A FOLLOW-UP TO THE UNBREAKABLE REPORT

INTRODUCTION AND SUMMARY

Achieving resilience
through stronger,
faster, and more
inclusive post-disaster
reconstruction

a I B U I L D I N G B AC K B E T T E R
Stephane Hallegatte, Jun Rentschler, Brian Walsh
INTRODUCTION AND SUMMARY

b I B U I L D I N G B AC K B E T T E R
© 2018 International Bank for Reconstruction and Development / The World Bank
1818 H Street NW
Washington, DC 20433
Telephone: 202-473-1000
Internet: [Link]

This work is a product of the staff of The World Bank with external contributions. The findings,
interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World
Bank, its Board of Executive Directors, or the governments they represent.

The World Bank does not guarantee the accuracy of the data included in this work. The boundaries,
colors, denominations, and other information shown on any map in this work do not imply any judgment
on the part of The World Bank concerning the legal status of any territory or the endorsement or
acceptance of such boundaries.

Rights and Permissions

The material in this work is subject to copyright. Because The World Bank encourages dissemination of
its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as
full attribution to this work is given.

Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank
Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625;
e-mail: pubrights@[Link].

Cover design by Brad Amburn.

INTRODUCTION AND SUMMARY

ACKNOWLEDGEMENTS

This report has been prepared by a team composed of Stephane Hallegatte, Jun
Rentschler, and Brian Walsh. It benefited from invaluable comments and advice
from Elif Ayhan, Francis Ghesquiere, Joe Leitman, Thomas Lennartz, and Alice
Mortlock throughout the drafting process. For their time and feedback the
authors also thank the peer reviewers of this report: Artessa Saldivar-Sali, Marc
Forni, and Niels Holm-Nielson. Further helpful comments were received from
participants of an internal policy research seminar organized by the Global
Facility for Disaster Reduction and Recovery (GFDRR).

Editorial services were provided by Nick Paul. Brad Amburn designed the
report. Visibility and launch of the report were supported by Elisabeth Mealey
and Mehreen Sheikh.

The report was sponsored by GFDRR and the Climate Change Group of the
World Bank, under the leadership of John Roome.

i I B U I L D I N G B AC K B E T T E R
FOREWORD

T he 2017 Unbreakable report made the case that disaster losses

disproportionately affect poor people. The report showed that
they have limited ability to cope with disasters, and estimated that
the impact on well-being is equivalent to consumption losses of
about $520 billion a year around the world—outstripping previous
estimates of pure asset losses by as much as 60 percent.

The Caribbean Hurricane season of 2017 was a tragic illustration of this. Two
Category 5 hurricanes wreaked destruction on numerous small islands, causing severe
damage in places like Barbuda, Dominica, and Saint Martin. The human cost of these
disasters was immense, and the impact of this devastation was felt most strongly by
poorer communities in the path of the storms.

And yet, amidst the destruction it is essential to look forward and to build back better.
In this report we follow up on the Unbreakable report and explore how countries
can strengthen their resilience to natural shocks through stronger, faster, and more
inclusive post-disaster reconstruction. It shows that reconstruction needs to be strong,
so that assets and livelihoods become less vulnerable to future shocks; fast, so that
people can get back to their normal life earlier; and inclusive, so that nobody is left
behind in the recovery process.

This report shows how the benefits of building back better could be greatest among
the communities and countries that are hit by disasters most intensely and frequently.
For a selection of small island states, this report shows that stronger, faster, and more
inclusive recovery would lead to an average reduction in disaster-related well-being
losses of 59 percent. For Antigua & Barbuda, the reduction is as large as 74 percent.

ii I B U I L D I N G B AC K B E T T E R
“This report shows how the benefits of
building back better could be greatest
among the communities and countries
that are hit by disasters most intensely
and frequently.”

Indeed, small island developing states have been at the forefront of the fight against
the devastating impacts of climate change. High exposure to natural hazards, high
event frequency, and the concentration of assets mean that many small islands face
far higher risks to assets and well-being than other countries. In these places, building
back better can mean the difference between standing tall or repeated devastation.

Of course, strategies for building back better cannot replace measures for disaster
prevention and preparedness. However, they can be integrated into comprehensive
disaster risk management frameworks and can help communities seize opportunities
for building resilience. This report provides estimates for the benefits of resilient
recovery. Various case studies tell the stories of countries that have emerged stronger
from devastating disasters and offer lessons to be shared with the rest of the world.

Francis Ghesquiere
Head, Global Facility for Disaster Reduction and Recovery

iii I B U I L D I N G B AC K B E T T E R
 INTRODUCTION
INTRODUCTION AND SUMMARYAND SUMMARY

T he 2017 hurricane season in the Caribbean will surely

be remembered for years to come. Hitting the region in
mid-September, Hurricane Irma was followed within days by
Hurricane Maria both of which reached the highest category of
intensity, allowing no respite to people across the Caribbean. They
wreaked destruction on numerous small islands, causing severe
damage in places like Barbuda, Dominica, and Saint Martin.

The human cost of disasters like these is immense: many lives were lost, although
early warning systems and timely evacuations were able to save many more. And
many survivors have lost lifelong savings, homes, and livelihoods. Destruction in
the infrastructure and residential sectors is likely to exceed 100 percent of GDP
on several islands. In Dominica, 70 to 80 percent of all houses and buildings
sustained major storm damage, from ripped-off roofs to total destruction. Entire
regions lost access to basic services, such as electricity and safe drinking water,
and some remote communities were cut off completely for days. Returning to
normality will take years.

The risk of hurricanes is a growing annual threat: the Caribbean hurricane

season extends from the beginning of June through the end of November
each year, and the frequency and intensity of storms may be exacerbated by
climate change in the decades to come. In addition, continued urbanization
and population growth in coastal areas around the world are putting more
communities in harm’s way.

Considering these trends, the destruction caused by disasters reveals the need,
but also the opportunity, to build back better. As highlighted in Priority 4 of the
Sendai Framework for Disaster Risk Reduction adopted in 2015, reconstruction
offers an opportunity to build more resilient societies. These are characterized
as more able to withstand future shocks by better managing the risks they face:
with new buildings located outside flood zones and with structures designed to
resist high winds; with roads, bridges, and electric grids that are able to endure
the next storm; and with human settlements that provide a better quality of life
and enable higher productivity. Such a stronger recovery can reduce the impact
and the cost to well-being associated with future disasters.

1 I B U I L D I N G B AC K B E T T E R
INTRODUCTION AND SUMMARY

Beyond this traditional understanding of building back better, there are many opportunities
to improve the recovery and reconstruction phase that follows a disaster, so that well-
being impacts can be minimized. A faster recovery can ensure that people restore their
income and assets as early as possible, making it possible to use their savings to maintain
consumption levels. And a more inclusive recovery can ensure that the poorest and the most
vulnerable can access the support they need to reconstruct. In the absence of such support,
they are the most likely to experience the long-term consequences caused by health issues
and disability, loss of schooling and education, or simply the inability to save or borrow to
rebuild or replace lost assets. A rapid and more inclusive reconstruction is key to preventing
poor people from falling into poverty traps that can magnify the impacts of disasters.

In this study, therefore, building back better means that the repaired or replaced assets are
more resilient, but also that the recovery process is shorter and more efficient, and that the
entire recovery process does not leave anyone behind—i.e. that even the poorest and most
vulnerable receive the support they need to fully recover. The study investigates the potential
benefits of building back better, building on the framework and model described in the
Unbreakable report, and considering the three dimensions, independently and together:

• Building back stronger reduces well-being losses by ensuring that reconstructed

infrastructure can resist more intense events in the future. If all countries were to “build
back stronger” in the next 20 years—ensuring that rebuilt assets can resist hazards with
a 50-year return-period—then global well-being losses due to natural disasters would
be reduced by 12 percent, a gain equivalent to US$65 billion annually. Stronger
reconstruction would reduce overall well-being losses due to natural disasters by
more than 40 percent in ten countries in particular: Antigua and Barbuda, Dominica,
Vanuatu, Myanmar, Laos, Tonga, Guatemala, Trinidad and Tobago, Peru, and Fiji.

• Building back faster reduces disaster impacts by accelerating reconstruction through

measures such as contingent reconstruction plans, pre-approved contracts, and financial
arrangements. Estimates in this report show that if the average reconstruction speed is
reduced by two thirds (without compromising the quality of reconstruction), global
well-being losses could be reduced by 14 percent—equivalent to increasing global
consumption by over US$75 billion per year. These gains are especially pronounced in
countries with frequent events, such as small island countries or Sub-Saharan countries.

• Building back more inclusively ensures that post-disaster support reaches all affected
population groups. This emphasizes the importance of providing reconstruction support

2 I B U I L D I N G B AC K B E T T E R
INTRODUCTION AND SUMMARY

to low-income households, which are typically more exposed, more vulnerable, and less
comprehensively supported. If all countries had the ability to provide the poorest people
with the post-disaster support found in developed countries, global well-being losses due
to natural disasters could be reduced by 9 percent, equivalent to a US$52 billion increase
in annual global consumption. The effect is particularly pronounced in countries with
high inequality, and where poor people have little access to social protection and financial
instruments. In Angola, Benin, Comoros, the Republic of the Congo, the Central African
Republic, the Democratic Republic of the Congo, Russia, Gabon, Haiti, and Lesotho,
building back more inclusively could reduce disaster losses by 27 percent or more.

If implemented together, these three strategies—rebuilding stronger, faster, and more

inclusively—could generate major benefits, totaling US$173 billion per year, or 31
percent of current well-being losses due to natural disasters.

Building back better is particularly important in small island countries, due to their high
current levels of vulnerability, and their small scale. In the small island states included in this
analysis, building back better could lead to an average reduction in disaster-related well-
being losses of 59 percent.

Using the reconstruction process to upgrade assets and increase their productivity—for
example, by using the most recent technologies or adapting old infrastructure systems
to current and future needs—would generate further economic benefits, making it even
more attractive to invest in a better recovery and reconstruction process.

Such resilient and effective recovery and reconstruction is possible only if the appropriate
policies and tools are made available to affected households, firms, and local and national
authorities before the disaster hits. These are usually incorporated into a disaster recovery
framework that include contingency plans and institutional arrangements with a clear
allocation of responsibility in the recovery period, access to practical knowledge and
information, and strong and inclusive financial protection provided by a combination of
disaster-response social safety nets, insurance mechanisms, and access to borrowing to
finance the reconstruction.

While a better recovery and reconstruction process cannot replace investments in

disaster risk reduction and prevention, this study provides many examples of policies and
interventions that have made countries better able to face the next disaster and that could be
replicated in the rest of the world to contribute to a more resilient future.

3 I B U I L D I N G B AC K B E T T E R
 BEYOND
BEYOND ASSET
ASSET LOSSES LOSSES:
Assessing socio-economic resilience
and losses in well-being

M ost assessments of losses due to natural disasters

focus on damages to assets—including buildings,
infrastructure, equipment, and production. According
to such estimates, in 2017, global economic losses due to
weather-related natural disasters—from hurricanes and
wildfires to droughts and floods—totaled more than $330
billion (Munich Re, 2018).

However, as highlighted in the Unbreakable report (Hallegatte et al., 2017),

the focus on asset losses fails to inform us on how disasters affect people’s
well-being. The report highlighted the fact that the overall well-being
impact of a disaster depends critically on two factors.

First, it depends on how asset losses affect income and consumption during
the recovery and reconstruction phase. For example, the same asset losses
have different impacts depending on whether reconstruction takes place over
a few months or several years, and on whether savings make it possible to
smooth the impact on consumption.

Second, the impact on well-being depends on who is affected. Clearly, a

one dollar loss is experienced differently by a rich person than by a poor
person. The same loss affects poor and marginalized people far more
because their livelihoods depend on fewer assets, their consumption is closer
to subsistence levels, they cannot rely on savings to smooth the impacts,
their health and education are at greater risk, and they may need more time
to recover and rebuild.

To account for this important difference between asset and well-being

losses, the Unbreakable report developed a new resilience metric to measure
how natural disasters affect people’s well-being. It accounts for the exposure
and vulnerability of people: how often they are affected, and how much they
lose when they are affected. But the framework also includes their socio-
economic resilience, defined as an ability to cope with a disaster, receive
support, and recover and reconstruct (Figure 1).

4 I B U I L D I N G B AC K B E T T E R
BEYOND ASSET LOSSES

Figure 1. People’s well-being losses due to a disaster depend on their resilience,

i.e. their ability to cope, receive support, and recover.

ASSET LOSSES
1. Hazard 2. Exposure 3. Vulnerability

WELL-BEING LOSSES
1. Hazard 2. Exposure 3. Vulnerability 4. Socioeconomic
resilience

By examining well-being instead of asset losses, the report provided a deeper (and
grimmer) view of natural disasters than does the usual reporting—and indeed, this
view takes better account of poor people’s vulnerability.

In doing so, the report also highlights new opportunities for interventions to minimize
disaster losses by boosting people’s resilience—for example, through appropriate
targeting of social expenditures, or improved access to financial instruments from
saving accounts and borrowing to insurance. While these measures do not reduce asset
losses—the typical metric used in disaster risk management—they can very efficiently
mitigate their impacts on people’s livelihood and well-being. This benefit can be
captured using well-being losses as an additional metric for disaster impacts.

Since the publication of the Unbreakable report, the model has been used in targeted
analysis of the vulnerability of transport systems in Small Island Development States
(World Bank, 2017a), and in country-level analysis in Fiji (Government of Fiji and
World Bank, 2017) and the Philippines, where the work is still under way.

Resilience to natural disasters is never constant; it depends on a wide range of

dynamic factors. These include changing exposure due to population growth
and rapid urbanization, intensifying hazard levels due to climate change, and
strengthened ability to cope and recover due to effective disaster risk management.

5 I B U I L D I N G B AC K B E T T E R
 Based on the resilience model presented in the Unbreakable report,
BEYOND ASSET LOSSES
this update report calculates the risk to well-being by considering the
four drivers of the loss in well-being as described above. These are:
BOX 01
QUANTIFYING • hazard (the probability an event occurs);

SOCIOECONOMIC • exposure (the population and assets located in the affected area);

RESILIENCE AND • asset vulnerability (the fraction of asset value lost when affected
WELL-BEING by a hazard), and;

LOSSES • socio-economic resilience, which is defined as the ratio of asset

losses to well-being losses:

asset losses
socioeconomic resilience =
well-being losses

Based on this definition, socio-economic resilience is a driver of the

risk to well-being, along with the three standard drivers:

Risk to well-being =
expected asset losses (hazard) * (exposure) * (asset vulnerability)
=
socioeconomic resilience socioeconomic resilience

The Unbreakable report used this approach to quantify the risk to

well-being in 117 countries, based on the latest available data in
2016, and estimates of asset losses from the United Nations Global
Assessment Report on Disaster Risk Reduction—the so-called GAR
(UNISDR, 2015). It calculated well-being losses for multiple hazards,
considering return periods from 2 to 1,500 years. These hazards
included river floods, coastal floods due to storm surge, windstorms,
earthquakes, and tsunamis. A detailed description of the model is
provided by Hallegatte, Bangalore, and Vogt-Schilb (2016), and the
model is publicly available.1

To understand how resilience to natural disasters is evolving in countries around the

world, these diverse factors must be tracked and evaluated continuously.

In this section, the computation of the resilience indicator from Unbreakable is

updated using the latest available data on socio-economic trends. More specifically,
updated data include:

• economic activity (GDP);

• urbanization;
• income inequality;

6 I B U I L D I N G B AC K B E T T E R
BEYOND ASSET LOSSES

• level and coverage of social protection schemes;

• contingent finance instruments (especially the World Bank’s Catastrophe
Deferred Drawdown Option (Cat-DDOs)); and
• sovereign credit ratings.

Credit ratings are not used here as a measure of the risk of defaults, but more generally
as a measure of the quality of public finance management in a country, and thus of its
ability to reallocate and mobilize financial resources in the case of a contingency.

This analysis also extends the calculation from 117 countries in the initial report to 149
countries today, covering 95.5 percent of the world’s population, and 94 percent of
global GDP.

Any indicator is only as good as the data it is based on, and the uneven pace at which
new data becomes available means that different variables are updated at different
times, in different countries. Measures of income inequality, for example, are based
on household surveys that are typically not conducted annually. While the updated
resilience indicator in this report is based on the latest available data, not all variables
in all countries will have been updated. Since data updates occur at different times for
different countries, any comparison across countries should be made with caution. The
resilience indicator and its subsequent updates primarily serve the purpose of tracking
the resilience of individual countries across time.

Socioeconomic resilience in 2017

In the 149 countries, and still based on the assessment from UNISDR (2015), the total
risk to assets reaches US$382 billion, the annual average cost of repairing and replacing
assets after natural disasters.2 Risk to assets, expressed as a share of GDP, disproportionately
affects low-income countries (Figure 2). This difference arises from better disaster
protection in high income countries—for example dikes and seawalls for floods—and the
higher quality of buildings and infrastructure. High-income countries tend to have risks
to assets of below 1 percent of GDP, while for low income countries risk to assets can
reach almost 10 percent of GDP for some small states.

Using the latest data available in 2018, the global socioeconomic resilience is
estimated at 69 percent. National-level estimates display a similar pattern to the
earlier estimates (Figure 3).3 Countries with higher average incomes tend to be
more resilient than lower income countries. At the same time, resilience levels in

7 I B U I L D I N G B AC K B E T T E R
BEYOND ASSET LOSSES

low-income countries vary substantially, indicating the importance of factors other

than income in determining resilience. Differences in socioeconomic resilience are
explained by differences in inequality, financial inclusion, housing quality across
income classes, access to financing, and social protection coverage and generosity.

Figure 2. Risk to assets as a share of GDP.

10

8
Risk to assets (% of GDP)

0
0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 100,000
GDP per capita (US$)

Figure 3. Socio-economic resilience to natural disasters.

100

90

80

70
Resilience (%)

60

50

40

30

20

10

0
0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 100,000
GDP per capita (US$)

The risk to well-being, measured as a share of GDP, is particularly high in low-

income countries (Figure 4). This high risk is mostly driven by higher risks to assets,

8 I B U I L D I N G B AC K B E T T E R
BEYOND ASSET LOSSES

but is magnified by a lower level of resilience. Globally, annual average well-being

losses are estimated at US$555 billion for the 149 countries included in this analysis.

Figure 4. Risk to well-being as a share of GDP.

25

20
Risk to well-being (% of GDP)

15

10

0
0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 100,000
GDP per capita (US$)

Small island developing states

For years, small island developing states have been at the forefront of the fight against
the devastating impacts of climate change. These countries experience tropical storms,
storm surges, floods, and landslides at a higher frequency than most countries—and
often when they are hit, a large share of their population and assets are affected. The
2017 hurricane season in the Caribbean is an illustration of this: several Caribbean
countries experienced loss of life, as well as severe damages to the majority of their
building stock and infrastructure. For some islands, asset losses alone were estimated at
over 100% of annual GDP.

Despite their high exposure to natural hazards, small island states are not regularly
featured in in-depth studies, such as the 2017 Unbreakable report—primarily due to a
lack of data. This follow-up report provides evidence for an additional 32 countries
(on top of the 117 in the original Unbreakable report), and many of these are small
island states from the Atlantic, Indian, and Pacific Oceans. The data on disaster asset
losses in small island states—as for all countries in this analysis—are based on the Global
Assessment Report 2015 (GAR15).

9 I B U I L D I N G B AC K B E T T E R
 BEYOND ASSET LOSSES

Figure 5. Risk to assets in small island states (orange).

10
Antigua &
Barbuda
9

7 Vanuatu
Risk to assets (% of GDP)

5 Haiti

4 Tonga
Madagascar
Solomon Islands
3
Fiji
Dominica
2 St. Vincent & the Grenadines
Jamaica St. Lucia
Dominican Republic
1 Guyana
Trinidad & Tobago
Comoros Mauritius
Kiribati Maldives
0
0 10,000 20,000 30,000 40,000 50,000
GDP per capita (US$)
Other countries are indicated in grey.

The estimates of socio-economic resilience, risk to assets, and risk to well-being confirm the
story highlighted above: high exposure to natural hazards, high event frequency, and the
concentration of assets means that many small islands face far higher risk to assets and well-
being (as a share of GDP) than other countries (Figures 5 and 7). In addition, among small
island states, a similar pattern emerges as for all countries: poorer island states, i.e. those with
lower per capita income, tend to face higher risks to well-being, as they are less able to cope
with and recover from natural shocks (Figure 6).

However, the estimates also show that not all small islands have high levels of risk to assets
and well-being. For instance, Kiribati and the Maldives are spared from one of the most
devastating natural hazards—tropical cyclones—and thus experience lower levels of risk
than many other islands. This analysis is however limited to five hazard types (floods, wind,
storm surge, earthquakes, tsunami), which means that other hazards such as droughts are
omitted, and losses are likely to be underestimated. Also, such analyses based on global
databases and models are limited, especially when considering small geographic areas such
as small islands. This means that the estimates offered in this analysis are indicative, and
cannot replace a detailed country level hazard assessment.

10 I B U I L D I N G B AC K B E T T E R
BEYOND ASSET LOSSES

Figure 6. Socio-economic resilience to natural disasters in small island states (green).

100

Madagascar
90 Kiribati
Solomon Islands
80 Mauritius
Vanuatu
70 Dominican Republic
Tonga Fiji
Maldives
60
Resilience (%)

Jamaica St. Vincent and

Comoros the Grenadines
50 Guyana

St. Lucia
40 Antigua &
Haiti Barbuda Trinidad &
Tobago
30 Dominica

20

10

0
0 10,000 20,000 30,000 40,000 50,000
GDP per capita (US$)

Figure 7. Risk to well-being in small island states (blue).

25
Antigua &
Barbuda

20
Risk to well-being (% of GDP)

15

Haiti

10 Vanuatu

Solomon
Islands Dominica
Tonga
5 Madagascar
Fiji
St. Lucia
Jamaica
St. Vincent & the Grenadines
Guyana Dominican Republic Trinidad & Tobago
Comoros Mauritius
Kiribati Maldives
0
0 10,000 20,000 30,000 40,000 50,000
GDP per capita (US$)

11 I B U I L D I N G B AC K B E T T E R
 RESILIENT
BEYOND RECOVERY
ASSET LOSSES

AND BUILDING BACK BETTER

T he long and difficult recovery process begins at the

moment a disaster strikes. This process is extremely
challenging, combining the usual issues of infrastructure and
building construction with the urgency and confusion of the
post-disaster context.

The recovery process is usually structured by three main stages: (i)

humanitarian relief, including search and rescue, and medical care; (ii)
restoration of basic services, including the supply of clean water, food,
and sanitation, basic energy, mobility, and health care needs; and (iii) the
reconstruction phase, including infrastructure reconstruction, the repair or
replacement of building and production equipment, and asset recovery by
households—typically the longest and most costly phase of recovery (Figure 8).

The concept of building back better is aimed at improving the three phases
of this recovery process, to ensure that the recovery contributes to a more
resilient society. It is defined by the United Nations Office for Disaster
Risk Reduction (UNISDR) as “the use of the recovery, rehabilitation and
reconstruction phases after a disaster to increase the resilience of nations
and communities through integrating disaster risk reduction measures into
the restoration of physical infrastructure and societal systems, and into the
revitalization of livelihoods, economies, and the environment.”4

The recovery and reconstruction phase after a disaster offers incomparable

opportunities to rebuild in a way that prevents the same hazards from leading
to the same impacts, through the improvement of land-use planning (e.g.,
deciding not to reconstruct in a highly-vulnerable area), the application
of construction norms (e.g., ensuring that rebuilt buildings can resist the
next earthquake better), or the deployment of prevention and preparedness
options (e.g., designing a neighborhood to facilitate evacuation) (UNISDR,
2017). These opportunities are present mainly in the last phase of the
recovery, when assets and infrastructure are repaired or rebuilt.

But a better recovery can do more than reduce the impact of future disasters: it
can also reduce the impact of the disaster that caused the damages in the first place:

12 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

Figure 8. An illustration of the post-disaster recovery and its three phases.

RESILIENT
RECOVERY
Reconstruction
and asset recovery
(BBB)
Restoration of
basic services

Humanitarian
relief

TIME
Disaster
The vertical axis refers to a range of aspects, from the stock of assets, to the income, consumption and well-being of the
affected population. Building back better means that the recovery process is stronger compared to pre-disaster levels, but also
faster and more inclusive.

13 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

• A faster recovery and reconstruction process can restore the income and assets
of the affected population earlier, reducing the cumulative income losses and
making it easier to rely on savings to maintain consumption levels. As a result, a
faster recovery minimizes the disaster's impact on economic growth and poverty
reduction (Hallegatte and Vogt-Schilb, 2017).

• Also, people in poverty and members of disadvantaged groups are the most likely
to experience long-term consequences—or even a poverty trap—as the result of
a disaster: such effects can result from health impacts, especially for children who
may be undernourished or forgo health care in a disaster aftermath, or simply
from the inability of poor people to use savings or borrow to repair or replace
their lost assets (Baez and Santos, 2007; Carter et al., 2007). A more inclusive
recovery—one that supports the most vulnerable populations to ensure they can
rebuild and do not suffer from long-term consequences—reduces the overall
impact of a disaster.

In this study, building back better means that the repaired or replaced assets are more
resilient, but also that the recovery process is shorter and more efficient, and that the
entire recovery process does not leave anyone behind—i.e. that even the poorest and
most vulnerable receive the support they need to fully recover.

Better recovery and reconstruction cannot replace risk reduction and prevention,
which remain the main instruments for the reduction of losses of assets and thus well-
being. However, this study explores the contribution that preparedness and building
back better can provide, as a complement to the other components of the disaster risk
management toolkit (see the Unbreakable report for a full review of this toolkit).

To assess the importance of building back better, we use policy simulations to

explore how policies that favor a better recovery and reconstruction after disasters
can reduce the current and future impact of disasters on people’s well-being. We use
the model described in the previous section to compare the current level of risk (to
assets and well-being) with the level of risk that would be attained under four policy
scenarios, corresponding to (i) building back stronger; (ii) building back faster; (iii)
building back more inclusively; and (iv) combining the three policies to build back
stronger, faster, and more inclusively.

14 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

Building back stronger

Reconstruction phases provide rare opportunities to reduce the vulnerability of

affected regions and countries (UNISDR, 2017). This could be achieved through
risk-informed construction standards and smart spatial planning. For example, the
large-scale physical destruction experienced in 2017 by several Caribbean island
states, including Dominica and Antigua & Barbuda, offers the opportunity to ensure
that destroyed assets are reconstructed to more resilient standards—i.e. that they can
withstand more intense events in the future.

In practice, the foundation for building back stronger is best laid before a disaster:
Strengthening the institutional and technical capacities of public and private sectors is
crucial for ensuring that there is sufficient design, construction, and quality assurance
capacity in a post-disaster situation. This should extend from individual builders
and carpenters, to contractors, to building officials at all levels of government. The
spotlight at the end of this section provides case studies to illustrate these measures.

Estimating well-being benefits of building back stronger

To estimate the benefits of building back stronger, we assume that destroyed assets
are reconstructed to a resilience standard that is able to withstand shocks of up to a
50-year return period.5 Building back stronger reduces the risk to assets, and thus the
risk to well-being.

The estimates suggest that if all countries were to build back stronger during a 20-
year window, then global asset losses due to natural disasters would be reduced by
11.2 percent from US$382 billion to US$339 billion annually. The benefits for well-
being are even greater: well-being losses due to natural disasters would be reduced by
11.7 percent from US$555 billion to US$490 billion annually.

The issue of building back better in island countries has received particular attention
in the wake of the 2017 hurricane season in the Caribbean. Figure 9 (left) provides
an overview of the countries where stronger recovery would lead to the greatest
reduction in average well-being losses. The estimates show that stronger recovery
could reduce overall well-being losses due to natural disasters by more than 40
percent in ten countries: Antigua & Barbuda, Dominica, Vanuatu, Myanmar, Laos,
Tonga, Guatemala, Trinidad & Tobago, Peru, and Fiji. Unsurprisingly several small
island states are among those with the highest potential to benefit from stronger

15 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

recovery. Figure 9 (right) provides estimates for ten selected small island states. The
well-being losses in these small island states would, on average, be reduced by 37.3
percent through stronger reconstruction (compared to 11.7 percent globally).

Figure 9. Left: Top 10 largest reductions in average well-being losses due to

building back stronger. Right: Reductions in average well-being losses for
selected small island states.

Antigua & Antigua &

Barbuda -62% -62%
Barbuda

Dominica -50% Dominica -50%

Vanuatu -48% Vanuatu -48%

Myanmar -46% Tonga -44%

Lao PDR -45% Trinidad & Tobago -42%

Tonga -44% Fiji -40%

Guatemala -42% St. Lucia -30%

Trinidad & -42% St. Vincent & -23%

Tobago the Grenadines

Peru -42% Solomon Islands -22%

Fiji -40% Jamaica -12%

-80 -70 -60 -50 -40 -30 -20 -10 0 -80 -70 -60 -50 -40 -30 -20 -10 0

16 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

Figure 10. Percentage reduction of well-being losses associated with stronger

recovery.

Countries are sorted into quartiles (i.e. same number of countries for each color).

These estimates show that post-disaster reconstruction offers an opportunity for

implementing resilience standards and reducing losses from future events. However,
this also implies that the same argument applies to all new infrastructure construction,
regardless of whether or not a disaster has occurred recently. Many developing
countries are in the process of rapidly developing their infrastructure in response to
population growth, urbanization, and economic growth. In order to avoid increasing
exposure and vulnerability, all of these infrastructure investments need to take natural
hazards and risks into account.

Moreover, when determining resilience standards, it is important to recognize that

climate change will mean that past disaster risk probabilities (and return periods) do
not offer robust guidance for the long-term future (Hallegatte, 2009). As extreme
weather events become more frequent, for example, a Category 5 hurricane that has
a 50-year return period today may have only a 20-year return period in the course of
the lifetime of an infrastructure investment. Such uncertainty must be accounted for
in the design and construction or reconstruction of infrastructure. Decision-making
under uncertainty (DMU) is one approach that can help to eliminate strategies with
catastrophic outcomes, and prioritize strategies that perform robustly under a wide
range of future climate change scenarios (Kalra et al., 2014).

17 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

SPOTLIGHT

Resilient housing and infrastructure

reconstruction in Nepal, China, Fiji, and Dominica
Major earthquakes rocked Nepal in April and May 2015, with devastating
consequences: the Government of Nepal reported the death toll at
approximately 8,700 and those injured at 25,000. In an early post-disaster
needs assessment (PDNA), total recovery needs were estimated at US$6.7
billion, or about a third of Nepal’s economy. The single largest need identified
in the PDNA (US$3.27 billion) was for “housing and human settlements”: 490,000
houses were destroyed and another 265,000 damaged to an extent that they
were at least temporarily uninhabitable.

The Government of Nepal, in partnership with a number of development

partners6, launched the Rural Housing Reconstruction Program (World Bank,
2016b). Building back stronger is at the heart of this housing reconstruction
program. It aims to ensure that houses destroyed in the most affected districts
of the country will be rebuilt using earthquake-safer building techniques,
through training, grants, and technical support to eligible households.
Specifically, the program will provide training to local artisans, facilitate the
building of material markets, and disseminate information on earthquake-
resilient construction techniques, all in coordination with partner organizations,
to assist the reconstruction process.

While consistently ensuring resilient building standards, the program does

not restrict the individual choices of households: beneficiary households may
choose to rebuild houses themselves and/or hire labor, such as masons and
carpenters, according to their needs. There is also no restriction in the use of
materials as long as it complies with the earthquake-resilient construction
techniques defined by the program. Overall, 55,000 households benefit directly
from reconstruction grants, while another 490,000 households receive technical
assistance. Moreover, the Government had several pre-approved reconstruction
designs ready before the earthquake struck. This meant that reconstruction was
not only stronger, but was also accelerated through foresight and preparedness.

Earthquake damage in Nepal.

18 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY
China, too, emerged stronger from a severe disaster. In 2008, an 8.01
magnitude earthquake struck southwestern China; with over 69,000 fatalities,
374,000 people injured, and about 18,000 missing, it was one of the deadliest
earthquakes in recent history (World Bank, 2018). In addition to the human toll,
the disaster destroyed or severely damaged 34,000 km of highways, thousands
of schools, hospitals, and wastewater systems, as well as more than 4 million
homes. In response to this disaster, the government of China adopted a building
back stronger approach.

It ensured that the reconstruction of affected infrastructure followed higher

seismic standards and flood risk management codes, while ensuring a balance
between reconstruction activities and laying a foundation for the longer-term
sustainable economic recovery and development of the affected areas. In fact,
the restored infrastructure was not only built to be more resilient to natural
hazards than before the disaster—it also greatly enhanced the service quality
and access to essential public services, including water, sanitation, roads,
health and education. For example, 300 roads were rebuilt or renovated to
new seismic standards, and upgraded through the addition of modern traffic
management and drainage systems.

A similar story comes from Fiji: Cyclone Winston in February 2016 was the
most intense cyclone to be recorded in the Southern Hemisphere—and to
make landfall in Fiji. 44 people lost their lives, and about 30,000 houses were
destroyed. In the aftermath of the storm, a massive rebuilding and recovery
effort got underway, led by the Government of Fiji and supported by numerous
relief and development organizations. As part of this effort, the government
placed strong emphasis on the need to build back stronger, working with
humanitarian actors to train hundreds of community carpenters and workers in
order to mainstream resilient building techniques to ensure that houses—and
people—could better withstand future cyclones. To this end, the Government
of Fiji operates the dedicated program “Help for Homes”, which offers grants,
materials, and technical training to assist people in the construction and
reconstruction of safer and more resilient homes.

In the Caribbean, the hurricane season of 2017 brought two category 5

hurricanes and, in addition to the loss of lives, caused severe damages to
the building stock of several island states. In Dominica, 70 to 80 percent of all
houses and buildings sustained major storm damages, ranging from ripped-off
roofs to total destruction. Supported by the World Bank, the Government of
Dominica has launched a housing reconstruction project that aims to follow the
principle of build back stronger. By rebuilding houses that can withstand future
storms of the same intensity as the 2017 hurricanes, Dominica’s reconstruction
efforts are helping to protect people and their livelihoods from future shocks.

19 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

Building back faster

People’s ability to generate income is fully dependent on the stock of assets, including
private assets, public infrastructure, and human and social capital. When these assets
have been lost or damaged, incomes are reduced, and with them consumption and
well-being, as well as the ability to save and invest in the future. And of course, the
longer it takes to recover these assets, the greater the impact on people’s well-being
and prospects (Hallegatte and Vogt-Schilb, 2017).

It is a different story to experience a reduction in income for a few months than to have
to cope with many years of depressed income. This is especially true because people
tend to rely on savings to smooth consumption in the months following a disaster,
but people’s savings rarely represent more than a few months of normal expenditures,
making it impossible to maintain consumption if reconstruction takes several years.
The speed at which assets and public infrastructure are reconstructed after a disaster is
thus a major factor determining the overall impact on people’s well-being.

In contrast with building back stronger, building back faster does not reduce future
asset losses. But it does reduce well-being losses immediately, by making it easier for
affected people to cope with the shock. It is therefore an important component of a
good recovery and reconstruction.

In countries affected by frequent events—such as many small island countries—the

length of the reconstruction period has another dimension: since these countries are
particularly vulnerable during post-disaster recovery periods—for example because
public services and infrastructure may still not be fully functional when the next
disaster hits—building back faster can make the difference between resilience and
stagnation (Hallegatte et al., 2007).

In practice, recovery speed is highly dependent on the level of preparedness and

resources available for a quick and resilient recovery. Measures that can ensure
rapid recovery and reconstruction include: (i) contingency plans to ensure that
the coordination of the recovery and reconstruction efforts is effective and that
responsibilities are clearly allocated among government agencies; (ii) contingent
financial arrangements—such as contingent credit lines or insurance products—to
ensure that financing is immediately available and is not delayed by budgetary
procedures; (iii) pre-arranged contracts to accelerate procurement, for example
ensuring that debris removal can start as soon as possible to facilitate reconstruction;

20 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

and (iv) international cooperation to share costs of staff and equipment needed for the
recovery and reconstruction, including the use of innovative technologies.

The main challenge to ensuring rapid recovery and reconstruction is that the process
depends on many factors that need to progress in parallel. The overall speed can be
constrained by a delay in one dimension, such as the procurement or the approval of
building permits, that can create a bottleneck regardless of the progress achieved in
other dimensions.

To illustrate this challenge, it is useful to consider one of the major bottlenecks to a

rapid reconstruction process: the difficulty of removing debris from reconstruction
sites. Particular challenges in this process can include the contracting of private
construction firms and the identification of suitable equipment. As Jha et al. (2010)
emphasize, various measures can significantly speed up debris removal—provided
that they have been prepared before the disaster: (i) pre-identifying public resources
that are available to assist with debris collection and management; (ii) pre-identifying
contractors that own heavy equipment needed for debris removal and collection
(including bulldozers, dump trucks) and that can provide skilled operators to run
the equipment; (iii) pre-qualifying firms, pre-arranging contracts, and pre-defining
contract scope, terms and prices; and (iv) analyzing the financial resources available
for debris management and developing a financial plan, which may include taxes,
user fees, donations, and resources from a higher level of government. With these
preparatory measures in place, governments can execute a fast and well-coordinated
debris management plan in the immediate aftermath of a disaster, and thus pave the
way for faster reconstruction.

Estimating the well-being benefits of building back faster

The potential benefits from these measures are substantial: faster recovery can
significantly reduce average well-being losses due to natural disasters. For a
reconstruction time of 3 years—which is the baseline considered in this study—losses
are US$555 billion. If the average recovery speed is reduced to one year, global well-
being losses could be reduced to US$480 billion—a 13.5 percent reduction compared
to the basis of 3 years (Figure 11)—or a 23.3 percent reduction compared with a
5-year reconstruction duration. Figure 12 shows the reduction in average well-being
losses due to faster recovery for the ten countries with the largest absolute losses.

21 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

Figure 11. Global well-being losses associated with different recovery durations.
800 billion

700 billion
Global well-being loss (US$)

600 billion

500 billion

400 billion

300 billion

200 billion

100 billion

0
5 years 4 years 3 years 2 years 1 year

Figure 12. Reduction in average well-being loss due to faster recovery for the
top ten countries in terms of losses.
120 billion

100 billion 19%

Average well-being loss (US$)

80 billion 24%
China

60 billion
22%
16% 32%
United States

40 billion

24% 22%
Philippines

25% 25% 17%

Japan

20 billion
India

Iran

Bangladesh

Peru

Russia

Italy

5 years
4 years
3 years
2 years
1 year

Note: The top point refers to the well-being loss associated with a five-year reconstruction period, while the bottom point
refers to a one-year reconstruction period. Percentages indicate the reduction in well-being losses by speeding up recovery
from five years to one.

22 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

As suggested above, the importance of ‘building back faster’ differs across countries.
Countries where indirect losses are the largest (i.e. countries with low socioeconomic
resilience) are bound to benefit most from accelerated recovery. Figure 13 presents
the ten countries with the largest relative reduction in well-being losses due to faster
recovery, which include in particular small island states and low-income African
countries. Figure 14 provides a global overview illustrating that the countries that
benefit most from faster reconstruction tend to be lower-income developing countries.

Figure 13. Ten largest reductions in average well-being losses due to faster recovery
(reduced from five years to one).

Dominica -54% St. Lucia -45%

Trinidad & -50% Chad

Tobago -42%

Sudan -47% Rwanda -41%

Nigeria -47% Guatemala -41%

Belize -46% Myanmar -40%

-60% -50% -40% -30% -20% -10% 0 -60% -50% -40% -30% -20% -10% 0

Figure 14. Percentage reduction of well-being losses associated with faster

recovery (reduced from five years to one).

Countries are sorted into quartiles (i.e. same number of countries for each color).

23 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

Figure 15 presents the reduction in average well-being losses due to faster recovery
for ten selected small island developing states. Unsurprisingly, small islands are among
the countries where the benefit of rebuilding faster is the greatest, with a reduction
of 34 percent on average, compared to 23 percent globally. In Trinidad and Tobago,
measures to accelerate recovery could even halve the annual well-being losses.

Figure 15. Reduction in average well-being loss due to faster recovery for ten
selected small island developing states.
$1.2B
50%

$1B
Average well-being loss (US$)

$800M
28%
30%
$600M
Trinidad & Tobago

$400M 26%
Jamaica

Barbuda
Antigua &

$200M
Fiji

22% 23% 45% 54%

26% 37%
$0
Vanuatu

Solomon
Islands

St. Lucia

the Grenadines
Dominica

St. Vincent &

Tonga

5 years
4 years
3 years
2 years
1 year

The top point refers to the well-being loss associated with a five-year reconstruction period, while the bottom point refers to
a one-year reconstruction period. Percentages indicate the reduction in well-being losses by speeding up recovery from five
years to one.

These results underestimate the value of building back faster, as they do not account
for the higher vulnerability of assets that have been damaged and cannot be fully
repaired before the next event. For example, houses with damaged roofs or power
systems operating thanks to “quick fixes” may be revealed as unable to cope with
another hurricane, leading to compounding losses.

24 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

SPOTLIGHT

Streamlined processes for rapid reconstruction

and recovery in Indonesia, Turkey, and Colombia
In December 2004 an earthquake and subsequent tsunami devastated much of
the coast of Aceh, Indonesia, and other coastal areas around the Indian Ocean. In
total, 286,000 people in 14 countries lost their lives, including 221,000 killed or missing
in Aceh alone. Nearly US$7 billion in contributions flowed in from the Indonesian
government and international donors, fueling a boom in reconstruction activity and a
successful reconstruction (World Bank, 2012). Nearly ten percent of these funds were
contributed through the Multi Donor Fund for Aceh and Nias (MDF).

This experience helped to establish strong policies and institutions, including the
newly-formed National Board for Disaster Management (BNBP) and the Indonesia
Disaster Fund (IDF) which is largely modeled on the Aceh MDF. These institutions have
helped to significantly streamline the post-disaster processes for rapid response and
recovery in the country.

Several key principles are now integral to Indonesia’s disaster risk management and
response strategy, which is based on the maxim of building back faster. For instance,
the government follows a phased approach that prioritizes the rapid rebuilding of
homes and basic infrastructure (e.g. sanitation), then progresses to infrastructure,
and finally to economic development. It has also established streamlined budgetary
processes to ensure that funds can be disbursed and transmitted quickly and
efficiently to where they are most needed. Moreover, the government emphasizes
community-based development for the rebuilding of homes and local infrastructure,
in order to mobilize local capacity and commitment to rapid community recovery. In
addition, cross-cutting elements are integrated into all recovery projects, including
disaster risk reduction, capacity building, gender inclusiveness, environmental
protection, and capacity development, in order to enhance not only the speed, but
also the quality of reconstruction.

In the years following the 2004 disaster, Indonesia was struck by earthquakes,
tsunamis, and volcanic eruptions (World Bank, 2012). Although these events led to

Flooding in Colombia.

25 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY
loss of life and destruction of property, the lessons learned in Aceh had equipped the
government with significant expertise in disaster management, and strengthened its
preparedness. The newly formed institutions and strategies enabled more effective
and faster recovery and reconstruction.

With the same objective of strengthening preparedness and accelerating

recovery processes, the Government of Turkey started implementing a seismic risk
management and emergency preparedness project in 2005, guided in part by
the lessons of the Marmara earthquake, which claimed over 17 000 lives in 1999.
The Government focused on strengthening the effectiveness and capacity of the
provincial and municipal public safety organizations in Istanbul (World Bank, 2016a).
The package of measures ranged from the improvement of public awareness
and institutional capacity, modernization of emergency communications systems,
establishment of an emergency management information system, and expansion of
existing emergency response capacity.

Following the principle of building back faster, the Government, at the central level,
prepared a National Response Plan and assigned responsibility to the provincial
directorates to prepare response plans at the local level. These response plans
define all service groups needed for an efficient response, including the private
sector. Information management and decision-support systems and databases
established for the plan also include a comprehensive inventory of construction firms,
detailing their capacities and equipment in major cities. The overall objective of this
emergency management system was to improve the ability of authorities at all levels
to mobilize resources rapidly and effectively, and to jump-start debris removal and
reconstruction in a pre-arranged and well-coordinated manner.

In addition, the preparation of contingent financing instruments can also help

governments to strengthen their ability to respond to shocks quickly and effectively.
Contingent finance instruments like the World Bank’s Catastrophe Deferred
Drawdown Option (Cat-DDO) provide countries with a pre-approved credit line that
can be accessed immediately after the declaration of an emergency following a
natural disaster. While the amounts provided by a Cat-DDO are usually well below the
full cost of reconstruction, the rapid mobilization of pre-approved funds is designed
to accelerate effective recovery measures and ensure that reconstruction starts as
quickly as possible. In addition to providing immediate liquidity following a disaster,
this instrument also supports policy programs to strengthen the preparedness and
response capacity of governments and risk management institutions.

In 2010, Colombia experienced one of the worst rainy seasons in decades, resulting
in severe flooding and landslides. However, before this disaster the Government
of Colombia and the World Bank had prepared a contingent credit (a Cat-DDO)
worth US$150 million. The disaster triggered the disbursement of the contingency
funds (World Bank 2010). This helped the government to signal its resilience and
preparedness to financial markets, and contributed to expediting the post-disaster
recovery. Several other countries—including low-income countries with high exposure
to natural hazards and more limited access to international credit markets—are
exploring or preparing such contingent finance instruments; they include Saint Lucia,
Cabo Verde, Kenya, Malawi, and the Maldives.

26 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY
The Government of Mexico created the Natural Disaster Fund (FONDEN) in
1996 in response to delays faced in the post-disaster financing of emergency
and recovery activities (GFDRR 2012). FONDEN is a financial mechanism that
provides federal agencies and the Mexican states with immediate liquidity to
finance recovery efforts. It has a mandate to (1) finance post-disaster emergency
assistance (through a revolving fund), and (2) provide the 32 Mexican states
and the line ministries (for example, the Ministry of Infrastructure, Ministry
of Health, Ministry of Education, and Ministry of Human Development) with
financial resources in case losses from natural disasters exceed their budget
capacity. FONDEN finances the post-disaster recovery and reconstruction
of public assets (100 percent of federal assets and 50 percent of state and
municipal assets) and low-income houses. In 1999 the FONDEN Trust Fund was
established to help finance the FONDEN program through a catastrophe reserve
fund that accumulates the unspent disaster budget of each year. Overall,
FONDEN is credited with playing a critical role in strengthening Mexico’s disaster
preparedness and ability to build back faster and better­—and can provide
valuable lessons for other countries.

Building back more inclusively

As the Unbreakable report has shown, poor people are not only more vulnerable
to natural hazards, they also tend to have access to fewer post-disaster support
mechanisms, such as insurance, borrowing, or remittances, and have fewer savings to
draw on. As a consequence, they tend to experience higher losses relative to income,
and often have to resort to “negative coping mechanisms” such as reducing food
intake, cutting down on health care, or reducing education spending.

For instance, in response to weather shocks in Sub-Saharan Africa, asset-poor

households provide children with lower-quality nutrition (Alderman, Hoddinott,
and Kinsey, 2006; Dercon and Porter, 2014; Yamano, Alderman, and Christiaensen,
2005), and they are less likely to take sick children for medical consultations (Jensen,
2000). These behaviors have short- and long-term impacts, particularly for children
younger than two. Within this group, in households reducing nutrition, children’s
average height fell by 0.9 centimeters within six months of a disaster (Yamano,
Alderman, and Christiaensen, 2005), and the stature of children in these households
was permanently reduced by 2–3 centimeters (Dercon and Porter, 2014).

Impacts on education are also prevalent. In Africa, enrollment rates have declined 20

27 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

percent in regions affected by drought (Jensen, 2000). Such findings are not restricted
to Africa; similar post-disaster impacts on health and education have been found in
Asia, Latin America, and elsewhere (Baez, de la Fuente, and Santos, 2010; Maccini
and Yang, 2009). In Mexico, once children have been taken out of school, even for a
temporary shock such as a flood, they are 30 percent less likely to proceed with their
education, compared with children who remain in school (de Janvry et al., 2006).

A single disaster can irreversibly affect a child’s life. After Hurricane Mitch hit Nicaragua
in 1998, the probability of child undernourishment in regions affected by the hurricane
increased by 8.7 percent, and child labor force participation increased by 5.6 percent
(Baez and Santos, 2007). In Guatemala, Storm Stan increased the probability of child
labor by 7.3 percent in departments hit by the storm (Bustelo, 2011).

Even when impacts on human capital can be avoided, major losses can have long-
term consequences for some households—such as when people become locked
in poverty traps because their asset base has fallen below a critical threshold. For
example, there is little chance that herders with only few animals left after a natural
disaster will be able to quickly regrow their herd (Carter and Barrett, 2006).
Marginalized and disadvantaged groups—including women—are especially at risk of
experiencing such poverty traps, not least because they tend to have fewer support
systems available to them.

The long-term cost of these negative coping mechanisms and poverty traps can be
very high, and this is why adaptive social protection and social safety nets that can
respond quickly to natural disasters have substantial benefit-cost ratios. Case studies
suggest that the cost of a drought to households can increase from zero to about
US$50 per household if support is delayed by four months, and to about US$1,300 if
support is delayed by six to nine months (Clarke and Hill, 2013).

However, providing rapid support to poor people cannot be easily improvised when
a disaster occurs. Instead, it requires the development—before the crisis—of social
safety nets that are adaptive and can react to shock, and of appropriate delivery
mechanisms that cover vulnerable populations.

It requires, for example, large and flexible social registries that include both potential
and existing beneficiaries, and contingency plans for when and how support will be
provided, and where the financial resources will come from.7 It also requires efficient

28 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

delivery mechanisms that can cover the poorest and most vulnerable people. For
instance, by providing more convenient and affordable financial services, mobile
money accounts offer promise for reaching unbanked adults traditionally excluded
from the formal financial system such as women, poor people, young people, and
those living in rural areas (Demirgüç-Kunt et al., 2015).

One example of a system enabling a more inclusive recovery is Kenya’s Hunger

Safety Net Program (HSNP), which is a safety net that is active every year but can
be scaled up in case of adverse weather conditions. In 2015, it delivered support to
more than 100,000 additional households in response to the drought. Transfers could
proceed two weeks after the decision to scale up was made, but this was possible
only through the use of satellite data and clear pre-determined thresholds to trigger
the scale-up, the pre-registration of all households in the covered counties, and the
provision of bank accounts to all potential beneficiaries. Only preparedness allowed
such a rapid and inclusive response.

Estimating well-being benefits of building back more inclusively

In this context, building back more inclusively means that post-support and recovery
and reconstruction measures can cover the entire population, including the poorest,
and that the amounts of transfers do not favor richer households as is sometimes
observed (Noy and Patel, 2014).

For the sake of illustration, the estimates in this section assume that all countries set up
a dedicated system that ensures that all affected households have access to post-disaster
support, regardless of their pre-disaster income level or social marginalization, and
receive the same amount, regardless of their wealth and losses. The amount of resources
available to all countries is assumed sufficient for governments to cover 80 percent of
the losses of the poorest in each country. The estimates suggest that if all 149 countries
could implement these measures, global well-being losses due to natural disasters could
be reduced by 9.4 percent from US$555 billion to US$502 billion annually.

Figure 16 provides an overview of the countries in which a dedicated system for a

more inclusive recovery would lead to the largest reduction in average well-being
losses. The estimates show that inclusive recovery could reduce overall well-being
losses due to natural disasters by up to 41 percent in the case of Angola. Such
dedicated systems tend to be particularly effective in lower-income developing

29 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

countries, countries with high inequality and limited coverage and generosity of
social protection, and countries with low sovereign credit ratings (which affects their
ability to finance post-disaster support measures in the absence of a dedicated system).
Figure 18 shows that building back more inclusively tends to yield higher reductions
in well-being losses in countries with high pre-existing inequality.

However, in high-income countries, a dedicated system shows little benefit—mainly

because social protection systems are already relatively well-equipped to target and
support even the lowest income groups. Deryugina (2016) shows for instance that
in the US, a large fraction of post-disaster support is provided by social protection
instruments that are not designed primarily as post-disaster systems—such as
unemployment and health insurance systems.

Figure 16. Left: Top 10 largest reductions in average well-being losses due to
more inclusive recovery. Right: Reductions in average well-being losses for
selected small island states.

Antigua &
Angola -41% -25%
Barbuda

Benin -35% Jamaica -14%

Comoros -30% Solomon Islands -13%

Congo, Rep. -29% Vanuatu -11%

Central African -29% St. Lucia -10%

Republic

Congo, -29% Dominica -8%

Dem. Rep.

Russian -28% Tonga -7%

Federation

Gabon -28% Trinidad & Tobago -7%

Haiti -27% Fiji -5%

St. Vincent &

Lesotho -27% the Grenadines -1%

-50 -40 -30 -20 -10 0 -50 -40 -30 -20 -10 0

30 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

Figure 17. Percentage reduction of well-being losses associated with more

inclusive recovery.

Countries are sorted into quartiles (i.e. same number of countries for each color).

Figure 18a. The benefits of more inclusive recovery increase with the level of
pre-existing income inequality.
14%
Share of national income earned by poorest 20%

12%

10%

8%

6%

4%

2%

0%
-40% -35% -30% -25% -20% -15% -10% -5% 0%

Change in average well-being losses due to dedicated system for more inclusive recovery

31 I B U I L D I N G B AC K B E T T E R
 SPOTLIGHT
RESILIENT RECOVERY

Adaptive social safety nets to ensure

inclusive recovery in Fiji
Fiji has a long history of providing social assistance to vulnerable populations
which are struggling to meet their basic needs—including in the aftermath of major
natural shocks. The country’s social protection system has evolved over the years
to be gender-sensitive and rapidly responsive. The core social protection programs
are the Poverty Benefit Scheme (PBS), Care and Protection Allowance (CPA), and
Social Pension Scheme (SPS). PBS, rolled out in 2013, targets the poorest 10 percent
of households in Fiji. In addition, the SPS was introduced to provide social pensions
to elderly people who fell outside the coverage of social assistance programs.
Since 2009, the government has significantly increased investment in its social
protection portfolio: funding increased from US$40 million in 2009 to US$66 million
in 2015, in recognition of the importance of social assistance programs.

In 2016, the category 5 tropical cyclone Winston struck Fiji, leaving behind severe
destruction: 44 lives were lost, and the assets losses alone were estimated to
be around US$1.4 billion. The strength of the existing social protection system
allowed the government to act swiftly and efficiently to provide support to the
affected population. The government scaled up its three main social assistance
programs to provide existing beneficiaries with top-up payments equivalent to
three months’ worth of their regular benefit amounts, and a total of US$5 million
was disbursed within one month of the disaster. Under the PBS, 22,802 households
were paid a lump sum of US$300. Some 17,782 pensioners in the SPS over the age
of 68 received an additional US$150. Finally, 3,313 families under the CPA received
a total of US$150. The cash top-up payments were intended to help people
meet immediate expenses following Winston and were provided to all existing
beneficiaries, irrespective of whether they resided in the affected areas.

A comprehensive impact evaluation of the government’s response to Winston

shows adaptive safety nets were effective in helping households cope (Mansur
et al., 2018). The evaluation shows that three months after the cyclone took
place, beneficiaries under the PBS (who belong to the poorest 10 percent of
the population) were more likely to have recovered from the shocks they faced
than comparable households that did not receive the additional assistance.
This includes having recovered from sickness or injury, repaired their dwelling,
replenished their food stocks, remedied the damage to their agricultural land,
repaired village or neighborhood infrastructure, and resolved problems of conflict,
violence, or insecurity. According to the analysis provided in Government of Fiji
and World Bank (2017), this intervention has a benefit-cost ratio higher than 4.

Through a strong focus on building back more inclusively, Fiji has ensured that the
poorest and most vulnerable population groups received the targeted support
they needed. In this way, the government was able to lessen the disproportionately
adverse impacts on poor people and ensure that recovery efforts do not
systematically overlook a significant fraction of the population.

32 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

Figure 18b. High inequality (i.e. low share of national income earned by poorest
20 percent) tends to reduce countries’ level of socio-economic resilience.
14%
Share of national income earned by poorest 20%

12%

10%

8%

6%

4%

2%

0%
0% 20% 40% 60% 80% 100%

Socio-economic resilience (%)

Building back better: stronger, faster, and more

inclusively

Taken separately, there are large benefits in making the post-disaster recovery stronger,
faster, or more inclusive. But of course, the benefits are maximized if these three
objectives can be achieved together.

Doing so creates specific challenges. For instance, while accelerating reconstruction

is crucial to minimize the well-being impacts of disasters, fast should not come at the
expense of strong­—i.e. efforts to increase the speed of reconstruction should not lead
to rebuilding assets that are more vulnerable. All the benefits from a faster recovery
could easily disappear if a hastened reconstruction made it impossible to improve the
vulnerability of the asset to future hazards.

If successfully achieved together and in the 149 countries analyzed here, building back
stronger, faster, and more inclusively would reduce global well-being losses by 31.2
percent from US$555 billion to US$382 billion. Figure 19 (left) provides an overview
of the countries in which the full building back better package would lead to the

33 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

largest reduction in average well-being losses. The estimates show that overall well-being
losses due to natural disasters would be more than halved in all of these countries. As in
the previous sections, several small island states are among the list with the largest benefit
potential from a better recovery; their reduction in well-being losses due to natural disasters
is on average 58.8 percent (Figure 19, right).

Figure 19. Left: Top 10 largest reductions in average well-being losses due to
building back stronger, faster, and more inclusively. Right: Reductions in average
well-being losses for selected small island states.

Antigua & -78% Antigua & -78%

Barbuda Barbuda

Dominica -72% Dominica -72%

Trinidad &
Guatemala -72% -66%
Tobago

Trinidad & -66% Vanuatu -61%

Tobago

Zimbabwe -65% St. Lucia -57%

Myanmar -65% Tonga -56%

Belize -61% Fiji -52%

Vanuatu -61% Solomon -41%

Islands

Peru -61% St. Vincent & -41%

the Grenadines

Angola -60% Jamaica -38%

-80 -70 -60 -50 -40 -30 -20 -10 0 -80 -70 -60 -50 -40 -30 -20 -10 0

The reconstruction after a disaster can offer further opportunities for improving the
post-disaster outcome. In particular, the destroyed assets can be replaced so that they are
not only more resilient, but also better adapted to current and future needs, and using
the most recent technologies, which have higher levels of productivity.

Examples of such upgrading of assets are: (i) for households, the reconstruction of houses with
better insulation technologies and better heating systems, allowing for energy conservation
and savings; (ii) for companies, the replacement of old production technologies by new ones,
like the replacement of paper-based management files by computer-based systems; and (iii)

34 I B U I L D I N G B AC K B E T T E R
RESILIENT RECOVERY

for government and public agencies, the adaptation of public infrastructure to new needs, like the
reconstruction of larger or smaller schools when demographic evolution justifies it.

Figure 20. Percentage reduction of well-being losses associated with building back better.

Countries are sorted into quartiles (i.e. same number of countries for each color).

Capital losses can, therefore, be compensated by higher productivity of the economy in

the aftermath of an event. This process, if present, could increase the pace of technical
change and represent a positive consequence of disaster. This effect, often referred to as the
“productivity effect”, has been mentioned by Albala-Bertrand (1993), Okuyama (2003)
and Benson and Clay (2004), inter alia, and highlighted in the case of the Chicago fire
in Hornbeck and Keniston (2017). While it has been shown that the productivity effect
cannot turn natural disasters into desirable events (Hallegatte and Dumas, 2009), it can
reduce their overall impacts on economic growth and well-being.

The large potential benefits of building back better—and using more-recent and higher-
productivity technologies—provides a strong incentive to introduce policies and tools that
help ensure the highest possible quality of the reconstruction. However, these policies and
tools need to be in place before the disaster hits for their benefits to be realized. They are
usually incorporated into a disaster recovery framework and include contingency plans and
a clear allocation of reconstruction responsibilities, financial protection—through insurance
or social protection systems—and more general financial inclusion, with access to credit to
finance the reconstruction, and the availability of information and knowledge on how to
implement modern solutions and technologies.

35 I B U I L D I N G B AC K B E T T E R
 BUILDING
RESILIENT RECOVERY MORE
RESILIENT SOCIETIES

I n 2017, Hurricanes Irma and Maria caused large-scale

destruction across the Caribbean. While these events were
extreme, they were by no means unique. These disasters
happened in the context of increasingly frequent and intense
extreme weather events—a trend that is not limited to the
Caribbean, as other countries have also been heavily affected by
exceptional disasters in recent years. Continued urbanization and
population growth, especially in coastal areas, as well as climate
change, are likely to intensify these trends in the next decades.

In this context, the destruction caused by disasters highlights the need to

build back better—and also the opportunities it brings. Reconstruction offers
an opportunity to build more resilient societies, better able to withstand
future shocks.

However, building back better is just one of many measures for increasing
resilience, and can only be truly transformative if integrated into a more
comprehensive strategy of disaster risk reduction and management. For example,
if resilient building standards are not systematically applied to new and existing
facilities and infrastructure, there are bound to be capacity constraints to
implementing and enforcing these in a post-disaster situation.

While post-disaster reconstruction offers opportunities for fast and significant

progress, ex-ante actions are of paramount importance. Financial inclusion,
retrofitting of buildings, social safety nets, and risk-informed land-use
planning are some of the ex-ante measures that not only help to reduce the
losses in the case of a disaster—but also contribute to sustainable and resilient
development more generally.

36 I B U I L D I N G B AC K B E T T E R
BEYOND ASSET LOSSES

ACRONYMS AND ABBREVIATIONS

BNBP Indonesia’s National Board for Disaster Management

Cat-DDO Catastrophe Deferred Drawdown Option

CPA Fiji’s Care and Protection Allowance

DMU Decision-making under uncertainty

FONDEN Mexico’s Natural Disaster Fund

GDP Gross Domestic product

GFDRR Global Facility for Disaster Reduction and Recovery

IDF Indonesia Disaster Fund

MDF Multi Donor Fund for Aceh and Nias

PBS Fiji’s Poverty Benefit Scheme

PDNA Post-disaster needs assessment

SPS Fiji’s Social Pension Scheme

UNISDR United Nations Office for Disaster Risk Reduction

37 I B U I L D I N G B AC K B E T T E R
ENDNOTES

1. The Unbreakable model is available at [Link]

2. In this report, local currencies are translated into US$ using 2011 Purchase Power Parity rates.
3. These results use the model of the Unbreakable report with a few minor changes—but all results shown in this report have
been produced with the same model, ensuring that changes over time arise from changes in parameters, not changes in
the model.
4. United Nations General Assembly. 2016. Report of the Open-Ended Intergovernmental Expert 2 Working Group on
Indicators and Terminology Relating to Disaster Risk Reduction. Seventy-First Session, Item 19(c). A/71/644.
5. Countries that already have assets able to resist the 50-year return period event do not benefit from this measure. A time
frame of 20 years is chosen for this purpose: every asset destroyed by a natural disaster within a 20-year time frame is
reconstructed to the new standard. The estimated reduction in well-being losses are the amount that can be achieved
at the end of this time frame. (To simplify the interpretation of the results, we assume that everything else remains un-
changed—i.e. we remove the effect of population and economic growth during this period.)
6. Including the US Agency for International Development (USAID), the Swiss Agency for Development and Cooperation
(SDC), the World Bank, and the Japan International Cooperation Agency (JICA)
7. In the absence of the required data, self-targeting methods (public work programs) and subsidies can be used, but they
also require preparation.

38 I B U I L D I N G B AC K B E T T E R
REFERENCES

Albala-Bertrand, J. M. 1993. Political Economy of Large Natural Disasters: With Special Reference to Developing
Countries. New York: Clarendon Press; Oxford: Oxford University Press.

Alderman, H., J. Hoddinott, and B. Kinsey. 2006. “Long Term Consequences of Early Childhood Malnutrition.”
Oxford Economic Papers 58: 450–74.

Baez, J. E., and I. V. Santos. 2007. Children’s Vulnerability to Weather Shocks: A Natural Disaster as a Natural
Experiment. New York: Social Science Research Network.

Baez, J., A. de la Fuente, and I. V. Santos. 2010. “Do Natural Disasters Affect Human Capital? An Assessment
Based on Existing Empirical Evidence.” Discussion Paper No. 5164, IZA, Bonn Germany.

Benson, C., and E. Clay. 2004. Understanding the economic and financial impact of natural disasters. The
International Bank for Reconstruction and Development. The World Bank, Washington DC.

Bustelo, M. 2011. “Bearing the Burden of Natural Disasters: Child Labor and Schooling in the Aftermath of the
Tropical Storm Stan in Guatemala.” University of Illinois at Urbana-Champaign.

Carter, M. R., and C. B. Barrett. 2006. “The Economics of Poverty Traps and Persistent Poverty: An Asset-Based
Approach.” Journal of Development Studies 42: 178–99.

Clarke, D. J., and R. V. Hill. 2013. “Cost-Benefit Analysis of the African Risk Capacity Facility.” Discussion Paper
01292, International Food Policy Research Institute, Washington, DC.

de Janvry, A., F. Finan, E. Sadoulet, and R. Vakis. 2006. “Can Conditional Cash Transfer Programs Serve as Safety
Nets in Keeping Children at School and from Working When Exposed to Shocks?” Journal of Development
Economics 79: 349–73.

Demirgüç-Kunt, A., L. Klapper, D. Singer, and P. Van Oudheusden. 2015. “The Global Findex Database 2014:
Measuring Financial Inclusion around the World.” World Bank, Washington, DC.

Deryugina, T. 2016. “The Fiscal Cost of Hurricanes: Disaster Aid Versus Social Insurance.” National Bureau of
Economic Research, Cambridge, MA.

Dercon, S., and C. Porter. 2014. “Live Aid Revisited: Long-Term Impacts of the 1984 Ethiopian Famine on
Children.” Journal of European Economic Association 12: 927–48.

GFDRR. 2012. FONDEN: Mexico’s Natural Disaster Fund – A Review. Washington DC: Global Facility for
Disaster Reduction and Recovery.

Government of Fiji and World Bank. 2017. Climate Vulnerability Assessment – Making Fiji Climate Resilient.
Washington DC: World Bank.

Hallegatte, S. 2009. Strategies to adapt to an uncertain climate change. Global environmental change, 19(2), 240-247.

Hallegatte, S., J.C. Hourcade, and P. Dumas. 2007. Why economic dynamics matter in assessing climate change
damages: illustration on extreme events. Ecological economics, 62(2), 330-340.

Hallegatte, S. and P. Dumas. 2009. Can natural disasters have positive consequences? Investigating the role of
embodied technical change. Ecological Economics, 68(3), pp.777-786.

Hallegatte, S. and A.C. Vogt-Schilb. 2016. Are losses from natural disasters more than just asset losses? the role of
capital aggregation, sector interactions, and investment behaviors. World Bank Policy Research Working Papers.
WPS7885.

Hallegatte, S., A.C. Vogt-Schilb, M. Bangalore, and J. Rozenberg. 2017. Unbreakable: Building the Resilience of the
Poor in the Face of Natural Disasters. Climate Change and Development Series. Washington, DC: World Bank.

Hallegatte, S., M. Bangalore, and A.C. Vogt-Schilb. 2016. Socioeconomic resilience: multi-hazard estimates in 117
countries. World Bank Policy Research Working Papers. WPS7886.

39 I B U I L D I N G B AC K B E T T E R
REFERENCES

Hornbeck, R. and Keniston, D., 2017. Creative destruction: Barriers to urban growth and the Great Boston Fire of
1872. American Economic Review, 107(6), pp.1365-98.

Jensen, R. 2000. “Agricultural Volatility and Investments in Children.” American Economic Review 90: 399–404.

Jha, A., J. D. Barenstein, P. Phelps, D. Pittit, and S. Sena. 2010. Safer Homes, Stronger Communities: A Handbook
for Reconstructing after Natural Disasters. Washington DC: World Bank

Kalra, N., S. Hallegatte, R. Lempert, R., C. Brown, A. Fozzard, S. Gill, and A. Sha. 2014. Agreeing on robust
decisions: new processes for decision making under deep uncertainty. World Bank Policy Research Working
Papers. WPS6906.

Maccini, S., and D. Yang. 2009. “Under the Weather: Health, Schooling, and Economic Consequences of Early-
Life Rainfall.” American Economic Review 99: 1006–26.

Mansur, A., J. Doyle, and O. Ivaschenko. 2018. “Cash Transfers for Disaster Response: Lessons from Tropical
Cyclone Winston.” Development Policy Centre Discussion Paper No. 67. Australian National University.

Munich Re. 2018. Hurricanes cause record losses in 2017 - The year in figures. Online. Available from: https://
[Link]/topics-online/en/2018/01/2017-year-in-figures

National Oceanic and Atmospheric Administration. 2018. International Best Track Archive for Climate
Stewardship – Storm lines. Online. Available from: [Link]

Noy, I., and P. Patel. 2014. “Floods and Spillovers: Households after the 2011 Great Flood in Thailand.” Working
Paper Series No. 3609, School of Economics and Finance, Victoria University of Wellington.

Okuyama, Y., 2003. Economics of natural disasters: a critical review. Research Paper 2003-12, Regional Research
Institute, West Virginia University, USA.

UNISDR (United Nations Office for Disaster Risk Reduction). 2015. United Nations Global Assessment Report
on Disaster Risk Reduction. Geneva: UNISDR.

UNISDR (United Nations Office for Disaster Risk Reduction). 2017. Build Back Better in recovery, rehabilitation,
and reconstruction. Geneva: UNISDR. [Link]

World Bank. 2010. Colombia disburses US$150 million Cat-DDO. Online. Available at: [Link]
org/en/news/press-release/2010/12/28/colombia-disburses-us-150-million-cat-ddo-catastrophe-risk-financing-
from-world-bank-to-help-with-humanitarian-aid-and-reconstruction

World Bank. 2012. Indonesia: A Reconstruction Chapter Ends Eight Years after the Tsunami. Online. Available
at: [Link]
after-the-tsunami

World Bank. 2016a. The Istanbul Seismic Risk Mitigation Project. Online. Available at: [Link]
org/en/country/turkey/brief/the-istanbul-seismic-risk-mitigation-project

World Bank. 2016b. Nepal - Rural housing reconstruction program: program overview and operations manual
summary. Washington DC: World Bank

World Bank. 2017a. Climate and Disaster Resilient Transport in Small Island Developing States: A Call for Action.
Washington DC: World Bank

World Bank. 2018. China: Building Back Stronger. Online. Available from: [Link]
results/2018/05/11/building-back-better-plus

Yamano, T., H. Alderman, and L. Christiaensen. 2005. “Child Growth, Shocks, and Food Aid in Rural Ethiopia.”
American Journal of Agricultural Economics 87: 273–88.

40 I B U I L D I N G B AC K B E T T E R
 The Global Facility for Disaster Reduction and Recovery
[Link]

(GFDRR) is a global partnership that helps developing

countries better understand and reduce their
vulnerabilities to natural hazards and adapt to climate
change. Working with over 400 local, national,
regional, and international partners, GFDRR provides
grant financing, technical assistance, training, and
knowledge sharing activities to mainstream disaster
and climate risk management in policies and strategies.
Managed by the World Bank, GFDRR is supported by 33
countries and 11 international organizations.