World Development Vol. 34, No. 12, pp.

2116–2136, 2006
Ó 2006 Elsevier Ltd. All rights reserved
0305-750X/$ - see front matter
www.elsevier.com/locate/worlddev
doi:10.1016/j.worlddev.2006.03.006

Socio-Economic Impact of Super Typhoon Harurot

in San Mariano, Isabela, the Philippines
*
MARCO G.A. HUIGEN and ISABELLA C. JENS
Leiden University, The Netherlands
Summary. — This paper reports on the socio-economic effects and coping mechanisms of farm
households affected by super typhoon Imbudo in San Mariano, Isabela, the Philippines. Estima-
tions of economic losses are given based on 150 interviews among the rural population. The relative
loss per crop as part of the annual household income for yellow corn, banana, and rice were 64%,
24%, and 27%, respectively. Unexpectedly, most farm households did not change their agricultural
strategies and continued with ‘‘business as usual’’ (78%). The main explanation for this lack of
adaptation is found in the cultural and societal structure of farm households and their traders. This
paper concludes with a short-term and long-term vulnerability and resilience analysis for the house-
holds, the socio-agricultural system, and the ecological system.
Ó 2006 Elsevier Ltd. All rights reserved.

Key words — South East Asia, the Philippines, typhoon, socio-economic analysis, household
analysis, disaster coping mechanisms

1. INTRODUCTION clones, as well as the highest number of remark-

The Philippines is considered, both geo-phys-
ically and meteorologically, one of the world’s
natural hazard ‘‘hot spots,’’ and suffers more
natural hazards (e.g., earthquakes, volcanic
eruptions, typhoons, floods, droughts, and
landslides) than any other country, with an
average of eight disasters per year (Bankoff,
2003).
Typhoons are tropical cyclones with very
strong winds, of over 118 kilometers per hour,
whose epiphenomena (e.g., landslides, storm
surges, and floods) cause the largest loss of life
and property in the Philippines (of all natural
disasters). Each year about 20 typhoons, equiv-
alent to 25% of the global occurrence of ty-
phoons, occur in the Philippine Area of
Responsibility. Most of the typhoons occur
during the rainy season (July until September).
The data analysis of the typhoon events during
1880–1994 seem to indicate a slow shift towards
the end of the year, that is, the number of ty-
phoons in the rainy season seems to decrease
over time, while the number of typhoons which
occur in December have increased. The island
Northern Luzon, in which the province Isabela
is located, receives by far the most tropical cy-
2116
able typhoons (Bankoff, 2003).
* This paper was written as part of the Land Use Tra-
nsition Modelling project (LUTM). This project is an
NWO-WOTRO funded four-year project whose aim is
to understand land use and land cover changes (LUCC)
at the micro- and meso-level. The project is a joint un-
dertaking of the University of Nijmegen, the Wagenin-
gen Agricultural University (WAU), the Institute for
Environmental Sciences (CML), and the Isabela State
University (ISU-C) in the Philippines. In 1989, the CML
and the ISU-C established the Cagayan Valley Pro-
gramme on Environment and Development (CVPED) in
the Philippines. The main objective of the CVPED is to
strengthen the scientific base for policy-making in the
region. The LUTM project is grateful for the infrastruc-
ture provided by the CVPED.
The authors wish to express their gratitude to the
‘‘Typhoon Team:’’ Fatima Bagunu, Roldan Dugay,
Kenet Guingab, and Jinky for their research assistance
and friendship. Furthermore, we would like to thank
Koen Overmars and Dr. Hans Bauer for their useful
comments on an earlier draft. We would also like to
thank two anonymous reviewers for their excellent
comments on our paper. Final revision accepted: March
13, 2006.
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES
Natural hazards such as super typhoons are
recurrent phenomena that may have devastat-
ing effects on the socio-economic development
and environmental conservation programs,
but foremost on the personal lives of the people
in the disaster area. In land use change studies,
environmental conservation studies, and eco-
nomic development studies, understanding the
effects of a disaster is pivotal, because they
can drastically change the agro-ecological and
political-economic landscape. The socio-eco-
nomic effects of a disaster change the short-
term and long-term motivations and options
of farming households. Hence, on the aggregate
level, they change the land use and land cover.
Furthermore, these socio-economic effects on
the farming household level change the eco-
nomic development perspective of the farming
household (micro-level) and of the region (the
meso- or macro-level).
In this paper, a disaster is defined as the con-
sequence of a process that involves a potentially
destructive agent (hazard) and a population in
a socially created state of vulnerability. Disas-
ters are seen as a measure of a society’s success-
ful adaptation to certain features of its natural
and socially constructed environment in a sus-
tainable fashion. Disasters emerge as a result
of human–environmental interactions; they
arise when there is a lack of ‘‘mutuality,’’ a
measure of how well a society is adapted to
the environment, and how well that environ-
ment fares at the hand of human activity. As
a society interacts with its environment, it en-
gages in a series of processes over which it
has incomplete control and incomplete knowl-
edge, particularly over long periods of time. It
is these conditions that turn a natural phenom-
enon or hazard into a social crisis, a disaster
(Bankoff, 1999, 2003).
This research primarily investigates the direct
economic effects of super typhoon Harurot
(international codename: Imbudo) and the so-
cio-economic implications in the municipality
of San Mariano, Isabela, the Philippines. This
area was hit by the super typhoon on July 22,
2003. Another focus of this micro-level, actor-
oriented research are the farmers’ decision mak-
ing and coping strategies under such extreme
circumstances, within their structural and cul-
tural context. The behavioral strategies of these
primary actors, as well as the behavioral strate-
gies of the secondary and tertiary actors are
analyzed by combining economic disaster study
theories (Bull, 1994) and the Action-in-Context
(AiC) framework (De Groot, 1992).
2117
After the introduction of the research back-
ground and the explanation of the theoretical
framework and methodology in Section 2, Sec-
tion 3 continues with the results of the analysis.
Section 4 discusses the outcome and implica-
tions of this analysis. In Section 5, we conclude
by placing this paper within the context of
ongoing scientific debate regarding hazards
and disasters.
In this scientific debate the focus of research
has been primarily on societal responses to haz-
ards and too little on the processes that pro-
duce disasters (Blaikie, Cannon, Davis, &
Wisner, 1994). This reasoning implies a shift
from a static effect approach (often with an eco-
nomic focus) toward a more dynamic, politico-
economic focus. Blaikie et al. (1994) further
suggest, supporting this dynamic, process-ori-
ented approach that root causes of a disaster
reflect the distribution of power in a society.
Dynamic pressures direct the root causes to-
ward particular forms of insecurity that have
to be considered in relation to the types of haz-
ards facing those people. Accordingly, disaster
is defined as an expression of social vulnerabil-
ity (disaster is the result of the underlying com-
munity structure; of an internal and social
process). The vulnerability of people is primar-
ily rooted in social processes and underlying
causes which may ultimately be quite remote
from the disaster event itself. It is a means for
understanding and explaining the causes of
disaster (Blaikie et al., 1994). As such, a disaster
cannot be regarded as a discrete, static event,
because by doing so it becomes separated from
the activities and processes that create its con-
text. Moreover, a disaster is not only part of
its context; it is in part also caused by its con-
text. This context is not only physical, but so-
cial, institutional and political as well.
Vulnerability is the opposite of resilience,
where resilience is low, vulnerability is high,
and vice versa. In agro-ecology, ‘‘resistance’’ de-
scribes the ability of a farming system to resist
the impact of a disturbance, while ‘‘resilience’’
is the ability of the land manager and its wider
socio-ecological system to recover from a distur-
bance. Resilience has been discussed within eco-
logical theory, systems analysis, and disasters
studies (Tobin, 1999). It is a quality that enables
an organization, ecological system, household, or
nation to recover quickly from a disaster. At the
community level, class, caste, gender, ethnicity,
age, level of education, and access to resources
all determine vulnerability (Blaikie et al., 1994;
McGuigan, Reynolds, & Wiedmer, 2002).
2118 WORLD DEVELOPMENT
In order to make use of these concepts in our
paper, we distinguish between vulnerability and
resilience on the household level and vulnera-
bility and resilience on the system level. In this
paper, the system level is the aggregate of agri-
cultural and social interactions between farmers
and traders in the research area. Besides these
two levels, the ecological aspect is also impor-
tant in the analysis of vulnerability and resil-
ience. A second logical distinction in this
analysis is to differentiate between a short-term
and long-term perspective on vulnerability and
resilience. The short-term analysis has been
based completely on the data presented in this
paper and on informal interviews with key ac-
tors. Logically, the long-term analysis cannot
be supported by the presented data. This anal-
ysis is therefore mainly based on informal inter-
views and expert opinions. Understanding
these perspectives is crucial, in order to find
their potential role in disaster management
practices. In the conclusion, we will come back
to this issue of interrelation between hazards
and governance for the San Mariano context,
where we follow Wisner (2001) who claims that
vulnerability studies and practice have not
given enough attention to local capacity, espe-
cially as social capital.
2. RESEARCH BACKGROUND AND
METHODOLOGY
(a) Theoretical framework
(i) Direct, indirect, and secondary economic
effects
We began the micro-level analysis of the farm
households with an analysis of the direct effects.
In this paper we define, following UNDRO
(1979), Jovel (1989), Bull (1994), and Zapata-
Marti (1997), direct effects as the economic
damage to property and the loss of income.
Examples of direct effects are loss of capital
(housing, farm-land), loss of stocks, costs of
emergency relief and repairs, and production
loss (poor harvest, destruction of crops, death
of livestock).
In disaster economics, two other types of
disaster effects are distinguished when evaluat-
ing disaster impact, namely the indirect and
secondary effects. Indirect economic effects
may be caused by direct losses, which are
resulting from the decline in production and
in the provision of services, for example, a
reduction in activity of suppliers. Furthermore,
both direct and indirect effects may result in
secondary effects which appear some time after
the disaster. Secondary effects include, for
example, an increase of disparity between indi-
vidual and family income, ecological changes,
or negative changes in the balance of payments
(Bull, 1994; Jovel, 1989; UNDRO, 1979; Zapa-
ta-Marti, 1997; Benson & Clay, 2004). Both
indirect and secondary effects of a disaster are
measured on a meso- or macro-level (regional
or national economic scale). This paper focuses
entirely on the micro-level effects; therefore
most of the indirect and secondary effects are
beyond the scope of this paper.
(ii) Behavioral aspects
Scientific literature shows that farm house-
holds affected by a calamity, such as a typhoon,
do not remain passive, but employ several cop-
ing strategies (Del Ninno, Dorosh, & Smith,
2003; Del Ninno, Dorosh, Smith, & Roy,
2001). A coping strategy of a farm household
is a response to a calamity whose aim is to se-
cure their livelihood system. In other words, it
is a fallback mechanism which is activated
when the normal means of subsistence are dis-
rupted. The strategies adopted by communities
to reduce the impact of hazard or avoid disaster
are known as coping mechanisms, and are
based on the assumption that what happened
in the past is likely to repeat itself, following
a familiar pattern. People’s earlier actions,
therefore, constitute a reasonable framework
for guidance during similar events in the future
(Blaikie et al., 1994). The term ‘‘coping capac-
ity’’ is a significant measure of response in a
post disaster situation. Coping is defined as a
manner in which people contest, struggle, com-
bat within the existing range of resources and
expectations in a situation, in order to achieve
various ends (Blaikie et al., 1994). Coping skills
and capacities can be physical, social, economic
and institutional.
Following and adjusting Del Ninno et al.
(2001, 2003) we identify the following coping
strategies: (1) borrowing, (2) modification of
lifestyle, (3) disposal of assets, and (4) finding
alternative sources of income. Farm households
will also prepare themselves for a possible fu-
ture calamity. This is meant to reduce the antic-
ipated negative effects. The following measures
can be taken to prepare for a future calamity
(Del Ninno et al., 2001): (1) change in agricul-
tural practices; (2) diversification of income
sources; (3) building up stock and inventories;
and (4) insurance. While preparedness repre-
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES
sents an ‘‘advance’’ action based on past expe-
riences and anticipated outcomes, coping strat-
egies represent actual measures to adjust once
the event has occurred.
(b) Methodology
In 2002, the LUTM (Land Use Transition
Modeling) project interviewed 151 farm house-
holds to get a clear picture of the general agri-
cultural situation in the research area (Huigen,
2004; Overmars & Verburg, 2005). These 151
respondents represent a 5% sample of the total
farm household population and are stratified
by ‘‘barangay’’ [village] (systematic random
sampling with sampling interval 20). This data-
set has been used in this paper as base data and
reference data.
During February–May 2004, in the cropping
cycle after super typhoon Harurot, the above-
mentioned farm households participated again
in a cross-sectional interview using a structured
questionnaire. The questionnaire consisted of
more than 80 questions with the following sec-
tions: (1) general information; (2) normal situa-
tion before Harurot; (3) situation during
Harurot; (4) situation after Harurot; (5) ty-
Figure 1. Research area.
2119
phoon information and assistance; and (6) fu-
ture of the farmers. The questionnaire was
constructed by Filipino researchers in both
the English language and the local dialect.
The retrieved data is stored in a relational data-
base. For various reasons, 14 respondents of
the original dataset could not be interviewed.
Besides structured interviews, we also con-
ducted semi-structured interviews with key-
actors such as traders, middlemen, and
government officials. The performed analysis
is both of a qualitative and a quantitative nat-
ure. The quantitative analysis in Section 3 is
supported by descriptive statistics.
(c) Research area
(i) Geographic description
San Mariano is one of the 37 municipalities
in the Isabela province in region 2, the Philip-
pines and covers an area of 1,469.5 square kilo-
meters. It covers 36 barangays, of which three
are classified as urban. The research area, the
watershed of the Disabungan River, is enclosed
by the Disulap River in the north and (partly)
by the Pinacanauan River in the west, and in-
cludes 13 of the 36 barangays (Figure 1).
2120 WORLD DEVELOPMENT
San Mariano lies in the foothills of the Sierra
Madre Mountain Range. The northern portion
of this mountain range, the Northern Sierra
Madre, is remote, inaccessible, and still largely
covered by undisturbed forest. It is the largest
remnant of undisturbed forest in the Philip-
pines. Established in 1997, the Northern Sierra
Madre Natural Park (NSMNP) is the largest of
10 priority protected areas for biodiversity con-
servation in the Philippines. The park encom-
passes 360,000 ha of which 270,000 ha are
terrestrial and 90,000 ha are marine habitats.
In this Natural Park 63 globally threatened
and near-threatened wildlife species have been
recorded. Among these threatened animals is
the Philippine crocodile (Crocodylus mindoren-
sis) (Van Weerd, 2002). Besides threatened flora
and fauna, the park is inhabited by several
indigenous tribes, such as the Agta (Persoon,
Minter, & Visorro, 2003) and Kalingha (Scott,
1975).
(ii) Socio-economic description and contempo-
rary land use system
In 2000, the total San Mariano population of
40,995 was divided over 8,001 households
(NSO, 2001). San Mariano is predominantly
rural, with 82% of its total municipal popula-
tion residing in the barangays outside the urban
core. Approximately 23,000 people or 3,200
households inhabited the research area (NSO,
2001). The majority of these households (88%)
are farming households.
The San Mariano municipality is a melting
pot of ethnic identities, each with their own
agricultural practices, that migrated to the area
in the last century. Currently, more than 30 eth-
nic identities can be found in San Mariano. The
most dominant ethnic identity is Ilocano (53%),
with the Ybanags as the second largest ethnic
group (24%). Another ethnic group, the Ifugao
(3%), migrated to the area just after World War
II and in the 1960s. The original inhabitants of
Table 1. Overview of the crops per farm household in the research area
Crop name No. HH (n = 137) No. parcels (n = 463)
Yellow corn 113 (81%) 234
Lowland rice 26 (19%) 28
Banana 81 (59%) 129
Fruit trees 11 (8%) 36
Gmelina 17 (12%) 17
Grass land
Forest
Water bodies
the area, the Kalingha (4%) and the Agta (1–
2%), who already lived in the San Mariano
vicinity around the time of the Spanish occupa-
tion (16th century) (Scott, 1975), are now only
a small part of the total population. During the
logging boom (1960–90), many Filipinos from
the southern islands also arrived in San Mari-
ano to work in the logging industry.
Two satellite images were used to investigate
land use. The land cover classes are shown in
Table 1 (for a detailed description see Overmars
& Verburg, 2005). Table 1 also shows that the
majority of the farming households (81%) in
the research area produce hybrid yellow corn,
which is a cash crop. It has a crop cycle of
approximately four months and is planted
twice a year. Some farmers (8%) plant yellow
corn during the dry season and rain fed rice
during the rainy season. The second important
crop in the research area is banana. In our sam-
ple, 59% of the farm households are reported to
have at least one plot with banana trees. The
bananas are usually harvested twice a month,
and are then picked up by banana middlemen
at the farm gate. The middleman either brings
the harvest to a larger trader or transports the
bananas to the regional and national markets
himself. The third crop with increasing impor-
tance is irrigated rice (19% of the farming
households). This irrigated or lowland rice is
mainly produced for own consumption (85%)
and is only sold sporadically at the local mar-
ket. Rice has a crop cycle of approximately
100 days. The farmers in the research area plant
irrigated rice twice a year.
Table 2 shows that the largest part of the
farm households (62%) is purely cash-oriented,
because their farming system consists of yellow
corn, banana, and Gmelina (Gmelina arborea)
or a combination of these cash-crops. A small
part (9%) combines cash crops and subsistence
crops (lowland rice). The remaining farming
households (other) have a mixture of farm sys-
% Land cover (RS and GIS) Main usage
(51%) 22.2 Cash
(6%) 1.5 Subsistence
(28%) 23.3 Cash
(8%) Cash
(4%) Cash
29.5
21.1
2.5
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES
Table 2. Overview of the farm systems in San Mariano, based on a random sample of 137 farm households
Farm system No. HH
Yellow corn and banana 48
Yellow corn 27
Yellow corn, banana, and lowland rice 12
Yellow corn, banana, and gmelina 9 7
No farm 8 6
Other 33
tems with some of the above mentioned crops
and upland rice, white corn, and/or vegetables.
Furthermore, many of the farming households
(66%) have a home garden (vegetables, fruit
trees) to complement the nutrient intake of
the produced or bought rice.
(iii) Socio-cultural context: Patronage
The cultural and structural context of the
farming households is of major importance to
the understanding of their disaster coping
behavior. One principal aspect of the Philippine
society is the social bond between the farming
household and the (corn) trader. In literature
this bond is described as ‘‘patronage’’ or
‘‘clientelism.’’ This bond between the traders
and farmers derives much of its stability from
the credit base which underlies it. Patronage de-
notes a specific type of social structure. These
ties of patronage involve dyadic bonds between
individuals of unequal power and socio-eco-
nomic status (Eisenstadt & Lemarchand,
1981). The basis for patronage relationships
are often elementary reasons like kinship,
shared ethnic background, and/or school con-
nection (Van der Werf, 1994).
At the beginning of the yellow corn crop cy-
cle, the farm households have to pay either
farm laborers or a tractor to plough the land,
as well as buy ‘‘inputs’’ (seeds, fertilizer, and
pesticides). Because most farmers lack the
financial means for such investments they will
approach their financer (a corn trader often re-
ferred to as ‘‘boss’’) to borrow the investment
money (‘‘credit’’) against an interest rate of
15–25% per five months. Some traders will
not let the farmers borrow directly from them.
In this case, a ‘‘guarantor’’ or middleman is a
channel for farmers to receive loans. Guaran-
tors are a guarantee for the traders that the
loan will be paid back. They receive a percent-
age (3–5%) of the harvest and 30 centavos 1 per
kilogram from the farmer. Next, the farm
households have to pay the farm laborers for
2121
% HH Average parcel Average land
(n = 137) size (Ha) size (Ha)
35 1.00 3.45
20 0.79 1.49
9 0.86 4.02
1.06 4.60
0 0
24 1.23 3.17
planting. The maintenance of the yellow corn
is mostly done by household members them-
selves. At harvest time, farm laborers are
needed again, as well as the threshing facility,
which in most cases has to be hired. After
sun-drying the corn, the farmer has to trans-
port it to the corn trader’s storage facility. Stor-
ing yellow corn by farmers themselves is
discouraged because of sensitivity to decay
due to moisture and rodents. Without modern
dry storage facilities, the risk of postharvest
losses outweighs the potential economic gains
to be made by storing the crop until market
prices recover (a 50% drop in price during the
harvesting season is not uncommon (Polet,
1991)). The farmers have no other alternative
than to sell their crop directly to the traders,
who have the necessary transport facilities
and warehouses (‘‘bodega’’). One of the
requirements for the farm households to receive
‘‘credit’’ is to sell the corn harvest to the corn
trader. Van den Top (1998) mentions the com-
pulsory sale of the borrower’s crop to the mon-
eylender as a common form of collateral in
informal credit. The traders store the dried corn
and when the crop prices have increased after
the harvesting period they transport and sell it
on the national market.
There are eight corn traders that provide
most of the San Mariano farming households
with farming inputs and who buy the harvested
crop. The largest crop trader is associated with
approximately 600 farm households, while the
smallest trader has approximately 150 farm
households. During a cropping cycle, the farm
household often buys food supplies at the corn
trader’s supermarket on ‘‘credit.’’ Most infor-
mal credit-lines encompass production loans,
consumption loans, and emergency needs
(Van den Top, 1998). This system of market
interlinkages combines in a single agreement
the purchase of agricultural inputs, the alloca-
tion of the farmer’s labor, and the sale of the
crop. In this way, the farmers cultivate a crop
2122 WORLD DEVELOPMENT
50
40
30
20
10
0 N = 102.00
0 20000 40000 60000
Loan amount per HH
Figure 2. The reported amount farm households borrow per cropping cycle. Note: 11 farm households indicated the type
of financier but did not give the borrowed amount.
that buys them a line of credit in the informal
money-lending market. By interlinking markets
for capital inputs, crops, consumer goods, and
labor, informal moneylenders are in a better
position than formal lending institutions to re-
cover these loans. When it comes to decisions
on priorities in debt payment, the informal len-
der, who also provides daily needs on credits,
naturally has to be paid back first.
In Figure 2, a histogram of the amount
‘‘normally’’ borrowed by a farm household
per cropping cycle is shown. Because of large
standard deviations, the median represents
the farm households’ general situation better
than the average. The median normally bor-
rowed amount per cropping cycle is approxi-
mately 13,300 pesos (n = 102, avg. 17,652,
std. dev. 15,933). Figure 2 also shows that
approximately 90% of the farm households
negotiate a loan every cropping cycle; of these,
64% have a direct link to the crop trader, while
15% use a middleman to get financing for their
crops. A middleman is often a relative or
neighbor who is in direct contact with the corn
trader. Only a small percentage (4%) of the
farm households in our sample did not use
any credit. Almost all these ‘‘non-credit’’ farm
households are of the Ifugao ethnic identity
Financier type Number of HHs
Corn/rice trader 87 (64%)
Middleman 21 (15%)
Family 5 (4%)
No loan 6 (4%)
No Data 18 (13%)
Std. Dev = 15933.48
Mean = 17652
80000 100000
and have a farm system with yellow corn, low-
land rice, and banana.
Traditionally, informal moneylenders have
played an important role in the rural credit
market of the Philippines. Government owned
financial institutions repeatedly found them-
selves unable to participate effectively in the
high-risk market of small scale agricultural
credit. Acquaintance, trust, and social control
mechanisms are critical in rural finance (Van
den Top, 1998), especially when the borrower
is a poor farm household with few assets that
can serve as collateral. Farmland ownership
and cultivation rights are not a popular form
of collateral for the professional lending institu-
tions.
(d) Typhoon Harurot
PAGASA, the Philippine Atmospheric, Geo-
physical and Astronomical Services Adminis-
tration uses a four-level warning system,
based on the wind speed of storms expected
within 12–18 h: Signal Number 1 for 30–
60 kph (kilometers per hour) winds; Signal
Number 2 for 60–100 kph; Signal Number 3
for 100–185 kph; and Signal 4 for over
185 kph.
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES
Figure 3. Trajectory of super typhoon Harurot (Imbudo). Source: EORC.
On July 22, 2003, a public storm warning sig-
nal number four was raised in the provinces of
Cagayan, Isabela, and northern Aurora con-
cerning super typhoon Harurot (meaning ‘‘to
move fast’’). Harurot was the strongest ty-
phoon to hit the country since 1998. Harurot
hit land in the vicinity of Palanan, Isabela at
10:00 am and moved inland in a west/north-
west direction. The super typhoon’s center
came straight over the municipality of San
Mariano. Harurot had peak winds of 190 kilo-
meters an hour and gusts of 230 kilometers an
hour (Manila Times, 2003). The trajectory of
the super typhoon is given in Figure 3 (EORC,
2003).
In Northern Luzon, around 100,000 families
were affected by typhoon Harurot and approx-
imately 40,000 houses were damaged. Six peo-
ple died, 29 people were injured, and one
went missing. Harurot caused destruction
worth 1.9 billion pesos in agriculture and infra-
structure in Isabela. The damaged crops and
2123
livestock were worth about 1.2 billion pesos,
while the remaining damage was in infrastruc-
ture and the social sector (ACT, 2003; Manila
Times, 2003; VIC, 2003).
3. RESULTS AND ANALYSIS
(a) Introduction
We will first give an overview of the most
important actors and their general options
and motivations regarding agricultural prac-
tices after a typhoon by presenting the actors’
field (De Groot, 1992). Next, we present the di-
rect economic effects on the farm households.
The data are presented as aggregates on the
individual household level, per crop, per plot,
and per land use system. Third, we analyze
the household’s behavioral aspects regarding
their disaster coping mechanisms and their
preparation strategies. In the final part, we
2124 WORLD DEVELOPMENT
report our findings on another relevant influ-
ence for the households; the preparation and
mitigation measures by the government.
(b) Actors’ field
The main actors who form the actors’ field,
with regard to our research question, are the
farm households, the middlemen, and traders.
The trader delivers to the national crop markets
and is therefore influenced by the supply and
demand of the market. The national market is
considered part of the societal structure of the
traders and hence indirectly of the farm house-
holds’ societal structure, because national corn
price fluctuations may influence the motiva-
tions of the primary actor via the traders.
Another influential actor in the area is the
New Peoples Army (NPA), a communist, mili-
tant faction which sometimes mediates in civil
disputes to protect farmers’ interests (for a de-
tailed overview, see Rutten, 2001). So, although
the traders have a complete monopoly on the
agricultural production in the area, they cannot
become too greedy because threats from the
NPA protect the farm households from overex-
ploitation. These threats are not often directly
communicated, but are implicitly understood
by both the farmers and traders.
After the typhoon, the NPA organized two
rallies in which they addressed the traders to
alleviate the farmers’ financial conditions. In
this way, they influenced the options of the
traders. Another condition which restricts over-
exploitation is that the traders have to comply
with the social mechanism of patronage. Both
these pressures are part of the contextual struc-
ture and culture. Another actor playing a role
in the actors’ field is the government. The gov-
Table 3. Overview of plot size, land size/no. of trees per household, and damage per plot specified per crop
Crop name Avg. plot size
Yellow corn 0.96 Ha (n = 234),
(std. dev. 0.55)
Banana 0.88 Ha (n = 112),
(std. dev. 0.77)
Lowland rice 0.55 Ha (n = 28),
(std. dev. 0.29)
Fruit trees
Gmelina
ernment (both national and local) has pro-
grams that should provide help after the
typhoon.
The farm households’ motivations in the
agricultural context are primarily to meet the
needs of their families and reduce the risk of
losing income due to a typhoon. Furthermore,
they want to improve their agricultural prac-
tices. The options and motivations of the trad-
ers are simply to make as much money as
possible given the restrictions of their situation.
(c) Economic effects
(i) Direct effects; profit loss
The farm households suffered severe financial
damage due to crop loss, livestock loss, and
damage to their houses. The largest part of
the damage was due to crop yield loss. As
shown in Table 3, the farm households reported
an average of 71% yellow corn crop loss due to
the super typhoon. This loss is the difference be-
tween the amount of crop harvested during a
normal cropping cycle and the amount har-
vested after the super typhoon. The damage
to the lowland rice crop is of the same magni-
tude (72%). Furthermore, they reported an al-
most complete destruction of their banana
trees (98%). Most of the Gmelina and fruit trees
(mainly mango, mandarin, and pomelo) were
damaged in the smaller plantations (with less
than 50 trees). Younger trees (up to approxi-
mately 3 three years) were especially susceptible
to the strong winds of the typhoon.
The farm households also lost part of their
livestock. For the most part, the smaller ani-
mals were killed or lost during the super ty-
phoon. The farmers reported an average loss
of 38% of their chicken, 30% of their ducks,
Avg. land size or avg. Avg. damage
no. of trees per hh (crop loss) per plot
1.99 Ha (n = 111), 71%
(std. dev. 1.44)
500 trees (n = 81), 98%
(std. dev. 581)
0.59 Ha (n = 26), 72%
(std. dev. 0.41)
Three plots: 100%
Rest: 0%
292 trees (n = 17), Seven plots: 100%
(std. dev. 514) (small (<50) plantations)
Rest: 0% (large plantations)
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES 2125

and 20% of their pigs. Almost no damage to the
larger livestock was reported (0% horses, 0%
cows, 1% water buffalo).
Houses were severely damaged as well; 72%
of the farmers reported damage to their home
(19% no data). The average cost of house repair
was 4,124 pesos (n = 99, std. dev. 5,238). In
particular, farmers in the poorer barangays
with wooden/bamboo housing were severely af-
fected by the typhoon.
The average of 71% yellow corn crop loss (see
Table 3) corresponds to an average income loss
per household of approximately 40,000 pesos.
Figure 4 shows a histogram of the calculated
loss of yellow corn income per farm household.
The calculation of the yellow corn income loss
is given in Eqn. (1).
Lost income ðYellow cornÞ
¼ ðnormal harvest  normal price
 normal harvest  transport costs per unitÞ
 ðsuper typhoon harvest  normal price
 super typhoon harvest
 transport costs per unitÞ
For lowland rice this calculation is a bit differ-
ent, because the lowland rice produced is
mainly used for own consumption. In the calcu-
lation (see Eqn. (2)), the price farmers get for
milled rice (21 pesos per kilogram) is used.
The average gross lowland rice income of a
farm household before the typhoon is 18,857
pesos (n = 26, std. dev. 14,910), while after
the typhoon the gross income is 5,018 pesos
(n = 26, std. dev. 8,127).
Lost income ðRiceÞ
¼ ðnormal harvest  super typhoon harvestÞ
 milling factor  price ð2Þ
For bananas, the income loss is calculated over
a whole year (see Eqn. (3)) because it will take
approximately a year for the destroyed trees to
produce again. Because the investments and
maintenance of bananas are insignificant, the
gross income of banana equals the net profit
of banana. The average loss due to destroyed
banana trees per household is 23,745 pesos
ð1Þ (n = 81, std. dev. 46,043).

30

25
Number of House Holds

20

15

10

5 Std. Dev = 37433.68

Mean = 39871

0 N = 110.00
0

20

40

60

80

10

12

14

16

18

20

22
00

00

00

00

00

00

00

00

00

00

00
0

00

00

00

00

00

00

00

Gross Income Loss per Household (in pesos)

Figure 4. The calculated damage for yellow corn per household.

2126 WORLD DEVELOPMENT

Lost income ðBananaÞ  ðtotal yield  market price

¼ 12  ðnormal harvest month  price
 normal harvest month
 transport costs per unitÞ
 ðsuper typhoon harvest  price
 super typhoon harvest
 transport costs per unitÞ
As seen above, the suffered damages per house-
hold are enormous. The average damage per
household equals 88% of the total normal net
profit of the household (see Figure 5). In this
calculation we summed all the direct damages
and divided it by the estimated annual net prof-
it per household. This net profit includes in-
come derived from both farm production and
off-farm labor. The farm production is the
sum of the profit of the various crops per
household. The costs of threshing are 1/12 of
the yield. These are already subtracted from
the yield figure used. Thus, for yellow corn we
calculated the profit as follows:
Annual net profit yellow corn
¼ number of cropping cycles
 interest rate  ðinput costs
þ ploughing costs þ planting costs
þ maintenance costs þ harvesting costs
þ drying costsÞ  normal harvest
 transport costs per unitÞ ð4Þ
For lowland rice, which is mainly used for own
ð3Þ consumption, the annual net profit calculation
differs from yellow corn in several ways. The
biggest difference is that the farm household
does not have many labor costs, because they
do the work themselves or get labor assistance
from neighbor farmers according to the
‘‘amuyu’’ principle. 2 Another important differ-
ence is that the annual profit is calculated as if
the farm household had to buy the food itself
(see Eqn. (5)), hence we included the milling
loss factor (50%).
Annual net profit lowland rice
¼ number of cropping cycles
 ðtotal yield  milling loss factor
 market price  interest rate
 ðinput costs þ harvesting costsÞÞ ð5Þ

Total Damage as a percentage of the

annual net profit per Household
45

40

35
Number of Households

30

25

20

15

10 Std. Dev = 78.19

5 Mean = 88

0 N = 136.00
0 100 200 300 400 500
50 150 250 350 450
Percentage (damage/annual net profit)

Figure 5. Total damage (in pesos) as a percentage of the annual net profit per household.
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES 2127

The net profit calculation for bananas (see Eqn.
(6)) is simple; banana trees require little mainte-
nance. The farm households do the planting
and the harvesting themselves. Therefore, we
do not include any labor costs in the investment
calculation for banana.
Annual net profit bananas
¼ number of harvesting months
 ðtotal yield per month
 market price
 normal harvest month
 transport costs per unitÞ
In Figure 5, the calculated outcome of the total
damage as a percentage of the estimated annual
net income per household is given.
Table 4 shows that the yellow corn produc-
tivity loss (64% of the total annual profit,
calculated according to Eqn. (7)) has the big-
gest financial impact on the farming house-
holds. This is explained by the high
investment costs in comparison to the other
crops. Another cause is that the investment
for corn is in most cases paid with ‘‘credit’’
for which the farm household has to pay back
interest.
Relative loss crop
¼ ðNormal annual profit crop
 profit crop in typhoon yearÞ
=total annual profit ð7Þ
The average relative loss due to livestock dam-
age is rather low, because the larger and more
expensive livestock animals are less vulnerable
to the strong winds than the smaller animals.
The typhoon did not cause much rain or flood-
ing, which would have caused much more live-
stock damage.
In Table 5, the relative losses for the farming
systems with yellow corn, banana, and lowland
rice is lower than those with only yellow corn
ð6Þ and/or banana. However, the absolute loss
for this farming system is higher. Independent
sample t-tests indicate that the size of land is
an important contributor to this difference. In
general, farms with more diversity have larger
land sizes; therefore the absolute profit and loss
are higher.
Unfortunately, no data was available on the
financial damage of the Gmelina or fruit trees,
because neither of these agro-forestry crops
have a marketing system yet. For both crops
the financial damage, as well as the profit, are
excluded from the calculations. This affects
mainly the outcomes for the farm system with
yellow corn, banana and Gmelina. We may

Table 4. Overview of average relative losses per damage category

No. HH Percentage crop profit of total Relative loss (damage as percentage
household profit of annual profit)
Min. Max. Avg. Std. dev. Min. Max. Avg. Std. dev.
Banana 81 0.3 85.7 24.0 20.9 0.3 85.7 23.6 20.8
Yellow corn 110 0.4 98.4 70.9 22.8 3.7 328.1 64.0 45.1
Rice 29 6.3 96.9 50.0 26.4 3.1 86.8 26.8 20.3
House 100 0.1 444.4 17.5 51.5
Livestock 61 <0.1 80.0 4.4 10.7

Table 5. Overview of average losses per household grouped by farm system

Farm system Loss typhoon Normal annual Relative loss (%)
(Pesos) profit (Pesos)
Avg. Std. dev. Avg. Std. dev. Avg. Std. dev.
No farm 13,959 35,183 22,500 0 62.0 156.4
Other 38,382 34,263 61,519 57,781 68.1 43.9
Yellow corn 35,349 29,596 51,548 53,830 104.3 86.6
Yellow corn and banana 72,288 82,447 89,226 102,482 93.7 50.0
Yellow corn, banana, and gmelina 78,165 64,238 103,003 112,055 101.6 61.4
Yellow corn, banana, and lowland rice 98,221 87,223 150,652 140,201 65.2 15.6
2128 WORLD DEVELOPMENT
expect a relatively higher absolute and relative
loss for this farm system, because the invest-
ments of this perennial are almost completely
lost without returns. Because all Gmelina cal-
culations are excluded, making the calculations
for the fourth and fifth farm system in Table 5
equal, the table indicates that especially the
farm systems with higher profits, and thus big-
ger lands, have Gmelina plantations.
(ii) Indirect and secondary effects
This paper focuses on the micro-level, while
indirect and secondary effects are mainly wit-
nessed at a meso- or macro-level. Yet, some ef-
fects need to be explained and analyzed to get a
clear picture at the micro-level. One of the men-
tioned indirect effects mentioned in the litera-
ture is the reduction in supplier activity. Like
the farm households, the input- and money
suppliers in San Mariano also suffered severely
from the effects of the typhoon. They lost a
large part of their annual income and they
had to take a risk by increasing the loans of
their farmers in order to get their previous
investments back. After Harurot, the median
balance 3 per household was 7,000 pesos
(n = 54, avg. 12,690, std. dev. 15,425), so in
the cropping cycle following ‘‘Harurot’’ the
median outstanding debt is 20,300 pesos per
household (balance plus normal amount bor-
rowed). The traders’ total outstanding loans
with the farmers are between 1 and 10 million
pesos per trader depending on the number of
‘‘connected’’ farm households. Before the ty-
phoon, the traders’ investments in outstanding
loans equaled approximately 10–15% of their
capital. After the typhoon this amount in-
creased to 20–35%. Increasing these credits is
risky for the traders, because they have no
guarantee that farmers will be able to pay it
back. In general the traders use, as a rule of
thumb, the farmer’s average yellow corn yield
multiplied by the market price as the maximum
amount to be borrowed by the farmer. The
financial buffer of the traders was large enough
to increase the outstanding loans of most of the
farmers, but the traders refused credit to
approximately 10–15% of the farmers, because
they lost trust in their ability to pay the loans
back. The total number of suppliers did not
change; all the traders could stay in business.
Another important secondary effect of the ty-
phoon is the ecological impact. San Mariano is
bordering the NSMNP, which has one of the
last undisturbed virgin forests in the Philip-
pines. After the typhoon, the forest edge was
on fire for three weeks. Reasons for the forest
fires as indicated by the farm household inter-
views were diverse and inconclusive. The three
main reasons were (1) ‘‘Hunters want to chase
out the animals and sell the meat,’’ (2) ‘‘Farm-
ers near the forest are making new ‘kaingins’
(slash-and-burn fields) and the winds burn the
forest,’’ (3) ‘‘It was the lighting that put the for-
est on fire.’’
As a reaction to their large financial loss,
many farm households (36%, n = 136) decided
to gather non-timber forest products and to go
logging. These collected forest products were
either used for house repair or sold to the local
wood-processing industry. The extent and ram-
ifications in terms of erosion risks and land-
slides are unknown.
(d) Farm households’ behavioral aspects
In the 2002 interviews, the farm household
identified the main damage risks per crop per
plot. In Table 6, the outcome of this question
is given. The answer that was given most fre-
quently per crop is indexed as 1.00 which allows
showing the relative importance of the other
answers. Note that the farmers did not indicate
the risk of a typhoon as an important one for
rice and corn, while in the previous section it
is shown that these contributed most to the
financial damage per household (see Table 4).
In the 2004 questionnaire, most of the farm-
ers (78%) reported that they did not adapt their
agricultural strategy; they did not change any-
thing in their agricultural approach and contin-
ued their farming as usual. A minority (17%)
reported to change some of the crops. Four
percent of the respondents reported that they
stopped farming.
(i) Borrowing
Literature shows that money borrowing is
one of the main coping mechanisms used by
households in response to a calamity. Del Nin-
no et al. (2001) found that households that suf-
fered severe crop losses of at least 24% of the
total value of anticipated production for the
year mainly adjusted to the shock of the 1998
Bangladesh floods by borrowing money. As
he noted:
‘‘. . .by far, borrowing was the major coping mecha-
nism of households sampled both in terms of the
value of the resources and the number of households
that borrowed. About 66% of poor, flood-exposed
households in the sample borrowed money in the
months immediately following the flood and of these
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES 2129

Table 6. Relative damage risk factor per crop per plot as indicated by the respondents (2002)
Trees Rice Corn Banana
Animals 0.78 1.00 1.00 0.37
Fire 0.44
Insects 0.11 0.64 0.78 0.04
Tungro 0.22 0.33
Typhoon 1.00 0.06 0.08 1.00
Birds 0.58 0.15 0.19
Drought 0.04 0.17 0.06
Floods 0.08 0.09
Stem borer 0.13
Geophysical 0.01 0.07
Heavy rains 0.01
Total number of answers 23 120 425 111
Note: The case with the highest number of answers is indexed as 1.00.

52% borrowed money for food. Average household
debt went to 1.5 times its average value.’’
We also find this tendency of increased debt;
the median loan per household before the ty-
phoon was 13,300 pesos, while after the ty-
phoon the median debt per household was
around 20,300 pesos (1.5 times higher). How-
ever, farm households did not increase the
amount of their loan, but they simply could
not pay back their previous loan because of
harvest failure. Hence, in our research borrow-
ing cannot be noted as a coping mechanism.
The increase in debt shows the negative impact
of the typhoon on the balance of payments of
the farmers.
(ii) Modification of lifestyle
Our findings show that one of the major cop-
ing strategies applied by households that are af-
fected by a calamity is to change their eating
habits. Almost all farm households (93%,
n = 126) reported that after the typhoon they
did not have enough money to meet the needs
of their family. More than half (53%) of the
households adjusted this financial effect by min-
imizing the meat intake and relying on less pre-
ferred and less expensive foods. Only 6% of the
households reported that they reduced the
number of meals eaten in a day.
Another possible coping mechanism regard-
ing lifestyle is the reduction of investments.
For a Filipino family child education is one
of the most important investments. Thirty-one
percent of the families with schoolchildren indi-
cated that they had to adjust the money spent
on the education of their children. Eight chil-
dren stopped going to school permanently,
while 13 children stopped temporarily so that
they could help increase the income of their
family.
(iii) Disposal of assets
While in literature the disposal of assets, such
as livestock and land, is indicated as an impor-
tant coping strategy for households exposed to
calamities, not many farmers in our sample re-
ported such disposal. Only four farmers (4%,
n = 103) reported the selling of livestock and
three farm households leased their land (3%,
n = 103). One of the main reasons indicated
by farmers and traders not to apply this strat-
egy is that households are losing a source of in-
come, which makes them more vulnerable to a
next disaster and that it will be harder to pay
back the outstanding debt.
(iv) Finding other sources of income
To obtain an additional income, farmers par-
ticipate in often illegal, non-farming activities,
such as hunting, fishing, and logging. Further-
more, the children were withdrawn from school
and required to work. Local employment is
preferred but limited; this results in migration
patterns to urban centers and nearby towns in
search of employment. Twelve farm households
reported that a member of their family (often
an unmarried daughter) went to Manila for
labor, and 16 households reported a member
starting to work outside the research area after
the typhoon. Hence, a total of 20% (n = 132) of
the farm households reported semi-permanent
migration for employment.
Ninety farm households (68%, n = 132) also
reported that the head of their family worked
2130 WORLD DEVELOPMENT
as an off-farm laborer. Before the typhoon only
73 farm household heads (55%, n = 132) re-
ported off-farm labor as a secondary source of
income. Therefore, the number of reported
off-farm labor households increased by 23%
after the typhoon.
Another important source of additional in-
come after the typhoon is illegal logging; 49
farm households reported that they gathered
bamboo, rattan, construction wood, and/or
furniture wood from the forest for house repa-
ration and additional income. Before the ty-
phoon, only six households reported illegal
logging as a source of income. The farm house-
holds that did not gather wood or non-timber
forest products indicated that the prohibition
of the activity and the distance to the forest
were the main reasons that discouraged them.
The farm households involved in logging are
mainly located at the outer barangays near
the forest. This activity to obtain an additional
income has devastating effects on the forest.
(v) Change in agricultural practices
In our original set of hypotheses, we expected
that due to the financial problems farm house-
holds were forced to change their cash crops,
especially yellow corn, to crops with lower
investment costs (subsistence). However, our
results show no sign that this hypothesis is true.
106 households (78%, n = 136) reported that
they did not change anything and continued
their farming system as usual.
A change to a more typhoon-resistant crop
(e.g., watermelon or sweet potatoes) is consid-
ered a typhoon preparation measure, yet such
a change was not observed in our research area.
Several farmers (17%) made changes in their
crop-choice, but typhoon resistance was not a
motivation for these changes. Most of these
changes (35%, six cases) consisted of changing
from banana to yellow corn. This change can
be explained by the fact that banana trees need
at least a year before they bear fruits again,
while yellow corn can be harvested sooner.
Hence, the motivation for this change is ‘‘quick
cash.’’
Approximately a quarter of the households
(26%) stopped planting a certain crop on their
plot; 36% of these households (12 households)
stopped their banana production and 36%
stopped planting yellow corn. The main moti-
vation given to stop planting yellow corn is
the lack of money for the inputs. To stop pro-
ducing bananas reflects a lack of trust in the
profitability of bananas.
Another preparation measure is crop diversi-
fication. However, our results do not suggest
that farmers diversified their crops. Most of
the farm households continued with the crops
they used before the typhoon.
Yet another typhoon preparation measure is
varying the planting time. The usefulness of this
measure is visible in our dataset, because most
of the yellow corn and lowland rice plots with
minimal damage were planted later than the
majority of the crops. The farmers indicated
that these crops were not yet in their flowering
stage and thus less susceptible to the strong
winds. However, farmers did not purposely ad-
just their time of planting to prevent typhoon
damage, because it is hard to predict a ty-
phoon. The risk to plant later in the cropping
cycle is taken to get a higher price for their
crop, but also involves a higher probability that
the crop is destroyed by the monsoon rains.
(vi) Diversification of income sources
In the literature, the diversification of income
sources is mentioned as a preventive measure to
deal with the economic shocks caused by ty-
phoons. As mentioned above, the farm house-
holds searched for other jobs after the super
typhoon. However, all respondents gave a
short-term motivation; they needed an addi-
tional income. Hence, it was merely a short-
term financial coping solution and not intended
as a structural change. In our research area,
diversification of income sources and hence
being less dependent on employment in only
one sector might be the aggregate result, but
is not a farm household’s motivation.
(vii) Building up stock and inventories
Building up a livestock or crop stock is re-
garded as a preparation strategy to deal with
the effects of future calamities, because such a
financial buffer might prevent farm households
from ending up in a downward spiral. How-
ever, our respondents indicated that building
up a stock is impossible, since they hardly have
enough money to meet their primary needs. If
they have any extra money it is used in most
cases to pay back loans. Only a few, more for-
tunate farmers are able to build up a stock. The
investment costs for the construction of a grain
silo is more than eight times the average annual
farm household income. The investment costs
for a piggery are less, approximately half of
the average annual income. Some better-off
farmers with savings reported that they con-
sider such an investment as a viable option.
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES
(viii) Insurance
In the early 1990s, formal financial institu-
tions, which were initiated by government ef-
forts, gave farmers an access to crop
insurance and agricultural loans. These efforts
failed however, because the farmers did not
pay back their loans and did not pay their con-
tribution fees. The main reason for this failure
is that the ‘‘paper trail’’ from farmer to bank
is much too complicated and ‘‘scary’’ for the
less educated farmers. They often lack the nec-
essary knowledge to fill in the paperwork and
the needed collaterals. Furthermore, the rela-
tion maintenance with the bank is too costly;
traveling to the bank in the city and back is a
two-day trip, which the farmers cannot afford.
(e) Information, government assistance, and rally
(i) Typhoon warning
Approximately, a third of the households
(34%, n = 116) were not informed beforehand
about the approaching typhoon, and therefore
could not take any precautionary measures at
all. Most of the people that were informed be-
fore the approaching typhoon heard it on the
radio (86%, n = 77). Typhoons are ranked
from signal number one to signal number four,
where the number indicates the strength of the
typhoon (with four being the strongest). Many
people mentioned that the typhoon was an-
nounced as signal number two, which is why
they did not take special precautions. However,
it turned out to be signal number four, doing a
lot more damage than was expected.
(ii) Government assistance
In most cases (75%, n = 137), the govern-
ment provided help after the typhoon by dis-
tributing relief goods such as rice, sardines,
and coffee. 13% of the households reported
not to have received any help from the govern-
ment, while we do not have data about the
remaining 12% of the households. Households
with completely damaged houses were sup-
posed to receive 1,000 pesos. However, only
few households (4%, n = 137) actually received
it and, according to the interviews, these were
not always the households that needed it most.
Free inputs would be supplied by the govern-
ment if the farmers would use these within 15
days. This was not a feasible condition for sev-
eral reasons. Firstly, it was in the middle of a
cropping cycle, and corn planting requires good
timing. Secondly, the parcels still contained
partial cropping and thirdly, it requires more
2131
than 10 days to prepare a plot for a new crop-
ping, which was made more difficult because
technical labor support (tractor) was difficult
to arrange and the farmers lacked the finances
for payment of the laborers.
(iii) Rally
After super typhoon Harurot, a march was
organized to protest against the traders’ low
corn prices and high interest rates. This rally
was called ‘‘Apo Daga’’ (‘‘My Beloved Land’’).
Because the national price was high and the
harvesting season had not yet started, the trad-
ers could easily react to the protest by increas-
ing the price to nine pesos per kilo. However,
during the harvesting season, the prices lowered
again to the normal price of six pesos per kilo.
Again, another rally was organized called
‘‘Maipanggep ti Harurot’’ (‘‘About Harurot’’).
In this rally, the organizers (church and NPA)
had various demands. The traders responded
by decreasing the interest rate of the already
outstanding loans to 13%; however, they also
increased the input costs. Most of the farmers
(94%, n = 124) thought that the protest
marches were effective and that the farmers
would benefit from the lower interest rates
and the higher corn prices.
4. DISCUSSION; THE LACK OF
AGRICULTURAL COPING
STRATEGIES EXPLAINED
In the last 15 years, three major typhoons
have struck the San Mariano area (in 1992,
1998, and 2003). Such calamities have devastat-
ing effects on the inhabitants’ personal lives, the
socio-economic system, and the environment.
Figure 5 shows that the farm households’ aver-
age loss due to the super typhoon Harurot
equals 88% of their calculated annual income.
Furthermore, the median normal net profit
per household per cropping cycle (six months)
is around 24,300 pesos (n = 135, avg. 39,165
std. dev. 45,585) per farm household. The bor-
rowed amount (‘‘credit’’) corresponds to
approximately 55% of the farm household’s
net profit. If we subtract an estimated ‘‘liveli-
hood cost’’ (=[average household size]
5.6 * [average cost per month] 600 * [number
of months] 6 = 20,200 pesos), then a household
has a net savings of 4,100 pesos per cropping
cycle. Hence, it will take the farming household
six to seven good harvests to pay back their
debts. The farmers report that they feel
2132 WORLD DEVELOPMENT
committed to pay back their loans to their trad-
ers. They also report that they do not feel
responsible for the loan when a trader discon-
nects them, or denies them further credit.
Contrary to the literature and our hypothe-
ses, the farm households showed little to no
coping strategies with regard to their farming
practices. The farm households mainly coped
with the hardships by: (1) changing their food
patterns (eating less meat) (53%), (2) gathering
wood (debris) for house reconstruction and sale
on the local market (38%), (3) doing more off-
farm labor (27%), and (4) decreasing human
capital investments such as education of chil-
dren (13%).
This leaves us with the question of why the
households do not change their agricultural
practices and why they do not adapt their
farming systems to systems with a higher ty-
phoon resistance and smaller risks. The main
reason for this lack of change is found in the
structure of the farm households’ society. As
mentioned in Section 2(c)(iii), the social tie be-
tween corn traders and farm households plays
an important role in the contemporary Fili-
pino society. Farm households borrow their
crop investments against, from a trader’s point
of view, an interesting interest rate and com-
mit themselves to sell their crop to their trad-
ers. However, this mechanism is more than a
series of economic transactions. It is embed-
ded in an exchange of social capital in which
trust plays an important role; traders lend
the crop investments to a farm household
without a contractual guarantee of loan pay-
back, while the farmers heavily depend on
the traders for the continuation of their liveli-
hood and for financial help during emergen-
cies such as hospitalization (Van den Top,
1998).
This societal structure required the traders to
lend money to the farm households again after
the typhoon to secure their already outstand-
ing loans and to carefully safeguard their so-
cial position. After the typhoon, the traders
increased the outstanding debts to the farmers
with a factor of 1.5. In cases where a trader
will not provide a new loan to a farm house-
hold, it is likely that the farmer will not pay
back its outstanding loan, but more impor-
tantly, the trader might lose the reputation of
being a ‘‘good boss,’’ which negatively affects
the trader’s relationship with the other farm-
ers. Without the farmers’ crops, their business
will cease to exist. Moreover, if the traders will
not provide a new loan, the farmer will have to
find another credit-provider and the guarantee
that the trader will get the investment back is
enormously reduced. Giving out new loans
also includes a large risk for the traders; if
the next cropping cycle does not yield a good
harvest the debt of the farmers will only in-
crease.
The traders’ increase of the outstanding loans
functions as a financial buffer for the farm
household. This buffer causes the financial
hardship for the farm households to be less
noticeable than the annual profit loss suggests.
Basically, this explains why the farm house-
holds did not feel the need to activate coping
mechanisms involving their farming practices.
The money for the next cropping cycle could
be borrowed and thus there was no need for
the farm households to change their practices
in order to minimize disaster risks. A possible
activation of a coping mechanism to change
agricultural practices is even discouraged by
the traders and restricted by the farmers’
‘‘debt-trap.’’ Traders do not want farmers to
change to another crop, because their busi-
nesses cannot adapt to other crops. The farmers
are in a ‘‘debt-trap’’ that prevents such an agri-
cultural adaptation; the outstanding loan gives
the trader the ‘‘right,’’ within the explained so-
cial context, to ‘‘force’’ farmers to continue yel-
low corn.
Another important reason for the lack of
implemented agricultural coping and prepara-
tion strategies is that most farm households
have few resources and hardly any knowledge
on typhoon risk reduction strategies. There
are no non-governmental organizations or gov-
ernmental institutions promoting agricultural
changes and instructing farmers to reduce ty-
phoon risks.
A third important reason why farm house-
holds did not change to other crops after the
typhoon is that they lack the necessary mar-
keting channels for possible new crops. An
even stronger reason is that the San Mariano
farmers have had negative experiences with
the implementation of new crops. In the past
10 years, several new products (Gmelina trees
mainly promoted via governmental programs
and fruit trees promoted via a long-term pro-
ject of PLAN INTERNATIONAL) have been
introduced to the farmers, but the necessary
marketing channels are still missing, making
these products unprofitable for the farm
households. Farmers have a great sense of
insecurity regarding the possibility of selling
new products.
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES
5. CONCLUSIONS
(a) Vulnerability and resilience
Between 1988 and 1997, an average of 88% of
the casualties caused by disasters due to natural
hazards each year were in developing countries.
Over the same period, on average, 98% of those
directly affected each year lived in developing
countries (IFRC, 1999). Most of the economic
losses are felt in wealthier countries, but the
economic impact of disasters on developing
countries is also severe, and the relative impact
is often greater (Benson, 1997). The impact of
disasters is not felt evenly within countries,
either. Studies have shown that, in general, it
is the weaker groups in society that suffer worst
from disasters: (in particular) the poor; the very
young and very old; women; the disabled; and
those who are marginalized by race or caste
(Blaikie et al., 1994).
At the household level, poverty is regarded as
the single most important factor determining
vulnerability, in part reflecting the location of
housing, primary types of occupation and the
level of access to financial and other resources.
The poverty-exacerbating nature of vulnerabil-
ity is attributable not only to post-disaster re-
lated damage, temporary loss of income
generating opportunities and increased indebt-
edness but also to deliberate risk-averting live-
lihood choices that poorer households may
make. At first sight, the short-term vulnerabil-
ity of the poor, individual households in our
study seems relatively low from an agricultural
point of view. The data does not show any ma-
jor agricultural changes as reaction to the
shock. Both the size of the agricultural area
and the composition of the crops after the ty-
phoon do not differ significantly compared to
the situation before the typhoon. However, this
agricultural resilience of the farmers can be
attributed almost entirely to the resilience at
the system level. The coping mechanisms exe-
cuted by the households illustrate that they
are vulnerable in the short run. Our research
approach does not allow a proper estimation
of the effects in changes in nutritional intake,
but informal interviews indicated that several
youngsters and older people became ill due to
lack of food.
Also, the long-term vulnerability of the indi-
vidual households was affected. Households
had to stop sending their children to school to
save money; while children with a good educa-
tion will in general be able to contribute more
2133
to the economic resilience of a household than
children who leave school at a young age.
Therefore, the long-term consequences of the
loss of human capital investments will primar-
ily affect the potential resilience of a household.
The increased debts with the traders is obvi-
ously the most important factor affecting both
the short-term and long-term vulnerability of
the individual households in our study. Their
increased debts with the traders increased the
pressure of the social debt trap the farm house-
holds are in, reducing even more their already
scarce potential of exploring other opportuni-
ties such as adapting their agricultural practices
into less hazard prone ones.
The households living near the edge of the
forest were able to cope with the economic vul-
nerability by compensating the economic loss
with income from illegal logging. This resilience
on the household level is the direct opposite of
the resilience on the ecological level; obviously,
logging negatively affects the long-term vulner-
ability of the ecological system. Following the
effects of this along the complete cycle, the
decreasing quality of the ecological system will
lead to an increase of natural hazards, such as
floods. Furthermore, it is mainly young people
who abandoned their education go to the forest
to illegally log and earn an additional income
for their household. Informal interviews show
that they do not return to school when the
financial shock has more or less been overcome,
but they continue to earn money in this ‘‘easy’’
way.
On the system level, the agricultural complex
has shown to be rather resilient in the short
run. Few changes took place in the functioning
of the system, especially in the relative sense;
everything continued as it was before. In an
absolute sense, the change is more difficult to
estimate. Traders indicated that they discon-
nected about 10–15% of the farmers from credit
provision, but, based on our data, most of these
farmers have been able to get credit from an-
other trader. The social structure seemed a little
more sensitive to the economic shock in the
short run. This can be concluded from the
spontaneous organization of the rallies. How-
ever, these rallies did not escalate into social
disruptions, and the agricultural system has
also shown to be resilient to this expression of
social stress. Everything went back to normal
on the system level. The only difference is that
outstanding loans have increased. In earlier
years, the traders had a larger financial buffer
because they made more profit. The effects of
2134 WORLD DEVELOPMENT
calamities were then less noticeable. However,
the price of corn barely changed over the years,
causing the profit and hence, the buffer to de-
crease. Now, the financial buffer has been af-
fected severely by the super typhoon. This
could have major consequences in case of a
next calamity and thus in the long run. This
has also been indicated clearly by the traders.
According to them, a lot of farmers will have
to be disconnected from the credit system if
such a disaster happens again.
We can only speculate about the conse-
quences of a second calamity soon after a first,
but it seems reasonable to expect that the resil-
ience limit of the agricultural system will be ex-
ceeded, leading to stress in the social system.
This fear is also being expressed by the inter-
viewed traders; if a lot of farmers do not qualify
for credit they might revolt, possibly organized
by the NPA. The interviewed traders also indi-
cated that a toughening of the interactions be-
tween them and the farmers can already be
seen. More often than before, they get the na-
tional police involved in collecting debts from
farmers that they do not trust, and demand col-
lateral as a payback for loans.
In the long run, it is likely that the contempo-
rary system level organization has serious limi-
tations, especially for shocks that are common
to all members of an informal insurance group.
The absence of an elective public safety net sys-
tem ultimately prevents households from ever
escaping from poverty or from reaping the ben-
efits of future economic growth. There is a
strong possibility that these households may
never be able to replenish their stocks of pro-
ductive assets, thus remaining in poverty per-
manently or for years after the typhoon.
Another unfortunate consequence is that they
may transmit poverty from the current genera-
tion to the future generation.
(b) The future
The need for action to reduce disasters’ im-
pact has been acknowledged for some time.
This acknowledgement lay behind the designa-
tion of the 1990s as the United Nations Inter-
national Decade for Natural Disaster
Reduction (IDNDR). Nevertheless, hazard risk
and disaster reduction remain marginal issues
in development planning, political commitment
appears to be weak, and actions on the ground
are as likely as not to be individual and one-off
rather than collective and long term (Benson &
Twigg, 2004; Twigg, 2001).
While vulnerability to natural hazards are an
unequally distributed public bad (since disas-
ters typically strike sub-sectors or whole com-
munities, not isolated individuals), measures
to reduce vulnerability are, to a large extent
public or collective goods. The collective nature
of vulnerability reduction (both in preventive
and response capacities) implies that effective
responses need to involve effective institutions
and governance. Governance is the complex
of mechanisms, processes, relationships, and
institutions through which citizens and groups
articulate their interests, exercise their rights
and obligations, and mediate their differences.
Based on our findings, we can conclude that
various governmental tasks and institutional
structures, that could provide help in case of
calamities such as a super typhoon, are not
yet operating at a suitable level in the San
Mariano region. Governmental actions offered
little structural help. Farmers expected and
were promised more help from the government
than they actually received. The government
does not provide structural help via informa-
tion programs, partly because they lack the
knowledge themselves and partly because they
have other priorities. Interviews with govern-
ment officials show that they have the same
attitude as farmers: ‘‘a typhoon just happens;
you cannot do anything about it.’’
Finally, natural hazards are only one of the
many problems that the San Mariano farmers
have to deal with. The San Mariano population
is growing rapidly, legal land availability is low,
alternative jobs are very scarce, and the tropical
rainforest resource depletion continues; all this
makes them more vulnerable to further chal-
lenges. In order to cope with the challenges
ahead, government agencies, NGOs, and civil
initiatives should plan ahead and mobilize
resources for risk mitigation measures. In the
Isabela province, an initiative such as the
Corporate Network for Disaster Response
(CNDR) is needed, whose mission it is to mobi-
lize the corporate sector in the Philippines to
respond to natural and man-made disasters by
facilitating the mobilization and distribution of
resources from the corporate sector to commu-
nities affected by disasters and by strengthening
linkages with government and non-government
organizations undertaking disaster response
programs. Important in such initiatives, is to
keep the planning and implementation objec-
tives within its context; without the farmers
and especially the traders’ acceptation and
cooperation, nothing will work.
 SOCIO-ECONOMIC IMPACT OF SUPER TYPHOON HARUROT IN THE PHILIPPINES
NOTES
1. One hundred pesos equal $ 1.77; 1 peso equals 100
centavos.
2. ‘‘Amuyu’’ [Ilocano] is a concept in which neighbor-
ing farmers reciprocally help each other in planting and
harvesting without the transfer of money.
REFERENCES
Action by Churches Together (ACT) (2003). Super
typhoon hits the Philippines, 23 July 2003. http://act-
intl.org/alerts/Alerts_2003/AlPhilTyph-1-03.html.
Bankoff, G. (1999). A history of poverty: the politics of
natural disasters in the Philippines, 1985–95. The
Pacific Review, 12.3.
Bankoff, G. (2003). Cultures of disaster: Society and
natural hazards in the Philippines. London: Routl-
edge.
Benson, C. (1997). The economic impacts of natural
disasters in the Philippines. ODI Working Paper No.
99. London: Overseas Development Institute.
Benson, C., & Clay, E. J. (2004). Understanding the
economic and financial impacts of natural disasters.
Disaster Risk Management Series No. 4. Washington
DC: World Bank. http://www-wds.worldbank.
org/servlet/WDS_IBank_Servlet?pcont=details&eid=
000012009_20040420135752.
Benson, C., & Twigg, J. (2004). Measuring mitigation;
Methodologies for assessing natural hazard risks
and the net benefits of mitigation – A scoping study.
http://www.reliefweb.int/rw/lib.nsf/db900SID/OCHA-
6BCM2U/$FILE/Measuring_Mitigation_report.pdf.
Blaikie, P., Cannon, T., Davis, I., & Wisner, B. (1994).
At risk: Natural hazards, people’s vulnerability and
disasters. New York: Routledge.
Bull, R. (1994). Disaster economics. Disaster Management
Training Programmes. Geneva: UNDP and DHA.
De Groot, W. T. (1992). Environmental science theory;
concepts and methods in a one-world, problem-oriented
paradigm. Amsterdam: Elsevier, 583 p.
Del Ninno, C., Dorosh, P., & Smith, L. (2003). Public
policy, markets and household coping strategies in
Bangladesh: avoiding a food security crisis following
the 1998 floods. World Development, 31(7),
1221–1238.
Del Ninno, C., Dorosh, P., Smith, L., & Roy, D. K.
(2001). The 1998 floods in Bangladesh: disaster impact,
household coping strategies and response. IFPRI
Research Report 122. IFPRI, Washington, DC.
Earth Observation Research Center (EORC) (2003).
http://www.eorc.nasda.go.jp/TRMM/typhoon/images/
track/jpg/bst_2003s.09W.IMBUDO.jpg.
Eisenstadt, S. N., & Lemarchand, R. (Eds.) (1981).
Political clientelism, patronage, and development.
Beverly Hills; London: Sage (Contemporary political
sociology, ISSN 0305-3482; 3).
Huigen, M. G. A. (2004). First principles of the
MameLuke multi-actor modelling framework for
2135
3. The balance is the part of the borrowed amount that
a household could not pay back to their trader.
land use change, illustrated with a Philippine case
study. Journal of Environmental Management, 72,
5–21.
International Federation of Red Cross And Red Cres-
cent Societies (IFRC) (1999). World Disaster Report
1999.
Jovel, R. (1989). Economic and social consequences of
natural disasters in Latin America and the Carib-
bean. United Nations Economic Commission for Latin
America and the Caribbean (UN-ECLAC) Santiago,
Chile.
Manila Times (2003). It’s final: ‘‘Harurot’’ damage rises
to P2B. http://www.manilatimes.net/national/2003/
jul/26/prov/20030726pro1.html.
McGuigan, C., Reynolds, R., & Wiedmer, D. (2002).
Poverty and climate change: assessing impacts in
developing countries and the initiatives of the interna-
tional community. London School of Economics
Consultancy Project for the Overseas Development
Institute (internal paper).
National Statistics Office (NSO) (2001). The Philippine
Yearbook, census 2000. Sto. Niño Catholic House,
Inc.
Overmars, K. P., & Verburg, P. H. (2005). Analysis of
land use drivers at the watershed- and household
level: linking two paradigms at the Philippine forest
fringe. International Journal of Geographical
Information Science, 19(1), 1–2, Taylor & Francis
Group.
Persoon, G. A., Minter, T., & Visorro, P. (2003).
Indigenous peoples and tropical rainforest manage-
ment: global discourses, local dilemmas. In J. van der
Ploeg, E. C. Bernardo, & A. B. Masipiquena (Eds.),
The Sierra Madre Mountain range: Global relevance,
local realities (pp. 209–240). Tuguegarao City: Gold-
en Press.
Polet, G. (1991), Rattan and bamboo utilisation in a
Sierra Madre community, University of Leiden.
Centre of Environmental Science; Isabela State
University. Cabagan Valley Programme on Environ-
ment and Development (CVPED), Environment and
development student report, nr. 3.
Rutten, R. (2001). Villagers into national revolutionar-
ies? Shifting ‘‘Communities’’ in a period of revolu-
tionary mobilization in the Philippines. Journal of the
Humanities and Social Sciences of Southeast Asia and
Oceania, 157, 629–659.
Scott William H., & C. S. (Eds.) (1975). German
travelers on the Cordillera: (1860–90). Publications
2136 WORLD DEVELOPMENT
of the Filipiniana Book Guild. Manila, The Filipinia
Book Guild.
Tobin, G. A. (1999). Sustainability and community
resilience: The Holy Grail of hazards planning?
Environmental Hazards, 1, 13–25.
Twigg, J. (2001). Sustainable Livelihoods and Vulnera-
bility to disasters, Benfield Greig Hazard Research
Centre Disaster Management Working paper 2.
http://www.benfieldhrc.org/disaster_studies/working_
papers/workingpaper2.pdf.
United Nations Disaster Relief Organization (UNDRO)
(1979). Disaster prevention and mitigation, Vol. 7,
Economic aspects. United Nations, New York, NY.
Van den Top, G. M. (1998). The social dynamics of
deforestation in the Sierra Madre, Philippines. Ph.D.
dissertation, Centre of Environmental Science Leiden
University, P.O. Box 9518, 2300 RA Leiden, The
Netherlands, 437 p.
Van Weerd, M. (2002). The status and conservation of
the Philippine Crocodile Crocodylus mindorensis in
the Northern Sierra Madre, Luzon, the Philippines,
pp. 97–107. In Crocodiles. Proceedings of the 16th
working meeting of the Crocodile Specialist Group.
IUCN – The World Conservation Union, Gland,
Switzerland and Cambridge, UK.
Van der Werf, I. (1994). The administrative context of
an environmental management plan: the impact of
policies, policy actors and their interactions on the
management of natural resources in Cagayan Valley,
the Philippines, University of Leiden. Centre of
Environmental Science; Isabela State University.
Cabagan Valley Programme on Environment and
Development (CVPED), Environment and develop-
ment student report, nr. 38.
Virtual Information Center (2003). Typhoon ‘‘Haru-
rot’’ Hits Philippines – A Special Press Summary.
http://www.vic-info.org/regionstop.nsf/c04a2bbd81cc-
53958a2568390005ec5f/cfb7d96a5296ffff0a256d6c-
0005fdff.
Wisner, B. (2001). Disasters: What the United Nations
and its world can do. United Nations Chronicle,
37(4), 6–9. http://www.un.org/Pubs/chronicle.
Zapata-Marti, R. (1997). Methodological approaches:
the ECLAC methodology. In Center for the Re-
search on the Epidemiology of Disasters (CRED),
Assessment of the economic impact of natural and
man-made disasters. Proceedings of the expert con-
sultation on methodologies, Brussels, 29–30 Septem-
ber, Universite Catholique de Louvain, Belgium,
10–12.