See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/9067931

Constructing Vulnerability: The Historical,

Natural and Social Generation of Flooding in
Metropolitan Manila

Article in Disasters · October 2003

DOI: 10.1111/1467-7717.00230 · Source: PubMed

CITATIONS READS

70 267

1 author:

Greg Bankoff
University of Hull
73 PUBLICATIONS 1,053 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

1. Natural Environment Research Council/Economic and Social Research Council: Earthquakes

without Frontiers: Resilience to Natural Hazards in Earthquake-prone regions, July 2012-June 2017
View project
3. Arts and Humanities Research Council: Risk, Hazards, Disasters and Cultures: Exploring an
Integrated Humanities, Natural Science, and Disaster Studies Approach, June 2016-Jan 2018 View
project

All content following this page was uploaded by Greg Bankoff on 10 May 2014.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
Disasters, 2003, 27(3): 95–109

Constructing Vulnerability: The Historical,

Natural and Social Generation of Flooding in
Metropolitan Manila

Greg Bankoff
Wageningen Univ and Univ of Auckland

Flooding is not a recent hazard in the Philippines but one that has occurred
throughout the recorded history of the archipelago. On the one hand, it is related to a
wider global ecological crisis to do with climatic change and rising sea levels but on
the other hand, it is also the effect of more localised human activities. A whole range
of socio-economic factors such as land use practices, living standards and policy
responses are increasingly influencing the frequency of natural hazards such as floods
and the corresponding occurrence of disasters. In particular, the reason why flooding
has come to pose such a pervasive risk to the residents of metropolitan Manila has its
basis in a complex mix of inter-relating factors that emphasise how the nature of
vulnerability is constructed through the lack of mutuality between environment and
human activity over time. This paper examines three aspects of this flooding: first, the
importance of an historical approach in understanding how hazards are generated;
second, the degree of interplay between environment and society in creating risk; and
third, the manner in which vulnerability is a complex construction.

Keywords: Philippines, vulnerability, flooding, urbanisation, history.

The state of the Philippine environment in the new millennium is not encouraging.
Prey to frequent seismic and climatic hazards, buffeted by the vagaries of 381 years of
colonial misappropriation followed by the much more ruthless exploitation of
international market forces and having a population of around 80 million, the country
suffers from severe environmental strain in part manifested by the increasing
magnitude of floods. On the one hand, this phenomenon is related to a wider global
ecological crisis originating in climatic change, ozone depletion and the rapid loss of
biological diversity, but it is also the effect of more localised human-induced activities.
A whole range of socio-economic factors such as land use practices, living standards
and policy responses are increasingly influencing the frequency of natural hazards such
as floods and the corresponding occurrence of disasters. In particular, the reason why
flooding has come to pose such a chronic risk to the residents of metropolitan Manila
has its basis in a complex mix of inter-relating factors that emphasise how the nature of
vulnerability is constructed through the lack of mutuality between environment and
human activity over time. Rainfall, topography and subsidence combine with
demographic increase, urban growth and the volume of waste products to prevent run-
off and impede drainage. Even the palliative measures taken often only shift the locale

© Overseas Development Institute, 2003.

Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148,
USA.
96 Greg Bankoff

of disaster from one area to another, usually to the disadvantage of the more socially
vulnerable without addressing the root causes of that condition.
Rather than regarding floods as purely physical events requiring largely
technological solutions, they can be viewed primarily as the result of human actions.
Terry Cannon persuasively argues that while hazards are natural, disasters are not.
Social systems generate unequal exposure to risk by making some people more prone
to disaster than others and that these inequalities in risk and opportunity are largely a
function of the power relations operating in every society. Critical to discerning the
nature of flooding in Manila, then, is an appreciation of the ways in which human
systems place people at risk in relation to the environment and to each other, a causal
relationship that is now mainly understood in terms of an individual’s, household’s,
community’s or society’s vulnerability (Cannon, 1994: 14–15, 19). Above all, though,
it is also an historical process. Only through a perspective that pays attention to
changes in topography, demographic growth and urban development over time can the
measure of flooding in the metropolitan area truly be gauged. It is the interplay
between these three — history, nature and society — that determines how the
vulnerability of the city’s inhabitants is constructed.

Climate and history

Any appreciation of the nature of floods in the Philippines first requires an

understanding of the impact of typhoons on the archipelago’s climate. Each year about
20 typhoons cross the Philippine Area of Responsibility (Brown et al., 1991: 196). A
distinction should be made between ‘remarkable’ or destructive typhoons and the more
ordinary variety of tropical cyclones. The former are ‘one of the greatest natural
calamities that may occur in any place’, while the latter are responsible for much of the
rain that makes the climate so ideal for agriculture (Coronas, 1920: 446). While
tropical cyclones can occur in any month of the year, they are much more frequent
between July and November and very rare between January and March. Just as
significant as this seasonality are the routes taken by tropical cyclones and the
consequences for various parts of the islands. At least five main tracks have been
identified: one that crosses to the north of Manila; one that traverses south of the
capital; one that passes east or north-east of the archipelago either disappearing or re-
curving in the Pacific; one that forms in the China Sea to the west of the Philippines;
and another that re-curves in the China Sea between the parallels 10º and 20º. As a
result, some provinces are more frequently exposed to typhoons than others. A
comparative historical analysis of the total number of tropical cyclones experienced by
each major region of the archipelago reveals that northern Luzon receives by far the
highest number but that there is little difference between central Luzon, southern Luzon
and the Visayas, although two islands in the latter, Samar and Leyte, figure among the
10 most exposed provinces. Mindanao, on the other hand, presents a very different
profile with fewer tropical cyclones and a higher percentage of milder events
proportionately than any other region (Census, 1920: 1, 462).
This pattern of tropical cyclones is important to understand both the intensity
and regularity of flooding in the Philippines as they constitute up to 38 per cent of
annual average rainfall in the archipelago, especially during the latter half of the year
(BAS, 2001: 9). Historically, floods are the source of much privation and suffering and
are largely of two types: the sudden raging torrent that peaks sharply and dies away
 Historical, Natural and Social Generation of Flooding in Manila 97

quickly as a result of localised rainfall, and those of a much more widespread nature
and longer duration usually associated with persistent rainfall. The minutes of local
‘town chronicles’ give frequent accounts of such hazards in the past (Bankoff, 2003:
41–51). A list drawn up from these sources found in the Archive of the Manila
Observatory constitutes a record of major floods that occurred between 1691 and 1911.
While almost certainly incomplete, it does provide an indication of the primary causes,
geographical predisposition and even the frequency of such events in specific areas. In
particular, the chronicles regularly refer to flooding in connection to the passage of
tropical cyclones; over 56 per cent of all recorded incidences are directly attributed to
typhoons (AMO, Box 10-37). On other occasions, floods are mainly attributed to
heavy rainfall and the monsoons. Moreover, the close association between flooding
and typhoons suggests a certain seasonality in the their occurrence that corresponds to
the peak in the latter’s annual cycle between July and November.
The geographical predisposition to flooding is even more tenuous to gauge as
the records largely reflect the principal centres of Spanish colonial power. Still, the
preponderance of the northern part of Luzon over the rest of that island is consistent
with the higher incidence of typhoons there. Some idea of what the incidence of floods
on local communities might mean can be estimated by closer scrutiny of the more
complete local chronicles that suggest how often people were faced with such
situations. The records for Nabua in Camarines between 1691 and 1856, and those for
Pangasinan 1768–1872 depict just how frequent a life event floods were. In Nabua, a
person experienced one such event every 9.7 years on average but once every 5.6 years
between 1733 and 1800 (AMO, Box 10-37). Other local histories present a glimpse of
the reality of this hazard for communities. The rising of the Abra River to a height of
more than 25 metres above its normal level caused over 1,800 deaths on 25–7
September 1867 or, again, the flood in Santa Maria, Ilocos Sur destroyed the barrio of
Sumagui, carrying away more than 22 houses and causing more than 100,000 pesos
worth of damages in 1911 (AMO, Box 9-35).
Nor has the number or severity of floods lessened in recent times. Heavy
rainfall led to the submergence of low-lying areas in Pampanga and Bulacan in July
1972, August 1974 and October 1978. Continuous rain killed over 200 people and
damaged 200,000 hectares of agricultural land in Mindanao in January 1981 (Brown et
al., 1991: 198–9). There was severe flooding during the summer of 1986 and again on
6 September 1995 when the sides of Lake Maughan burst, killing 57 and affecting
more than 60,000 people in South Cotabato (NDCC, 1995).1 But the flash flood that
overtook Ormoc on 5 November 1991 was exceptional even by Philippine standards.
Torrential rainfall at the height of a weak tropical cyclone dumped nearly 150
millimetres on to the denuded mountain slopes above the small eastern Visayan city in
a short period. Water levels rose to over six metres in a matter of minutes, while an
unusually high tide prevented the flood from running straight out to sea. The whole
episode was over in less than 15 minutes but the flood destroyed between 80–90 per
cent of buildings, left 5,365 people dead and a further 2,046 missing (Vitug, 1993: 8;
Laurel, 1991: 7).
Moreover, statistical trends suggest that floods have become more numerous
and more devastating in recent years (see Figure 1). Certainly, the frequency of events
and the number of people affected have increased steadily as human-related activities
such as deforestation, overgrazing and urbanisation aggravate environmental
conditions, making communities more vulnerable. The toll, too, has been rising.
Floods have been responsible for 1,353 deaths, 423 injuries and 433 missing since
98 Greg Bankoff

60 3,500,000

50 3,000,000

2,500,000
40

Persons affected
2,000,000
Events

30
1,500,000
20
1,000,000

10 500,000

0 0
1973 1978 1983 1988 1993 1998
Years

Events Persons

Figure 1 Number of people affected by floods, 1973–2001

Source: Disaster Data 1973 to 1999 Floodings/Flashfloods (NDCC, 1999) and Disasters in the
Philippines 2001 (CDRC, 2001a: 2, 10).

1973. No fewer than 1,563,342 families comprising 8,021,739 people have been
affected, 5,133 houses destroyed and a further 14,984 damaged and property and
agriculture losses estimated at P7.575 billion ($303 million) (NDCC, 1999). Floods
were the sixth-most-common disaster in 2000, affecting 227,306 families or 1,115,344
individuals (CDRC, 2001b: 1–2).

Flooding in Metro Manila

Metro Manila is situated in a semi-alluvial floodplain formed by sediment flow from

the Meycauayan and Malabon-Tullahan river basins in the north and the Marikina river
basin in the east. It encompasses a land area of 636 square kilometres, measuring about
20 kilometres in length along a north-south axis and stretching more than 22 kilometres
at its widest. The conurbation is open to Manila Bay on the west and to Laguna de
Bay, a large lake, on the south-east. As such, the metropolitan area now constitutes a
vast urbanised drainage basin that experiences frequent inundations from overflowing
rivers and storm waters that render the existing system of esteros (modified natural
channels) and canals constructed during the Spanish and American colonial periods
inadequate (Liongson, 2000). Despite the growing vulnerability of much of the
metropolitan area, however, rapid urbanisation has continued unabated with residential
homes, industries and commercial sites increasingly exposed to flood-related
destruction. The administratively designated National Capital Region (NCR) now
comprises 12 cities and five towns.
Leonardo Liongson and Peter Castro classify flooding in Manila into three
types: local, moderate and regional. Local street flooding is the result of intense,
sudden thunderstorms over a few city blocks that cause inundations of 20–50
centimetres, light property damage and heavy traffic congestion. Typically flooding
 Historical, Natural and Social Generation of Flooding in Manila 99

______________________________________________________________________

Pateros
M'tinlupa
Pasig
Taguig
City & municipalities

Malabon
Paranaque
Marikina
M'aluyong
Valenzuel
Makati
Las Pinas
Manila
Quezon
Navotas
San Juan
Caloocan
Pasay
Metro
0 10 20 30 40 50 60 70 80 90 100

Per cent flooded

Figure 2 Per cent of Barangays in Metro Manila by city affected by flood in

2000
Source: Hazard Prone Areas in Metro Manila (OCD, 2000a).

happens because curb inlets, drainage culverts or natural upland ditches are insufficient
to handle the sudden volume of water. Such events occur all over the metropolitan area
but are more frequent in low-lying areas. Moderate flooding is produced by intense
rainfall of over an hour’s duration that is often associated with tropical cyclones and
affects a wide area of the city with inundations of more than 30 centimetres. Damage
to property can be considerable; there may be isolated cases of personal injury or even
loss of life; and whole districts become impassable to vehicle movement. Again,
moderate flooding mainly occurs in low-lying areas of the capital. Regional flooding,
on the other hand, affects whole cities and may cover several river basins. It is a large-
scale condition consequent upon typhoons that ensues from heavy rainfall lasting over
several days and may be aggravated by high tides or storm surges. All gainful
activities are severely curtailed until the waters subside (Tabios et al., 2000: 19–20).
While flooding variously affects all areas of Manila, some cities and
municipalities are more vulnerable than others due to their location and height relative
to sea level. Some 20 per cent of the capital’s 63,600 square hectares is designated as
flood prone, of which 5,385 square hectares (41 per cent) are served by pumping
stations and the remaining 7,715 square hectares (59 per cent) suffer frequent and long-
lasting inundation (MMDA, n.d./a). Areas to the east, south-east and south of the
capital around Marikina, Pasig, Mandaluyong, Muntinlupa and Parañaque and
especially those adjacent to Laguna de Bay such as Taguig and Pateros are acutely
susceptible to flooding (see Figure 2). In some particularly exposed cities such as
Muntinlupa and Taguig, all barangays (the basic unit of local government) are
regularly inundated and the coverage of flooding extends to 88 per cent and 83 per cent
of their respective land areas (OCD, 2000a). Flooding affected 162 of Manila City’s
897 barangays in November 1998 causing hardship to an estimated 61,104 families or
approximately 305,520 people, fully 18 per cent of the total population (OCD, 1998).
100 Greg Bankoff

Unsurprisingly, there is often a positive correlation between flood-prone areas and

those cities and municipalities that have larger proportions of low-income residents.
Flooding has been a feature of daily life in Manila since at least the 19th
century. When rains were particularly strong or tides high, the natural drainage system
was unable to cope with the sudden volume of water and the city was turned into one
vast lake that its inhabitants had to traverse by canoe. The gradual expansion of the
urban area into the surrounding marshes only increased the frequency and magnitude of
such inundations: workers in the booming cigar-making factories of Arroceros were
distinguished by their high-heeled sandals worn to gain access safely to their places of
work (Huetz de Lemps, 2001: 495). More recently, rapid population growth,
differences in land usage and environmental factors have combined to magnify the
nature and extent of flooding in the metropolitan area. In the 1950s, areas below the
flood line of 12.5 metres above sea level in Manila (Tondo, Sampaloc and Santa Mesa)
as well as those low-lying barangays in Quezon City, San Juan and Mandaluyong were
regularly inundated. Flooding was mainly attributed to natural causes such as flat
terrain, rainfall intensity and high tides in Manila Bay. However, even by the late
1950s, activities associated with modernisation such as the spread of urbanisation in the
hills to the east of Manila and the encroachment of structures along the banks of the
Pasig River led to severe reductions in the width and depth of tidal channels. By the
1960s, almost 70 per cent of the city was subject to floods that ranged in depth from 3.6
to 4.5 metres. The spread of congested neighbourhoods, especially along the banks of
rivers and creeks, proved to be especially susceptible to floodwaters that regularly
swept away the flimsy housing of the urban poor. Floodwaters reportedly reached a
height of 6.1 metres in the squatter colony of Tatalon on one occasion and even lapped
across the floors of Malacañang, the presidential residence, on another. The severity of
flooding was undoubtedly aggravated by the fast pace of urban expansion, in particular
the rise in the number of paved roads and the growth in building complexes that
enlarged impervious surface areas and increased the incidence of flash floods. The
massive deforestation of the Marikina and Montalban watersheds and the further
encroachment and even disappearance of esteros (an estimated loss of 21 kilometres)
severely restricted the dissipation of floodwaters (Zoleta-Nantes, 2000: 43–4).
By the 1970s, the degree of soil erosion turned floodwaters brown and murky
while silt and garbage clogged remaining waterways, further restricting the flow of
water. Flooding was perennial in Malabon and Navotas and simply a fact of life to
many living in Manila, Quezon City, Pasay and San Juan. Heavy floods left thousands
of families homeless and severely disrupted power, communications and transport,
forcing many people to have recourse to rafts and small boats as a mode of urban
conveyance. Silt deposits reduced the water-holding capacity of Laguna de Bay by 64
per cent while surface run-off from denuded watersheds increased its level by 2.7
metres during the 1980s with dire consequences for low-lying communities along its
shores. Matters were only aggravated by the development of new sub-divisions on
former agricultural lands in Marikina, Cainta, Pasig, Pateros and Taguig. Floodwater
depths continued to rise in the 1990s, especially in the south and south-east and in the
northern cities of Kalookan, Malabon, Navotas and Valenzuela. Hundreds of
thousands of families regularly required evacuation to higher ground and the
subsequent traffic congestion following a torrential downpour turned commuting into a
nightmare, trapping some inside their vehicles for up to 16 hours. Floods were usually
accompanied by outbreaks of infectious diseases such as cholera and leptospirosis,
power outages that paralysed industrial and manufacturing operations and property
 Historical, Natural and Social Generation of Flooding in Manila 101

losses that were estimated at P900 million ($36 million) over the period (Zoleta-
Nantes, 2000: 45–8).
The first widely recorded instance of serious flooding dates to 1942 when
unprecedented high water levels were reported that inundated the city for several days
(Fano, 2000: 53). Major floods also occurred in 1948, 1966, 1967, 1970, 1972, 1977,
1986 and 1988 as water overflowed the main rivers and canals. The flooding caused
by Typhoon Miding in 1986 inflicted the most serious damage in recent years with
floodwaters extending to 103.6 square kilometres or over 16 per cent of the total area of
Metro Manila. Flooding in 1988 caused by Typhoon Unsang also inflicted serious but
more localised damage in the Marikina river basin and along the low-lying shoreline of
Laguna Lake (MMDA, n.d./b). Nor have matters improved much of late: thousands of
Manila residents were stranded on the streets or trapped in vehicles all night after heavy
rains on 28 July 1995 and major flooding incidents happened again on 28 May 1996
and 18 August 1997 (Page, 2000: 86). Over the past half-century, floods have become
both more extensive and more severe. Areas regularly inundated have spread from the
low-lying coastal areas to encompass suburban neighbourhoods, newer urban
developments and along the shores of Laguna de Bay. Nor have the expensive
subdivisions built on former agricultural lands been spared and flooding has even
become prevalent in locations with higher elevations (Zoleta-Nantes, 2000: 49).
Moreover, water depths have also steadily risen over time making flooding a major
hazard to residents and prompting a recent newspaper article to quip that: ‘Though the
deluge mentioned in the Bible may be argued as allegorical, the flooding that occurs in
Metro Manila streets after a downpour is definitely not’ (Afuang, 2001: B4).

Constructing vulnerability

Topography predisposes Metro Manila to flooding. Climate makes the city still more
vulnerable: average annual precipitation over the area between 1961 and 1990 varied
from 1,834 to 2,257 millimetres, with rainfalls of 200 millimetres or more over a two-
day period not uncommon and substantially higher rates recorded in the event of
typhoons. The volume of run-off is greatly intensified by the widespread use of
bitumen and concrete associated with urbanisation. In many low-lying coastal or
riverine areas, flooding is also associated both with tidal variations that can inundate
lands above mean sea levels and with poor drainage of alluvial soils in the Marikina
valley (Zoleta-Nantes, 2000: 35–8). Over recent decades, moreover, the extent of
flooding has been considerably aggravated by the subsidence of the land fringing
Manila Bay. The sediments that underlie river deltas have a high water content that is
subject to ‘squeezing’ by the weight of succeeding deposits, a process that is greatly
accelerated when groundwater is extracted faster than it can be replenished by natural
recharge from rain seeping back into the ground. Records taken at Manila’s South
Harbour show that mean sea levels rose about two millimetres a year between 1902 and
the early 1960s but accelerated sharply thereafter, reaching a rate of approximately
three centimetres by 1991. Such an increase cannot be explained as solely a
consequence of global warming and bears a marked correlation to the rise in both
groundwater extraction and population growth.2 As the land around Manila Bay sinks
and the level of the sea rises, flooding has become more prevalent not only in Metro
Manila but in the surrounding provinces (Siringan and Rodolfo, 2002: 4; Rodolfo,
2003).
102 Greg Bankoff

These climatic and environmental factors that themselves are evolving over
time partly as a result of natural processes and partly from human-induced ones also
inter-relate in complex ways with the changes brought about as a result of human
activity to make Metro Manila more flood prone. Principal among these latter
developments is the sheer rise in the size and density of population. Since 1903, the
number of people living in the metropolitan area has risen from 328,939 inhabitants to
10,491,000 in 2000, an increase of 3,189 per cent. Population density in the region
also rose accordingly from 517 people per square kilometre in 1903 to 3,872 in 1960
and 16,495 by 2000 (see Figure 3). These averages, however, conceal an enormous
variation between cities whose densities range from a low of 10,717 people per square
kilometres in some areas of Quezon City to 50,042 persons in more congested areas
(Zoleta-Nantes, 2000: 40).
The sheer weight of human numbers puts considerable pressure on resources
that, in turn, has substantial consequences on the environment and intensifies both the
severity and duration of floods. Much of Metro Manila’s population increase has come
about as a result of massive urban migration since the Second World War as rural folk
are attracted to the city by higher incomes and greater livelihood opportunities.
Approximately 300 families have migrated to the NCR each month since the 1980s
(Zoleta-Nantes, 2000: 40). The level of urbanisation in the Philippines, that is the
proportion of urban population to the total, rose from 21.7 per cent in 1950 to 47.6 per

12,000,000 18,000

16,000

10,000,000

14,000

8,000,000 12,000 Population density (sq km)

Total population

10,000

6,000,000

8,000

4,000,000 6,000

4,000

2,000,000

2,000

0 0
1903 1939 1960 1980 2000

Years

Population Density

Figure 3 Population data for Metro Manila, 1903–2000

Sources: Zoleta-Nantes, 2000: 41; OCD, 2000b.
 Historical, Natural and Social Generation of Flooding in Manila 103

cent in 1997 (Rebullida, 1999: 16; Ibon, 2000: 66). Most of these people were poor
and could not afford the high costs of land, housing materials and construction. Land
values throughout the country increased 12 to 15 times between 1940 and 1969 but 27
times in the NCR. In the 1980s, land prices rose 35 to 40 times in Quezon City, 50 to
80 times in Makati, 250 to 400 times in Diliman and a staggering 2,000 times in
Escolta. In 1996, the CBD was registering an annual increase of 50 per cent and even
the value of land in peripheral areas rose by 25 per cent annually (Rebullida, 1999: 16–
17). As a consequence, migrants have generally had to find accommodation in the
informal housing sector becoming interchangeably squatters (illegal occupiers of land),
slum dwellers (residing in blighted urban communities) and makeshift dwellers (living
in shelters made of scrap materials). Such neighbourhoods are often situated on the
urban fringes or wastelands that proliferate in Third World cities, especially near to
areas that provide work and along major transport hubs and links.
In Metro Manila, the banks of rivers, canals and esteros have frequently
served in this capacity. Since the demand for land is at such a premium, spaces that are
vacant or only nominally owned by national, city or municipal authorities prove
particularly attractive as locales for squatters. The result is that makeshift housing
often encroaches on to available waterways, blocking the access of maintenance
personnel and equipment from the Department of Public Works and Highways
(DPWH) or the Metropolitan Waterworks and Sewerage System (MWSS) and, by a
gradual process of accretion, narrows their flow capacity and diminishes the volume of
discharge they are able to handle (Labrador and Bualat, 1996–7: 44).3 A recent report
submitted by the Office of the City Engineer details the specific histories of how such
processes actually occur. How the Estero de Tanque in Paco was partly filled in and is
now occupied by adjacent lot owners and squatters; how the Estero Tripa de Gallina in
Santa Ana was subdivided and disposed of by the Land Authority to tenants during the
administration of President Macapagal; and how a portion equivalent to 1,788.30
square metres of the Estero de Aviles in San Miguel was titled to a certain Arsenio Dy
of 3011 Nagtahan, Sampaloc as TCT Nos. 67,425 and 67,426. Nor are squatters and
the urban poor the only ones responsible for these encroachments. Thus the Estero de
Maytubig in Malate that has been dried up since 1925 is currently occupied by the
Agno-Leveriza government subdivision, Bank Plaza and the Manila Zoo among others,
or the Estero de Concordia in Paco that is currently included in the private land title of
the Manila Gas Corporation (Aboy, n.d.). The extent of the esteros problem is
indicated by the number of what are termed informal settler families (ISFs) that were
deemed by the DPWH to be in urgent need of relocation before the onset of the rainy
season in 2001. Priority areas included 279 families living near or on the Estero de
Valanecia, 336 on the Estero de San Miguel, 200 on the Estero de Aviles, 151 on the
Estero de Santibañez, 858 on the Estero de Magdalena and 650 on the Estero de San
Lazaro totalling 2,474 families and costing an estimated P208 million to relocate
(DPWH, 2001). Recent estimates of ISFs living along the banks of the metropolis’s
waterways are 27,300 families — or some 164,000 people (Fano, 2000: 59).
Unfortunately, the construction of informal housing is not the only
environmental problem related to the encroachment and gradual infilling of esteros.
Residential communities are primarily responsible for generating about half the total
volume of the metropolitan area’s solid wastes. Metro Manila inhabitants disposed of
6,050 tons of garbage daily in 1995 with an annual increase estimated at 2 per cent and
were generating on average 0.71 kilograms of waste per person per day by 2000 (Ibon,
2001: 4–5). Only 71 per cent of this rubbish is collected by trucks and taken to landfill
centres. The remaining 1,750 or so tons are simply left on street corners, dumped on
104 Greg Bankoff

vacant lots or thrown into storm drains, canals, creeks or rivers. Anything from rubber
tires to dead animals are disposed of in this manner. A study completed in 1995 found
that from 0.9 to 2.6 per cent of people in the NCR, according to age, disposed of their
trash in the esteros, the under-30s being far less environmentally conscious than the
over-60s (Perez et al., 1995). Altogether, this represents an amount of from 55 to 157
tons of solid wastes each day that clog the network of drainage canals, posing a
considerable risk to health and greatly increasing the likelihood of flooding. That such
a quantity of waste represents a serious threat to flood-control measures is borne out by
the quantity of garbage recorded at the metropolitan area’s various pumping stations
and related facilities. Approximately 15,500 cubic metres of rubbish accumulates
annually around the capital’s 16 pumping stations and two trash screens, although the
volume at any one time can vary enormously from the average 29 cubic metres at
Makati to the 4,217 cubic metres at Tripa de Gallina. Although the total amount of
accumulated garbage actually fell in 2000, more worrying is the trend that suggests a
steady decline in the percentage of garbage that is collected with corresponding risks to
the maintenance and efficiency of these stations and their pumps (see Figure 4).
The building of pumping stations to facilitate the dispersal of floodwaters
represents only one of the ways in which city and municipal authorities have attempted
to deal with the problem of chronic flooding within the metropolitan area. As early as
1882, a plan was drawn up for the refurbishment of the drainage system based on the
existing network of esteros. Regarded as the only practical scheme given the financial
constraints of the time, the project had only been partially realised by the eve of the
1896–98 Philippine Revolution (Huetz de Lemps, 2001: 502–3). A sewerage system
was had been constructed under the American administration between 1904 and 1911
but was only designed to serve an urban population envisaged at half a million people

18,000
16,000
Garbage (cubic metres)

14,000
12,000
10,000
8,000
6,000
4,000
2,000
0
1998 1999 2000
Collected 15,170 12,184 11,040
Uncollected 1,455 4,786 2,156
Metro 16,625 16,970 13,196

Collected Uncollected Metro

Figure 4 Accumulated garbage recorded at NCR pumping stations, 1998-

2000
Source: MMDA, n.d./c.
 Historical, Natural and Social Generation of Flooding in Manila 105

(Zoleta-Nantes, 2000: 40). Currently responsibility for flood control in Metro Manila
depends on the scale and extent of the problem. Localised flooding associated with
drainage basins of a few square kilometres are handled by city or municipal officials
usually through the construction of gutters and storm drains that direct rain into
existing natural or artificial waterways. Major flood-control problems involving issues
that affect the entire metropolitan region and nearby watersheds were made the
responsibility of the Department of Public Works and Highways in 1987 and the Metro
Manila Development Authority since 2002.4
After the widespread flooding of 1972, a major flood-mitigation programme
was undertaken, financed by the Overseas Economic Cooperation Fund it resulted in
the raising of the Pasig River walls and the erection of the first seven pumping stations,
two floodgates and four drainage mains between 1974 and 1978. In 1980, excavation
of a 10-kilometre long diversion channel known as the Mangahan Floodway Project
(MFP) was begun linking the Marikina River at Pasig (Barangay Mangahan) to Laguna
de Bay at Taytay, Rizal. Using the lake as a temporary catchment basin in times of
intense rainfall over the metropolitan area to divert 70 per cent of the flow coming
down the Marikina River and reduce the likelihood and extent of inundation along the
Pasig River, the channel also acts as a reverse flow at times when the water level in
Laguna de Bay is higher than that of the river. The Marikina River is one of 13
tributaries that discharge into the lake but there is only a single outlet: the Napindan-
Pasig River that carries water from the lake to the sea at Manila Bay. The flow of
water out of Laguna de Bay has been subsequently regulated by the construction of the
Napindan Hydraulic Control Structure — locks that open to control the reverse
overflow into the lake through the reinforced channel of the Napindan River to
accelerate the lowering of its level and alleviate flooding along its shoreline (MMDA,
n.d./d). A major dyke is now envisaged to protect the neighbourhoods adjacent to the
lakeside running from General Santos Avenue in the south to the Mangahan Floodway
in the east. The core of the DPWH flood mitigation works, however, remains its
network of drains, canals, esteros and rivers that discharge into Manila Bay. This
system consists of 44 kilometres of main drains, 1,200 kilometres of lateral drains, 290
kilometres of canals and esteros and 153 kilometres of rivers and major streams (Fano,
2000: 66).
Problems, however, arise with this purely technological solution to the
question of flooding in Metro Manila. In the first place, the cleaning and rehabilitation
of existing flood-control structures and facilities is both costly and difficult. Some
P152 million is spent annually to cover such services and a full-time staff of 446
maintenance crew employed. The impossibility of keeping the esteros clear near major
market places and sites heavily occupied by squatters has led to the innovative
recruitment of local Bantay estero teams, composed of five to seven labourers, whose
sole responsibility is to maintain a specified section of waterway free of garbage
(Fano, 2000: 67; MMDA, n.d./e). Nor, of course, can the question of ISFs be simply
regarded as a matter of flood mitigation. The eviction and demolition of such
communities has serious socio-economic implications and important political
ramifications that few elective office holders choose to ignore. In many cases, drainage
or maintenance work is suspended because there are informal dwellings. A recent
MMDA document cites three examples of flood-control projects not completed for this
reason: the dredging of the Estero de Magdalena only 40 per cent accomplished and
then suspended on presidential instructions to defer demolition and relocation of
informal settlers; the cyclone wire fencing of the Vitas pumping station only 35 per
cent accomplished and then suspended because of the presence of informal settlers; and
106 Greg Bankoff

the steel sheet pile driving as part of the improvement works on the Estero de Sunog
Apog only 54 per cent accomplished and then suspended for the same reason (MMDA,
n.d./f). While squatters and ISFs provide useful media scapegoats and are largely
blamed for encroachment on the existing waterways system and of impeding cleaning
operations, industry, business, government and fish-pen operations are also major
contributory agents and just as culpable (Page, 2000: 87–8; Rodolfo, 2003). Moreover,
the prevailing assumption that the problem of flooding in Metro Manila necessitates
purely technological solutions is often itself a factor in making some sectors of the
population more vulnerable to hazard as in the case of the Mangahan Floodway
Project: the MFP’s diversion of the overflow from the Pasig River only exacerbates the
rising water levels in Laguna de Bay and contributes to the frequency and severity of
flooding among shoreline municipalities such as those in Taguig. Nor is it completely
incidental that these neighbourhoods also house a large percentage of the
metropolitan’s urban poor, many of whom are squatters without legal title to the land
their dwellings are sited on.

Conclusion

There are three aspects of flooding in Metro Manila that this study illustrates: first, the
importance of an historical approach in understanding how hazard is generated; second,
the degree of interplay between environment and society in the creation of risk; and
last, the manner in which vulnerability is a complex construction. A fuller appreciation
of the nature of vulnerability is still often hampered by the lack of an adequate
historical perspective from which to understand the contexts and roots of disaster
causality (Oliver-Smith, 1986: 18; Lees and Bates, 1984: 146). It is not simply the
occurrence, frequency and intensity of environmental events that are significant but
their sequence that is of critical importance (Winterhaler, 1980). The relationship
between climate, topography, resource use and culture over time is the basis to
determining the nature of flooding in Metro Manila as it is to any other location under
review. History, far from being a scholarly topic with little practical application,
actually provides both the framework as well as sets the conditions within which these
other factors relate. Without proper consideration of the temporal dimension, hazards
remain random, disasters unaccountable and societies simply exposed. Certain
communities or segments of populations are often situated in more perilous settings
than others due to the consequences of political, economic and social forces that change
over time and in relation to the landscape and its use and to the nature of hazards that
also have their own cycles and historical evolutions. Just as history reveals that
vulnerability may take centuries in the making, so is the sense of its complex
construction important to the way in which disaster is increasingly conceptualised. The
notion that hazards are not merely physical phenomena but that human agency is also a
contributing factor to their occurrence has led to a considerable extension in
understanding how disasters come about and how the root causes of vulnerability lie in
a variety of relational exchanges. It is the dynamics between stakeholders (human
agency and animal behaviour), ecosystem (the specifics of the environment) and nature
(extreme physical phenomena) that determine the increasing complexity of these
events.
Flooding is not a recent hazard in the Philippines but one that has occurred
throughout the recorded history of the archipelago: it is the result of the low-lying
 Historical, Natural and Social Generation of Flooding in Manila 107

nature of much of the terrain and the frequency of typhoons. These typhoons do not
necessarily constitute hazards as such and are, in fact, responsible for a significant
percentage of the annual rainfall that makes the islands so fertile and thus ideal for
agriculture. This fertility, in turn, encourages or at least permits demographic growth
and the location of people in large cities such as Metro Manila. The demands of this
steadily expanding urban population for basic amenities such as water, together with
the nature of modern development, generate environmental problems like accelerated
subsidence and garbage disposal that, in conjunction with the torrential rainfall
associated with tropical storms, aggravates the incidence and severity of flooding in the
metropolitan area. The concentration of employment, educational and health among
other opportunities in the ‘big city’ only spurs more rural-to-urban migration, creates
shortages of suitable residential land and encourages the occupation of areas more
prone to flooding or that have important drainage functions. The actions of
governments and the technological solutions they mainly favour have only limited
outcomes and may actually aggravate conditions usually to the disadvantage of the
most vulnerable poor, whether urban or rural. The construction of vulnerability to
flood in Metro Manila, therefore, clearly shows how societies and destructive agents
are very much mutually constituted and embedded in natural and social systems as
unfolding processes over time.

Notes
1. Kelvin Rodolfo claims that the crater was deliberately breached with high explosives by a
cartel of miners, local government and military personnel seeking treasure in the lake and
that the number of casualties was much higher (2003).
2. The withdrawal of groundwater greatly accelerates the compaction of delta sediment. When
water is withdrawn faster than it can be replenished, grains of sand and gravel that constitute
an aquifer begin to draw moisture from the surrounding layers of clay. As both layers begin
to settle and lose volume, the land surface above them subsides.
3. Shanty dwellers must also rely on shallow wells for their water needs and so further con-
tribute to the problem of subsidence and flooding in yet another way.
4. A Metro Manila Flood Control and Drainage Council was establishment by PD 18 on 7
October 1972 that among other powers was authorised to collect an Adjustment Tax of P0.25
levied on all film tickets.

References
Aboy, R. (n.d.) Report Submitted by Mr Ricardo C. Aboy, Engineer V. Office of the City
Engineer. Manila.
Afuang, B. (2001) Floods and the City. Philippine Star 15(290): B4.
AMO (n.d.) List of Typhoons. Archive of the Manila Observatory, Box 9-35.
AMO (n.d.) Floods in the Philippines, 1691–1911. Archive of the Manila Observatory, Box 10-
37.
Bankoff, G. (2003) Cultures of Disaster; Society and Natural Hazard in the Philippines.
Routledge Curzon, London.
BAS (2001) Documentation on the Impacts of and Responses to Extreme Climate Events: Food
and Agricultural Sector. Extreme Climate Events Technical Working Group for Agriculture
(chaired by Dr Romeo Recide), Bureau of Agricultural Statistics, Quezon City.
Brown N., L. Amadore and E. Torrente (1991) Philippine Country Study. In Disaster
Mitigation in Asia and the Pacific. Asian Development Bank, Manila.
108 Greg Bankoff

Cannon, T. (1994) Vulnerability Analysis and the Explanation of ‘Natural’ Disasters. In A.

Varley (ed.) Disasters, Development and Environment. John Wiley & Sons, Chichester.
CDRC (2001a) Disasters in the Philippines 2001; Drenchers and Blazers. Citizens’ Disaster
Response Center, Quezon City.
—— (2001b) Disasters in the Philippines 2000. Citizens’ Disaster Response Center, Quezon
City.
Census (1920) Census of the Philippine Islands: 1918. Bureau of Printing, Manila.
Coronas, J. (1920) The Climate and Weather of the Philippines, 1903–1918. In Census of the
Philippine Islands: 1918. Bureau of Printing, Manila.
DPWH (2001) CCPS Sub-Cluster on Relocation of Informal Settlers As of 27 June 2001.
Department of Public Works and Highways, Manila.
Fano, N. (2000) Flood Control and Drainage Network/Operations/Plans in Metro Manila. In L.
Liongson, G. Tabios and P. Castro (eds.) Pressure of Urbanization: Flood Control and
Drainage in Metro Manila. UP-CIDS, Diliman: 53-74.
Huetz de Lemps, X. (2001) Waters in Nineteenth Century Manila. Philippine Studies 49(4):
488–517.
Ibon (2000) The State of the Philippine Environment. Ibon, Manila.
—— (2001) Ecological Solid Waste Management Act of 2000 (RA 9003). Ibon 24(20): 31
October.
Labrador, L. and B. Bualat (1996–7) Privatization Industrial Relations and Human Resource of
the Metropolitan Waterworks and Sewerage System. Philippine Journal of Labor and
Industrial Relations 17(1, 2): 44–61.
Laurel, R. (1991) The Great Flood of Ormoc. National Midweek 27 November: 7–11.
Lees, S. and D. Bates (1984) Environmental Events and the Ecology of Cumulative Change. In
E. Moran (ed.) The Ecosystem Concept in Anthropology. Westview Press, Boulder.
Liongson, L. (2000) The Esteros of Manila: Urban Drainage A Century Since. In L. Liongson,
G. Tabios and P. Castro (eds.) Pressure of Urbanization: Flood Control and Drainage in
Metro Manila. UP-CIDS, Diliman.
MMDA (n.d./a) Briefing Materials Flood Control III (Metro-Manila Flood Control Plans).
Metro Manila Development Authority, Pasig City.
MMDA (n.d./b) DPWH Develops NCR Flood Control Plan. Philippine Infrastructure Journal,
paper obtained from the Metro Manila Development Authority, Pasig City: 18–24.
MMDA (n.d./c) Accumulated Garbage at the Various NCR-DPWH Pumping Stations and
Related Facilities in Metro-Manila From CY1998 to CY 2000. Metro Manila Development
Authority, Pasig City.
MMDA (n.d./d) The Mangahan Floodway Project. Metro Manila Development Authority, Pasig
City.
MMDA (n.d./e) No. of Manpower Utilized by the Regional/District Offices All Year Round For
the Maintenance/Cleaning of Drainage Systems in Metro Manila. Metro Manila
Development Authority, Pasig City.
MMDA (n.d./f) Issues and Concerns of Flood Control and Drainage Division. Metro Manila
Development Authority, Pasig City.
NDCC (1995) 1995 Disaster Incidents. Republic of the Philippines, National Disaster
Coordinating Council, Office of Civil Defense, Camp General Emilio Aguinaldo, Quezon
City.
NDCC (1999) Disaster Data 1973 to 1999 Floodings/Flashfloods. Republic of the Philippines,
National Disaster Coordinating Council, Office of Civil Defense, Camp General Emilio
Aguinaldo, Quezon City.
OCD (1998) Manila City Flood Prone Areas. Office of Civil Defense, Camp General Emilio
Aguinaldo, Quezon City.
OCD (2000a) Hazard Prone Areas in Metro Manila. Office of Civil Defense, Camp General
Emilio Aguinaldo, Quezon City.
OCD (2000b) Total Population, Population Growth Rate, Population Density Philippines and
National Capital Region. Office of Civil Defense, Camp General Emilio Aguinaldo, Quezon
City.
 Historical, Natural and Social Generation of Flooding in Manila 109

Oliver-Smith, A. (1986) Disaster Context and Causation: An Overview of Changing Perspectives

in Disaster Research. In A. Oliver-Smith (ed.) Natural Disasters and Cultural Responses.
College of William and Mary, Williamsburg.
Page, J. (2000) Metro Manila Flooding: The Sociocultural Dimension. In L. Liongson, G.
Tabios and P. Castro (eds.) Pressure of Urbanization: Flood Control and Drainage in Metro
Manila. UP-CIDS, Diliman.
Perez, A., M. Baltazar, M. Esquivias and F. Arguillas (1995) Waste Management Differentials in
Filipino Households. National Statistics Office, Manila.
Rebullida, L. (2000) Resources Recovery in Solid Waste Management: Strategies, Initiatives,
Polict Issues. UP Center for Integrative and Development Studies, Diliman.
Rodolfo, K. (2003) Personal communication to author.
Siringan P. and K. Rodolfo (2002) Subsidence and Flooding of the Delta Plain North of Manila
Bay: Causes, Trends and Possible Solutions. Paper presented at Convention Hall, Bureau of
Soils and Water Management, 21 October, Quezon City.
Tabios, G., L. Liongson and P. Castro (2000) Flooding Issues and Concerns in Metro Manila: An
Overview. In L. Liongson, G. Tabios and P. Castro (eds.) Pressure of Urbanization: Flood
Control and Drainage in Metro Manila. UP-CIDS, Diliman.
Vitug, M. (1993) The Politics of Logging; Power from the Forest. Philippine Center for
Investigative Journalism, Manila.
Winterhaler, B. (1980) Environmental Analysis in Human Evolution and Adaptation Research.
Human Ecology 8: 137–70.
Zoleta-Nantes, D. (2000) Flood Landscapes of Metro Manila. In L. Liongson, G. Tabios and P.
Castro (eds.) Pressure of Urbanization: Flood Control and Drainage in Metro Manila. UP-
CIDS, Diliman: 35-49.

Address for correspondence: Wageningen University, 1 Hollandsweg, 6700 EW

Wageningen, Netherlands and School of Asian Studies, University of Auckland,
Private Bag 92019, Auckland, New Zealand. E-mail: << g.bankoff@auckland.ac.nz or
gregory.bankoff@wur.nl>>

View publication stats