public disclosure authorized progress...shaking up economic progress aggregate shocks in latin...
TRANSCRIPT
SHAKING UPECONOMICPROGRESS
Aggregate Shocksin Latin America
and the Caribbean
Javier E. BáezAlan Fuchs
Carlos Rodríguez-Castelán
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
© 2017 International Bank for Reconstruction and Development / The World Bank
1818 H Street NW, Washington DC 20433
Telephone: 202-473-1000; Internet: www.worldbank.org
Some rights reserved
This work is a product of the staff of The World Bank with external contributions. Note that The World
Bank does not necessarily own each component of the content included in the work. The World Bank
therefore does not warrant that the use of the content contained in the work will not infringe on
the rights of third parties. The risk of claims resulting from such infringement rests solely with you.
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. Nothing herein shall constitute or be considered to be a limitation
upon or waiver of the privileges and immunities of The World Bank, all of which are specifically
reserved.
Rights and Permissions
This work is available under the Creative Commons Attribution 3.0 Unported license (CC BY 3.0)
http://creativecommons.org/licenses/by/3.0. Under the Creative Commons Attribution license, you
are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under
the following conditions:
Attribution—Please cite the work as follows: Javier E. Báez, Alan Fuchs and Carlos Rodríguez-
Castelán. 2017. Overview: Shaking Up Economic Progress: Aggregate Shocks in Latin America and
the Caribbean. Washington, DC: World Bank. License: Creative Commons Attribution CC BY 3.0.
Translations—If you create a translation of this work, please add the following disclaimer along with
the attribution: This translation was not created by The World Bank and should not be considered
an official World Bank translation. The World Bank shall not be liable for any content or error in this
translation.
All queries on rights and licenses should be addressed to the Office of the Publisher, The World Bank,
1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: [email protected].
SHAKING UPECONOMICPROGRESS
Aggregate Shocksin Latin America
and the Caribbean
Javier E. BáezAlan Fuchs
Carlos Rodríguez-Castelán
3
Contents
1. Executive Summary 9
2. Why Care about Aggregate Shocks in Latin America and the Caribbean? 17
The region has made impressive strides in the struggle against poverty and income inequality 17
But the gains are not assured partly because of the substantial exposure to multiple shocks 18
Aggregate shocks can reverse economic and social development 24
3. Analytical Framework: The Link between Shocks and Household Welfare 29
4. Aggregate Shocks Threaten Progress in Household Welfare 35
Aggregate shocks often signify income shocks 36
The effects of aggregate shocks on incomes can persist 37
Income shocks translate into consumption shortfalls and greater vulnerability to poverty 39
5. The Vulnerability of Assets and Livelihoods to Aggregate Shocks 45
Progress in the region has translated into remarkable welfare-enhancing asset accumulation 45
Yet, progress in asset accumulation is at risk because of the effects of aggregate shocks 46
Aggregate shocks disrupt and constrain asset accumulation 49
Not everyone’s assets are affected equally 51
6. Aggregate Shocks Undermine Incentives to Use Factors of Production and Invest 57
The supply of labor is sensitive to major shocks, increasing at times and at times falling 57
Migration is a common risk mitigation strategy 58
Households tend to rely on the labor of their children to cope with aggregate shocks 60
The demand for labor is responsive to aggregate shocks 61
Shocks and perceptions of risk distort the allocation of financial capital 63
Risk pushes households to use productive assets inefficiently and forgo some returns 64
Because of shocks or risk management strategies, production factor returns are often lower 66
7. Risk Management Policy: Preparing for and Coping with Aggregate Shocks 69
The rationale for public action 69
A policy framework to prepare for and cope with aggregate shocks 71
References 79
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
4
ACKN
OW
LED
GM
ENTS
5
Acknowledgments
This sTudy, parT of The LaTin america and caribbean Regional Studies Program of the World
Bank, is the work of a team led by Javier E. Báez, Carlos Rodríguez-Castelán, and Alan Fuchs.
The team has been ably assisted by Germán Caruso, who also contributed substantially to
the background material. Important additional contributions have been made by Andres
Garcia-Suaza, Marcelo Paiva, and Shital Beejadhur. Background papers were provided
by Leonardo Lucchetti, Monserrat Bustelo, Javier E. Báez, Andres Zambrano, Leopoldo
Fergusson, María E. Genoni, Oscar Ishizawa, Johana Rivera, Juan J. Miranda, Juliana Suarez,
Luis F. Jimenez, Mary Boyer, Gonzalo Pita, Rafael Santos, Valentina Duque, Germán Caruso,
Pablo Acosta, Eduardo Malásquez, Maria C. Acevedo, Mateo Salazar, Valerie Mueller, José
Luis Toro, Carlos Carcach, Emma Monsalve, Chiyu Niu, Gonzalo Reyes, Javier Bronfman,
Maria F. Rosales, and Thiago Scot. We also thank the many World Bank colleagues, peer
reviewers, and other experts who have contributed ideas and guidance: Augusto de la
Torre, Daniel Lederman, Michael Carter, Richard Akresh, Emmanuel Skoufias, Kathleen
Beegle, Laura Chioda, Leonardo Lucchetti, Marianne Fay, Stephane Hallegatte, Margaret
Grosh, Louise Cord, Oscar Calvo-Gonzalez, and Marcela Sanchez-Bender. Editorial help has
been provided by Robert Zimmermann in English. The team also extends many thanks to
Karem Edwards and Ruth Delgado Flynn for valuable administrative support.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
6
About the Authors
Javier e. baez is a Senior Economist for the Africa Region and a Global Lead of Risk
Management in the Poverty and Equity Global Practice of the World Bank. Javier is also a
Research Fellow at the Institute for the Study of Labor (IZA). His areas of interests are in
the fields of development economics and impact evaluation. His research focuses on the
short- and long-term effects of shocks, either natural (hurricanes, floods earthquakes) or
man-made (civil conflicts), on human development outcomes in countries like Tanzania,
Mozambique, Guatemala, El Salvador and Nicaragua. He has also led impact evaluations of
the long-term effects of conditional cash transfers in Colombia and Pakistan. His research
has been featured in several press articles and published in the Journal of Development
Economics, Journal of Development Studies, American Economic Review (Papers and
Proceedings), Climatic Change and World Bank Research Observer. Javier holds a PhD in
Economics from Syracuse University, an MPA/ID from Harvard University and B.A. and
M.Sc. degrees in Economics from Universidad de Los Andes in Colombia.
aLan fuchs TarLovsky is a Senior Economist in the Poverty and Equity Global Practice
for Latin America and the Caribbean of the World Bank. He co-leads the Global Solutions
Group for Fiscal and Social Policies for Poverty and Shared Prosperity. Alan has led
operations and technical work in Chile and Uruguay, including the First Social Inclusion
DPL for Chile, the SCD for Chile and Uruguay and analysis of the distributional effects of
the 2014 Chilean fiscal reform. Alan’s research focuses on development economics, applied
microeconomics, insurance and energy economics and has peer reviewed publications at
the American Economic Review, American Economic Journal: Economic Policy and the
American Journal of Agricultural Economics. Prior to joining the World Bank Alan worked
for the UNDP and for the Mexican Government at the Ministry of Finance, Mexican Institute
for Social Security and Ministry for Social Development. He holds a PhD from UC Berkeley,
an MPA/ID from Harvard University and B.A. in Economics and a MPP from Instituto
Tecnológico Autónomo de México (ITAM).
ABO
UT
THE
AUTH
ORS
7
carLos rodríguez-casTeLán is a Senior Economist and since 2014 a Global Lead of
Markets and Institutions in the Poverty and Equity Global Practice of the World Bank. He
has led operations and analytical work on social inclusion and economic productivity
in Chile, Colombia, Mexico and Uruguay. Between 2012 and 2014 he also led the Latin
America Team for Statistical Development. Prior to joining the bank he was a postdoctoral
fellow at Brookings (2011), and held research positions at the UN-ECLAC (2009)
and the IDB (2008). He also worked for the Government of Mexico at the Ministry of
Education (2004-2006) and the Ministry of the Environment (2003-2004). His research
interest are in the fields of development economics and public economics. He has
published in peer reviewed academic journals and is co-author of World Bank books,
including: Shared Prosperity and Poverty Eradication in Latin America and the Caribbean,
and Inequality in a Lower Growth Latin America. As of January 2017, he was ranked as one
of the Top 100 Young Economists by IDEAS/RePEc. Carlos holds a PhD in Economics and
MA from Cornell University, as well as a B.A. in Economics, a B.A. in Political Science and
a MPP from ITAM, Mexico.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
8
CHA
PTER
1Ex
cecu
tive
Sum
mar
y
9
1. Executive Summary
The region has made impressive strides in the struggle against poverty and income inequality
The Latin America and Caribbean region has achieved remarkable economic and
social progress over the last decade, gradually shifting toward middle-income status.
Economic growth reached an average annual rate of 3.2 percent over 2000–14, noticeably
higher than in previous decades. This favorable context contributed to significant poverty
reduction and expansion of the middle class. The proportion of the region’s 600 million
people living in extreme poverty, defined in the region as life on less than $2.50 a day, was
cut in half between 2003 and 2012, to 12.3 percent. Similarly, the share of Latin Americans
living in moderate poverty, corresponding to living on less than $4.00 a day, fell from 41.1
percent to 25.3 percent. Since 2011, there have been more Latin Americans in the middle
class than in poverty, and the middle class is projected to become the largest group in the
region (World Bank 2014a). The gains attained span other areas of human development
such as increased access to basic services and lower child and maternal mortality.
But the gains are not assured partly because of substantial exposure and vulnerability to multiple shocks
While a significant share of households moved upward in socioeconomic class, the
largest group of the population remained vulnerable to poverty. Many Latin Americans
escaped poverty buoyed by more than a decade of strong economic growth and inequality
reduction. However, most of the new nonpoor did not move directly to the middle class,
but continued to be vulnerable, facing economic insecurity and likely to experience spells
of poverty in the future.1 Nearly 4 households in 10 in the region belong to this group,
making it the largest socioeconomic class. Owing to their substantial vulnerability, many
of these households may be one shock away from sliding back into poverty.
1 López-Calva and Ortiz-Juárez (2014) propose a 10 percent probability of falling into poverty as a dividing line between economic security and vulnerability and define the predicted income associated with that probability as the upper bound of vulnerability or the lower bound of the middle class. The lower bound is the $4.00-a-day poverty line. The resulting per capita incomes among nonpoor individuals facing a 10 percent probability of falling into poverty was $9.80 based on panel data for Chile and Mexico.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
10
The substantial vulnerability is compounded by exposure to a numerous shocks that
affect the region, particularly natural hazards. The incidence of these phenomena grew
threefold regionally and globally between 1970 and 2014. Extreme rainfall and drought
are a near-constant threat across the region. Of every 10 natural events recorded in the
region, 7 arise because of storms and floods (Holt 2014). In the Caribbean, at least one
country—and often more than one—is hit by a strong hurricane or cyclone every year. The
dry corridor, a region of dry tropical forests in Central America spanning parts of Costa
Rica, El Salvador, Guatemala, Honduras, Nicaragua, and Panama, is regularly affected
by recurrent droughts that endanger crops, livestock, and food security. The Andean and
Central American countries are within the Ring of Fire, a string of volcanoes and sites of
seismic activity in the Pacific Ocean that accounts for roughly 90 percent of the world’s
earthquakes and more than 75 percent of the world’s active and dormant volcanoes. Over
a quarter of all earthquakes of magnitude 8.0 or higher occur in western South America.
The region must also address major economic and social risks. A more restrictive
macroprudential regulatory framework across most countries has made them more
resilient to economic downturns. Yet, the region is edging toward a new equilibrium of
lower growth, coupled with a rise in current account deficits and greater exposure to
other external factors. Crime and violence continue to top the list of challenges in many
countries; the incidence of crime is comparable with the rates recorded in war-torn
countries. An individual born in Caracas, San Salvador, or Tegucigalpa—the three most
violent large cities in the region—has about one chance in eight of being murdered during
his lifetime. Parts of the region show a proliferation of violent youth gangs, drug trafficking,
money laundering, and domestic violence, whereas other parts have endured civil conflict
and unrest. Infectious disease and viruses thrive in warm and humid conditions such as
those in the tropics, which span most countries in the region. Major epidemics such as
Chikungunya and Zika are a serious risk to public health.
Aggregate shocks often signify income shocks
Large natural and man-made shocks undermine economic expansion and household
incomes. The first manifestation of a major negative shock at the macro level is typically
a drop in output. Gross domestic product (GDP) fell by 6 percent and 11 percent during
the crises that hit Mexico (1994–95) and Argentina (2001–02), respectively. At the micro
level, household incomes in the region are equally likely to experience hefty declines in
the aftermath of major shocks. Hurricanes in Central America cut household incomes by 3
percent for each standard deviation in the intensity of the hurricane windstorm (Ishizawa
and Miranda 2016). Similarly, two earthquakes that struck El Salvador in 2001 reduced
a third of the preshock median per capita income of households in areas exposed to the
strongest ground shaking (Báez and Santos 2009).
CHA
PTER
1Ex
cecu
tive
Sum
mar
y
11
Income reductions caused by aggregate shocks may persist
Some households recover quickly following aggregate shocks, but recovery may take
much longer among many other households, particularly the poor and vulnerable. Access
to well-functioning credit and insurance markets, as well as savings, liquid assets, and
informal risk-sharing mechanisms such as private transfers and community-based financial
systems position households to withstand the negative income effects of major shocks. This
is often not the case among the poor and vulnerable, who may be obliged to undertake a shift
to a lower equilibrium of welfare and income growth. In Honduras, for example, two and
half years after Hurricane Mitch (1998) had ravaged Central America, evidence shows that
wealthier households were able to bounce back quickly, but less well off households went on a
downward trajectory of sustained asset depletion and low income growth (Carter et al. 2005).
Income reductions translate into consumption shortfalls and greater vulnerability to poverty
Income risks help determine consumption, especially among vulnerable, low-income
households. Incomplete insurance means vulnerable households manage to protect their
consumption only partially from the negative income shocks caused by major hazards.
Plentiful evidence shows that consumption (including basic expenditures) tends to
fall in households and communities affected by natural disasters. Median per capita
consumption fell by 7.7 percent in households affected by Agatha, a major tropical storm
that hit Guatemala in 2010 (Báez et al. 2016). Man-made shocks potentially lead to similar
outcomes. Households that were obliged to move from conflict zones in Colombia suffered
a 22 percent decline in consumption, affecting their food and calorie intake (Ibáñez and
Moya 2006). Moreover, transient impacts on consumption can become chronic. Panel data
on rural families in El Salvador show lower consumption growth between 1995 and 2001
among less well off households that had dealt with substantial income shocks (Rodríguez-
Meza and González-Vega 2004).
Because of consumption volatility, vulnerable households may be a disaster away
from falling below the poverty line or sliding further back into poverty. If insurance
and risk management are lacking or incomplete, income or consumption declines caused
by major shocks push some households into poverty or deepen the levels of deprivation
among the poor. Poverty rates rose by 5.5 percentage points in areas of Guatemala flooded
by Agatha (2010), equivalent to nearly 80,000 more families falling below the poverty
line (Báez et al. 2016). In 2003–04, the Dominican Republic experienced a bank collapse,
rapid currency depreciation, and inflation, which hastened a domestic crisis that led to
economic contraction. The poverty rate soared from 32 percent in 2002 to over 50 percent
at the peak of the crisis (World Bank 2014b).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
12
The significant vulnerability is influenced by the enduring negative effects of shocks on current asset stocks and future
asset accumulation
Major shocks reduce asset ownership. Human capital is particularly sensitive. Natural
disasters, civil conflict, and widespread epidemics result in human casualties, leaving
negative human capital impacts in their wake. The 2010 earthquake that struck Haiti
caused close to 250,000 fatalities. Colombia’s internal conflict, lasting for over 50 years,
has claimed a similar number of lives. In addition to human casualties, these events
often result in the destruction of the infrastructure required for the acquisition of human
capital, such as schools, hospitals, and clinics. Likewise, shocks destroy private property
(housing, machinery, crops and livestock), other critical public infrastructure (roads and
bridges), and natural capital. Hurricane Mitch (1998) decimated over 80,000 hectares of
agricultural land, the large majority of which was used by small farmholders for subsistence
agriculture (Ishizawa and Miranda 2016).
Shocks also undermine investments in assets. Major shocks can pose a financial burden
on households, often forcing them to cut back on food and health care expenditures,
thereby raising the risk of malnutrition and other negative effects on health. For example,
Nicaraguan children ages 0–5 in households located in the path of Hurricane Mitch (1998)
were 30 percent less likely than children in unaffected parts of the country to be taken for
medical consultation after the shock (Báez and Santos 2007). Households may also need
to pull their children out of school to benefit from the extra labor, but at the cost of lower
human capital. Once Mexican children are withdrawn from school in response to large
shocks, they become nearly 30 percent less likely to reenroll relative to children who stay
in school (Sadoulet et al. 2004).
Not everyone’s assets are affected equally—the poor and the vulnerable often bear the greater burden
Aggregate shocks affect broad economic and social systems, but their impacts on
assets, groups, and communities are not evenly distributed. The earthquakes that
struck Chile and Haiti in 2010 were of similar magnitude, but were associated with
starkly contrasting outcomes: 525 people were killed in the former case compared with
over 200,000 in the latter. The variations in impact are determined by the circumstances
of the affected population, such as gender equality, geographical location, educational
attainment, and type of job. In Mexico, the poor are almost three times more likely
than the vulnerable or the middle class to be affected by a natural disaster that results
in loss of dwellings, crops, and livestock (de la Fuente, Ortiz-Juárez, and Rodríguez-
Castelán 2017).
CHA
PTER
1Ex
cecu
tive
Sum
mar
y
13
Shocks undermine the incentive to use factors of production optimally
The supply of labor is sensitive to major shocks, at times increasing and at times
declining. Aggregate shocks destroy household wealth and assets, reducing the income
opportunities that can be generated from wealth and assets. They change prices and thus
relative wages. In an effort to protect private expenditures or prevent expenditures from
falling below subsistence needs, affected individuals may respond by supplying extra labor.
Households in northern Colombia, for example, attempted to cope with extreme flooding
that struck the country in 2010 by boosting their labor participation (Acevedo 2016). In
contrast, unfavorable economic circumstances or major health setbacks can result in less
labor supply in the market. Forced migration, a typical risk mitigation strategy, imposes
high costs on populations by unsettling their engagement in labor and other markets. Over
half the household heads internally displaced by the conflict in Colombia were unemployed
three months after settlement at destination sites (Ibáñez and Moya 2006).
Households often rely on the labor of their children to cope with shocks, and this
comes at a cost: lower human capital and productivity. Children represent a buffer,
particularly among resource-constrained households. In tough times, households are often
forced to resort to the labor of their children, who can contribute to income or free up time
for adults. However, children’s work entails long-run costs as it often interferes with the
human capital accumulation of the children and leads to reduced earnings potential. For
instance, Brazilian children, particularly girls, were more likely to abandon school and
enter the labor force during several of the economic slowdowns recorded in the 1980s and
1990s. Once out of school, they became 10 percentage points less likely to progress across
grades later on, offsetting short-term smoothing gains with long-term losses in human
capital (Duryea, Lam, and Levison 2007).
Major shocks disrupt job creation, reducing the demand for labor. Large natural
disasters or civil conflicts do more than wipe out homes. They decimate local economies,
thereby distressing businesses, supply chains, and markets. Close to 100,000 jobs
were lost in Chile after the 2010 earthquake, pushing up the unemployment rate by
1 percentage point. Economic recessions lead to a slowdown in output, a slump in
consumer spending, and a reduction in investments in capital and the expansion of
credit. As this broad-spectrum aggregate supply slowdown unfolds, firms are likely
to freeze hiring and lay off employees, which ultimately results in lower demand for
labor, higher unemployment, and diminished job quality. The global financial crisis of
2008–09 raised unemployment rates in Mexico by more than 50 percent (Freije, López-
Acevedo, and Rodríguez-Oreggia 2011).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
14
Uninsured risk pushes households to use productive assets inefficiently and forgo
significant returns. Credit- and insurance-constrained, risk-averse households are
inclined to mitigate ex ante the effects of shocks by making conservative employment and
production choices. Thus, fertilizers or enhanced seeds that boost crop productivity and
raise expected profits are typically used less intensively by risk-prone households for fear
of incurring investment losses if shocks result in poor harvests. Vulnerable households self-
insure by diversifying toward portfolios with safer, but less profitable activities. Farmers in
the Peruvian district of Cuyocuyo, for instance, diversify their crops spatially by working on
smaller fields rather than on a more efficient, consolidated plot. However, this diminishes
the overall yields by 7 percent (Goland 1993). Likewise, evidence shows that risk-reduction
income diversification strategies among vulnerable households often translate into lack of
specialization, small scale, informality, and even little income smoothing (the tactic of
leveling expenditures across income highs and lows, that is, saving in good times to be
able to spend normally in bad times).
Efficiency and equity losses linked to inadequate insurance, protection, and coping justify government intervention
Many of the negative impacts and development setbacks that arise from shocks are the
result of inadequate risk management. Households follow multiple strategies to manage
risk, but these are only partially effective. Inadequate risk management discourages
households from taking on the risks that are encountered in pursuing economic gain,
undermining income generation. Income- and asset-based strategies to self-insure smooth
incomes, but also skew them, clustering households into low-return, low-risk activities. The
poor functioning of asset markets constrains the access to alternative economic activities
and depresses asset values if these are liquidated during a crisis. Informal risk-sharing
mechanisms are prone to collapse if aggregate shocks occur. The welfare losses caused
by these inefficiencies are not trivial. In addition, because the shocks tend to affect the
poor and vulnerable disproportionally owing to a combination of greater exposure, weaker
internal conditions, and unsatisfactory risk management, the resulting equity losses are
equally large. The costs of inaction are substantial.
Risk management policy needs to address four objectives to help households effectively
prepare for and cope with shocks. World Development Report 2014 argues that preparing
for aggregate shocks involves three of the objectives: (1) gaining knowledge to understand
the shocks, conditions, and potential outcomes, thereby reducing the uncertainties faced
by people, societies, and countries; (2) building protection to reduce the probability
and size of losses, while increasing the chances and size of positive outcomes; and (3)
acquiring insurance to transfer resources. Successfully coping with aggregate shocks ex
post is covered in the fourth objective: (4) applying proven coping mechanisms to recover
from the losses inflicted by the shocks.
CHA
PTER
1Ex
cecu
tive
Sum
mar
y
15
Strengthening the preparation and coping aspects of risk management requires
policy actions on five main fronts, paying special attention to efficiency and equity
considerations:
1 Address market failures and the underprovision of public goods. A major factor
explaining suboptimal risk management is the lack of critical markets for credit,
insurance, and jobs. These markets may exist, but fail to develop completely. Better
financial inclusion contributes directly to faster recovery after a disaster, while also
supporting asset diversification, which reduces vulnerability. Basic services and public
goods that are essential in managing risks such as safe water and sanitation, education,
key infrastructure, weather warning systems, economic and political stability, and the
rule of law are often missing or substandard.
2 Internalize social and economic externalities. Economic activities adopted by some
agents, including risk management strategies, can impose costs on others. Lack of
land-use and building regulations can lead to infrastructure development in unsafe
places under unsafe construction codes. Free-riding among agents who benefit from
risk prevention or mitigation without contributing to the costs likely discourages risk
management. Regulatory policies that enable collective action and help internalize
relevant externalities need to be adopted.
3 Reform weak government incentives. Intertemporal political incentives are an
impediment to adequate risk management. Risk preparation requires investment,
often costly, that potentially yields discernible returns only in the medium and long
term. Governments prefer to devote resources to policies and programs that produce
gains in the short term even if the gains are smaller or to rely on international aid,
which translates into a widespread inability to prepare for crises before they occur.
Addressing these failures requires the development ex ante of coordinated plans for
postdisaster action and agreements on standby financing.
4 Deal with the lack of resources and information. Investments in risk management
infrastructure and technology usually involve large upfront costs. Households and
governments on tight budgets may favor current expenditures over investments in
risk reduction and mitigation. Lack of information on the relevant risks and on the
benefits of risk management constrains the ability of public and private agents to price
risks, undermining public and private efforts to insure against shocks and provide
compensation at actuarially fair rates. An option is to leverage private resources and
official development assistance for large, better investments in resilience.
5 Develop and strengthen rapidly scalable social protection. Households cannot be
fully insured against every type of shock. Ex post mechanisms such as social transfers
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
16
are necessary to ensure a minimally acceptable standard of living, especially among
the poor. Adaptive and scalable social protection can provide this type of insurance
during a crisis by increasing the amount transferred to beneficiaries, relaxing eligibility
rules and conditionalities, extending coverage to new beneficiaries, or creating a new
program. Latin America and the Caribbean has succeeded in building safety nets to
alleviate poverty. It is now time to use them to enhance the resilience of nonpoor
households that face a substantial risk to slide back into poverty if struck by a shock.
CHA
PTER
2W
hy C
are
abou
t Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n?
17
Figure 2.1. The Region Has Recorded Remarkable Poverty Reduction and Expansion of the Middle Class
Source: World Bank (2016).
Poverty $2.50 (2005 PPP)
Porc
enta
ge (%
)
35
42.8
25.5
21.2
23.3
10.8
0
10
20
30
40
50
201420132012201120102009200820072006200520042003200220012000
Poverty $4 (2005 PPP) Middle Class $10-$50 (2005 PPP)
2. Why Care about Aggregate Shocks in Latin America and the
Caribbean?
The region has made impressive strides in the struggle against poverty and income inequality
The Latin America and Caribbean region has achieved remarkable economic and
social progress over the last decade, gradually shifting toward middle-income status.
Economic growth reached an average annual rate of 3.2 percent over 2000–14, noticeably
higher than the rates in previous decades. This favorable context contributed to significant
poverty reduction and expansion among the middle class. The share of the region’s 600
million people living in extreme poverty, defined in the region as life on less than $2.50 a
day, was cut by half between 2003 and 2012, to 12.3 percent. Similarly, the share of Latin
Americans living in moderate poverty, corresponding to living on less than $4.00 a day,
fell from 41.1 percent to 25.3 percent. The Gini coefficient has declined by over 10 points
since 2003. Since 2011, there have been more Latin Americans in the middle class than in
poverty (figure 2.1). The middle class has been growing rapidly and is projected to become
the largest group in the region, exceeding the share of the vulnerable, namely. those who
are not poor, but who face a higher risk of falling back into poverty (World Bank 2014a).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
18
The laudable gains also span other areas of human development. Access to basic services
such as education, health care, electricity, water and sanitation not only increased, but
also became more even across the population. Under-5 mortality declined by 28 percent
during the first decade of the 21st century. Literacy rates are at an all-time high: 92 percent
and 97 percent among adults and youth, respectively. Life expectancy, which was 67 years
in 1990, has improved significantly, to 75 years today.2 Workers have become more skilled;
the quality of jobs has risen; and more women have joined the labor force.
But the gains are not assured partly because of the substantial exposure to multiple shocks
While a large share of households moved up the socioeconomic ladder, a larger share
risked falling back into poverty. Many Latin Americans escaped poverty buoyed by more
than a decade of strong economic growth and inequality reduction. However, most of
the new nonpoor did not move directly into the middle class, but joined the vulnerable,
that is, individuals facing substantial economic insecurity and likely to experience spells
of poverty.3 Nearly 4 households in 10 belong to this group and represent the largest
socioeconomic class in the region. Owing to their exposure and vulnerability, many of
these households may be one shock away from sliding back into poverty, a risk that is
particularly great in in the region.
Large shocks continue to affect the region. Natural hazards may be the greatest risk.
Across the world, natural disasters have become more frequent; the incidence of these
phenomena grew threefold in the region and globally between 1970 and 2014 (figure 2.2).
Environmental hazards, such as droughts, floods, landslides, hurricanes, earthquakes, and
volcanic eruptions are a near-constant threat across the region. In the Caribbean, at least
one country—and often more than one—is hit by a strong hurricane or cyclone every year.
In most countries, poor households are more regularly exposed to floods (Hallegatte et al.
2016). The Andean and Central American countries are within the Ring of Fire, a string
of volcanoes and seismic activity sites in the Pacific Ocean that accounts for roughly 90
percent of the world’s earthquakes and more than 75 percent of the world’s active and
dormant volcanoes (map 2.1). Natural hazards are also converging into more intense, more
severe phenomena. The 2014 report of the Intergovernmental Panel on Climate Change
underlines that the intensity of extreme weather and climate in the region is expected to
increase in coming years (Magrin et al. 2014).
2 WDI (World Development Indicators) (database), World Bank, Washington, DC, http://databank.worldbank.org/data/reports.aspx?source=world-development-indicators.3 López-Calva and Ortiz-Juárez (2014) propose a 10 percent probability of falling into poverty as a dividing line between economic security and vulnerability and define the predicted income associated with that probability as the upper bound of vulnerability or the lower bound of the middle class. The lower bound is the $4.00-a-day poverty line. Based on panel data for Chile and Mexico, the resulting daily per capita incomes among nonpoor individuals facing a 10 percent probability of falling into poverty was $9.80.
CHA
PTER
2W
hy C
are
abou
t Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n?
19
Figure 2.2. The Incidence of Natural Disasters Has Been on the Rise Globally and Regionally
LAC
0
100
200
300
400
500
201020052000199519901985198019751970
World
Source: World Bank calculations based on data of EM-DAT (International Disaster Database), Centre for Research on the Epidemiology of Disasters, Université Catholique de Louvain, Brussels, http://www.emdat.be/database.Note: Index 1970 = 100.
Map 2.1. Several Types of Natural Disasters Are Geographically Clustered in Latin America and the Caribbean
Sources: World Bank and United Nations 2010. Earthquakes above magnitude 6 on the Richter scale, 1950–Feb-ruary 2010: NCEDC (Northern California Earthquake Data Center) (database), Berkeley Seismological Laboratory, University of California, Berkeley, Berkeley, CA, www.ncedc.org. Droughts: based on SPI (Standardized Precipita-tion Index) (database), National Center for Atmospheric Research, Boulder, CO, https://climatedatagu-ide.ucar.edu/climate-data/standardized-precipitation-index-spi (larger values indicate a higher probability of precipitation deficits). Compiled for UNISDR 2009.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
20
Extreme rainfall and drought are commonplace in the region. Of every 10 natural
events recorded in the region, 7 arise because of storms and floods (Holt 2014). Most
cyclones and storms begin in the eastern Atlantic, but develop into deadly hurricanes by
the time they make landfall, often in the Caribbean or Central America. Extreme weather
events can become manifest on both sides of the spectrum; thus, severe, prolonged water
shortages are also common. The dry corridor, a region of dry tropical forests in Central
America spanning parts of Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, and
Panama, is regularly affected by recurrent droughts that endanger the crops, livestock, and
food security of over a million families engaged in subsistence farming. In the last couple
of years, southeast Brazil has experienced long droughts, leading to water scarcity in São
Paulo, the country’s most densely populated state (Seth, Fernandes, and Camargo 2015).
Floods and droughts are aggravated by the warmer- and cooler-than-normal sea surface
temperatures in the Pacific Ocean caused by the El Niño and La Niña phenomena.
The region is sensitive to earthquakes and other geological risks. Over a quarter
of all earthquakes of magnitude 8.0 or higher occur in western South America. The
strongest earthquake ever recorded (magnitude 9.5) struck Chile in 1960. Vulnerability
to earthquake-related disasters is mostly confined to the active fault lines where the
Cocos and Nazca plates converge with the South American plate on the western coasts
of Central and South America. The 7.0 magnitude earthquake that ravaged Haiti in 2010
killed over 200,000 people, left 1.5 million homeless, and destroyed billions of dollars in
infrastructure and the recent 7.8 magnitude earthquake that struck Ecuador and instantly
generated damage that represented over 3 percent of the country’s GDP remind us of the
destructive potential of these catastrophes.
The region is edging toward a new equilibrium of lower growth, coupled with a rise in
current account deficits. The adoption of a more restrictive macroprudential regulatory
framework across most countries has made the region more resilient to economic
downturns. The region experienced a recession during over a third of the 1980s, but the
frequency of recession fell to 10 percent in the 1990s and much lower in the 2000s (World
Bank 2013a). However, the context has changed in the last few years, and the region seems
to be edging toward a new equilibrium of greater exposure to external factors. Countries
benefited in the late 1990s and 2000s from growing external demand for commodities,
allowing for increases in consumption. Yet, the end of the commodity super cycle in recent
years was not paired with restraint in consumption, leading to growing current account
deficits across the region, mainly financed through debt. A sudden rise in interest rates
could stop capital inflows and make debt untenable, thereby causing a potential balance
of payments crisis and adding pressure to fiscal imbalances (figure 2.3) (World Bank 2015).
Commodity price fluctuations add to macroinstability, especially in countries in which
growth is fueled by international commodity markets. Among world regions, the Latin
CHA
PTER
2W
hy C
are
abou
t Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n?
21
America and Caribbean region shows the strongest correlation between international
commodity prices and terms of trade, which has a great influence on other macroeconomic
variables. Indeed, over the last decade, the commodity boom boosted growth in the region.
Recently, however, the sluggishness of the Chinese economy, coupled with weaker global
growth, has put downward pressure on commodity prices. The price of oil, which fell by
more than half between September 2014 and January 2015, as well as the subsequent
strong appreciation of the U.S. dollar, squeezed fiscal and current accounts, suppressing
growth in the region. Fluctuations in food prices disproportionally hurt households at the
bottom of the income distribution, given that a larger share of their incomes are devoted
to food consumption and that their ability to substitute cheaper sources of calories is often
limited. The 2007–08 food price crisis and resulting food inflation were barriers along the
region’s path toward reducing food insecurity (Robles and Torero 2010). In Ecuador, Haiti,
Nicaragua, Panama, and Peru, calorie intake shifted below sufficiency in poor households,
raising malnutrition rates.4
Crime and violence are among the top challenges. Nearly a third of the population in
the region ranks the lack of physical security as the main threat to basic rights.5 In stark
contrast to the global decline in crime and homicide rates in the last two decades, the
region has experienced a steady rise (figure 2.4). It is considered the most dangerous
4 “Nutritional Impacts of Rising Food Prices,” Food Security Portal, International Food Policy Research Institute, Washington, DC, http://www.foodsecurityportal.org/nutritional-impacts-rising-food-prices?print.5 Latinobarómetro Database, Latinobarómetro Corporation, Santiago, Chile, http://www.latinobarometro.org/latContents.jsp.
Figure 2.3. The Region Has Seen Increasing Current Account Deficits in Recent Years
External balance on goods and services (% of GDP)
-4-3-2-101234567
2014
2011
200
8
200
5
200
2
1999
1996
1993
1990
1987
1984
1981
1978
1975
1972
1969
1966
1963
1960
Source: World Bank calculations based on data of WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/data-catalog/world-development-indicators.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
22
Table 2.1. Longest, Deadliest Internal Conflicts, Latin America and the Caribbean, 1964–present
Country Period Insurgent Group Consequences
Colombia 1964–2016 FARC, ELN, M-19 Casualties: 220,000Displaced people: 4.7–5.7 million
Guatemala 1965–95 FAR I, FAR II, ORPA, EGP, URNG
Casualties: 140,000–200,000
Peru 1965, 1982 MIR, Sendero Luminoso Casualties: 70,000+
El Salvador 1972, 1979–91 FMLN, ERP, FPL Casualties:70,000–80,000Displaced people: 550,000
Nicaragua 1978–89 Contras/FDN Casualties: 20,000–50,000
Sources: Data of UCDP (Uppsala Conflict Data Program) (database), Department of Peace and Conflict Research, Uppsala University, Uppsala, Sweden, http://ucdp.uu.se/?id=1; data of ACLED (Armed Conflict Location and Event Data Project) (database), Robert S. Strauss Center for International Security and Law, Austin TX, http://www.acleddata.com/.Notes: EGP: Ejército Guerrillero de los Pobres. ELN: Ejercito de Liberación Nacional. ERP: Ejército Revolucionario del Pueblo. FAR: Fuerzas Armadas Rebeldes. FARC: Fuerzas Armadas Revolucionarias de Colombia. FDN: Fuerza Democrática Nicaragüense. FMLN: Frente Farabundo Martí para la Liberación Nacional. FPL: Fuerzas Populares de Liberación Farabundo Martí. M-19: Movimiento 19 de Abril. MIR: Movimiento de Izquierda Revolucionaria. ORPA: Organización del Pueblo en Armas. URNG: Unidad Revolucionaria Nacional Guatemalteca. The Government of Colombia signed a peace accord with FARC in 2016. However, the agreement does not include ELN.
Figure 2.4. Latin America and the Caribbean Is the Most Dangerous Region in the World, circa 2013 (average homicide rate, by subregion per 100,000 population)
0 5 10 15 20 25 30 35 40
Western EuropeAustralia and New Zealand
Southern EuropeNorthern Europe
Western AsiaEastern Asia
PolynesiaEastern Europe
MelanesiaNorthern Africa
MicronesiaNorthern America
Southern AsiaSouth-Eastern Asia
Western AfricaCentral Asia
Eastern AfricaMiddle Africa
South AmericaCaribbean
Southern AfricaCentral America
Source: UNODC 2014.
CHA
PTER
2W
hy C
are
abou
t Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n?
23
part of the world and in the grips of a social pandemic. Violence is widespread, and crime
rates are comparable with rates recorded in war-torn countries. With only 8 percent of
the world’s population, the region accounts for one-third of global homicides. Since 2010,
more than 10 percent of the population and over 30 percent of businesses have been
direct victims of crime (Jaitman 2015). An individual born in Caracas, San Salvador, or
Tegucigalpa—the three most violent large cities in the region—has about one chance in
eight of being murdered during his lifetime. Parts of the region also show a proliferation of
violent youth gangs, drug trafficking, money laundering, and domestic violence. The cost of
violence is significant; by some estimates, it reaches the equivalent of 3.5 percent, close the
total invesments made by the region on infrastructure (IDB 2017). In addition to the direct
human costs, the scale of crime and violence thwarts economic, social, and institutional
development and undermines political stability. Domestic and foreign investors are often
hesitant to do business in the most violent parts of the region.
The region has also endured long, violent, and destabilizing civil conflict. Since the
end of World War II, the region has recorded 22 conflicts lasting an average of seven years.
The civil conflict in El Salvador spanned more than a decade between the 1980s and 1992
(table 2.1). Although there are no official numbers, the United Nations estimates that more
than 75,000 people were killed. The Peruvian civil conflict between Maoist terrorist groups
and the government lasted two decades, from 1980 to 2000, and killed around 70,000
people. Colombia has been struggling with a 50-year domestic conflict, the longest armed
conflict ever recorded in the region, which has internally displaced more than 5.7 million
people, hindering growth and social development.
Major epidemics represent a serious risk to public health. Infectious diseases and
viruses thrive in warm and humid conditions such as those in the tropics, which span
most of the region. After a period of relative calm, recent years have seen an emergence
of epidemic threats (map 2.2). The outbreak of the Zika virus, which was first reported in
the Americas in 2014, has spread rapidly. As of June 2016, 39 countries and territories
reported confirmed cases of the virus, which can lead to a surge in microcephaly among
newborns.6 More than five million children are expected to be born in 2016 in areas
prone to Zika transmission in the region (Messina 2016). Another vector-borne virus,
Chikungunya, began spreading in the Americas in December 2013. More than 1.7 million
suspected cases in 45 countries and territories have been reported. Other infectious
diseases, such as dengue, dengue hemorrhagic fever, and equine encephalitis also
present a latent risk of recurrence in several areas. Large health shocks likely result in
human casualties, illness and disability, a rise health expenditures, lost labor income,
and disruption in economic activity.
6 “Geographic Distribution of Confirmed Autochthonous Cases of Zika Virus (Vector-Borne Transmission) in Countries and Territories of the Americas, 2015–2016,” Pan American Health Organization and World Health Organization, Washington, DC (data as of March 31, 2016), http://ais.paho.org/phip/viz/ed_zika_countrymap.asp.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
24
Aggregate shocks can reverse economic and social development
Natural and man-made shocks will continue to disrupt economic performance. Large,
sudden natural disasters such as earthquakes or hurricanes can cause great damage to
physical and human capital and may push communities, cities, or nations into slower
growth, even if only temporarily. The 2010 earthquake in Haiti is a striking example of
the impact these events can have on people’s livelihoods and economic circumstances:
over one-third of Haitians were severely affected; 2.5 percent of the population was
killed; and 80 percent of the country’s schools were destroyed. Growth shrank in the
aftermath of three major earthquakes that hit countries in the region: by 8.8 percentage
points in Haiti (2010), 1.1 percentage points in Chile (2005), and 1.2 percentage points in
Guatemala (2012) (figure 2.5). Similarly, in addition to the death toll and the destruction
of infrastructure, conflict imposes a large cost on economic output. Regression analysis
shows that relative to the historical performance, annual growth has been an average 2.8
percentage points less in countries in the region that have undergone a civil conflict in the
last 50 years (third colmun, table 2.2). Disasters affect the wider economy. Price inflation
skyrocketed in Peru (163 percent) and El Salvador (32 percent) during the bloodiest
periods of civil conflict (figure 2.6). High levels of crime and violence make investments
riskier and undermine the legitimacy of the state, weakening institutions and the rule of
law. Large chunks of public and private capital and the labor force are idle as economic
recession, disease outbreaks, conflict, or natural disasters unfold.
Map 2.2. Infectious Diseases and Viruses Thrive in the Tropics (areas at risk of Dengue in 2013)
Sources: World Health Organization, 2014.Notes: The contour lines of the January and July isotherms indicate areas at risk, defined by the geographical limits of the northern and southern hemispheres for year-round survival or Aedes Aegypti, the principal mosquito vector of dengue viruses. Areas in orange show parts where dengue was reported in 2013.
CHA
PTER
2W
hy C
are
abou
t Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n?
25
Human welfare losses are amplified in a riskier world. Natural disasters have been
associated with substantial damage to human capital. Nicaraguan children affected by
Hurricane Mitch in 1998, one of the most powerful hurricanes in the tropical Atlantic
basin in the 20th century, were four times more likely to become undernourished in
the aftermath of the disaster (Báez and Santos 2007). School attendance fell by almost
7 percent among children in households that were more heavily hit by two consecutive
earthquakes that struck El Salvador in 2001 (Santos 2007). Per capita expenditures,
including food consumption, fell by 8.2 percent in the urban areas that bore the burden,
in 2010, of the strongest tropical storm ever to strike Guatemala (Báez et al. 2016). The
droughts in 1997–98 and 2000 in Nicaragua increased by 10 percent the probability that
households would remain trapped in poverty (Premand and Vakis 2010).
Figure 2.5. Growth Slowed Down Following Major Earthquakes in Chile, Guatemala, and Haiti
Chile 2005
GD
P gr
owth
(%)
Mom
ent m
agni
tud
sacl
e
-8-7-6-5-4-3-2-10123456
4
5
6
7
8
2014201320122011201020092008200720062005200420032002
Haiti 2010 Guatemala 2012 MM
Source: World Bank calculations.
Table 2.2. Civil Conflict in the Region Have Thwarted Economic Growth
Incidence of conflict Conflict intensity
Conflict -0.0189***(0.004)
-0.0289***(0.005)
Minor conflict -0.0075*(0.004)
-0.0150***(0.006)
War -0.0604***(0.008)
-0.0662***(0.008)
Country fixed effects No Yes No Yes
Source: World Bank calculations.Notes: Results based on 858 (country-year) observations.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
26
Figure 2.7. Living on the Edge: The Riskier Environment of the Poor(households reporting impacts of large shocks during the 12 months preceding the survey)
0
10
20
30
Q5Q4Q3Q2Q1
Source: World Bank calculations.
a. Colombia
0
5
10
15
20
25
Q5Q4Q3Q2Q1
b. Peru
0
20
40
60
80
Q5Q4Q3Q2Q1
c. Haiti
0
1
2
3
Q5Q4Q3Q2Q1
d. Honduras
Figure 2.6. Civil Conflicts Also Disturb Other Economic Fundamentals Such as Inflation
Peru El Salvador
0
100
200
420
600
3,000
8,000
1999
1997
1995
1993
1991
1989
1987
1985
1983
1981
1979
1977
1975
-5%
0%
5%
35%
30%
25%
20%
15%
10%
1989
1987
1985
1983
1981
1997
1995
1993
1991
1979
1975
1977
1973
1971
1969
1967
1965
Source: World Bank calculations.Note: years in dark gray correspond to the periods with the largest number of casualties.
CHA
PTER
2W
hy C
are
abou
t Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n?
27
The negative impacts on household well-being are not confined to natural hazards and
can endure. Economic crisis inflicts a similar burden on affected populations. Evidence
on a sample of 59 countries shows that negative shocks on per capita GDP increase infant
mortality, especially among girls (Baird, Friedman, and Schady 2011). School-age cohorts
directly exposed to the Salvadoran civil war (1980–92) exhibited lower school attainment
(two fewer years of education) than slightly older or younger cohorts (Acosta et al. 2016).
The consequences on human welfare are also likely to persist. Mexican children withdrawn
from school because of aggregate and systemic shocks are about 30 percent less likely to
continue studying compared with children who stay in school (Sadoulet et al. 2004). The
civil conflict in Peru (1980–93), which caused about 70,000 deaths, reduced by 0.3 years
the school attainment of children who were exposed to conflict (León 2012). Likewise, the
cohorts that grew up during the civil war in El Salvador are today less likely to work than
counterfactual cohorts (Acosta et al. 2016).
Not everyone is equally sensitive to shocks; the poor are the most vulnerable.
Because of their physical scale, major shocks such as earthquakes, hurricanes, volcanic
eruptions, or civil wars often affect large geographical areas and vast populations.
Economic recessions are, by definition, national or global phenomena. Their impacts,
however, are not evenly allocated across the population: poor people show the greatest
vulnerability (figure 2.7). Calculations for this study indicate that, across the world, the
higher the GDP per capita, the lower the relative number of people affected by natural
disasters. Household surveys in four countries in Latin America and the Caribbean
show that low-income households, such as those in the bottom two quintiles of the
distribution, are more likely to report the occurrence of large shocks, including storms,
floods, landslides, droughts, crop plagues, and animal diseases. Empirical analysis of
the impact of a natural disaster on dwelling losses in Mexico between 2002 and 2005
was two times more likely among the poor than among the middle class (de la Fuente,
Ortiz-Juárez, and Rodríguez-Castelán 2017).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
28
CHA
PTER
3A
naly
tica
l Fra
mew
ork
29
3. Analytical Framework: The Link between Shocks and
Household Welfare
A simple conceptual framework helps in understanding the multiple ways shocks
affect household welfare. The framework conceptualizes in a simple manner the main
channels through which aggregate shocks can undermine the income-generating capacity
of households, hindering their ability to maximize welfare. The framework is comprised
of three main elements: (1) the capacity of households to earn fair returns through the
ownership and use of productive assets at certain points in time and dynamically; (2)
the inability of credit-constrained households to separate their consumption and income
generation decisions in a context of large, frequent uninsured risks; and (3) the implications
of insured risks on household behavior and decision making.
The income and consumption possibilities of households are largely determined by
the assets they possess, the intensity in which the assets are used, and the returns
to the assets. The framework follows an asset-based approach that illustrates the
dynamic decisions that households make in managing their resources to maximize income
and achieve an optimal consumption pattern.7 Households own assets that are used to
generate income, such as skills employed in the labor market, land utilized for agricultural
production, or properties rented out.8 Assets can be used more or less intensively, namely,
the rate of utilization (for instance, the employment rate for labor, or the depreciation of
capital). In addition to asset ownership and utilization, the income generation capacity of
households is affected by the returns to these productive assets: the real wages paid in the
labor market, the real interest rate awarded on savings, or the rent charged for property.
The income generation function of households is also influenced by transfers (public or
private), the prices of goods and services, and the probability of being affected by positive
or negative (idiosyncratic or aggregate) shocks. The ultimate welfare maximization among
households will also depend on the specific preferences of households, particularly in
consumption and leisure activities (figure 3.1).
7 On the asset-based approach, see Attanasio and Székely (2001); Bussolo and López-Calva (2014); Carter and Barrett (2006); Carter and Janzen (2015); Janzen, Carter, and Ikegami (2015); and López-Calva and Rodríguez-Castelán (2016).8 Household assets can be grouped into five broad categories: human capital (educational attainment, health, nutrition, skills and professional experience), physical capital (machinery, property, housing, and equipment), financial capital (savings, stocks and bonds), natural capital (land, soil, forests, and water), and social capital (networks and social trust).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
30
The accumulation of productive assets is a dynamic process that influences household
well-being. Households seek to maximize their welfare by adjusting consumption based
on the returns to the assets they own and the investments made or planned (López-Calva
and Rodríguez-Castelán 2016).9 The future level of productive assets depends on today’s
consumption decisions. Households can delay consumption through an increase in their
current asset base by, for instance, raising the educational attainment of household
members, strengthening nutrition, or acquiring physical capital.10 These investments
are expected to produce benefits later through higher productivity and, therefore, higher
incomes and better standards of living. Consequently, households will accumulate assets to
the point where the marginal benefit of present consumption and the flow of consumption
(in present value) derived from the stock of productive assets used in the future are equal.
Risks of shocks and shocks can aggravate poverty traps because households with
lower asset endowments are structurally more highly exposed and vulnerable to the
negative effects. Households with a high stock of productive assets are often more able to
move into higher consumption paths by taking up more profitable economic opportunities.
They are also more well endowed to withstand the effects of negative aggregate shocks
by taking out loans, selling assets, or receiving insurance compensation. The prospects
are different among vulnerable, credit-constrained households with traditionally lower
asset holdings. Even a relatively small shock can hinder the accumulation capacity of
these households, rendering them unable to accumulate the critical assets necessary
to enjoy higher standards of living. In the absence of full and well-functioning credit
and insurance markets, these households typically engage in economic activities and
9 While the emphasis in this framework is to maximize utility based on consumption, which is a function of assets, other models link assets more directly to livelihoods. See Scoones (1998) for such a sustainable livelihoods framework.10 Assets are fungible under this framework. It may thus be argued that poor people should spend their incomes on nutrients instead of savings, as this would provide higher potential assets in the future. This would still raise their productive assets in the future in terms of enhanced human capital.
Figure 3.1. Shocks Can Disrupt a Household’s Intertemporal Utility Maximization
Inter-temporalHousehold Utility
Maximization
HouseholdIncome
ProductiveAssets
Intensityof use
of Assets
Returns to Assets
Prices
AggregateShocks
TIME DISCOUNTINGPREFERENCES
= x x x
CHA
PTER
3A
naly
tica
l Fra
mew
ork
31
Box 3.1. Putting Faces to the Framework
The background
The cases of Isabel in rural Honduras and Alejandro in urban Peru illustrate the capacity of households to generate income based on available assets and the potential impact of aggregate shocks on welfare. The illustration focuses on three types of assets: physical, human, and financial capital.
Isabel has a small plot of land, where she grows modest amounts of produce to sell in the local market. She holds traditional rights over her land, respected for generations, but no formal deed. She completed elementary school and works for a moderate salary, a few months a year, in a small grocery store. Isabel keeps her cash savings at home. She hopes to save enough to open her own grocery store one day.
Alejandro went to school in Arequipa and finished secondary school. He has a job in a local textile manufacturing plant. Using some savings and money he inherited from his father, he bought a small apartment, where he lives and where he also rents a room to a work colleague. Alejandro manages to save part of his salary regularly in the bank.
Both individuals generate income from a different combination of productive assets, the utilization of these assets, and the real market value they thereby obtain. Alejandro has accumulated a higher stock of human capital; he uses it more intensively; and earns a higher return than Isabel, who uses her lower human capital to farm her own land and work at the store. In terms of physical capital, Isabel exploits her land and obtains a return by selling the produce at the market. If the legal property deeds were available, however, she could use the land more intensively, as collateral to access credit and start a business, diversifying her sources of income. Alejandro also obtains an inflow of income through his physical capital, the rent of the second room in the apartment. In terms of financial capital, Isabel does not use her cash savings intensively, thereby forgoing potential returns, unlike Alejandro who saves in the formal banking sector.
Table B3.1.1. The Story of Isabel and Alejandro
Capital Stock Intensity of use Returns
Isabel’s household
Physical capital Plot of land Medium: missing formal land titles; may not be used as collateral for credit
Moderate
Human capital Primary schooling
Low: works a few months a year Low
Financial capital Cash savings at home
Low Low
Alejandro’s household
Physical capital Apartment Medium: rents a room Medium
Human capital Secondary school
High: manufacturing plant; works each day Medium
Financial capital Savings in a bank
Medium Medium
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
32
Box 3.1. (continued)
The impacT of aggregaTe shocks on isabeL and aLeJandro’s weLfare
Isabel and Alejandro are exposed to idiosyncratic and systemic risks. In terms of the former, both individuals are exposed to the risk of illness. Isabel’s household, with a less diversified asset-stock and lower intensity of use, is more vulnerable to aggregate shocks.
Consider the occurrence of a severe weather event, such as a hurricane, that strikes Honduras. The shock could take a direct toll on Isabel’s stock of physical assets, both immediate and long-lasting, by flooding her land and leaving it unfit for farming. It could also impact her ability to utilize these assets indirectly if the hurricane affected infrastructure, such as the roads to take the produce to the market or to go to the grocery shop to work. To compensate for the loss in income and to smooth consumption, the household could even be forced to take a child out of school, affecting the accumulation of human capital of the next generation. Furthermore, the latent risk of hurricanes affecting the region and her asset base could be making Isabel lose out on higher income-generating opportunities, such as choosing to grow low-risk (low-profit) subsistence crops over higher-risk (higher-profit) cash crops.
Now consider the effect of a global economic recession, including a steep drop in the external demand of Peruvian manufactures. While an economic crisis would not have a direct effect on Alejandro’s stock of physical assets, it could likely impact asset utilization. Demand for labor could drop, leading to lower salaries in the manufacturing industry, making Alejandro decide to work fewer hours and depressing the returns to his human capital. The crisis could also affect his tenant’s income-generating capacity, which could lead to a decrease (or a loss) in Alejandro’s returns from renting the room in his apartment. Economic activity could decline enough to trigger a decline in interest rates, which would lower returns to his financial capital.
While both individuals would be affected by the shocks, Alejandro appears to be better equipped to deal with the consequences than Isabel. Alejandro uses his assets more productively, which provides the opportunity to have more savings to cope with the effects of the crisis and thus smooth his consumption pattern. His link to the financial system—his current savings account—would make it easier for him, relative to Isabel, to take out a loan or access other forms of credit to smooth consumption. His formal job at a manufacturing plant may provide more job security, at least in the short run, than farming in the face of a natural disaster. By virtue of living in an urban setting, even if he is subjected to a weather-related shock, it is likely that Alejandro would encounter fewer difficulties in accessing infrastructure, because damage and service interruptions are likely to be addressed more quickly in urban areas than in rural areas.
This illustration sketches out some of the possible impacts of aggregate shocks on household welfare. To study the full impact on household income generation capacity and consumption would require a more sophisticated analysis of the effects of shocks on assets, intensity of use, and returns, as well as the interaction of these effects and their duration.
CHA
PTER
3A
naly
tica
l Fra
mew
ork
33
investments to reduce risk exposure rather than to maximize profits.11 However, ex ante
risk mitigation through low-return, low-risk production and employment choices leads to
large inefficiencies and contributes to the miring of households along low-income paths,
which implies deepening inequities. For example, risk-prone, low-income farmers are
often reluctant to invest in fertilizers, even though these would boost the expected returns,
because they prefer to retain savings as a cushion to deal with bad weather or other shocks.
Similarly, aggregate shocks can impact household welfare through the negative distorting
effects on markets, prices, and the allocation of resources (box 3.1).
Income uncertainty can reinforce poverty traps through a variety of behavioral
mechanisms that affect household decision making. In addition to undermining the
capacity of households to generate income, uninsured risks also influence household
preferences in consumption, leisure activities, and time use in ways that weaken asset
accumulation and the ability of households to rise out of poverty. A behavioral explanation
of poverty traps involves the issue of lack of self-control, namely, the inability of individuals
to follow through on an envisioned plan such as saving and reinvestment. Recent theories
show that a lack of self-control can lead to poverty traps and that market imperfections
and lack of commitment devices (such as lockbox accounts designed to promote savings),
both of which are aggravated during serious shocks, reinforce the preference for temptation
goods (Banerjee and Mullainathan 2010).12 A related strand in the literature indicates that
the burden of scarcity also influences how poor individuals allocate attention, pushing
them to devote more mental effort to solving some problems, while neglecting others that
may more directly lock them in poverty (Shah, Mullainathan, and Shafir 2012). Information
biases can also shape behavior in ways that make poverty more persistent. Lacking accurate
knowledge about risks and effective ways to deal with them may adversely encourage
households to make decisions that increase their exposure and vulnerability to shocks
(Dupas 2011). Finally, a high-risk context can accentuate behavior that leads to aspirational
failures, defined as the failure to aspire to one’s own potential (Dalton, Ghosal, and Mani
2016).13 In sum, aggregate shocks and exposure to risks could also affect household welfare
by affecting preferences and time discounting in an inefficient way (see figure 3.1).
11 This is also known as the nonseparable household model. See de Janvry and Sadoulet (2006); López (1984, 1986); Singh, Squire, and Strauss (1986).12 Bernheim, Debraj, and Yeltekin (2013) explore how the ability to exert self-control (such as to save more or borrow less) relates to economic circumstances. Using a model of intertemporal allocation with an imperfect credit market, they find that poverty in itself can be harmful to self-control. In a similar line, Banerjee and Mullainathan (2010) look at the link between the consumption of households and temptation goods in the context of shortsightedness and time-inconsistent preferences. Their framework explores whether the impact of temptation declines with income.13 Recent empirical studies focus on behavioral effects in the presence of risk. Using data from Kenya, Lybbert and McPeak (2012) calculate the risk aversion and intertemporal preferences of herders in the northern region. Their results indicate that poorer households are more risk averse, but also more ready to disrupt consumption than wealthier households. This is in line with the notion that, at certain levels of wealth, asset smoothing (over consumption smoothing) is an equilibrium behavior. This is confirmed in Carter and Lybbert’s (2012) study, which uses data from Burkina Faso. Rural households below an estimated threshold (a critical herd size) do not insulate their consumption, while, for the most part, those above the threshold do insulate their consumption.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
34
The conceptual framework outlined above provides an opportunity to organize the
discussion that follows. The next chapters synthesize the available evidence that links
aggregate shocks to household welfare, as well as the evidence on the two main channels
of transmission through which shocks affect welfare: the accumulation of assets and
decisions to utilize productive assets based on the returns observed in the market. Starting
with the main outcomes of interest, chapter 4 examines the responsiveness of household
income and consumption to the variety of effects arising from aggregate shocks, placing
special interest on the likely increase in poverty vulnerability among affected households.
Chapter 5 focuses on the disruption of household asset ownership and accumulation that
is caused by large shocks. Chapter 6 pays attention to the other broad elements of the
framework, namely, the influence of aggregate shocks on the ability of households to put
their assets to good use and earn a fair return, as well as the effects of the nonseparability
of production and consumption decisions in the context of pervasive market failures.
CHA
PTER
4A
ggre
gate
Sho
cks
Thre
aten
Pro
gres
s in
Hou
seho
ld W
elfa
re
35
4. Aggregate Shocks Threaten Progress in Household Welfare
Household welfare—linked to the affordability of goods and services, the opportunities
for human and physical capital investment, and sustainable economic expansion—has
progressed in Latin America and the Caribbean. People require an adequate minimum
level of income or consumption to meet basic needs in food, clothing, shelter, and health
care. In addition to this monetary criterion, well-being is also associated with satisfactory
levels of health and education, access to basic services such as clean water, sanitation, and
electricity, and physical security and voice to enable the capabilities and opportunities of
the individual to function in society (Sen 1997). Countries across the region have recorded
remarkable progress in their command of resources and their enhanced capacities and
economic choices. Income per capita grew an average of 2 percent a year between
2000 and 2014. Around 80 million people escaped poverty over the same period, and
households—including those at the bottom of the distribution—are now endowed with
sufficient human and physical capital to participate effectively in society and unleash their
economic potential (World Bank 2015). Notwithstanding the greater prevalence of social
assistance, labor income is the main driver of progress among most of the population in
the region.14
However, a large share of households continue to face substantial income and
consumption risks, many of which are tied to large natural and man-made hazards.
Despite substantial growth in living standards at the bottom end of the income distribution,
the median household is barely nonpoor and hence still vulnerable to becoming poor.
Aggregate shocks place a huge burden on the incomes of households, especially those that
are poor and vulnerable. They may be unable to maintain a stable level of consumption
once large negative shocks occur and incomes fall. Markets either do not exist or function
imperfectly so that households may no longer be able to protect themselves against the
risks by relying on insurance. Households may lack liquid assets or other formal strategies
to cope. Some informal risk-sharing mechanisms such as private transfers and community-
based financial systems are useful, but empirical evidence indicates that they shield
consumption only partially. Local risk-pooling arrangements are prone to collapse in the
face of spatially covariant risk. Despite all the formal and informal risk management and
14 According to household data on 17 countries in Latin America, labor income accounts for 60–80 percent of total income among households in the bottom 40 percent of the distribution, while the corresponding share is even higher among households in the top 60 (Cord, Genoni, and Rodríguez-Castelán 2015).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
36
coping strategies available to households, income shocks still count in the determination
of consumption, especially among vulnerable, low-income families, sometimes trapping
them in persistent poverty.
Aggregate shocks often signify income shocks
The first manifestation of a major negative shock at the macroeconomic level is a
drop in output. Financial crises, earthquakes, floods, droughts, and civil conflicts disrupt
the routine functioning of economic systems and institutions. Macroeconomic crises are
often characterized and even defined by the losses they generate in the main economic
aggregates such as GDP growth, investment, and employment. The Latin America and
Caribbean region has experienced steep declines in GDP. For instance, GDP fell by 6.0
percent in Mexico in 1995, while the drop was 11.0 percent in Argentina in 2002, and
this also had a profound impact on the economy of Uruguay, where GDP shrank by 17.5
percent. Wars and civil conflict inflict a similar cost. In a global context, World War II
has historically been considered the most serious economic shock: declines in GDP in
the aftermath of the war averaged 34 percent in 22 countries on which reliable data are
available (Barro and Ursúa 2008). The Latin America and Caribbean region has not been
spared from this type of impact. The GDP of Nicaragua collapsed by 6 percent during the
most violent period of the revolutionary war (1962–90). The potential economic effect of
the H1N1 flu outbreak in Mexico in April 2009 is estimated at 2.2 percent of GDP (World
Bank 2009). Sometimes, the effects of shocks become more manifest in local economies.
Two major back-to-back earthquakes that struck Colombia’s coffee-producing departments
in 1999 destroyed over 8,000 farms and heavily affected the coffee industry, a strategic
economic sector of the area.
Household incomes show hefty declines after major natural disasters, particularly
in agriculture. Weather shocks are the most important source of income volatility in
developing countries, chiefly in rainfed agriculture, which is prevalent in almost three-
quarters of the total area allocated to agriculture in the world.15 Income risk because of
extreme weather is equally common in Latin America and the Caribbean. Data collected in
Honduras shortly after Hurricane Mitch show that 1 household in 3 incurred a loss of crops
that translated into a loss in income, mostly among poor households, and 1 household in
10 lost wages or business income (Morris et al. 2001). These examples are consistent with
other evidence on the effects of shocks on income. Thus, hurricanes in Central America
have cut household incomes by 3 percent for each standard deviation in the intensity of
the hurricane windstorm (Ishizawa and Miranda 2016). Similarly, individuals in the region
who were in utero when major floods occurred in the areas where their mothers were
living eventually earned 12.6 percent less income as adults (Caruso 2017).
15 Data of FAOSTAT (database), Food and Agriculture Organization of the United Nations, Rome, http://faostat3.fao.org/.
CHA
PTER
4A
ggre
gate
Sho
cks
Thre
aten
Pro
gres
s in
Hou
seho
ld W
elfa
re
37
Household incomes collapse in the aftermath of earthquakes, volcanic eruptions,
tsunamis, and avalanches. The multiple effects of earthquakes translate into sizable
declines in income. El Salvador was struck by two consecutive strong earthquakes (7.7
and 6.6 on the Richter scale) in 2001. Detailed data on the seismic parameters of the
earthquakes (depth and energy released), the geographical coordinates of the dwellings
damaged, and the soil types of the villages allow measurements of the intensity of the
shocks among individual households. Empirical analysis of microdata reveals that the
total effect of both events led to an average income reduction by almost a third relative
to the median preshock per capita income of households in the upper half of the ground-
shaking distribution (Báez and Santos 2009). An earthquake of like magnitude that hit the
central coast of Peru in 2007 was found to have similar impacts (Lucchetti 2011).
Civil conflict jeopardizes short-term income. Civil conflict disrupts market transactions
and economic activity immediately. Income among entrepreneurs and shop owners
dropped by 47 percent and 59 percent, respectively, after the 2007–08 election crisis in
Kenya and the ensuing massive killings (Dupas and Robinson 2012). The civil war in Peru
led to an earnings decline of 5 percent for every standard deviation of early-life exposure
(at the district level) to conflict-related violence (Galdo 2013). Municipalities with higher
levels of drug-related crime in Mexico in 2007 observed a decline in annualized household
income growth of 0.2 percentage points between 2005 and 2010 (Enamorado, López-
Calva, and Rodríguez-Castelán 2014).
Other types of shocks such as economic crises also heavily affect household incomes
and expenditures. Average household labor earnings fell by 37 percent and 30 percent,
respectively, in the year following the economic crisis in Argentina (2001–02) and in the
Dominican Republic (2003–04). Mostly because of the 2001–02 crisis, the share of people
living below the poverty line doubled in Argentina in only two years, and it took almost
10 years for the country to rebound to the precrisis poverty rates (Barriga Cabanillas et
al. 2015). Mean household expenditures in Mexico fell by Mex$15 between 1994 and 1996
during the peso crisis. Although households were unable to protect their consumption
during the crisis, there is evidence that they altered the composition of consumption by
reducing spending on nonessentials and durable goods to allocate a higher share of their
budgets to basic food items (McKenzie 2006).
The effects of aggregate shocks on incomes can persist
Some households improve quickly following shocks, but recovery may take much
longer among other households, particularly the poor and vulnerable. Access to
well-functioning credit and insurance markets, savings, and other liquid assets position
households to withstand the negative income effects of major shocks. This is often not the
case among the poor and vulnerable, who may be obliged to undertake a shift to a lower
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
38
equilibrium of welfare and income growth.16 In Honduras, two and half years after Hurricane
Mitch (1998) had ravaged Central America, evidence shows that wealthier households were
able to bounce back quickly, but less well off households went on a downward trajectory
of sustained asset-depletion and low income growth (Carter et al. 2005). An increase of
about 10 homicides per 100,000 inhabitants doubles the probability that a municipality
in Mexico remains chronically poor 5 to 10 years later (Martinez-Cruz and Rodríguez-
Castelán 2016). In Colombia and Mexico (in 2010–13 and 2002–05, respectively), upward
economic mobility was less prevalent among households affected by large shocks. Almost
half the unaffected households were able to move out of the bottom 40 percent of the
distribution, almost twice as large as the share of shock-affected households that showed
upward mobility (figure 4.1).
The effect of major hazards such as natural disasters on short- and long-term
economic growth is less straightforward. Growth theory does not offer a clear-cut
answer on the question of whether natural disasters should affect economic growth.
Traditional neoclassical growth models predict that the destruction of capital does
not affect the rate of technological progress, and it might thus only enhance short-
term growth prospects. Endogenous growth models based on Schumpeter’s creative
16 A large body of evidence indicates that the effect of aggregate shocks on the incomes of the bottom of the income distribution is larger, but the potential influence on income inequality is depends mainly on the type of shock and the impact on particular sectors of the economy. Thus, part of the recent narrowing in income inequality in Latin America has been driven by a compression in earnings structure, especially in Mexico and Central America during the global financial crisis of 2008–09. The crisis had a significant negative effect on high-skilled workers through spells of unemployment and lower real earnings (Cord et al. 2017).
Figure 4.1. Aggregate Shocks Reduce Upward Economic Mobility
0%
20%
40%
60%
80%
100%
Shock 2002and 2005
Shock 2005No shocks0%
20%
40%
60%
80%
100%
Shock 2010 and 2013Shock 2013No shocks
Source: Data of the Colombian Longitudinal Survey of Wealth, Income, Labor, and Land and the Mexico Family Life Survey.Notes: Based on probabilities from transition matrices of 2002–05. Stayers correspond to households that remain in the bottom 40, and climbers to households transitioning from the bottom 40 to the top 60.
a. Mexico b. Colombia
Climbers Stayers Climbers Stayers
CHA
PTER
4A
ggre
gate
Sho
cks
Thre
aten
Pro
gres
s in
Hou
seho
ld W
elfa
re
39
destruction process may even ascribe higher growth as a result of negative shocks
because these shocks can be catalysts for reinvestment and the upgrading of capital
goods. AK-type endogenous growth models in which technology exhibits constant
returns to capital predict no change in the growth rate following a negative capital
shock; meanwhile, endogenous growth models that exploit increasing returns to scale in
production generally predict that a destruction of part of the physical or human capital
stock results in a lower growth path. However, empirical analysis on 196 countries over
1970–2008 suggests that GDP growth in the short- and long-run is rarely affected by
major floods, storms, or earthquakes (Cavallo et al. 2010).17
Income shocks translate into consumption shortfalls and greater vulnerability to poverty
Vulnerable households manage to protect their consumption only partially from
the negative income shocks caused by major hazards. Households are resourceful in
handling the effects of negative shocks. If they lack full access to credit and insurance,
they rely on informal mechanisms ranging from income smoothing to self-insurance and to
social insurance arrangements such as borrowing from friends and family. Despite these
strategies, a good deal of evidence shows that households, particularly the less well off,
are unable to allocate risk optimally across space (for example, mutual insurance within
villages) and income cycles. Several studies show a strong relationship between changes
in household consumption and measures of aggregate risk in villages such as rainfall
variability (Alderman and Paxson 1994; Báez, de la Fuente, and Santos 2010). Empirical
analysis of consumption growth in rural villages in southern India, Mexico, and Thailand
confirms the existence of a significant amount of risk-sharing, but rejects the hypothesis of
full insurance (García-Verdú 2002; Paxson 1992; Townsend 1994).18
Incomplete insurance means that consumption falls in the aftermath of aggregate
shocks. Drought-affected households in Burkina Faso and other countries in Sub-Saharan
Africa reduce their consumption relative to those not affected. The evidence points to little
or no risk-sharing, as well as little consumption smoothing (Kazianga and Udry 2004).
While households attempted to offset the fall in income by selling livestock, these sales
covered only 20–30 percent of the shortfall in crop income (Fafchamps, Udry, and Czukas
1998). In Ethiopia, a 10 percent reduction in rainfall resulted in a 5 percent decline in
food consumption (Dercon 2004). In addition, a 10 percent reduction in rainfall reduced
future consumption growth by 1 percentage point. Households that were obliged to move
17 A recent study has also shown the potential trade-offs in aggregate welfare of positive trade shocks in Mexico. While increasing real earnings among the less highly skilled, a positive shock that fostered growth in export manufacturing in Mexico in 1986–2000 also raised the opportunity costs of schooling, which then raised drop-out rates at the margin (Atkin 2016).18 Underlying the relationship between individual income and consumption is the high covariance between income sources and asset values that is distinctive of aggregate shocks (Báez, de la Fuente, and Santos 2010).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
40
from conflict zones in Colombia suffered a 22 percent decline in total per adult equivalent
consumption, reducing their food and calorie intake (Ibáñez and Moya 2006).19
Urban households also have limited scope to smooth consumption. It is often argued
that individuals in urban areas are more effective at protecting their consumption
against income risk. The literature suggests, however, that this is not always the case. In
Guatemala, a range of risks are prevalent throughout the territory, affecting rural areas
as well as the main cities. Empirical analysis of the effects of Agatha, a major tropical
storm that hit the country in 2010 and dropped the largest amount of rainfall on the
country since 1963, shows that per capita consumption fell in the affected areas by an
average of 7.7 percent relative to the median consumption of the control group at baseline
(figure 4.2). Yet, further inspection of the data indicates that the drop in consumption was
concentrated in urban centers, where the storm happened to have dropped the largest
amount of precipitation, and, consequently, the affected households were unable to offset
the fall in income (Báez et al. 2016). Similarly, longitudinal data on urban households in
Colombia reveal that households that experienced major adverse shocks were more likely
to see their consumption fall, which pushed some of the affected households from the
bottom half of the distribution into a poverty trap (Fergusson and Zambrano 2016).
Transient impacts on consumption and growth can become chronic. This implies that
the welfare losses are well beyond the cost capacity of lower consumption in the short
term. Most of the available evidence on this subject is derived from studies on Africa
and South Asia. One of these studies investigates the determinants of food consumption
growth in rural Ethiopia between 1989 and 1997, when most households were dependent
on rainfed agriculture, and droughts were commonplace. The results show that a drop by
10 percent in historical rainfall reduces consumption growth by 1 percentage point four
to five years later. Furthermore, the households most affected by the large-scale famine
crisis of the 1980s endured 16 percent lower consumption growth in the following decade
(Dercon 2004). In Tanzania, farmers who experienced crop failures induced by severe
droughts between 1991 and 1995 suffered consumption growth losses ranging from 17
percent to 40 percent 10 years later (Beegle, Dehejia, and Gatti 2006). The evidence on
Latin America and the Caribbean is thinner, but points in a similar direction. Rural Brazilian
workers affected by strong rainfall shocks migrate permanently to urban areas where they
earn lower wages and can probably afford only lower consumption following the depletion
of their productive assets (Mueller and Osgood 2009). Panel data on rural families in El
Salvador between 1995 and 2001 also show low income and consumption growth among
less well off households that experienced substantial income shocks (Rodríguez-Meza and
González-Vega 2004).
19 This population was found to allocate 65 percent of their total consumption to food expenditures.
CHA
PTER
4A
ggre
gate
Sho
cks
Thre
aten
Pro
gres
s in
Hou
seho
ld W
elfa
re
41
Consumption volatility means that vulnerable households may be a disaster away
from falling below the poverty line or sliding further into poverty. Although households
may adopt various risk management and coping strategies, the growth rate of household
consumption is largely determined by the growth rate of household income. From a policy
perspective, the lack of full insurance is particularly relevant among households that are
below or slightly above the level of income or consumption deemed necessary to afford
minimal standards of food, clothing, health care, and shelter. Translating this into numbers
for Latin America and the Caribbean means that more than half the total population faces
substantial consumption risk: as of 2014, around 24 percent of the people are poor,
whereas 39 percent are vulnerable (a group located between the poor and the middle
class that faces high levels of economic insecurity).20 The vulnerable encompass the modal
(typical) household in the region, that is, the household with the income that occurs with
the highest frequency in the distribution (World Bank 2013b).
The inability of households to insure their consumption effectively against large
shocks increases their vulnerability to poverty. Some studies have investigated whether
the fall in income or consumption caused by major shocks pushes some households
into poverty or deepens the deprivation among households that were already poor.
This was the case among a subset of the households affected by tropical storm Agatha
in Guatemala (2010). The drop in consumption put these households below the poverty
20 This refers to households with a 10 percent probability or higher of experiencing spells of poverty over a five-year interval (World Bank 2013b).
Figure 4.2. In the Wrong Place at the Wrong Time: Household Consumption Fell after Tropical Storm Agatha, Guatemala, 2010
-0.2
0.0
0.2
0.4
0.6
10987654
Source: Báez et al. 2016.
a. Before the shock (2006) b. After the shock (2011)
Control Treatment
Aggregate Consumption Aggregate Consumption
Den
sity
-0.2
0.0
0.2
0.4
0.6
10987654
Den
sity
Control Treatment
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
42
threshold, increasing the overall poverty rate among urban households in the sample by
5.5 percentage points, equivalent to roughly 80,000 more families in poverty (figure 4.3).
Likewise, the 2007 earthquake in the Ica Region of Peru increased objective and subjective
poverty by roughly 4 percent and 15 percent, respectively, among households within 80
kilometers of the epicenter (Lucchetti 2011). A series of droughts and floods in Mexico
increased poverty by between 1.5 percent and 3.7 percent in the affected municipalities in
2000–05 (Rodríguez-Oreggia et al. 2013). The impact of economic stagnation on poverty
is noticeable. In 2003–04, the Dominican Republic experienced a bank collapse, rapid
currency depreciation, and inflation, which hastened a domestic crisis that caused the
economy to contract. Poverty soared from 32 percent in 2002 to over 50 percent at the
peak of the crisis (World Bank 2014b) (figure 4.4).
Despite the region’s notable progress on many fronts, the risk of large shocks
threatens the recent income and consumption gains. The next two chapters expand on
these notions, unpacking the channels of transmission from shocks to household welfare
through three main drivers: asset ownership, asset utilization, and returns to assets.
Natural shocks can damage or destroy physical and natural capital, posing another layer
of disruption on households that are either already living in poverty or living on the edge
of poverty. Labor, the main source of income across the region, may behave adversely in
response to shocks. Households may lower their labor supply because of large disruptions
that cause negative income effects or increase it if the shock causes consumption to fall
(the substitution effect). In the absence of social insurance, children may join the labor
Figure 4.3. Poverty Increased among Households Affected by Tropical Storm Agatha, Guatemala, 2010
Parameter estimate
Perc
ent
95% Conf. Interval
Source: Báez et al. 2016.Notes: Parameter estimates of a double difference analysis of the effect of the shock on household poverty. The dummy captures the effect using a binary treatment, while the other parameters are obtained from a treatment intensity specification. The 95 percent confidence intervals are shown.
-0.05
0.00
0.05
0.10
0.15
z-score <= 53 < z-score <= 52 < z-score <= 5Dummy
CHA
PTER
4A
ggre
gate
Sho
cks
Thre
aten
Pro
gres
s in
Hou
seho
ld W
elfa
re
43
Figure 4.4. Substantial Reversals in Poverty Reduction Are Likely to Arise during an Economic Crisis
a. Colombia
b. Argentina
GDP growth
GD
P gr
owth
(%)
Pove
rty
-8
-6
-4
-2
0
2
4
6
4
10
5
20
15
25
35
30
200
5
200
4
200
3
200
2
200
1
200
0
1999
1998
1997
1996
1995
1994
1992
1993
1991
1990
Poverty
GDP growth
GD
P gr
owth
(%)
Pove
rty
-15
-10
-5
0
5
10
15
10
20
15
30
25
35
50
45
40
35
30
200
5
200
4
200
3
200
2
200
1
200
0
1999
1998
1997
1996
1995
1994
1992
1993
1991
1990
Poverty
c. Dominican Republic
GDP growth
GD
P gr
owth
(%)
Pove
rty
-4
-2
0
2
4
6
8
10
0
10
30
20
40
60
50
200
5
200
4
200
3
200
2
200
0
200
1
2010
200
9
200
8
200
7
200
6
2013
2012
2011
Poverty
Source: World Bank calculations.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
44
force because of aggregate shocks if more income is needed during tough times or if the
opportunity cost of sending the children to school becomes too high during good times.
Tough times caused by major crises often dampen the returns to household and firm assets,
and risk distorts incentives in ways that may lead to a series of nonoptimal decisions,
thereby ultimately dampening household welfare.
CHA
PTER
5Th
e Vu
lner
abili
ty o
f Ass
ets
and
Live
lihoo
ds to
Agg
rega
te S
hock
s
45
5. The Vulnerability of Assets and Livelihoods to Aggregate
Shocks
Progress in the region has translated into remarkable welfare-enhancing asset accumulation
Recent progress in inclusive growth in Latin America and the Caribbean has been
paired with the accumulation of physical, human, and financial capital. A wider fiscal
space has allowed an expansion of public investment in infrastructure, basic services, and
social programs. The share of adults with more than six years of schooling rose from 56
percent to 66 percent between 2003 and 2013. Life expectancy increased from 67.0 years
in 1990 to 74.6 years in 2014, and the under-5 mortality rate fell by nearly half between
2000 and 2013. Investments in physical infrastructure enhanced access to critical services
such as electricity. In 2014, 86.1 percent of rural households were connected to the grid,
up from nearly two-thirds in 2000.21 By 2010, 18 countries in the region had been able
to implement conditional cash transfer schemes, which represented the equivalent of an
average 0.4 percent of GDP and benefited almost a quarter of the population in the region,
while also contributing to objectives in poverty reduction and asset building (Paes-Sousa,
Regalia, and Stampini 2013).
This progress was particularly beneficial for poor households and low-skilled
workers. The economic advances helped poorer households. Access to basic services
expanded significantly among households at the bottom of the distribution. These
households were able to purchase a wide range of durable goods, including refrigerators,
stoves, televisions, bicycles, and motorcycles. Mobile phone ownership also increased,
averaging 80 percent throughout the region and reaching almost 100 percent in some
countries (figure 5.1). Although access to the Internet is much lower across the region,
data suggest that coverage rates among the bottom 40 in Brazil and Chile rose from
almost zero in 2000 to 21 percent and 25 percent in 2012, respectively (Cord, Genoni,
and Rodríguez-Castelán 2015). Asset accumulation brought about improvements in the
quality of life, but also enhanced the opportunities available to low-income individuals.
21 WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/data-catalog/world-development-indicators.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
46Thus, access to electricity can help improve educational attainment; air conditioning
can lower heat-related mortality; refrigeration can improve child health outcomes;
stoves (gas or electric) can reduce indoor air pollution; durables can strengthen the
financial position of households that are able to use the durables as collateral; and
mobile phones and the Internet can facilitate access to crucial information, from pricing
on crop markets to weather alerts.
Yet, progress in asset accumulation is at risk because of the effects of aggregate shocks
Recurring shocks or shocks that are more severe interrupt positive trends and may
reduce the stock of accumulated assets. Natural disasters, widespread violence, civil
conflict, or economic crises have a negative effect on the stock of assets owned by
individuals and households. These phenomena disrupt human lives and destroy human
capital (by weakening nutrition, health, and school participation), private property
(housing, machinery, crops and livestock), critical public infrastructure (roads,
bridges, schools, and health facilities), and natural capital (through environmental
degradation). Similarly, droughts, floods, storms, and plant diseases wipe out crops,
while civil conflict erodes social cohesion, trust, and stability. In addition to the
rather obvious negative effects on aggregate output, economic crises disrupt financial
markets (thus reducing the availability of capital), discourage labor supply, and thwart
productive skills.
Human capital is sensitive to the direct effects of major shocks. Natural disasters,
civil conflict, and widespread epidemics result in human casualties, leaving a wake
Figure 5.1. A Remarkable Increase in Asset Ownership among the Bottom 40
Air conditioning
Porc
enta
ge o
f hou
seho
lds
26
25
76
10
43
5 8
8491
14
60
0
20
40
60
80
100
43
60
04
76
11
91
11
25
84
6 82 5
20142000
Source: Tabulations of Equity Lab, Team for Statistical Development, World Bank, Washington, DC, based on data in SEDLAC (Socio-Economic Database for Latin America and the Caribbean).
Car Cellphone Electricity Motorcycle Refrigerator
CHA
PTER
5Th
e Vu
lner
abili
ty o
f Ass
ets
and
Live
lihoo
ds to
Agg
rega
te S
hock
s
47
of negative human capital impacts. The 2010 earthquake that struck Port-au-Prince,
Haiti’s densely populated capital, caused between 200,000 and 250,000 fatalities.
Colombia’s internal conflict, lasting for over 50 years, has claimed a similar number of
lives, most of them civilian noncombatants. Hurricane Mitch accounted for over 14,000
deaths across four countries in Central America. In addition to human casualties, these
events often result in the destruction of the infrastructure required for the acquisition
of human capital, such as schools, hospitals, and clinics. Major shocks may pose a
financial burden on households, often forcing them to cut back their expenditures on
food and health care, thereby raising the risk of malnutrition and other negative health
impacts. Economic crisis also erodes human capital through the sharp rise in the rate
and duration of unemployment and the subsequent depreciation of the human capital
of workers.22
Shocks destroy other forms of capital such as productive agricultural assets, durable
goods, and basic public infrastructure. Droughts and floods typically destroy key
farming inputs, leading to declines in crop production that compromise food security
and other productive and consumption goals. Hurricane Mitch (1998) decimated over
80,000 hectares of agricultural land, the large majority of which was used by small-
scale farmers for subsistence agriculture (Ishizawa and Miranda 2016). As the severity
of extreme weather or other comparable shocks increases, households are more likely
to deplete livestock, including livestock that contributes to farm production, such as
transport, draft, and breeding animals. After Hurricane Mitch hit Honduras, poorer
households lost close to 20 percent of their productive assets (land, livestock, and
plantations), thereby compromising their capacity to generate earnings and livelihoods;
in contrast, higher-income households lost a lower share of assets (7.5 percent) and
recovered more quickly (Carter et al. 2005).23 The effects are also evident in other types
of assets. Peruvian households affected by large shocks are less likely to own durables
such as refrigerators or cars (figure 5.2). Physical infrastructure that is central to the
provision of basic services is habitually destroyed or damaged. Data on Haiti in 2012
show that households hit by major shocks are less likely to have access to electricity.
Because assets and infrastructure are destroyed, the negative effects tend to linger well
into the medium and long term.
22 Evidence based on Swedish data shows that a full year of unemployment led to a 5 percentile downward shift in the distribution of skills, including in the ability of adults to read and make practical use of printed information (Edin and Gustavsson 2008).23 The long-run effect of weather shocks on household productive assets depends on the household’s sensitivity to and resilience in the face of the shocks. Sensitivity and resilience are likely to depend on the household’s wealth prior to the shock and on the household’s access to employment and capital, as mediated by either market or social mechanisms.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
48
Social capital is a risk-sharing mechanism commonly used by households.24 Mutual
interactions through norms and networks (for instance, kinship or friendship) that support
each other based on principles of trust, cooperation, and reciprocity are a key instrument
in risk-sharing and mitigate the effects of aggregate shocks. Indeed, private transfers of
food, clothing, and cash are common in the aftermath of a shock in developing countries.
Recovery after Hurricane Mitch was less challenging among poor rural household in
24 Social capital is thus a set of customary principles that enable collective action. Putnam (1995, 67) defines it as the “features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit.”
Figure 5.3. Individuals Prone to Aggregate Shocks Have Stronger Ties to Social Capital, Peru
Indi
vidu
als
that
par
tici
pate
in s
ocia
l com
mun
itie
s (%
)
13%
29%
0
5
10
15
20
25
30
35
Affected by a shockNot affected by a shock
Source: Calculations based on the 2012 National Household Survey (Encuesta Nacional de Hogares sobre Condiciones de Vida y Pobreza); tabulations of Equity Lab, Team for Statistical Development, World Bank, Washington, DC, based on data in the SEDLAC database.
Figure 5.2. Ownership of Durable Goods Is Lower among Households Hit by Aggregate Shocks, Peru, 2012
49%
10%
11%
4%
0
10
20
30
40
50
60
Affected by a shockNot affected by a shock0
2
4
6
8
10
12
Affected by a shockNot affected by a shock
Source: Calculations based on the 2012 National Household Survey (Encuesta Nacional de Hogares sobre Condiciones de Vida y Pobreza).
Hou
seho
lds
that
hav
e a
refr
iger
ator
(%)
Hou
seho
lds
that
ow
n a
car
(%)
CHA
PTER
5Th
e Vu
lner
abili
ty o
f Ass
ets
and
Live
lihoo
ds to
Agg
rega
te S
hock
s
49
Honduras that relied upon the support of their communities for financial aid (Carter and
Castillo 2004). In the same way, interhousehold transfers among Nicaraguan households
that were affected by the same hurricane acted as an informal insurance mechanism,
benefiting mostly those who had experienced the largest economic losses (Santos 2006).
Aware of the mutual benefits of private transfers, households in Peru that are more prone
to being affected by a shock engage more actively in community activities focusing on the
application of social capital (figure 5.3).
Yet, social capital can evaporate during major shocks, especially among the poor.
Shocks may affect the entire social fabric, disrupting links that might have helped
households endure in the aftermath. Civil wars, other forms of widespread violence and
destruction, and the subsequent involuntary displacement of large groups of people alter
networks and tear apart ties developed through mutually supportive community relations.
Over three million people have been displaced internally in Colombia in the last 50 years
because of incursions by armed groups. The social capital of the poor tends to be weaker.
Detailed analysis of networks for mutual support in rural Tanzania show that wealth is a
strong determinant of network formation. Consequently, the poor have less dense networks
relative to the more well off (De Weerdt 2004).
Aggregate shocks disrupt and constrain asset accumulation
Human capital formation, a process that takes place over years, is deeply disrupted
by large shocks. The impacts of shocks may extend beyond the direct damage to the stock
of assets; they also undermine future investment in assets. Extreme weather, civil strife,
price spikes, and even economic crisis often lead to reduced investment in food, health
care, and other critical family inputs among children. Evidence from rural Colombia in
1999–2008 shows that fetal exposure to temperature shocks during the first trimester
of pregnancy reduces the average birthweight of infants by about 4.1 grams and has
negative effects on Apgar tests (Andalón et al. 2016). These two indicators are strong
predictors of socioeconomic outcomes during adult life, such as labor earnings (Behrman
and Rosenzweig 2004). If they showed signs of illness, Nicaraguan children ages 0–5 in
households located in the path of Hurricane Mitch (1998) were 30 percent less likely than
sick children in unaffected parts of the country to be taken for medical consultation after
the shock. Likewise, malnutrition rates among affected children almost quadrupled (Báez
and Santos 2007). Similarly, the food price spike in 2007–08 led to a deterioration in food
expenditures and, possibly, nutrition. A 1.0 percent increase in the price of wheat flour
was associated with a 0.2 percent decline in real monthly per capita food consumption
(D’Souza and Jolliffe 2014).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
50
The negative effects on human capital accumulation are particularly evident among
school-age children. If they face tight budgets, households sometimes need to pull
their children out of school to benefit from the extra labor. This was the case among
households in the areas most affected by the 1999 earthquake that struck the west-
central part of Colombia’s coffee belt. These households were less likely to send their
children ages 6–11 and 11–15 to primary and secondary school, respectively, chiefly
because of the massive destruction of housing and infrastructure (Bustelo, Arends-
Kuenning, and Lucchetti 2012). Similarly, school attendance fell by almost 7 percent
among households more heavily affected by the two serious earthquakes that struck El
Salvador in 2001 (Santos 2007). Because schools play a central role in the creation of
human capital, the destruction of school facilities and other complementary resources,
such as roads and other transport infrastructure, hamper learning and reduce instruction
time, thereby lowering school quality.25
The detrimental effects on human capital are typically long-lasting. Once Mexican
children are withdrawn from school in response to large shocks, they become nearly 30
percent less likely to reenroll relative to children who stay in school (Sadoulet et al. 2004).
Similarly, resource-constrained households in Venezuela resorted to the labor of their
children to cope with the 2002–03 economic crisis. Many of these children never went
back to school after the economy recovered (Blanco and Valdivia 2006). Interruptions in
school participation associated with aggregate shocks translate into lower educational
attainment among young adults. Individuals born and raised in the areas most affected by
the 1979–92 armed conflict in El Salvador showed a reduction in educational attainment
later in life equivalent to 0.4 to 1.7 years of education, which also weakened the engagement
of these people in the labor market (Acosta et al. 2016). In this way, aggregate shocks may
also affect future generations.
The persistence of the negative effects are manifest in other, economically meaningful
ways, particularly among the poor. The Ecuadoran children affected by the 1997–98 El
Niño floods while in utero were found to be 0.09 standard deviations shorter in stature
seven years later and 0.13 standard deviations lower on cognitive tests (Rosales 2014).
Infants and school-age children affected by the 7.5 magnitude earthquake that struck
Guatemala in February 1976 accumulated an average of 0.2 to 0.4 fewer years of schooling
in adulthood for each additional standard deviation in the intensity of the shock. Moreover,
they averaged around 0.4 centimeters shorter in height than others (Hermida 2011).
Analysis of prenatal stress caused by large shocks finds a negative impact on children’s
cognitive outcomes. This effect is acutely stratified by income levels: it is large among
25 However, these types of shocks do not always slow human capital accumulation. An analysis of the effects of the 1988–92 economic crisis in Peru finds no evidence that school attendance fell during the crisis. Instead, there was a significant decline in the share of children who were both in school and employed, implying that unemployment lowered the opportunity costs of schooling (Schady 2004).
CHA
PTER
5Th
e Vu
lner
abili
ty o
f Ass
ets
and
Live
lihoo
ds to
Agg
rega
te S
hock
s
51
poor families, but disappears among high-income families. The cognitive ability of poor
children in places affected by the 2005 Tarapacá earthquake in Chile turned out to be
significantly lower seven years later than that of poor children in places that were not
affected by the shock or children in high-income households (Torche 2016).
Not everyone’s assets are affected equally
Aggregate shocks affect broad economic and social systems, but their effects on assets,
groups, and communities are not evenly distributed. Although the incidence of major
disasters in some developed countries is comparable with the incidence in developing
countries, developed countries tend to report substantially fewer human casualties and
much less material damage because economic development typically provides implicit
insurance against shocks (Kahn 2005; World Bank 2013b). The earthquakes that struck
Chile and Haiti in 2010 were of similar magnitude, but were associated with starkly
contrasting effects: 525 were killed in the former case compared with over 200,000 in the
latter. As of 2010, GDP per capita was around 17 times larger in Chile than in Haiti. Cross-
country analysis shows that the death toll of natural disasters falls as the GDP per capita of
countries rises (figure 5.4). There is likewise variation in the impacts within countries and
communities across different socioeconomic and demographic groups (box 5.1). The poor,
women, geographically isolated households, and unskilled workers are frequently affected
the most. The differences in exposure and vulnerability are equally evident in developed
countries. Less availability of economic resources and mobility in the neighborhoods of
New Orleans prior to Hurricane Katrina were strong predictors of the negative effects that
unfolded during and after the catastrophic event (Masozera, Bailey, and Kerchner 2007).
Figure 5.4. Natural Disasters Kill More People in Poor Countries Than in Rich Countries
Tota
l pop
ulat
ion
affe
cted
(log
s)
1.5
2.0
2.5
3.0
3.5
4.0
4.543.532.52
Source: World Bank calculations based on data in EM-DAT (International Disaster Database), Centre for Research on the Epidemiology of Disasters, Université Catholique de Louvain, Brussels, http://www.emdat.be/database.Note: Each observation corresponds to the initial GDP per capita and total population affected in each decade between 1960 and 2010.
All Regions LAC
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
52
Box 5.1. The Multifaceted Vulnerability of Haiti to Natural Disasters
Haiti exhibits one of the highest levels of exposure to natural hazards in the world. It is located in the path of frequent tropical storms, which are becoming more frequent and more intense. As this box was being drafted, Hurricane Matthew (October 4, 2016) ravaged Haiti’s remote southwestern peninsula, killing more than 1,000 people, leaving 15 percent of the population in need of humanitarian assistance, and destroying infrastructure and the livelihoods of a large share of Haitians.
Measured as the ratio of economic losses to GDP, Haiti’s economic vulnerability to natural disasters is significant. Household surveys reveal geographical variations in the exposure and vulnerability to natural disasters. The Sud-Est (southeast) Department is the most vulnerable given that 78 percent of the population reported effects of a natural shock in 2012 (figure B5.1.1). In contrast, the Ouest (west) Department reported the lowest share of people affected by natural hazards (43 percent).
Historically, the deadliest hazard in Haiti has been seismic activity, the consequences of which are related to human decisions on where and how to build. Institutional weaknesses such as deficient urban planning and unenforced building codes aggravate the impact of disasters. With a magnitude of 7.0 on the Richter scale, the 2010 earthquake killed over 200,000 Haitians and injured many more. Thousands of people were left homeless. Water and electricity infrastructure, roads, and port systems in the capital, Port-au-Prince, and surrounding areas were destroyed. Haiti was subsequently struck by a cholera epidemic that led to more deaths. Although 75 percent of households throughout the country reported their living standards were degraded after the earthquake, there were stark differences in the impact of shocks on vulnerable and more resilient households.
Figure B5.1.1. The Close Relationship between Poverty and Vulnerability to Natural Disasters, Haiti (affected people and poverty headcount, by department)
Shar
e of
pop
ulat
ion
livin
g in
pov
erty
30%
50%
60%
70%
80%
90%
40%
65% 70% 75% 80%60%55%50%45%40%
Source: Calculation based on L'Enquête sur les Conditions de Vie des Ménages Après Séisme (postearthquake living conditions survey).Note: Climatic shocks include cyclones (hurricanes), floods, drought, and irregular rainfall. The poverty line was set at G 29,138.74. Departments are classified by level of vulnerability, which is based on the percentage of people affected by a climatic shock. The bubble size indicates total population in 2009.
Percentage of population affected by a climatic shock (Vulnerability)
Ouest
Nord
Artibonite
NippesSud
Sud-Est
Centre
Nord-OuestGrand-AnseNord-EstR2=0.6175
CHA
PTER
5Th
e Vu
lner
abili
ty o
f Ass
ets
and
Live
lihoo
ds to
Agg
rega
te S
hock
s
53
The poor and other vulnerable groups often bear the largest burden. The interactions
of shocks and individual and community vulnerabilities lead to large heterogeneity in the
impacts. Economic crises, for example, do not affect all groups equally; inequalities in
exposure and sensitivity to risk as well as inequalities in access to resources, capabilities,
and opportunities disproportionately affect certain groups or communities, rendering
them more vulnerable to the impact of major shocks. Natural disasters exhibit a similar
pattern: deeper effects on the most vulnerable (figure 5.5). Women are remarkably less
well protected from weather shocks in Ethiopia and Zimbabwe, whereas under-5 mortality
is worse among landless households in India (Dercon and Hoddinott 2004; Rose 1999).
Globally, the life expectancy of women has been found to fall more than that of men in the
context of natural disasters in a sample of up to 141 countries in 1981–2002 (Neumayer
and Plümper 2007). Wars and political oppression tend to affect particular demographic
groups more than others. In Peru, the civil war systematically affected certain political
groups, and, in Guatemala, the death toll from the country’s civil war was relatively larger
among the indigenous population (Malasquez 2016). Major shocks are generally rather
regressive: the monetary and nonmonetary poor are the most affected in most cases.
Box 5.1. (continued)
The poor, representing 57 percent of the population, are disproportionately affected by natural disasters. Analysis using data collected by the World Bank and the government shows that, in Haiti, the poorer individuals are, the more vulnerable they are to natural disasters. Households in extreme poverty reported they had experienced an average of nearly three shocks, whereas less vulnerable households experienced 2.5 shocks.a Likewise, rural households were more deeply affected, experiencing almost twice the number of shocks as the average household in Port-au-Prince. Overall, however, a typical household in Haiti, regardless of poverty status, faces multiple economic shocks annually, often because of natural disasters.
In Haiti, the main economic impact of shocks is a reduction in income, followed by a reduction in food expenditures, as well as a depletion in assets. Impacts are heterogeneous, exhibiting differences between the vulnerable and more resilient households: 53 percent of households in extreme poverty experience a reduction in food consumption following a natural shock, whereas 34 percent of resilient households reported they cut back on these expenditures. Poor households are also less likely to rely on private transfers and other risk-sharing mechanisms: 54 percent of the resilient households are able to tap into the monetary support of others to counter the effects of shocks, while nearly a quarter of the households (26 percent) in extreme poverty are able to do so. Other coping strategies commonly seen among poor and vulnerable households such as selling assets, pulling children out of school, and reducing calorie intake undermine future income opportunities.
a. The total shocks may be slightly underestimated because the survey did not collect information on the number of times a particular type of shock is experienced.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
54
Exposure to shocks and the size of the impacts are determined largely by the
characteristics of the population. The physical deprivations that characterize most
vulnerable groups, such as their location, the low-quality of their housing, and their
restricted access to hazard buffer mechanisms also raise the importance of the impact
of major hazards on these groups. Data on Colombia and Mexico show a high correlation
between lower incomes and location in areas that are more likely to be affected by natural
disasters. For instance, households among the bottom 40 in Mexico are nearly four times
as likely to report they have been affected by a systemic shock relative to households in
the top 20. A similar relationship between income level and the incidence of a shock is
observed among households in Colombia. In both countries, major hazards are systemically
more common in rural areas than in urban areas (table 5.2). Additional data from panel
surveys in Mexico show that, in 2002–05, the poor were almost three times more likely
than the vulnerable or the middle class to be affected by a natural disaster resulting in loss
of dwelling, crops, and livestock (de la Fuente, Ortiz-Juárez, and Rodríguez-Castelán 2017).
The size of the effects of aggregate shocks on assets can increase more than
proportionally as the intensity of the shocks increases. Empirical studies document a
host of nonlinear relationships between aggregate shocks (particularly extreme weather
events), physical and human capital damage, and reductions in household well-being. This
was the case of tropical storm Agatha in Central America in 2010. Spatially disaggregated
Figure 5.5. The Poor Often Suffer the Most from Natural Disasters
a. Exposure b. Vulnerability
0
20
40
60
80
100
Vietnam
Tegu
cigalp
a
San Salv
ador
Nepal
Mumbai
Middle Eas
t and
North Afri
caKenya
Honduras
Haiti
Guyana
Guatemala
Bangla
desh, 2
Bangla
desh, 1
0
20
40
60
80
100
Mumbai
Honduras
Bangla
desh, 3
Bangla
desh, 2
Bangla
desh, 1
Sources: Based on del Ninno et al. 2001 for Bangladesh (1) and Akter and Mallick 2013 for Bangladesh (2); Tesliuc and Lindert 2003 for Guatemala; Pelling 1997 for Guyana; Fuchs 2014 for Haiti; Carter et al. 2007 for Honduras; Opondo 2013 for Kenya; Wodon et al. 2014 for the Middle East and North Africa; Baker et al. 2005 and Ranger et al. 2011 for Mumbai; Gentle et al. 2014 for Nepal; Fay 2005 for San Salvador and Tegucigalpa; and Nguyen 2011 for Vietnam. Taken from Hallegatte et al. 2016.Note: Each study has different definitions of poor and nonpoor, and exposure differs based on the type of hazard and the context in which it occurs.
Surv
eyed
hou
seho
lds
affe
cted
by
nat
ural
dis
aste
r (%
of t
otal
)
Asse
ts o
r in
com
e lo
st fo
r af
fect
edha
ouse
hold
s (%
of a
nnua
l inc
ome)
Poor Nonpoor
CHA
PTER
5Th
e Vu
lner
abili
ty o
f Ass
ets
and
Live
lihoo
ds to
Agg
rega
te S
hock
s
55
rainfall records suggest that consumption fell more than proportionally among Guatemalan
households exposed to rainfall 6 standard deviations (extremely high intensity) above the
historical mean relative to households in areas that recorded rainfall anomalies between
2 and 3 standard deviations (relatively lower intensity) (Báez et al. 2016). Accounting for
different threshold effects is crucial for policy purposes. Under different global warming
forecasts, global distributions of temperatures are projected to shift, and the frequency
and severity of extreme shocks are expected to increase. Several analyses reveal similar
tipping points in the relationships between temperature and agricultural yields, health
status, or mortality (for instance, Deschênes and Greenstone 2011; Deschênes, Greenstone,
and Guryan 2009; Schlenker and Roberts 2008). Therefore, assuming linearities in
extrapolating from existing evidence to project and plan against future damage will likely
misinform policy.
Table 5.2. Income and Location Are Strong Predictors of the Exposure to Shocks
Mexico 2009–12 Colombia 2013
Total 5.6 13.4
Quintile 1 9.0 19.2
Quintile 2 7.1 18.9
Quintile 3 6.3 13.1
Quintile 4 2.8 9.2
Quintile 5 2.5 7.1
Rural 13.7 17.8
Urban 3.0 9.5
Source: Calculations based on the Colombian Longitudinal Survey of Wealth, Income, Labor, and Land and the Mexico Family Life Survey.Notes: The relevant survey questions were “Over the last two years, has this house been affected by floods, landslides, earthquakes, or similar events?” (Colombia) and “Did the household lose the home, businesses, crops, or agricultural production animals due to a natural disaster?” (Mexico).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
56
CHA
PTER
6A
ggre
gate
Sho
cks
Und
erm
ine
Ince
ntiv
es to
Use
Fac
tors
of P
rodu
ctio
n
57
6. Aggregate Shocks Undermine Incentives to Use Factors of
Production and Invest
The supply of labor is sensitive to major shocks, increasing at times and at times falling
Aggregate shocks are expected to influence the supply of labor. Among most households,
labor is the main asset. Labor income in the region accounts for 60 percent to 80 percent
of the total income for a typical household in the bottom two quintiles (Cord, Genoni,
and Rodríguez-Castelán 2015). Simple neoclassical theory predicts that the labor supplied
by individuals is the result of a utility maximization process that reflects trade-offs in
preferences between income (consumption) and leisure time. Given that large shocks
often have a sizable effect on factors such as wealth, consumption, and relative wages,
as well as the demand for workers, it is likely that they influence household decisions on
the amount of time to allocate to working and to leisure activities. Yet, as the evidence
summarized below suggests, whereas a change in the amount of labor supplied—or even
a shift in the aggregate supply curve—is expected in the context of a major shock, it is
difficult to forecast the direction of the net effect, particularly because aggregate shocks
often also disrupt labor demand. Thus, potential impacts vary with the type of shock, the
country context, and local markets and institutions.
A typical coping response of households involves labor supply adjustments as an
attempt to earn extra income. Aggregate shocks can destroy household wealth and assets,
reducing the income opportunities that can be generated from wealth and assets. They are
also expected to change prices and thus relative wages. In an effort to protect private
expenditures or to avoid a fall in expenditures below subsistence, affected individuals may
respond by supplying extra labor. Some individuals who were not participating in the labor
market before the shock may decide to seek employment, while those already employed
may attempt to work longer hours. Mexican households attempted to cope with the 1994–
95 crisis driven by the sharp devaluation of the peso and high inflation by increasing the
labor force participation of adult women, including the wives of household heads who
had been transitioning from employment to unemployment (Skoufias and Parker 2006).
Extreme floods in the Caribbean areas of Colombia encouraged a similar coping response in
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
58
urban areas (Acevedo 2016). Reconstruction efforts in the aftermath of major catastrophes
can create new job opportunities, partly offsetting job cuts in some economic sectors.
Unfavorable economic circumstances or major health setbacks can result in lower labor
supply in the market. Labor market conditions frequently deteriorate at the peak of an
economic crisis. Firms cut jobs; wages fall; and unemployment rises, denting employment
growth. Individuals may boost their labor supply to compensate for job losses or to respond
to changes in relative wages. However, the lack of suitable employment options and the
depreciation of skills because of long spells of unemployment may discourage workers
from looking for jobs even if they are eligible and able to work. Serious systemic health
shocks can result in lower labor supply and productivity. For instance, individuals exposed
to the Dutch Famine (1944) before birth were less likely to be employed in adulthood partly
as a result of reduced cognitive ability (figure 6.1). Crop diseases exert similar pressure
on economic activity and employment. The 2012–13 coffee season in Central America was
ravaged by the rust fungus disease, erasing over 20 percent of the production and almost
15 percent of the employment in the sector, equivalent to 250,000 jobs (Piñeiro, Morley,
and Elverdin 2015).
Migration is a common risk mitigation strategy
The relocation of labor through migration is a household strategy for dealing with
shocks. Families may decide to diversify by supplying their labor in markets and economic
activities less exposed to risk as a sort of insurance. Analysis of internal migration in
Nicaragua shows that the level of insurance increases if migrants and households of
Figure 6.1. Individuals Affected by the Dutch Famine (1944) in Utero Are Less Likely to be Employed as Adults
Individuals non-exposed
23%
24%
24%
27%
20% 24% 28%
3
24
27
233
24Individuals born inaffected aereas
Individuals born innon-affected aereas
Source: Calculations based on Scholte, van den Berg, and Lindeboom 2012.
Exposure during first trimester of in utero gestation
CHA
PTER
6A
ggre
gate
Sho
cks
Und
erm
ine
Ince
ntiv
es to
Use
Fac
tors
of P
rodu
ctio
n
59
origin are exposed to less correlated rainfall shocks or are engaged in different economic
sectors (Molina 2014). Similar insurance motives encourage other forms of migration.
Marital arrangements among households in south Indian villages function as implicit
interhousehold contracts to mitigate income risks in the face of spatially covariant shocks.
Households exposed to higher income risks are more likely to invest in longer-distance
migration-marriage arrangements (Rosenzweig and Stark 1989).
Migration decisions are also shaped by the effects that unfold after the occurrence
of major shocks. Internal and external migration is often substantial in difficult times
as households try to cope with the effects of negative shocks.26 Migration flows between
Mexico and the United States slowed after the global financial crisis of 2008–09. Indeed,
for the first time in recent history, more Mexican immigrants returned to Mexico from the
United States than left Mexico for the United States (Gonzalez-Barrera 2015). An analysis
of eight countries in Central America, which is characterized by exceptionally high internal
and international migration rates and substantial exposure to natural hazards, finds that
individuals, particularly youth, are more likely to migrate in response to these disasters,
especially if confronted with droughts (Báez et al. 2016). In the absence of agricultural
shocks, Salvadoran households are 24 percent less likely to send at least one member to
the United States (Halliday 2006). Research on Mexico shows that a 10 percent reduction
in crop yields driven by variations in climate is associated with an increase of 2 percent in
the rate of migration to the United States (Feng, Krueger, and Oppenheimer 2010).
Natural disasters, civil conflict, and repression frequently lead to forced population
displacements. People often have to migrate as a survival strategy in the aftermath of
major earthquakes, floods, or droughts. Similarly, entire families are forced to flee their
homes to seek refuge in their own countries or across international borders because of civil
conflict, violence, or systematic human rights violations. As of 2014, close to 60 million
people worldwide were living under forced displacement, the highest levels since World
War II (UNHCR 2015). An additional 20 million are estimated to have been displaced
because of natural disasters (IFRC 2012). While at a lower scale relative to other regions,
forced migration has also been prevalent in Latin America and the Caribbean. More than six
million people have been forcibly displaced during 50 years of internal conflict in Colombia.
Similarly, over a million Guatemalans and Salvadorans abandoned their homes to flee from
civil war, most of whom represented indigenous groups (Gammage and Fernandez 2000;
IDMC 2011). One-sixth of the population of Port-au-Prince (close to a half million people)
left the capital immediately after the 2010 earthquake to seek refuge in other provinces.
Forced migration imposes high costs on civilian populations, unsettling their
engagement in labor and other markets. Households are left without access to basic
26 Internal migration rates are also strongly procyclical (Monras 2015; Saks and Wozniak 2011).
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
60
health care, education, and welfare services. Incomes are weakened, while assets, such
as housing and land, are destroyed or illegally seized, and social capital is fractured.
Approximately half the households internally displaced by the conflict in Colombia lost
their homes. The aggregate land loss was over 2 million hectares, 3.5 times the amount of
land granted in agrarian reform programs in 1993–2002. Because most households were
active in agriculture prior to displacement, they struggled to integrate into urban labor
markets. As a result, over half the household heads were unemployed three months after
settlement at destination sites. Moreover, upon return, few households can recover or use
their previously occupied land because of weak systems of property rights (Ibáñez and
Moya 2006).
Households tend to rely on the labor of their children to cope with aggregate shocks
Children represent a buffer mechanism, particularly among resource-constrained
households. In tough times, households are often forced to resort to the labor of their
children, who can thus contribute to income or free up time for adults. However, children’s
work entails long-run costs if it interferes with the human capital accumulation of the
children and leads to reduced earnings potential. Brazilian children, particularly girls,
were more likely to abandon school and enter the labor force during several of the
economic slowdowns recorded in the 1980s and 1990s. Once out of school, they became
10 percentage points less likely to progress across grades later on, offsetting short-term
smoothing gains with long-term losses in human capital (Duryea, Lam, and Levison 2007).
Salvadoran children were more likely to be taken out of school to look for work following
the 2008–09 global crisis (Duryea and Morales 2011). Unemployment among household
heads driven by the 1994–95 Mexican peso crisis was significantly associated with a higher
probability that teenage girls were not attending school (Skoufias and Parker 2006). Natural
disasters prompt similar responses. Children’s labor participation swelled in Colombian
municipalities hit by record flooding in 2010 because of La Niña (Acevedo 2016).
Children and teenagers may be pushed to work if favorable economic circumstances raise
the opportunity costs of sending them to school. Households send their children to work
out of necessity, but they also do so in relatively better economic times to take advantage of
good opportunities. If the substitution effect of unexpected temporary changes in income
exceeds the income effect, transitory improvements in economic conditions and household
incomes can increase children’s labor participation and deter schooling. Aggregate events
such as economic booms in specific sectors or sharp increases in commodity prices can
raise the value of children’s time. Coffee production, an important sector in many countries
in the region, provides a good example of the effects of local economic conditions on the
value of children’s time. Increases in the coffee harvest and prices in Brazil led to more
work among low- and middle-income boys and girls, but only poor children were withdrawn
CHA
PTER
6A
ggre
gate
Sho
cks
Und
erm
ine
Ince
ntiv
es to
Use
Fac
tors
of P
rodu
ctio
n
61
from school as a result of this time reallocation (Kruger 2007). The sharp increase in gold
prices in 2002–11 led to a similar response. Child labor rose by 12 percent in Colombian
villages close to gold mines. This happened in tandem with a drop in school attendance
of 28 percent, mostly among children attending public schools (Santos 2015). Similarly,
the engagement of children in labor markets increased following the strong earthquakes in
Haiti and Peru in 2010 and 2007, respectively (figure 6.2).
The demand for labor is responsive to aggregate shocks
Major shocks disrupt job creation, reducing the demand for labor. Large natural
disasters or civil conflicts do more than wipe out homes. They decimate local economies,
thereby distressing businesses, supply chains, and markets. Simulations of the economic
impacts of electricity disruptions caused by catastrophic earthquakes show that the
potential output loss could amount to as much as 7 percent of the gross regional product
in areas near Memphis in the United States (Rose et al. 1997). Economic recessions lead to
a slowdown in aggregate output, a slump in consumer spending, and a fall in stocks, and
they put the brakes on investments in capital, marketing, research and development, and
the expansion of credit. As this broad-spectrum aggregate supply slowdown unfolds, firms
are likely to freeze hiring and lay off employees, which ultimately results in lower demand
for labor and higher unemployment. Aggregate shocks can leave long-lasting scars among
those workers who are more prone to bear the brunt of major crises. For instance, jobs
remain relatively more difficult to find among young workers even as economies recover.
This is especially relevant in the region, where one individual in five ages 15–24—totaling
Figure 6.2. More Children Join the Labor Market after a Major Earthquake
After Earthquake
0% 10% 20% 30% 40% 50%
2010 Haiti earthquake
2007 Peru earthquake
Source: Calculations based on WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/data-catalog/world-development-indicators.Notes: Children ages 7–14 years in employment refer to children involved in economic activity for at least one hour in the reference week of the corresponding household survey. Values shown correspond to the indicators available closely before and after each earthquake.
Before Earthquake
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
62
more than 20 million people—is neither working nor in school (de Hoyos et al. 2016).
More broadly, major crises may lower the output potential of a country whenever they
structurally undermine labor and productivity growth.
More jobs are destroyed than created as a consequence of major shocks, particularly
economic crisis. The global financial crisis of 2008–09 is probably the most recent
illustration of the weakening effects of crisis on net job creation across developed and
emerging economies. Globally, it is estimated that firms cut nearly 67 million jobs (ILO
2015). The unemployment rate in the United States rose sharply, from 4.5 percent in 2007
to over 10.0 percent in 2010, whereas it reached levels on the order of 25.0 percent in
Greece and Spain. The same crisis destroyed a large number of formal and informal jobs
in Mexico, lifting unemployment rates by more than 50 percent and depressing real wages
across almost all economic sectors (Freije, López-Acevedo, and Rodríguez-Oreggia 2011).
Unemployment rates also surged dramatically following previous macroeconomic crises in
Argentina (1995) and the Dominican Republic (2003–04) (Fallon and Lucas 2002; World
Bank 2015).
Natural disasters lead directly to job losses. The 2010 earthquake in Haiti and the
devastating tropical storms that hit the country in 2012 crippled the agricultural sector, the
main employer, leaving thousands jobless and causing unemployment to skyrocket. Close to
100,000 jobs were lost in Chile after the 2010 earthquake, pushing up the unemployment
rate by 1 percentage point (Elsahai et al. 2010). Women in Colombia became 7 percentage
points more likely to lose their jobs in municipalities flooded by extreme rainfall in 2010
(Acevedo 2016). The impacts are not always instantaneous. Empirical analysis on a large
number of countries in the region shows that the probability of unemployment rises by up
to 10 percent among individuals exposed to a major natural disaster at some point during
their first 15 years of life (Caruso 2017).
A related concern is the increase in the duration of unemployment. In some cases, long-
term unemployment results from a combination of higher labor participation and lower
demand for labor. In other cases, however, even if labor force participation has tumbled,
longer spells of unemployment are likely driven by a high rate of job loss because of real
wage adjustments or lower labor demand. This was the case during the global financial
crisis, which was associated with an unprecedented increase in the mean and median
time of unemployment across most advanced and developing countries.27 The duration of
unemployment almost doubled during the Argentine crisis, from 6.6 months in 2001 to 11.7
months in 2003. Spells of unemployment may become structurally longer if firms decide
to shift jobs away from areas prone to natural disasters, which are typically inhabited by
vulnerable populations.
27 In the United States, the average duration of unemployment reached 39 weeks, up from less than 20 weeks in previous recessions between the early 1980s and early 2000s.
CHA
PTER
6A
ggre
gate
Sho
cks
Und
erm
ine
Ince
ntiv
es to
Use
Fac
tors
of P
rodu
ctio
n
63
Many of the jobs that remain or are created after a major shock are lower in quality.
Underemployment, a manifestation of the underutilization of workers and skills, tends to
rise during economic recessions, natural disasters, and civil conflict. The global financial
crisis led to a sharp rise in the incidence of precarious employment in many countries
despite the desire of workers to remain on the job for more hours and have more stable
contracts. Of 35 advanced economies tracked by the International Labour Organization, 29
recorded an intensification in involuntary part-time and temporary employment between
2007 and 2010. Informal employment, a common feature of labor markets in the region,
also increased in some countries during the global crisis. In Honduras, the ratio of informal
jobs to total nonagricultural jobs rose from 52 percent (precrisis) to 58 percent (2010).
Increases in informality—although of smaller magnitude—were also seen in Colombia,
the Dominican Republic, El Salvador, Mexico, and Venezuela over the same period (ILO
2012). Individuals exposed during childhood to the direct effects of internal conflicts in
El Salvador and Peru show a larger probability of working at an informal job later in life
(Acosta et al. 2016; Malasquez 2016). Moreover, after crises, newly created jobs are more
likely to be provided on a short- rather than long-term basis.
Shocks and perceptions of risk distort the allocation of financial capital
Capital often becomes scarcer in the wake of major financial crises. Negative
macroeconomic and financial shocks trigger vast losses, dislocating capital markets.
Precautionary liquidity and risk-aversion on the supply side following a crisis distort
the optimal allocation of capital across an economy. Banks and investors respond
by reallocating their investments or even reducing their portfolios, which often
results in a credit crunch. As a result, private firms face greater financial constraints
because of a lack of accessibility to debt and equity capital, which limits their ability
to invest. In Mexico, the debt crisis of 1982 and the peso crisis of 1994–95 eroded
trust in financial institutions among savers who lost large portions of their deposits
and among borrowers who lost assets and other collateral (IMF 2015). These effects
are usually not equally distributed. Credit tends to dry up more rapidly among small
and medium enterprises, a sector that contributes substantially to employment and
income. Likewise, foreign investment weakened in the region following the Argentine
crisis in 2001 and the most recent global crisis: it dropped by 38 percent and 23
percent, respectively (figure 6.3).
Risk-averse, shock-prone households are less likely to use financial and insurance
instruments. Standard neoclassic theory predicts the opposite, namely, that this type
of household should show a stronger demand for insurance instruments to cope with
shocks. The evidence speaks to the opposite view. In Ethiopia, the demand for weather
index insurance is markedly lower among smallholder farmers who are characterized
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
64by their higher degree of risk aversion and vulnerability to shocks (McIntosh, Sarris,
and Papadopoulos 2013). Another possible explanation for the low take-up is ambiguity
aversion, namely, that households do not perfectly understand the distribution from which
the relevant risk probabilities are drawn (Bryan 2010). Another contributory factor may be
the presence of basis risk, which means households may end up not receiving indemnities
even if they are hit by a serious shock (Clarke 2016). Ambiguity aversion and basis risk are
likely to be aggravated in a context of stronger hazards at greater frequency.
Risk pushes households to use productive assets inefficiently and forgo some returns
Aversion to latent risk distorts the productive choices of households at the cost of
forgoing returns. Credit- and insurance-constrained, risk-averse households are inclined
to mitigate ex ante the effects of major shocks by making conservative employment and
production choices. These strategies reflect the desire of households to choose income-
generating activities and use their assets to reduce risks rather than to maximize profits.
Shock-prone farmers in south India use labor relatively more intensively than profit
maximization behavior alone would dictate (Antle 1987). Inputs such as fertilizers or
enhanced seeds that boost crop productivity and raise expected profits in low-income
agricultural environments are used less intensively by risk-prone households in Ethiopia
and India for fear of incurring investment losses if shocks result in poor harvests (Bliss and
Stern 1982; Dercon and Christiaensen 2011; Lamb 2003).
The nature of aggregate shocks and market imperfections forces households to own
assets that help them reduce risk rather than maximize profits. An essential feature of
Figure 6.3. Foreign Investment Fell during the Argentine Crisis (2001) and the Global Financial Crisis (2008–09)
Fore
ign
dire
ct in
vest
men
t, ne
t inf
low
s(B
oP, c
urre
nt M
illio
ns U
S$)
GlobalFinancial Crisis
ArgentineanCrisis
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
180,000 Crisis nancial Crisi
201520122009200620032000
Source: Calculations based on data in WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/data-catalog/world-development-indicators.
CHA
PTER
6A
ggre
gate
Sho
cks
Und
erm
ine
Ince
ntiv
es to
Use
Fac
tors
of P
rodu
ctio
n
65
communities characterized by incomplete markets and high exposure to large risks is asset
portfolios that contribute directly to production and indirectly to smoothing consumption.
The covariate nature of aggregate shocks that simultaneously affect geographical clusters
of households and villages tend to exacerbate borrowing, insurance, and rental market
constraints. Indian farmers own bullocks to till the land in the short time window between
the onset of the monsoon rains and the optimal sowing date. However, while it would be
more efficient for farmers to rely on a bullock rental market, accumulating and using other
profit-enhancing assets instead, the systemic uncertainty of the monsoon onset date and
the positive covariance of rainfall within villages force farmers to own bullocks to meet
their demand for animal traction. Simulations in this context reveal large welfare gains
arising from policies such as credit, subsidized weather insurance, and cash transfers that
would help stabilize income flows among farmers (Rosenzweig and Wolpin 1993).
Households prone to weather variability have a greater propensity to change the
timing of their investments, undermining the returns. A consequence of climate
change or phenomena such as El Niño and La Niña is to shift global precipitation patterns
dramatically, delaying the onset of rains, broadening or shortening rainfall seasons,
making the path of storms and hurricanes less predictable, and increasing the intensity
of the storms and hurricanes. In the face of rising uncertainty about the timing of rainfall,
poor households may resort to postponing investments (for instance, sowing seeds) until
they have better information on expected weather conditions. If rains are poor, farmers
may cut back production to reduce possible losses (Morduch 1995). Evidence shows that
low-income households do not plant crops if the soil lacks sufficient water. As a result,
the variability of areas cultivated in low-income settings is larger than the variability
in agricultural yields. This explains why consecutive drought years may result in larger
reductions in cultivated area than shortfalls in yields (Walker and Ryan 1990). Delaying
the use of assets can involve high costs.
In dealing with shocks, vulnerable households opt to diversify toward portfolios with
safer, but less profitable activities. Vulnerable rural households devote larger shares of
their land to more conservative crops rather than to riskier, higher-yielding crops. Poor
Indian farmers in villages of the International Crops Research Institute for the Semi-Arid
Tropics are more likely to plant safer, but less lucrative varieties of rice (Morduch 1990).
Likewise, households in the lowest wealth quintile in western Tanzania allocate close to
10 percent of their land to sweet potatoes (compared with less than 2 percent among
the well off), a low-risk yield that is difficult to market and is thus characterized by low
economic returns (Dercon 1996). Vulnerable households also diversify their crops spatially
to work on a number of small fields rather than on a more efficient consolidated plot.
This is the traditional risk management strategy among shock-prone Peruvian farmers in
the district of Cuyocuyo who disperse their agricultural fields over the landscape (Goland
1993). The riskiness of agricultural incomes, intensified in a context of aggregate shocks,
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
66
also influences tenancy contracts (such as sharecropping) and long-term labor contracts
(for example, permanent arrangements at stable, but low wages) in ways that favor risk
management over expected returns.
Engagement in often less profitable nonfarm jobs provide another typical form of
diversification among vulnerable households. A large share of farm households across
the developing world earn a significant share of their incomes from nonfarm sources,
often motivated by the need to manage risk and cope with shocks. In Guatemala, 86
percent of the rural households living in extreme poverty report engagement in activities
outside agriculture. A similar share also report that they participate in two or more
activities to earn a living. Even if the levels are lower, a comparable pattern is observed
in other countries in the region, such as Mexico, Nicaragua, and Panama (Banerjee
and Duflo 2007). Rural households in areas highly affected by the Colombian conflict
reduced the amount of time spent on onfarm work and increased their supply of labor
to off-farm activities. While evidence shows that the poor are more likely to engage in
off-farm activities where earnings have a low positive covariance with the returns they
obtain in agriculture, their overall earnings outside agriculture are lower more often
than not (Reardon et al. 2006).
Because of shocks or risk management strategies, production factor returns are often lower
Workers in areas affected by major shocks often experience substantial earnings
losses, which can extend into the medium and long run. Individuals often increase their
labor supply in times of aggregate shocks, whereas firms and businesses are likely to shed
more jobs than they create. Taken together, both changes in labor supply and labor demand
lead to a drop in real wages. This seems to be the rationale behind the wage income losses
suffered by urban households located in the path of Agatha (2010), the strongest tropical
storm to strike Guatemala since rainfall records have been kept. While the decision
whether to work remained unchanged among adults in the affected households, they
were found to work an average of 2.2 more hours per week in the aftermath of the event.
Possibly signaling a general equilibrium effect, workers commanded wages 5.4 percent
lower relative to the median hourly wage at baseline (Báez et al. 2016). Similarly, a careful
analysis of the nature of the jobs created during the global financial crisis (2008–09)
shows that the majority of new jobs in Argentina and Mexico were remunerated at a rate
below average wages. In fact, the number of jobs paying below two times the minimum
wage rose during the 2008–09 crisis in Mexico and now make up the largest share of jobs
(ILO 2012; World Bank 2012a).
Forced migrants are among those people who bear the largest costs imposed by
large, negative shocks. The returns to human and physical capital usually drop after the
CHA
PTER
6A
ggre
gate
Sho
cks
Und
erm
ine
Ince
ntiv
es to
Use
Fac
tors
of P
rodu
ctio
n
67
occurrence of aggregate shocks because the skills of most displaced people tend to be
concentrated in agricultural activities, which are noticeably less in demand in urban labor
markets. Income among household heads affected by the Colombian conflict is less than
half among the people who are employed following displacement. The net present value
of forgone agricultural revenue over a lifetime is estimated at $15,787 per household,
nearly twice Colombia’s GDP per capita (Ibáñez and Moya 2006). More broadly, labor
market conditions can deteriorate because of forced migration flows, usually expanding
the informal economy (Calderón and Ibáñez 2009). Empirical research has shown that
children born and raised in areas affected by conflict and natural shocks typically grow
up to be shorter in stature, possibly implying negative effects on labor productivity in
adulthood, particularly if they are engaged in physically demanding activities such as
agriculture (Akresh, Caruso, and Thirumurthy 2014; Báez 2011; Báez and Santos 2007).
The latent risks of potential shocks undermine economic returns, particularly among
the poor. Households accumulate and use their assets to balance two objectives: maximize
profits and reduce risks. This link between production and consumption decisions entails
sizable efficiency losses because expected profits are sacrificed for lower risk. The
excessive use of labor by farmers in south India implies a risk premium of 14 percent of the
expected net returns (Morduch 1995). Low-income Indian farmers forgo up to 35 percent
of the expected profits if the impact of risk aversion on the use of a broad set of inputs is
considered (Rosenzweig and Binswanger 1993).28 The crop portfolio of Tanzanian farmers
in the wealthiest quintile yields 25 percent more per adult than that of the poorest quintile
(Dercon 1996). In Palanpur, India, delaying the onset of production by two weeks to gain
extra information on expected weather is found to decrease yields by 20 percent (Bliss
and Stern 1982). Field-scattering among farmers in the Peruvian district of Cuyocuyo is
estimated to diminish yields by 7 percent (Goland 1993). Conservative estimates in several
surveys suggest that the median return from the use of fertilizers is above 75 percent; yet,
few risk-averse and shock-prone households use fertilizers (Banerjee and Duflo 2007).
Overall, the lower returns do not follow from differences in risk preferences, but from the
limited options available to insured households.
Income diversification among vulnerable households often translates into lack of
specialization, small scale, informality, and little income smoothing. Households tend
to have multiple jobs, partly as a result of risk-spreading strategies. Short-term migration
is widespread. This multiplicity of occupations and the substantial amount of time spent
commuting often mean less specialization as households trade off opportunities to
master tasks or jobs that fit their skills. Businesses created by these households operate
at remarkably small scale. The average number of paid employees of a business created
by someone who lives on less than $2.00 a day range from 0.14 in rural Nicaragua
28 In contrast, farmers in the top wealth quartile take on more risk without sacrificing expected returns.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
68
to 0.53 in urban Panama (Banerjee and Duflo 2007). Available evidence contradicts
the conventional wisdom that earnings from labor migration—another common income
diversification strategy—exceed the earnings obtained from local nonfarm activities.
Empirical studies in five countries in the region (Brazil, Colombia, Ecuador, Mexico, and
Nicaragua) show that local nonfarm earnings surpassed the earnings of migrant family
members by a ratio of over 10 to 1 (figure 6.4). Diversification does not always lead
to income smoothing when most needed. Whereas incomes from jobs across different
sectors are uncorrelated in normal times, they are likely to collapse in tandem during
major crises. Droughts in West Africa reduce crop incomes, but also nonfarm incomes
(Fafchamps, Udry, and Czukas 1998).
Figure 6.4. The Share of Earnings from Labor Migration in the Region Is Low
Composition of nonfarm earnings: Local
0%
10%
20%
30%
40%
50%
60%
LAC averageNicaraguaMexicoEcuadorColombiaBrazil
Source: Calculations based on Reardon et al. 2006.
Composition of nonfarm earnings: Migration
CHA
PTER
7Ri
sk M
anag
emen
t Pol
icy:
Pre
pari
ng fo
r an
d Co
ping
wit
h A
ggre
gate
Sho
cks
69
7. Risk Management Policy: Preparing for and Coping with
Aggregate Shocks
The rationale for public action
Many of the negative impacts and development setbacks that arise from aggregate
shocks are the result of risk management systems unable to shield household well-
being. Aggregate shocks can lead to sizable losses in human lives, lower standards of
living, destruction of physical and human capital, disruptions of markets, and sacrifices
in economic opportunities. Many of these effects even last for a long time. Yet, the losses
associated with shocks do not arise primarily from the actions of nature, but from the
choices of people. In spite of notable recent economic and social progress, risk management
among individuals, households, communities, villages, and countries in the Latin America
and Caribbean region remains inadequate. Because of the large scale of aggregate shocks,
households alone are unable to handle the impacts and need support systems to withstand
them without forgoing present and future well-being.
Efficiency losses linked to weak insurance, protection, and coping against aggregate
shocks justify public intervention. Households follow multiple strategies to manage risks,
but these are only partially effective. Inadequate risk management discourages households
from taking on the risks that are encountered in pursuing economic gain, undermining
income generation. Income- and asset-based strategies to self-insure smooth incomes, but
they also skew incomes, clustering households into low-return, low-risk activities. The
poor functioning of asset markets constrains the access to alternative economic activities
and depresses asset values if these are liquidated during a crisis. Informal risk-sharing
mechanisms are prone to collapse when aggregate shocks occur. The welfare losses caused
by these inefficiencies are not trivial. The median return from the use of fertilizers is above
75 percent, but poor, shock-prone households rarely use fertilizers. Aggregate shocks also
tend to affect the poor and vulnerable disproportionally owing to a combination of greater
exposure, weaker internal conditions, and unsatisfactory risk management. The resulting
equity losses are equally large. The poor typically earn lower returns on their portfolios
than more well off households at similar levels of risk exposure.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
70
The benefits of adequate risk management are substantial. The risks of economic
stagnation can be reduced if countries follow sound fiscal and monetary policies.
Seismic building codes can protect vital infrastructure against the destructive power of
earthquakes. Spending on flood defenses and warning systems can help households and
businesses withstand floods. Farmers can boost agricultural productivity through increased
use of irrigation and fertilizers. Immunization among children provides protection against
certain diseases and contributes to healthier and more productive lives. Each dollar spent
on improving water and sanitation systems yields around 4 dollars in economic benefits
(figure 7.1).
Successfully addressing aggregate shocks requires an enabling environment for
collective coordination, action, and responsibility. World Development Report 2014
argues that effective risk management can be a powerful instrument for development by
building people’s resilience and allowing them to seize opportunities for improvement.
It calls for a “shift from unplanned and ad hoc responses when crises occur to proactive,
systematic, and integrated risk management” (World Bank 2013b, 4). Aggregate shocks
require coordination ex ante to prepare for the risks and shared action and responsibility
ex post to cope with the impacts. Public policy plays a central role in promoting the
environment needed for this sort of collective action to occur across society, from
households and firms to local communities, nongovernmental organizations, the financial
system, countries, and the international community.
Figure 7.1. The Benefits of Formal Risk Management Often Offset the Costs
Bene
fit-
cost
rat
ios
Break-evenpoint
0
2
4
6
8
10
12
Bevpo
Measures to reduce damage from:Tropicalstorms
FloodsEarthquakesNutritional
interventionsEarly
warningsystems
Improvedwater andsanitation
Vaccinations
Source: World Bank 2013b.Note: The figure shows the median benefit-cost ratios across a range of studies in each category (with a minimum of at least four estimates in each category). Above the dotted line, expected benefits exceed expected costs. The range of estimates within each category can be substantial, reflecting a diversity of intervention types and locations and the sensitivity of estimates to variations in underlying assump-tions. However, in almost all cases, even the 25th percentile of the ranges are above the break-even point.
CHA
PTER
7Ri
sk M
anag
emen
t Pol
icy:
Pre
pari
ng fo
r an
d Co
ping
wit
h A
ggre
gate
Sho
cks
71
A key message of this report is that aggregate shocks vary considerably in nature,
scale, and intensity, calling for careful identification of the risks and predetermined
action plans. Equally diverse is the spectrum of impacts the shocks produce. An
exhaustive policy strategy to confront the impacts requires careful identification of the
risks and the environment in which they arise, the resilience of various groups (such as
families, communities, and countries), the interactions with internal and external factors,
and the obstacles to successful risk management. This chapter draws extensively from
World Development Report 2014 (World Bank 2013b) to outline a simple framework and
derive a set of guiding principles and insights to strengthen risk management policy
in Latin America and the Caribbean, paying special attention to efficiency and equity
considerations and focusing on household resilience and coping ability. Because of the
variety of risks and local conditions, holistic risk management strategies will necessarily
include a range of instruments targeted on specific shocks and social groups.
A policy framework to prepare for and cope with aggregate shocks
Risk management policy needs to address four objectives to help households effectively
prepare for and cope with aggregate shocks. Preparing for aggregate shocks involves
three of the objectives: (1) gaining knowledge to understand the shocks, conditions, and
potential outcomes to reduce the uncertainties faced by people, societies, and countries;
(2) building protection to reduce the probability and size of losses, while increasing the
chances and size of positive outcomes; and (3) acquiring insurance to transfer resources.
Successfully coping with aggregate shocks ex post is covered in the fourth objective: (4)
applying proven coping mechanisms to recover from the losses inflicted by shocks. Box 7.1
explains the advantages of considering both ex ante and ex post objectives.
Strengthening the preparation and coping aspects of risk management to confront
aggregate shocks requires policy actions on five main fronts. They include addressing
(1) market failures and the underprovision of key public goods, (2) social and economic
externalities, (3) weak government incentives, (4) limited resources and information, and
(5) underdeveloped and inflexible social protection (figure 7.2).
(1) Addressing market failures and the underprovision of public goods. One of the
major factors explaining suboptimal risk management is the lack of critical markets
for credit, insurance, and jobs. In some cases, these markets may exist, but fail to
develop completely. Household surveys across the developing world often show that
the demand for credit from farmers is rarely met by financial institutions. In other
contexts, the opposite is also true. Experiments in Ghana and India show that the
demand for weather index insurance is low (around 6–18 percent) at market prices
(Karlan et al. 2014). Better financial inclusion contributes directly to faster recovery
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
72
Box 7.1. Ex Ante or Ex Post Risk Management
Each shock is unique, and the impacts may be diverse. Nonetheless, most components of the response to shocks should be planned before a shock strikes. Even a clear and structured ex post decision-making process can be established beforehand. This will promote a sense of confidence and commitment, increasing the effectiveness of the ex post intervention. Plans and any decision making they entail are kept as clear and simple as possible (Clarke and Dercon 2016).
Several ex ante decisions influence ex post policy responses. The participation of strong and dynamic institutions requires policy agreements and negotiations that take time. First, it is necessary to undertake a rigorous assessment of the potential risks that a country is facing and establish the institution that is to be responsible for each risk response.
Three main dimensions of the ex post response to aggregate shocks can be planned ex ante to avoid negative impacts (Clarke and Dercon 2016). First, a synchronized plan for ex post action that is established previously reduces coordination costs. Typically, a lack of coordination among the government, international organizations, and the population after a shock strikes generates inefficiencies in the delivery of assistance to those who are most affected by the shock. Without adequate planning, the interactions among these three sets of actors may become unnecessarily difficult. Detailed planning ex ante also represents an opportunity to coordinate shock risk reduction.
Second, identification of a dynamic, evidence-based decision-making procedure assures the efficacy of response policies. The efficiency of the response depends on the implementation of policies that truly ameliorate the impact of the shock. The inclusion of scientists in the planning stage may increase the efficacy of the reaction to the shock. There is a particular need to invest in science-based risk information and clearly communicate this information to ensure that the stakeholders and actors are well aware of the contingencies. Moreover, if decisions are to be made based on rules that rely on data, the reliability of the data and statistical systems is crucial to identifying the potential at-risk populations and the responsibilities of those who will be delivering the assistance.
Financing the response and the assistance should also be planned ex ante. Identifying the necessary contingency resources guarantees that the plan can be financially implemented. To have financial and budgetary instruments on standby requires ex ante planning as well as financial discipline given that governments may have incentives to use contingency funds. To avoid these incentives, triggers to enable the use of funds can be established within the financial strategy and matched with initiatives set out in the shock response plan. Traditional reinsurance can be particularly useful in locking in plans for reconstruction, and indexed reinsurance can play the same role in financing indexed early actions. The support of international humanitarian agencies can be sought and agreed on as a backup if plans fail.
CHA
PTER
7Ri
sk M
anag
emen
t Pol
icy:
Pre
pari
ng fo
r an
d Co
ping
wit
h A
ggre
gate
Sho
cks
73
after a disaster while also supporting asset diversification, which reduces vulnerability
in turn. In Guatemala, for instance, universal access to financial institutions would
reduce welfare losses associated with natural disasters by 7.6 percent (Hallegatte et al.
2017). Similarly, basic services and public goods that are essential in managing risks,
such as safe water and sanitation, education, key infrastructure, weather warning
systems, economic and political stability, and the rule of law, are often missing or of
substandard quality. Healthier and more well educated people are better at adopting
behaviors and investing in technologies that contribute to building resilience. Early
warning systems enable preparation against natural hazards, reducing death tolls and
assets losses. Macroeconomic stability acts as a buffer against interest and currency
fluctuations and provides a framework for the improved performance of markets and
for steady growth.
(2) Helping internalize social and economic externalities. Economic activities adopted
by some agents, including risk management strategies, can impose costs on others.
In the absence of banking regulations, excessive risk taking by the banking system
can increase the risks of systemic shocks to the overall economy. Lack of land use and
building regulations can lead to the development of infrastructure in unsafe places
and under unsafe construction codes. The depletion of forests, mangroves, and other
natural assets exacerbate the effects of floods, landslides, storms, and droughts.
Contractionary monetary policy in one country can reduce the local inflation rate,
but disrupt capital flows in other parts of the world. Overuse of antibiotics is causing
some bacteria to become drug-resistant. Free riding among agents who benefit from
risk prevention or mitigation without contributing to the costs likely discourages risk
management. Flood defenses can provide significant protection to infrastructure in
vast areas; yet, they are less likely to be installed and maintained if areas close to the
coastline have to bear most of the burden of the costs. Governments have to establish
regulatory policies and enable collective action that contribute to internalizing the
relevant externalities.
(3) Reforming weak government incentives. Intertemporal political incentives are a
major impediment to adequate risk management. Risk preparation requires a number of
investments, often costly, that potentially yield discernible returns only in the medium
and long term. Governments prefer to devote resources to policies and programs that
produce gains in the short term, even if the gains are smaller. International aid, while
beneficial from a humanitarian perspective, leads to moral hazard, creating incentives
for governments to underinvest in risk preparation and coping. Disaster aid is often
associated with few conditions, and efforts to provide debt forgiveness in times of
distress are unlikely to benefit the poor. Corruption, elite capture, and distortionary
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
74
policies undermine risk management. Government failures across the world and the
lack of a regulatory and institutional framework for risk management translate into a
widespread inability to prepare for crises before they occur. Addressing these failures
requires the development ex ante of coordinated plans for postdisaster action and
agreements on standby financing to ensure that the plans can be implemented (Clarke
and Dercon 2016) (see box 7.1).
(4) Addressing the lack of resources and the lack of information. Investments in risk
management infrastructure and technology usually involve large upfront costs. The
costs of the new protection systems in New Orleans are on the order of $15 billion
(Hallegate et al. 2017). Meeting these costs may be difficult in low-income contexts
where asset ownership is limited, access to credit is lacking, and other market failures
and externalities exist. Whereas the benefits are likely to surpass the costs, the
former are usually realized only in the long term, complicating the decision-making
process. Protection infrastructure rarely attracts sufficient capital largely because
Figure 7.2. A Framework to Switch from Fighting Crises to Managing Risks Effectively
Missingmarkets &
public goods
Cons
trai
nts
Risk
man
agem
ent c
ompo
nent
s
Social &economic
externalities
Weakgovernmentincentives
Resource andinformationconstraints
Adaptive andscalable social
protection
• Basic statistics• Administrative data• Weather forecasts• Mobiles phones• Research
• Immunization• Sound macro policy• Building codes• Safety nets• Early warning systems
• Credit markets• Insurance markets• Safety nets• Self-insurance• Village risk-sharing
Reducing the uncertaninties when confronting risk
Lower the probabilityand size of negativeoutcomes
Transferring resourcesacross agents and over time
Knowledge Protection
Preparation (ex-ante)
Insurance
Coping(ex-post)
Source: Adapted from World Bank 2013b.
CHA
PTER
7Ri
sk M
anag
emen
t Pol
icy:
Pre
pari
ng fo
r an
d Co
ping
wit
h A
ggre
gate
Sho
cks
75
the benefits mostly take the form of avoided losses. Households and governments
on tight budgets may favor current expenditures over investments in risk reduction
and mitigation. Information on the relevant risks and on the benefits and costs of
various actions to prepare and cope with the risks may not exist or, if it does, may
not be available, accurate, or understood. Data from the United States shows that
less than a third of homeowners in flood-prone areas are cognizant of the risk (World
Bank 2013b). Lack of information constrains the ability of public and private agents
to price risks, undermining public and private efforts to insure against shocks and
provide compensation at actuarially fair rates. An option for countries is to leverage
private resources and official development assistance for large and better investments
in resilience.
(5) Developing and strengthening rapidly scalable social protection. Ex ante measures
to increase knowledge, protection, and insurance raise resilience to aggregate shocks,
but, ultimately, households cannot be fully insured. Ex post mechanisms such as
social transfers are necessary to ensure a minimally acceptable standard of living,
especially among the poor. Adaptive and scalable social protection can provide
this type of insurance. Flexible public safety nets can be scaled up during a crisis
by increasing the amount transferred to beneficiaries, relaxing eligibility rules and
conditionalities, extending coverage to new beneficiaries, or creating a new program.
These changes are scaled down after economic and social systems recover. Recent
experiences in the region offer valuable lessons (box 7.2). Mexico’s Ministry of Social
Development, for instance, designed a rapidly scalable national program of temporary
employment through public works. Familias en Acción, a standard conditional cash
transfer program launched in Colombia after the economic crisis of the late 1990s,
relaxed program rules and conditionalities in 2008 after a series of natural events
destroyed critical infrastructure, including schools and health facilities. Expansion of
social safety nets, coupled with prudent macroeconomic management, increased the
resilience to external shocks in Uruguay. Mexico’s Prospera conditional cash transfer
program expanded coverage to one million families to alleviate the effects of the food
price surge. Chile and Guatemala introduced new or redesigned programs to respond
to the 2008–09 global financial crisis. The Latin America and Caribbean region has
succeeded in building safety nets to alleviate poverty. However, many of the nonpoor
are not covered, while remaining vulnerable to sliding back into poverty if they are
hit by a major shock. In the face of this challenge, the many conditional cash transfer
programs in place in the region offer an opportunity to enhance the resilience of shock-
prone and vulnerable households.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
76
Box 7.2. Mexico’s Fund for Natural Disasters
Mexico’s Fund for Natural Disasters (FONDEN) is a good example of a successful risk management and risk copying policy. Aggregate shocks require ex ante coordination to prepare for the risks and ex post action and responsibility to cope with the impacts. FONDEN strikes a good balance between these objectives by committing contingent resources ex ante and setting clear rules to disburse them for reconstruction efficiently and transparently. Designed in the 1990s, FONDEN’s main objective is to support the rapid rehabilitation and reconstruction of infrastructure affected by adverse natural disasters and also act as the contracting authority for market-based risk transfer mechanisms, including insurance and catastrophe bonds. It is funded through the Federal Expenditure Budget and can be used for the rehabilitation and reconstruction of public infrastructure, low-income housing, and certain components of the natural environment. These efforts include the reconstruction of state and federal roads, the provision of funds to rebuild hydraulic, health, and educational infrastructure. Furthermore, FONDEN attempts to avoid recreating vulnerabilities, meaning that funding can be used to rebuild infrastructure at higher standards and to relocate public buildings or communities to safer zones. These actions allow FONDEN to address market failures, the lack of critical public goods, and limited resources and information, as well as formal institutional needs in risk management.
Another critical factor that successful risk management policies must address is related to government incentives. FONDEN tackles this issue by setting clear preestablished rules on execution to balance the need for time-efficient disbursement and the need for accountability and transparency. Whenever a shock or natural disaster occurs, the Federal Ministry of the Interior first issues a declaration of a natural disaster so FONDEN resources become accessible to affected federal agencies or state governments. Once this declaration is made, federal agencies and state governments can apply for funding, and a damage assessment process begins. The damage assessments makes use of innovative information technology, such as geocoding and digital imagery, to ensure efficiency and accuracy, but also entail complementary physical site visits by state and federal officials to assess damage. Based on the findings, the Ministry of the Interior determines the appropriate allocations and requests the Ministry of Finance and Public Credit to authorize the transfer of funds to the service providers implementing the reconstruction works. Moreover, FONDEN may also provide a double gain in economic development, first, by coping with the losses created by natural disasters and, second, by enabling local governments and households to reallocate resources from safer but inefficient low-risk, low-return productive activities to more risky higher-yielding activities, thereby enhancing risk management decisions.
Sources: de Janvry, del Valle, and Sadoulet 2016; World Bank 2012b.
CHA
PTER
7Ri
sk M
anag
emen
t Pol
icy:
Pre
pari
ng fo
r an
d Co
ping
wit
h A
ggre
gate
Sho
cks
77
Three other elements should be considered in risk management policy: careful design,
the identification of risk vulnerabilities, and innovation in market-based products.
First, badly designed policies may make matters worse. Public programs may dislocate
informal risk-sharing institutions by crowding out private transfers across households
within an insurance network. Evidence on Mexico shows that private transfers to households
covered by the Prospera conditional cash transfer program fell after the households joined
the program (Albarran and Attanasio 2004). While the provision of subsidized weather
insurance may provide incentives for households to embrace more risk-taking (a desirable
objective), the extra risk can undermine the access of the households to credit markets and
other financial instruments. Cautious assessment of the trade-offs arising from different
policy options is critical. Second, the effectiveness of public interventions hinges on the
ability to target them on those most in need of insurance and protection against major
shocks. While challenging, emerging evidence (including panel data studies) suggest that
vulnerability to shocks is largely determined by access to and ownership of human and
physical capital as well as by the functioning of and opportunities in labor, product, and
asset markets (Dercon 2004). Most household surveys collect this type of information
to varying degrees. Third, like most policy areas, risk management calls for innovation
in the design of market-based mechanisms to increase knowledge and insurance against
major crises. While instruments have emerged, such as derivative financial instruments,
catastrophe bonds, weather index insurance, and interlinked credit and insurance
contracts, substantial untapped opportunities exist to increase the protection available to
households in Latin America and the Caribbean, especially to the two-thirds of households
that are either poor or experience a relatively high probability of falling back into poverty.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
78
REFE
REN
CES
79
References
Acevedo, Maria Cecilia. 2016. “The Effect of Extreme Hydro-Meteorological Events on Labor Market Outcomes: Evidence from the Colombian Caribbean.” Unpublished paper, World Bank, Washington, DC.
Acosta, Pablo, Javier E. Báez, Germán D. Caruso, and Carlos Carcach. 2016. “The Scars of Civil War: The Long-Term Welfare Effects of the Salvadoran Armed Conflict.” Unpublished paper, World Bank, Washington, DC.
Akresh, Richard, Germán Daniel Caruso, and Harsha Thirumurthy. 2014. “Medium-term Health Impacts of Shocks Experienced in Utero and after Birth: Evidence from Detailed Geographic Information on War Exposure.” NBER Working Paper 20763, National Bureau of Economic Research, Cambridge, MA.
Akter, Sonia, and Bishawjit Mallick. 2013. “The Poverty-Vulnerability-Resilience Nexus: Evidence from Bangladesh.” Ecological Economics 96: 114–24
Albarran, Pedro, and Orazio P. Attanasio. 2004. “Do Public Transfers Crowd Out Private Transfers?: Evidence from a Randomized Experiment in Mexico.” In Insurance against Poverty, edited by Stefan Dercon, 281–304. UNU-WIDER Studies in Development Economics. Helsinki: United Nations University–World Institute for Development Economics Research; New York: Oxford University Press.
Alderman, Harold, and Christina H. Paxson. 1994. “Do the Poor Insure? A Synthesis of the Literature on Risk and Consumption in Developing Countries.” In Economics in a Changing World, vol. 4 Development, Trade, and the Environment, edited by Edmar L. Bacha, 48–78. New York: St. Martin’s Press.
Andalón, Mabel, João Pedro Azevedo, Carlos Rodríguez-Castelán, Viviane Sanfelice, and Daniel Valderrama-González. 2016. “Weather Shocks and Health at Birth in Colombia.” World Development 82: 69–82.
Antle, John. 1987. “Econometric Estimation of Producers’ Risk Attitudes.” American Journal of Agricultural Economics 69 (3): 509–22.
Atkin, David. 2016. “Endogenous Skill Acquisition and Export Manufacturing in Mexico.” American Economic Review 106 (8): 2046–85.
———, eds. 2001. Portrait of the Poor: An Assets-Based Approach. Washington, DC: Inter-American Development Bank; Baltimore: Johns Hopkins University Press.
Báez, Javier E. 2011. “Civil Wars beyond Their Borders: The Human Capital and Health Consequences of Hosting Refugees.” Journal of Development Economics 96 (2): 391–408.
Báez, Javier, Alejandro de la Fuente, and Indhira Santos. 2010. “Do Natural Disasters Affect Human Capital? An Assessment Based on Existing Empirical Evidence.” IZA Discussion Paper DP 5164, Institute for the Study of Labor, Bonn, Germany.
Báez, Javier E., Leonardo Lucchetti, María Eugenia Genoni, and Mateo Salazar. 2016. “Gone with the Storm: Rainfall Shocks and Household Well-Being in Guatemala.” Journal of Development Studies online. http://dx.doi.org/10.1080/00220388.2016.1224853.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
80
Báez, Javier E., Germán Caruso, Valerie Mueller, and Chiyu Niu. 2017. “Droughts Augment Youth Migration in Northern Latin America and the Caribbean.” Climatic Change, vol. 140, (3), pp 423–435.
———. 2017. “Heat Exposure and Youth Migration in Central America and the Caribbean.” American Economic Review, Papers and Proceedings (forthcoming).
Báez, Javier E., and Indhira V. Santos. 2007. “Children’s Vulnerability to Weather Shocks: A Natural Disaster as a Natural Experiment.” Unpublished paper, World Bank, Washington, DC.
———. 2009. “On Shaky Ground: The Effects of Earthquakes on Household Income and Poverty.” Research for Public Policy: MDGs and Poverty, MDG-02-2009, Regional Bureau for Latin America and the Caribbean, United Nations Development Programme, New York.
Baird, Sarah, Jed Friedman, and Norbert Schady. 2011. “Aggregate Income Shocks and Infant Mortality in the Developing World.” Review of Economics and Statistics 93 (3): 847–56.
Baker, Judy, Rakhi Basu, Maureen Cropper, Somik V. Lall, and Akie Takeuchi. 2005. “Urban Poverty and Transport: The Case of Mumbai.” Policy Research Working Paper 3693, World Bank, Washington, DC.
Banerjee, Abhijit V., and Esther Duflo. 2007. “The Economic Lives of the Poor.” Journal of Economic Perspectives 21 (1): 141–68.
Banerjee, Abhijit V., and Sendhil Mullainathan. 2010. “The Shape of Temptation: Implications for the Economic Lives of the Poor.” NBER Working Paper 15973, National Bureau of Economic Research, Cambridge, MA.
Barriga Cabanillas, Oscar, María Ana Lugo, Hannah Nielsen, Carlos Rodríguez-Castelán, and María Pía Zanetti. 2015. “Is Uruguay More Resilient This Time? Distributional Impacts of a Crisis Similar to the 2001/02 Argentine Crisis.” Journal of Banking and Financial Economics 2 (4): 64–90.
Barro, Robert J., and José F. Ursúa. 2008. “Macroeconomic Crises since 1870.” NBER Working Paper 13940, National Bureau of Economic Research, Cambridge, MA. http://www.nber.org/papers/w13940.
Beegle, Kathleen, Rajeev H. Dehejia, and Roberta Gatti. 2006. “Child Labor and Agricultural Shocks.” Journal of Development Economics 81 (1): 80–96.
Behrman, Jere R., and Mark R. Rosenzweig. 2004. “Returns to Birthweight.” Review of Economics and Statistics 86 (2): 586–601.
Bernheim, B. Douglas, Ray Debraj, and Sevin Yeltekin. 2013. “Poverty and Self-Control.” Technical Report, National Bureau of Economic Research, Cambridge, MA.
Blanco, Federico, and Cristina A. Valdivia. 2006. “Child Labour in Venezuela: Children’s Vulnerability to Macroeconomic Shocks.” UCW Project Working Paper, Understanding Children’s Work Project, Faculty of Economics, University of Rome “Tor Vergata,” Rome.
Bliss, C. J., and Nicholas Herbert Stern. 1982. Palanpur: The Economy of an Indian Village. Oxford, United
Bryan, Gharad. 2010. “Ambiguity and Insurance.” Job market paper (November 14), Department of Economics, Yale University, New Haven, CT.
REFE
REN
CES
81
Bronfman, Javier and Gonzalo Reyes. 2010. “Poverty, Vulnerability and Resilience to Economic Shocks: Simulating the Impact of Social Programs in Jamaica.” Background paper, World Bank, Washington, DC.
Bussolo, Maurizio, and Luis F. López-Calva. 2014. Shared Prosperity: Paving the Way in Europe and Central Asia. Europe and Central Asia Studies. Washington, DC: World Bank.
Bustelo, Monserrat, Mary Arends-Kuenning, and Leonardo Lucchetti. 2012. “Persistent Impact of Natural Disasters on Child Nutrition and Schooling: Evidence from the 1999 Colombian Earthquake.” IZA Discussion Paper DP 6354, Institute for the Study of Labor, Bonn, Germany.
Calderón, Valentina, and Ana María Ibáñez. 2009. “Labor Market Effects of Migration-Related Supply Shocks: Evidence from Internal Refugees in Colombia.” MICROCON Research Working Paper 14, Documento CEDE 2009–14, Centre for Economic Development Studies, University of Los Andes, Bogotà, Colombia.
Carter, Michael R., and Christopher B. Barrett. 2006. “The Economics of Poverty Traps and Persistent Poverty: An Asset-Based Approach.” Journal of Development Studies 42 (2): 178–99.
Carter, Michael R., and Marco Castillo. 2005. “Coping with Disaster: Morals, Markets and Mutual Insurance, Using Economic Experiments to Study Recovery from Hurricane Mitch.” In The Social Economics of Poverty: On Identities, Communities, Groups and Networks, edited by Christopher B.. Barrett, 256–75. Priorities in Development Economics Series. New York: Routledge.
Carter, Michael R., and Sarah A. Janzen. 2015. “Social Protection in the Face of Climate Change: Targeting Principles and Financing Mechanisms.” Policy Research Working Paper 7476, World Bank, Washington, DC.
Carter, Michael R., Peter Little, Tewodaj Mogues, and Workneh Negatu. 2005. “Shocks, Sensitivity, and Resilience: Tracking the Economic Impacts of Environmental Disaster on Assets in Ethiopia and Honduras.” Staff Paper 489, Department of Agricultural and Applied Economics, University of Wisconsin, Madison, WI.
———. 2007. “Poverty Traps and the Long-Term Consequences of Natural Disasters in Ethiopia and Honduras.” World Development 35 (5): 835–56.
Carter, Michael R., and Travis J. Lybbert. 2012. “Consumption Versus Asset Smoothing: Testing the Implications of Poverty Trap Theory in Burkina Faso.” Journal of Development Economics 99 (2): 255–64.
Caruso, Germán. 2017. “The Legacy of Natural Disasters: The Intergenerational Impact of 100 Years of Natural Disasters in Latin America.” Forthcoming in Journal of Development Economics.”
Cavallo, Eduardo, Sebastián Galiani, Ilan Noy, and Juan Pantano. 2010. “Catastrophic Disasters and Economic Growth.” IDB Working Paper DB-WP-183, Inter-American Development Bank, Washington, DC.
Clarke, Daniel J. 2016. “A Theory of Rational Demand for Index Insurance.” American Economic Journal: Microeconomics 8 (1): 283–306.
Clarke, Daniel J., and Stefan Dercon. 2016. Dull Disasters?: How Planning Ahead Will Make a Difference. London: Oxford University Press.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
82
Cord, Louise, Oscar Barriga Cabanillas, Leonardo Lucchetti, Carlos Rodríguez-Castelán, Liliana D. Sousa, and Daniel Valderrama. 2017. “Inequality Stagnation in Latin America in the Aftermath of the Global Financial Crisis.” Review of Development Economics 21 (1): 157–81.
Cord, Louise, María Eugenia Genoni, and Carlos Rodríguez-Castelán, eds. 2015. Shared Prosperity and Poverty Eradication in Latin America and the Caribbean. Washington, DC: World Bank.
Dalton, Patricio S., Sayantan Ghosal, and Anandi Mani. 2016. “Poverty and Aspirations Failure.” Economic Journal 126 (590): 165–88.
de Hoyos, Rafael, Halsey Rogers,, and Miguel Székely. 2016. “Out of School and Out of Work: Risk and Opportunities for Latin America’s Ninis.” World Bank, Washington, DC.
de Janvry, Alain, Alejandro del Valle, and Elisabeth Sadoulet. 2016. “Insuring Growth: The Impact of Disaster Funds on Economic Reconstruction in Mexico.” Policy Research Working Paper 7714, World Bank, Washington, DC.
de Janvry, Alain, Frederico Finan, Elisabeth Sadoulet, Renos 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 (2): 349–73.
de Janvry, Alain, and Elisabeth Sadoulet. 2006. “Progress in the Modeling of Rural Households’ Behavior under Market Failures.” In Poverty, Inequality, and Development: Essays in Honor of Erik Thorbecke, edited by Alain de Janvry and Ravi Kanbur, 155–82. New York: Springer.
de la Fuente, Alejandro, Eduardo Ortiz-Juárez, and Carlos Rodríguez-Castelán. 2017. “Living on the Edge: Vulnerability to Poverty and Public Transfers in Mexico.” Oxford Development Studies (forthcoming).
del Ninno, Carlo, Paul A. Dorosh, Lisa C. Smith, and Dilip K. Roy. 2001. “The 1998 Floods in Bangladesh: Disaster Impacts, Household Coping Strategies and Response.” Research Report 122. International Food Policy Research Institute, Washington, DC.
Dercon, Stefan. 1996. “Risk, Crop Choice, and Savings: Evidence from Tanzania.” Economic Development and Cultural Change 44 (3): 485–514.
———. 2004. “Growth and Shocks: Evidence from Rural Ethiopia.” Journal of Development Economics 74 (2): 309–29.
Dercon, Stefan, and Luc Christiaensen. 2011. “Consumption Risk, Technology Adoption, and Poverty Traps: Evidence from Ethiopia.” Journal of Development Economics 96 (2): 159–73.
Dercon, Stefan, and John Hoddinott. 2004. “Health, Shocks, and Poverty Persistence.” In Insurance against Poverty, edited by Stefan Dercon, 124–36. UNU-WIDER Studies in Development Economics. Helsinki: United Nations University–World Institute for Development Economics Research; New York: Oxford University Press.
Deschênes, Olivier, and Michael Greenstone. 2011. “Climate Change, Mortality, and Adaptation: Evidence from Annual Fluctuations in Weather in the US.” American Economic Journal: Applied Economics 3 (4): 152–85.
De Weerdt, Joachim. (2004). “Risk-Sharing and Endogenous Network Formation.” In Insurance against Poverty, edited by Stefan Dercon, 197–216. UNU-WIDER Studies in Development Economics. Helsinki: United Nations University–World Institute for Development Economics Research; New York: Oxford University Press.
REFE
REN
CES
83
D’Souza, Anna, and Dean Jolliffe. 2014. “Food Insecurity in Vulnerable Populations: Coping with Food Price Shocks in Afghanistan.” American Journal of Agricultural Economics 96 (3): 790–812.
Dupas, Pascaline. 2011. “Do Teenagers Respond to HIV Risk Information? Evidence from a Field Experiment in Kenya.” American Economic Journal: Applied Economics 3 (1): 1–34.
Dupas, Pascaline, and Jonathan Robinson. 2012. “The (Hidden) Costs of Political Instability: Evidence from Kenya’s 2007 Election Crisis.” Journal of Development Economics 99 (2): 314–29.
Duque, Valentina. 2016. “The Hidden Cost and Lasting Legacies of Violence on Education: The Case of Colombia.” Working paper, Columbia University.
Duque, Valentina and Maria Rosales. 2016. “Can Conditional Cash Transfers Build up Resilience? Evidence from Colombia’s Familias en Acción.” Working paper, CAF, Bogota, Colombia.
Duryea, Suzanne, David Lam, and Deborah Levison. 2007. “Effects of Economic Shocks on Children’s Employment and Schooling in Brazil.” Journal of Development Economics 84 (1): 188–214.
Duryea, Suzanne, and Melisa Morales. 2011. “Effects of the Global Financial Crisis on Children’s School and Employment Outcomes in El Salvador.” Development Policy Review 29 (5): 527–46.
Edin, Per-Anders, and Magnus Gustavsson. 2008. “Time Out of Work and Skill Depreciation.” Industrial and Labor Relations Review 61 (2), 163–80.
Enamorado, Ted, Luis F. López-Calva, and Carlos Rodríguez-Castelán. 2014. “Crime and Growth Convergence: Evidence from Mexico.” Economics Letters 125 (1): 9–13.
Fafchamps, Marcel, Christopher Udry, and Katherine Czukas. 1998. “Drought and Saving in West-Africa: Are Livestock a Buffer Stock.” Journal of Development Economics 55 (2): 273–305.
Fallon, Peter R., and Robert E. Lucas. 2002. “The Impact of Financial Crises on Labor Markets, Household Incomes, and Poverty: A Review of Evidence.” World Bank Research Observer 17 (1): 21–45.
Feng, Shuaizhang, Alan B. Krueger, and Michael Oppenheimer. 2010. “Linkages among Climate Change, Crop Yields, and Mexico–US Cross-Border Migration.” Proceedings of the National Academy of Sciences 107 (32): 14257–62.
Fergusson, Leopoldo and Andres Zambrano. 2016. Endogenous Persistent Shocks and Households’ Welfare. Working paper, Universidad de Los Andes, Bogota, Colombia.
Freije, Samuel, Gladys López-Acevedo, and Eduardo Rodríguez-Oreggia. 2011. “Effects of the 2008–09 Economic Crisis on Labor Markets in Mexico.” Policy Research Working Paper 5840, World Bank, Washington, DC.
Fuchs, Alan. 2014. “Shocks and Poverty in Haiti.” Paper presented at the World Bank’s “Poverty and Climate Change Flagship: An Overview of the LAC Region” workshop, Washington, DC, September 14.
Galdo, Jose. 2013. “The Long-Run Labor-Market Consequences of Civil War: Evidence from the Shining Path in Peru.” Economic Development and Cultural Change 61 (4): 789–823.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
84
Gammage, Sarah, and Jorge Fernandez. 2000. “Conflict, Displacement, and Reintegration: Household Survey Evidence from El Salvador.” New Issue in Refugee Research, Working Paper 25, International Center for Research on Women, Washington, DC.
García-Verdú, Rodrigo. 2002. “Evaluation of Conditional Income Support Programs: The Case of Mexico’s Progresa.” PhD dissertation, Department of Economics, University of Chicago, Chicago.
Gentle, Popular, Rik Thwaites, Digby Race, and Kim Alexander. 2014. “Differential Impacts of Climate Change on Communities in the Middle Hills Region of Nepal.” Natural Hazards 74 (2): 815–36.
Goland, Carol. 1993. “Agricultural Risk Management through Diversity: Field Scattering in Cuyo Cuyo, Peru.” C&AFE: Culture, Agriculture, Food and the Environment 13 (45-46): 8–13.
Gonzalez-Barrera, Ana. 2015. “More Mexican Leaving Than Coming to the U.S.” November 19, Pew Research Center, Washington, DC.
Hallegatte, Stephane, Mook Bangalore, Laura Bonzanigo, Marianne Fay, Tomaro Kane, Ulf Narloch, Julie Rozenberg, David Treguer, and Adrien Vogt-Schilb. 2016. Shock Waves: Managing the Impacts of Climate Change on Poverty. Climate Change and Development Series. Washington, DC: World Bank.
Hallegatte, Stephane, Adrien Vogt-Schilb, Mook Bangalore, and Julie Rozenberg. 2017. Unbreakable: Building the Resilience of the Poor in the Face of Natural Disasters. Climate Change and Development Series. Washington, DC: World Bank.
Halliday, Timothy. 2006. “Migration, Risk, and Liquidity Constraints in El Salvador.” Economic Development and Cultural Change 54 (4): 893–925.
Hermida, Priscila. 2011. “The Long-Term Effect of Natural Disasters: Health and Education in Guatemala after the 1976 Earthquake.” Paper presented at the Fifth Annual PopPov Conference on Population, Reproductive Health, and Economic Development, Marseille, January, 19–21.
Holt, William G., ed. 2014. From Sustainable to Resilient Cities: Global Concerns and Urban Efforts. Research in Urban Sociology Series 14. Bingley, United Kingdom: Emerald.
Ibáñez, Ana María, and Andrés Moya. 2006. “The Impact of Intra-State Conflict on Economic Welfare and Consumption Smoothing: Empirical Evidence for the Displaced Population in Colombia.” HiCN Working Paper 23, Households in Conflict Network, Institute of Development Studies, University of Sussex, Brighton, United Kingdom.
IDMC (Internal Displacement Monitoring Center). 2011. “Summary: Guatemala, Internal Displacement in Brief.” December 31. IDMC, Geneva. http://www.internal-displacement.org/americas/guatemala/summary.
IFRC (International Federation of Red Cross and Red Crescent Societies). 2012. World Disasters Report, 2012: Focus on Forced Migration and Displacement. Geneva: IFRC.
ILO (International Labour Organization). 2012. “World of Work 2012: Better Jobs for a Better Economy.” International Institute for Labour Studies, ILO, Geneva.
———. 2015. World Employment and Social Outlook: The Changing Nature of Jobs. Geneva: ILO.
IMF (International Monetary Fund). 2015. “Mexico: Selected Issues.” IMF Country Report 15/314 (November), IMF, Washington DC.
REFE
REN
CES
85
Inter-American Development Bank. 2017. “The costs of crime and violence: new evidence and insights in Latin America and the Caribbean” (Laura Jaitman, ed.). Washington, DC.
Ishizawa, Oscar A., and Juan Jose Miranda. 2016. “Weathering Storms: Understanding the Impact of Natural Disasters on the Poor in Central America.” Policy Research Working Paper 7692, World Bank, Washington DC.
Jaitman, Laura, ed. 2015. “The Welfare Costs of Crime and Violence in Latin America and the Caribbean.” Inter-American Development Bank, Washington, DC.
Janzen, Sarah A., Michael R. Carter, and Munenobu Ikegami. 2015. “Asset Insurance Markets and Chronic Poverty.” Unpublished working paper (December), Montana State University, Bozeman, MT. http://www.montana.edu/sjanzen/valuingassetinsurancev11.pdf.
Kahn, Matthew E. 2005. “The Death Toll from Natural Disasters: The Role of Income, Geography, and Institutions.” Review of Economics and Statistics 87 (2): 271–84.
Karlan, Dean, Robert Osei, Isaac Osei-Akoto, and Christopher Udry. 2014. “Agricultural Decisions after Relaxing Credit and Risk Constraints.” Quarterly Journal of Economics 129 (2): 597–652.
Kazianga, Harounan, and Christopher Udry. 2006. “Consumption Smoothing? Livestock, Insurance, and Drought in Rural Burkina Faso.” Journal of Development Economics 79 (2): 413–46.
Kruger, Diana I. 2007. “Coffee Production Effects on Child Labor and Schooling in Rural Brazil.” Journal of Development Economics 82 (2): 448–63.
Lamb, Russell L. 2003. “Inverse Productivity: Land Quality, Labor Markets, and Measurement Error.” Journal of Development Economics 71 (1): 71–95.
León, Gianmarco. 2012. “Civil Conflict and Human Capital Accumulation: The Long-Term Effects of Political Violence in Perú.” Journal of Human Resources 47 (4): 991–1022.
López, Ramón E. 1984. “Estimating Labour Supply and Production Decisions of Self-Employed Farm Producers.” European Economic Review 24 (1): 61–82.
———. 1986. “Structural Models of the Farm Household That Allow for Interdependent Utility and Profit-Maximization Decisions.” In Agricultural Household Models: Extensions, Applications, and Policy, edited by Inderjit Singh, Lyn Squire, and John Strauss, 306–26. Washington, DC: World Bank; Baltimore: Johns Hopkins University Press.
López-Calva, Luis F., and Eduardo Ortiz-Juárez. 2014. “A Vulnerability Approach to the Definition of the Middle Class.” Journal of Economic Inequality 12 (1): 23–47.
López-Calva, Luis F., and Carlos Rodríguez-Castelán. 2016. “Pro-Growth Equity: A Policy Framework for the Twin Goals.” Policy Research Working Paper 7897, World Bank, Washington, DC.
Lucchetti, Leonardo. 2011. “Subjective Poverty and Natural Disasters: Evidence from the 2007 Earthquake in Peru.” Working paper, Department of Economics, University of Illinois at Urbana-Champaign, Urbana, IL.
Lybbert, Travis J., and John McPeak. 2012. “Risk and Intertemporal Substitution: Livestock Portfolios and Off‐Take among Kenyan Pastoralists.” Journal of Development Economics 97 (2): 415–26.
Magrin, Graciela O., José A. Marengo, Jean-Phillipe Boulanger, Marcos S. Buckeridge, Edwin Castellanos, Germán Poveda, Fabio R. Scarano, and Sebastián Vicuña. 2014. “Central
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
86
and South America.” In Regional Aspects. Part B of Climate Change 2014: Impacts, Adaptation, and Vulnerability, edited by Vicente R. Barros, Christopher B. Field, David Jon Dokken, Michael D. Mastrandrea, Katharine J. Mach, T. Eren Bilir, Monalisa Chatterjee, et al., 1499–1566. Geneva: Intergovernmental Panel on Climate Change; New York: Cambridge University Press.
Malasquez, Eduardo. 2016. “Long-Term Effects of Civil War on Education and Labor Market Outcomes: Evidence from the Peruvian Civil Conflict.” Working paper, Department of Economics, University of Illinois at Urbana-Champaign, Urbana, IL.
Martinez-Cruz, Adan L., and Carlos Rodríguez-Castelán. 2016. “Crime and Persistent Punishment: A Long-Run Perspective on the Links between Violence and Chronic Poverty in Mexico.” Policy Research Working Paper 7706, World Bank, Washington, DC.
Masozera, Michel, Melissa Bailey, and Charles Kerchner. 2007. “Distribution of Impacts of Natural Disasters across Income Groups: A Case Study of New Orleans.” Ecological Economics 63 (2–3): 299–306.
McIntosh, Craig, Alexander Sarris, and Fotis Papadopoulos. 2013. “Productivity, Credit, Risk, and the Demand for Weather Index Insurance in Smallholder Agriculture in Ethiopia.” Agricultural Economics 44 (4–5): 399–417.
———. 2006. “The Consumer Response to the Mexican Peso Crisis.” Economic Development and Cultural Change 55 (1): 139–72.
Messina, Jane P., Moritz U. G. Kraemer, Oliver J. Brady, David M. Pigott, Freya M. Shearer, Daniel J. Weiss, Nick Golding, et al. 2016. “Mapping Global Environmental Suitability for Zika Virus.” eLife 5 (April 19): e15272. https://elifesciences.org/content/5/e15272/article-info.
Molina, Teresa. 2014. “Regional Migration, Co-Insurance and Economic Shocks: Evidence from Nicaragua.” Working paper, Paris School of Economics, Paris.
Monras, Joan. 2015. “Economic Shocks and Internal Migration.” IZA Discussion Paper DP 8840, Institute for the Study of Labor, Bonn, Germany.
Morduch, Jonathan J. 1990. “Risk, Production, and Saving: Theory and Evidence from Indian Households.” Working paper, Department of Economics, Harvard University, Cambridge, MA.
———. 1995. “Income Smoothing and Consumption Smoothing,” Journal of Economic Perspectives 9 (3): 103–14.
Morris, Saul S., Oscar Neidecker-Gonzales, Calogero Carletto, Marcial Munguia, Juan Manuel Medina, and Quentin Wodon. 2001. “Hurricane Mitch and the Livelihoods of the Rural Poor in Honduras.” World Development 30 (1): 39–60.
Mueller, Valerie A., and Daniel E. Osgood. 2009. “Long-Term Consequences of Short-Term Precipitation Shocks: Evidence from Brazilian Migrant Households.” Agricultural Economics 40 (5): 573–86.
Neumayer, Eric, and Thomas Plümper. 2007. “The Gendered Nature of Natural Disasters: The Impact of Catastrophic Events on the Gender Gap in Life Expectancy, 1981–2002.” Annals of the Association of American Geographers 97 (3): 551–66.
Nguyen, van K. 2011. “Building Livelihood Resilience in Changing Climate.” Paper presented at the Asia Regional Conference, Kuala Lumpur, Malaysia, March 3–5.
REFE
REN
CES
87
Opondo, Denis Opiyo. 2013. “Erosive Coping after the 2011 Floods in Kenya.” International Journal of Global Warming 5 (4): 452–66.
Paes-Sousa, Romulo, Ferdinando Regalia, and Marco Stampini. 2013. “Conditions for Success in Implementing CCT Programs: Lessons for Asia from Latin America and the Caribbean.” Policy Brief IDB-PB 192, Social Protection and Health Division, Inter-American Development Bank, Washington, DC.
Paxson, Christina H. 1992. “Using Weather Variability to Estimate the Response of Savings to Transitory Income in Thailand.” American Economic Review 82 (1): 15–33.
Pelling, Mark. 1997. “What Determines Vulnerability to Floods: A Case Study in Georgetown, Guyana.” Environment and Urbanisation 9 (1): 203–26.
Piñeiro, Valeria, Samuel Morley, and Pablo Elverdin. 2015. “Los Efectos de la Roya en las Economías Centroamericanas.” IFPRI Documento de Discusión 01457SP, International Food Policy Research Institute, Washington, DC.
Premand, Patrick, and Renos Vakis. 2010. “Do Shocks Affect Poverty Persistence? Evidence Using Welfare Trajectories from Nicaragua.” Well-Being and Social Policy 6 (1): 95–129.
Putnam, Robert D. 1995. “Bowling Alone: America’s Declining Social Capital.” Journal of Democracy 6 (1): 65–78.
Ranger, Nicola, Stéphane Hallegatte, Sumana Bhattacharya, Murthy Bachu, Satya Priya, K. Dhore, Farhat Rafique, P. Mathur, and Nicolas Naville. 2011. “An Assessment of the Potential Impact of Climate Change on Flood Risk in Mumbai.” Climate Change 104 (1): 139–67.
Reardon, Thomas, Julio Berdegué, Christopher B. Barrett, and Kostas Stamoulis. 2006. “Household Income Diversification into Rural Nonfarm Activities.” In Transforming the Rural Nonfarm Economy: Opportunities and Threats in the Developing World, edited by Steven Haggblade, Peter B. R. Hazell, and Thomas Reardon, 115–40. Washington, DC: International Food Policy Research Institute; Baltimore: Johns Hopkins University Press.
Robles, Miguel, and Máximo Torero. 2010. “Understanding the Impact of High Food Prices in Latin America.” Economia 10 (2): 117–64.
Rodríguez-Meza, Jorge, and Claudio González-Vega. 2004. “Household Income Dynamics and Poverty Traps in Rural El Salvador.” Paper presented at the Agricultural and Applied Economics Association’s annual meeting, Denver, August 1–4.
Rodríguez-Oreggia, Eduardo, Alejandro de la Fuente, Rodolfo de la Torre, and Hector A. Moreno. 2013. “Natural Disasters, Human Development, and Poverty at the Municipal Level in Mexico.” Journal of Development Studies 49 (3): 442–55.
Rosales, María Fernanda. 2014. “Impact of Early Life Shocks on Human Capital Formation: El Niño Floods in Ecuador.” IDB Working Paper IDB-WP-503, Inter-American Development Bank, Washington, DC.
Rose, Adam, Juan Benavides, Stephanie E. Chang, Philip Szczesniak, and Dongsoon Lim. 1997. “The Regional Economic Impact of an Earthquake: Direct and Indirect Effects of Electricity Lifeline Disruptions.” Journal of Regional Science 37 (3): 437–58.
Rose, Elaina. 1999. “Consumption Smoothing and Excess Female Mortality in Rural India.” Review of Economics and Statistics 81 (1): 41–49.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
88
Rosenzweig, Mark R., and Hans Peter Binswanger. 1993. “Wealth, Weather Risk, and the Composition and Profitability of Agricultural Investments.” Economic Journal 103 (416): 56–78.
Rosenzweig, Mark R. and Oded Stark. 1989. “Consumption Smoothing, Migration, and Marriage: Evidence from Rural India.” Journal of Political Economy 97 (4): 905–26.
Rosenzweig , Mark R., and Kenneth I. Wolpin. 1993. “Credit Market Constraints, Consumption Smoothing, and the Accumulation of Durable Production Assets in Low-Income Countries: Investment in Bullocks in India.” Journal of Political Economy 101 (2): 223–44.
Sadoulet, Elisabeth, Frederico Finan, Alain de Janvry, and Renos Vakis. 2004. “Can Conditional Cash Transfer Programs Improve Social Risk Management? Lessons for Education and Child Labor Outcomes.” SP Discussion Paper 420 (December), Social Protection, World Bank, Washington, DC.
Saks, Raven E., and Abigail Wozniak. 2011. “Labor Reallocation over the Business Cycle: New Evidence from Internal Migration.” Journal of Labor Economics 29 (4): 697–739.
Santos, Indhira V. 2006. “Risk-Sharing and the Role of Inter-Household Transfers: Evidence from Hurricane Mitch in Nicaragua.” Unpublished PhD Dissertation, April, Kennedy School of Government, Harvard University, Cambridge, MA.
———. 2007. “Disentangling the Effects of Natural Disasters on Children: 2001 Earthquakes in El Salvador.” PhD Dissertation, Kennedy School of Government, Harvard University, Cambridge, MA.
Santos, Rafael J. 2015. “All that Glitters Is not Gold: The Gold Boom, Child Labor, and Schooling in Colombia.” Centre for Economic Development Studies, University of Los Andes, Bogotà, Colombia.
Santos. 2016a. “La Niña y los niños: Effects of an Unexpected Winter on Early Life Human Capital and Family Responses.” Working paper, Universidad de los Andes.
Santos. 2016b. “Not all that Glitters Is Gold: Gold Boom, Child Labor and Schooling in Colombia.” Working paper, Universidad de los Andes.
Schady, Norbert R. 2004. “Do Macroeconomic Crises Always Slow Human Capital Accumulation?” World Bank Economic Review 18 (2): 131–54.
Schlenker, Wolfram, and Michael J. Roberts. 2008. “Estimating the Impact of Climate Change on Crop Yields: The Importance of Nonlinear Temperature Effects.” NBER Working Paper 13799, National Bureau of Economic Research, Cambridge, MA.
Scholte, Robert S., Gerard J. van den Berg, and Maarten Lindeboom. 2012. “Long-Run Effects of Gestation during the Dutch Hunger Winter Famine on Labor Market and Hospitalization Outcomes.” Working Paper 2012:15, Institute for Evaluation of Labour Market and Education Policy, Uppsala, Sweden.
Scoones, Ian. 1998. “Sustainable Rural Livelihoods: A Framework for Analysis.” IDS Working Paper 72, Institute of Development Studies, Brighton, United Kingdom.
SEDLAC (Socio-Economic Database for Latin America and the Caribbean), Center for Distributive, Labor, and Social Studies, Universidad de La Plata, La Plata, Argentina; World Bank, Washington, DC, http://sedlac.econo.unlp.edu.ar/eng/index.php.
Sen, Amartya. 1997. Resources, Values, and Development. Cambridge, MA: Harvard University Press.
REFE
REN
CES
89
Seth, Anji, Kátia Fernandes, and Suzana J. Camargo. 2015. “Two Summers of São Paulo Drought: Origins in the Western Tropical Pacific.” Geophysical Research Letters 42 (24): 10,816–23.
Shah, Anuj K., Sendhil Mullainathan, and Eldar Shafir. 2012. “Some Consequences of Having Too Little.” Science 338 (6107): 682–85.
Singh, Inderjit Lyn Squire, and John Strauss, eds. 1986. Agricultural Household Models: Extensions, Applications, and Policy. Washington, DC: World Bank; Baltimore: Johns Hopkins University Press.
Skoufias, Emmanuel, and Susan W. Parker. 2006. “Job Loss and Family Adjustments in Work and Schooling during the Mexican Peso Crisis.” Journal of Population Economics 19 (1): 163–81.
Tesliuc, Emil, and Kathy Lindert. 2003. “The Impact of Vulnerability and Shocks on the Poor: The Guapa Program in Guatemala.” en breve 22 (April), World Bank, Washington, DC.
Torche, Florencia. 2016. “Unequal Trajectories: Prenatal Stress, Intergenerational Stratification, and Children’s Outcomes.” Working paper, New York University, New York.
Toro, Jose. 2016. “Ex ante and ex post effects of Natural Disasters on Migration: Evidence from Peru.” Working paper, Pontificia Universidad Catolica de Peru, Lima, Peru.
Townsend, Robert. 1994. “Risk and Insurance in Village India.” Econometrica 62 (3): 539–91.
UNHCR (United Nations High Commissioner for Refugees). 2015. UNHCR Global Report 2014. Geneva: UNHCR.
UNISDR (United Nations Office for Disaster Risk Reduction). 2009. Global Assessment Report 2009: Disaster Risk Reduction, Risk and Poverty in a Changing Climate. Geneva: UNISDR. http://www.preventionweb.net/english/hyogo/gar/2009/.
UNODC (United Nations Office on Drugs and Crime). 2014. Global Study on Homicide 2013: Trends, Contexts, Data. Vienna: United Nations.
Walker, Thomas S., and James G. Ryan. 1990. Village and Household Economics in India’s Semi-arid Tropics. Baltimore: Johns Hopkins University Press.
Wodon, Quentin, Andrea Liverani, George Joseph, and Nathalie Bougnoux, eds. 2014. Climate Change and Migration: Evidence from the Middle East and North Africa. World Bank Study Series. Washington, DC: World Bank.
World Bank. 2009. Global Development Finance 2009: Charting a Global Recovery. 2 vols. Washington, DC: World Bank.
———. 2012a. “Labor Markets for Inclusive Growth.” Mexico Policy Note 4, World Bank, Washington DC.
———. 2012b. “FONDEN, Mexico’s Natural Disaster Fund, a Review.” World Bank, Washington, DC.
———. 2013a. United Mexican States: Mexico Policy Notes. Policy Note, Report ACS3529 (May). Washington, DC: World Bank.
———. 2013b. World Development Report 2014, Risk and Opportunity: Managing Risk for Development. Washington, DC: World Bank.
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
90
———. 2014a. “Social Gains in the Balance: A Fiscal Policy Challenge for Latin America and the Caribbean.” Poverty and Labor Brief, Report 85162 rev (February), Latin America and Caribbean Region, World Bank, Washington, DC.
———. 2014b. “When Prosperity Is Not Shared: The Weak Links between Growth and Equity in the Dominican Republic.” January, World Bank, Washington, DC.
———. 2015. “Working to End Poverty in Latin America and the Caribbean: Workers, Jobs, and Wages.” LAC Poverty and Labor Brief (June), World Bank, Washington, DC.
World Bank and the United Nations. 2010. Natural Hazards, UnNatural Disasters: The Economics of Effective Prevention. Washington, DC: World Bank.
AN
NEX
1Su
mm
ary
of th
e Fi
ndin
gs o
f the
Bac
kgro
und
Pape
rs
91
Annex 1. Summary of the Findings of the Background
Papers
Human welfare impacts of aggregate shocks
Outcomes Country Sources Main findings
Consumption, income and poverty
Central AmericaIshizawa and Miranda (2016)
One standard deviation in the intensity of a hurricane windstorm leads to a contraction in the growth of total per capita GDP of between 0.9 and 1.6 percent and a decline in total income and in labor income by 3 percent, which raises the moderate and extreme poverty rates by 1.5 percentage points
ColombiaFergusson and Zambrano (2016)
Shocks are persistent: experiencing an adverse shock increases future vulnerability by about 9 to 11 percentage points, making households in the middle of the wealth distribution significantly decrease their consumption when they are hit by an adverse shock; this generates a poverty trap among households in the first two quartiles and implies large welfare losses among the first quartile when it is hit by an adverse shock
Asset accumulation
Latin America and the Caribbean
Caruso (2015)
In utero and young children are the most vulnerable to natural disasters and suffer the most long-lasting negative effects, including less human capital accumulation, worse health, lower income, and fewer assets when they are adults; furthermore, the results provide evidence of the intergenerational transmission of the impacts of the shocks, indicating that children born to mothers who were exposed to natural disasters achieve less education and run a greater risk of performing child labor
Colombia
Duque (2016)Children exposed to violence in utero and in childhood experience significant developmental declines in health and cognitive abilities
Santos (2016a)In utero shocks on birthweight translate into lower weight and height among children in middle-income families
Santos (2016b)After commodity price increases: unskilled employment expands; more children begin working and fewer children attend school; educational attainment shrinks; public-school enrollment declines
SHA
KIN
G U
P EC
ON
OM
IC P
ROG
RESS
Agg
rega
te S
hock
s in
Lat
in A
mer
ica
and
the
Cari
bbea
n
92
Human welfare impacts of aggregate shocks
Outcomes Country Sources Main findings
Asset utilization and returns
El SalvadorAcosta et al. (2016)
Exposure to war at birth reduces the likelihood of employment, the skill level of available jobs, and long-term human capital accumulation
Colombia Acevedo (2016)
After exposure to shocks: intrahousehold reallocation of resources; conditional on generating positive income, more teenagers take up work (rather than adults) in the presence of extreme weather than in its absence
PeruMalásquez (2016)
Exposure to violence during the first year of life reduces average height; the body mass index rises if exposure occurs during the first year of life; educational attainment declines if the exposure to violence occurs at 3–5 years of age; exposure to violence in utero reduces the probability of completing primary or secondary education; the effect of exposure to violence on labor market outcomes is negligible
The effectiveness of ex ante risk management
Outcomes Country Sources Main findings
Income diversification, occupational choice, and migration
Central America and the Caribbean
Báez et al. (2017)
Extreme events, such as hurricanes, affecting infrastructure tend to discourage people from migration; longer-term events, such as drought, affect agricultural labor markets, have an insignificant or a positive impact on migration decisions in Central America
Peru Toro (2016)Disasters do not incentivize people living to emigrate across provinces
Social protection and labor market programs
ColombiaDuque and Rosales (2016)
Although conditional cash transfers have an overall positive impact on children's educational outcomes, they do not have a differential effect on children exposed to early-life shocks; however, the overall effect of such transfers is large enough to mitigate the negative impact of weather shocks
JamaicaBronfman and Reyes (2016)
Labor market programs attenuate the effects of employment shocks; while a simulated employment shock would increase the prevalence of vulnerability to 31 percent in urban areas and 47 percent in rural areas in the absence of Program of Advancement through Health and Education cash transfers, the impact of the employment shock on vulnerability is reduced to 20 percent and 15 percent, respectively, with the transfers; this impact is stronger (ex ante) among the vulnerable than among the poor