more efficient risk retention, microcredit
to support prudent risk taking and offers insurance (risk transfer). An innovative aspect of the scheme is that
it allows some premia payments to be made in kind through undertaking risk reduction investments. In 2018,
around US$ 1.5 million of insurance payouts were distributed through the initiative in Ethiopia, Kenya, Malawi,
Senegal and Zambia.
DISASTER RISK FINANCE – A TOOLKIT46
TAKAFUL & MUTUAL INSURANCE
All scales
Lives and livelihoods,
operational
and physical assets.
Response
and recovery
Most cost effective when
used to respond to low
frequency, high-intensity
events e.g. beyond
1 in 10-year events
OVERVIEW
Mutual insurance offers products that are very similar to standard insurance, but rather than the
insurance company being owned by shareholders, it is owned by its policyholders. This means that
any surplus income generated by the insurance company is returned to customers or used to reduce
future premia. In this sense, strictly speaking, risk is shared among policyholders, rather than
transferred to third parties.
Takaful insurance is closely linked to the concept of mutual. It responds to an ethical concern
with Islamic jurisprudence that conventional insurance provides benefits that are too uncertain
or no benefit at all if there is not a risk event) and that insurance company often invest premia in
interest-bearing instruments. As such Takaful insurance involves members who, rather than pay
premia, make regular ‘donations’ and receive a pre-defined pay-out in the event of loss, plus a return
on the investments made in the insurance fund (which is invested in Sharia compliant instruments).
47
DESIGN OPTIONS
Mutual insurance companies can offer a wide range of insurance products, with different trigger
mechanisms, covering different perils, and with varying designs as to whether insurance is mandated
to be compulsory and the premium charged/subsidy offered. As mutual insurance companies do not
access external capital, they are sometimes not able to offer as much cover against high impact, low
probability events as conventional insurers.
Many of these variants are also applicable to Takaful insurance, although takaful principles mean
that pay-outs are typically fixed, rather than being based on a detailed assessment of the loss or
damage incurred.
CHALLENGES
Mutual and takaful insurance face broadly the same challenges as for insurance products
offered by shareholder-owned companies
REQUIRE-
MENTS
The regulatory structures for mutual insurance may sometimes be explicitly articulated and different
from those for insurance companies owned by shareholders. Countries as diverse as China, Chile, Iran
and Indonesia have separate laws for mutual insurance. However, many other countries do not have
dedicated laws for mutual insurers, which can impede their market development. As of 2016, it was
estimated that 45% of countries, and 63% of low-income countries, did not have a mutual insurance
law
48
.
Takaful insurance requires monitoring by a Sharia advisory board in order to ensure that Sharia
principles are being respected both in relation to operational practices and in how donations are
then invested. Further statutory and regulatory provisions may be needed to allow Takaful insurance
companies to access external finance in a way that is Sharia compliant.
42
TAKAFUL INSURANCE EXAMPLE
Takaful Insurance of Africa offers an Index-Based Livestock Insurance (IBLI) product, branded as Index-Based
Livestock Takaful (IBLT). This offers protection against prolonged lack of pasture as a result of severe drought
and offers protection in the event of limited vegetation for cattle, camel, sheep and goats
49
.
In 2014, the company made the first Takaful insurance livestock payment for livestock insurance to 30 women
and 71 men in Wajir County in Kenya
50
.
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 47
INSURANCE & REINSURANCE
All scales
Lives and livelihoods,
operational and physical
assets.
Response
and recovery
Most cost effective when
used to respond to lower
frequency, high-intensity
events e.g. beyond
1 in 10-year events
OVERVIEW
Insurance cedes the risks associated with a disaster to a third party (insurance company),
in exchange for premium payments. If a qualifying event takes place, the insurance company
is contractually obliged to make payments to cover some or all of the losses. Governments,
infrastructure and property owners, to farmers, firms and households can all take out insurance
vi
.
Insurance contracts typically last one year, but, on occasion, longer term contracts are available.
Insurance companies, in turn, will purchase re-insurance to cover part of their loss exposure.
Both parametric and indemnity products are relatively expensive: previous estimates suggest that
they may cost more than 2 times the expected pay out. They are this better suited for less frequent
but more damaging events, where the long-term impacts from not otherwise having reliable access
to capital would be most damaging.
DESIGN OPTIONS
A critical design feature of any insurance contract is the trigger mechanism, as discussed in the
Box D above. Other key factors include the cover provided, which perils are covered, whether
payment is compulsory, and the premia amount (including whether the premia is subsidised).
Different types of insurance cover different phases of disaster financing needs. The ability of
parametric cover to pay out quickly – in just less than 2 months – makes it a useful instrument for
the response phase, supporting debris removal, funding temporary living solutions etc. However, the
higher basis risk it less suited for funding longer term reconstruction. In these cases the fact that
indemnity products are more likely to pay out what is needed to cover reconstruction costs makes
them more desirable, even if the speed of pay-out is much slower (6 months or more)
34
.
CHALLENGES
One of the most significant is the cost associated with the products which may make them
prohibitive, especially for individuals and households where distribution channels and intense
marketing efforts may be required. These costs may be exacerbated by a lack of data, which
makes it harder to price risks, and a lack of understanding/mistrust among potential consumers
on the role and efficacy of the product. A further challenge, especially for indemnity products, is
a concern that their reduce the incentive to undertake risk reduction investments (moral hazard).
Parametric products offer opportunities to both reduce the cost and moral hazard problems, but,
in many cases, increase the basis risk embedded in the product.
REQUIRE-
MENTS
Insurance regulation varies across countries. In some countries, regulations may restrict the
type of cover that can be provided, the extent to which premia can vary between customers
or the nationality of the firms that can offer insurance. Greater regulatory harmonisation offers
opportunities for growth in insurance markets51.
INSURANCE PENETRATION
Insurance of public assets for disaster losses is compulsory in countries such as Colombia, Peru, the Philippines
and for some assets in Vietnam. Indemnity insurance is much more common than parametric insurance. Reserve
funds may purchase reinsurance to ensure that they can remain solvent if they face large payouts: for example,
since 2011 Mexico’s disaster fund, FONDEN, has purchased reinsurance cover on international markets. However,
generally insurance levels, especially in developing countries are low. Previous analysis suggests that just 3%
of the annual losses of around $30bn from natural catastrophes in low and lower-middle income countries are
covered by insurance.
52,53
v The specific features of insurance for individuals – microinsurance – is explored further below.
DISASTER RISK FINANCE – A TOOLKIT48
CATASTROPHE BONDS
Sovereign
(and large corporates
and municipalities)
Life and livelihood,
operational costs
and physical assets
Response
and recovery
Most cost effective when
used to respond to low
frequency, high-intensity
events e.g. beyond
1 in 10 year events
OVERVIEW
Catastrophe (‘cat’) bonds are short term bonds (see → section 3.1 above) (3–5 years) issued by a
sponsor to investors in the capital markets. However, in contrast to normal bonds, they are ‘triggered’
by a catastrophe. Once triggered, the bond sponsor maintains a portion of the principal and
consequently investors lose a portion of principal and interest payments. In this way, they transfer
natural catastrophe risk to investors. The bond issuer will typically be a state or large infrastructure
owner. Insurers, reserve funds or risk pools might also issue catastrophe bonds, as an alternative
to purchasing reinsurance, to lessen their risk exposures. They can be attractive instruments to
investors as cat bond risks are uncorrelated with other risks investors face.
The cost of catastrophe bonds (see below), plus the fact that bonds can be issued for large amounts,
means that they are best suited for low frequency, high impact events; this is also the perspective of
investors who would not be interested in bonds that were triggered on a frequent basis.
DESIGN OPTIONS
As with many of the other instruments discussed above, the most important design considerations
is whether the trigger mechanism is indemnity-based or parametric. This has the same trade-offs
between speed and basis risk identified for other instruments, and accordingly means that cat bonds
can either be used to cover near term response costs or longer term reconstruction efforts. Other
design questions include the size of the issuance and the coupon on the bond.
One specific- at present, hypothetical - version of a cat bond is a resilience bond. This would work
in the same way as a cat bond except that, if risk reduction investments are undertaken, interest
rates on the bond would fall to reflect the risk reduction
vii
. 2 This anticipated interest rate reduction
could help the financing of the risk-reduction investment. However, this is a complicated instrument –
for instance, requiring reliable estimates from modellers of how much the investments reduce risks.
As yet, there are no examples of this instrument reaching the market
54
.
CHALLENGES
Catastrophe bonds are relatively expensive risk transfer instruments: previous work suggest they
may cost more than two times their expected pay out
34
. This reflects both that structuring and
issuance cost of these bonds are more expensive than for other bonds and because they tend to
require relatively high interest rates to generate investor interest
viii
. This means that they tend to
be more appropriate for larger organisations or more developed governments.
REQUIRE-
MENTS
As sophisticated instruments, cat bonds are subject to a range of regulatory requirements. Cat bonds
are usually set up by Special Purpose Vehicles (SPVs) that will require licensing and having various
capital, reserve and solvency requirements. The issuer may also be required to demonstrate that a
meaningful transfer of risk has taken place.
IBRD NOTES 2018-1
With support from the World Bank, Mexico, Peru, Chile and Colombia all issued cat bonds for earthquake
risk in 2018. Collectively, these bonds had a value of around $1360m and with coupons of between 2.5% and
8.25% depending on the risk. They were all designed with tier structures, with the proportion of the principal
that investors lost in the event of an earthquake, varying in discrete steps depending on the severity of the
earthquake – for example in Peru the payout amounts were set at 30%, 70% or 100% of the bond principal.
55
.
vii This would be easier to implement with an indemnity-based trigger mechanism.
viii Artemis report that average coupon returns range from 3% to 6%, but can sometimes be as high as 15% or
higher.
58
DISASTER RISK FINANCE TOOLKIT: 4. Disaster Risk Finance Instruments 49
RISK POOLS
Municipals
and sovereign
Operational costs and
physical assets
Response
and recovery
Most cost effective when
used to respond to low
frequency, high-intensity
events e.g. beyond
1 in 10 year events
OVERVIEW
Risk pools are structures where a selection of organisations (typically administrative units) come
together to purchase insurance. The pool effectively becomes the ‚captive insurer‘ (bespoke insurance
company) for the units in question. The pool retains some of the risks itself and transfers other risks,
through reinsurance, or other instruments, to third parties. The pool is able to purchase insurance
more cheaply than if its members purchased it individually, as it offers a more diversified risk
portfolio, and because of economies of scale and greater buyer power. Pool membership may be
conditional on having a disaster response plan.
Risk pools typically use parametric triggers, allowing pay-out within 1–2 weeks, making them
suitable instruments for providing liquidity during the response phase of a disaster.
As with other insurance instruments, risk pools are better suited for the less frequent, high impact
events where relatively larger amounts of response costs need to be covered (which it will be more
difficult for risk retention mechanisms to reliably provide) and where the economic and welfare
costs of not having access to these resources will be very damaging.
DESIGN
OPTIONS
The parametric trigger needs to be designed carefully to avoid excessive basis risk
ix
. Other key
design features include which products the pool might offer and the extent to which the pool retains
risks on its balance sheet versus transferring them to reinsurance markets or through purchase of
cat bonds (see below).
CHALLENGES
Risk pools face the same types of challenges as other forms of insurance, namely that the premium
costs may be too high, and not justifiable (given the risks they are expressed to) for potential
members. There are also sometimes concerns expressed regarding whether citizens of jurisdictions
within the pool benefit from pay-outs and that such schemes fail to incentivise and change behaviour
among those who are at the frontline of facing climate impacts.
REQUIRE-
MENTS
As pools typically work with parametric triggers, the regulatory environment needs to allow
for the use of parametric products.
SOVEREIGN RISK POOLS
In recent years, sponsors have created a number of risk pools for disasters. One of the most famous is the
Caribbean Catastrophic Risk Insurance Facility Segregated Portfolio Company (CCRIF-SPC)
56
. Setup in 2007 with
World Bank assistance, this is a risk pool for small Caribbean island nations and, more recently, some Latin
American countries. It offers insurance cover for earthquakes and hurricanes. Each country in the pool members
pays a premium ranging from $200,000-$4.5 million, depending on the size of the pay-out they consider they
require following an event. Possible pay outs range from $1–100 million. The scheme is parametric and pays out
within two weeks when triggered. To date, CCRIFF has paid out around $138M to member governments.
Other examples at the national level include the Pacific Catastrophic Risk Assessment and Financing Initiative
(PCRAFI) providing coverage against tropical cyclones, earthquakes and tsunamis and the Africa Risk Capacity
(ARC) providing coverage against droughts, floods and tropical cyclones across various countries in Africa.
While many risk pool examples operate at the sovereign level, they could also work at a regional or city level.
Recent analysis for the ADB has helped to inform the development of a risk pool for different cities in the
Philippines
36
.
ix For example, Africa Risk Capacity initially failed to make a pay-out to Malawi following droughts in the 2015/16 growing
season as the model on which the modelled loss parametric trigger was based assumed that a different model of maize
to that which was actually being grown, and out-of-date information on farming practices prevented the model from
accurately replicating conditions on the ground
59
.
DISASTER RISK FINANCE – A TOOLKIT50
5. RISK MANAGEMENT STRATEGY
is section describes the linkages and interdependencies
between the DRF instruments described in section 4 and
how they can be combined to create an ecient DRF
strategy.
Section 5.1. explores the critical importance of risk
reduction in enhancing the eectiveness of a disaster risk
nance strategy and how the benets from risk reduction
might be captured.
→ Section 5.2 then explains how to combine dierent
DRF instruments and the cost and coverage benets that
can be achieved when this is done well. It particularly
focuses on risk retention and risk transfer mechanisms.
5.1. Complementarity
e benets from combining instruments increase
further when policymakers and other actors take into
account risk reduction opportunities and the various
policy mechanisms (e.g. subsidies, crediting mechanisms
and impact bonds) and nancial instruments (loans,
microcredit, bonds) that can support these investments.
Examples of risk reduction include investment in coastal
barriers (including green infrastructure), upgrading
buildings to make them more structurally resilient to
wind or ood damage, or altering the design of critical
infrastructure like roads and ports, reduce the damage
done by disasters (retrot). Such benets reduce the
damage to physical assets that events cause and, in turn,
increase the ability of the people to continue to access the
essential services that the assets provide (shelter, health,
education).
By reducing the damage caused by events, the cost of both
risk retention or risk transfer instruments fall. is means,
in turn, that the budget needed to reach a given resilience
target is lower than before the investment is made, or that
a higher level of resilience can be targeted.
‘Annual expected loss’ is a metric which is typically used
to inform the prices of retention and transfer instruments.
is metric describes the annual losses that a risk holder
would experience on average.
Figure 15 provides an illustrative example of how a
program of residential retrotting can reduce the annual
expected damage and loss from typhoon risk. e
horizontal bars represent the contribution to the total
annual expected loss across the range of return periods
(impact frequency and severity shown along vertical axes).
It shows a key feature of risk reduction – that, in
economic terms, risk reduction generates the greatest
combined cost savings by reducing the risks associated
with lower severity and more frequent events. is also
makes sense intuitively – if a risk holder builds a 10 ft.
ood barrier, the risk reduction benet of the lower half
of wall is greater than the highest half. is is because the
lower half protects against ood waters more frequently,
and therefore generates higher expected savings.
Figure 15. the resilience dividend: wind retrofit
example, showing resilience benefits in terms of
reduction in annual expected loss.
DISASTER RISK FINANCE TOOLKIT: 5. Combining instruments to create a DRF Strategy 51
e implication from this result is that risk retention
and transfer instruments which cover lower loss levels,
will see the greatest benet in terms of cost savings, and
instrument which cover the more remote layers will see a
smaller relative benet.
However, note that the analysis in Figure 15 only
demonstrates the benet of physical risk reduction in
terms of annual average savings. It does not fully reect
the fact that risk reduction can also create distinct benets
for higher severity events.
Risk reduction generates benets that extend well beyond
reducing only the economic costs of disaster. In addition,
the greater condence that extreme events will not causes
losses encourages risk taking and entrepreneurship;
while risk reduction measures can also bring important
co-benets, such as using disaster shelters as schools or
community spaces, when not being used as a shelter.
Consistent with this, a recent report for Lloyds of London
in association with the UK’s Centre for Global Disaster
Protection found that measures to boost resilience might
typically have benet cost ratios of 4:1, and in some cases
this ratio is substantially higher.
An economic analysis such as this can therefore help to
quantify the cost-benet of risk reduction, but it should
not be used in isolation.
e possibility that risk reduction investments can reduce
the cost of risk retention or risk transfer opens up an
important complementarity between these instruments
in terms of designing innovative nancial instruments,
known as resilience-linked nancing that is only
beginning to be explored. Box F below discusses this
concept in more detail.
DISASTER RISK FINANCE – A TOOLKIT52
Box F. Resilience-linked Finance
‘Resilience-linked finance’ refers to the idea that the business case for risk reduction investments could be
made through monetizing the reduction in the cost of risk retention or risk transfer. There are a number of
models through which this could be achieved including:
• Insurance-linked loan package. This would involve a loan, most probably provided by an international finance
institution to a sovereign or municipality, towards infrastructure programs where resilience is explicitly built
into the design. The loan would cover both the construction of the resilient infrastructure program and of a
parallel multi–year insurance product. However, the loan amount to cover insurance would be based on the
expected insurance premiums without the resilience measures. By contrast, the actual cost of insurance
would take account of the resilience measures built into the infrastructure design. The result would be a series
of savings on the insurance premiums which could be used to partially pay down the loan.
• Resilience bond. As described in section 4, this is a version of a cat bond where, once risk reduction investments
are undertaken, the interest rates on the bond falls to reflect the fact that investors in the bond are now less
likely to suffer such large losses in the event of a disaster
x
. This anticipated interest rate reduction could help
the financing of the risk-reduction investment
xi
.
• Resilience service company (ReSCO) could offer to finance the cost of retrofitting buildings at its own risk. This
risk reduction could result in lower insurance premium (assuming these are risk–based). The ReSCo would then
realise a return by receiving some proportion of the savings that are realised due to reduced insurance costs.
This builds on the concept of energy savings companies (ESCOs) who develop, build, and finance projects that
create energy savings. They pay for the project upfront and rely on receiving some proportion of the savings that
are realised due to the reduced energy usage to make a return on their initial investment.
These product concepts aim to both promote resilience, but also capitalize on the economic benefits of risk
reduction. If effect, capturing the expected savings and using this to part fund the additional investment required
to build resilience.
As is demonstrated in Figure 15, the greatest annual expected savings are generated in aggregate from higher-
frequency lower-severity events. The absolute savings generated through risk reduction will therefore be greater
for DRF instruments which target lower loss levels.
Another challenge is that the annual expected loss for a set of exposures is typically much smaller than the
total value of the exposure, so for the savings on expected loss to contribute significantly to the additional
cost of resilience, the risk must be high to begin with.
These types of instrument are therefore most appropriate in very high-risk regions, where low-cost resilience
measures can significantly reduce the vulnerability. Risk models can help to identify where risk reduction can
have greatest impact.
x This would be easier to implement with an indemnity-based trigger mechanism.
xi The expected in interest rate reductions could potentially even be securitised.
DISASTER RISK FINANCE TOOLKIT: 5. Combining instruments to create a DRF Strategy 53
5.2. Risk Layering
An eective risk management strategy will use risk
management actions and appropriately selected DRF
instruments in combination. e way in which the
instruments are combined has implications for both
the cost-eciency of the DRF, and the overall
eectiveness
of the DRM strategy.
As a rule of thumb, an economic and pragmatic
approach is to aim to reduce risk rst, then to arrange
risk retention, followed by risk transfer. is is known
as risk layering – Figure 16 provides an illustrative
example.
In this example:
– e resilience target is set at set at the 1 in 200-year
return period. is denes the level up to which the
risk holder will account for risk using risk retention
and transfer instruments. Losses that exceed this
target will not be actively managed using ex-ante
mechanisms.
– For the most frequent risks, with return periods
of up to about 1 in 3 years, and estimated to causes
losses of up to $9m, risk retention through reserve
funds might be most appropriate
– In this strategy, for risks with return periods of
between 1 in 3 years and 1 in 12 years, contingent
credit can be used
– Insurance then covers losses for events with return
periods between 1 in 12 years and 1 in 50 years
– Catastrophe bonds cover the residual risks up to
the 1 in 200 year return period.
Figure 16. Risk layering diagram.
300
290
280
270
260
250
240
230
220
210
200
190
180
170
160
150
|
140
|
130
|
120
|
110
|
100
|
90
|
80
|
70
|
60
|
50
|
40
|
30
|
20
|
10
|
10
135
|
130
|
125
|
120
|
115
|
110
|
105
|
100
|
95
|
90
|
85
|
80
|
75
|
70
|
65
|
60
|
55
|
50
|
45
|
40
|
35
|
30
|
25
|
20
|
15
|
10
|
5
|
0
RESILIENCE TARGET
CATASTROPHE BOND
INSURANCE
CONTINGENT CREDIT
RESERVE FUND
0% – 10% – 20% – 30% – 40% – 50% – 60% – 70% – 80% – 90% – 100%
Coverage (%)
Return Period (years)
Modeled Loss ($ million)
DISASTER RISK FINANCE – A TOOLKIT54
e most cited benet of a risk layering strategy is
reducing costs. For risk retention instruments, the
most important costs are the opportunity costs associated
with not being able to make use of the funds held in
reserve and the costs of having to pay back contingent
lines of credit. For risk transfer instruments, the key costs
are of premia payments and/or of interest rates on the
cat bonds. ese costs are captured in technical pricing
formula. Box G below explains these technical pricing
formulae and practically illustrates how a risk layering
approach reduces costs.
While technical pricing formula provide useful
general guidance on costs, the actual costs of dierent
instruments can vary over time – for example, for risk
transfer, insurance markets alternate between phases of
‘soft’ and ‘hard’ pricing (phases marked by expansion
and contraction of insurance availability, and associated
deation or ination of insurance premium costs). ere
is also near-term volatility in instrument prices that might
be driven with recent disaster losses, and changes in the
underlying perception of risk. Any actual analysis of costs
should ideally be carried with context-specic pricing
formulae that account for market and pricing dynamics.
is assessment should also take account of the costs of
setting up the instruments (frictional or transactional
costs) as well as practical considerations.
In addition to being cost eective, risk-layering also
facilitates reliable access to funds. Reliability relates to the
condence the risk holder can have that they will have
access to adequate levels of funds at the point of need.
It may be compromised if there is signicant basis risk.
As is demonstrated in the analysis in Box G, high
frequency events (1–10 year return period) tend to
drive the majority of the expected losses, implying that
reliability of funding is most important for these events.
ereby, by emphasizing the use of reserve funds for these
most frequent losses, or other instruments with limited
basis risk, risk layering also promotes reliability.
DISASTER RISK FINANCE TOOLKIT: 5. Combining instruments to create a DRF Strategy 55
Box G. Optimised Risk Layering
The analysis presented below shows how the risk-based costs of DRF instruments can be quantified, compared,
and ultimately used to help structure a layered risk financing strategy. This analysis compares the relative cost
efficiency of instruments across a range of risk levels for an illustrative risk profile.
Cost Efficiency =
Risk Premium
Expected Loss
The cost-efficiency ratio is equivalent to the cost multiple, i.e. the risk-based price of the DRF instrument
(risk premium) is the cost multiple multiplied by the modelled risk (expected loss). For a DRF instrument,
the cost multiple varies according to the underlying risk, such that one instrument may become relatively
more cost effective than another for lower frequency losses. An analysis such as this can help to inform
the structuring of an ‘optimised’ risk layering strategy.
Figure 17 illustrates the relative cost efficiency
for five DRF instruments. The analysis quantifies
the risk premium for each instrument, for a modelled
risk profile (high-frequency low-severity to
low-frequency-high severity).
The black bars represent the underlying expected
loss (risk). The sum of the bars equals the total
annual expected loss. Each bar represents a $1 million
loss band, i.e. the bottom bar is $0-1m, the second
$1-2m. etc. There is a non-linear relationship between
loss and return period, which is why the return periods
are broader at the lower losses.
The analysis highlights how the majority of annual
expected loss is contributed by high-frequency events.
The coloured lines represent the risk-based cost
estimates, or risk premium, for each DRF instrument.
The risk premiums have been estimated using
indicative pricing formula outlined in Table 1.
Risk Premium = a
j
+ b
j
· Expected Loss
The points where the lines intersect highlight the return
periods where one instrument becomes relatively
more cost-effective than the other. This can guide the
development of a layered risk financing strategy. In this
example, the most cost-efficient approach is outlined
below:
• 1–3 year return period: Reserve Fund
• 3–12 year return period: Contingent Credit
• 12–50 year return period: Insurance
• 50+ year return period: Catastrophe Bond
The 5x and 10x benchmarks are also included
for reference.
Figure 17. Comparison of DRF Cost-efficiency
(SOURCE: RMS)
DISASTER RISK FINANCE – A TOOLKIT56
Table 1. Indicative technical pricing formulare for DRF instruments
Technical Pricing Formulare
DRF Instrument Overview a
j
b
j
1 Reserve Fund
The pricing formulae for the Reserve Fund and Contingent
Credit are replicated from ‘Evaluating Sovereign Disaster
Risk Finance Strategies: A Framework’, D. Clarke (2017).
(δ-r)
(1+i)
1+r
1+i
2 Contingent
Credit
κ+ λ ·
(i-c)
(1+i))
1+c
1+i
3 Insurance Prices for insurance and reinsurance policies vary significantly,
and are influenced by many factors including the underlying risk,
how much capital the (re)insurer needs to hold relative to the
risk, desired level of return, how correlated the risk is to the
rest of the portfolio. The technical pricing formula used is based
on the expected loss, and a function of the standard deviation.
The loads on EL and standard deviation will vary according to
the specific use case.
max(0.5
· EL, 0.15
· σ
EL
)
1
4 Catastrophe
Bond
The technical pricing formula for catastrophe bonds is empirically
derived from historic bond prices and modelled expected losses.
Note that cat bond prices vary by peril, region, and trigger type
among other factors – these factors have not been isolated in the
pricing formulae.
α β
Table 2. Pricing parameter values.
Description Parameter Assumed value
Annual expected loss
EL
Variable, based on simulated loss data
from RMS catastrophe risk models.
Marginal interest rate on sovereign debt, assumed
to be average borrowing rate on government debt
portfolio
i(=δ)
5.5%
Investment return on unspent reserves
r
1%
Arrangement fee for contingent credit
κ
1%
Treatment of outstanding concessionary loans
λ
1
Interest rate on contingent credit
c
2,5%
Standard deviation of losses
σ
EL
Variable, based on simulated loss data from
RMS catastrophe risk models.
Base cost for indemnity catastrophe bond
α
2,9%
Risk load for indemnity catastrophe bond
β
1.4
DISASTER RISK FINANCE TOOLKIT: 6. Illustrative Urban Use Case 57
6. ILLUSTRATIVE URBAN USE CASE
In order to demonstrate how this toolkit could be applied
in practice, the principles outlined in the report have been
applied to realistic, but ctional, use case.
e use case provides a simplied illustrative example.
In reality, the process of developing a disaster risk
management strategy is unlikely to be so clean.
A limitation of the framework presented in this report is
that it assumes the ideal situation, in which all risk can
be measured and DRF strategies can be developed to
completely match the nancing to the underlying need.
When applying this toolkit in practice it is important
to recognize that the reality of risk management is more
complex, though this toolkit should help to provide
guidance when assessing disaster risk management, and
the disaster risk nance tools that can be used to fund it.
Situational Context
A city in South-east Asia is aiming to develop a disaster
risk nancing strategy that helps it to manage signicant
typhoon wind risk, and ooding from excess rainfall.
eir initial focus is on managing the risk to their
municipally owned physical assets, including road, water
and energy infrastructure, schools and hospitals, and
public oces.
e following illustrative example shows how the
authority could use the DRF toolkit in order to guide
their risk management strategy design, and disaster risk
nance selection.
e DRF toolkit has been designed as a guideline to
inform how disaster risk nancing instruments can be
used to support a risk management strategy. In practice,
the specic situational context will inform appropriate use
of DRF, and this illustrative example does not provide the
only possible solution for an urban use case.
DISASTER RISK FINANCE – A TOOLKIT58
Phase Task Process
1
RISK AUDIT
Exposure
definition
The city carries out an exposure data collection exercise – the output from this
is a database that contains asset-level information which describes the assets.
The database contains exposure information which includes occupancy type
(e.g. residential, commercial, highway), the construction method and materials
(e.g. masonry, timber-frame), age (year-built), and a value estimate that is based
on reconstruction value. The city also collects information on where people with
different incomes levels (and other characteristics determining vulnerability) live,
work and access essential services.
This exposure data forms the input to catastrophe risk models. More detailed
exposure information can help to create more accurate results, but may take more
time and effort and can cost more to collect. Simple input data can provide good
enough results to make initial risk-based decisions. The city decides to err on the
side of simplicity, with the aim to enhance the exposure data later if necessary.
Peril
identification
The city had recently experienced a very damaging flood event, which had
motivated the city authority to manage its risk more actively.
The flood damage from the recent event was fresh in the minds of the local
population, and this is the primary focus. However, stakeholder engagement
also identifies that severe typhoons are a key concern to residents and
businesses. Despite the fact that flood risk is a more frequent issue, the decision
is made to also investigate ways to manage the typhoon risk. A multi-peril
approach to risk management also allow the city to make most use of the
collected exposure data, and potentially ‘bundle’ more risk to (potentially) be
transferred to others.
Risk
quantification
The city approaches the national risk management agency, who has access
to risk modeling capabilities. The national risk management agency is supported
by development partners in utilizing and interpreting this data.
The flooding and typhoon risk are quantified, and the city is provided with a
risk analysis which uncovers some new insight into which assets and people are
most at risk, as well as an overall risk profile – this provides the foundation
for risk-informed decision making.
Resilience
targeting
The city finds out that the recent flood event was approximately equivalent to a
1 in 150-year return period loss. Using this experience as a benchmark, a decision
is made to ensure that the risk management strategy is able to manage all
disaster losses up to this level.
The initial resilience target is therefore set at the 150-year return-period loss.
The city therefore looks to develop a strategy of risk reduction, retention, and
transfer that ensures they are able to actively manage all losses up to the
resilience target.
DISASTER RISK FINANCE TOOLKIT: 6. Illustrative Urban Use Case 59
Phase Task Process
2
RISK MANAGEMENT ACTIONS
Reduction Based on the risk audit, the city learns that the hospitals contribute the majority
to the overall risk for the city. Community engagement also reveals that it is also
a risk that particularly affects vulnerable people, who tend to make more use of
secondary care facilities.
As such, the city prioritizes investment in risk reduction actions for hospitals.
It uses risk models to quantify the possible risk reduction benefits for a range
of risk reduction options, and uses this to inform a cost-benefit calculation.
The analysis indicates that a cost-effective solution is to raise the electrical
equipment from the basement to higher floors.
The proposed risk reduction activity reduces the 1 in 150-year return period
loss by 20%, which leaves 80% of the resilience target loss left to be managed
using risk retention and risk transfer instruments. The city now proceeds to better
define its capacity and needs, so that it can decide how much risk it wishes to
retain, and how much it should transfer.
Retention
Transfer
3
DIMENSIONS OF INSTRUMENT DESIGN
Risk holder The city raises capital through local taxes, and also receives an allocated
budget for disaster risk management from the national finance ministry.
They have an annual budget that is approximately equivalent to 2x the average
annual modeled loss.
Purpose Funding is required to support the costs of retrofitting for the hospitals.
The city also requires operational funds for restoring essential services
immediately following any disaster, and funding for repair and reconstruction
of the physical assets.
Timing Funds for risk reduction are required immediately. Funds for restoring essential
services should be available as soon as possible following impact, if not before.
Funds for reconstruction will be required over the longer term, the speed of
financing is not so important, but the funding needs to closely match the loss.
Risk level The city is aiming to make itself resilient for all risk levels, from the frequent
attritional losses, up to the 1 in 150-year return period resilience target.
4
DRF INSTRUMENT OPTIONS
Having identified the risk management actions, and further defined the needs
for the funding, the city now assesses the range of DRF options that are available
to it.
To fund the risk reduction activities, it identifies the following instruments
from the DRF taxonomy; loans, bonds and impact bonds.
For risk retention, the city identifies budget contingency, and reserve funds
as possible options.
For risk transfer, the city identifies insurance, catastrophe bonds, and risk
pools as appropriate mechanisms.
With a range of possible options identified, the city now needs to select from
these options.
DISASTER RISK FINANCE – A TOOLKIT60
5
RISK LAYERING
The city has identified that the risk reduction exercise should lower the costs
of risk retention and transfer instruments. Selecting an instrument to fund the
risk reduction activity is therefore a priority.
The city determines that impact bonds would take a long time to arrange, so
given the time constraints a bank loan or bond are more attractive. The cost
estimate for the hospital retrofits is significant, so the city decides to issue bonds
to raise capital for the project. It also commits to exploring in the medium term
how impact bonds could improve the efficiency and resilience of its hospitals,
building on the experience of the Humanitarian Impact Bond experience in
improving health care performance.
The hospital retrofit has reduced the overall 1 in 200 year return period loss
by 10%. There are also additional benefits in terms of reducing the expected
disruption to hospital services.
The city carries out a risk layering analysis and determines that the optimal
level to start taking insurance is around the 1 in 10 year return period loss.
The city already has a budgetary allocation for flood impacts, which was used
effectively during the recent events – it is decided to build on this with an
additional reserve fund, which is allocated to pay for the costs of clearing the
roads immediately following disaster. The regional transportation authority is
consulted, and a rapid response plan is designed to make most effective use
of the reserve fund.
The total risk of the city owned assets is too small to justify the additional
costs required to implement a catastrophe bond or risk pool. The city elects
to purchase insurance for the remainder of the risk. The insurance premium
quoted to cover all of the assets up to the 1 in 200-year resilience target
exceeds the funds that the city has available. The city renegotiates for a
reduced amount of cover.
This completes the disaster risk financing strategy for this year, but the city
has now identified where there are protection gaps within its own strategy, and
identified a number of additional initiatives which may help to reduce the cost
of DRF. The city begins to engage with other cities in neighboring regions to
share its DRF experience, and explore options for a city-level risk pool which
might help it to achieve its resilience targets in future years.
MODULE 1: GENERAL GUIDE 61
REFERENCES
1. IPCC. Summary for Policy Makers In Climate Change 2014: Impacts, Adaptation and Vulnerability. Contribution
of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. 2014.
2.
INFORM Index for Risk Management [Internet]. [cited 2019 Apr 5]. Available from: http://www.inform-index.org/
3.
United Nations. The Sustainable Development Goals Report [Internet]. 2018. Available from:
https://unstats.un.org/sdgs/files/report/2018/TheSustainableDevelopmentGoalsReport2018-EN.pdf
4. Ranger N, Surminski S, Mitchell T, Jones L, Lovell E, Comba E. Disaster resilience and post-2015 development
goals: the options for economics targets and indicators. Policy Pap Grantham Res Inst Clim Chang Environ
London, UK. 2013;
5.
United Nations. Sendai Framework for Disaster Risk Reduction 2015 - 2030 [Internet]. 2015. Available from:
https://www.unisdr.org/files/43291_sendaiframeworkfordrren.pdf
6.
UNFCCC. Warsaw International Mechanism for Loss and Damage [Internet]. 2013 [cited 2019 Apr 5].
Available from: https://unfccc.int/topics/adaptation-and-resilience/workstreams/loss-and-damage-ld/warsaw-
international-mechanism-for-loss-and-damage
7. United Nations. Agenda for Humanity - Annex to the Report of the Secretary-General for the World
Humanitarian Summit [Internet]. 2016. Available from: https://agendaforhumanity.org/sites/default/files/
AgendaforHumanity.pdf
8. Clarke DJ, Dercon S. Dull Disasters?: How planning ahead will make a difference. Oxford University Press;
2016.
9.
Molinario G. Demystifying machine learning for disaster risk management [Internet]. 2019 [cited 2019 Apr
5]. Available from: https://blogs.worldbank.org/opendata/demystifying-machine-learning-disaster-risk-
management
10. Twigg J. Disaster risk reduction: mitigation and preparedness in development and emergency programming.
Overseas Development Institute (ODI); 2004.
11.
International Federation of Red Cross and Red Crescent Societies. Forecast-based Financing - A new era
for the humanitarian system [Internet]. 2019. Available from: https://www.forecast-based-financing.org/wp-
content/uploads/2019/03/DRK_Broschuere_2019_new_era.pdf
12. Lloyds’s of London, Centre for Global Disaster Protection. Innovative finance for resilient infrastructure
[Internet]. 2018. Available from: https://www.lloyds.com/news-and-risk-insight/risk-reports/library/
understanding-risk/innovative-finance-for-resilient-infrastructure?utm_source=Lloyds_posts&utm_
medium=social&utm_campaign=emergingrisks_innovative-finance-for-resilient-infrastructure
13.
European Bank for Reconstruction and Development. The Tajikistan Climate Resilience Financing Facility -
CLIMADAPT [Internet]. [cited 2019 Feb 22]. Available from: https://www.climadapt.tj/
14.
VisionFund International. Recovery Lending in Africa [Internet]. [cited 2019 Apr 5]. Available from:
http://www.visionfund.org/2430/recovery-lending-in-africa/disaster-resilient-microfinance/microfinance/
15. Stewart R, van Rooyen C, Dickson K, Majoro M, de Wet T. What is the impact of microfinance on poor people?
A systematic review of evidence from sub-Saharan Africa. London; 2010.
16.
Pilot Programme for Climate Resilience. Microfinance for Climate Resilience [Internet]. 2018. Available from:
https://ppcrja.org.jm/download/microfinance-in-the-ppcr-jamaica-case-study/?wpdmdl=566
17.
Climate Bonds Initiative. Bonds and Climate Change: The State of the Market 2018 [Internet]. 2018. Available
from: https://www.climatebonds.net/resources/reports/bonds-and-climate-change-state-market-2018
18.
Ernst and Young. Green Bonds: A Fresh Look at Financing Green Bonds [Internet]. 2016. Available from:
https://www.ey.com/Publication/vwLUAssets/Green_bonds-a-fresh-look-at-financing-green-projects/$FILE/EY-
Green bonds-a-fresh-look-at-financing-green-projects.pdf
19. Mecagni M, Kriljenko JIC, Gueye CA, Mu I, Yabara M, Weber S. Issuing international sovereign bonds:
Opportunities and challenges for Sub-Saharan Africa. International Monetary Fund; 2014.
20.
The World Bank. Fiji Issues First Developing Country Green Bond, Raising $50 Million for Climate
Resilience [Internet]. 2017 [cited 2019 Feb 18]. Available from: https://www.worldbank.org/en/news/press-
release/2017/10/17/fiji-issues-first-developing-country-green-bond-raising-50-million-for-climate-resilience
21.
City of Cape Town. Green pays: City’s R1 billion bond a resounding success in the market [Internet]. 2017 [cited
2019 Apr 5]. Available from: http://www.capetown.gov.za/media-and-news/Green pays City
FINANCE & INSURANCE TOOLKIT For the Renewable Energy Sector in Barbados62
22. Moody’s Investors Service. Green bond market poised to hit $200 billion in 2019
[Internet]. 2019 [cited 2019 Feb 18]. Available from: https://www.moodys.com/research/
Moodys-Green-bond-market-poised-to-hit-200-billion-in--PBC_1159526?WT.mc_
id=AM~RmluYW56ZW4ubmV0X1JTQl9SYXRpbmdzX05ld3NfTm9fVHJhbnNsYXRpb25z~20190131_PBC_1159526
23.
Infrastructure Canada. Disaster Mitigation and Adaptation Fund [Internet]. 2018 [cited 2019 Feb 18].
Available from: https://www.infrastructure.gc.ca/dmaf-faac/index-eng.html
24.
Adaptation Fund. Flood Resilience in Ulaanbaatar Ger Areas – Climate Change Adaptation through
community-driven small-scale protective and basic-services interventions [Internet]. 2018 [cited 2019
Apr 5]. Available from: https://www.adaptation-fund.org/project/flood-resilience-ulaanbaatar-ger-areas-fruga-
climate-change-adaptation-community-driven-small-scale-protective-basic-services-interventions-2/
25.
Green Climate Fund. Scaling-up Multi-Hazard Early Warning System and the Use of Climate Information
in Georgia [Internet]. 2018 [cited 2019 Feb 18]. Available from: https://www.greenclimate.fund/projects/
fp068?inheritRedirect=true&redirect=%2Fwhat-we-do%2Fprojects-programmes%3Fp_p_id%3D122_
INSTANCE_VKj2s9qVF7MH%26p_p_lifecycle%3D0%26p_p_state%3Dnormal%26p_p_mode%3Dview%26p_p_
col_id%3D_118_INSTANCE_4ZRnUzRWpEqO__column-2%26p_p_col_count%3D2%26p_r_p_564233524_
resetCur%3Dtrue%26p_r_p_564233524_categoryId%3D846529
26.
United States Environmental Protection Agency. Mitigation Banking Factsheet [Internet]. [cited 2019 Feb 18].
Available from: https://www.epa.gov/cwa-404/mitigation-banking-factsheet
27.
African Development Bank Group. Adaptation Benefit Mechanism (ABM) [Internet]. 2018 [cited 2019 Feb 18].
Available from: https://www.afdb.org/en/topics-and-sectors/initiatives-partnerships/adaptation-benefit-
mechanism-abm/
28.
Oroxom R, Glassman A, McDonald L. Structuring and Funding Development Impact Bonds for Health:
Nine Lessons from Cameroon and Beyond. 2018; Available from: https://www.cgdev.org/sites/default/files/
structuring-funding-development-impact-bonds-for-health-nine-lessons.pdf
29.
Gustafsson-Wright E, Gardiner S, Putcha V. The potential and limitations of impact bonds:
Lessons from the first five years of experience worldwide [Internet]. 2015. Available from:
https://www.brookings.edu/wp-content/uploads/2016/07/Impact-Bondsweb.pdf
30.
International Committee of The Red Cross. The world’s first “Humanitarian Impact Bond” launched to transform
financing of aid in conflict-hit countries [Internet]. 2017 [cited 2019 Feb 18]. Available from: https://www.icrc.
org/en/document/worlds-first-humanitarian-impact-bond-launched-transform-financing-aid-conflict-hit
31.
International Federation of Red Cross and Red Crescent Societies. Forecast-based Financing [Internet].
[cited 2019 Mar 19]. Available from: https://www.forecast-based-financing.org/
32.
Start Network. Start Fund Crisis Anticipation Window Annual Report 2017 [Internet]. 2017. Available from:
https://startnetwork.org/resource/start-fund-crisis-anticipation-window-annual-report-2017
33.
The World Bank. World Bank Launches First-Ever Pandemic Bonds to Support $500 Million Pandemic
Emergency Financing Facility [Internet]. 2017 [cited 2019 Mar 19]. Available from: https://www.worldbank.org/
en/news/press-release/2017/06/28/world-bank-launches-first-ever-pandemic-bonds-to-support-500-million-
pandemic-emergency-financing-facility
34. Ghesquiere F, Mahul O. Financial protection of the state against natural disasters: a primer. World Bank Policy
Res Work Pap. 2010;(5429).
35. Wang X, Mahul O, Stutley C. Weathering the storm: options for disaster risk financing in Vietnam. World Bank,
Washingt DC. 2010;
36.
Asian Development Bank. Philippine City Disaster Insurance Pool: Rationale and Design [Internet]. 2018.
Available from: https://www.adb.org/sites/default/files/publication/479966/philippine-city-disaster-insurance-
pool-rationale-design.pdf
37.
Food and Aggriculture Organization of the United Nations. Community Contingency Funds: Savings as a means
of resilience [Internet]. Available from: http://www.fao.org/3/a-i5623e.pdf
38. Schijman A, Haarsager U, Hernández MJ, Khadr A, Jansson M. Corporate Evaluation: Contingent Lending
Instruments. Inter-American Development Bank; 2016.
39.
The World Bank. IDA Catastrophe Deferred Drawdown (Cat DDO) [Internet]. 2018. Available from:
http://pubdocs.worldbank.org/en/563361507314948638/product-note-cat-ddo-ida-english-2018.pdf
40.
The World Bank. IBRD Catastrophe Deferred Drawdown Option (Cat DDO) [Internet]. 2018. Available from:
http://pubdocs.worldbank.org/en/526461507314946994/product-note-cat-ddo-ibrd-2018.pdf
MODULE 1: GENERAL GUIDE 63
41. The World Bank. Disasters, funds, and policy: Creatively meeting urgent needs and long-term policy goals
[Internet]. 2017 [cited 2019 Feb 18]. Available from: https://reliefweb.int/report/world/disasters-funds-and-
policy-creatively-meeting-urgent-needs-and-long-term-policy-goals
42.
Crawford D, Kroessin M, Loeffler K. Mitigating climate change with Takaful insurance [Internet]. 2018.
Available from: https://www.gbgfund.org/mitigating-climate-change-takaful-insurance/
43.
ClimateWise. ClimateWise Compendium of disaster risk transfer initiatives in the developing world [Internet].
2011. Available from: https://www.cisl.cam.ac.uk/business-action/sustainable-finance/climatewise/pdfs/
overview-of-compendium.pdf/view
44. Jensen ND, Barrett CB, Mude AG. Cash Transfers and Index Insurance: A Comparative Analysis from Northern
Kenya. J Dev Econ. 2015;129:14–28.
45. Mahul O, Stutley CJ. Government support to agricultural insurance: challenges and options for developing
countries. The World Bank; 2010.
46.
World Food Programme. The R4 Rural Resilience Initiative [Internet]. [cited 2019 Mar 29]. Available from:
https://www1.wfp.org/r4-rural-resilience-initiative
47. Gönülal SO. Takaful and mutual insurance: alternative approaches to managing risks [Internet]. Washington, DC:
The World Bank; 2012. Available from: https://openknowledge.worldbank.org/handle/10986/13087
48.
icmif. Access to mutual and cooperative insurance infographic [Internet]. [cited 2019 Mar 29]. Available from:
https://www.icmif.org/publications/strategic-research/access-mutual-and-cooperative-insurance-infographic
49.
Takaful Insurance of Africa. Index Based Livestock Takaful (IBLT) [Internet]. [cited 2019 Mar 29].
Available from: https://takafulafrica.co.ke/takaful-insurance-products/iblt/
50.
Business Daily. Insurer to compensate livestock farmers [Internet]. 2014 [cited 2019 Apr 5]. Available from:
https://www.businessdailyafrica.com/Insurer-to-compensate-livestock-farmers/-/539552/2256242/-/gw6ico/-/
index.html
51.
King P, Rami S. Insurance regulation in Africa: CIMA – the benefits and challenges [Internet]. 2018 [cited
2019 Apr 5]. Available from: https://www.commercialriskonline.com/insurance-regulation-in-africa-cima-the-
benefits-and-challenges/
52. OCDE (Organización para la Cooperación y el Desarrollo Económicos). OECD Reviews of Risk Management
Policies: Mexico 2013: Review of the Mexican National Civil Protection System. OCDE París; 2013.
53.
Risk Management Solutions. Mapping the Role of Insurance in Managing Disaster Losses: A Study of Low and
Low-Middle Income Countries [Internet]. 2017. Available from: https://www.preventionweb.net/files/54328_
mappingtheroleofinsuranceinmanaging.pdf
54.
Re:Focus Partners. A Guide for Public-Sector Resilience Bond Sponsorship [Internet]. 2017. Available from:
http://www.refocuspartners.com/wp-content/uploads/pdf/RE.bound-Program-Report-September-2017.pdf
55.
Artemis. Artemis Catastrophe Bond & Insurance-Linked Securities Deal Directory [Internet]. [cited 2019 Feb
18]. Available from: http://www.artemis.bm/deal-directory/
56.
CCRIF SPC: The Caribbean Catastrophe Risk Insurance Facility [Internet]. [cited 2019 Feb 18]. Available from:
https://www.ccrif.org/
57.
Government Outcomes Lab. An introduction to development impact bonds (DIBs) [Internet]. [cited 2019 Apr 5].
Available from: https://golab.bsg.ox.ac.uk/knowledge/basics/introduction-development-impact-bonds/
58.
Evans S. Cat bond attraction as investment & risk transfer to persist, perhaps intensify [Internet]. 2019 [cited
2019 Apr 5]. Available from: http://www.artemis.bm/news/cat-bond-attraction-as-investment-risk-transfer-to-
persist-perhaps-intensify/
59.
African Risk Capacity. African Risk Capacity – Response to ActionAid’s Flawed Claims [Internet]. 2017 [cited
2019 Apr 5]. Available from: http://www.africanriskcapacity.org/2017/07/10/african-risk-capacity-response-to-
actionaids-flawed-claims/
FINANCE & INSURANCE TOOLKIT For the Renewable Energy Sector in Barbados64
ACKNOWLEDGEMENTS
is paper was commissioned by the Deutsche
Gesellschaft für Internationale Zusammenarbeit (GIZ)
GmbH under the “Advancing Climate Risk Insurance
plus” (ACRI+) project. e ACRI+ project is part of a
larger programme entitled “Promoting Integrated
Climate Risk Management and Transfer” (ICRM)
which is funded through the International Climate
Initiative (IKI) of the German Federal Ministry for
the Environment, Nature Conservation and Nuclear
Safety (BMU). e project is jointly implemented by
the Munich Climate Insurance Initiative e. V. (MCII)
and GIZ GmbH.
e report was compiled by Conor Meenan
(conor.meenan@rms.com), John Ward
(john.ward@pengwernassociates.com), and
Robert Muir-Wood. Although it would not have been
possible without the keen insights from Daniel Stander,
Charlotte Acton, Laurence Carter, Stephen Moss,
eresa Lederer and Deborah Leahy, whose
contributions have been invaluable.
Reviewers
To ensure the relevance and acceptance of this toolkit,
deep appreciation goes to our reviewers from:
– OASIS Loss Modelling Framework
– Oxfam UK
– World Food Programme (WFP)
– German Federal Foreign Oce
– International Federation of Red Cross
and Red Crescent Societies (IFRC)
– UNDRR
– UNDP
– African Risk Capacity (ARC)
Disclaimer
is report, and the analyses, models, methodologies,
and predictions contained herein („Information“),are
based on RMS research, expertise, data provided by third
parties, and where applicable, proprietary computer risk
assessment technology of Risk Management Solutions,
Inc. („RMS“).
e ndings and recommendations detailed in the
Information are based on the RMS’ analysis of the
available data sources as described in the document.
eInformation in this report may not be copied,
photocopied, reproduced, translated, reduced to any
electronic medium or machine-readable form for
presentation in any other format without the prior
written consent of RMS. is Information should not
be used under any circumstances in the development
or calibration of any product or service oering that
competes with RMS.
e recipient of this Information is further advised that
RMS is not engaged in the insurance, reinsurance, or
related industries, and that the Information provided is
not intended to constitute professional advice.
RMS specifically disclaims any and all responsibilities,
obligations, and liability with respect to any decisions
or advice made or given as a result of the information
or use thereof, including all warranties, whether
express or implied, including but not limited to,
warranties of non-infringement, merchantability, and
fitness for a particular purpose. In no event shall
RMS (or its parent, subsidiary, or other affiliated
companies) be liable for direct, indirect, special,
incidental, or consequential damages with respect to
any decisions or advice made or given as a result of
the contents of this information or use thereof.
About ACRI+
ACRI+ is implemented by Munich Climate Insurance Initiative (MCII)
and GIZ and nanced by BMU and is part of the Promoting Integrated
Mechanisms for Climate Risk Management and Transfer programme (ICRM),
implemented by GIZ.
For more information, please visit:
http://www.climate-insurance.org/projects/advancing-climate-risk-
insurance-acri/
About GIZ
e Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH
oers customized solutions to complex challenges. GIZ is an experienced
service provider and assists the German government in achieving its objectives
in the eld of international cooperation. GIZ oers demand-driven,
tailor-made and eective services for sustainable development.
For more information, please visit: www.giz.de/en/
About RMS
For almost 30 years, RMS’ mission has remained constant: to make
communities and economies more resilient to disasters through a deeper
understanding of catastrophes. Today, RMS is a leading provider of
products and services that help governments, public agencies, corporations,
nancial institutions and their clients to evaluate and manage catastrophe
risks throughout the world.
For more information, please visit: https://www.rms.com/
About Pengwern Associates
Pengwern Associates is a UK-based consultancy specialising in the economics
of climate change, the environment, international development and the
linkages between them. Across these areas, it provides advice to support
strategy development, decision-making and implementation, drawing on
both quantitative and qualitative analysis.
For more information, please visit: www.pengwernassociates.com
IMPRINT
Published by
Deutsche Gesellschaft für
Internationale Zusammenarbeit (GIZ) GmbH
Registered offices
Bonn and Eschborn, Germany
Friedrich-Ebert-Allee 40
53113 Bonn, Germany
Phone: +49 228 44 60-0
Fax: +49 228 44 60-17 66
Dag-Hammarskjöld-Weg 1-5
65760 Eschborn
Phone: +49 61 96 79-0
Fax: +49 61 96 79-11 15
Email: [email protected]
Internet: www.giz.de
Responsible for the content:
Matthias Range (matthias.range@giz.de)
Eike Behre ([email protected])
Design:
www.kromativ.de, Romy Glende
Programme:
Advancing Climate Risk Insurance Plus
(ACRI+)
Picture credits:
cover - © shutterstock_705475312
page 5 -© pixabay 2181892
page 9 - ©pixabay-4168234
page 15 - ©pixabay-2770340
Articles attributed to named authors
do not necessarily reflect the views of
Deutsche Gesellschaft für Internationale
Zusammenarbeit (GIZ) GmbH
Deutsche Gesellschaft für
Internationale Zusammenarbeit (GIZ) GmbH
Registered offices
Bonn and Eschborn, Germany
Friedrich-Ebert-Allee 36 + 40
53113 Bonn, Deutschland
T +49 228 44 60-0
F +49 228 44 60-17 66
E info@giz.de
I www.giz.de
Dag-Hammarskjöld-Weg 1 - 5
65760 Eschborn, Deutschland
T +49 61 96 79-0
F +49 61 96 79-11 15
On behalf of:
