33822 NATURAL DISASTER RISK MANAGEMENT IN THE PHILIPPINES: ENHANCING POVERTY ALLEVIATION THROUGH DISASTER REDUCTION THE WORLD BANK NATIONAL DISASTER COORDINATING COUNCIL EAST ASIA AND PACIFIC REGION REPUBLIC OF THE PHILIPPINES RURAL DEVELOPMENT Table of Contents Acknowledgements ii Acronyms iii Executive Summary v Chapter 1. Introduction 1 Chapter 2. Hazard Exposure and Disaster Impacts in the Philippines 3 Chapter 3. Disaster Management Capacity in the Philippines 24 Chapter 4. Sharing the Costs of Disasters 39 Chapter 5. The Way Forward: Summary of Findings and Recommendations 44 References 58 List of interviewees 60 Figures ii Acknowledgements This study would not have been possible without the funding support provided by the Regional VPU. The study is part of a broader study that is looking at " Comprehensive Disaster Risk Management For East Asia And The Pacific Region" , also funded by the Regional VPU. In particular the team would also like to thank Messieurs: Homi Kharas, Chief Economist, EASPR; Jeffrey Gutman, Director Strategy and Operations (EAPVP); Mark Wilson, Sector Director, EASRD; Van Pulley, Director, EACPF; Stephen Mink, Lead Economist, EASRD; Llyod Mckay, Lead Economist, EASPR, Susan Hume, Country Program Coordinator, EACPQ and Wael Zakout, Lead Operations Officer, EASRD for their guidance and support. The team also extends a very warm thanks to NDCC, in particular, General M. Rosales for their guidance, availing of important information and considerable patience in the long process. A special thanks to all those who helped make the report a reality, even though they may not be explicitly named here. iii Acronyms AFP Armed Forces of the Philippines BDCC Barangay Disaster Coordinating Council/s BDOC Barangay Disaster Operations Center/s CBDM Community-Based Disaster Management CBO Community Based Organization/s CMDCC City/Municipal Disaster Coordinating Council DBM Department of Budget & Management DCs Developed Country/ies DCG Disaster Control Group DOC Disaster Operations Center/s DPM (Thailand) Department of Disaster Prevention and Mitigation DPWH Department of Public Works and Highways DRC Disaster Resilient Community/ies DLGI Department of Local Government & Interior DSWD Department of Social Welfare & Development EdM ­ (Japan) Earthquake Disaster Mitigation Research Center EqTAP ­ (Japan) Development of Earthquake and Tsunami Disaster Mitigation Technologies and their Integration for the Asia-Pacific Region) Project FEMA (US) Federal Emergency Management Agency EO Executive Order EMIP Emergency Management Institute of the Philippines IDNDR International Decade for Natural Disaster Reduction IDRM Integrated Disaster Risk Management JICA Japan International Development Agency LDCs Less Developed Country/ies LDCC Local Disaster Coordinating Council/s LGU local government unit LI Letter of Instruction LUM land use management MAAs Mutual Aid Agreements MMDA Metro Manila Development Authority MMDCC Metro Manila Disaster Coordinating Council MMEIRS Metro Manila Earthquake Impact Reduction Study NCDA National Civil Defense Administration NEDA National Economic Development Authority NGO Non-Government Organization NPO Not-for-Profit Organization NDCC National Disaster Coordinating Council OCD Office of Civil Defense PAGASA Philippine Atmospheric, Geophysical and Astronomical Services PD Presidential Decree PDCC Provincial Disaster Coordinating Council PDOC Provincial Disaster Operations Center PHIVOLCS Philippine Institute of Volcanology & Seismology PNP Philippine National Police RA Republic Act RDCC Regional Disaster Coordinating Council/s RDOC Regional Disaster Operations Center RoP Republic of the Philippines UP University of the Philippines WB The World Bank iv Executive Summary The Philippines by virtue of its geographic circumstances is highly prone to natural disasters, such as earthquakes, volcanic eruptions, tropical cyclones and floods, making it one of the most disaster prone countries in the world. These hazards cost the Government an average of P15bn per year in direct damages, or more than 0.5% of the national GDP, and indirect and secondary impacts further increase this cost. In addition, to this significant economic cost, there are also substantial social and environmental impacts. This high level of risk is what prompted the Bank and the Government to mount an informal study to: document the impacts of natural disasters on the social and economic development of the Philippines; assess the country's current capacity to reduce and manage disaster risk; and identify options for more effective management of that risk. The primary audience of this report is the Government, at all levels, the donor community and stakeholders involved in disaster management. The report is also of interest to the Bank primarily as it provides a good basis to determine where its assistance could best be used. The frequent disasters hinder the Philippine Government's efforts to reduce the incidence of poverty and reduce the number of people and assets vulnerable to these hazards. There are indications of close linkages between poverty and vulnerability to natural disasters and of their mutually re-enforcing effects. The poorer communities tend to be the most vulnerable. Data show that at the household level, poverty is the single most important factor determining vulnerability. This situation is exacerbated by rapid urbanization, environmental degradation and the increasing risk of environmental disasters, whether as a result of direct human impact and or from climate change. The Philippine institutional arrangements and disaster management systems tend to rely on a response or reactive approach, in contrast to a more effective proactive approach, in which disasters are avoided, by appropriate land-use planning, construction and other pre-event measures which avoid the creation of disaster-prone conditions. There is a widespread emphasis on post-disaster relief and short-term preparedness (forecasting, evacuation, etc.) rather than mitigation or post-disaster support for economic recovery, such as livelihood regeneration or tax breaks to affected businesses. This much shorter term focus does not adequately emphasize natural hazards as a potential obstacle to long-term sustainable development. To evolve to this more proactive role, it is important that a national framework for comprehensive disaster risk management be prepared and implemented. The framework would provide for political leadership and policy support at the highest levels, while facilitating the active engagement and implementation of all relevant stakeholders at the national, local, and household levels. The actors should include public agencies, the private sector, and civil society. The framework should incorporate the essential steps of integrated risk management, which include risk identification, risk reduction, and risk sharing/financing. The study identified some specific areas under these key themes that would need to be addressed to improve the current system and these are briefly presented below. Risk Identification. In this area, the fundamental requirements are reliable data on the type and amount of Philippine economic activity at risk which if not available complicate planning and risk reduction and risk sharing activities. To achieve this, high quality comprehensive hazard and vulnerability maps for major natural hazards would need to be produced or updated. In addition, there is scope for knowledge enhancement and understanding of the nature and scale of impact of disasters and forms of vulnerability. This, in turn would aid the implementation of good risk management practices, including greater consideration of hazard-related issues in broader sustainable development and poverty reduction policies and programs as well as of appropriate, cost-efficient post-disaster relief and rehabilitation efforts. As part of the strategy to reduce risk, it will be important not only to increase the focus on mitigation, but also on post- disaster support for economic recovery, such as livelihood regeneration or tax breaks to affected businesses. Risk Reduction. Once the risk has been adequately identified, measures would need to be taken to prevent, mitigate and reduce the inherent risks. The type of risk reduction measures to be implemented must be supported by appropriate institutional arrangements, including legislative and policy changes, as needed. The disaster management system in the Philippines is based on a decree that has not been updated in the last v 20 years, and to create an enabling environment for a comprehensive disaster management strategy, it will be essential that the governing decree is updated. Changes proposed will need to take into account evolving roles of the central and local governments and directly bring the private sector, specifically utilities, into both emergency management strategic decision-making and operational contexts, while promoting the sustainable management of hazards and risks in a way that contributes to the well being and safety of the public and property. However, before this work can be carried out, a more detailed review of institutional arrangements and capacities for disaster risk management to identify gaps and priorities must be carried out. Changes in the roles and responsibilities of the various actors will need to be accompanied by adequate coordination and implementing capacity in agencies involved in disaster management. In keeping with the shifting focus to a more proactive role, the type of coordination needed should be less of a top-down oversight function to one that is more participatory. Greater organizational, management and task synchronization would be prerequisites at both national and international levels. As risk reduction also requires that the resilience of the most vulnerable communities to hazard impacts be enhanced to help them cope with the hazards when they occur, the approach taken must emphasize a bottom-up approach with participation of all stakeholders. The Philippines system tends to be more of a centralized top-down administrative system than a community-based system and there are few incentives for local level initiatives. The system needs to be strengthened to encourage some local government units (LGUs) to initiate sustainable development practices just as much as it is to encourage other LGUs to continue along this path, often against bureaucratic obstacles. With very few exceptions, local level systems are response-driven ­ there is no obvious effort to initiate proactive hazard management/risk reduction coordination. Risk Sharing/Financing. The study found that presently the Government of the Philippines and individual households bear the majority of costs caused by natural disasters. More effective options for financing disaster risk and relieving the burden of disasters from the public sector should be explored, including the idea of a catastrophe insurance pool, and/or contingent credit facilities. As a start, the fiscal vulnerability of the federal and local governments to natural disasters can be reduced significantly by introducing institutional incentives1 for better risk management and institution building. To reduce the funding gap and the vulnerability of the poorest segments of the population and critical infrastructure to natural disasters, the existing national system of financing disaster losses could be redesigned to provide strong fiscal incentives to LGUs for more proactive risk management. More effective than diverting funds from ongoing development projects, a more efficient option for catastrophe risk financing would be the use of contingent credit facilities. The contingent credit facility could be extended to the National Calamity Fund to finance disaster management activities at the LGU level, Federal Government level as well as support an insurance pool. The credit facility could be designed to disburse quickly on an as-needed basis, triggered by the occurrence of a disaster of magnitudes agreed to by the Government and based on agreed criteria. If drawn down, the lending facility backing the National Calamity Funding could then be replenished without any major costs for the Bank or the Borrower. This funding approach for natural disasters would enable the Government to switch to a proactive mode of financing natural disasters by replacing multiple ex-post future emergency lending operations with a single line of credit, and to obtain access to immediate liquidity to meet reconstruction needs in the aftermath of a disaster. The study found that despite the high hazard risk in the Philippines, the insurance coverage for catastrophic perils for residential dwellings is almost non-existent. In addition, there is a limited risk bearing capacity of the domestic insurance market and an over-dependence on international reinsurers for claims paying capacity. In this regard, the idea put forward by several local insurers to create a Philippines Catastrophe Insurance Pool warrants serious attention. A Working Group would need to be created to explore the establishment of a Philippines Catastrophe Insurance Pool. 1The level of post-disaster funding to the LGUs is based more on the size of disaster losses and the local economy. The current system does not provide incentive for those LGUs that may have taken proactive steps in risk reducing measures. vi Potential Areas for World Bank Support To support the Philippines' effort to better manage disaster risk and enhance poverty reduction, the World Bank should examine the ongoing portfolio to identify how its ongoing development projects can support the goal of disaster risk reduction. In addition, the Bank should consider more direct support to the development of an integrated disaster management risk approach, through the provision of technical assistance and lending. The World Bank has developed a significant amount of expertise over the years through its provision of technical and lending support for disaster risk management. It should apply this to the Philippines in their efforts to improve poverty reduction. This assistance can take the form of a technical assistance loan to facilitate the development of a national framework for disaster risk management and institutional capacity building to implement the strategy. As described above, the Bank could also provide support in the development of contingency credit facilities to reduce the economic and financial impacts of disasters in the Philippines. vii Chapter 1. Introduction 1. The social and economic cost of natural disasters has increased in recent years due to population growth, change in land use patterns, migration and unplanned urbanization, environmental degradation and global climate change. About 75 percent of the world's major natural disasters between 1970 and 1997 occurred in the Asia and Pacific Region. In the East Asia and Pacific Region, there are several countries that are located in disaster prone zones and are exposed to a number of natural disasters annually. The list includes Vietnam, the Philippines, Cambodia, and most of the Pacific Island countries. In addition to the loss of lives and major destruction of economic and social infrastructure, natural disasters set back poverty reduction programs and cause diversion of government funds to pay for reconstruction and recovery efforts. 2. The Philippines is known as one of the most hazard prone countries in the world. Disasters have destroyed human, social, and physical capital, and they have derailed social and economic development, as funds are reallocated from ongoing programs to finance relief and reconstruction assistance. While the documented losses caused by disasters are formidable, the full impacts are not known, and are probably much more significant. 3. Due to its geographic circumstances, the Philippines has an unusually high exposure to natural hazards. The most catastrophic of these hazards include earthquakes, volcanic eruptions, tropical cyclones, floods and droughts. These events have had grave social and economic consequences on the country, costing an average of P15bn per year in direct damages between 1970 and 2000, and killing about 866 people per year in that same period. These figures do not consider the equally devastating indirect and secondary impacts of disasters. In addition to the large catastrophes that command a considerable humanitarian assistance response from the international community, there are an even larger number of small hazard events that do not register on the global or even national news scene but that consistently wipe out the few assets of the poor, locking them into the poverty cycle. Data on disaster losses also does not fully consider the impacts on the poor, who are the most vulnerable to disasters. At the household level, poverty is the single most important factor determining vulnerability, in part reflecting location of housing (e.g., on floodplains, riverbanks, steep slopes or contaminated land previously occupied by industrial facilities); level of access to basic services (e.g., refuse collection) particularly for illegal squatters; sources of livelihood; and level of access to financial and other assets and resources, leaving limited recourse to inter-temporal consumption smoothing. In the case of the Philippines, linkages between poverty and vulnerability to natural hazards are clearly evident. Despite this, however, and despite Government's central objective to reduce poverty, efforts to reduce vulnerability to natural hazards are not systematically included as a central component of the government's poverty reduction strategy. 4. In addition to the large catastrophes that command a considerable humanitarian assistance response from the international community, there are an even larger number of small hazard events that do not register on the global or even national news scene but that consistently wipe out the few assets of the poor, locking them into the poverty cycle. 5. A rapid assessment of disaster management in the country indicates that disaster management systems tend to rely on a reactive approach, i.e., institutional arrangements and hardware for post-disaster emergency response; preparedness measures such as evacuation plans and stockpiling of emergency supplies; and the provision of relief. Financing arrangements also tend to be reactive, with a strong reliance on humanitarian support from the international community, calamity funds for relief activities, and the diversion of funding for development activities for post-disaster recovery. International experience has shown that this is an ineffective manner to deal with disaster risk which detracts from development objectives. This reactive approach is in contrast to a more effective proactive approach, in which disasters are avoided by appropriate land-use planning, construction and other pre-event measures which avoid the creation of disaster-prone conditions. 1 6. The Philippines' main challenge for the medium-term remains to reduce poverty. Poverty is still largely a rural phenomenon. Its causes include: high dependence on agriculture where productivity has been declining and per capita economic growth is low; lack of adequate social safety nets, especially for poor women and children; and lack of educational attainment in terms of school drop-outs and low educational quality. Population pressures and a declining natural resource base have exacerbated these problems. To alleviate poverty, the Philippines has to overcome external and internal challenges. On the external side, it must cope with fluctuations in global markets on which it depends heavily. On the internal side, it needs to win back the confidence of investors in order to raise the currently low volume of investment that is necessary for the country's development. Moreover, institutional capacity must improve, especially at the level of local governments whose responsibility for delivering development services is growing. 7. Natural disasters are a key cause of increased poverty and environmental degradation, and reducing the risk of disasters will be key to achieving the strategic goals laid out for the nation. This study was undertaken by the Rural Development Department of the East Asia and Pacific Region (EASRD) and the Disaster Management Facility (DMF) of the World Bank in partnership with the Government of the Philippines (GoP). The study aims to document the impacts that natural disasters have had on the social and economic development of the Philippines; assess the country's current capacity to reduce and manage disaster risk; and identify options for more effective management of that risk. A research mission of the World Bank was carried out in collaboration with the GoP in May 2003. This report details the findings of that mission, and provides a starting point for a dialogue with the GoP on how the World Bank can best support the country's efforts to better manage disaster risk and thus more effectively reduce poverty. 2 Chapter 2. Hazard Exposure and Disaster Impacts in the Philippines 8. The process of managing disaster risk effectively begins with risk identification. Risk identification comprises assessing the hazard exposures of a country, as well as understanding the vulnerabilities of the targets of extreme events in order to determine the potential impacts and losses. In this way, informed decisions can be made regarding steps to reduce the impacts of disasters. The development of a clear understanding of the impacts and losses of past events is also important to comprehending the impacts of future events. As this chapter will show, the Philippines has a high exposure to a number of natural hazards, and these hazards have had grave social and economic impacts on the country's development over the years. Natural Hazard Exposure 9. Due to a combination of the Philippines being at the junction of several tectonic plate boundaries, as well as being in an area of frequent tropical cyclones, the nation is unusually susceptible to natural hazards, with earthquake, volcanic eruptions, tropical cyclones, and floods being the most catastrophic hazards, see Table 1. This section provides an overview of each of these hazards. Table 1. Ranking of Disaster-related 20th century Philippine Fatalities Hazard event Number Killed Damage (USD millions) Typhoon 28812 5653 Earthquake 9572 517 Volcano 6331 228 Flood 2545 431 Source: CRED, 1998 10. Earthquakes: The Philippines is an archipelago of over 7,000 islands and islets with a total land area of 300,000 square kilometers. Geologically, the archipelago is the product of accretionary wedges and volcanic eruptions resulting from the collision of the Philippine Sea, Pacific and Eurasian plates, as well as smaller platelets (Sulu, Celebes Sea) being forced northward by the large India-Australia plate. The plate motions are accompanied by earthquakes and volcanism. As a result of its geological exposure, the country experiences on average 887 earthquakes every year, some of which have proved to be damaging. 11. The US Geological Survey lists 168 significant damaging events in the Philippines since 1599, equivalent to a damaging event every 2.5 years on average. However, this record is biased, in that earlier centuries had fewer damaging events due to lower population and assets at risk as well as poorer communications and little or no scientific instrumentation. Two major damaging earthquakes have occurred in the Philippines in recent years ­ the 1976 Mindanao event, which killed approximately 6,000 and caused about USD 400 million in damage (in present value), and the 1990 Central Luzon event, which killed over 1,000 and caused damages of about USD 400 million (in present value). 12. A comprehensive seismic hazard analysis for the entire Philippines is beyond the scope of this study and, indeed, such analyses for the Philippines are relatively scarce2. Few exist, such as seismic hazard analyses by Bautista et al (n.d.) and that by Thenhaus et al (1994) in which a probabilistic analysis, which shows peak ground accelerations (PGA) on medium soil was expected to be exceeded with probability of 10% in a 50-year period3. More recently, Torregosa et al (2001) performed a similar analysis, confirming 2R. Solidum, personal communication. 310% in 50 year probability of exceedance is a standard measure of ground motion, used in the Philippine, US and other building codes as a design basis ground motion. It is also referred to, and seen as equivalent to, a `475 year return period' ground motion, although strictly speaking this is not an accurate equivalence. 3 Thenhaus' findings. A review of these studies shows accelerations in the Metro Manila area to be about 0.4g, comparable to those in San Francisco, Tokyo and other high seismicity areas. 13. The Philippines are bisected longitudinally by the Philippine fault, a major tectonic feature which has many subsidiary faults, one strand of which (the Valley fault system) bisects Metro Manila. This strand is capable of magnitude 6-7 events occurring on average every 200-300 years, and has not ruptured in over 200 years (Nelson et al, n.d.). PHIVOLCS, in association with the MMDA and JICA is currently carrying out a detailed analysis of the potential for damage in the Metro Manila area, focusing on this risk. Preliminary results indicate on a deterministic basis ground motions of about 0.4g or greater, depending on soil conditions, in the Metro Manila region. On a probabilistic basis, their preliminary analysis (which included recent paleoseismic data) shows lesser ground motions, in general. 14. In addition to shaking, earthquakes cause damage in other ways, the most significant of which are liquefaction, landslides, tsunami and fires following earthquakes. Liquefaction typically occurs in loose saturated sandy ground, and is the loss of strength of that ground due to strong ground motion ­ it was particularly damaging in Dagupan in the 1990 earthquake. 15. Volcanoes: Out of 220 volcanoes in the archipelago, 22 are classified as active. Simkin and Siebert (1994) document literally hundreds of historic eruptions ­ Mt. Mayon, for example, is indicated to have erupted 12 times in the 20th century alone. The most active volcanoes are probably Pinatubo, Taal, Mayon, Canlaon and Ragang. Currently, PHIVOLCS, who have the responsibility for monitoring volcanic activity, can currently monitor only six of the 22 active volcanoes for possible eruptions. 16. Volcanic eruptions are accompanied with a wide variety of destructive effects including release of magma, explosive effects, expulsion of large projectiles, pyroclastic flows (flow of hot molten material), ashfall, release of clouds of very hot gases, lahars, mudflows and ground shaking, to name the most common. 17. The Mount Pinatubo eruption in June 1991 provided an example of how rapidly and destructively a volcano can erupt. That event was the second largest volcanic eruption of this century, and by far the largest eruption to affect a densely populated area. Fortunately PHIVOLCS had forecast the eruption, resulting in the saving of at least 5000 lives and USD 250 million. However, the eruptions have dramatically changed the face of central Luzon, and the impacts of the event continue to this day. About 200,000 people who evacuated the lowlands surrounding Pinatubo have returned home but face continuing threats from lahars that have already buried numerous communities4. Rice paddies and sugar cane fields that were buried by lahars will be out of use for years. Other volcanoes, such as Mt. Mayon, are even more active, and surrounded by many communities at risk to lahars and other hazards. 18. Probabilistic analysis of volcanic hazards is very difficult, and beyond the scope of this study. A review of the historic record indicates that a general approximation would be that central and southern Luzon are likely to see a significant eruption about once every 3 years, with a major eruption perhaps every few decades. Most active sources are Mayon and Taal. Mindinao appears to have a somewhat lower likelihood of a major eruption, based on the historic record. However, any probabilistic estimate of volcanic hazard needs considerably more study, with a close involvement of PHIVOLCS. 19. Tropical cyclones: The climate of the Philippines is tropical and is strongly affected by monsoon (rain-bearing) winds, which blow from the southwest from approximately May to October and from the northeast from November to February, although there is considerable variations in the frequency and amount of precipitation across the archipelago. From June to December typhoons often strike the archipelago. Most of these storms come from the southeast, with their frequency generally increasing from south to north. On 4A lahar is a mixture of water, mud and rock, resulting in a landslide or mudflow of volcanic fragments on the flanks of a volcano. 4 average, about 20 typhoons occur annually, with the months of June to November averaging approximately 3 typhoon strikes per month. Luzon is significantly more at risk than more southern areas. Typhoons are heaviest in Samar, Leyte, eastern Quezon province, and the Batan Islands, and when accompanied by floods or high winds they may cause great loss of life and property. Mindanao is generally free from typhoons. Figure 1 attached shows portions of coasts prone to storm surge, based on the historical record. 20. Typhoons have killed about 29,000 people in the country in the 20th century, including about 6000 in the 1991 typhoon. As Table 2 indicates, about 500 people are killed each year, and about P4 billion are lost due to tropical cyclones. In certain cases, a single event can kill as many as 6000, and cause P20 billion in damage. In addition to high winds, a major damaging element of tropical cyclones is storm surge. Table 2. Tropical Cyclones Impacts in the Philippines, 1970-2002 Persons Affected Displaced Houses Destroyed Properties Bn Pesos (~54.5 Pesos to Dead Injured Families Persons Families Persons Total Partial US$1) Tot 16,654 20,311 16,843,467 83,575,773 2,111,246 9,398,995 2,140,938 4,876,749 140 Per Yr 520 635 526,358 2,611,743 65,976 293,719 66,904 152,398 4 Source: http://www.ndcc.gov.ph/home_typhoon%20data%2070-2002.html 21. Flooding. Flooding is yet another hazard facing the Philippines, and has annual impacts on the country. However, relatively little information is available on flooding, and a detailed mapping of flood prone areas in the Philippines has yet to be done. Figure 2 attached shows major flood prone areas in the Philippines. Economic Impacts of Past Disasters in the Philippines 22. Between 1970 and 2000, the Philippines incurred an average annual direct damage of P15bn per annum (in real 2000 prices) as a direct consequence of natural disasters5, equivalent to an average 0.7% of GDP every year (Table 3). In 1991 alone losses totaled P65bn (in 2000 prices), equivalent to 2.6% of GDP, as the country experienced both a major earthquake and exceptionally heavy typhoon-related damages. Over the period 1970 to 2000, an average of 866 lives was also lost every year. Typhoons alone accounted for 65% of lives lost (56% excluding the year of the Ormoc floods in which some 5,000 perished) and 76% of total damage, reflecting their high, annual frequency. 23. As regards the country's principal food staple, rice crop losses equivalent to 2.6% of actual production (in volume terms) were experienced as a consequence of typhoons and flooding between 1991 and 2000 (1.8% excluding 1998) (Table 4). Typhoons, floods and drought collectively caused losses equivalent to 3.3% of total actual production (2.0% excluding 1998). 24. It is generally accepted that the Philippines is one of the most hazard prone countries in the world, a ranking surely warranted by the above evidence. Yet, for a country with such a reputation, it scores extremely low on global disaster indices. It does not appear at all on the UNDRO ranking of the world's 50 "most disaster-prone" countries, defined as total damage from 'significant disasters' (exceeding 1% of GDP over the period 1970-89) (UNDRO, 1990); ranks 31st on the Commonwealth Secretariat index based on population affected over the period 1970-96 (Atkins et al, 2000); and 25th according to the Commonwealth Secretariat Index based on number of disasters relative to land mass over the period 1970-96 (ibid). 5 Defined in terms of direct losses to agriculture, infrastructure and the private sector. 5 Table 3. Estimated damage, in real 2000 prices and numbers of deaths from natural disasters 6 7 Volcanic Typhoons Earthquakes8 eruptions & lahars9 Other10 Total Estimated No. of Estimated No. of Estimated No. of Estimated No. of Estimated No. of Year damage deaths damage deaths damage deaths damage deaths11 damage deaths (m Peso) (units) (m Peso) (units) (m Peso) (units) (m Peso) units) (m Peso) (units) 1970 2,472 1,328 137 17 0 27 3 2,636 1,605 1971 1,058 59 1 586 43 1,646 146 1972 4,395 298 294 28 4,688 1044 1973 5,295 74 106 15 743 51 6,143 180 1974 5,807 153 61 2 57 41 5,925 373 1975 524 39 126 17 650 248 1976 9,970 313 8,398 3,782 1,143 40 19,811 4202 1977 4,157 99 63 1 201 14 4,422 128 1978 69,002 663 0 2 69,002 665 1979 4,102 69 51 1 4,153 67 1980 12,674 143 22 51 3,169 336 15,865 530 1981 9,875 484 33 125 9,908 609 1982 11,821 337 820 27 12,640 364 1983 3,256 126 92 19 4,760 41 8,108 186 1984 23,144 1,979 264 10 0 23,418 1979 1985 9,424 211 0 14 25 59 9,449 284 1986 5,966 171 31 4 5,997 175 1987 12,759 1,020 2,209 0 14,968 1020 1988 24,982 429 2,459 0 0 27,443 429 1989 11,746 382 1,027 101 12,774 483 1990 29,333 670 28,285 1283 7,958 53 65,585 2006 1991 9,102 5,199 21,094 850 3,318 72 33,515 6121 1992 9,331 117 335 1 1,014 6 6,433 31 17,112 155 1993 34,417 794 126 80 1,912 4 36,455 878 1994 5,007 242 806 83 16 20 568 31 6,397 376 1995 22,207 1,204 2,041 149 24,248 1353 1996 3,832 124 378 34 4,210 158 1997 759 77 1,410 60 2,169 137 1998 19,609 490 9,696 28 29,305 518 1999 2,749 100 4 5 1,554 300 4,307 405 2000 7,469 345 90 1,779 206 9,248 641 1970-2000 average 12,137 572 1,315 170 726 34 1,699 61 15,877 886 Five-year averages 1970-4 3,805 382 61 7 - - 341 33 4,208 670 1975-9 17,551 237 1,692 757 - - 364 15 19,608 1,062 1980-4 12,154 614 23 14 53 - 1,758 106 13,988 734 1985-9 12,975 443 492 3 - - 659 33 14,126 478 1990-4 17,438 1,404 5,885 273 4,450 191 4,038 38 31,811 1,907 1995-9 9,831 399 1 1 - - 3,016 114 12,848 514 1990-94 excl. year of Ormoc flood 456 25. The Philippines' relatively low ranking in part reflects the indices' focus on single catastrophic events rather than the cumulative impact of annual, if individually often relatively localized, events in a country the size of the Philippines. Figures on reported damages in the Philippines are also almost certainly an underestimate of loss for several reasons. First, they are based on government damage assessment reports that only cover selected types of damage, focusing on damage to public property and to assets of lower-income households potentially eligible for state assistance. Second, it is not clear if disasters that are not officially 6 Damage to agriculture, infrastructure and private property. 7 Based on GDP at market price deflator as GDP deflator at factor cost is not available for the 1990s. 8 From 1994 onward, "small" volcanic and earthquake events are reported under `other'. 9 From 1994 onward, "small" volcanic and earthquake events are reported under `other'. 10 Floods and droughts. From 1994 onward also includes "small" volcanic and earthquake events. 11 Deaths were almost entirely as a consequence of flooding. 6 declared are included in the figures. As illustrated in the case of Navotas (Box 4) not all qualifying events are necessarily officially declared as disasters, depending on whether or not an LGU wishes to draw on Local and National Calamity Funds. Third, there are an unknown number of lesser events that fall below the threshold for an event that can be deemed to constitute a `disaster'. Evidence collated by the DesInventar initiative in Latin America suggests that in some countries the impact of these `everyday disasters' may be much greater than those of the larger events formally recorded as disasters (IFRC, 2002). Table 4. Philippines- annual rice production and losses arising as a consequence of natural disasters, 1991-2000 Typhoons and floods Drought Total losses Total palay Loss in Loss as Loss in Loss as Loss in Loss as production Production % actual Production % actual Production % actual Production Production Production Year 000 t 000 t % 000 t % 000 t % 1991 9,673 152 1.6 16 0.2 168 1.7 1992 9,129 11 0.1 79 0.9 90 1.0 1993 9,434 246 2.6 0 0.0 246 2.6 1994 10,538 112 1.1 48 0.5 160 1.5 1995 10,541 328 3.1 45 0.4 373 3.5 1996 11,284 73 0.6 1 0.0 74 0.7 1997 11,269 75 0.7 15 0.1 90 0.8 1998 8,555 1,048 12.3 462 5.4 1,511 17.7 1999 11,787 321 2.7 1 0.0 322 2.7 2000 12,389 390 3.1 0 0.0 390 3.1 Total 104,599 2,757 2.6 667 1.5 3,424 3.3 Less 1998 96,044 1,709 1.8 204 1.2 1,913 2.0 Source: Philippine DA 26. The very fact that the Philippines experiences annual typhoon-related losses has also created an apparent dis-regard to their macroeconomic ramifications. Disaster losses are conventionally categorized as: · direct costs - physical damage to capital assets, including social infrastructure; · indirect costs - knock-on disruption to the flow of goods and services (e.g. reduced output, loss of earnings and job losses). · secondary effects - short- and long-term impacts of a disaster on the overall economy and socio- economic conditions (e.g. fiscal and monetary performance, levels of indebtedness, the distribution of income and scale and incidence of poverty). 27. Yet only severe El Niño episodes and major geophysical disasters are generally regarded as economically significant events while the economic ramifications and impacts of, cumulatively, considerable direct damage as a consequence of annual typhoons are ignored. 28. Key events identified retrospectively in the literature are droughts in 1984 and 1987, the combined impacts of the 1989/90 drought and 1990 Baguio earthquake, the 1991 Mount Pinatubo eruption and the combined impacts of a further drought and the Asian financial crisis in 1997-98. All of these events are reported to have had a discernible impact on GDP performance (Figure 3). 29. The series of disasters in the early 1990s, in particular, were widely claimed to be responsible for a period of poor economic performance.12 Drought in the latter part of 1989 and first half of 1990 reduced 12 The following three paragraphs draw heavily on Benson (1997). 7 agricultural production, damaging some P 365m worth of crops and causing an estimated opportunity loss of P1.2bn in palay production and P 808m in corn production (Philippine NEDA, no date). It also resulted in severe power shortages, causing estimated daily losses to the industrial sector of P 25-75m and reducing the rate of industrial growth for the year from 7.4 percent in 1989 to 2.5 percent in 1990 (Philippine NEDA, no date). The severe earthquake in Luzon in July 1990 (see Box 1), caused a further economic setback, forcing NEDA to adjust GDP growth forecasts for 1990 and 1991 down from 4.8 to 3.8% and from 5.5 to 5.2% respectively. Actual performance was even lower, reflecting the combined impacts of the earthquake the 1989/90 drought and also the Gulf Crisis, which in turn increased the cost of oil imports and displaced many Filipino overseas workers. Box 1: The July 1990 Luzon Earthquake The Philippines suffered a major earthquake of intensity 7.7 on the Richter scale in July 1990 with an epicenter near San Jose City, Nueva Province, Luzon. Some 100,000 km2 including all of North and Central Luzon as well as parts of Central Luzon were affected by the earthquake, with most serious damage over an area of some 15,000 km2. The cities or towns of Baguio, Dagupan, Agoo, Aringay and Pura were particularly badly affected whilst Tarlac, Cabantuan, Rizal and Manila were marginally damaged (Rantucci, 1994). Total damage was estimated at P 12.2bn according to NDCC data. Some 56% of damages occurred to infrastructure, 12% damage to agriculture and 32 % to private property, principally non-housing properties. Some 1,293 lives were also lost. The earthquake resulted in temporary major disruptions to transportation, communications and power networks and to the supply of water for agricultural purposes, affecting agricultural production capabilities and marketing and distribution arrangements in the short to medium term. As of 1993, road communications had still not yet been fully restored (Alatec- Harris-Tym Group, 1993). Fishponds and almost 87,000ha of rice land was destroyed, including 60,000ha of irrigated land and some fourteen national and 174 communal irrigation systems. The livestock population declined rapidly as animals were killed either accidentally as a consequence of the earthquake or deliberately to meet short-term food requirements (Fernandez and Gordon, 1993). Post-harvest and storage facilities and several dams were also damaged. Several dams suffered damage. In particular, Ambuklao Dam, located near Baguio, was silted up to a few meters below water level forcing the shutdown of electricity production and hampered irrigation activities. The dam had been expected to have a useful life of around 50 years but this was reduced to 28 years as a consequence of the earthquake (Rantucci, 1994). Production capacity at the Baquio Export Processing Zone, at the time one of only four in the country, was also severely affected. One building in the EPZ housing two garment firms, one plastics firm, one electronics firm and one pipe fittings firm collapsed entirely whilst another building housing seven firms was partly damaged. Losses from the Baguio EPZ together with those from the mining sector were estimated at around US$35m (P 851m) between mid-July and the end of August 1990 alone (Philippine NEDA, no date). Source: After Benson (1997) 30. The government responded with the introduction of a stabilization program at the beginning of 1991, including efforts to improve the fiscal deficit and control the money supply. However, any hopes of improved economic performance were thwarted by further disasters, including the eruption of Mt. Pinatubo in July 1991 (Box 2), Super Typhoon Trining in late October and the Ormoc flood towards the end of the year. These disasters contributed to negative GDP growth rates of 0.6 % year-on-year, in turn largely reflecting lower growth in Central Luzon and the Metro Manila region and tight monetary and fiscal policy as the government strove to meet pre-determined IMF economic targets. 8 Box 2: Eruption of Mt. Pinatubo Mt. Pinatubo, a volcano located on the west coast of Central Luzon some 100 km northwest of Manila, violently erupted in July 1991. The eruption, one of the largest globally in the 20th century, caused severe damage. Some 6 km3 of pyroclastic material was deposited in river basins whilst a further 1 km3 of ash was deposited across an area up to 40 km from the volcano, effectively altering the hydrology of the whole region (USACE, 1994). Fallout affected a total area of 340,000 km2 (PHIVOLCS, 1991). The impact of the eruption was exacerbated by Typhoon Diding, which occurred immediately after the eruption scattering water-soaked ash over a very large area and causing massive mudflows. These, in turn, covered large areas of agricultural land and destroyed buildings, bridges and roads and other infrastructure. Minor eruptions continued until 4 September 1991. The provinces of Zambales, Pampanga, Tarlac and part of Bataan, all in Region III, were most severely affected by the initial eruption of Mt. Pinatubo. Some 80,000 ha of agricultural lands and fishponds was buried by ash and the initial lahars; transport, communications, power, irrigation and other infrastructure as well as houses and public buildings were damaged; drains and other water conduits were blocked, increasing the risk of flooding; commercial and industrial operations in the cities of Angeles and Olongapo suspended; some 600,000 jobs lost, equivalent to around a quarter of total employment in Central Luzon; and 200,000 people evacuated at the height of the eruption (PHIVOLCS, 1991). Subsequent lahars generated by heavy rainfall have occurred in every year since the eruption, although lahar dykes defenses have gradually been constructed to contain lahars and reduce losses. The lahars have caused extensive damage to agriculture and infrastructure, for example, lahars in 1991 and 1992 alone affected almost 260,000 persons and destroyed 4,190 houses. The continued annual threat of lahars has created certain problems in designing appropriate rehabilitation programs for livelihood creation and for the reconstruction of infrastructure. Prior to July 1991, Central Luzon had been the Philippine's prime rice growing region, accounting for 20.5% of national palay production and 15.7% of gross acreage. Ten years on, in 2000, it accounted for 15.2% of national production and 13.% of gross acreage. Sugar production and aquaculture had also been important. For example, had Central Luzon produced some 45 percent of the country's total fishpond production in 1990. By 1994 aquaculture output from the region was still only some 60 percent of previous levels due to the obstruction of water flows and tidal exchange. Coastal and freshwater fishing operations as well as fragile eco-systems, including mangroves and coral, were also disrupted by increased siltation, changing river flow patterns and destruction of breeding areas (USACE, 1994). These problems have continued as more ash is washed downstream. Source: After Benson (1997) 31. From 1992, tighter domestic policies began to attract external capital flows whilst inflation fell and foreign exchange reserves rose. However, GDP growth remained very weak as a direct consequence of continuing tight monetary and fiscal policies. The economy also faced particular difficulties relating to electricity shortages in 1992, in part the consequence of a drought in the first four months of the year. 32. Towards the end of the decade, in 1998, the combined impacts of a further widespread drought ­ widely claimed the most severe ever (e.g., PCARRD/DA, 2001) ­ and the Asian financial crisis resulted in a 0.6% decline in real GDP, a 2.6% decline in real per capita GDP and a 6.4% decline in agricultural GDP. Unemployment and inflation both rose to double digits, the latter fuelled by food price increases. 33. However, direct damage resulting from the eruption of Mt. Pinatubo was equivalent to only 1.2% of GDP, and that from the 1991 Luzon earthquake was 0.9%. Damage from typhoons averaged 0.5% every year over the period 1970-2000, implying far greater cumulative losses. Despite this, typhoons, which occur annually in the Philippines, are apparently widely accepted as a fact of life and attract little interest from policy and decision makers in either the public or private sectors. The very fact that they occur so frequently also makes it more difficult to measure their impact because the benefits of a typhoon-free year are not readily visible, in turn again apparently contributing to the perception that they are of little macroeconomic consequence. At the local level, however, their impacts can be devastating, destroying key infrastructure and reinforcing poverty. 9 34. There has been little formal analysis of the longer-tem impacts of disasters in the Philippines but annual infrastructure losses and related diversion of scarce public resources into their replacement must ultimately have an impact on the country's long-term sustainable development. For instance, only modest success in efforts to improve the country's transportation systems and increasing difficulties in meeting the social infrastructure needs of the country's rising population have been in part attributed to the fact that a large proportion of available public resources earmarked for such purposes have had to be re-directed in response to calamities (Benson, 1997). 35. A fuller analysis of both the impact of individual disasters and of current and likely future vulnerability to natural hazards from a macroeconomic perspective is required. Careful analysis would shed light on the extent to which and how risks emanating from natural hazards could be better managed. It would help facilitate the consideration of risks emanating from natural hazards both in the broad development planning process and in the design of individual projects as well as help ensure provision of appropriate post disaster economic support. Assessing Impacts of Disasters 36. According to rules and regulations under the provision of PD 1566, in the aftermath of a disaster all concerned agencies should undertake a survey of damage within their scope of responsibility and submit detailed damage assessments and requirements for relief and rehabilitation assistance. The reports should include information on numbers of casualties, deaths, persons affected and losses to housing, public infrastructure and crops. 37. These assessments should be undertaken by local disaster coordinating councils (DCCs) and submitted to the regional and then the national council. The NDCC should collate the various assessments together with reports prepared by various regional government departments and submit a single document to the President of the Philippines recommending whether or not a disaster should be declared and the amount of national funds that should be released in support of the relief efforts.13 38. Various efforts have been made to improve the quality and coverage of government post-disaster damage assessments in recent years, including via the provision of some limited training in assessment processes. Individual departments have also drawn up their own forms for reporting damage to help improve reporting practices. 14 39. Additional efforts have been made to corroborate damage assessments, in part by ensuring that damage estimates are validated by the relevant department at national Government level before being released. The DA has also undertaken considerable additional effort to improve reporting of crop damages, including by developing a matrix for validating rice and corn losses resulting as a consequence of strong winds, floods and drought. These calculations take into account the growth stage of the crop; expected losses from winds of varying strength and length of exposure to them; and days submerged under clear and muddy flood waters. The DA is thus able to compare reported losses against expected ones and adjust figures accordingly if necessary. 13The UN is not involved in the damage assessment process but could be in the event of a big disaster and so has recently provided some training to UN staff. 14Under an initiative in the mid-1990s involving the NSCB and a disaster statistics task force, a detailed assessment checklist was additionally prepared to complement the much briefer assessment forms already in use. However, this initiative was latterly shelved following a change of administration. 10 40. However, a number of problems still remain with the damage assessment process and the quality of reports is variable. · Most fundamentally, there are no comprehensive guidelines on how damage assessments should be undertaken. Different government departments follow different practices and there appears to have been little attempt to coordinate either their scope of coverage or assessment methodologies. In particular, different departments use different methodologies for valuing damage. For example, agricultural losses are valued at replacement value. Destroyed houses are also valued at replacement value but based on the cost of provision of core shelters to eligible households only.15 Partially damaged houses are valued at amounts to be made available for repair (maximum of P5,000 per house).16 Damage and destruction of agricultural infrastructure is valued at book value. Damage to other infrastructure for which DPWH is responsible is valued at re-establishment or cost of repair. NDCC has developed some standard formats for reporting overall damage, including some information on costs, but there is no guidance on how this damage should be valued.17 · The scope of coverage and detail of reporting is limited, focusing on damage experienced by lower income groups and sub-sectors that may be eligible for public assistance. Each agency is concerned only with losses within its own mandate of activity while private sector losses are ignored. This reflects the fact that, as already noted, the primary purpose of damage assessments is to determine emergency response, relief and rehabilitation requirements. · Assessors typically lack relevant training, adding to the uncertainty concerning the accuracy of assessments. · There is no baseline data for use in undertaking assessment. · There is often apparently poor coordination between agencies. Some government departments report (contradictory) information on the same category of loss. Requirements on the timing of reporting also vary. A series of reports at various points after a disaster is required by various government Departments to provide validation of impacts reported in the earlier reports but these appear to follow different schedules. According to PD 1566, for instance, reports should be submitted two weeks and two months after the occurrence of a calamity. The NDCC itself has drawn up standard reporting formats on impacts to be submitted one hour, six hours and 12 hours after a disaster, but apparently not beyond that period. Meanwhile, the DA requires reports on agricultural damage upon advise of an impending disaster (detailing forecast impacts), within 24, 36 and the 72 hours of a disaster, within a `subsequent' period and finally within 10 days of an event. · There is no clear system for reconciling differences in data reported by more than one agency. For example, figures on casualties and mortalities are reported by both the DoH and DSWD while the DSWD and DA both report data on numbers of people affected. To do this will also require that an assessment of the current data collection and analysis is carried out. 15Post disaster, the DSWD runs a core housing project, providing cyclone proof housing to families who have lost their homes. The cost of a single unit ranges from P3,500 to P25,000, depending on the type of construction materials used. They are 15m2 and can house a family of around six. Approximately 35,000 core shelters have been built since the program began in 1998, funded out of the QRF. 16The DSWD's standard damage assessment form also requires information on the numbers of houses totally and partially damaged and on the `cost of assistance', which appears to be defined as monetized assistance provided by DSWD, LGUs and NGOs and other governmental organizations. 17A flash report outlining basic information on the type of disaster, location of incidence and so forth should be submitted to the NDCC within one hour of a disaster; a damage and needs assessment, primarily focusing on medical, health, food, clothing and shelter requirements and on damage to lifelines, within six hours; and an information guide on damage assessment within 12 hours, giving further information on the damage to infrastructure (including commercial facilities) and agriculture, provision of essential utility services and so forth. 11 · Individual agencies sometimes report directly to their regional and national counterpart agencies rather than to the LGU, further contributing to problems in integrating agency assessments and prioritizing needs. Meanwhile, some LGUs report directly to the President before regional and national agencies have had time to validate figures. · Damage assessments are typically completed in a very short period of time, with no later attempt to assess longer-term socio-economic impacts or lessons learned in reducing future vulnerability. Post-disaster damage assessments often represent little more than stocktaking exercises, focusing on direct costs relating to damage to buildings, other infrastructure, capital equipment and standing crops and human losses. · There can be some political massaging of figures, particularly at the LGU level. 40. In summary, there is scope for considerable improvement in post-disaster damage assessment methodologies and practices. It is recommended that existing guidelines developed by the Economic Commission for Latin America and the Caribbean (ECLAC) should be adapted for use in the Philippines. More comprehensive assessment would enhance knowledge and understanding of the nature and scale of impact of disasters and forms of vulnerability. This, in turn would aid the implementation of good risk management practices, including greater consideration of hazard-related issues in broader sustainable development and poverty reduction policies and programs as well as of appropriate, cost-efficient post- disaster relief and rehabilitation efforts. Fiscal impacts 41. Natural disasters can have important implications for public finance. Disasters are likely to result in additional expenditure and/or the partial reallocation of already committed financial resources, with implications for planned investment and other expenditures. Public revenue may also decline as levels of economic activity fall. In consequence, a government may face increasing budgetary pressures which could in turn increase levels of domestic and/or external borrowing or increasing the money supply, each, in turn, with potentially significant knock-on effects (Benson and Clay, 2003). 42. In the case of the Philippines, national budgetary resources in the form of calamity funds, as well as local government resources, are annually appropriated for emergency relief and rehabilitation activities. This constitutes good budgetary practice in a country where disaster-related expenditure occurs every year. Annual budgetary allocations help strengthen both financial planning and also fiscal discipline more broadly (Benson and Clay, 2003). 43. However, available evidence suggests that, at least at the national level, existing calamity funds may be insufficient to meet much of the costs of rehabilitation and reconstruction even in years of lower loss and are likely to be grossly inadequate in the event of a major disaster. Despite this, it is difficult to discern much impact of natural disasters on either government expenditure or revenue at the aggregate level in the Philippines. This is not an uncommon or surprising finding. When broad fiscal aggregates, such as central government's recurrent and capital expenditure, revenue and the budgetary deficit are examined, disasters are often found to have little discernible impact (Benson and Clay ,2003). Instead, a fuller understanding of fiscal consequences of disasters requires a careful, more disaggregated examination. This examination should be undertaken in the context of overall budgetary performance, recent government policies and budgetary targets, and the structure of government revenue and expenditure itself (ibid). 44. For example, even major disasters in the early 1990s may have had little apparent impact on broad fiscal aggregates in the Philippines. However, this reflected a gradual underlining increase in direct tax revenues that had started in the mid-1980s as a direct consequence of tax reforms, and a concerted and successful effort to reduce the consolidated public sector deficit. National government expenditure actually fell in 1991 due to efforts to satisfy conditions under an IMF stabilization program, despite particular pressures placed on 12 government spending by both the July 1990 Luzon earthquake reconstruction program and by the Mt. Pinatubo relief and rehabilitation efforts. 45. The composition of government expenditure, particularly the share of non-discretionary expenditures, is also relevant in analyzing the budgetary implications of natural disasters. In the case of the Philippines, there is little flexibility in the national budget due to very high levels of non-discretionary spending. Total expenditure on personal services, the allotment to LGUs under the IRA and interest payments alone totaled 73% of the 2001 actual national government budgetary obligation; 73% of its adjusted expenditure in 2002; and 76% of proposed expenditure in 2003. Total capital expenditure (including capital transfers to LGUs) accounted for only 14% of total expenditure in 2001, 14% in 2002 and 13% of planned expenditure in 2003. Capital expenditure as a percentage of GDP is also very low, standing at a mere 2.9% of GDP (at factor cost) in 2001. Operations and maintenance (O&M) is similarly under-funded, with implications for vulnerability to natural hazards. Thus, disaster-related expenditure has to compete for very limited budgetary resources. The World Bank (2000: 74) comments that limited capital and O&M expenditure `cannot but damage the long-term development prospects of the country', a problem that must be exacerbated by annual damage and destruction of existing infrastructure. With some loss of international revenue expected as trade is gradually liberalized, limited fiscal flexibility seems set to remain a problem in the short to medium term. This limited flexibility leaves the Philippine government potentially highly exposed to economic shocks such as major natural disasters. 46. Indeed, the government needs to develop an explicit policy on the financing of post-disaster response; beyond making some relatively limited budgetary allocations for use in the event of a disaster and relying on annual unscheduled reallocations of resources. In developing this policy, the Government should explore risk transfer options as well as the precise scale and nature of annual reallocations and the appropriateness of current levels of expenditure on mitigation and preparedness. It should also bear in mind that disaster-related external assistance for immediate response or reconstruction cannot be assumed to be additional (Benson and Clay, 2003). Disasters and Poverty in the Philippines 47. Poverty and vulnerability to natural hazards are closely linked and mutually reinforcing. Poor and socially disadvantaged groups are usually the most vulnerable to hazards, reflecting their social, cultural, economic and political environment. Disasters, in turn, are a source of transient hardship and distress and a factor contributing to persistent poverty. Indeed, at the household level, poverty is the single most important factor determining vulnerability, in part reflecting location of housing (e.g., on floodplains, riverbanks, steep slopes or contaminated land previously occupied by industrial facilities); level of access to basic services (e.g., refuse collection) particularly for illegal squatters; sources of livelihood; and level of access to financial and other assets and resources, leaving limited recourse to inter-temporal consumption smoothing. The covariate nature of natural hazards also implies that there is limited scope for community level support systems. The poverty-exacerbating nature of vulnerability can be further reinforced by deliberate risk- averting livelihood choices that poorer households may make. For example, poorer households may choose to forego the potential benefits of higher yielding crops in favor of more hazard-tolerant ones, implying more stable and secure but, in most years, lower earnings. 48. The Philippines has a high incidence of poverty, standing at 26% in 2000 and poverty reduction is a central theme of development policy. The country had achieved considerable progress in reducing the level of poverty between 1990 and 1997, falling from 34% to 25% (World Bank, 2001). However, in 1998 it rose again to 28% and was still higher than the 1997 level in 2000. Poverty is largely a rural phenomenon in the Philippines. The rural poor accounted for about 77% of the poor in 1997 and the agriculture, fishing and forestry sector alone for two-thirds of the poor. 13 49. In the case of the Philippines, linkages between poverty and vulnerability to natural hazards are clearly apparent, despite the fact that they have not been systematically analyzed. Rapid urban growth and lack of tenure, for instance, have forced many to live and work in high-risk areas, such as on the shores of Navotas (Box 3) or flanks of active volcanoes.18 Families may have little choice but to return to such areas post disaster even when resettlement options are available because of the importance of proximity to place of work.19 Disasters can be associated with spiraling debt, reflecting limited provision of rehabilitation assistance for repair of houses and restoration of livelihoods, poor access of lower income households to lending facilities and extremely low rates of saving.20 In the aftermath of a disaster, poor families may be forced to rely on borrowing from informal sector `five-six' moneylenders, forcing them into further poverty, and/or on remittances from relatives elsewhere in the country and overseas.21 Other post disaster coping mechanisms include diversification of livelihood strategies. Household responses to the 1997-98 El Niño event, for instance, included cultivation of different crops, such as vegetables and rootcrops and increased engagement in activities such as sewing, carpentry, construction and domestic services) as well as the withdrawal of children from school and a reduction in the quality and quantity of food intake (PCARRD/DA, 2001). Much of the increase in poverty in 1998 was attributable to the El Niño shock, reflecting its particular impact on the agricultural sector where the poor are concentrated, rather than the financial crisis, which primarily affected relatively better-off wage earners (World Bank, 2001). Balisacan (1999) comments that the withdrawal of children from school following disasters is particularly disturbing given the role education attainment plays in determining income and the implications this finding has for second-generation poverty. 18For example, marginalized farmers continue to live and work on the flanks of Mt Mayon, an active volcano, because it provides them an opportunity to produce food without secure land titles. People tend to ignore evacuation orders, typically only moving when the highest level of alert is reached (Heijmans, 2001). 19For example, a survey of 90 victims of the tragic Ormoc flood found that of the 30 respondents remaining on the riverside after the flood, 80% had returned to this location because they had no other place to live, despite the fact that two-thirds of them now recognized the dangers of living on the riverside. Although 60% were awaiting resettlement at the time of the survey, 50% indicated that they were happy to remain on the riverside, most commonly because of ease of access to their place of work and to markets (JDI/ECFA/DEVMAN, 1993). 20As a nation, the Philippines has one of the lowest domestic savings rates in East Asia and the Pacific. The lowest three deciles of the population do not save (World Bank, 2000). 21Transfers are a significant feature of the Philippine economy and have been increasing over time, accounting for an average 13% of pre-transfer household income in 1997 with 57% received from abroad (World Bank, 2001). Transfers are also highly progressive, benefiting lower per capita income households. 14 Box 3: Navotas Municipality ­ poverty and natural hazards Navotas is a small coastal municipality located on the extreme northeastern shore of Metro Manila. It regards itself as the `fishing capital of the Philippines', with about 70% of the population directly or indirectly dependent on fishing in Manila Bay and related industries. Its overall vision is to emerge as a marina city. The municipality has a population of approximately, 230,000 persons, of which an estimated 70% are below the poverty line. Physically, Navotos is comprised of a long narrow strip of land, with shorelines both along the sea to the east and the Navotos- Malabon river to the west. It has a total land area of 10.8km2 and a shoreline of 12.5 km. The municipality is elevated only 2 meters above sea level and around 50% of the total land area is vulnerable to flooding. Regular low-level flooding occurs approximately ten times a month, during high tides (1.8 meters above sea level). Two or three times a year more serious flooding, up to waist height, occurs as storms or typhoons coincide with high tides, necessitating temporary evacuation of many homes for several days. Fishing operations cannot resume for about a week, leaving many families with no source of livelihood. Over a quarter of the population are informal settlers mainly located along the shorelines. They were attracted to Navotos by the `availability' of land in the form of easements along the bay and river, and fishing opportunities that required little educational attainment. The municipality estimates that a total of 2,500 households are located along rivers and 57,500 people along the coastline. Assuming, say, that the average household contains 8 people, this implies a total shoreline population of about 77,500 people ­ that is, 77,500 people who are potentially particularly vulnerable to flooding. The municipality now has a firm policy barring new squatters from settling in the municipality but demographic growth implies that these numbers are likely to rise. Flooding is evidently exacerbated by a solid waste management problem, with considerable amounts of garbage thrown into the river, sea and drainage channels despite the existence of a fine for dumping in waterways. The construction of houses also increases the risk of flooding where dwellings along riverbanks constrict the flow of water. Navotas is a 1st class municipality. Its 2002 IRA was around P130m and total revenue, including locally raised taxes (primarily from business and property taxes), around P350m. This implies an annual 5% Local Calamity Fund of around P18m (US$350,000).22 Despite annual flooding, however, in most years Navotas spends very little of its 5% calamity fund and does not make any claims on the National Calamity Fund. Some funds are spent on stockpiling relief supplies. Post-disaster, the municipality provides only food and clothing. No assistance is forthcoming for repair of homes or livelihood support, perhaps in part to avoid encouraging people to remain on the land. Poorer households apparently scavenge for materials to repair their homes. Limited use of the Local Calamity Fund also reflects reluctance on the part of the local government to declare a calamity because it would be `bad' for business. However, expenditure on efforts to reduce structural vulnerability to flooding do account for a significant share of total LGU expenditure. The LGU estimates, for instance, that it spends around 60-70% of its total annual infrastructure budget of P15m (excluding municipal roads) on flood control. Meanwhile, the ten-year investment plan for 2003-2013 envisages a P677 million investment program including a riverbanks protection and management project (P30m); a mangrove reforestation project (P6m); and a community-based coastal protection project (P6m). Another proposed project to develop and implement alternative coastal livelihoods (P65m) could also play in important role in reducing vulnerability to natural hazards. A separate P3.9bn JICA tidal defense project is also intended to close off the Navotos-Malabon river. Also of relevance in reducing vulnerability to flooding, DPWH and the National Housing Authority (NHA) have resettled 500 families, with further resettlement planned. Fishponds account for around 40% of land area in Navotos but these are now largely unproductive, replaced by large-scale sea fishing. The municipality has therefore embarked on a program to reclaim some of this land for resettlement. It has purchased 8.4ha of land, at a total cost of P26m, which it intends to divide into 1,680 lots (35 m2 per household). Construction costs are estimated at around to P5,000 per m2 although the municipality is also exploring pre-fabricated housing as a cheaper option. It is looking to devise a 25-year payment scheme for relocated families to buy their homes and land, but estimated monthly payments of P2,000-3,000 month would be required, well beyond the means of much of those requiring resettlement. The Navotas MDCC has developed a municipal disaster plan, preparedness and response plan for typhoon and flood preparedness and an earthquake preparedness plan. However, the emphasis is on preparedness and post-disaster relief, while mitigation or rehabilitation is not covered. Earthquake risk is considered `low' because Navotas does not have any high-rise buildings and is not located on reclaimed land. 22 A copy of summary municipal accounts was obtained but this reports expenditure on a functional basis, implying that it is not possible to determine actual amounts spent on disaster-related activities other than fire fighting equipment and accessories. 15 50. Disasters can also contribute to longer-term states of poverty by delaying development of poorer areas. For instance, Balisacan et al (2002: 24) in an initial poverty mapping exercise of the Philippines report that the results from the rapid appraisal demonstrate the importance of road conditions and distances to "centers of trade" as a determinant of poverty. Yet disasters destroy roads and many, particularly feeder roads, may not be repaired for several years after a disaster. 51. Despite both the high incidence of disasters in the Philippines and the government's central objective to reduce poverty, however, efforts to reduce vulnerability to natural hazards are not systematically included as a central component of the government's poverty reduction strategy. There is recognition of the need to support the poor post disaster as victims, as for instance, reflected in post-disaster relief activities implemented by DSWD. Similarly, the government's central poverty reduction engine, the National Anti- Poverty Commission (NAPC), established in 1998, is composed of 14 Commissioners each representing the 14 basic sectors (farmers, workers, indigenous Filipinos, women, and so on), including the victims of disasters and calamity sector (VDC). There was some discussion at the time NAPC was created that vulnerability to natural hazards should be considered as a cross-cutting issue instead, in part because `victims' eventually return to other sectors (fisherfolk etc). However, there was concern that hazard vulnerability concerns could then end up being overlooked as each sector chose to focus on issues of most pressing concern to the group it represented and also that it would lose out in the allocation of budgetary resources. 52. Elsewhere, again, there appears to be little if any discussion of efforts to reduce vulnerability to natural hazards as a central component of poverty reduction initiatives. For instance, there is no explicit mention of this issue in the current Medium Term Development Plan 2001-04. Similarly, the five cross- cutting priorities of a major new poverty-reduction initiative, the Kalahi project, launched in May 2002 with NAPC as lead agency are asset reform; the human development services; employment and livelihood opportunities; participation in governance of basic sectors; and social protection and security against violence. Achievements in all of these areas could play an indirect ­ and in some cases direct ­ role in reducing vulnerability to natural hazards but reduction in vulnerability itself is not seen as a priority. 53. The World Bank's 2001 Philippines Poverty Assessment does identify climate and economic instability as the likely main sources of vulnerability and notes the role that effective public policies and regulation in areas such as watershed management, water impounding, drainage, flood control, forestry regulation, housing and zoning standards and trash collection can play in reducing impacts. It also comments that public safety nets are also needed. However, this analysis focuses primarily on drought and its inclusion is clearly directly a consequence of the severe El Niño event occurring just before the report was drafted. There is no discussion of the impact that typhoons or other natural hazards can have. 54. As a starting point in promoting greater consideration of vulnerability to natural hazards in anti- poverty strategies, more research needs to be undertaken on the linkages between poverty, vulnerability and the environment or the socio-economic impact of disasters. There appears to have been surprising little research in this area to date. In addition, disaster management needs to be more systematically mainstreamed into all aspects of national economic planning, Sectoral Plans and at a more operational level into assessments such as those on the environment. Future Vulnerability to Natural Hazards 55. A comprehensive multi-hazard loss estimation for the Philippines is beyond the scope of this study. Based on the above review of hazards and vulnerabilities, the following observations can be made regarding expected impacts of natural disasters on the Philippines. 56. The population of the Philippines was 76 million per the 2000 census, and is currently estimated to be approximately 81 million, with an average population density of 270 per sq. kilometer. Manila, the capital and largest city, is located on Luzon, the largest island. The National Capital Region, generally coinciding 16 with greater Manila, has a population of approximately 10 million, and a population density of 15,000 persons per sq. kilometer. 57. Philippine Gross National Product (GDP) is estimated for 2001 to be USD 80.8 billion, or approximately USD 1,000 per capita. While only about 12% of the population, the Metropolitan Manila area accounts for 31% of the Philippines' GDP. 58. The 2000 National Census of Population and Housing indicates building construction for the period 1977-1998 in the Philippines to have been 1.1 million buildings, with a total floor area (TFA) of 177.5 million sq. m., and an indicated value of 657 billion Pesos (average value per sq. m. is 3,700 Pesos). A recent study by JICA (2003) provides a complete building count for the Metro Manila area (see Table 5 below). Note that the 90 million sq. m. for the Metro Manila region equates to about 9 sq. m. per capita. As a rough approximation, based on 90 million sq. m. of total floor area in the Metro Manila region, the total building value of the Metro Manila region is estimated to be approximately USD 20 billion. Infrastructure would be approximately the same, for a total value at risk in the Metro Manila region of about USD 40 billion23. Table 5. Building Exposure Data, Metro Manila (Source: NSO, 2002, per JICA, 2003) LGU Area Total No. of Density Total No. of Total No. of Total Floor Nam (ha) Building (bldgs/ha) Household Habitant Area (Buildings) m2 Manila 4,130 168,528 41 333,546 1,569,581 11,475,903 Mandaluyong City 1,107 32,942 30 59,681 275,106 2,149,173 Marikina City 2,265 53,422 24 80,159 389,758 4,217,033 Pasig City 3,189 72,143 23 107,834 503,674 4,856,306 Quezon City 16,539 302,818 18 480,623 2,166,314 22,194,266 San Juan 588 11,793 20 24,604 117,392 1,253,073 Kalookan City 5,314 168,480 32 249,566 1,174,669 9,116,205 Malabon 1,597 51,694 32 74,136 336,511 2,498,690 Navotas 1,095 35,124 32 49,449 229,710 1,537,689 Valenzuela City 4,454 62,778 14 106,381 481,039 3,866,230 Las Pinas City 3,228 73,919 23 97,961 471,764 5,600,672 Makati City 3,197 50,381 16 103,980 470,304 5,031,045 Muntinlupa City 3,814 55,522 15 78,015 370,329 3,398,054 Paranaque City 4,563 72,230 16 94,108 447,901 6,084,705 Pasay City 1,779 39,968 22 78,179 354,011 2,597,026 Pateros 195 8,726 45 12,028 57,389 470,536 Taguig 2,753 65,428 24 102,722 464,552 3,825,264 MM: Total 59,809 1,325,896 22 2,132,972 9,880,004 90,171,870 59. Regarding infrastructure, the Philippines has about 200,000 km. of roads, 20% of which are paved. The Philippines produces about 41 billion kWh of electricity per annum, or about 1.4 kWh per person per day. 60. The Philippines built environment, particularly Metro Manila, is a study in contrasts, with some of the most modern high-rises in Asia within sight of some of the worst poverty in the region. Metro Manila's built 23 As much as possible, all estimates are based on Philippines replacement costs and converted to USD at current exchange rates. As appropriate, values of losses in earlier years are inflated to present values, although no attempt is made to systematically adjust the tabulated loss or other data, nor to normalize it for population. 17 environment is particularly stressed by rapid population growth, and migration of the rural poor to the metropolitan region. Population and economic pressures undoubtedly create a situation in which high hazard areas are built on, and substandard building practices proliferate. Poor or non-existent land use exacerbates, to an extreme degree, all natural hazards risks ­ houses are built in floodways, and on soils prone to failure in earthquakes and/or high rains, not to mention development on the slopes of volcanoes. The following paragraphs discuss briefly the vulnerability of typical construction and expected impacts of each of the hazards described above. 61. Earthquake risk: High quality planning and engineering is readily available in the Philippines, and is clearly employed in the construction of high-rises and some transportation infrastructure. Encouragingly, seismic retrofitting of selected highway structures was observed, indicating a responsible capability and recognition of risk in that sector. The Philippine structural engineering code is taken literally from the California code (DeGracia, 1994; also personal communications with local engineers)24, and was observed to be followed in better quality buildings in Manila. However, code enforcement for small to mid-size buildings is unknown, and many of these buildings may well evade the code. Small to moderate sized buildings have proved very vulnerable in past earthquakes, see Figure 4. 62. On an annual basis, physical damage due to earthquakes are comparable to those due to flooding ­ that is, about 10~20% of typhoon losses (see section below on typhoon risk). However, the Valley fault system, in Marikina City is capable of a magnitude 7 event, posing a major threat to the Metro Manila region. This situation is further exacerbated by the findings of the recent paleoseismic work (Nelson et al, n.d.), which found the system capable of magnitude 6-7 events occurring on average every 200~300 years, and that it has not ruptured in over 200 years (i.e., it is due!). An ongoing PHIVOLCS-JICA study is in the process of estimating the potential impacts of such an event, and their results will be available next year (N. Ikenishi, personal communication). In the interim, it is estimated that a magnitude 7 event on the Valley fault system may likely result in about damage equivalent to about 20% of total building value at risk, and 5% of infrastructure value at risk ­ that is, approximately USD 4 billion in building damage, and USD 1 billion in infrastructure damage, for a total direct property loss of about USD 5 billion, in terms of present values. This is approximately 6% of the Philippines GNP. This compares to a total loss in the 1995 Kobe earthquake of USD 100 billion, which was magnitude 7 event directly in the middle of a highly urbanized area (comparable in some ways to the Metro Manila ­ Valley fault situation). When adjusted for GNP per capita, these are comparable losses. Not accounted for in these numbers are secondary economic impacts due to the loss and disruption of Metro Manila as a center of economic activity (and Government center) for an extended period. 63. Volcanic risk: Building and infrastructure vulnerability to volcanic effects is quite high in the Philippines. In most cases, pyroclastic flows and lahars are localized, and devastating, and the only feasible mitigation is land use controls, which typically prove difficult to enforce. Review of the pattern of settlements, indicates the high degree of vulnerability to lahars in the area surrounding Mayon volcano. Given PHIVOLCS' excellent record of monitoring and prediction at Pinatubo, it is possible that future eruptions may not be accompanied by large loss of life, but the surrounding area would still be very vulnerable to economic impacts, due to ash fall and other effects. In this regard, some observations on the Pinatubo disaster are worth quoting: The Mount Pinatubo eruptions...have wrought havoc to the infrastructure and to economic activities of Central Luzon. Damage to crops, infrastructure, and personal property totaled at least 10.1 billion pesos (USD 374 million) in 1991, and an additional 1.9 billion pesos (USD 69 million) in 1992. In addition, an estimated 454 million pesos (USD 17 million) of business was foregone in 1991, as was an additional 37 million pesos (USD 1.4 million) of business in 1992....The actual destruction, coupled with the continuing threat of lahars and ash fall, has disrupted the otherwise flourishing economy of Central Luzon, slowing the region's growth momentum and altering key development activities and priorities. Major resources have 24 See JICA (2003) for a good discussion of the National Structural Code of the Philippines. 18 been diverted to relief, recovery, and prevention of further damage. The costs of caring for evacuees (including construction of evacuation camps and relocation centers) was at least 2.5 billion pesos (USD 93 million) in 1991/92, and an additional 4.2 billion pesos (USD 154 million) was spent during the same period on dikes and dams to control lahars. (Mercado et al, n.d.) 64. Volcanic eruptions are a frequent event in the Philippines, although not usually as catastrophic as Pinatubo in 1991, which caused perhaps USD 500 million in losses, in present value terms25. A major eruption of Mayon or Taal might result in comparable losses. On an annual basis, physical damage due to volcanic eruptions are perhaps 50% of those due to flooding ­ that is, about 5-10% of typhoon losses (see section below). 65. Typhoon and flood risk: Typhoon and flooding are treated together here since they are often correlated. Typhoons damage buildings via wind pressure, especially damaging to roofs, via wind-born missiles, wind- driven water penetration of building envelopes and via storm surge. Typhoons are often accompanied by heavy rains, which can cause localized or widespread flooding. Flooding damages buildings via inundations and, in fast moving waters, via velocity effects. Flooding can have significant long-term human effects due to contamination of drinking water as well as providing sites for disease vectors. Damage to infrastructure is similar to that of buildings although infrastructure is also often damaged by ground failure due to rain or flood saturated ground. 66. Typical modern homes in the Philippines consist of concrete block construction with light metal roofing .If protected from storm surge, this type of housing is not especially vulnerable to typhoon effects, with the worst effect being damage to the roofing in high winds. Only very high typhoon winds will be catastrophic to this type construction, though some damage can occur at significantly lower winds. Flooding can be more damaging, but still not catastrophic. Bamboo and wood construction in rural areas is more vulnerable to typhoon, but loss and rapid replacement of this light and inexpensive construction has historically been culturally acceptable. 67. Examination of data on typhoons reveals a gradual increase in five yearly losses (ignoring particularly high losses of 1978) until the latter part of the 1990s when they fell (and remained relatively low in the 2000, 2001 and 2002). Losses exceeding P20bn (in real 2000 prices) were experienced only once during the 1970s, twice in the 1980s and 4 times in the 1990s. 68. Recent much lower typhoon-related losses appear to have been more a matter of luck than successful mitigation. Manila, for instance, has not been hit by a severe typhoon since 1995. The marginal decline in relative importance of the agricultural sector, which averaged 22.6% of total GDP in 1990-94 compared to 20.6% in 1995-99, could also have played a minor role. Many of the poor are located in rural areas and heavily dependent on agriculture, as already noted. Gradual growth of non-agricultural sectors is implicitly reducing relative dependence on the particularly hazard-sensitive agricultural sector and thus the macroeconomic consequences of hydro-meteorological disasters. Indeed, a continued structural shift of employment and output out of agriculture is necessary both to reduce vulnerability to natural hazards and also poverty. The extent of efforts to reduce the sensitivity of the agricultural sector itself to natural hazards will also be critical in determining future vulnerability. Another, not entirely implausible partial explanation of the decline in typhoon-related losses is that with devolution disasters are also being increasingly under- reported. 69. Large portions of Metro Manila are vulnerable to flooding, as well as storm surge should a major typhoon strike Manila. Informal settlements are in or adjacent to extremely flood-prone areas, and are regularly inundated for several months each year, see Figure 5. 25Its not clear if the loss of Clark Air Base is included in these numbers. 19 70. Table 6, provides detail on one year's experience with typhoon-related and non-typhoon floods in the Philippines. Modern high rise construction in Metro Manila should not be especially vulnerable to typhoon, with the exception that wind-borne missiles may cause a significant amount of glass damage, followed by water penetration and damage. Table 6. Philippine Flood Experience in Year 2000 D etailed Loc ations B egan - D uration K nown N um ber D am age F lood Ty pe G eographic E nded (day s ) D ead D is plac ed E s tim ate F lood E x tents (m /d/y r) (U S $) (s q k m ) Northern and eas tern M anila c ity - dis tric ts 09/04/00 - 1 3 5,000 nd 90 m illiliters 1636 of C alooc an, Q uez on, M alabon 09/04/00 of rain Is land of Luz on - M anila and north; Tow ns : 05/18/00 - 6 112000 nd Tropic al 11740 V alenz uela, 21 tow ns in the provinc es of nd depres s ion B ataan, B ulac an, N ueva E c ija and Luz on Is land - P rovinc es : R iz al, Laguna, 11/03/00 - 2 43 668,000 2,600,000 Ty phoon 6000 Cagay an and Nueva V iz c ay a. M anila area, 11/04/00 rains s uburban Taguig P hilippines - S outheas tern Luz on: 10/28/00 - 3 41 100000 12,500,000 F las h floods 7000 P rovinc es : Cavite, S ors ogon, 10/30/00 triggered by Catanduanes , S am ar, A lbay . M anila and c ontinuing nearby areas heavy rains Northern P anay Is land - Capiz provinc e 12/08/00 - 3 11 55,000 nd S trong rains 2200 12/10/00 brought about by a Source: Dartmouth Flood Observatory tropic al depres s ion Source: Dartmouth Flood Observatory 71. On a regular basis, typhoon (with associated flooding) represents by far the largest loss natural hazard in the Philippines, killing about 500 people per year and causing about USD 200 million per year26, in present value terms. A major typhoon striking Metro Manila could cause larger losses than this. Warning would presumably save lives in the informal settlements near the Pasig and Bay, but would do little to prevent catastrophic economic impacts in these communities. A major typhoon striking Metro Manila might result in total losses for the region of several percent, with USD 1 billion being perhaps an upper bound. Flooding is also a regular disaster in the Philippines, although typically on a smaller and more localized basis. Based on review of the data, non-typhoon immediate flooding impacts are perhaps 10~20% of typhoon losses. 72. Overall Vulnerability. Taking into account the annual losses and probability-weighted losses for catastrophic events, such as a major earthquake in Metro Manila, as discussed above, the total direct economic impacts of typhoons, earthquakes, volcanoes and floods in the Philippines are estimated to be about USD 500 million, with a coefficient of variation of about 70%, Figure 6. These annualized losses are significantly higher than the Philippines have experienced historically, estimated to be approximately USD 300 million when (very approximately) adjusted for inflation and exchange variation. This difference is dominated by the potential for a very large loss in Manila due to earthquake. The magnitude of this loss is highly dependent on the probability of the event, which is very uncertain. Therefore, the estimates presented here have substantial uncertainty associated with them. More precise quantification of mean annual losses, and variation around this mean, is necessary. 73. Moreover, there are a number of factors which seem to imply potentially far greater losses over the medium and long term if appropriate mitigation measures are not stepped up significantly. These include: increasing urbanization and demographic growth; environmental degradation and the threat of environmental disasters; and, climate change. 26USD 200 million is estimated by taking the P4 billion per annum during the 1970-2002 period in Table 7, and doubling it to account for exchange rate variation and deflated values in earlier years. 20 74. Urbanization. The urban population accounted for 59% of the total population in 2001 and is forecast to rise to 75% by 2030. Urbanization is creating large concentrations of people and physical capital that are potentially exposed to natural hazards. For instance, rapid urbanization, low levels of income and lack of areas allocated for affordable residential developments have led to the proliferation of unplanned, informal and overcrowded settlements (NLUC, 2000), often in more hazard-prone areas as illustrated in the case of Navotas, and this trend is set to continue. The rapid growth of Metro Manila, in particular, is placing increasing numbers of people and physical assets at direct risk from potential seismic events. The National Capital Region already accounted for 30.8% of national GDP in 2001 and this figure seems set to rise. A major earthquake in Metro Manila would thus have substantial short and longer-term economic ramifications, far beyond those experienced as a consequence of the 1990 Baguio earthquake. Demographic growth and urbanization is also placing considerable pressure on the provision of basic services. Resulting shortfalls in provision ­ for example, deteriorating solid waste management and related siltation of rivers and drainage channels ­ could have direct implications for the future incidence and severity of flooding.27 Informal settlers are no longer welcomed in some cities and municipalities, as seen in the cases of Marikina and Navotas. However, others continue to allow them in, in some cases as a deliberate political, vote-raising strategy. 75. Environmental degradation. Environmental degradation is also playing a significant role in increasing the incidence of natural disasters. Demographic growth and poor land-use planning have led to the massive depletion of natural resources and destruction of the environment (World Bank, 2002). Declining forest cover, in particular is contributing to increased run-off, resulting in more frequent flash flooding, landslides and droughts. Upland communities in Panay, for instance have started to experience periods of drought since the 1980s, becoming more frequent in the 1990s, as a direct consequence of deforestation. Reduced forest cover has also left the area also increasingly exposed to typhoons, which in the past had little impact (Heijmans, 2001). Urban flooding is also on the increase. Bagiuo City, for instance, has begun to experience problems of flooding in recent years, again due to deforestation. 76. In order to help overcome these trends, it is important to address environmental degradation directly as well as consider land use plans and building codes in addressing physical vulnerability to natural hazards. Indeed, the wider importance of improving environmental management is well recognized. The World Bank (2002: 29), for example, writes that `unsustainable resource management and environmental quality problems have become a critical constraint to economic development and social cohesion, necessitating a closer linkage of environmental strategies with key economic objectives and poverty reduction efforts.' 77. Degradation of the environment is also increasing the risk of environmental disasters, another type of event with the potential for devastating results. A number of these disasters have been experienced mainly in densely populated and developed areas of the country. A recent example is the March 24, 1996 mine tailing tragedy in Marinduque, which affected 20,700 people when 1.6 million cubic meters of copper mine waste spilled into the Boac and other nearby rivers. This was by far the country's worst industrial pollution disaster. In urban areas, incessant rains brought about a tragic landslide in Cherry Hills, Antipolo on August 3, 1999 claiming 58 lives and injuring 31. The following year, on July 10, 2000 the same set of monsoon rains caused lethal landslides in the Payatas dumpsite, claiming 223 lives and affecting 680 urban poor families. The undocumented cumulative toxic loading of the agricultural soils, the gradual but progressive warming of the fishing waters, and the perennial flooding of Metro Manila are only some examples of this type of disaster that can be attributed not entirely to force majeure but to socio-economic and political forces. 78. Public concern for this type of disaster is still minimal. While early warning systems for natural disasters such as typhoons and volcanic eruptions are in place, pre-disaster alarm systems on environment- related catastrophes are largely non-existent. 27As of 2000, of an estimated 16,400 tons of waste generated by the urban population, only 2,600 tons was properly disposed in controlled dumpsites. Projections for the following 15 years indicated that an additional 12 million poor people in cities would generate an additional 4,000 more tons of daily waste (World Bank, 2000). 21 79. The complex nature of environmental disasters can be described by a matrix that categorizes their impact according to spatial and temporal scale, as shown below: Short term Long term Local A B Global C D 80. These categories can be distinguished from each other by defining "local" as that which spans the barangay to the national scale, while "global" is taken to mean the wider regional (transnational or continental) and worldwide context. "Short term" is used to refer to time scales of the order of weeks to a year, while "long term" is predicated of disaster events that can span decades or even centuries. 81. Of these categories, Type A environmental disasters (EDs) are more commonly understood and acted upon. The impacts of this type of disasters are more direct and the causes more immediate in time. Examples of this type of EDs are the annual flooding of Metro Manila caused by the congestion of garbage in the city's sewage system. Needless to say, the direct and indirect damages wrought by this ecological scourge can be significant. 82. Environmental disasters of Type B are often not as well understood as the previous type. The domain of environmental damage may be localized, as in deforestation or mangrove depletion in a particular area, but the adverse impacts could well extend into the long term. The potentially tragic, long-term consequences of this type of assault on the environment are easily blurred by short-term economic benefits. For subsistence communities in a developing country context such as those of the Philippines, what is known as "long-term" remains an abstraction vis a vis the exigencies of poverty and survival. 83. Type C EDs are relatively recent, having come to the fore of public concern only in recent decades. An example of this type of disaster is the semi-annual El Nino Southern Oscillation (ENSO), which, in the case of the Philippines, brings about massive drought lasting several months to almost a year. In other years, ENSO's twin sister, La Nina (LN), brings about intense rainfall and flooding to the country. The impacts of these phenomena are not confined to the Philippines alone, but are nearly global in extent. While ENSO and LN events are natural phenomena caused by the oscillatory warming of the eastern Pacific Ocean, there may be reason to expect that the additional warming of the sea surface due to the increasing anthropogenic release of carbon into the atmosphere could affect this cycle. Even without this supposition, it is acknowledged that ENSO or LN events already exacerbate the impacts wrought by present environmental stress. Preliminary maps showing the vulnerability of the country to both ENSO and LN are presented in Figures 7 and 8. These indicate areas such as the central and western parts of Luzon island which are most vulnerable to the onset of either event. Traditional agricultural areas such as the plains of central Luzon are thus most likely to be affected by recurring ENSO or LN events in the future. 84. While environmental disasters that are global in scope and long term in duration (i.e. Type D) can be the most controversial due to scientific uncertainties, geopolitical forces, economic and equity considerations, the potential damages associated with this type of ED can be quite extensive and protracted, cutting across critical sectors such as agriculture, human health, forest ecosystems and biodiversity. A major cause of impact in type D disasters is climate change, discussed in more detail below. 85. Climate change. The world's leading scientific body on climate change, the Intergovernmental Panel on Climate Change (IPCC), has concluded that the human release of carbon dioxide and other greenhouse gases due to industrial, agricultural, and urban activities has contributed to the increase in global surface temperature in the last 50 years. Climate change is expected to increase the frequency and intensity of weather related hazards in the longer-term. 86. Over the last century, the observed rise of surface temperatures in the Philippines follows the global trend. This is shown in Figure 9. Even while the extent of the anthropogenic contribution to this warming 22 trend continues to be debated in scientific circles, warming impact signals are already being observed in the archipelago. 87. Temperature increases in a tropical country may seem innocuous at first, but the recent crop failure in Region 2 (Summer, 2003) point to the critical importance of temperature in agricultural productivity. The observed warming of the sea during the 1997-98 ENSO event led to the bleaching of corals off Southern Luzon, which in turn is expected to adversely affect marine productivity. 88. Rainfall variability, particularly during the wet season, is on the rise making it more difficult to predict precipitation events on an inter-seasonal basis. Sea level rise (SLR), although a complex phenomenon, has been documented for various ports in the country such as Manila, Legaspi, and Davao. Such a plot is shown in Figure 10, where the SLR rate in Manila Bay is about 15 cm per decade. This plot is an example of how the confluence of many environmental factors such as silting, land reclamation, subsidence, and warmer sea surface temperatures, can lead to a rapid rise in sea level. Figure 11 shows a map of areas in Metro Manila that are likely to be inundated further by a rise in sea level from 0.3 to 2.0 meters. Increases in sea level will eventually affect freshwater supply from aquifers, storm surge vulnerability of coastal communities, and the archipelago's rich coastal resources. 89. Initial studies of the impact of climate change in the Philippines for the 21st century show that temperatures are bound to rise by as much as 2.2 degrees in the last quarter of this century. Rainfall is more difficult to predict. Many of the results modeled in the tropics confirm an increase of rainfall during the wet seasons but a marked decrease during the dry seasons. Initial maps showing the vulnerability of geographic areas to temperature and rainfall change are shown in Figures 12 and 13, respectively. These maps were generated by intersecting the human development index of the UNDP with climate model results. Although preliminary, the maps show that Mindanao is most vulnerable to projected increases in temperature while Luzon is likely to be affected the most in the face of changing rainfall patterns in this century. 90. Historically typhoon-vulnerable areas in the country (shown in Figure 14) all need to be reevaluated as typhoon trajectories, intensities, and frequency could change unexpectedly. 91. However uncertain these projected impacts are, a potential environmental disaster caused by global climate change cannot be summarily dismissed from the discourse of current disaster management. Critical areas such as food and water security, natural resources, ecosystems and public health are bound to be stressed even further, not only by the impact of largely transient natural disasters, but also by the advent of environmental disasters of varying spatial extent and duration. Conclusion 92. While data on the full social and economic impacts of disasters on the Philippines is incomplete, the existing data clearly indicates that these events have had a grave impact on the development of the country over past decades. Moreover, current trends such as urbanization, environmental degradation, and climate change will bring even more severe impacts. Poverty and disaster vulnerability are inextricably linked, and efforts to reduce poverty in the Philippines will simply not be sustainable without addressing disaster risk reduction. 93. A more comprehensive risk analysis is necessary to effectively determine the potential impacts of disaster events in the Philippines. Undertaking such an analysis will allow more informed decisions to be taken in terms of priority actions and investments to reduce these potential impacts and ensure the social and economic growth of the country in years to come. 94. The Philippines does have in place institutions and arrangements for dealing with disaster risk. The following chapter examines those arrangements. 23 Chapter 3. Disaster Management Capacity in the Philippines Background 95. Following risk identification, risk reduction mechanisms need to be developed and implemented. However, their degree of success and or applicability is dependent on the institutional structure that is in place. This Chapter examines the disaster management in the Philippines, and to an extent possible, given the limited time, propose areas for improvement. 96. The disaster management system of the Philippines originated from World War II, and was very much geared towards preparations for war. President Manuel Quezon created the Civilian Emergency Administration (Executive Order EO335) in 1941 to prepare the population in case war shifted to the Pacific, and to adopt measures to control and coordinate civilian participation to meet grave emergencies. Under EO335 a National Emergency Commission was established. At the same time, a Provincial Emergency Committee was also established in each province, which in turn had supervision and control over Municipal Emergency Committees and City Emergency Committees. 97. In 1954 Republic Act RA1190 (the Civil Defense Act) came into being, establishing the National Civil Defense Administration (NCDA), under the Office of the President. RA1190 also created civil defense councils at national and local levels. 98. Executive Order EO159 (November 1968) required the establishment of a disaster control organization by all government offices including departments, bureaus, offices, agencies, instrumentalities and political subdivisions of government, including all corporations owned and/or controlled by government. The NCDA was designated the national coordinator to oversee and implement EO159 and to report on the degree of preparedness of all government offices to the Office of the President. As a planning body, however, observers noted that NCDA was constrained due to budgetary constraints, apathy and indifference by the public and government officials alike. After a series of natural hazard impacts in 1970, including major flooding in Metro Manila, President Marcos approved a Disaster and Calamities Plan prepared by an Inter- Departmental Planning Group on Disasters and Calamities. 99. The Office of Civil Defense (OCD) was established in 1972, by Letter of Instruction LI19, which was mandated to coordinate national level activities and functions of the national government, private institutions and civic organizations. The tasks of OCD included formulating plans and policies to protect the general public; estimate the total material, manpower and fiscal requirements for carrying out a national civil defense and civil assistance program; allocate to provinces, cities, municipalities and barangays such aid in facilities, materials and funds as may be available from the national government; to coordinate programs for informing, educating and training the general public; and provide guidance to sub-national levels on planning, organization and operations for their civil defense requirements. Institutional Set-up for Disaster Management 100. Presidential Decree PD1566, promulgated on 11 June 1978 is the current basis of the Philippines' disaster management arrangements. PD1566 provides for the National Disaster Coordinating Council (NDCC) as the highest policy-making body on matters pertaining to disasters, advising the President. PD1566 stipulates: · State policy on self-reliance among local officials and their constituents in responding to emergencies and disasters · Each political and administrative subdivision to utilize all available resources in the area before requesting assistance from neighboring or higher authority 24 · Primary responsibility rests on government agencies in the affected areas in coordination with the people themselves · Government departments, bureaus and agencies to have documented plans · Planning and operation shall be done at the barangay level in an inter-agency, multi-sectoral basis to optimize resource utilization · Responsibility for leadership rests on the Provincial Governor, City/Municipal Mayor and Barangay Chairman · When an emergency covering several towns or cities occurs the Provincial Governor assumes operational control · Periodic exercises to be conducted at all levels, principally in barangays. 101. National Disaster Coordinating Council (NDCC) member agencies are responsible for carrying out respective tasks and responsibilities, which include preparedness, mitigation, response and rehabilitation. NDCC does not have a regular budget; but operates through member-agencies and their local networks, namely the regional and local disaster coordinating councils (DCCs). Members of the NDCC and their respective responsibilities are as follows: · Chairman ­ Secretary of National Defense, convenes NDCC as necessary and calls on other government agencies and private sector when need arises; · Secretary of Public Works & Highway, restores destroyed public structures, such as flood control, waterworks, roads, bridges and other vertical/horizontal facilities; provides equipment for rescue, relief and recovery; · Secretary of Transportation & Communications, restores destroyed communication and transportation facilities such as railroads and vertical structures; organizes national transport services; · Secretary of Science & Technology, (Philippine Atmospheric, Geophysical and Astronomical Services PAGASA ­ continuing watch on environmental conditions to prepare daily weather forecasts, typhoon watches and flood outlooks. Philippine Institute of Volcanology & Seismology PHIVOLCS ­ issues advisories on earthquakes, volcanic activity and tsunamis; identifies appropriate evacuation sites and organizes disaster control groups and reaction teams. Philippine Nuclear Research Institute ­ issues advisories on radioactive fallout, contamination and radiation incidents; organizes disaster control groups and reaction teams; · Secretary of Social Welfare & Development, extends relief assistance and social services to victims and provides rehabilitation; · Secretary of Agriculture, undertakes surveys in disaster-prone areas and actual disaster areas to determine extent of damage of agricultural crops, livestock and fisheries; technical assistance to disaster victims; · Secretary of Education, Culture & Sports, provides assistance in public education and campaigns regarding disaster preparedness, prevention and mitigation through integration of relevant subjects in school curriculum; makes school buildings available as evacuation centers; trains education staff in disaster preparedness; · Secretary of Finance, issues rules and regulations regarding funding by local governments of DCC requirements; with DBM issues rules and regulations on preparation of local government budget and utilization of the 2% reserve for disaster operations; · Secretary of Labor & Employment, organizes and trains Disaster Control Groups in factories and industrial complexes; provides emergency employment opportunities to disaster victims and implements industrial civil defense programs and measures; · Secretary of Trade & Industry, maintains normal level of commodity prices during emergencies and organizes disaster control groups and reaction teams in large commercial and recreational premises; · Secretary of Interior & Local Government, oversees organization of local DCCs, the establishment of Disaster Operations Canters (DOCs) of all local governments, and the training of DCC members in coordination with OCD, DSWD and other relevant agencies; 25 · Secretary of Health, provides health services during emergencies and organizes reaction teams; also issues public health warning notices; · Secretary of Environment & Natural Resources, responsible for reforestation and control of areas prone to flood, landslide, mudflow and ground subsidence; also technical assistance on environmental pollution; · Secretary of Tourism, organizes and trains disaster control groups and reaction teams in hotels, pension houses, restaurants and other tourist-oriented facilities; · Secretary of Budget & Management, releases funds required by departments for disaster operations; · Secretary of Philippine Information Agency, provides public information service through dissemination of mitigation and preparedness measures; · Secretary-General, Philippine Red Cross, conducts disaster leadership training courses, assists in DCC training at all levels; helps in provision of emergency relief; · National Housing Authority, assessment of housing requirements of displaced persons; provision of temporary housing and rebuilding of destroyed areas; · Chief of Staff, Armed Forces of the Philippines (AFP), responsible for provision of security in disaster area and assistance in reconstruction; provides transportation for relief supplies and personnel; · Director-General, National Economic Development Authority, responsible for determination and analysis of effects of disasters on socio-economic programs, and the development of damage assessment schemes; and · Administrator, Office of Civil Defense, acts as NDCC Executive Officer; coordinates activities and functions to implement policies and programs, and advises Chairman on disaster management matters). The OCD serves as the operating arm of NDCC. 101. The Chairman convenes the NDCC when necessary, calling on member agencies or the private sector when the need arises. The NDCC is replicated at regional and local levels, with each tier operating and utilizing its own resources. 102. Regional Disaster Coordinating Councils 103. (RDCCs) coordinate activities of national government agencies at the regional level. The NDCC Chair assigns the chairman of a RDCC to the post by presidential designation, although under the present arrangement the Philippine National Police (PNP) regional directors are the designated chairpersons. In autonomous regions, the Chief Executives automatically become chairmen of the RDCC. In Metro Manila, the Chairman of the Metro Manila Development Agency (MMDA) is also chairman of the Metro Manila Disaster Coordinating Council (MMDCC). 104. RDCCs have no budget of their own and can operate only through mutual coordination of the member agencies. The OCD Regional Director acts as the Executive Officer of the RDCC. The RDCC is expected to: · Establish a Regional Disaster Operations Center (RDOC) · Implement guidelines within the regions set by the NDCC · Advise local disaster coordinating councils (LDCCs) on disaster management · Submit recommendations to the NDCC as necessary 105. The term `Local Disaster Coordinating Council' appears to cover all non-national level activities, thereby embracing provincial, city/municipal and barangay system levels. Chief executives are, by law, chairmen of their respective LDCCs. The primary functions of LDCCs are detailed in an NDCC document Calamities and Disaster Preparedness, the details of which are: 26 106. Provincial Disaster Coordinating Council (PDCC) · Establish a Provincial Disaster Operations Center (PDOC) · Coordinate disaster operations within the province from the PDOC · Implement within the province guidelines set out by the RDCC · Advise City/Municipal and Barangay DCCs regarding disaster management · Submit recommendations to the RDCC as necessary 107. City/Municipal Disaster Coordinating Council (CMDCC) · Establish a City/Municipal Disaster Operations Center and coordinate disaster operations from it · Implement guidelines set by the PDCC within the city/municipality · Advise Barangay Disaster Coordinating Councils regarding disaster management 108. Barangay Disaster Coordinating Council (BDCC) · Establish a Barangay Disaster Operations Center (BDOC) and coordinate disaster operations from it · Implement guidelines established by the CMDCC · Advise BDCC member son disaster management. 109. A summary diagram of the institutional arrangements for disaster management is provided below: President of the RoP and Cabinet National Disaster Coordinating Council (NDCC) Office of Civil Defense Regional Disaster Coordinating Councils (16) Metro Manila Coordinating Council Provincial Disaster Coordinating Councils (76) Highly Urbanized Cities Coordinating Councils City/Municipal Disaster Coordinating Councils (about 1400+) Barangay Disaster Coordinating Councils (44,000) 27 Funding Support for Disaster Management 110. As mentioned in the previous chapter on economic and fiscal impacts of disasters, the Philippines Government provides for annual appropriations of calamity funds for emergency relief and rehabilitation efforts. National government28 111. Government expenditure on disaster related activities is provided from a number of budget lines held by various government departments and various levels of government. 112. National calamity fund At the national level, the government's primary fund for responding to disasters is the National Calamity Fund, which is intended to supplement and complement Local Calamity Funds. Priority in use is given: firstly, for urgent and emergency relief operations and emergency repair and rehabilitation of vital public infrastructures and lifelines damaged by calamities occurring within the budget year; secondly, for the repair, rehabilitation and reconstruction of other damaged public infrastructures; and thirdly, for pre-disaster activities outside the regular budgets of line agencies and proposed capital expenditures for pre-disaster operation. 113. Part of the annual appropriation is immediately, a priori, allocated to the Quick Response Fund, with allocations within that to particular agencies. Under the 2002 budget, for instance, 37.5% of the total P800bn calamity fund was pre-allocated as QRF. Some 12.5% of the total National Calamity Fund was pre-allocated as QRF to the DSWD, 7.5% to OCD, 10% to the DPWH and 7.5% to DND. QRFs are intended as rapid stand-by funds for use in the immediate aftermath of a disaster. The NDCC recommends the release of the NCF, including QRFs. 114. Most applications for National Calamity Funds come from 4th, 5th and 6th class LGUs, who have relatively little income and so rely only on small Local Calamity Funds.29 LGUs have to put up a local counterpart fund ranging from 50% of the total project cost for 1st class LGUs to 30% for 4th class LGUs. Those LGUs in the 5th and 6th classes are exempt but no single request is allowed to exceed P3bn a year. 115. According to PD 477, two per cent of the budgetary reserve should be allocated to the National Calamity Fund. In reality, appropriations to the National Calamity Fund have been far lower and, moreover, have gradually declined in recent years, falling from as high as P2.8 billion in 1996 to only P800 million in 2002 and P700 million in 2003 ­ a drop of over 80% in real terms (Table 3). This decline reflects increasing fiscal difficulties combined with a lower incidence of major disasters in recent years, allowing the government to justify falling appropriations. The DBM actually reported a surplus on the calamity fund budget in 2002. However, the NDCC maintains that resources are far lower than required and that this has led to a backlog of claims on the National Calamity Fund. For instance, part of the 2003 National Calamity Fund will be used to meet requests received for damage incurred in 1998 and 1999. In March 2003, use of the calamity fund was extended to cover insurgency, including the conflict in Mindanao, and acts of terrorism, placing additional pressures on already limited resources. 116. Evidence on the use of resources confirms that calamity funds are extremely stretched and able to meet only a small proportion of relief and rehabilitation costs. Between 1996 and 2000, total disaster-related damage amounted to P 49.2 bn (in real 2000 prices) according to the NDCC whilst total expenditure under the calamity fund totaled P 8.9 bn, or just P 7.5 bn excluding expenditure on mitigation and preparedness. 28There is also some off-budget external assistance both for mitigation and preparedness and post-disaster response. However, consolidated data on this assistance could not be readily obtained for the purposes of this report. 29The DILG/DOF classify LGUs in descending order from class 1 to 6 based on average annual total income. 28 117. Other sources of post-disaster funding Other sources of funding have also been drawn upon ­ and sometimes specifically created - for use in responding to more severe events, beyond the capacity of the National Calamity Fund. These include: · Special rehabilitation funds ­ following both the 1990 Baguio earthquake and the 1991 eruption of Mt Pinatubo, special relief and rehabilitation funds were established to fund reconstruction and rehabilitation operations. · Property replacement fund ­ this used to operate as a sinking fund for the restoration of national government buildings, equipment and transport vehicles damaged by fire and national calamities. However, its role was reduced to high-risk public properties (primarily aircraft and ships) only in the late 1990s. · Line agency standby funds - some government agencies have separate standby funds under their own budgets for stockpiling relief supplies and so forth. For instance, the DSWD's standby fund stood at P200m in 2002. · Reserves Control Account ­ this account, comprising 5% of maintenance expenditure and 5% of capital outlay of each government department, is intended to meet unforeseen requirements, in some cases arising as a consequence of disasters. It was drawn upon, for example, as part of the initial government response to the Mt. Pinatubo eruption. · Allocation of Congressmen - in responding to the 1998 El Niño event, each of the country's 224 congressmen allocated P5m for construction of deep-water wells, water tubes, procurement of seeds and other measures. · Presidential Social Fund ­ this is sometimes used by the DA to fund post-disaster operations. 118. Following the 1998 drought, further relief and rehabilitation funding was also allocated from the Agricultural Production Fund and from the National Government lump sum whilst resources from the Poverty Alleviation Fund were used to fund food distribution. NDCC's own costs are separately funded out of the general appropriations. 119. Aggregate data on total disaster-related expenditure under these various budget heads is not available. However, in some years expenditure may be substantial, placing significant pressure on other areas of government spending. For instance, in response to the Mt Pinatubo eruption the GoP released a total amount of P32.3 billion (in nominal terms) over five years, of which P2.99bn was for various foreign-assisted projects relating to infrastructure. This was equivalent to 1.5% (excluding foreign-assisted projects) of total nominal expenditure over the period 1991 to 1995; and 4.7% of expenditure excluding interest payments on domestic and external debt and wages and salaries.30 120. Additional rehabilitation spending is drawn from existing operational budgets of relevant agencies, most obviously the DPWH, DA and DENR on a regular basis. DPWH, for instance, has a rehabilitation fund within its annual budgetary appropriation. Upon declaration of a calamity, concerned national agencies and LGUs are permitted to program or reprogram funds for the repair and safety upgrading of public infrastructures and facilities and reallocations occur on an annual basis. However, such reallocations are largely unrecorded and thus difficult to capture without detailed examination of budgetary operations as they occur internally, within budget heads. Some surplus funding can also currently be `borrowed' from projects that are being implemented more slowly than anticipated in determining budgetary appropriations, although the DBM is trying to streamline this process and only issue funding for projects that are ready. There may be substantial additional unrecorded reallocations within the recurrent budget, relating to re-deployment in kind ­ of government staff, vehicles and equipment, supplies of drugs and other items and so forth. 30Figures cannot be reported in real, rather than nominal, terms because only the total, rather than annual, figure on expenditure in response to the Mt Pinatubo eruption was available. 29 121. Ideally, any reallocations of budgetary resources post-disaster should occur through a formal process, in the context of a careful strategic review. Four conditions are required to ensure that this process is rational and cost-effective: a clearly defined and applied policy framework; an already established system of prioritization for investment projects; up-to-date and reliable information on the current demand for and availability of resources; and a considered assessment of the broader macroeconomic impact of the event (Benson and Clay, 2003). In the case of the Philippines (and many other countries), however, few if any of these conditions hold. 122. In the event of more extreme events, total government expenditure may also be reduced to compensate for lower revenue, as in 1998 when the government was forced to implement cuts of up to 15% on the 1998 appropriations, equivalent to 38% of non-personnel appropriations (World Bank, 1998). The government was apparently successful in protecting expenditure in the social sectors but investment and maintenance expenses were squeezed (ibid). 123. Finally, longer-term rehabilitation activities for specific purposes may be formally incorporated into the annual budgetary process, at least implying an opportunity for careful assessment of the relative economic returns to different investment projects and efficient allocation of resources. For example, in the mid-1990s it was ruled that any further expenditure relating to the eruption of Mt Pinatubo and related lahar mitigation measures should come out of regular budgetary appropriations to relevant government departments. 124. Mitigation and preparedness expenditure Although disaster mitigation and preparedness is a devolved responsibility, national government maintains responsibility for provision of national infrastructure and provision of services and information in support of LGU activities. However, mitigation and preparedness responsibilities are contained within overall budgetary allocations to relevant departments and total mitigation and preparedness expenditure across the different government agencies is not reported, with the exception of a small level of expenditure under the calamity fund (Table 7). Some could be relatively easily identified regardless, for instance expenditure on flood control, but aggregating total spending in this area is more complicated. In some cases, for instance, risk reduction measure may be included as just a small component of a project, such as hazard proofing of buildings as part of a school building project. In others, there may be indirect, rather than direct, risk reduction benefits ­ for instance, in the area of environmental management or poverty reduction or even operations and maintenance of infrastructure.31 Nevertheless, lack of readily accessible data even on dedicated mitigation and preparedness projects has important policy implications, making it difficult to assess the overall adequacy of current spending in this area relative to the scale of disaster losses. Better documentation of expenditure is needed. This area of work could be the work of a follow-on study that would aim to understand whether there are increases in Government expenditures to finance relief efforts, which, if it were the case, then there are other implications and costs to fiscal and debt management. 31Badly maintained infrastructure can increase vulnerability to natural hazards, as well as impose larger post-disaster costs, even where original standards of work were high. 30 Table 7 Philippines - annual appropriations and expenditure under the National Calamity Fund, 1996 - 2002 (real 2000 million peso) Actual Expenditure Appropriation Relief Rehabilitation Mitigation Preparedness Others Total (QRF) 1990 2,082 n/a n/a n/a n/a n/a 2,078 1991 1,986 n/a n/a n/a n/a n/a 214 1992 2,575 n/a n/a n/a n/a n/a n/a 1993 2,755 n/a n/a n/a n/a n/a n/a 1994 3,131 n/a n/a n/a n/a n/a n/a 1995 2,911 n/a n/a n/a n/a n/a n/a 1996 3,786 160 2,467 116 119 547 3,408 1997 2,546 33 371 150 35 611 1,200 1998 2,305 240 1,251 124 201 478 2,293 1999 1,067 23 381 550 70 3 1,027 2000 999 238 452 55 0 253 998 2001 936 139 302 252 0 239 932 2002 711 92 87 112 0 259 549 Total 12,349 925 5,310 1,359 424 2,389 10,407 Source: NDCC Based on GDP at market price deflator as GDP deflator is not available for the 1990s. Local Government Units 125. The 1991 Local Government Code established three levels of local government ­ provinces, cities and municipalities and barangays ­ and assigned them major responsibilities in the areas of agriculture, environment and natural resources management, health, public works and highways (World Bank, 2000). A considerable share of public national resources, amounting to a total 40% of internal revenue collections, is now allocated to local government to fund devolved functions.32 LGUs have also been given substantial revenue-raising capabilities, allowing them to levy certain taxes, fees and other charges for their own use, including through business and real property taxes. 126. Post disaster response. As part of the 1991 program of devolution, LGUs are mandated to set 5% of their estimated revenue from regular sources aside as an annual lump sum appropriation for use in relief, rehabilitation, reconstruction and other works or services in connection with calamities occurring during the budget year in the LGU or other LGUs affected by a disaster or calamity. Funds are released within 24 hours of the declaration of a disaster.33 Any resources remaining at the end of the fiscal year revert to the unappropriated surplus for reappropriation the following year. 32 Actual allocations are based on collections in the third year preceding the current budgetary year. Of this, 22% is allocated to provinces, 23% to cities, 34% to municipalities and 20% to barangays. Allocations to individual LGUs are then calculated according to population (50%) and land area (25%), with the remaining 25% shared equally. 33Local Sanngunians are empowered to declare a state of calamity, and thus access Local Calamity Funds, if at least two or more of the following conditions apply for at least four days: - (a) 20% of the population are affected and in need of assistance, or 20% of the dwelling units have been destroyed; and (b) 40% of the means of livelihood are destroyed (e.g. bancas, fishing boats, vehicles), major roads and bridges are destroyed and impassable and there is widespread destruction of fishponds, crops, poultry and livestock and other agricultural products. Lifelines such as electricity, potable water system, transport system, communications, and other related systems are disrupted. If two or more barangays are affected, an entire municipality or city can be declared under a state of calamity. If two or more municipalities or cities are affected the entire province or a portion thereof can be declared under a state of calamity. 31 127. Information on the actual use of the calamity funds, including unused funds remaining at the end of a year, can only be readily accessed directly from individual LGUs. LGUs are not required to report use of calamity funds to the NDCC, DBM or any other part of national government.34 In theory, LGUs are supposed to submit a utilization report on their Local Calamity Funds to the NDCC if they want to apply for National Calamity Funds but in practice typically fail to do so. 128. However, an apparently substantial part ­ perhaps as much as 50% or more - of Local Calamity Funds for the country as a whole go unused each year. By the same token, when a disaster does occur Local Calamity Fund resources may be far from adequate, requiring additional funds. A proportion of these are likely to be met from national, and perhaps international, sources. However, LGUs may be left with some residual demands on their own resources. In such situations, it would appear that LGUs are more likely to reallocate resources or postpone rehabilitation - or in some cases even ignore some rehabilitation needs (e.g., repair of feeder roads) ­ rather than increase short-term expenditure as LGUs normally keep their fiscal position in balance, cutting O&M expenditure, and in more serious situations instituting cross-the-board cuts, when resources are constrained (World Bank, 2003).35 National government restrictions on rates of disbursement of LGU revenues can also be lifted in the event of a natural disaster.36 However, disbursements can only be made for purposes and amounts included in the approved annual budget, effectively implying little official flexibility in the re-allocation of resources following a natural disaster. LGUs apparently rarely borrow to finance their current expenditure but arrangements are also sometimes made with local businesses to provide credit facilities for the procurement of relief supplies when scheduled releases of the IRA are delayed. 129. An alternative option that could be explored is the establishment of a Local Calamity Fund pool, with individual LGUs placing their annual calamity fund into a collective pool that is rolled over between years. Assuming all LGUs do, indeed, earmark funds for calamities at the start of each year and that these funds are carefully ring-fenced, nationally a considerable volume of resources remain unutilized each year. This is not an economically optimal use of resources. By making budgetary allocations for use in the event of disasters, LGUs are raising the general level of budget expenditure and/or squeezing allocations to other projects and programs (Benson and Clay, 2003). If these resources could be held in a collective pool instead, then criteria could be established for ensuring that a greater share of funds are utilized, at the same time also providing LGUs access to larger levels of funding in the event of a disaster. The latter would help strengthen LGU budgeting and planning exercises by reducing unplanned reallocations of resources. 130. Mitigation and preparedness expenditure. There are no mandatory requirements relating to LGU expenditure on mitigation and preparedness nor dedicated funding. Currently, Local Calamity Funds cannot be used for mitigation and preparedness, although an amendment allowing such use has been awaiting Congress approval for six years. Instead, LGUs are expected to finance mitigation and preparedness activities out of other budget heads. Some, primarily richer, LGUs have spent part of their 20% development fund on disaster mitigation. In other cases, mitigation and preparedness activities have been funded, for instance, from the MSWD budget (for emergency equipment); by utilizing unexpended balances on other projects; via the establishment of a Disaster Administrative Fund for the MDCC; and from the Local Calamity Fund itself via creative labeling of mitigation activities as rehabilitation (e.g., for a drainage project) (Luna, 2000). Some have even established disaster management offices, with a dedicated budget for preparedness, training, 34This appears to be symptomatic of a broader weakness in the monitoring of LGU financial performance and a lack of financial transparency, as reported by the World Bank (2003). 35Any overdraft outstanding at the end of a fiscal year must be met from the first collections of the following year's revenue (Nolledo, 1991), again implying a disincentive to go into deficit. 36Under normal circumstances, total disbursements must not exceed 50 percent of the uncollected estimated revenue for that year. 32 equipment and other expenditures.37 However, more generally there is apparently inadequate support for mitigation and preparedness (ibid). 131. Again, as at the level of national level, there is a case for seeking improved reporting on LGU mitigation and preparedness expenditure. 132. Broad LGU performance in the area of disaster management There has been no systematic evaluation of LGU performance in delivering devolved functions (World Bank, 2001). There is `considerable anecdotal evidence of decentralization having stimulated local authorities to improve service delivery' but some have failed to assume a number of the functions devolved to them and questions have also been raised about their continuing capacity to deliver certain key services, such as agricultural extension, primary health and local roads maintenance (World Bank, 2001: 33). Moreover, `with the role of oversight and review of local government decisions to other levels of local government (rather than to the national level...), little assurance is there that they actually assume their responsibilities under the law' (World Bank, 2000: 84). 133. In the case of disaster management, available evidence suggests very mixed performance. LGUs are mandated to draw up disaster management plans, for instance, but in practice may not, particularly in the case of 4th, 5th and 6th class LGUs. Even where plans do exist, they often focus largely on relief and rescue operations (CNDR/USAID, 2000), as illustrated in the case of Navotos. LGU development plans are also meant to cover disaster management, but, again, vary widely in the extent to which they do. Some plans even contain a specific chapter on disaster management but coverage in others is very limited. 134. There are also issues of continuity, as LGU agendas are to a large extent driven by incumbent administrations, while a sizeable proportion of LGU posts are political.38 Mayoral terms of office are for three years, with a maximum service of three terms. Thus, political commitment on the part of LGUs to particular issues can change abruptly, as evidenced in the case of disaster management by experience in implementing the CNDR Bayanihan project (Luna, 2000). In the case of Marikina City (Box 5) the current mayor estimates that the strong disaster management policy, initially embarked on under his leadership a decade ago, has become institutionalized and will continue beyond the end of his term of office. However, this case is exceptional. 135. Devolution has also created certain problems relating to economy of scale, with many LGUs yet to seek cooperative solutions. For example, there are concerns about limited emergency services capacity. An examination of selected LGUs found that few of them had firefighting equipment or heavy-duty vehicles for evacuation, rescue and clearance purposes (CNDR/USAID, 2000).39 Similarly, inter-municipality collaboration is required to develop management strategies for spatially inter-related hazards such as flooding. Neighboring LGUs would need to work together on such issues, perhaps supported by the development of legislative procedures. LGUs could also consider transferring the responsibility of particular services to other LGUs, helping those who invest in emergency equipment in recovering costs. Marikina City, for instance, has been an exception in investing in emergency equipment and takes pride in the fact that it has been able to use these resources to support other LGUs, but so far provides this service free of charge. 37For instance, Albay (1st Class), Bulacan (1st Class), Negros (Occidental) (1st Class), Isabela (1st Class), San Fernando (3rd Class) and Marikina (1st Class) (Luna, 2000). 38The number of political appointees in LGUs is almost as large as in the national executive (World Bank, 2001). 39An exception, Marikina City takes pride in the fact that it has been able to use its emergency equipment to support calls for assistance post disaster in other LGUs, free of charge. 33 Box 5: Marikina City ­ reaping the benefits of flood mitigation policies The city of Marikina is located along a valley in eastern Metro Manila, with a total land area of 21.5km2, It is comprised of 14 barangays, of which 12 lie along the Marikina river, a tributary of the Pasig river. In 2003, Marikina's population stood at approximately 480,000-500,000 persons, of which only an estimated 5% were below the poverty line, far below the national average. It is primarily a residential area, with around half of the population working outside the city. Marikina's broad vision is to become a `little Singapore'. Marikina has placed considerable emphasis on mitigation and preparedness since its current mayor was first elected in 1992. At the time of his election, Marikina faced considerable environmental problems, with limited garbage collection, dumping of waste on open land and water ways, considerable informal riverbank settlements (accounting for around a quarter of the city's total population) and an only partially functioning drainage system. In 1993, however, the `Save the Marikina river' program was launched. Existing commercial buildings and informal settlers were removed from easements, with 109 ha of land and a community mortgage program provided for the re-settlement of 10,500 families. A strict refuse collection policy was also introduced, with garbage only allowed out of houses on the day and at the time of collection and a penalty of P2,000 introduced for non-compliance. In addition, Marikina City decided to undertake by itself the task of river dredging ­ a national government responsibility - as national government funding was not available. This program has been hugely successful. Areas prone to flooding have been significantly reduced, from 20% to 5%, and floodwaters recede faster now. Property values have risen around 10-fold in areas previously vulnerable to flooding, with indirect benefits for the LGU in higher property tax revenues. There are no longer any squatters located along the riverbanks and Marikina City is continuing to move more homes onto higher ground. The city lost some P10m in taxes and revenues by removing slaughterhouses and other industry from the side of the river but recreational activities have instead been established. Marikina has been judged as the cleanest city in the country. However, problems still remain relating to the inflow of garbage from upstream. Marikina also faces considerable seismic risk as it is located on a major fault, its namesake the Marikina fault. It has an ongoing program to improve preparedness but reports that it has not undertaken many measures to reduce structural vulnerability. Key Issues Pertaining to the Disaster Management System 136. Current global thinking and practice on disaster risk management has witnessed a change in emphasis from a reactive, preparedness/post-disaster recovery approach to a more proactive, risk reduction approach to managing disasters. In this context, achieving sustainable hazard mitigation through risk reduction and creating community resilience are key concepts in contemporary disaster management. Current thinking holds that resilience in communities comes from reducing vulnerability to hazard impact and building capacity to deal with them when they occur. Hence, disaster resilient communities have five distinguishing characteristics: · They have a capacity to coordinate response and recovery, wherein responsibilities and contingencies have been debated, agreed on and allocated. · They have taken measures over time to reduce vulnerability through the application of risk management approaches. · These mechanisms are undertaken within a whole-of government (local, regional, national) and a whole-of-nation (public, private) approach. · These actions are underpinned by a strategic framework and associated enabling legislation. · Local communities are the fundamental building blocks. 137. Effective disaster management can ensure that communities have systems to reduce the probability and the impact of emergency events; ensure that sound decisions on investments are made in physical and social infrastructure; and, ensure efficient and effective resource allocation by reducing the need to expend resources on emergency response, reconstruction, and rehabilitation. 138. While there are positive aspects to the Philippines; disaster management system, its full potential has yet to be realized. Improving the current system would make a very significant contribution to poverty 34 reduction efforts and the development of the country. To achieve this goal, four main areas should be addressed: · emphasis on proactive, risk management approach · emphasis on bottom-up, decentralized approach · consolidation of current system · improved coordination among actors Emphasize proactive disaster risk management 139. The disaster management system in the Philippines is based on an outdated decree (dating back more than 20 years), and would greatly benefit from integrating lessons learned in the country around the globe over the past decades. In general, there is a need to recognize natural hazards as a potential obstacle to long- term sustainable development. Reflecting this, there is a widespread emphasis on post-disaster relief and short-term preparedness (forecasting, evacuation, etc.) rather than mitigation or post-disaster support for economic recovery, such as livelihood regeneration or tax breaks to affected businesses. For instance, the macroeconomic goals outlined in each of a succession of national government medium term development plans dating back to the early 1990s have failed to include any mention of the need to reduce vulnerability to natural hazards beyond specific measures along the lines of effective and adequate relief, rehabilitation and preparedness, improved government response capacity and various flood control measures. LGU medium term development plans also typically focus on flood control and post-disaster relief concerns. Within the private sector, some companies have provided disaster training and developed preparedness plans. Some also undertake some element of mitigation, such as dredging of rivers and waterways in the vicinity of their property. However, the design and implementation of comprehensive hazard management policies are not widely regarded as good business practice. 140. This focus on short-term preparedness and humanitarian relief partly links back to budgetary constraints on the part of government. For instance, the DA reports that post-disaster seed distribution programs are typically limited due to funding constraints and, in accordance with national food security considerations, tend to focus solely on rice and corn, the country's two principal food crops. The DA does not have any specific strategies or programs for supporting farmers in reducing vulnerability to natural hazards, beyond irrigation programs. 141. While a more proactive, risk management approach in the Philippines is in its beginning stages, there are examples of efforts to integrate risk reduction into development. Box 5 on Marikina City provides one example of good practice, and there are others. In 1991 the Philippines Government started a process to integrate disaster mitigation and sustainable development issues within the Medium Term Philippine Development Plans, under the Development Sector Administration. Within this framework, local governments are required to integrate their disaster management plans into their respective local development plans. Presidential Decree PD1096 (the National Building Code) specifies minimum requirements and standards on building designs to protect against fires and natural disaster. To improve the effectiveness of the Code, some study on the code's adequacy and enforcement, particularly in the area of retrofitting for earthquakes is needed. 142. Local Government Code 1991 contains provisions for disaster mitigation. A provision of the Code stipulates the increase of budgetary requirements earmarked for disaster preparedness from 2% to 5%. This 5% appropriation, which is estimated from the revenue from regular sources must be set aside for relief, rehabilitation, reconstruction or other works and services in connection with calamities occurring during the budget year. While revisions to this fund have been undertaken, it is still only available for post-impact activities within the specific local government unit affected by impact. 35 143. Since 2000, at least two proposals for new legislation have been submitted: a second draft bill is currently within the legislature, although this one appears to have stalled. Emphasize more bottom-up approach with participation of all stakeholders 144. The Philippines system tends to be more centralized and top-down administrative system than a community-based system and there are few incentives for local level initiatives. For example, under NDCC arrangements, while each local level can `submit recommendations as necessary', all are required to adhere to guidelines established by the level above. 145. In practice, the Philippine disaster management system is a disaster response system at local levels. While there are measures to link disaster with development, this connection is very loose: it needs strengthening. This is necessary to encourage some local government units (LGUs) to initiate sustainable development practices just as much as it is to encourage other LGUs to continue along this path, often against bureaucratic obstacles. With very few exceptions, local level systems are response-driven ­there is no obvious effort to initiate proactive hazard management/risk reduction coordination. Providing incentives for proactivity that reinforces the goals of integrated risk management, which itself is linked directly to sustainable development, is desirable. 146. Within this, the private sector needs to be directly brought into the disaster management policy arena. There are several models available internationally to show how this could be undertaken. The recent Employers Confederation of the Philippines forum held in Manila (and reported in the Business World newspaper 23-4 May, 2003), a senior spokesperson was reported as stating that the Confederation called for stronger partnerships with government in a bid to move the country forward: involving the business community in IDRM will result in major potential enhancements. 147. There is widespread acknowledgement that the current legislation, passed in 1978, is outdated and is constraining progress. Assuming the draft Bill is passed (it is titled the Comprehensive Disaster Management Bill), many issues discussed in this report might be remedied. The Bill passed its First Reading in 2002, but now appears to have been sidelined ­ it is apparently not a pressing issue for the Cabinet. While this is not unusual, it is worth noting that an earlier draft replacement Bill `died two years ago'. Updated legislation is an imperative if the nation is to move forward, but to do so the new legislation needs to focus on IDRM ­ integrated disaster risk management. What this implies is that the new legislation needs to redefine the duties of central and local governments, directly bring the private sector, specifically utilities, into both emergency management strategic decision-making and operational contexts, and promote the sustainable management of hazards and risks in a way that contributes to the well being and safety of the public and property. There are examples of this type of legislation for the Philippines to consider. As such, further examination of the Comprehensive Disaster Management Bill is warranted to identify gaps, assess more clearly what can be realistically expected to be accomplished and move the Bill forward. 148. With reference to current legislation, discussions with key personnel revealed that a key issue requiring adjustment is the provision of incentives or deterrents to encourage compliance. Consolidation of current system 149. Related to promoting a proactive risk reduction approach to disaster management, is the issue of developing a strategic, holistic framework. Many stakeholders interviewed in the country expressed that the current system is still very much ad hoc. The development of a strategic framework is widely regarded as being necessary. 150. This system currently comprises six levels encompassing almost 46,000 separate administrative units. This is a difficult system to operate given the top-down approach that characterizes Philippines disaster management. Merging tiers and some local groups at the barangay level into clusters (perhaps aligned to 36 hazardscape clusters, such as floodplains or fault traces) might warrant consideration: this has been successfully undertaken recently in New Zealand for the purpose of disaster risk management at both regional and local government levels. 151. The issue of consolidation at the barangay level seems especially important for disaster management since barangay captains are obliged to make certain their community develops preparedness plans. However, 10 years after the initiation of the Local Government Code, the barangay system has not been sufficiently empowered to achieve its tasks. The question has to be asked whether the barangay is too small and/or inherently weak to be the basic building block to be the planning and implementing unit for government policies (this is an issue that goes beyond disaster management). 152. Another area that requires consolidation pertains to the education and training of local officials in disaster management. Currently, training programs are not compulsory for personnel designated with disaster management tasks. Moreover, the ODC is unable to provide systematic training for local government unit (LGU) personnel primarily because the EMIP, linked to the Armed Forces of the Philippines, has a first priority to train AFP staff; it does not have the resources to extend far beyond this group. Consideration should be given to expanding existing facilities or alternatively to create a new facility for LGUs. The latter suggestion might be better placed within DLGI, since it too has a training wing, the Local Government Academy. If this facility does not have existing disaster-related courses, perhaps consideration could be given as to how they might be developed: there are many relevant programs and trainers internationally that could provide insights. In similar manner, OCD doesn't know how the training material it does provide LGUs is being used. LGUs could also benefit from participating in regional initiatives in disaster management which would help them to share lessons of experience and seek external collaboration. Improved coordination among actors 153. The need for greater coordination is well recognized by almost all the agencies involved in disaster management. The type of coordination needed is less a top-down oversight function but rather a proactive, participatory and encouraging orientation type of function. Greater organizational, management and task synchronization are regarded as prerequisites at both national and international levels. These factors extend across-the-board, at all levels of the Philippines disaster management institutional arrangements. Two examples can illustrate the diversity: · The issue of uncoordinated and non-cumulative technical studies resulting in `project findings often going nowhere after initial funds dry up' cited above is a case in point at the international level. · The potential for national-local conflicts over hazard management regimes is likely to cause problems unless a coordinated effort is undertaken. A cited case relates to river flood management: rivers are managed nationally by DPWH, although they flow through LGU jurisdictions. In Metro Manila this responsibility was recently handed over to MMDA for river control and flood management, but it claims it does not have the capacity to do this, so it is staying with DPWH for the present. This situation has put flood management in Manila into a state of limbo. At the same time, individual LGUs in Metro Manila wanting to develop flood management regimes within their jurisdiction are wrangling with neighboring LGUs, which do not regard the issue as a priority. Conclusion 154. Experience has shown that for nations with a decentralized political structure, diverse local arrangements need to be accommodated within an integrated national framework, which can be managed to accommodate changes in knowledge, the community, government and the hazard environment. Legislation alone is insufficient to manage such a framework. Experience also demonstrates the value of building on 37 existing institutions and placing these expectations within legislation, thereby underpinning a fundamental tenet that everyone has a role in managing public safety. Since risk is defined by the hazardscape, communities with significant common threats (such as a floodplain, fault line or tsunami-prone coastline) should function as an enlarged unit (a community of interest) and develop an operational planning framework as a cluster or group. Local community teams (CBOs, NGOs, NFPs, private sector) supported by awareness and training programs coordinated by regional initiatives are linked through a national-level planning and information framework. 155. In short, an effective disaster risk management system comprises: · The development of a national strategy that provides statements on the strategic direction for emergency management, establishes a platform and framework to ensure national and/or sub- national capability for managing an emergency of national significance by setting out expectations, targets and objectives of major actors and agencies. · Legitimizing agreed strategic outcomes through the enactment of enabling legislation. · Development of an operational planning framework, including resource requirements (inputs), for all actors and agencies that is fully articulated and reciprocal so that each actor/agency knows how they fit into the wider operational frame, what is expected of them, and what they can expect from others in return. 156. The Philippines has the potential develop such a system. International support should be directed at reinforcing such arrangements. 38 Chapter 4. Risk Sharing and Transfer 157. The third and final step in a comprehensive risk management strategy is risk sharing and transfer, including financing. Once the hazards have been accurately identified, the vulnerabilities to those hazards understood, and steps taken to reduce the risk of losses caused by hazard events, ex ante arrangements should be made for transferring, sharing, and/or financing the risk that cannot be further reduced or eliminated in order to protect local communities and the country from the financial and economic impacts of disasters. At the household level, such coping strategies may include personal savings, microfinance instruments, temporal financing and or informal risk sharing mechanisms developed by local communities in case of emergencies. At higher levels, these arrangements can take the form of calamity funds, insurance and reinsurance instruments, contingency loans, catastrophe bonds, etc. In most industrialized countries, for example, the private insurance sector contributes significant funding for post-disaster reconstruction. Private insurance markets in emerging economies are less developed, leaving the state and individual to bear the brunt of the costs of disasters. Existing calamity funds are commonly not sufficient to support post-disaster reconstruction, resulting in a diversion of funds aimed at other development goals towards relief and reconstruction. 158. Risk sharing and transfer mechanisms can also enhance the other components of the risk management strategy (risk identification and risk reduction) by providing important incentives for the development of a better understanding of risk, and investments to reduce risk. This chapter takes a rapid look at the state of the insurance sector in the Philippines as a key potential mechanism for sharing and transferring the cost of risk. In addition, the chapter also examines the potential of using contingency funding as a flexible financing mechanism that can be drawn upon quickly and on an as needed basis. Annex 2, International Experience with National Catastrophe Risk Management Programs presents international examples where some of the mechanisms discussed in the chapter have been used. Catastrophe Insurance in the Philippines 159. Despite significant exposure of the economy to natural disasters, insurance penetration in the Philippines is very low. In 2000, the non-life insurance premium collected was USD 458 million, which amounts to only 0.6 percent of GDP. It is estimated that less than 10% of all residential property policies cover natural perils. The total number of fire policies with endorsements for catastrophic perils is estimated at less than 50,000, which appears to be extremely low for a nation of 80 million people. As a result, the cost of natural disasters is largely borne by government and homeowners when disasters strike. While no single factor can explain such a low insurance penetration for natural hazards, the underdeveloped state of the insurance market and low demand for insurance products seem to be key factors. 160. The Philippines non-life insurance market is known to suffer from excessive fragmentation, fierce competition, low capital base, and, as a result, excessive dependence on international reinsurers for claims paying capacity. The market is currently served by 98 domestic and 12 foreign-controlled direct insurance companies. As a result, competition has led to low primary rates in almost all classes, but particularly for natural perils. Despite the existence of a regulated earthquake tariff of 0.1% through the bundling of earthquake and fire coverage (with the latter being subject to market competition), local insurers until recently effectively bypassed the earthquake tariff regulation, which led to the under-pricing of earthquake risk. In an attempt to reduce the effects of competition on the earthquake tariff, it has been recently increased by the Insurance Commissioner to 0.15%. In addition, companies are now allowed to sell earthquake, typhoon, and flood coverage on a stand alone basis and separately from fire. Despite the recent requirement by the Insurance Supervisor to increase the amount of minimum paid up capital for all non-life companies to USD 1 million, many smaller companies managed to dodge the regulation and are still under-capitalized, carrying as little as USD 250,000, which is the minimum amount required to be put on deposit by the regulator. 39 161. The Philippines' local insurance market is also characterized by a low risk bearing capacity. Local companies' retentions on the treaty basis rarely exceed USD 200,000. This is due to relatively low surpluses of domestic insurers, in addition to the heavy concentration of companies' risk exposures in Manila (little business is written outside the capital). As mentioned previously, the low risk bearing capacity of the local market makes it heavily dependent on international reinsurance, with the result being that tremendously volatile premium rates are charged to local consumers. Such pricing volatility is a direct reflection of the pricing trends in the global reinsurance market which is notorious for its extreme 4 to 5 year pricing swings. With the continuing hardening of the international reinsurance market since September 11, 2001, the amount of reinsurance capacity available to the Philippines has been reduced, causing major rate increases for reinsurance capacity across the board. Insurance professionals interviewed for this study informed that on average, rates on line (ROL) of 25% or higher for reinsurance coverage (bottom-up) in excess of companies' own retentions are rather common. 162. The local reinsurance capacity is represented by three companies, out of which National Re is the largest. The overall reinsurance premium from the market in 1999 was USD 44.8 million, of which the companies retained USD 14.8 million, or 33.5%. With about USD 16 million of surplus capital, National Re has been retroceding most of its risk to foreign reinsurers as well, on average retaining over USD 8 million of premium annually. National Re is collectively owned by 116 local insurance companies, with GSIS being the single largest shareholder, with an equity stake of 19.3%. 163. Low demand for insurance remains among the key barriers to the growth of the industry. While per capita GDP is a low USD 1000, income alone does seem to explain the problem, as countries with similar per capita GDP have higher insurance penetrations. While an in-depth analysis on the demand side is needed to fully understand the problem, it appears that a general lack of insurance culture, poor track record of some insurance companies in paying claims, public reliance on government aid for reconstruction of dwellings destroyed by natural disasters and a 25% tax on the insurance premium are among the key factors to consider. 164. Philippines Catastrophe Insurance Pool? In 2002, the hardening of the international reinsurance market prompted vigorous discussions among local insurers about the creation of a catastrophe reinsurance pool which overtime could build up a surplus and thus reduce the dependency of the Philippines' insurers on global reinsurers for catastrophe coverage. These discussions have not yet resulted in any concrete proposals. The largest market players interviewed for this study and the Insurance Commissioner appear open to further exploring the idea of creating a specialized catastrophe insurance pool in the Philippines. 165. Such a catastrophe pooling arrangement can be highly beneficial for the Philippines and the development of the local insurance market. By adding the additional risk absorbing capacity, the creation of a Philippines Catastrophe Insurance Pool (PCIP) may lead to a major increase in insurance penetration for catastrophe insurance coverage and thus would greatly reduce the shocks of natural disasters on the local homeowners, and the economy as a whole. The launch of a PCIP is also likely to result in the reduced government fiscal exposure to the consequences of natural disasters as the government liability to the affected homeowners would be phased out over time. Finally, the creation of a PCIP is likely to give an impetus to the development of the domestic insurance industry as it would result in more stable premium rates for catastrophe coverage thus benefiting local consumers and making insurance more attractive and affordable in the long run. 166. Despite the fact that the track record of past insurance pools in the Philippines has been poor, there are some positive exceptions, with the Philippines Machinery Insurance Pool being the case in point40. It appears that key factors underlying the success of a specific insurance pool invariably could be traced to the 40The PMVLP was designed for providing third party auto liability coverage and shortly after was disbanded due to unsatisfactory operational performance. The PMVLP's destiny was shared by the SSS (the Social Security Pool) and the Nuclear Pools. 40 quality of the management team and the composition of the Board. It would therefore be essential that the Board of a PCIP be representative of the main writers of catastrophe insurance coverage in the market and be designed with a view to avoiding giving any single company the controlling share of the Board regardless of its market share. A balanced and well represented Board consisting of key public and private market players would be essential to making such a program a success. Government influence over the PCIP could be exercised through representation of the Insurance Commissioner and/or GSIS on the Board and, possibly, by bringing in as a shareholder the Philippines Development Bank. In this context, in designing a workable governance structure for the PCIP, one could use National Re as a model which withstood the test of time due to its ability to align the interests of different market players with that of the company. 167. The performance of the PCIP's operations would be inseparable from the quality of its management team. Therefore it would be crucial to ensure that only high caliber professionals are retained for the Pool's senior management team. Again, National Re and the Philippines Machinery Pool's recruitment practices provide good examples. 168. In designing the PCIP it would be important to address the issue of adverse selection, which could loom large if the purchase of catastrophic coverage is left to consumers' discretion. Making catastrophe insurance coverage compulsory for homeowners in certain income brackets would be a way to address the problem. A "softer" way to address the issue of adverse selection would be to require catastrophe insurance coverage on all residential mortgage loans originated in the country. In addition, it will be important to ensure that provisions for lower income families who have limited purchasing power are explored so that they are not marginalized. Policy Options for More Effective Catastrophe Risk Management 169. While further study is needed, a brief overview of the insurance market and the existing risk financing arrangements indicates that the Philippines is facing some potentially very large funding gaps in the case of a catastrophe. In order to manage risk more effectively, Government action to reduce risk should be accompanied by a formal approach to risk financing. Such a risk financing strategy would consist of: (i) formal risk assessments at the LGU and the federal levels; (ii) identification of funding gaps; and (iii) development of national and LGU risk management plans aimed at closing the identified funding gaps over time. Such risk management plans could consist of a combination of instruments such as catastrophe insurance, contingent credit facilities, and other mechanisms. This section lays out some policy options to consider and recommendations for further exploration. 170. Fiscal Incentives for Proactive Risk Management at the LGU Level. As a start, the fiscal vulnerability of the federal and local governments to natural disasters can be reduced significantly by introducing institutional incentives41 for better risk management and institution building. To reduce the funding gap and the vulnerability of the poorest segments of the population and critical infrastructure to natural disasters, the existing national system of financing disaster losses could be redesigned to provide strong fiscal incentives to LGUs for more proactive risk management. The following combination of policy actions at the federal and LGU levels is proposed as a part of any risk financing reforms: · In the case of ex-post sources of risk financing, the GoP may consider rewarding LGUs pursuing proactive catastrophe risk management with additional fiscal resources for rehabilitation of destroyed state-owned assets in the aftermath of natural calamities. It would be desirable to make the quantity of such additional financial assistance known in advance. · The GoP may also consider introducing an output driven budgeting of national disaster management agencies. For instance, a submission and achievement of certain risk management 41The level of post-disaster funding to the LGUs is based more on the size of disaster losses and the local economy. The current system does not provide incentive for those LGUs that may have taken proactive steps in risk reducing measures. 41 goals stated in agency risk management plans could be required as a precondition for budgetary allocations in support of their work programs. · It is also worth exploring the making of infrastructure investments financed by IFI's and the Development Bank of the Philippines contingent upon LGUs submitting a comprehensive risk management plan for the proposed investment. This would require government policy action that would not only safeguard the contemplated public investments but also would promote broader active risk management approaches, including loss reduction measures, on the part of LGUs. 171. Working Group to Explore a PCIP. It is recommended that a Working Group be created for the purpose of exploring the establishment of a Philippines Catastrophe Insurance Pool (PCIP), as described above. Such a group could consist of the representatives of PIRA, GSIS, National Re and the Office of Insurance Supervisor, with the latter providing intellectual leadership. The group could work closely with the insurance market, the Philippines Government, the World Bank and international reinsurers in developing a workable proposal for a business plan, institutional design, and regulatory framework for the operations of the PCIP. The Working Group could also engage in several studies that would be beneficial prior to the launch of a PCIP: · Risk Management Study. A detailed earthquake risk management study was commissioned by the PIRA to RMS and its findings should be available shortly. This is a welcome development which would enable the market to receive an independent verification of earthquake reinsurance rates and the primary rates and improve the management of aggregate risk accumulations in individual insurance companies. Nevertheless, it is understood that the study is based on the 1997 inventory data and thus would need to be updated. A similar risk management study should be carried out for three other major perils, i.e., typhoon, flood, and volcanic eruption with a view to producing an integrated national natural hazard GIS database for each of the perils. Such a detailed analysis would be instrumental for producing ratings schedules for each peril ­ an essential precondition for the PCIP launch. The analysis should start with base maps, to produce high hazard, vulnerability and risk maps, which are constantly updated, serves as a real-time response tool in disasters as well as the repository for damage and needs data, which are then input back into the database to improve the accuracy of (a) response, and (b) future hazard, vulnerability and risk maps. · Demand Study. Different opinions exist in the insurance market regarding the low consumer demand for catastrophe insurance covers. Poverty, religious beliefs, lack of risk awareness and low insurance culture are cited among the main constraints on the demand side. Given a disproportionately small number of fire policies carrying endorsements for earthquake and typhoon, a survey of consumer demand would need to be carried out. · Quantification of Catastrophic Risk Exposure for Commercial Banks. Interviews with commercial banks conducted for this study revealed a very low awareness of catastrophe risk exposures. At present, lenders do not require borrowers (neither commercial or residential) to obtain catastrophe insurance coverage for financed assets. Catastrophe risk is not factored in the banks' lending rates and is not accounted for in banks' risk management. However, given the high concentration of bank loans in Manila, a severe earthquake in the city can have devastating results for the banking lending portfolios. A detailed study of banks' exposures to risks of natural disasters is important to gauging this risk. 172. Mandated Catastrophe Insurance Coverage Requirements. These incentives/ requirements could be part of the requirements for additional post-disaster funding for LGUs from the center and are likely to vary between the regions according to local realities. Approaches that could be considered include: · Requiring that replacement cost insurance is purchased when mortgage financing is granted. This in some cases could be effected through a relatively small addition to the interest rate, and could 42 even be accompanied by a slight reduction in the underlying interest rate reflecting the slight resultant reduction in credit risk. · Making it clear, if necessary through regulation, that households in the upper or middle income brackets are not eligible for government reconstruction funding (although they would continue to be eligible for relief). · Linking catastrophe insurance into the land tax or land registration systems. 173. Contingent Credit Facilities. To finance reconstruction of public infrastructure and housing, the GoP has been utilizing World Bank and ADB emergency reconstruction loans, as well as reallocations from ongoing loans in the lending portfolio. Although this type of support is processed more quickly than standard lending operations, the funds do disburse slowly when compared to ex ante mechanisms. As a result, these lending instruments are not appropriate for meeting the Government's immediate liquidity needs in the aftermath of a disaster, which may have far reaching negative social and economic implications if they are not satisfied. A more efficient option for catastrophe risk financing would be the use of contingent credit facilities. The contingent credit facility could be extended to the National Calamity Fund to finance disaster management activities at the LGU level, Federal Government level as well as support the PCIP. 174. The credit facility could be designed to disburse quickly on an as needed basis, triggered by the occurrence of a disaster of magnitudes agreed to by the Government and based on agreed criteria. If drawn down, the lending facility backing the National Calamity Funding could then be replenished without any major costs for the Bank or the Borrower. This funding approach for natural disasters would enable the GoP to switch to a proactive mode of financing natural disasters by replacing multiple ex-post future emergency lending operations with a single line of credit, and to obtain access to immediate liquidity to meet reconstruction needs in the aftermath of a disaster. The facility could be applied at several different levels: · Contingent Liquidity Facility for the Federal Government. In this case, the contingent facility could be designated to finance post-event urgent repairs of federally owned infrastructure or the set-up of life-line facilities. Similarly, criteria for financing could also be set for disbursement, such as: (i) occurrence of a catastrophic event resulting in a major loss to federally owned infrastructure as certified by independent loss adjusters; and (ii) achievement of risk management objectives or specific institution reforms in the area of national catastrophe risk management as set out in the loan document. · Contingent Liquidity Facility for LGUs. Such a contingent credit line could be extended to the National Calamity Fund which would enable it to operate as an effective, fully funded provider of liquidity to disaster stricken LGUs. The facility would be available to meet claims of the LGUs in the aftermath of natural disasters, provided criteria, among them acceptable risk management programs are in place. In addition, through the use of criteria such as the presence of an acceptable risk management program before fund release to LGUs, the Government can build in other incentives. · Contingent Credit Facility in Support of the PCIP. The contingent credit facility could also be extended to the PCIP, similar to that supporting the Turkish Catastrophe Insurance Pool, and could be used to finance insurance claims from homeowners whose properties were damaged by natural calamities. 43 Chapter 5. The Way Forward: Summary of Findings and Recommendations 175. Efforts to reduce poverty in the Philippines, and investments in social and economic development, will simply not be sustainable without more effective disaster risk management. This chapter summarizes the recommendations on how to achieve this goal, and discusses potential areas for World Bank support in this regard. These recommendations are also presented in a table in Annex 1. 176. The main recommendation emerging from this study is that the Government of the Philippines should develop a national framework for comprehensive disaster risk management. This framework should provide for political leadership and policy support at the highest levels, while facilitating the active engagement and implementation of all relevant stakeholders at the national, local, and household levels. These actors should include public agencies, the private sector, and civil society. The framework should incorporate the three essential steps of integrated risk management, which include risk identification, risk reduction, and risk sharing/financing. Such a framework will take time to prepare and must be part of a medium to long-term objectives where results from implementation of priority activities would contribute to the development of such a strategy. Below are more specific recommendations for each of these areas. Risk Identification 177. Produce hazard and vulnerability maps for major natural hazards. In the Philippines, there are inadequate reliable data on the type and amount of Philippine economic activity at risk from natural hazards. A fundamental obstacle is the absence of accurate hazard and vulnerability maps. An integrated national natural hazards data and mapping system needs to be designed and implemented. This would consist of the following: o Design of an integrated GIS database for the Philippines, for use in natural hazards data collection and mapping. This GIS database would have many benefits beyond natural hazards. o Consistent modern maps of natural hazards zones (flood, geologic hazards, earthquake faults and shaking, volcanic hazards, etc) should be generated for the entire Philippines, using the integrated GIS database. These maps would be very beneficial for land-use zoning, project design, insurance, emergency response and many other applications. o Risk analyses should be developed for high hazard areas, using the GIS database, which would then form the basis for a comprehensive understanding of the impacts of natural hazards on the nation. Loss estimates would prove useful to decision-makers and the public, and could be integrated into a national economic development plan. o The data and mapping system could form the backbone for the national emergency response plan, and would serve during emergency periods as a database for needs assessment and data collection. The data so collected would then be in a consistent uniform format, for use in improving hazard maps and risk analyses. o Currently, NAMRIA, NDCC, PHIVOLCS and others are all implementing limited GIS databases. Their efforts should be encouraged, and coordinated into a common format, so as to begin the building of this integrated national natural hazards GIS database. 178. This information is important as a lack of national or regional vulnerability analyses impede the effectiveness of post-impact needs assessments, which in turn impede recovery. In addition, proper land use is not possible without accurate base maps. Similarly, the strengthening of risk transfer mechanisms such as insurance is unlikely without these analyses, as they are necessary to produce rating schedules for insurance pricing. 44 179. Improve capacity for damage and needs assessments. There is scope for considerable improvement in post-disaster damage assessment methodologies and practices. It is recommended that existing guidelines developed by the Economic Commission for Latin America and the Caribbean (ECLAC) should be adapted for use in the Philippines. More comprehensive assessment would enhance knowledge and understanding of the nature and scale of impact of disasters and forms of vulnerability. This, in turn would aid the implementation of good risk management practices, including greater consideration of hazard-related issues in broader sustainable development and poverty reduction policies and programs as well as of appropriate, cost-efficient post-disaster relief and rehabilitation efforts. 180. Integrate Environmental Disasters into Risk Management Framework. In light of the increasing risk of environmental disasters, it is recommended that the concept of disaster management, and the development of a national framework to manage disaster risk be expanded to include environment-related disasters, encompassing: (a) the identification of environment-related hazards and their potential impacts; (b) the assessment of local vulnerability to these impacts; and (c) the development of adaptation mechanisms and measures. A greater role for the Department of Environment and Natural Resources and for environment- related groups in civil society and the private sector is envisioned if environment-related disasters are to be integrated into this new framework. Risk Reduction 181. While a more detailed review of the institutional arrangements and capacities for disaster risk management should be undertaken in order to identify gaps and priorities, this study has revealed that the country's management of disaster risk would be greatly improved by addressing several areas: 182. Emphasize proactive disaster risk management. The disaster management system in the Philippines is based on an outdated decree (dating back more than 20 years), and would greatly benefit from integrating lessons learned in the country around the globe over the past decades. In general, there is a failure to recognize natural hazards as a potential obstacle to long-term sustainable development. Reflecting this, there is a widespread emphasis on post-disaster relief and short-term preparedness (forecasting, evacuation, etc.) rather than mitigation or post-disaster support for economic recovery, such as livelihood regeneration or tax breaks to affected businesses. 183. There is also widespread acknowledgement that the current legislation, passed in 1978, is outdated and is constraining progress. Assuming the draft Bill is passed (it is titled the Comprehensive Disaster Management Bill), many issues discussed in this report might be remedied. Updated legislation is an imperative if the nation is to move forward, but to do so the new legislation needs to focus on integrated disaster risk management (IDRM). What this implies is that the new legislation needs to redefine the duties of central and local governments, directly bring the private sector, specifically utilities, into both emergency management strategic decision-making and operational contexts, and promote the sustainable management of hazards and risks in a way that contributes to the well being and safety of the public and property. There are examples of this type of legislation for the Philippines to consider. 184. Emphasize a more bottom-up approach with participation of all stakeholders. The Philippines system tends to be more of a centralized top-down administrative system than a community-based system and there are few incentives for local level initiatives. For example, under NDCC arrangements, while each local level can `submit recommendations as necessary', all are required to adhere to guidelines established by the level above. In practice, the Philippine disaster management system is a disaster response system at local levels. While there are measures to link disaster with development, this connection is very loose: it needs strengthening. This is necessary to encourage some local government units (LGUs) to initiate sustainable development practices just as much as it is to encourage other LGUs to continue along this path, often against 45 bureaucratic obstacles. With very few exceptions, local level systems are response-driven ­there is no obvious effort to initiate proactive hazard management/risk reduction coordination. 185. Consider consolidating elements of the disaster management system. This current disaster management system comprises six levels encompassing almost 46,000 separate administrative units. This is a difficult system to operate given the top-down approach that characterizes Philippines disaster management. Merging tiers and some local groups at the barangay level into clusters (perhaps aligned to hazardscape clusters, such as floodplains or fault traces) might warrant consideration: this has been successfully undertaken recently in New Zealand for the purpose of disaster risk management at both regional and local government levels. 186. The issue of consolidation at the barangay level seems especially important for disaster management since barangay captains are obliged to make certain their community develops preparedness plans. However, 10 years after the initiation of the Local Government Code, the barangay system has not been sufficiently empowered to achieve its tasks. The question has to be asked whether the barangay is too small and/or inherently weak to be the basic building block for the planning and implementing unit for government policies (this is an issue that goes beyond disaster management). 187. Another area that requires consolidation pertains to the education and training of local officials in disaster management. Currently, training programs are not compulsory for personnel designated with disaster management tasks. Moreover, the ODC is unable to provide systematic training for local government unit (LGU) personnel primarily because the EMIP, linked to the Armed Forces of the Philippines, has a first priority to train AFP staff; it does not have the resources to extend far beyond this group. Consideration should be given to expanding existing facilities or alternatively to create a new facility for LGUs. The latter suggestion might be better placed within DLGI, since it too has a training wing, the Local Government Academy. If this facility does not have existing disaster-related courses, perhaps consideration could be given as to how they might be developed: there are many relevant programs and trainers internationally that could provide insights. In a similar manner, OCD doesn't know how the training material it does provide to LGUs is being used. 188. Improve Coordination Among Actors. The need for greater coordination is well recognized by almost all the agencies involved in disaster management. The type of coordination needed is less a top-down oversight function but rather a proactive, participatory and encouraging orientation type of function. Greater organizational, management and task synchronization are regarded as prerequisites at both national and international levels. These factors extend across-the-board, at all levels of the Philippines disaster management institutional arrangements. Two examples can illustrate the diversity: · The issue of uncoordinated and non-cumulative technical studies resulting in `project findings often going nowhere after initial funds dry up' cited above is a case in point at the international level. · The potential for national-local conflicts over hazard management regimes is likely to cause problems unless a coordinated effort is undertaken. A cited case relates to river flood management: rivers are managed nationally by DPWH, although they flow through LGU jurisdictions. In Metro Manila this responsibility was recently handed over to MMDA for river control and flood management, but it claims it does not have the capacity to do this, so the responsibility stays with DPWH for the present. This situation has put flood management in Manila into a state of limbo. At the same time, individual LGUs in Metro Manila wanting to develop flood management regimes within their jurisdiction are wrangling with neighboring LGUs, which do not regard the issue as a priority. 46 Risk Sharing/Financing: 189. As discussed in the last Chapter, the Government of the Philippines and individual households currently bear the majority of costs caused by natural disasters. More effective options for financing disaster risk and relieving the burden of disasters from the public sector need to be explored, including the idea of a catastrophe insurance pool, and/or contingent credit facilities. The options available to the Government are discussed in more detail above and are only summarized here: Development of fiscal Incentives for Proactive Risk Management at the LGU Level: o to implement proactive disaster management, thereby balancing out the present strong rehabilitation and reconstruction emphasis. However, this will require a more careful analysis of the appropriateness of the present funding system, funding gap, and actions that would need to be implemented to reduce risk. o that encourage LGUs to include mitigation and prevention activities in their development plans. Based on the achievement of agreed output indicators, the LGUs could be rewarded with additional funding to improve their overall disaster management capacity. o which, in the constantly hard hit area of infrastructure, could also be used by IFI's when financing LGU-led infrastructure development, for example, by requiring that the LGUs submit a comprehensive risk management plan for the proposed investments. However, this type of incentive must be accompanied by government policy action to cause it to be implemented. Explore the potential for a PCIP: o carry out a study to assess why the area of insurance remains under-developed in the Philippines as well try to determine its potential for increase. To do this, it would be most effective to create a technical working group including representatives of PIRA, GSIS, National Re and the Office of Insurance Supervisor to develop a workable proposal for a business plan, institutional design, and regulatory framework for the operations of the PCIP. Explore Contingent Credit Facilities: o to allow the rapid mobilization of funding after a disaster to assess damage, rehabilitate and reconstruct. Given the frequency of disasters in the Philippines, it is important that the Government explore the use of a contingent credit facility to fund post and pre-event activities. To develop a contingent facility requires that there is sufficient understanding of risk and agreement on the criteria that would trigger withdrawal. o determine the appropriate level at which the contingent facility could be developed: Federal Government, LGU level and/or PCIP. Potential Areas for World Bank Support 190. To support the Philippines' effort to better manage disaster risk and enhance poverty reduction, the World Bank should examine the ongoing portfolio to identify how its ongoing development projects can support the goal of disaster risk reduction. In addition, the Bank should consider more direct support to the development of an integrated disaster management risk approach, through the provision of technical assistance and lending. 191. The World Bank has developed a significant amount of learning over the years through its provision of technical and lending support for disaster risk management. It should apply this learning to the Philippines in their efforts to improve poverty reduction. This assistance can take the form of a technical assistance loan 47 to facilitate the development of a national framework for disaster risk management and institutional capacity building to implement the strategy. As described above, the Bank could also provide support in the development of contingency credit facilities to reduce the economic and financial impacts of disasters in the Philippines. 192. A definite timeline of how the various recommendations outlined in this proposal can be implemented cannot be presented at this time because of the need to get all stakeholders involved in the planning process. Designing an action plan should be an integral part of the integrated disaster management framework. Also, there is need to prioritize the various recommendations taking into account the Government constraints, donor and financing agency priorities, capacity of the implementing agencies and related policy and legal changes needed to create an enabling environment. Through the workshop that the Bank is co-financing with the NDCC at which various stakeholders will participate, it is expected that the beginnings of the prioritization process will begin. 193. Despite this, a few key areas where the Bank could add value in the short, medium and long-term: o Short-term (0-5 years). Support the country in carrying out gap analyses in comprehensive disaster management and particularly in areas of risk reduction and mitigation and include these in a comprehensive strategy; identify capacity building needs of the key implementing stakeholders with the aim of strengthening them; aim to better understand vulnerability of those most at risk and develop targeting mechanisms to ensure that they are not sidelined when disasters actually happen. As part of the CAS development include disaster management as an important consideration in poverty reduction mechanisms and strategies. The Bank can also start to engage other donors active in disaster management to develop collaboratively a disaster management strategy that is not only comprehensive but also has the buy-in of all concerned. With regards to risk financing, the Bank can assist the Government develop contingency financing and insurance options. o Medium term (10-15 years). More systematically include disaster management in the Bank's portfolio, not necessarily as stand-alone projects but as integral parts of the projects developed; include disaster management measures into the CAS and poverty reduction strategies. However, solutions should not be standardized but rather tailored to the issues identified. o Long-term (over 20 years). Encourage regional cooperation for disaster management to reduce the dependence on dole outs from the developed countries, and to benefit the participating country personnel from training and other capacity building activities. 48 Annex 1 Summary of Key Recommendations Area of Disaster Management Recommendations Hazard Exposure Identification Carry out a more comprehensive analysis of both the impact of individual disasters and of current and likely future vulnerability to natural hazards from a macro-economic perspective Produce hazard and vulnerability maps for the major hazards Comprehensive risk analysis to determine potential impacts of disaster events Develop comprehensive guidelines, methodologies and practices for damage assessments Build capacity for improved assessments Improve coordination between assessing agencies Improve communication and data capture and manipulation techniques Conduct research on the linkages between poverty, vulnerability and environment or the socio- economic impact of disasters More careful and disaggregated examination of fiscal consequences of disasters Risk Reduction Emphasize proactive risk management and develop a national framework for comprehensive disaster risk management to give strategic direction Emphasize a bottom-up decentralized approach Consolidate the current disaster management system Improve coordination among actors Development of a more explicit policy on financing of post-disaster response Risk Sharing/Financing Develop fiscal incentives for pro-active risk management Establish a working group to determine feasibility of increasing insurance coverage in the country Explore contingent facilities Explore risk transfer options Implementing Disaster Management Determine Policy and Legal changes requires to support implementation of integrated strategy Develop an operational planning framework for all actors and agencies to improve coordination and prioritization of activities Include environmental disasters into the risk management framework 49 Annex 2 International Experience with National Catastrophe Risk Management Programs42 Introduction While there is a wide range of views on the rationale for government interventions in the insurance markets in the pertinent literature, in general they can be reduced to two major groups. The first category emphasizes the high cost and limited supply of private capital.43 According to the proponents of this view, a shortage of risk- bearing capital leads to an inadequate supply of insurance, which keeps prices steep relative to projected losses for low frequency high severity events, which is socially sub-optimal. In 1994, for instance catastrophe reinsurance premiums were more than seven times the expected loss although that multiple has dropped to between four and five by now.44 Proponents of this view also contend that government, with its vast capacity to tax and borrow, has an advantage over private insurers in bearing catastrophe risk because it does not need to hold explicit capital to pay off claims and avoid bankruptcy.45 To free insurers from the costly burden of holding huge amounts of capital, proponents suggest that the government act as a residual provider of reinsurance for so-called mega-catastrophes. The government could set premiums below those charged by private insurers, thus lowering the cost of insurance while protecting taxpayers from losses. The second group of arguments emphasizes that the biggest barrier to an adequate supply of insurance, especially immediately after a catastrophe, is insurers' heightened uncertainty about the frequency and size of future losses. After Hurricane Andrew, the Northridge earthquake, and the World Trade Center attacks, insurers were not certain that they could assess the risks they were being asked to assume. Without such knowledge, they were unwilling to commit capital by underwriting the coverage. In time, insurers are usually able to recalibrate their estimates and reenter the market. Thus, proponents contend that the government needs to intervene to supply insurance while insurers reassess risk after a disaster, but they argue for a temporary government role.46 Despite these alleged differences of opinion on the government role in catastrophe insurance markets that exist until today, the mounting uninsured losses from natural disasters such as hurricanes, earthquakes and floods pressed governments in disaster prone countries and regions to look for practical solutions for catastrophe risk management which has spurred the formation of national and regional catastrophe insurance programs. To date, twelve national catastrophe risk management programs have been established and operate successfully in ten different countries with the sole purpose of providing affordable catastrophe insurance 42 This annex is a reprint of technical working paper "Managing the Risk of Natural Disasters in India" by Rodney Lester, Eugene Gurenko, and Vijay Kalavakonda, World Bank 2003. 43 For example, see David M. Cutler and Richard Zeckhauser, "Reinsurance for Catastrophes and Cataclysms," in Kenneth A. Froot, ed., The Financing of Catastrophe Risk (Chicago: University of Chicago Press, 1999), pp. 233-269. 44 Premiums for the highest layers of coverage (the lowest probability layers) were between 20 and 30 times expected losses in 1994, according to estimates. See Kenneth A. Froot, The Market for Catastrophe Risk: A Clinical Examination, Working Paper No. 7286 (Cambridge, Mass.: National Bureau of Economic Research, February 2001), p. 7 and Figures 3 and 4. However, research emphasizes the imprecision of the estimates of actuarial losses for the least likely events. See James F. Moore, Tail Estimation and Catastrophe Security Pricing: Can We Tell What Target We Hit If We Are Shooting in the Dark? Working Paper No. 99-14 (Philadelphia: Wharton Financial Institutions Center, 1999), available at http://fic.wharton.upenn.edu/fic/. 45 Statement of Lawrence H. Summers, Deputy Secretary, Department of the Treasury, before the House Banking and Financial Services Committee, April 23, 1998. 46Proposal on Federal Reinsurance for Disasters, Congressional Budget Office, September 2002 50 coverage for homeowners. While design and coverage features provided by these insurance programs varies, the underlying rationale for their introduction has been the same, i.e., to address the challenges faced by the private insurance markets in insuring the risk of natural disasters. Table 1 lists the most well known of these programs which include TCIP in Turkey, FONDEN in Mexico, the FHCF in Florida, the HHRF in Hawaii, CEA in California, EQC in New Zealand, NatCat in France, and Norway's Norsk Naturskadepool. Table 1: Government Sponsored Catastrophe Insurance Programs Name of the Fund Country Year Established & Risk Covered Turkey Catastrophe Insurance Pool Turkey 2000/ Earthquake (TCIP) Catastrophe Naturelles (CatNat) France 1982/ All Natural Disaster except for Windstorm, ice and snow Japanese Earthquake Reinsurance Japan 1966/ Earthquake, tsunami, and volcanic Company (JER) damage Earthquake Commission (EQC) New Zealand 1994/ Earthquake, tsunami, volcanic damage, landslide. Norsk Naturskadepool Norway 1980/ Floods, storms, earthquakes, avalanches, tidal waves Consorcio de Compensacion de Seguros Spain 1954/ Earthquakes, tidal waves, floods, volcanic eruptions, and cyclonic storms. Taiwan Residential Earthquake Taiwan 2002/ Earthquake Insurance Pool (TREIP) Florida Hurricane Catastrophe Fund USA 1993/ Windstorm during a hurricane (FHCF) Hawaii Hurricane Relief Fund (HHRF) USA 1993/ Windstorm during a hurricane California Earthquake Authority (CEA) USA 1996/ Earthquake Note: None of the above mentioned insurance programs insure public infrastructure assets or provide/finance emergency relief services. Catastrophe risk is unique due to its highly systemic nature which manifests itself in large aggregate concentrations of risk and consequently in infrequent but severe losses for national or regional economies. For instance, since 1989, there have been 15 natural disasters in the United States alone which resulted in $43 billion of insured losses. It is no longer unusual for the global insurance industry to sustain losses from a single catastrophic event in excess of US$1 billion due to the growing concentration of assets in major urban centers. The management of these catastrophe risk exposures is highly capital intensive and it is difficult, if not impossible, to diversify risk at the level of primary insurers. As a result, after sustaining major losses in the aftermath of natural disasters, private insurance markets tend to ration or, in some cases, discontinue offering their catastrophe insurance coverage for homeowners or small business unless some sort of risk sharing arrangement with the government is put in place. Design Features of Catastrophe Insurance Programs47 A survey48 of twelve national programs reveals significant similarities that exist among the programs. Most programs (1) tend to focus on providing coverage against a specific natural hazard; (ii) tend to have a regional focus; (3) offer coverage mainly for dwellings and contents; (3) have premium rates that reflect the characteristics of the risk and contain an element of solidarity which effectively provides for cross-subsidies 47Apart from catastrophe insurance programs presented in Table 1, some countries opt for public sector managed and financed disaster funds with the primary objective of providing ex-post disaster assistance to a) low-income households, and b) to carry out immediate repairs (but not necessarily replacement ) of damaged infrastructure assets in the wake of natural disasters. An overview of these disaster relief funds however is beyond the scope of this paper. 48World Catastrophe Reinsurance Market 2002, by Guy Carpenter. 51 from better risks to worse; (iv) receive no direct government subsidies; (v) typically do not emphasize mitigation, although some programs encourage retrofitting and safer construction practices by offering premium discounts; and, finally, (vii) carry out sales and servicing through the established distribution networks of private primary insurance companies and their agents. Table 2 provides a convenient overview of key design choices available to policy makers and insurance practitioners involved in the creation of national catastrophe insurance programs. A more detailed discussion of these design options follows. Table 2. Catastrophe Program Design Variables Program Variables Design Choices Management Public/ Private Governance Public/ Private/ Mixed Funding Public/ Private/ Mixed Insurance Vehicle Insurance Pool/ Reinsurance Pool/ Insurance Companies Coverage Buildings; Contents; Business Interruption Lines of Business Residential/ Commercial (SME's) Rates Flat/ Risk-based/ Mitigation Incentives Distribution Insurance Companies/ Alternatives Retentions by Insurance Companies 0 to risk based solvency margin Geographical Coverage National/ Regional/ Inter country Participation Compulsory/ Voluntary Reinsurance Private/ Public mix Management and Governance Less than a third of catastrophe insurance programs mentioned in Table 1 are managed by the government, with NatCat of France or CEA of California being among the primary examples. Some pools however are managed privately, like the Turkish TCIP. Nevertheless, it should be noted that even in privately run programs government influence and control remains strong through some form of government representation on their Board, which provides an element of accountability to the public. In most cases an independent professional fund/pool manager is employed to carry out day-to-day operations. Typically, the primary functions of the fund manager include but are not limited to a) collection of premium; b) claims management; (c) asset management; and, (d) placement of reinsurance. Investment functions are carried out in accordance with the guidelines established by the Governing Boards of Directors. While in some cases, like the Florida Hurricane Insurance Pool, catastrophe programs have their own direct distribution channels in addition to those of participating private insurers, most of these entities tend to rely heavily on the distribution and servicing capabilities of primary insurers. For instance, in the case of TCIP, the pool manager is the country's largest reinsurer, Milli Re, which markets earthquake coverage through the distribution channels of the Turkish insurers. For claims adjustments, the TCIP relies on independent loss adjusters mobilized by insurance companies responsible for handling respective claims. The pool managers are typically compensated for their services with a management fee which varies widely ­ from 0.8 percent of the Net Written Premium in the case of FHCF to 2 percent in the case of JER - and which in some cases is contingent upon achieving certain performance benchmarks such as a certain level of insurance penetration for the Pool's major business line, for instance. 52 Besides direct involvement in the operations of a pool or through representation on the Board, government has another important role to play, e.g., to be a reinsurer of last resort. In the case of NatCat (see Figure 1), for instance, government provides a sovereign guaranty to the state-owned reinsurer, CCR, for all claims in excess of its claims paying capacity. In the case of New Zealand's EQC program, government guarantees its policyholders that it will assume the financial responsibility for meeting the EQC's residual claims which are over and above its claims paying capacity. Figure 1. French NatCat System French Government Unlimited Private Reinsurers Guarantee CCR* P & C Premium Extra-Charge CatNat Decided by the government Property and Casualty Contracts Private Insurers *CCR ­ Caisse Centrale de Reassurance, the public reinsurer . The French government offers the CCR a non-limited guarantee, meaning that the government is the reinsurer of last resort. Source: The Public Private Sector Risk-Sharing in the French "Cat. Nat System" by Marcellis-Warin and Michel- Kerjan, November 2001. Funding The primary sources of funding risk exposures for catastrophe pools are insurance premium from the insured homeowners, reinsurance premium, in cases when pools act as reinsurers themselves, reinsurance coverage from their own reinsurers, pool's own surplus capital, assessments on private insurance companies, commercial backstop facilities, contingent credit lines, and, finally, direct government contributions in excess of programs' claims paying capacity. In addition, over the last few years, some catastrophe insurance pools, such as CEA, have also obtained access to international capital markets by issuing catastrophe insurance bonds. Figure 2 provides an example of a mixed structure of funding for catastrophe risk in case of JEE. 53 Figure 2. Japanese Earthquake Reinsurance Program Liability of J.E.R.;(A1) + (A2) + (A3) = Y 75,000m. + Y 219,400m.+ Y 85,630m Liability of private ins. and The Toa Re.; (B1)+(B2) = Y 281,800m. + Y 85,500m Liability of Government; (C1) + (C2) = Y 501,200m. + Y 3,251,470m Insurance Vehicle In addressing the inherent underlying constraints of the domestic private insurance market in case of catastrophe insurance coverage, countries opted for either of two choices, e.g. specialized direct catastrophe insurance or reinsurance vehicles. This choice in many respects has been predetermined by the development of the local insurance market and its willingness to retain any catastrophic risk underwritten by the program. A combination of the two approaches is also possible, with the Florida Hurricane Fund being the prime example. Currently, of the twelve programs discussed here, four of them (CCR, FHCF, JER, and Norsk Naturskadepool) are designed around the reinsurance concept and in the remaining eight programs government plays a critical role by providing an "implicit" or an explicit guaranty to honor all claims against the pools, which in essence amount to an excess of loss reinsurance contract. Table 3. Insurance Vehicle Fund Insurance Vehicle Characteristics TCIP/ EQC/ CEA/ Insurance Pool Lack of capacity on the part direct insurers HHRF to underwrite Cat Risks. FHCF/ CCR/ Norsk Reinsurance Pool Reinsurance provided at both below market Naturskadepool rate and with minimum volatility in reinsurance prices JER Insurance Company Risk spreading among insurance companies who are shareholders of JER and also reinsurance capacity provided by the Government of Japan. Coverages While all the catastrophe insurance programs listed above offer coverage for buildings and usually contents, only one third covers of Business Interruption risk. Several of the surveyed insurance programs also include emergency living expenses in the immediate aftermath of a disaster in their coverage. While all programs offer personal catastrophe risk coverage, only a few cover commercial risks. One of the reasons behind such a strong focus of these institutions on providing residential coverage is their explicit social commitment to ensuring that adequate catastrophe insurance coverage exists for the population. 54 Besides, commercial industrial risks as a rule are well covered even in the least developed markets and thus are rarely a subject of a public policy concern. Nevertheless, there is certainly a scope for extending catastrophe insurance coverage provided by catastrophe pools to small and medium enterprises (SMEs), which are typically underinsured. Rates As the primary objective of most catastrophe insurance programs is to ensure the availability of affordable insurance coverage for homeowners, their premium rates for worst risks tend to be capped at some level. Some programs such as the HHRF and Norsk Naturskadepool charge a flat rate irrespective of location or construction quality of covered properties, which of course takes the "solidarity" principle to the extreme and offers no mitigation incentives. The advantage of having the flat rate is its administrative simplicity. The majority of programs, however, charge variable rates which depend on a property's risk zone and the type of construction. All in all, about a half of the programs have premiums which are risk-based and none are subsidized. On average, all programs appear to collect enough premium to cover claims and expenses. Table 4. Rates Charged and Mitigation Incentives Fund Rate Mitigation Incentives TCIP Depends on the location of property Can refuse coverage for new construction which is non-compliant with the building code. NatCat 9% of underlying policy Insurance companies can refuse to (auto, fire) premium extend the "natural disaster" guarantee to buildings built in high-risk zones and/or built in violation of administrative rules in effect at the time of construction. This encourages mitigation. JER 5.0% to 4.3% of insured None value, depends on location and construction type Norsk FLAT RATE on insured None Naturskadepool values Consorcio de 0.09% to 0.25% of insured None Compensación value TREIP NT$1459 per policy, flat rate None FHCF Premium based on location US$10 million earmarked for funding and construction type. mitigation activities. Premium remains constant but coverage level changes. HHRF US$1.50 per US$1,000 Rate credits available for roof-wall and roof foundation clips and storm shutters. CEA Ranges from ­ 1.1% , and Depending on date of construction, a capped at 5.25%. house that has been retrofitted may be entitled to a 5% premium discount. 55 Voluntary vs. compulsory Most of the programs are voluntary, with only three being compulsory or semi-compulsory (TCIP, FHCF, and JER). In the case of compulsory programs, compliance is generally low, with around 20 percent of insurable housing stock covered in the case of TCIP and JER. Yet the level of insurance penetration achieved under the compulsory programs is undoubtedly considerably higher than under the programs with voluntary participation. In the case of India, for instance, where insurance coverage for natural disasters is optional, the insurance premium for natural disasters represents less than 1 percent of the total premium written. Reinsurance Catastrophe events are the greatest single threat to the solvency of insurers. Rating agencies generally require that insurers have enough capital to pay for at least a 100-year loss. To attain the top rating, insurers may need to maintain enough claims paying capacity for surviving a loss from a 250-year loss event. Reinsurance is the traditional method used by insurers to boost their claims paying capacity , but capital markets are becoming a growing source of alternate risk capacity, particularly lately in light of the global upswing in catastrophe reinsurance prices (Figure 3). This recent upward trend in reinsurance pricing has also spurred a series of reinsurance initiatives at various levels of sub-national and national governments. Since the insurance premium charged to property owners is to a greater extent a function of global reinsurance prices, national governments are becoming more and more concerned with the availability and affordability of such reinsurance coverage. Figure 3: The Development in Catastrophe Reinsurance Prices [ Index: 1990 = 100 400 350 300 250 200 150 100 50 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Source: Guy Carpenter & Company, The world catastrophe reinsurance market: 2001 While some programs such as FHCF, CEA, TCIP, HHRF, and Norsk Naturskadepool rely on private reinsurance markets for their reinsurance coverage, others are directly backed by their governments, as is the case with CCR. There are also cases when both private and government reinsurance capacities are used (JER). Another critical consideration in the design and management of a reinsurance program is the level of reinsurance to be purchased. This decision has an impact on the expected survivability of a catastrophe insurance pool, on the speed at which it would accumulate its surplus, and on the affordability of reinsurance or insurance coverage it provides. For instance, CEA is among the safest insurance programs in the world, 56 however, it is also one of the most expensive ones, as it maintains enough claims paying capacity for surviving a 1-in-800 year event. Being the least expensive catastrophe insurance program in the world, TCIP is on the other end of the spectrum, as it buys just enough reinsurance to survive a 1-in-170 year event (falling on the lower end of the investment grade scale for commercial insurers). Summary Creation of additional domestic reinsurance capacity for catastrophe risk and risk aggregation at sub- national or national level prior to its transfer to international reinsurance markets appear to be among the most important functions played by specialized catastrophe insurance programs. Most programs focus on the natural hazard of the greatest concern, often ignoring lesser risks of natural disasters. A program can be regional, or national, depending on the geographical spread of the respective risk across the country. Coverage for residential buildings and contents against the risk of natural disasters typically represents the main focus of catastrophe insurance programs. As a rule, coverage offered carries rather low deductibles but caps the limits at a low level as well to avoid crowding out the private insurance market and limit program's reinsurance costs. Although rates tend to vary insignificantly and in some cases are kept flat, they are most frequently tied to the characteristics of the risk, and are not subsidized. Compulsory catastrophe coverage does not guarantee full compliance. However, it is likely to significantly raise insurance penetration among homeowners compared to alternative solutions. In countries where homeowners are free to choose whether to participate or not, most remain uninsured. While mitigation is typically not a major focus of catastrophe insurance programs, they play an important role in the overall reduction of risk in disaster prone countries. Their positive impact on mitigation is achieved through price discovery of risk and an option to deny coverage to dwellings which do not comply with building codes. Management of catastrophe insurance programs is in one way or another held accountable to government and affected by government policies, although the extent of government influence varies. It appears that the most common way for government to exercise its authority over such programs is through majority representation on the Board, with the responsibility for day-to-day operations delegated to a private insurer. Sales and servicing are typically conducted through the established distribution networks of private primary insurance companies and their agents. Catastrophe coverage is often tied to the basic fire policy (through an endorsement), with the natural hazard risk `passed through' to a special insurance or reinsurance pool. There are some cases, however, when catastrophe coverage is decoupled from other standard coverages to increase its affordability. 57 REFERENCES Alatec-Harris-Tym Group and Angel Lazaro and Associates. 1993. `Interim Report: Baguio - Volume 1'. Earthquake Rehabilitation Program, Urban Planning Studies: Baguio and Dagupan. Project reference ALA/PH/9021. Baguio City. Atkins, J P, S Mazzi and C D Easter, 2000. A Commonwealth Vulnerability Index for Developing Countries: The Position of Small States. Commonwealth Economic Paper Series 40, Commonwealth Secretariat, London. Balisacan, A M. 1999. Causes of Poverty: Myths, Facts and Policies. Quezon City: School of Economics. University of the Philippines. Balisacan, A M, R G Edillon and G M Ducanes. 2002. `Poverty mapping and targeting for KALAHI- CIDSS'. Final report. Benson, C. 1997. `The Economic Impacts of Natural Disasters in the Philippines'. ODI Working Paper No. 99. London: Overseas Development Institute. Benson, C and E J Clay. 2003. `Economic and Financial Impacts of Natural Disasters: an Assessment of Their Effects and Options for Mitigation: Synthesis Report'. London: Overseas Development Institute, May. CNDR. 2000. `Bayanihan: Building Multi-Sectoral Networks for Disaster Prevention, Mitigation and Preparedness'. Terminal Report. Manila: Corporate Network for Disaster Response. CNDR/USAID. 2000. `Bayanihan: A Project Management Manual on Multi-Sectoral Approach to Disaster Prevention, Mitigation and Preparedness'. Manila: Corporate Network for Disaster Response and United States Agency for International Development. Fernandez, C and J Gordon, 1993. `Natural Disasters and their Human Consequences - Overcoming the Vacuum between Humanitarian Aid and Long Term Rehabilitation'. In P A Merriman and C W A Browitt (eds), Natural Disasters: Protecting Vulnerable Communities. Conference Proceedings. London: Thomas Telford. Heijmans, A. 2001. `Vulnerability: A Matter of Perception', Disaster Management Working Paper 4/2001. London: Benfield Greig Hazard Research, University College of London. IFRC. 2002. World Disasters Report 2002. Geneva: International Federation of Red Cross and Red Crescent Societies. JDI/ECFA/DEVMAN. 1993. Disaster Management: The Ormoc Experience ­ Analysis of Survey Results Conducted among Victims of the Flash Flood in Ormoc City. Luna, E M. 2000 "Bayanihan": Building Multi-Sectoral Partnership For Sustainable Disaster Prevention Mitigation and Preparedness - An Impact Evaluation.' Manila: Corporate Network for Disaster Response. NLUC. 2000. National Framework for Physical Planning 2001-2030. Manila: National Land Use Committee, National Economic and Development Authority. Nolledo, J N. 1991. The 1991 Local Government Code with Basic Features. Manila: National Book Store, Inc. PCARRD/DA. 2001. El Niño Southern Oscillation: Mitigating Measures. Los Banos, Laguna: Philippine Council for Agriculture, Forestry and Natural Resources, Research and Development, Department of Science and Technology and Department of Agriculture. Philippine NEDA. No date. Economic Impact of Disasters. Manila: National Economic Development Authority. PHIVOLCS. 1991. Pinatubo Volcano Wakes from Four Century Slumber. Quezon City: Department of Science and Technology, Philippine Institute of Volcanology and Seismology. Rantucci, G. 1994. Geological Disasters in the Philippines: The July 1990 Earthquake and the June 1991 Eruption of Mount Pinatubo ­ Description, Effects and Lessons Learned. Report in cooperation with the Philippine Institute of Volcanology and Seismology (PHIVOLCS). Rome: Directorate General for Development Cooperation, Italian Ministry of Foreign Affairs. 58 UNDRO (1990) Preliminary Study on the Identification of Disaster-Prone Countries Based on Economic Impact, United Nations Disaster Relief Organization, New York/Geneva. USACE. 1994. Mount Pinatubo Recovery Action Plan Abbreviated Version: Eight River Basins, Republic of the Philippines. Portland District: US Army Corps of Engineers. World Bank. 1998. Philippines Social Expenditure Priorities. Report No. 18562-PH. Washington, DC: East Asia and Pacific Region, The World Bank. World Bank. 2000. Philippines Growth with Equity: The Remaining Agenda. Washington, DC: The World Bank. World Bank. 2001. Philippines Poverty Assessment. Volume 1: Main Report. Washington, DC: Poverty Reduction and Economic Management Unit, East Asia and Pacific Region, World Bank, May. World Bank. 2002. Memorandum of the President of the International Bank for Reconstruction and Development and the International Finance Corporation to the Executive Directors on a Country Assistance Strategy of the World Bank Group for the Republic of the Philippines. Report No 24042-PH. Washington, DC: Philippines Country Management Unit, East Asia and Pacific Region, The World Bank. 59 List of Interviewees Department of Agrarian Reform (DAR) · Hon. Roberto M. Pagdanganan, Secretary Department of Agriculture (DA) · Mr. Edward Sana, Undersecretary · Ms. Bess Lim, Division Chief, Management Information Division Department of Budget and Management (DBM) · Mr. Mario Relampagos, Undersecretary · Ms. Carmencita N Delantar, Director, Regional Coordination Service Department of Environment and Natural Resources (DENR) · Ms. Linda SD. Papa, Director, Information Management Department, National Mapping and Resource Information Agency (NAMRIA) Department of Finance (DOF) · Ms. Juanita D. Amatong, Undersecretary · Mr. Roberto B. Tan, Assistant Secretary, International Finance Group Department of Interior and Local Government (DILG) · Ms. Teresita M. Mistal, Director, Bureau of Local Government Development (BLGD) · Ms. Angelina Coitz, Division Chief, BLGD · Ms. Thelma Abdulrahman, Section Chief, BLGD · Ms. Matilde Go, Local Government Operations Officer, BLGD Local Government Units (LGUs) Navotas · Mr. Tobias Reynaldo M. Tiangco, Mayor · Mr. Danilo S. Dar Santos, Barangay Affairs Assistant, Disaster Action Officer · Mr. Rosendo D. Calayag, Former Municipal Development Planning Officer & Consultant for Special Development Project · Mr. Chito M. Antonio, Assessment Officer Marikina · Hon. Marides Fernandos, Mayor · Mr. Alfonso Espiritu, City Engineer Department of National Defense (DND) · Brig. Gen. Melchor P. Rosales, Executive Officer, National Disaster Coordinating Council and Administrator, Office of Civil Defense · Mr. Arnel C. Capili, Regional Director, National Capital Region, National Disaster Coordinating Council · Mr. Allan S. Virtucio, Chief, Mitigation and Preparedness Section, Operations Division, Office of Civil Defense · Mr. Edgar Ollet, Chief, Planning Division, Office of Civil Defense 60 Department of Science and Technology (DOST) Philippine Institute of Volcanology and Seismology (PHIVOLCS) · Mr. Renato U. Solidum, Jr., Ph.D., Director · Ma. Leonila P. Bautista, Associate Scientist · Dr. Raymond Punongbayan, Former Director, and NDCC member Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA) Department of Social Welfare and Development (DSWD) · Ms. Rhoda Yap, Assistant Secretary · Mr. Delfin Salamat, Budget Officer V, Chief of Budget Division · Mr. Rey Camnin, Kalahi-CIDSS · Mr. Milosil Cruz, National Operations Office Government Service Insurance Systems (GSIS) · Mr. Amalio A. Mallari, SVP-General Insurance Group Metro Manila Development Authority (MMDA) · Mr. Ramon J. Santiago, Director for Special Operations Insurance Commission · Mr. Eduardo Malinis, Commissioner · Ms. Vida Chong, Deputy Commissioner National Anti-Poverty Commission (NAPC) / Philippine National Red Cross (PNRC) · Mr. James Sian, Manager, Disaster Management Service, PNRC, and Sectoral Representative, NAPC · Mr. Blenn Hueglas, former Sectoral Representative, NAPC · Dr. Clarita Carlos, Director, PNRC (and Professor, Department of Political Science, University of the Philippines; former President, National Defense College of the Philippines) National Economic and Development Authority (NEDA) · Mr. Raphael Lotilla, Undersecretary · Mr. Emmanuel Torrent, Supervising Economic Development Specialist, Social Development Staff · Mr. Ramon M. Falcon, Supervising Economic Development Specialist, Housing Division, Social Development Staff United Nations Development Programme (UNDP) · Mr. Terence Jones, Resident Representative · Ms. Ricarda L. Rieger, Deputy Resident Representative · Ms. Jay-Ann A. Arandia, Programme Associate, Environment · Ms. Jean Chu, Consultant · Ms. Amelia D. Supetran, Portfolio Manager, Environment 61 Asian Development Bank (ADB) · Mr. Patrick Safran, Planning and Policy Officer, Strategy and Policy Department · Mr. Joseph A. Weinstock, Senior Environmental Specialist, Regional and Sustainable Development Department Development Bank of the Philippines (DBP) · Mr. Vitaliano N. Nañagas, II, Chairman Academe · Dr. Clarita Carlos, Professor, Department of Political Science, University of the Philippines (former President, National Defense College of the Philippines) (and Director, Philippine National Red Cross) Private Sector · Mr. Redentor Magat, BPI/MS Insurance Corporation · Mr. William Cua, Malayan Insurance · Mr. Manuel Quijano, Insurance Co. of North America · Mr. Celestino L. Ang, Prudential Guarantee and Associates, Inc. · Mr. Lito Madrasto, Executive Director, Philippine Constructors Association · Ms. Aguenza, President, Republic Planters Bank · Atty. Senga, Republic Planters Bank · Mr. Manuel `Noli' Acosta, First VP ­ Credit Supervision Division, Rizal Commercial Banking Corporation · Mr. Alberto Lim, President, Corporate Network for Disaster Response · Ms. Mayet Alcid, Corporate Network for Disaster Response · Mr. Kenneth Y. Hartigan Go, M.D., Executive Director, The Zuellig Foundation, Inc. · Ms. Nonette C. Climaco, Editor-in-Chief, Philippine Business, Makati Business Club · Fr. Jose T. Villarin, PhD, Head, Climate Studies Division, Manila Observatory · Mr. Daniel J. McNamara, S.J.P.H.D., Director, Manila Observatory · Mr. Manolito P. Madrasto, Executive Director, Philippine Constructors Association, Inc. Others · Mr. Noboru Ikenishi, Development Planning Dept., Division Planning, Pacific Consultants International · Mr. Yutaka Koike, Deputy Team Leader / Geotechnical, Engineering, Earthquake Disaster Management, MMEIRS Earthquake Impact Reduction Study for Metropolitan Manila 62 33822 Figure 1. Areas at high risk due to storm surge v 2 Tarul, 1908 Lasay, 1972 San Vicente, 1922 Curriman, 1962 Viyan, 1970 Banger, 1969 Dilasag, 1978 Orami, 1982 Saler, 1981 Merung, 1978 Infante, 1983 Manila, 1954, Paracale, 1925, 1970 1965, 1971, 1972 Garshiturane, 1981 Batangas, 1981 Tinembas, 1981 Calapan, 1984 Gigmets, 1968 Sta. Cruz, 1968 Lagaspi, 1975 Orr. Mindoro, 1981 Sarsagen, 1970, 1983 Busuanga, 1981 Dolores, 1914 Catholagen, 1978 Culien, 1981 Herungan, 1987 Hernani, 1897 Sacalad, 1974 Yacluban, 1912 Abuyay, 1972 Tangub, 1971 Nenuz Is., 1984 Consulacian, 1912 Surigan, 1984 Argon, 1975 Tanday, 1976 Dipulag, 1974 Mabaini, 1984 (Source: Gonzalez, 1994) Figure 2. Major Flood Prone Areas in the Philippines (Source: Balce et al, 1994) Figure 3: Philippines GDP performance and major natural disasters Mt P inat ubo El Nino and Asian El Nino Eart hquake eruption ElNino financialcrisis 10.0 5.0 0.0 ) (% rate -5.0 growth Real -10.0 -15.0 -20.0 GDP at market price Agricultural GDP Non-agricultural GDP Figure 4. Examples of damage, 1990 Earthquake, Philippines Figure 5. Settlements along Lower Pasig River Credit: Scawthorn Figure 6. Probability of Exceeding Loss due to Typhoon, Earthquake, Flood and Volcano Figure 7. El Niño Southern Oscillation vulnerability map of the Philippines. Darker shade indicates greater vulnerability (Source: Bureau of Soils and Water Management, Department of Agriculture). Figure 8. La Nina vulnerability map of the Philippines. Darker shade indicates greater vulnerability (Source: Bureau of Soils and Water Management, Department of Agriculture). Figure 9. Temperature increase in the Philippines in the 20th century. Red line indicates a running 10-year mean. 26.8 26.6 26.4 C) o 26.2 ( eru 26.0 25.8 at erp 25.6 emT25.4 25.2 25.0 24.8 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 Figure 10. Observed sea level rise in select ports in the Philippines. 7400 7300 7200 ) m m( veleL 7100 eaS 7000 Manila 6900 Legaspi Davao 6800 1965 1970 1975 1980 1985 1990 1995 2000 Figure 11. Affected areas in Metro Manila due to projected sea level rise. Green and red regions indicate inundated areas for 1 and 2-meter sea level rise scenarios, respectively. (Source: U.S. Country Studies and NAMRIA). Figure 12. Vulnerability of the Philippines due to temperature increase in the latter half of the 21st century. Darker shade indicates greater vulnerability. Figure 13. Vulnerability of the Philippines due to changes in rainfall patterns in the latter half of the 21st century. Darker shade indicates greater vulnerability. Figure 14. Annual pattern of typhoon incidence from 1950 to 1995. Units indicate typhoon hits (Source of typhoon trajectory data: PAGASA).