WPS5269
Policy Research Working Paper 5269
Enterprise Recovery Following Natural Disasters
Suresh de Mel
David McKenzie
Christopher Woodruff
The World Bank
Development Research Group
Finance and Private Sector Development Team
April 2010
Policy Research Working Paper 5269
Abstract
Using data from surveys of enterprises in Sri Lanka after uncorrelated with reported losses of business assets.
the December 2004 tsunami, the authors undertake the Business recovery is found to be slower than commonly
first microeconomic study of the recovery of the private assumed, with disaster-affected enterprises lagging behind
firms in a developing country following a major natural unaffected comparable firms more than three years
disaster. Disaster recovery in low-income countries is after the disaster. Using data from random cash grants
characterized by the prevalence of relief aid rather than provided by the project, the paper shows that direct aid is
of insurance payments; the data show this distinction more important in the recovery of enterprises operating
has important consequences. The data indicate that aid in the retail sector than for those operating in the
provided directly to households correlates reasonably manufacturing and service sectors.
well with reported losses of household assets, but is
This paper--a product of the Finance and Private Sector Development Team, Development Research Group--is part of a
larger effort in the group to study microenterprise dynamics. Policy Research Working Papers are also posted on the Web
at http://econ.worldbank.org. The author may be contacted at dmckenzie@worldbank.org.
The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development
issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the
names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those
of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and
its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.
Produced by the Research Support Team
Enterprise Recovery following Natural Disasters
Suresh de Mel, David McKenzie and Christopher Woodruff*
JEL Classification Codes: O12, Q54, C93.
Keywords: Microeconomic Impact of Aid, Enterprise Recovery Following Natural
Disasters.
*
University of Peradeniya, World Bank, and University of Warwick, respectively. The authors thank
Susantha Kumara and Jayantha Wickramasiri and Michael Callen for outstanding research assistance and
Kathleen Beegle, Saroj Jha and Apurva Sanghi for comments. AC Nielsen Lanka administered the surveys
on which the data are based. Financial support from NSF grant # SES-0523167, and the World Bank
GFDRR and Norway Governance Trust Fund is gratefully acknowledged.
"What seems more typical are the comments appearing six months to a year after a
disaster, expressing surprise at the speed with which the community has recovered, and
the prosperity that now reigns."
(Dacy and Kunreuther, 1969)
A series of catastrophic events in recent years has drawn increased attention of
both the public and researchers to the plight of those impacted by natural disasters. The
number of natural disasters reported by the press and included in the most comprehensive
disaster database is increasing. The Intergovernmental Panel on Climate Change (2007)
believes that the frequency of natural disasters is "likely" to increase as a consequence of
global warming. Some argue that the current increase results from more complete
reporting of disasters, and others say that the future trend driven by climate change is
uncertain. But there is little disagreement that the impact of natural disasters is felt most
severely by households already living on the margins in low-income countries.1 The
death toll from disasters, for example, is typically much higher in low-income countries
(Kahn 2005).
There is a large literature on how households in developing countries cope with
and respond to disasters and aggregate shocks. (See Skoufias (2003) for an overview.)
The poor suffer disproportionately because missing credit and formal insurance markets
limit their ability to smooth aggregate shocks. The informal risk-coping strategies that are
used to smooth idiosyncratic shocks break down when all members of a risk-sharing
group are affected (Morduch, 1999). As a result, transient shocks can have permanent
effects, either through a lowered ability of households to provide nutrition or schooling
for their children (Maccini and Yang, 2009; Ferreira and Schady, 2009), or through the
inability to repurchase productive assets, such as livestock, which are sold to smooth
consumption (Carter et al, 2007).
1
The trend of an increasing number of disasters shown in the EM-DAT data is likely due in part or in
whole to a more complete reporting of disasters. (See, for example, the discussion in Str�mberg 2007.)
2
These same market failures that can limit the ability of households to recover
quickly from disasters are also likely to inhibit the recovery of microenterprises and small
businesses. Yet the existing literature has not looked at the process of enterprise recovery
in developing countries, and even in developed countries there are only a handful of
qualitative case studies. Until recent disasters such as Hurricane Katrina and the Asian
tsunami, the conventional wisdom, reflected in the quote from Dacy and Kunreuther's
book, is that the economy recovers surprising quickly.2 Based in qualitative interviews
with small business owners shortly after Hurricane Katrina, Runyan (2006) concludes
that insurance is key: firms with insurance quickly replace destroyed assets, but that those
without insurance could not, often because business records lost in the flooding and
destruction were required to access federal aid. Large government and non-profit aid
flows are also common in developing countries after natural disasters. The literature has
examined the determinants of how international financial flows such as aid respond to
disasters (e.g. Eisensee and Str�mberg, 2007; Yang, 2009), but there is little about how
and whether this aid reaches enterprise owners.
This paper provides the first microeconomic study of the recovery of the private
sector in a developing country following a major natural disaster. We use firm-level panel
data gathered from micro enterprises in southern Sri Lanka following the December 2004
Asian tsunami. In addition to surveys, the post-tsunami project involved a field
experiment providing grants to randomly selected enterprises. The grants allow us to
assess the importance of capital in the recovery process and measure the return to capital
immediately following a disaster.
2
Most of the research by economists examining the aftermath of disasters focuses on short-term recovery
process, rather than the longer-term rebuilding process (Okuyama 2003). Analysis of longer term recovery
generally uses aggregate data such as building permits (Dacy and Kunreuther 1969). Two exceptions using
micro-level data are Smith and McCarty (1996), who analyze the demographic changes in southern Dade
Country, Florida following Hurricane Andrew in 1992, and Dolfman et al (2007) who estimate losses in
employment and wages in New Orleans following Katrina.
3
Recovery in low-income countries differs from recovery in high-income countries
most notably in the flow of cash to households and enterprises. A large share of
households and businesses in high-income countries are covered by insurance against
disasters. While about 50 percent of losses resulting from Hurricane Andrew in Florida
and the Northridge earthquake in California were covered by insurance, for example, less
than 15 percent of the losses resulting from the tsunami were covered (Ferguson 2006).
Moreover, insurance coverage in low-income countries is typically limited to the largest
enterprises. Few households or small businesses have insurance to cover losses. Relief
aid flows serve as a substitute for insurance flows, both in paying for the recovery and in
stimulating economic activity. But as our data show, there are important differences
between insurance and aid flows. Only a small part of disaster relief aid flows as cash
directly to households and small businesses.3 The larger share of aid comes in kind, with
the majority channeled into infrastructure projects. Moreover, aid agencies are seldom in
a position to verify actual losses of households or small-scale enterprises. So while
insurance payments are closely related to the insured entity's losses, aid payments to
individuals, households, or firms may not reflect the actual losses suffered. Where do
households and enterprises obtain funds to rebuild? How long does the process of
rebuilding take? Our data are uniquely suited to shed light on these questions.
The December 2004 Indian Ocean tsunami produced catastrophic damage along
Sri Lanka's eastern and southern coastlines. Official estimates put total deaths in Sri
Lanka at more than 35,000. More than half a million people were displaced when their
houses were damaged or destroyed. Estimated damage to infrastructure and other assets
3
See Oxfam (2005) for an argument that cash aid flowing directly to households should be more common
following disasters. Harvey (2005) reviews the state of knowledge on cash aid following disasters. These
discussions focus largely on household recovery rather than microenterprise recovery. There is also a
debate among microfinance practitioners about the role of loan forgiveness in disaster recovery. See
Mathison (2003) for a discussion of these issues.
4
exceeded $1.3 billion, around 7 percent of the country's GDP.4 The tsunami's impact was
concentrated in a narrow strip along the coast, and in the fishing and tourism sectors.
Aggregate Sri Lankan GDP fell only by between 0.5 and 1.0 percent (Jayasuriya et al,
2005). But two-thirds of the island's fishing fleet was destroyed (Asian Development
Bank et al, 2005), and--in spite of the fact that the arrival of aid workers dampened the
blow--hotel bookings fell by 60 percent on Sri Lanka's southern coast in 2005.
The international response to the disaster was rapid and strong. Governments,
international NGOs and the international financial institutions committed over $2 billion
in relief and recovery funds, of which $1.1 billion had been dispersed and $0.6 billion
expended 18 months after the disaster (Government of Sri Lanka, 2006). Of the total aid
pledged, $184 million was categorized as "relief" aid. The remainder was for recovery of
the housing ($370 million), transportation ($245 million committed), and water ($190
million) infrastructure and for livelihood restoration ($219 million).
Though the prevalence of insurance and aid differs in high- and low-income
countries, businesses face a similar set of shocks following a natural disaster regardless of
their location. We discuss three factors that typically affect the recovery process and give
us reason to believe the findings here would apply more broadly to recovery from other
large disasters. First, labor and capital are destroyed. In Sri Lanka, it appears the impact
on the capital stock was proportionately larger than the impact on the labor force, almost
certainly the usual pattern. Second, demand shifts. In aggregate, the shift will generally
be inward. But for some sectors (e.g., construction materials), demand may shift outward.
Third, many trading relationships are destroyed, either temporarily or permanently. The
time required to re-form relationships depends on the nature of the relationships in a
given sector. This effect is analogous to the "disorganization" effects discussed in the
4
This is the figure reported in the EM-DAT database. An assessment by the Asian Development Bank,
World Bank and Japanese Bank for International Cooperation estimated losses of roughly $1.0 billion and
replacement costs of $1.5 billion. The Sri Lankan government estimated recovery costs to be $1.8 billion.
5
literature on transition economies (Blanchard and Kremer 1997; Roland and Verdier,
1999). We use our panel data to examine enterprise recovery from each of these shocks.
We begin in Section 2 by discussing the data and examining the sources of funds
available to enterprises and households to pay for repair or replacement of lost assets. In
Section 3, we use random capital shocks we generated as a part of the project to assess
the value of access to liquid capital in the recovery process. We study the timeline of
recovery in Section 4, and conclude in Section 5 section with a discussion of policies that
might quicken the recovery of microenterprises following a major disaster.
Section 2: Firm Losses and Sources of Recovery Funds
Even in normal circumstances, credit and insurance market failures loom large for
business owners in developing countries (e.g. Banerjee and Duflo, 2005). A variety of
informal mechanisms have evolved to overcome these market failures, but aggregate
shocks such as economic crises and natural disasters can limit the effectiveness of
informal financing and risk-coping mechanisms. In particular, group-based informal
insurance arrangements are ineffective, as the incomes of a household's risk-pooling
partners also fall (Lustig, 2000). Already limited access to formal credit may become
even more limited. As a result, it is of interest to ask: What resources do owners of
microenterprises and small firms use to replace buildings and equipment that are
destroyed or damaged in major disasters? And, how quickly are the enterprise owners
able to replace assets lost in the disaster?
The survey data
We provide some answers to these questions using three surveys we conducted
with enterprise owners and wage workers along the southern coast of Sri Lanka after the
tsunami. Each of the three samples contains information on 400 to 600 individuals: 130
to 200 individuals suffering asset losses from the tsunami, an equal number of individuals
6
living or working in the nearby neighborhoods but not suffering any asset damage, and an
third set of individuals living and working outside the tsunami-affected area. We refer to
the first group of individuals as "directly affected," the second as "indirectly affected,"
and the third as "unaffected." We believe these data are unique in allowing us to examine
the recovery process of small scale enterprise in a low-income country.
The first survey is a panel of 618 household enterprises, interviewed quarterly
between April 2005 and April 2007, and again in October 2007 and April 2008. The
April 2005 baseline survey asked owners which assets were damaged or destroyed by the
tsunami, and all waves of the survey ask about the repair or replacement of those assets.
The July 2005 survey has questions on grants and loans obtained to replace assets.
Enterprises in the panel were also subject to random capital injections in May and
November 2005, as described in more detail in Section 3 below. Note that all of the data
on asset damage and aid received are self reported. The small scale enterprises and
households which comprise the majority of the samples seldom keep written records of
assets or business transactions. Thus, while the surveys provide very detailed micro-level
data, we have no way to verify any of the responses. We discuss this issue in more detail
later in the paper.
In July 2007 we conducted two additional surveys--one with a sample of 456
wage workers and the other with a sample of 424 enterprises with between five and 50
employees. The wage workers all live in the same neighborhoods as the microenterprise
owners.5 Because we did not find enough larger enterprises in these neighborhoods, the
enterprise sample was drawn from surrounding areas as well. These surveys asked
households and enterprise owners about damage suffered from the tsunami, the extent to
which the damaged assets had been replaced or repaired, and the sources of funds used to
5
Sri Lanka is divided into 25 districts, which are divided further into 324 Divisions. These are further
divided into just over 14,000 Grama Nalidaris, or GNs, which are the smallest administrative units. There
are about 400 households in a typical GN. The microenterprise sample was drawn from 25 GNs in the
districts of Kalutara, Galle and Matara.
7
pay for those repairs. We asked households and business owners if they received
payments from insurance claims or aid in the form of grants, or loans.
Asset recovery
Summary data on the extent of damage, insurance coverage, and aid are shown on
Table 1. The reported damages reflect differences in wealth and income among the three
samples. The larger firm (SME) owners report losing an average of just over $40,000 in
business assets and $6,000 in household assets. Wage workers report household losses
averaging $5,000, while the microenterprise owners report losses of $897 in business
assets and just over $2,400 in household assets. The table demonstrates the lack of
insurance coverage in all three samples. Even among the larger enterprise owners, only
13% of those suffering losses reported having any insurance coverage for business assets,
and their policies generally did not cover damage from tsunamis. Those with insurance
report that only 7.5 percent of business losses were covered. Similar patterns hold for
household assets. Among the larger enterprise owners, only 4.1 percent said they had any
insurance on household assets, and only 5.1 percent of household losses were covered.
Among wage workers, only one of 153 directly affected by the tsunami report having
insurance to cover losses.6
Owners and wage workers report receiving more funds from grants than from
insurance; more than three-quarters of each group report receiving one or more grants.
Loans are less common, especially among the microenterprise owners. Only in the case
of wage workers, however, does the average respondent report that grants and loans
combined covered as much as half of their losses. For SME owners, where the survey
asked how the grants and loans were used, grants were more likely to cover housing
6
In contrast, 78% of households suffering damage from hurricane Andrew in Florida had insurance, and
the insured received average payouts of $32,000 (Smith and McCarty, 1996). The percentage of insured
was likely higher among businesses. Besides providing funds for the recovery of individuals' assets, the
$14 billion dollars of insurance aid following Hurricane Andrew provided a significant boost to the local
economy that is lacking in the Sri Lankan case.
8
losses and loans more likely to cover business losses. For the typical SME owner, grants
covered almost one-quarter of housing losses, but less than 2 percent of business losses.
Loans covered an additional 13.7 percent of business losses for the SME owners.
In Sri Lanka, cash aid came primarily through four programs. First, the Sri
Lankan government paid surviving family members 15,000 Sri Lankan Rupees (LKR,
about $150) for each person killed by the tsunami, to offset funeral expenses. Second, the
government and the World Bank provided four grants of 5,000 LKR (about US $50) to
220,000 households suffering direct damage from the tsunami.7 Third, the government
provided aid to rebuild houses destroyed by the tsunami. Finally, the government and
numerous NGOs sponsored cash-for-work programs, which typically paid workers
around 300-350 LKR ($3.00-$3.50) per day to participate in cleanup and rebuilding
activities. So while much of the aid came in-kind, there was a significant amount of cash
aid provided.
Our April 2005 microenterprise survey indicates that the households of owners
received an average of $115 in aid during the first three months of the tsunami, of which
$101 came from government programs, $8 from NGOs and $6 from other sources. Table
1 shows that by July 2005, 94 percent of those reporting damage had receive some form
of aid. However, the mean aid received ($332 among those receiving something) was less
than 10 percent of the reported damage.8 The grants reported by microenterprise owners
cover only a small portion of the losses they incurred. Loans were much less common.
Less then 4 percent of microenterprise owners reported receiving loans, and these
covered less than 1 percent of losses in aggregate. The data on Table 1 are self reported,
and we should be concerned that the reported losses are exaggerated and the reported aid
7
According to Jayasuriya et al (2005), the number of beneficiaries was reduced from over 800,000 to
around 200,000 after the first two payments. The smaller number represents only around 50,000
households.
8
The enterprises appear to have reported grants provided by our project itself, as described below, so these
amounts likely overstate the aid received by typical microenterprises. Those receiving our grants reported
cash aid averaging just over $100 more than those not receiving our grants.
9
understated. However, one measure suggests that the data are not far from reality. The
program to replace or repair housing made payments totaling $2,500 per household,
almost exactly the same average amount as the grants reported by wage workers. SME
owners reported grants averaging around $1,600.
Perhaps the most interesting aspect of the reported aid is the lack of correlation
between reported losses and reported grants and loans. For SME owners, where we are
able to separate business from household aid, the correlation between the reported loss of
household assets and the grants (loans) received is 0.70 (0.27). For business assets, on the
other hand, there is essentially no correlation between the reported losses and the (very
small level of) grants received; the correlation between losses of business assets and
loans to replace those assets is 0.16. For wage workers, the correlation between grants
and household losses is 0.31. Finally, for microenterprise owners, there is no correlation
between the level of losses on the one hand and grants or loans on the other. So while
overall aid flows into Sri Lanka following the tsunami were large, and while at least some
aid there flowed directly to households suffering losses, the data suggest that targeting is
particularly poor with regard to damage suffered by business owners.
Enterprise recovery
Given the lack of insurance and low aid flows, we find it surprising that owners
report the majority of damaged assets had been replaced or repaired by the summer of
2007. Households report having replaced 60% of lost assets, while SME and
microenterprise owners report having replaced more than two-thirds of their assets. By
April 2008, 75 percent of microenterprise owners reported having replaced all of the
housing assets damaged by the tsunami. Where did they obtain funds to repair or replace
damaged assets? The survey data allow us to provide some insight on this question, and
to say something about the speed of recovery, with regard to the microenterprises.
10
Among the 204 firms suffering tsunami damage, 168 (82 percent) repaired or
replaced some damaged enterprise assets within three months of the tsunami. On average,
owners suffering damage reported spending US$111 on enterprise recovery in the three
months following the tsunami. Excluding those spending nothing in this period, the
average expenditure was $135. While a substantial sum relative to pre-tsunami income
levels, the amount represents less than 15 percent of the assets lost or damaged by the
tsunami. The largest initial effort went to equipment and working capital: an average of
US$39 was spent on equipment, representing 25 percent of the losses in this category.
Relative to the losses suffered, the smallest investment was in land and buildings, where
owners reported spending US$29 compared with losses of US$362.9
What were the sources of funds which enterprises used to replace lost or damaged
equipment? On average, just over half of the funds spent in the first three months came
from own savings (51 percent). An additional 15 percent were obtained through loans
from family members (9 percent) or friends (6 percent). One-fifth of the resources (20
percent) came from grants or loans from tsunami relief agencies. The remaining 14
percent was spread among credit from suppliers (6 percent), loans from microfinance
organizations (2 percent), moneylenders (2 percent), banks (less than 1 percent),
remittances from relatives abroad (1 percent) and other sources. The largest spenders
relied more heavily on loans from family members and credit from suppliers, while those
spending the least relied more on own savings. As a percentage of the total amount spent
by all 176 entrepreneurs with complete data, own savings represents only 36 percent of
expenditures. Loans from family members (13 percent) and friends (20 percent) are
together almost as important, and credit from suppliers, at 11 percent of the total funds
spent, is also more important than the unweighted averages suggest.
9
The data on assets lost in this paragraph differ slightly from those reported for microenterprises in Table
1, because the data here come from the baseline survey while those in Table one come from the October
2007 retrospective survey. The two means are actually quite close, however, with owners reporting
business losses of $814 in the April 2005 survey and $897 in the October 2007 survey.
11
Given the breadth of damage in local areas, we might expect those with family
and social networks extending outside the direct impact zone to have recovered more
quickly, since their networks extend to unaffected areas. The evidence available to us on
this, while mixed, is not particularly strong. In the April 2008 survey, we asked
entrepreneurs whether their relatives were affected by the tsunami. Questions covered
parents, siblings, parents- and siblings-in-law, and adult children who were living at the
time of the tsunami. We do not find that those with parents or children who were
unaffected by the tsunami spent a larger amount on recovering assets in the months
following the tsunami, or that they obtained a larger share of the funds they did spend
from family members. We do find that entrepreneurs who lived in a different
administrative district (DS Division) at age 12 report using family members as a source
for a slightly higher portion of their recovery funds (14 percent vs. 9 percent), but the
difference is not significant. Those who lived outside the same DS Division at age 12 also
report having spent more to replace and repair assets during the first three months, both at
the mean ($185 vs. $102, p=0.12) and the median ($40 vs. $33). These data suggest a
surprisingly large contribution to the recovery comes from the owners' savings and loans
from family and friends.
Figures 1 and 2 show that enterprises suffering damage had recovered to their pre-
tsunami size within about 15 months of the tsunami. The figures show the value of
equipment, tools and vehicles used in the enterprise (Figure 1) and profits (Figure 2)
across time, as a percentage of the reported values in the month before the tsunami
(November 2004). Mean profits of directly affected firms fell to 65 percent of the pre-
tsunami level in March 2005 before recovering to 107 percent of the pre-tsunami level by
March 2006. Capital stock shows a similar pattern. Both figures compare the directly
affected enterprises with the unaffected enterprises.10 Note that while the affected
10
We use the set of firms not receiving one of the random grants provided by the project (as discussed
below) and reporting profits in all 11 rounds of our panel survey. We trim firms with profits in the top and
12
enterprises recovered to pre-tsunami levels, the profits and capital stocks of the
unaffected firms grew by 30 percent to 50 percent over this same period. Indeed, the
affected firms remain smaller in both profits and investment more than three years after
the tsunami.
In sum, the data paint a fairly consistent picture of the process of recovery of
business assets. The magnitude of the total aid flow to the affected area was roughly
comparable to what might be expected following a disaster in the United States. But the
flow was of a very different character, with less insurance and more aid. The portion of
the aid and insurance flowing directly to households and small scale enterprises appears
to be modest compared to the size of the losses incurred. Moreover, while we find some
correlation between the magnitude of the aid flows and the magnitude of the losses with
respect to losses of household assets, we find no correlation between losses of business
assets and aid flows. Consistent with this, the recovery of household assets is complete
within three years for a majority of households, while the affected enterprises lag behind
a comparable group of unaffected enterprises more than three years after the event.
Section 3: Profitability and the Incentive to Recover Assets
Given the low levels of official aid firms report receiving, we find even the partial
recovery of assets surprising. The households of microenterprise owners have low levels
of wealth, and few liquid assets. The recovery of the majority of their assets at a time
when competing demands for available funds must have been great suggests the
microenterprise owners had strong incentives to replace their assets. Did extraordinary
opportunities for profit drive enterprise recovery in the tsunami impact zone? We provide
bottom 1 percent, and then plot an index of mean real profits over the period November 2004 to September
2007. Nominal profits were converted to real profits using the monthly Sri Lanka Consumer's Price Index,
available at http://www.statistics.gov.lk/price/slcpi/slcpi_monthly.htm. Annual inflation was 4.0 percent
between March 2005 and March 2006, and 18.6 percent between March 2006 and March 2007. Inflation
was an additional 10.3 percent in the six months from March 2007 to September 2007.
13
evidence on this question by providing randomly allocated grants to a portion of
microenterprises in a panel survey, as described in more detail below.
A simple framework results in two predictions, which we test with the data.
Assume firms have Cobb-Douglas production functions and operate in perfectly
(1 )
competitive output markets. The output of firm i is Yi AK i Li , with K, and L
representing capital and labor, and Y representing the output of goods with unit price.
Prior to the shock,
(1 )
L
A i
K ri
i
( )
L
(1 ) A i
K wi ( w)
i
We allow the opportunity cost of both capital and labor to be firm specific, reflecting less
than perfect input markets. Since the enterprises in our sample employ few workers
outside the family, wi (w) reflects the individual's opportunity cost of time, which
depends on the market wage rate, w .
The tsunami appears to have destroyed a larger share of the capital stock than the
labor force. The estimated value of the lost assets was US$1 billion, just over 1 percent of
the country's pre-tsunami capital stock. About 35,000 people were killed, less than 0.2
percent of the population.11 A decrease in capital relative to labor leads to an increase in
the returns to capital at the margin, and a decrease in the return to labor at the margin.
But at least three factors might offset this. First, the psychological trauma
following the devastation may have increased at least temporarily the opportunity cost of
the owner's time. Owners reported (to us and to others) that they did not feel like
working in the weeks and months following the event. Second, the tsunami caused shifts
in the demand for output produced by the firms. Some evidence suggests that the
11
Because of the timing of the event and other factors, women were disproportionately likely to be killed.
Since women have lower labor force participation rates, the effect on the labor force was likely less than
0.2 percent.
14
immediate shift was inward. Fishing and tourism are the main industries along the
southern Sri Lankan coast. Tourism fell precipitously, replaced partially a few weeks
later by the inflow of relief workers. The demand for fish also fell. Much of the initial
relief funding came in the form of in-kind aid. After the first few weeks, more of the aid
appears to have come in cash or through the purchase and donation of goods produced
locally such as fishing boats. Third, the tsunami caused other disruptions in production.
Trading relationships were severely disrupted. The businesses of customers or suppliers
of firms, particularly those located near the coast, were destroyed and in some cases, their
owners were killed or severely injured. In some cases, the entire supply chain was
disrupted. This is the case in the coir industry, for example, where the pits used to soak
the coconut husks were filled with debris that took many months to clear. Even where
alternative suppliers for an input did exist, market frictions might have led to production
difficulties in some sectors. The simultaneous severing of many relationships is not
unlike the "disorganization" following the breakup of the Soviet Union (see Blanchard
and Kremer 1997; Roland and Verdier 1999).
This discussion leads to two testable predictions. First, the disproportionate
destruction of capital relative to labor should result in very high returns to capital among
firms, at least where demand is restored by aid flows and relief workers. Second, the
profitability is likely to be lower for manufacturers than for retailers, because
manufacturers rely more on the availability of specific inputs, and generally have more
intensive relationships with a smaller number of customers. For the same reasons,
manufacturers are likely to recover more slowly.
Other predictions of heterogeneity in the speed of recovery come from the broader
development literature. We explore three: education, gender, and the extent to which
immediate family members living in other households also were affected by the tsunami.
Schultz (1975) argues that education and ability are particularly important in dealing with
changes in economic conditions and economic disequilibria. We test to see if education
15
and other ability measures are associated with more rapid recovery. With respect to
gender, the shock to the household may affect household bargaining and the allocation of
resources. The recovery of males and females may differ as a result. Finally, gifts or
loans from family members may be one source of capital for recovery. These are likely to
be most commonly received from parents or adult children. Of course, aid from family
members is less likely if family members were also affected by the tsunami.
Using Experimental Data to Identify Returns to Recovery Funds
The suddenness of disasters and logistical issues involved in coordinating the aid
response make collecting data for households or enterprises receiving assistance a very
difficult and rarely undertaken task. Even where such data are collected, identifying the
role of capital in the recovery process is complicated by the presence of a number of
unobserved factors which are likely to be correlated with both the speed of capital stock
replacement and future profitability. Firms anticipating faster profit recovery may be
more inclined to replace capital stock or find it easier to persuade family members to lend
them resources. More politically connected firms may be better able to access aid flows.
And the direction of causation may flow from profit recovery to capital stock, if firms
replace damaged capital by reinvesting profits.
To investigate whether high profits provided an incentive for replacing damaged
assets, we carried out an experiment in which firms were randomly given grants of cash
or in-kind grants to purchase material or equipment, selected by the owner for the
enterprise. We describe in detail the results of the experiment among undamaged firms in
de Mel, McKenzie and Woodruff (2008). The experiment gives us a clean measure of the
role of capital in the disaster recovery process.
Our baseline survey included 227 enterprises in directly affected areas along the
coast. The enterprises were selected with a screening survey administered door-to-door in
residential neighborhoods in the districts of Kalutara, Galle and Matara. Our intention
16
was to draw a sample of enterprises with less than 100,000 LKR (US$1000) in capital
stock, excluding land and buildings. A screening survey eliminated enterprises hiring
paid employees, those owning a motorized vehicle, and those engaged in professional
services, fishing, and agriculture. After reviewing the baseline data, we eliminated 18 of
the 227 enterprises either because they exceeded the 100,000 LKR maximum size ceiling
we had set, or because a follow-up visit could not verify the existence of the enterprise.
The remaining 209 firms constituted the baseline sample. These firms almost evenly split
across two broad industry categories, with 107 in manufacturing or services and 102 in
retail, and by gender of the owner, with 107 firms owned by females, 98 by males, and 7
jointly owned.
After the baseline survey, we randomly selected some of the enterprises and gave
them either 10,000 LKR (about US$100) or 20,000 LKR (about US$200) either in cash
or in-kind grants. In the latter case, the items to be purchased were selected by the owner,
and purchased by research assistants working for the project. Cash treatments were given
without restrictions; recipients were told they could purchase anything they wanted with
the cash. The treatment was framed as compensation for participating in the panel survey,
and enterprise owners were told that they would be eligible to win the grant only once.
The aim of our experiment was to provide firms with an exogenous shock to
capital stock, and to measure the impact of this on business profits. Within the affected
zone 120 firms were assigned to treatment (57 percent), with 90 firms assigned to receive
treatment after the baseline survey in May 2005 and a further 30 firms assigned to receive
treatment after the third survey round in November 2005. This split frontloaded
treatments so that more of the randomly allocated aid could reach tsunami victims sooner.
The 120 treatments were made up of 77 of the 10,000 LKR treatments (39 cash, 38 in-
kind) and 43 of the 20,000 LKR treatments (21 cash, 22 in-kind).
Our initial plan was to survey firms for five quarterly waves only. Receipt of
further funding enabled us to continue the panel, with four additional quarterly waves
17
collected from July 2006 through April 2007, and a tenth and eleventh waves collected in
October 2007 and April 2008. In order to compensate firms for the additional burden of
staying in the study longer than we had anticipated, we gave 2,500 LKR (~$US25) in
cash to each of the remaining untreated firms after round five of the survey.
Attrition in the data is relatively low. Of the 209 baseline firms, 186 report profits
in round five and 173 in round 11 (83 percent of the initial sample). However, only 197
firms report profits in the baseline survey, and firms move in and out of the sample. One
hundred thirty-five firms report profits in all 11 rounds, and 182 report profits in eight
rounds or more. We restrict our analysis to the 200 firms reporting profits in three rounds
or more.12 Appendix 1 compares the characteristics of the treated and untreated firms
among these 200. The randomization was done by computer, so any differences can only
be due to chance or to the elimination of these nine firms that report less than three waves
of profits. The two groups appear to be balanced on the key observable characteristics,
but we will also include individual fixed effects to account for any baseline differences in
levels remaining.
The Impact of Grants on Profits
We begin by estimating the mean impact of the grants on real profits of tsunami-
affected firms, via the following fixed effects regression for firm i in period t:
10
PROFITS i ,t AMOUNTi ,t s s i i ,t (1)
s2
Where AMOUNTi,t is an indicator of the amount of treatment received by firm i at time t,
coded in terms of 100 LKR. Firms receiving 2,500 LKR after round 5 will thus have
AMOUNT of 25 in rounds six through 11 (and 0 before this). The s are wave dummies.
12
This eliminates six control firms and three firms assigned to receive the 10,000 LKR treatment.
18
Treatments are coded 0, 100, or 200 in the regressions so the coefficient shows
the increase in profits in rupees from a 100 LKR treatment. Thus, the coefficients can be
interpreted as the percentage return on the treatment. The interpretation of the
coefficients from the regressions merits several comments. First, for all regressions we
deflate profits by the all-island consumer price index. Second, since we use the amount of
the treatment as the independent variable, we are measuring the intention to treat. As not
all of the grants found their way into the enterprise, this may differ from the return to
incremental investments. The intention to treat seems the more policy-relevant effect in
the disaster recovery context. Third, we pool all waves of the survey, so the coefficient
measures an average treatment effect for the three years following the first set of
treatments. We find no significant time trend in the treatment effects, justifying the
pooling of the data.13
The first column of Table 2 shows the effect of the treatment on real profits. A
100 LKR grant increases average monthly profits by 9.90 LKR, representing a 9.9
percent real monthly return on the treatment. The treatment effect is significant at the 5
percent level. In column 2, we trim on extreme changes in profits. We eliminate
observations lying above the 99th percentile in either percentage or absolute changes in
profits across waves of the sample. We trim the top but not the bottom of the distribution
because rapid falls in profits are more likely to be due to owner illness, lack of demand,
or other negative shocks that we do not want to trim from the data.14
In column 2, we trim observations where profits increased by more than 800
percent or 15,000 LKR from one period to the next. Ideally this trimming should not
change the size of the estimated coefficient, but should increase its precision. In fact, we
13
Returns are slightly higher more than five quarters after treatment, but the difference is not significant at
the 0.10 level.
14
After receiving the data, we asked the survey firm to verify the records for all enterprises showing very
large changes from one wave to the next. Several data points were corrected for keypunch errors. The
survey firm also confirmed that several cases of large drops in profits were due to negative shocks suffered
by the enterprise. Their opinion was that large increases were more likely due to incorrect recording of data
in the field. On the basis of this exercise, we trim only the top tail of the data.
19
see a modest drop in the coefficient, to 8.9 percent, and a drop in the standard error. We
trim the remaining regressions reported in the paper in a similar fashion. The treatment
effect of 8.9 percent when we trim on changes compares to a treatment effect of 5.4
percent for the indirectly affected and unaffected firms (de Mel, McKenzie and
Woodruff, 2008). That is, grants have a far bigger impact on damaged firms, by a margin
of half again as much as the effect on undamaged firms.15
In column 3 of Table 2 we examine whether the cash and in-kind grants have
different effects. A 100 LKH cash grant increases profits by 5.3 LKR, and the in-kind
grant by 12.2 LKR. But given the standard errors, we cannot reject equality of the two
treatment effects (p=0.211). Therefore, evidence does not strongly support a preference
for aid in kind over cash grants in terms of their ability to raise firm profits.
Column 4 of Table 2 examines whether firm owners adjusted their labor hours in
response to the treatment. A priori, the direction of any effect is unclear � repaired capital
stock may allow owners to produce more, and hence increase complementary labor
inputs, or may enable owners to substitute capital for labor, leading them to work less.
The results in Table 2 show no strong effect in either direction. The point estimates
suggest that owners reduce labor hours by 0.3 hours per week as a result of the treatment,
and we cannot reject that the change in labor hours is zero. Thus, any effect on profits
would appear to be attributable to the injection of capital rather than any associated
changes in labor input.
15
Note, however, that the standard errors are large enough that we cannot reject that the treatment effect for
tsunami-affected enterprises is equal to that of unaffected enterprises, despite the large difference in point
estimates. Table III, column 6 of De Mel et al. 2008 is the only part of that paper which involves the
tsunami-affected firms: it reports a treatment effect of 9.08 percent for the tsunami-affected firms using
only nine waves instead of the 11 waves of data here. The paper then refers to the current paper for
analysis.
20
Demand, market friction, or production complementarity?
There are several dimensions along which we might expect to find heterogeneity
in the post-disaster returns to capital. The tsunami resulted in the closure of many
businesses on a temporary or (in the case of death) permanent basis. Where enterprises
purchase from or sell to only a few trading partners, replacing these relationships might
be expected to take time. Second, demand shocks may vary with the product of the
enterprise. McKenzie (2006) shows that one way credit-constrained consumers respond
to aggregate shocks is to cut back on their purchases of semi-durables to a much larger
extent than would be predicted just from the income effect. As a consequence, we would
expect demand to recover much more quickly in retail sales, and less quickly in
manufacturing, as consumers shift their expenditure patterns to protect food consumption.
Third, production assets may have stronger complementarity in manufacturing than in
retail. Owners may need to replace all of their capital stock before they can produce.
Retailers, on the other hand, may be able to sell goods even without assets such as display
cases, refrigerators, and so forth. Finally, supply chains may have been disrupted.
Manufacturers of products made from coconut husks (coir), for example, reported finding
supplies difficult, as the lagoons used to process the husks took many weeks or months to
clean. Without a supply of inputs, replacing machinery may be irrelevant.
We investigate how the impact of the grants varies across sectors, and by the
importance of individual customer and supplier relationship. We do this by estimating the
following fixed effects regression in which the treatment variable and wave dummies are
each interacted with sector or trading partner characteristics of the firm:
PROFITS i ,t AMOUNTi ,t AMOUNTi ,t X i
10 10 (2)
s s s s X i i i ,t
s 2 s 2
21
where Xi indicated the sector, the presence of a trading partner buying more than 25
percent of output or supplying more than 25 percent of inputs, or some other
characteristic of the enterprise or owner.
The results by broad sector and trading partners are shown on Table 3. We show
results with the sample trimmed at the 99th percentile on percentage and absolute changes
in profits, but the results are very similar without trimming. We find very significant
differences in the impact of the grant on manufacturing firms compared to retail firms.
Column 1 shows that the mean effect for retail firms is to increase real monthly profits by
19.6 LKR for every 100 LKR received, equivalent to a 19.6 percent monthly return on
the grant. In contrast, the interaction of amount with the manufacturing dummy is
significant, large, and negative. Adding the interaction effect to the amount coefficient
results in mean treatment effect which is slightly negative, and not significantly different
from zero for manufacturing. The grant therefore had a large effect on retail, and no
average effect on manufacturing. This is strikingly different from the results of applying
the treatment to indirectly affected and unaffected firms, where there is no significant
interaction with manufacturing.16 It therefore appears that after the tsunami a lack of
capital was not the main barrier to recovery of manufacturing, but that capital did
significantly impact on the recovery of retail.
Given the sample size, we are unable to say why capital is more important for
retailers than for manufacturers. We suspect the explanation varies with the nature of the
product. Both enterprise owners and NGOs told us, for example, that lack of inputs was
the main constraint in the coir industry. The primary customers for producers of lace are
tourists, so in that sector, a lack of demand may haven been the critical factor. Given our
sample of about 100 manufacturers in the directly affected area, we are unable to
16
In particular, when we trim on changes in profits, the coefficient on retail*amount is 6.43 (s.e. 3.08) and
the coefficient on manufacturing*amount is 4.08 (s.e. 2.80) when we estimate equation (4) for firms not
damaged by the tsunami. The difference is insignificant (p=0.57).
22
differentiate between these explanations. We can say that we find no evidence that
returns are lower among firms with suppliers or customers who account for a large share
of trade. In Columns 2 and 3 of Table 3, we interact the treatment amount with a variable
indicating the enterprise has a single customer accounting for at least 25 percent of sales
(Column 2) or one supplier accounting for at least 25 percent of input purchases (Column
3). Neither interaction is significant, suggesting that friction in trading relationships is not
a cause of slow recovery. However, we take even this limited evidence with a grain of
salt. The questions on which these variables are based were asked at the time of the
baseline survey. As such, they reflect the situation after the tsunami rather than before.
Other Dimensions of Heterogeneity
Appendix 1 compares the characteristics of manufacturing and retail firms in our
sample. The manufacturing firms are more likely to be run by females, have lower profits
before the tsunami, and are more reliant on a single customer and a single supplier of
inputs. They sell to far fewer customers per day, on average. The three largest sub-
industries within our manufacturing sample among directly affected firms are sewing
clothes, spinning lace, and making food such as string hoppers (Sri Lankan rice noodles
which are a dietary staple). These are all industries dominated by female owners. With
the exception of food preparation, all other products made by the manufacturing firms are
semi-durables, for which demand is likely to recover more slowly. Furthermore, a greater
reliance on a major customer and/or a major supplier increases the likelihood that a
disruption in this relationship as a result of the tsunami will have a large effect on the
firm.
In de Mel, McKenzie, and Woodruff (2008, 2009) we find that returns are
significantly lower in enterprises owned by females than in enterprises owned by males.
Indeed, we cannot rule out the possibility that the overall mean effect of the treatment is
zero for female owners. In the tsunami-affected zone, however, we find no significant
23
difference between returns in male- and female-owned enterprises. Column 4 Table 3
suggests the mean female effect is about 70 percent of the male effect. We posit that the
difference reflects the value of capital in recovery compared with the value of capital in
expansion. In the directly affected area, the grants helped owners return their business to
their pre-tsunami size. In the unaffected areas, the grants allowed owners to expand their
businesses. The higher returns for females in the directly affected area suggests that
recovering the enterprise is as valuable for females for males, while the low returns for
females in unaffected areas suggests that expanding the businesses from the steady state
size may not be profitable.
An additional hypothesis is that recovery may be faster for more educated, able
business owners. Schultz (1975) has argued that an important role of education is
providing the ability to deal with changes in economic conditions and economic
disequilibria. Column 5 of Table 3 shows that the treatment has a positive, but
insignificant interaction with the years of education of the firm owner. The point estimate
is positive and similar in size to the significant coefficient found among indirectly
affected and unaffected firms (de Mel, McKenzie and Woodruff, 2008). The results are
therefore consistent with the view that among small firms, more able firm owners are
further away from their optimal capital stock level (even prior to the disaster), and hence
have higher returns to capital. They do not suggest that human capital can serve as a
substitute for physical capital in the recovery process � which would require a negative
interaction between the grant and human capital.
Finally, we asked owners if they had parents or adult children who were not
affected by the tsunami. Unaffected family members might serve as an alternative source
of aid in the absence of insurance or grants tied directly to losses. Though we did not find
that those with unaffected family members received larger loans from family members,
we did find that those with parents or adult children who were unaffected by the tsunami
had lower returns from the treatments. Indeed, the results suggest that those with both
24
adult children and parents who were untouched by the tsunami had zero returns to the
treatment. Though we did not stratify the treatment on this variable, the results are
consistent with social networks being an important source of recovery funds. Those
without family networks in the position to provide help after the disaster have particularly
high-valued uses for recovery funds.
Profitability and investment
If profitability is higher in retail, we should expect to find that retailers invest a
larger share of the grants than manufactures. Table 4 presents evidence weakly consistent
with this expectation. We regress capital stock reported in each wave of the survey
against the treatment amount and the treatment amount interacted with a variable
indication the firm is a manufacturer. The regressions also include firm and wave fixed
effects, and wave / manufacturing interaction effects. The regression is trimmed on the
absolute and percentage change in reported profits. Retailers appear to invest all of the
grant in the enterprise. A 10,000 LKR grant is associated with an increase in the average
retailer's capital stock of 10,600 LKR. Manufacturers appear to invest a lower portion of
the grant in their business, about 6,400 LKR of a 10,000 LKR grant. Thus, the
investments behavior of manufacturers appears consistent with lower returns in the
manufacturing sector.
Section 4: The Speed of Recovery
The high returns to capital in the recovery zone provide a strong incentive for
reinvesting capital. If investment is in fact responding to that profit incentive, then we
should find that retailers recover their capital stock, sales and profits more rapidly than
manufacturers, even absent grants provided by the experiment. In this section, we
examine this correspondence. In particular, we are interested in the question: Do
25
untreated manufacturers recover more slowly than untreated retailers? That is, is the
behavior of the treated firms consistent with the data from the untreated sample?
Before looking at the microenterprise panel data, we note that the SME sample
includes both retailers and manufacturers. There are relatively few retailers among the
larger enterprises. The sample includes only 28 retailers (including hotels and restaurants)
that were damaged by the tsunami. But these retailers had replaced 78 percent of their
assets on average (85 percent at the median), compared with 64 percent (60 percent at the
median) among the manufacturers. The difference in means is significant at the .05 level,
in spite of the small sample size. On the other hand, we find no significant difference
between retailers and manufacturers with respect to replacing household assets. By the
summer of 2007, larger retailers had replaced 71 percent of the household assets damaged
in the tsunami, compared with 67 percent for manufacturers. These data suggest that
manufacturers had as many resources available to them as retailers for use in replacing
household assets. These data are thus consistent with retailers having stronger incentives
to replace or repair business assets.
We do not have good measures of inventories prior to the tsunami. This is a
concern, because a larger share of retailers' investment is made in inventories.
Admittedly, even our measures of profits and sales are retrospective. However, leaving
aside concerns about deliberate misreporting, we believe these are likely to more
accurately reflect the pre-tsunami situation. Using the untreated part of the
microenterprise sample, we now examine the characteristics of firms that recovered more
quickly. We estimate the following random effects model for log real profits for untreated
firm i in time period t=2,...,11:
ln PROFITS i ,t ' X i 1 ln PROFITS i , March 2005 2 ln PROFITS i , November 2004
11 (3)
t t i i ,t
t 3
26
where Xi is a vector of baseline characteristics of the owner and the firm, PROFITSi,March
2005 and PROFITSi,November 2004 are the firm's profits in March 2005 and November 2004
respectively, the t are wave effects, and i is a firm random effect. We estimate this for
the unbalanced sample of untreated firms experiencing business damage. This gives 702
observations on 80 firms. Since real profits are expressed in logs, the vector of
coefficients multiplied by 100 can be interpreted as the percentage growth in profits
associated with a one unit change in Xi. It is likely that the association with Xi will vary
from quarter to quarter, in which case is giving the average effect of Xi over the three
years post-tsunami studied.
The choice of variables to include in Xi in equation (3) is guided by several
competing hypotheses about the factors that might be most strongly linked to the speed of
enterprise recovery. Schooling and gender were discussed above. We also include
variables indicating the age of the owner and enterprise and the marital status of the
owner. We have shown that access to capital may be important, at least for retailers.
Since the tsunami destroyed physical capital of the business, access to credit is likely to
be one determinant of how quickly firms can replace this capital and recover. Our
baseline survey asks whether firms have ever had a loan from a private bank, government
bank, microfinance program, or government program. This category includes 30 percent
of untreated firms with business damage. However, 26 percent of firms still had
outstanding loans dating from before the tsunami, for which payments generally
continued to be paid. Previous use of credit may therefore not be a very good indicator of
the ability to raise new capital after the tsunami.
Another characteristic that might affect a firm's recovery is its formality status. A
firm's registration (in our case with the Pradeshiya Saba (municipal organization) or
District Secretariat) arguably offers a potential advantage by establishing a record of the
firm's existence prior to the disaster, and providing some basic information on its sales
27
and assets. Also, formalization enables firms better access to credit, according to often
heard claims. However, in our discussions with NGOs and government officials working
on tsunami recovery programs, registration of the firm apparently was not used to
identify firms as potential aid recipients, as a precondition for loans, or for verification of
pre-tsunami asset levels. We therefore expect to see no significant effect of formalization
on recovery, but include this variable as a check of the formality hypothesis.
Table 5 reports the results of estimating equation (3). Column 1 shows our base
specification. Baseline and pre-tsunami profits are positively and significantly associated
with profits in latter waves of the panel. Conditional on these, we see that profit recovery
in the absence of the capital provided by our grants is slower for manufacturing firms,
female-owned firms, and firms located closer to the coast. The speed of recovery is not
significantly associated with the owner's education, age, or marital status, or the age of
the enterprise. We also find no evidence that formally registered firms recover more
quickly. There is a sizeable positive point estimate on having previously had a loan,
consistent with greater access to capital aiding recovery. However, the standard errors are
large, so we cannot reject zero effect.
Columns 2 and 3 control for whether the firm has a major supplier or major
customer accounting for 25 percent or more of inputs or sales respectively. We see a
strong negative and significant effect of having a major customer. Firms with a major
customer average 33 percent less profit growth over this period. Overall, the recovery of
profits among the untreated sample suggests a consistent with the experimental results.
Manufacturers appear to recover more slowly than retailers.
Section 5: Conclusions
In large, well-publicized disasters affecting low-income countries, the flow of
relief and recovery aid is often very large. This paper uses data from surveys of private
enterprises and households in Sri Lanka in an attempt to understand how that aid might
28
be more effectively administered. We find that the aid flow to households for the purpose
of recovering household assets damaged by the tsunami was large and positively
correlated with the damage suffered by the household. In contrast, the aid flow to
enterprises was small and not well correlated with reported damage.
In spite of the lack of aid flowing to businesses, enterprises reported having
replaced or repaired two-thirds of the damaged assets 30 months after the tsunami, our
findings show. Surprisingly, among the smallest enterprises, the majority of the funds
used to pay for recovery came from personal savings or from loans or gifts from family
members and friends. Recovery aid and formal loans were a less important source of
finance for recovery. We find this surprising because these households are generally
thought to be the most capital constrained. Nonetheless, despite this partial recovery of
capital stock, tsunami-affected firms still had lower profits and capital stock three years
after the tsunami than similar firms not damaged by the tsunami, suggesting the recovery
process is slower than often assumed.
Using the random allocation of cash grants to enterprises, we find that returns to
capital among retailers in the recovery zone are very large--much larger than in inland
areas less affected by the tsunami. But among manufacturers, the incremental capital
provided by the grants did not result in higher profits. The high returns, of course,
provided an incentive to invest in the enterprises during the recovery period. The data
also show that the recovery was more rapid among those operating in the retail sector
than among those in manufacturing or services.
We believe that many of the findings from the tsunami experience would apply
more broadly to the recovery of firms from other disasters, especially those arising from
infrequent natural phenomena such as tsunamis and earthquakes. Arguably, firms may
take more ex ante actions when exposed to more frequent natural disasters such as
hurricanes. But even where disasters are recurrent, the majority of microenterprises are
likely to be uninsured. The general pattern of large aid flows and poor targeting toward
29
business recovery seems generalizable, as does the change in demand and supply chain
disruption leading to faster recovery for retail than manufacturing.
We interpret the data as supporting the use of cash grants in disaster recovery, but
only in limited cases. Grants to firms in retail trade stimulate more rapid recovery of
these enterprises. The data also support a greater use of cash aid in household recovery.
The spending by households in local shops provides a stimulus that is lacking with in-
kind aid. Finally, we believe the experiment has demonstrated the ability of random
grants to generate knowledge about how to increase the speed with which small
enterprises recover. If global warming leads to more frequent and more severe water-
related disasters, as climate change experts predict, this knowledge will be increasingly
valuable in hastening recovery from the growing devastation.
References
Asian Development Bank, Japan Bank for International Cooperation, and World Bank,
2005, "Tsunami Recovery in Sri Lanka: Damage and Needs Assessment," World Bank
Report.
Banerjee, Abhijit and Esther Duflo. 2005. "Growth Through the Lens of Development
Economics", in Philippe Aghion and Stephen Durlauf (eds), Handbook of Economic
Growth, Volume 1, pp. 473-552. Amsterdam: Elsevier.
Blanchard, Olivier and Michael Kremer, 1997, "Disorganization," Quarterly Journal of
Economics, Vol. 112(4), pp. 1091-1126.
Carter, Michael, Peter Little, Tewodaj Mogues and Workneh Negatu. 2007. "Poverty
Traps and Natural Disasters in Ethiopia and Honduras", World Development 35(5): 835-
856.
Dacy, Douglass, and Howard Kunreuther, 1969, The Economics of Natural Disasters:
Implications for Federal Policy, New York: The Free Press.
De Mel, Suresh, David McKenzie and Christopher Woodruff, 2008 "Returns to Capital in
Microenterprises: Evidence from a Field Experiment", Quarterly Journal of Economics,
123(4): 1329-72.
30
De Mel, Suresh, David McKenzie and Christopher Woodruff, 2009. "Are Women More
Credit Constrained? Experimental Evidence on Gender and Microenterprise Returns",
American Economic Journal: Applied Economics 1(3): 1-32.
Dolfman, Michael, Solidelle Fortier Wasser, and Bruce Bergman, 2007, "The Effects of
Hurricane Katrina on the New Orleans Economy, Monthly Labor Review, June 2007.
Eisensee, Thomas, and David Sromberg, 2007, "New Droughts, News Floods, and U.S.
Disaster Relief," Quarterly Journal of Economics, Vol. 122(2), pp.
Ferreira, Francisco and Norbert Schady. 2009. "Aggregate Economic Shocks, Child
Schooling, and Child Health", World Bank Research Observer 24(2): 147-181.
Ferguson, Roger, 2006, "Insurance Solutions for Emerging Markets," paper prepared for
London Economic Forum, December 2006.
Government of Sri Lanka, 2005 (May), "Post Tsunami Recovery and Reconstruction
Strategy."
Government of Sri Lanka, 2006, "DAD Date Report: Tsunami-Related projects as of 15
May 2006," Reconstruction and Development Authority, Sri Lanka.
Harvey, Paul, 2005, "Cash and Vouchers in Emergencies," Humanitarian Policy Group
Discussion Paper.
Jayasuriya, Sisira, Paul Steele and Dushni Weerakoon, 2005, "Post-Tsunami Recovery:
Issues and Challenges in Sri Lanka," Asian Development Bank working paper.
Kahn, Matthew 2005. "The Death Toll from Natural Disasters: The Role of Income,
Geography, and Institutions", The Review of Economics and Statistics 87(2): 271-84.
Lustig, Nora L. (2000) "Crises and the Poor: Socially Responsible Macroeconomics",
Economia 1(1): 1-30.
Maccini, Sharon and Dean Yang. 2009. "Under the Weather: Health, Schooling, and
Economic Consequences of Early-Life Rainfall", American Economic Review 99(3):
1006-1026.
Mathison, Stuart, 2003, "Microfinance and Disaster Management," memo, The
Foundation for Development Cooperation.
McKenzie, David J. 2006. The Consumer Response to the Mexican Peso Crisis.
Economic Development and Cultural Change 55(1): 139-172.
Morduch, Jonathan, 1999. "Between the state and the market: can informal insurance
patch the safety net?", The World Bank Research Observer 14(2):187�207
31
Okuyama, Yasuhide, 2003, "Economics of Natural Disasters: A Critical Review,"
working paper West Virginia University.
Oxfam, 2005, "Making the Case for Cash: Humanitarian Food Aid Under Scrutiny,"
Oxfam Briefing Note.
Roland, Gerard and Thierry Verdier, 1999, "Transition and the Output Fall," Economics
of Transition, Vol 7(1), pp.
Runyan, Rodney C. 2006. "Small Business in the Face of the Crisis: Identifying Barriers
to Recovery from a Natural Disaster", Journal of Contingencies and Crisis Management
14(1): 12-26.
Schultz, T.W. 1975. "The Value of the Ability to Deal with Disequilibria", Journal of
Economic Literature 13(3): 827-46.
Skoufias, Emmanuel. 2003. "Economic Crises and Natural Disasters: Coping Strategies
and Policy Implications", World Development 31(7): 1087-1102.
Smith, Stanley K and Christopher McCarty, 1996, "Demographic Effect of Natural
Disasters: A Case Study of Hurricane Andrew," Demography Vol 33(2), pp. 265-275.
Str�mberg, David, 2007, "Natural Disasters, Economic Development and Humanitarian
Aid," Journal of Economic Perspectives, Vol. 21(3), pp. 199-222.
Yang, Dean. 2008. "Coping with Disaster: The Impact of Hurricanes on International
Financial Flows, 1970-2002" The B.E. Journal of Economic Analysis & Policy, 8(1)
(Advances), Article 13.
32
Table 1
Sources of Recovery Funds for Small Businesses and Households
Wage
SME owners workers Panel of microenterprises (1)
Business Household Household Business Household
Number suffering damage 139 97 153 197 176
Mean damage (USD at 100Rs/$) $40,200 $6,282 $5,079 $897 (2) $2,421
% of those with loss covered by insurance 12.9% 4.1% 0.7% 0.0% 0.0%
Mean % of losses paid by insurance (conditional on
being insured) 7.5% 5.1% 33.3% NA NA
% of total losses covered by insurance 1.7% 0.2% 0.2% 0.0% 0.0%
% receiving grant for repair 23.7% 75.3% 86.4% 94.4%
% receiving loan for repair 25.9% 9.3% 20.5% 3.5%
Mean aid received from government / NGOs (3) $2,101 $2,068 $1,361 $332
Mean loan received from government or banks (3) $21,215 $3,122 $1,738 $496
Mean % losses covered by grants 1.2% 24.8% 48.9% 20.8%
Mean % of losses covered by loans 13.7% 4.6% 25.3% 0.4%
Correlation of losses and aid received 0.02 0.70 0.31 -0.02
Correlation of losses and loans received 0.16 0.27 -0.07 0.09
% of all losses covered by insurance, grants or loans 16.9% 30.2% 30.4% 21.3%
% of losses replaced/repaired by July/Oct 2007 67.8% 68.4% 60.6% 72.3% 64.5%
(1) Information on losses and insurance from October 2007 survey. Information on aid and loans from July 2005 survey.
(2) Data from the October 2007 survey. Similar questions asked in the April 2005 survey yielded very similar losses for
enterprise assets ($814), but losses of just under $900 for household assets.
(3) Mean conditional on reporting some aid.
Table 2: Effect of Grants on Profits Among Damaged Firms
Dependent Variable: Real Profits
Real Profits Own hours
(1) (2) (3) (4)
Amount 9.90** 8.96** -0.324
(4.25) (3.89) (1.84)
Cash Amount 5.27
(4.30)
Equipment Amount 12.21**
(5.04)
Trimming No Yes Yes Yes
Observations 2024 1993 1587 2095
Number of firms 200 200 172 200
Notes: Fixed effects estimation, estimated over 11 waves for firms reporting
profits in 3 or more waves.
Standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1
All regressions also include wave effects.
Trimmed samples delete the top 1% of percentage and absolute changes in
profits from one wave to the next.
Table 3: Heterogeneity in Treatment Effects
(1) (2) (3) (4) (5) (6)
Amount 19.6*** 14.53 12.88 10.17* 8.79** 19.91***
(6.16) (15.7) (11.61) (6.14) (3.74) (6.75)
Amount*Manufacturing -22.1**
(6.85)
Amount*MajorCustomer -3.28
(8.71)
Amount*MajorSupplier -2.66
(7.76)
Amount*Female -3.19
(7.70)
Amount*Years Education 1.69
(1.08)
Parents / children -9.77**
unaffected by tsunami (4.39)
Observations 1993 1993 1993 1925 1993 1790
Number of firms 200 200 200 193 200 172
Notes: Fixed effects estimation, estimated over 11 waves for firms reporting profits
in 3 or more waves.
Standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1
All regressions include wave effects and interactions of waves and variable shown.
Trimmed samples delete the top 1% of percentage and absolute
changes in profits from one wave to the next.
Table 4: Treatment and capital stock
(1)
Amount 10638***
(5831)
Amount*Manufacturing -4175*
(6910)
Observations 1857
Number of firms 200
Notes: Fixed effects estimation, estimated over 11 waves for
firms reporting profits in 3 or more waves.
Standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1
35
Table 5: Which Damaged Firms Recovered Fastest?
Dependent Variable: Log Real Profits
(1) (2) (3)
Manufacturing dummy -0.286* -0.254* -0.266*
(0.147) (0.150) (0.145)
Female owner -0.663*** -0.636*** -0.619***
(0.164) (0.165) (0.162)
Years of education of owner 0.00305 -3.02e-05 -0.00301
(0.0291) (0.0292) (0.0288)
Age of owner -0.00707 -0.00750 -0.00664
(0.00734) (0.00733) (0.00723)
Owner is married 0.0925 0.102 0.0985
(0.183) (0.183) (0.180)
Business<=3 years old 0.0468 0.0197 0.0294
(0.154) (0.156) (0.152)
Firm had a loan at baseline 0.230 0.200 0.179
(0.158) (0.161) (0.158)
Firm is registered -0.0546 -0.0351 -0.0917
(0.190) (0.191) (0.188)
Log March 2005 profits 0.245** 0.232** 0.245**
(0.101) (0.101) (0.0990)
Log November 2004 profits 0.216* 0.243** 0.217*
(0.113) (0.116) (0.111)
Log distance to the coast 0.212** 0.230** 0.239***
(0.0898) (0.0914) (0.0893)
The firm has a major supplier -0.146
(0.147)
The firm has a major customer -0.326**
(0.152)
Observations 702 702 702
Number of firms 80 80 80
Notes: Random effects estimation on untreated firms suffering business damage,
estimated over waves 2 to 11.
Standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1
Regression also contains wave effects.
36
Figure 1
Recovery of Capital Stock (Equipment)
160%
140%
120%
Pre-tsuanmi=100%
100%
Directly Affected
80%
Unaffected
60%
40%
20%
0%
Nov Mar Jun Sep Dec Mar Jun Sep Dec Mar Sep Mar
04 05 05 05 05 06 06 06 06 07 07 08
37
Figure 2: Mean Profits for Untreated Firms by Tsunami Exposure
180
160
140
Index (November 2004=100)
120
100
80
60
40
20
0
Aug-04 Feb-05 Sep-05 Mar-06 Oct-06 Apr-07 Nov-07 Jun-08
Directly Affected Indirectly Affected Unaffected
38
Appendix 1: Subsample Comparisons
39
Appendix 2: Suitability of the Inland firms as a comparison sample.
The strongest predictor of whether or not a firm was damaged by the tsunami is,
not surprisingly, how close it was located to the coastline. Table A2.1 shows the
proportion of firms in our sample that experienced asset damage from the tsunami, by
distance from the coastline. Three-quarters of firms located within 250 meters of the
shoreline experienced damage, compared to 37 percent of firms located 500 to 750
meters, and almost no firms claim to have experienced damage if located more than 750
meters from the shoreline. In some areas, the shape of the coastline mitigated the impact
of the tsunami, while in others, the tsunami caused a surge in rivers linked to the coast,
resulting in more damage slightly inland.
Did the characteristics of damaged enterprises differ systematically from
undamaged enterprises located similar distances from the shoreline? This question is
especially relevant for defining an appropriate comparison group against which to
measure recovery. If the affected firms are substantially different from unaffected firms,
the latter group will not be a relevant comparison group. We ran probit regressions on the
likelihood a microenterprise suffered some damage as a function of owner, firm, and
location characteristics. The results, shown in Table A2.2, indicate that enterprises with
more educated owners, those with higher levels of pre-tsunami profits and those which
had operated for more than three years were somewhat more likely to have suffered
damage.17 But we find no association between tsunami damage and the age, marital
status, and gender of the firm owner, nor with the industry or legal status of the firm. In
general, the characteristics of affected and unaffected enterprises appear to be similar
enough to use the latter as a comparison sample for the former.
Microenterprises in our sample were grouped into three groups: directly affected,
meaning they had suffered asset damage from the tsunami; indirectly affected, which are
firms in the same geographic areas as the directly affected, which didn't suffer damage;
and unaffected, which were firms located further inland in the same districts. The median
distance to the coastline is 261 meters for directly affected firms, 495 meters for
indirectly affected firms, and 7.2 kilometers for unaffected firms.
17
These results thus suggest that it is not the case that richer firms owned by more able owners were able
to avoid damage by virtue of better construction of premises.
40
41
42