Protection Gaps in Insurance for Natural Hazards and Retirement Savings in Asia

1. Protection gaps in insurance for natural hazards

Abstract

This chapter examines the drivers of protection gaps in natural hazard insurance in Asia and possible approaches to addressing these gaps. It provides an overview of exposure to natural hazard risk in emerging and developing Asia and Pacific countries and the magnitude of protection gaps; outlines the main drivers of protection gaps in this region; and discusses possible policy, regulatory, supervisory and market approaches to addressing protection gaps in natural hazard insurance.

There are significant “protection gaps” in insurance for natural hazard risks in many countries, particularly in emerging markets and developing countries. Natural hazards such as floods, earthquakes and cyclones can lead to significant damage to buildings and infrastructure, and losses of revenues and income linked to economic disruptions. The damages and losses resulting from these types of relatively infrequent but large-scale events have generally been challenging to insure as households and businesses will often underestimate the need for financial protection and as insurers may be reluctant to provide coverage for large-scale, correlated and uncertain risks.

Protection gaps can burden households and businesses with losses that they are unable to absorb, and recovery and rebuilding needs that they are unable to fund. Insufficient funding can slow recovery and reconstruction and prolong economic disruption, increasing the ultimate cost. Protection gaps can also create pressures on public finances as a result of a reduction in revenues linked to economic activity and an increase in expenditures linked to recovery and reconstruction. There could also be implications for financial stability if uninsured damages and losses lead to widespread defaults on outstanding loans and mortgages.

An increase in the frequency and/or intensity of many types of natural hazards could exacerbate protection gaps leading to larger and more regular losses. Increased losses could force insurance and reinsurance companies to charge more for the coverage they provide, making insurance coverage less or un-affordable for an increasing share of the population.

This chapter examines the drivers of protection gaps in natural hazard insurance and possible approaches to addressing these gaps, with a focus on Asia.1 Asia is one of the regions most exposed to natural hazards and vulnerable to climate risks.2 The chapter examines protection gaps in natural hazard insurance for households and businesses specifically, and does not consider the role of insurance markets in providing financial protection to governments or public assets3, nor does it examine disaster risk reduction or climate adaptation, beyond noting their important contribution to supporting insurability and reducing protection gaps.

The first section provides an overview of the exposure to natural hazard risk in Asian countries and the magnitude of protection gaps. The second section outlines the main drivers of protection gaps in insurance for natural hazard risks in this region. The third section discusses possible policy, regulatory, supervisory and market approaches to addressing protection gaps in natural hazard insurance. A final section provides some conclusions.

1.1. Natural hazards and protection gaps

Emerging and developing Asia and Pacific countries are exposed to a broad range of natural hazards, including earthquakes, floods, cyclones, convective storms and wildfires.4 Between 2000 and 2023, natural hazards have led to annual average losses of approximately USD 48.4 billion across the region, driven by floods (46% of losses), earthquakes (23%) and cyclones (22%) (Figure 1.1).

Figure 1.1. Average annual economic losses by hazard and subregion (2000-23)

Note: Southeast Asia includes losses reported in Cambodia, Indonesia, Lao People’s Democratic Republic, Malaysia, Myanmar, the Philippines, Thailand, Timor-Leste and Viet Nam. South Asia includes losses reported in Afghanistan, Bangladesh, Bhutan, India, the Maldives, Nepal, Pakistan and Sri Lanka. Pacific includes losses reported in Fiji, the Marshall Islands, Micronesia, Papua New Guinea, Samoa, the Solomon Islands, Tonga, Tuvalu and Vanuatu. East Asia includes losses reported in the People’s Republic of China, the Democratic People’s Republic of Korea and Mongolia. Central Asia includes losses reported in Kazakhstan, Kyrgyzstan and Tajikistan. No economic losses were reported in other countries during this period. “Other secondary perils” includes droughts and landslides.

Source: OECD calculations based on data provided by Swiss Re (Swiss Re, sigma database. All rights reserved.).

Countries classified as Upper Middle Income or Lower Middle Income incurred most of the reported losses, 75.7% and 24.2%, respectively. This reflects higher asset values, and potentially data gaps in lower income countries.5 Twelve emerging and developing Asia and Pacific countries accounted for practically all (98.9%) of the reported losses (Figure 1.2).

Figure 1.2. Economic losses by country (2000-23, cumulative)

Source: OECD calculations based on data provided by Swiss Re (Swiss Re, sigma database. All rights reserved.).

Average annual economic losses from natural hazards were 165% higher in the most recent five-year period (2019-23) relative to the first five years of the century (2000-2004), driven by a significant increase in losses resulting from weather-related natural hazards. Average annual economic losses from weather-related losses increased from USD 10.8 billion (2000-2004) to USD 42.4 billion (2019-2023) (Figure 1.3).

Figure 1.3. Economic losses by hazard (2000-23, emerging and developing Asia and Pacific)

Note: “Other” includes wildfire, convective storms, droughts and landslides.

Source: OECD calculations based on data provided by Swiss Re (Swiss Re, sigma database. All rights reserved.).

There is an expectation that the frequency and intensity of many natural hazards across the region is going to increase. Flood risk could increase as a result of a projected large increase in annual precipitation across North, South and East Asia; more frequent heavy and intense precipitation episodes across South, Southeast and East Asia and the Pacific; an increase in precipitation accompanying tropical cyclones; sea-level rise in coastal areas; and the potential for glacial lake outburst floods in the Hindu Kush Himalaya region (Figure 1.4) (Shaw et al., 2023[1]). Wind and flood damage from tropical cyclones could increase due to an increase in the frequency of intense tropical cyclones in the western North Pacific (Shaw et al., 2023[1]). A number of emerging and developing Asia and Pacific countries are ranked among the countries most exposed6 to significant climate change, including Bangladesh, India, Indonesia, the Democratic People’s Republic of Korea, the Maldives, Myanmar, Pakistan, Papua New Guinea, the Philippines, Sri Lanka, Timor-Leste, , Viet Nam and a number of small island states (Kiribati, the Marshall Islands, Micronesia, Nauru, Palau, the Solomon Islands, Tonga, Tuvalu and Vanuatu) (Chen et al., 2025[2]).

Figure 1.4. Projected impact of climate change on natural hazards in Asia

Source: OECD based on the findings from Shaw et al. (2023[1]) and ADB (2024[3]).

There are significant protection gaps in most emerging and developing Asia and Pacific countries, often defined as the difference between overall or economic losses and insured losses.7 Between 2000 and 2023, approximately 5%-7% of overall natural hazard losses in emerging and developing Asia and Pacific countries were insured.8 There was no significant difference in terms of the share of natural hazard losses insured in countries classified as Upper Middle Income and Lower Middle Income Countries9 although there were substantial differences across individual countries (Figure 1.5). The estimated share of losses insured was below 5% in just over half of the countries examined.

There were also some differences in the share of losses insured among the different types of natural hazards. The share of earthquake losses insured was significantly lower than other types of natural hazards (Figure 1.6).

Figure 1.5. Estimated share of economic losses insured by country (2000-23)

Note: The reported value is the average between two approaches to measuring the share of losses insured (i.e. an approach including all available data on insured and economic losses and an approach including only loss years where both insured and economic loss estimates are provided). There were no reported insured losses in Afghanistan, Bhutan, Kazakhstan, Kyrgyzstan, the Marshall Islands, Micronesia, Mongolia, Samoa, the Solomon Islands, Tajikistan, Timor-Leste or Tuvalu.

Source: OECD calculations based on data provided by Swiss Re (Swiss Re, sigma database. All rights reserved.).

Figure 1.6. Estimated share of economic losses insured by hazard (2000-23, emerging and developing Asia and Pacific)

Note: The reported value is the average between two approaches to measuring the share of losses insured (i.e. an approach including all available data on insured and economic losses and an approach including only loss years where both insured and economic loss estimates are provided). “Other secondary perils” includes wildfire, droughts and landslides (although there were no reported losses from wildfires). Insured and economic losses due to floods in Thailand in 2011 were excluded above given the magnitude of the event and relatively high level of insured losses. Including those losses would lead to an estimate of 8.1% of flood losses that were insured.

Source: OECD calculations based on data provided by Swiss Re (Swiss Re, sigma database. All rights reserved.).

There has been some progress in reducing protection gaps across the region (Figure 1.7). In 2014-23, 6.6% of economic losses were insured, up from 2.6% in 2000-09.

Figure 1.7. Share of economic losses insured (2000-23, emerging and developing Asia and Pacific)

Note: The reported value is the average between two approaches to measuring the share of losses insured (i.e. an approach including all available data on insured and economic losses and an approach including only loss years where both insured and economic loss estimates are provided).

Source: OECD calculations based on data provided by Swiss Re (Swiss Re, sigma database. All rights reserved.).

1.2. Drivers of protection gaps in natural hazard insurance

Several factors that affect the demand for and supply of insurance coverage drive the limited insurance coverage of losses resulting from natural hazards. Limited demand reduces the amount that households and businesses will be willing to pay for natural hazard insurance coverage or their willingness to purchase any coverage at all. Limited supply reduces the amount of coverage that insurance companies are willing to offer or leads them to charge higher premiums. The gap between the demand for coverage and the supply of coverage drives protection gaps. The following sections outline the factors that could limit households’ and businesses’ willingness-to-pay for insurance coverage for natural hazard risks and the factors that could limit insurance companies’ willingness to offer such coverage.

1.2.1. Factors that limit the willingness-to-pay for natural hazard insurance coverage – demand factors

Several factors impact the demand for natural hazard insurance coverage, including the level of awareness among households and businesses of their exposure to natural hazard risks and the availability and benefits of – and need for – specific coverage to protect against these risks. Households and businesses that do not perceive themselves to be at risk of being impacted by natural hazards or are unaware of the need for insurance coverage are less likely to acquire such insurance coverage. Households and small businesses with limited income or revenue may not have the financial capacity to allocate any significant amount of their spending towards insurance premiums for natural hazard insurance coverage. Insurance companies, intermediaries and regulators in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka identified some or all of these factors as drivers of protection gaps in insurance for natural hazards during meetings and/or in their responses to a survey circulated for the development of this chapter.

Limited awareness of exposure to natural hazard risks

Limited awareness among households and businesses of their exposure to natural hazard risk and the potential damages or losses that they could face should they be impacted by a natural hazard is likely to affect the demand or willingness-to-pay for natural hazard insurance coverage. Households and businesses that are unaware of their exposure to natural hazard risks – or that perceive there to be a low risk that they will face damage and losses due to natural hazards – are likely less willing to seek financial protection for these risks or pay any substantial amount for insurance coverage. For example, a recent survey of consumers in Europe (Belgium, Germany, Romania and Spain) found that the perceived low likelihood of being impacted by a natural hazard was a significant reason for not acquiring insurance coverage in all of these countries and the most significant reason in three of them (more important than other factors, such as cost, product complexity or a lack of awareness about coverage availability) (EIOPA, 2024[4]).

A higher level of risk awareness or risk perception usually leads to a greater willingness to implement preparedness measures, including acquiring insurance coverage for natural hazard risks (Yang, Tan and Peng, 2020[5]; Reynaud, Nguyen and Aubert, 2018[6]; Royal and Walls, 2019[7]; Seifert et al., 2013[8]; Cisternas et al., 2024[9]; EIOPA, 2024[4]).10 An individual’s perception of natural hazard risk is influenced by factors such as the person’s level of education as well as age, gender and wealth, although direct experience, particularly recent direct experience with the impacts of a natural hazard event may be one of the most significant drivers of risk perception and demand for insurance (Wachinger et al., 2013[10]; Bronfman et al., 2020[11]; Ren and Wang, 2016[12]; Wang et al., 2012[13]; Gallagher, 2014[14]). For example, one recent survey of consumers in Europe (Belgium, Germany, Romania and Spain) found that the share of households that acquired natural hazard insurance was significantly higher (almost two times higher) among respondents that had had a recent experience with a natural hazard event (64%) relative to those who had not had such an experience (36%) (EIOPA, 2024[4]).

In several emerging and developing Asia and Pacific countries, including Mongolia, Afghanistan, Bangladesh, the Philippines and many of the small island states in the Pacific, a potentially large share of the population has been affected by natural hazards in recent years (Figure 1.8). This is likely to lead to a relatively higher level of risk awareness and perception in Asia than in many other parts of the world where there is more limited experience with natural hazard impacts.

Some insurance companies, intermediaries and/or regulators in Indonesia, Malaysia, Pakistan and Sri Lanka indicated that a lack of risk awareness limits the demand for insurance coverage. In Indonesia, there is some evidence that risk awareness is relatively high, particularly in communities with recent experience (Viverita et al., 2014[15]), although a relatively high level of awareness has not led to the broad implementation of preparedness measures (including the purchase of insurance coverage) (Adhi et al., 2019[16]). In Sri Lanka, there is also some evidence of a high-level of risk perception. For example, a survey by the Disaster Management Centre in Sri Lanka found high levels of awareness and experience with natural hazards among micro, small and medium sized enterprises (MSMEs), while a study that surveyed residents of a highly-exposed community also found high-levels of risk perception among survey respondents (Weerasuriya and Rajapaksha, 2023[17]). In Malaysia, some regions have faced frequent flooding11 which has likely led to a relatively high-level of flood risk awareness in those areas, although some insurance companies noted that households and businesses outside of those regions often perceive themselves to be at low risk. In Pakistan, despite significant exposure to natural hazard risks and recent experience with major floods,12 there is some evidence that risk perception is generally low outside of the areas with recent experience (Shah, Rana and Ali, 2023[18]), including urban areas that have faced few major flooding events in recent years. In Pakistan, earthquake risk perception also tends to be lower than flood risk perception as earthquakes have occurred much less frequently. A lack of risk awareness was not identified as a major impediment to the demand for insurance coverage in the Philippines, as natural hazard risk perception is generally considered to be high given the frequency of natural hazard events. For example, one survey across regions and socio-economic groups in the Philippines found that almost half of the respondents perceived there to be a very high or extremely high likelihood that climate change would significantly impact their household within the next five years (Vinck et al., 2024[19]).

Figure 1.8. Estimated share of the population affected by natural hazards annually in recent years

Note: Calculated as the average number of people affected annually by a natural hazard between 2020 and 2025 (Q1) as a share of total population.

Source: OECD calculations based on data on number affected from Centre for Research on the Epidemiology of Disasters (CRED) (n.d.[20]) and population from World Bank (n.d.[21]).

Limited awareness of availability, need for or benefit of insurance coverage for natural hazard risks

A high-level of awareness or perception of exposure to natural hazard risks may not always translate into the implementation of preparedness measures or the acquisition of insurance coverage. Households and businesses may not be aware that coverage for natural hazard risks is available or that they need to acquire specific coverage to protect against damages and losses from natural hazards. They may be unaware of - or may under-value - the benefits of insurance coverage in providing financial protection against these risks, potentially as a result of a lack of trust in insurance companies. A lack of awareness about the need for – and benefits of – financial protection against natural hazard risks was identified by insurance supervisors, intermediaries and/or insurance companies as a driver of protection gaps in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka.

A lack of awareness of the availability of insurance coverage for natural hazard risks is unlikely to be a significant factor in limiting demand for insurance coverage in most countries, although it could impact demand among some segments of the population. For example, a survey of European consumers (Belgium, Germany, Romania and Spain) found that 18% of the respondents that had not acquired insurance for natural hazard risks were generally unaware that such coverage exists (EIOPA, 2024[4]). One of the insurance companies in Malaysia that responded to a survey for the development of this report indicated that some segments of the population are unaware of the coverage for natural hazard risks that is available from insurance companies. A survey of residents of the Klang Valley in Malaysia found that 68% would be willing to acquire flood insurance if available (Khairunisa et al., 2023[22]), suggesting that some households may not be aware of the availability of flood insurance coverage (or may not have access, as discussed below).

Households and businesses may be unaware of the need for specific coverage against natural hazard risks if they mistakenly believe that they are covered against these losses through their standard property or fire insurance coverage. This is more likely to occur in countries where coverage for some or all natural hazard losses is only available as an optional addition or endorsement to property insurance coverage (“optional add-on”) or as a separate policy (“stand-alone coverage”), rather than as a standard inclusion in property (fire) insurance (“automatic inclusion”). For example, in the United States, where flood insurance is only available as a stand-alone coverage, a recent survey of households found that 56% mistakenly believed that their standard home insurance policy included coverage for flood damages (Insurance Journal, 2024[23]). In the European Union, one consumer survey found that 22% of consumers were unaware of whether their insurance coverage included coverage for natural hazard risks (EIOPA, 2024[4]). For some risks and customers. coverage for natural hazard risks is offered as an optional add-on in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka (Figure 1.9):

• In Indonesia, insurance coverage for natural hazard risks is offered to households as an optional add-on to property insurance coverage, structured as an optional coverage for earthquake, volcanic eruption and tsunami (EQVET), and an optional coverage for tempest, storm, flood and water damage (TSFWD) (which also includes coverage for losses resulting from landslides). Coverage for losses resulting from wildfires is often excluded (although at least one insurer provides this coverage automatically in property insurance for households). Natural hazard insurance coverage for businesses may be offered as property (fire) insurance with the same optional EQVET and TSFWD add-ons available or as an all-risk coverage (“Industrial All-Risk”) that includes coverage for all natural hazard perils unless explicitly excluded. Insurance coverage for wildfire losses is often included in coverage provided to businesses although approaches differ among insurers.

• In Malaysia, property insurance coverage offered to households tends to take one of two main forms: (i) a standard (or “basic”) fire insurance policy where coverage for specific natural hazards can be provided as an optional addition; or (ii) a “houseowner”/“householder” insurance policy that includes coverage for specific natural hazards (hurricane, cyclone, typhoon, windstorm; earthquake and volcanic eruption; and flood). Coverage offered to businesses (particularly SMEs) tends to offer coverage for natural hazard risks as an optional add-on, although a few insurers automatically include coverage for natural hazard risks with an option to opt out of that coverage. There is also an Industrial All-Risk coverage available to businesses which includes coverage for natural hazard risks, unless specifically excluded. Coverage for losses resulting from landslides and wildfires is usually only offered to households and businesses as an optional addition to property insurance coverage.

• In Pakistan, coverage for flood, storm and earthquake losses is automatically included in the property insurance coverage provided to households by some insurers (usually the larger insurers), while smaller insurers may only provide that coverage as an optional add-on. Coverage for landslides, volcanic eruptions and wildfire losses is generally not available, although some insurers offer coverage to households for some or all of these perils, either as an automatic inclusion or optional add-on. For businesses, coverage for flood, storm and earthquake losses may be included automatically or offered as an optional add-on to property insurance coverage, although only a few large insurers offer optional add-on coverage for landslides, volcanic eruptions and wildfire losses, and insurers may choose to not offer coverage for some hazards in high-risk areas.

• In the Philippines, coverage for natural hazard risks is offered to households and businesses as an optional add-on although some insurers indicated that they tend to automatically include coverage for flood, storm and earthquake in the property insurance that they offer. Coverage for losses from landslides and volcanic eruptions is also available as an optional add-on to property insurance policies, although insurers are less likely to include it automatically and may not offer coverage for these hazards in some areas. Insurance coverage for wildfire losses is not generally available (and is also a more limited risk) in the Philippines.

• In Sri Lanka, coverage for natural hazard risks is generally included automatically in the property insurance coverage offered to households and businesses, although the policyholder or insurer may choose not to include coverage for specific perils and practices differ among different insurers. The specific natural hazard perils included in coverage (whether automatically or as an optional add-on) varies across insurers, with some offering bundled coverage (e.g. a “Specific Natural Perils Cover” that typically covers flood, storm and earthquake losses) and others providing coverage for named perils (which may include flood, storm, earthquake as well as tsunami and volcanic eruption). Coverage for landslide and wildfire losses is generally not available to businesses or households.

Figure 1.9. Approaches to offering natural hazard insurance coverage to households and businesses

Awareness of the need for coverage against natural hazard risks may also be lower if households and businesses expect that the government will take on responsibility for recovery and reconstruction and provide financial assistance or compensation to those affected by natural hazards. This is often cited as a concern in OECD countries and there is some evidence that consumers decide not to acquire insurance because they expect to be compensated by governments for any damages due to natural hazards. For example, surveys undertaken in a number of European countries found that between 4% and 15% of uninsured households13 did not acquire insurance because of an expectation that the government will provide compensation (EIOPA, 2024[4]). This concern was raised among some insurance supervisors, intermediaries and/or insurance companies in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka, although it was generally limited to vulnerable segments of society (e.g. rural or low-income households) that are more likely to receive government assistance and less likely to acquire insurance coverage even without an expectation of government assistance.

Awareness of the availability and need for insurance coverage against natural hazard risks may be insufficient to achieve broad levels of coverage if households and businesses have limited awareness of the benefits of insurance coverage for providing financial protection. Awareness of the benefits of insurance is likely to be higher where households and businesses have previous experience with insurance. For example, positive previous experience with insurance, whether for natural hazard losses or other types of losses or expenses, is likely to increase the tendency of a household or business to consider insurance as a means to protect against natural hazard risks (EIOPA, 2024[4]).

Households and businesses in countries with lower levels of insurance penetration will have less previous experience with insurance. The non-life insurance penetration rate in most countries in emerging and developing Asia and Pacific, including in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka, is less than half of the average penetration rate in OECD countries (Figure 1.10). Specific experience with insurance coverage for natural hazard risks is likely to also be low where there are significant protection gaps. For example, in Pakistan, there has been limited experience with insured natural hazard losses, particularly as recent losses have occurred outside of the industrial areas where insurance coverage is more likely to have been acquired.

Figure 1.10. Non-life insurance penetration

Source: Swiss Re Institute (2024[24]), World insurance: strengthening global resilience with a new lease of life, https://www.swissre.com/institute/research/sigma-research/sigma-2024-03-world-insurance-global-resilience.html.

An understanding of the potential benefits of insurance might not be sufficient to encourage take-up of coverage if households and businesses do not trust that insurance will provide an effective form of financial protection at a fair price. For example, a survey of consumers in a number of European countries (France, Greece, Italy and Spain) found that 7.4% to 13.1% of respondents without insurance did not acquire insurance because they do not trust that insurance companies will pay, while a further 9.0% to 11.3% remained uninsured because they do not trust insurance companies to offer fair prices (EIOPA, 2024[4]). In Indonesia, a national survey used for the development of the National Strategy on Indonesian Financial Literacy found that while 62% of respondents understood the benefits of insurance, only 10% of respondents had confidence in the insurance sector (OJK, 2021[25]). A lack of trust in insurance companies’ capacity or willingness to pay claims for natural hazard-related losses in a timely manner was raised by public and private sector stakeholders in Sri Lanka. Insurance companies indicated that misconceptions about coverage, policy terms and claims processes contributed to low levels of trust in insurance. However, a survey by the Insurance Regulatory Commission of Sri Lanka published in 2024 found that 49% of respondents were confident or extremely confident in the general insurance sector (IRCSL and University of Colombo, 2024[26]).

Factors such as gender, age, education, insurance literacy and, particularly, experience with insurance affect trust in insurance (Courbage and Nicolas, 2019[27]). Cultural factors such as religion can also have an impact. For example, some studies have found a higher level of trust in sharia-compliant and takaful insurance coverage in countries with significant Muslim populations (Hassan and Abbas, 2019[28]; Souiden and Jabeur, 2015[29]).

Limited capacity or willingness to pay for natural hazard insurance coverage

Even where there is high awareness of natural hazard risks and an understanding of the availability and benefit of specific insurance coverage for these risks, limited income and competing spending demands may limit the ability of households and businesses to afford insurance coverage. The cost of insurance was identified by insurance supervisors, intermediaries and/or insurance companies as a driver of protection gaps in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka, particularly for low-income households and SMEs. There is some evidence that low-income households are less likely to acquire insurance coverage. For example, in Europe, households with less than EUR 30 000 in annual income were found to be less likely to purchase home insurance than households with higher income (EIOPA, 2024[4]). In one survey of urban residents in Indonesia that had experienced climate-related extreme events, a lack of resources was among the most common reasons for not taking preparedness measures, such as acquiring insurance (Adhi et al., 2019[16]).

The premiums that insurance companies charge for natural hazard insurance coverage depends on: (i) their assessment of risk (i.e. the probability of incurring losses that would be covered by the policy); (ii) distribution and operating expenses (including commissions, underwriting and loss adjustment expenses as well as the cost of reinsurance); and (iii) market and competitive considerations. Lower risk, lower distribution and operating expenses, and more competitive markets can support affordability and reduce insurance costs.

In most countries, insurance supervisors also impose some requirements on insurers that are applicable when establishing premiums for the coverage that they provide. In some countries, supervisors take a principles-based approach that requires insurance companies to charge premiums that are adequate and not unfairly discriminatory – which may be subject to supervisory review either as a condition for approval or at the discretion of the supervisor. In other countries, supervisors impose quantitative requirements, such as a fixed premium (usually based on sum insured), a minimum or maximum premium, or a range of acceptable premium rates. Box 1.1 provides an overview of supervisory requirements and market approaches to premium setting in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka.

Box 1.1. Pricing approaches and supervisory oversight in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka

The insurance supervisors in Indonesia, the Philippines and Sri Lanka require insurers to submit new insurance products providing coverage for natural hazard risks for prior approval. In Malaysia, prior approval is only required for products with premium rates that deviate by more than 30% from the Revised Fire Tariff (see below). In Pakistan, there are no prior approval requirements for insurance products that provide coverage for natural hazard risks.

In Indonesia and the Philippines, insurance coverage for natural hazard risks is subject to pricing guidance or references. In Indonesia, pricing for property and natural hazard insurance coverage is based on occupancy and construction characteristics (for fire), hazard zone (for EQVET and TSFWD) and number of floors (for EQVET) – although insurance companies can deviate from the pricing guidance. In the Philippines, a minimum premium rate is required for property (fire) and natural hazard (typhoon, flood and earthquake) insurance coverage and insurers are required to set premiums at or above the minimum rate (subject to fines for non-compliance). Due to market dynamics (competitive pressure) and risk assessment challenges (see below), the minimum rates are very often applied as fixed rates by insurers in Indonesia and the Philippines. Insurers in Indonesia and the Philippines do not generally charge higher premiums in higher risk areas, although some Indonesian insurers impose higher deductibles in high-risk areas (which is not permitted in the Philippines).

In Malaysia, as part of a broader initiative to eliminate fixed (tariff) pricing, pricing consistent with the Revised Fire Tariff must be offered although insurers can also offer products that are not subject to those rates (i.e. non-tariff products). As noted above, without prior approval, pricing of non-tariff products cannot be more than 30% lower than the pricing of the tariff products. Many insurers tend to price coverage for natural hazard risks in both tariff and non-tariff products at the fixed rate of sum insured defined for tariff products, although insurers do have the flexibility to price natural hazard coverage at rates that are 30% above or below the fixed tariff rate (without prior approval). Some insurers have been introducing higher rates and (particularly) higher deductibles for flood insurance coverage in high-risk areas.

In Pakistan and Sri Lanka, insurers are free to choose the premium rates that they apply for property and natural hazard insurance coverage, although pricing approaches may be subject to supervisory review in Sri Lanka as part of product approval. In Sri Lanka, some insurers charge higher premiums in high-risk areas (and lower premiums in low-risk areas) although higher deductibles (or sub-limits and exclusions) appear to be more commonly applied to address higher levels of risk. In Pakistan, higher deductibles are often required for property insurance coverage that includes coverage for natural hazard risks. Insurers in Pakistan also vary premium pricing based on risk characteristics, particularly for commercial clients.

Households and small businesses with limited income or revenue may not be able to afford the cost of insurance coverage given other spending demands while, where premiums vary with risk, those located in high-risk areas or in poorly constructed buildings may face high premium costs beyond their capacity to pay.

The cost of coverage including coverage for natural hazard risks appears unlikely to create “affordability stress” for most households.14 This is based on a limited review of online offers of household property insurance coverage in Indonesia, Malaysia, Pakistan and the Philippines.15 Average household property insurance premiums would appear to cost less than two weeks of average household (gross) income in all four countries (and less than one week in the case of Malaysia). However, the estimated average household insurance premium could cost slightly more than four weeks of average household income for the 20% lowest income households in the Philippines and just under four weeks of the minimum wage income in Indonesia for homes with lower constructions standards (Figure 1.11, left-side), although this would likely be offset by lower property values that would result in lower premiums.

However, an “affordable” premium may not be sufficient to incentivise purchase where property insurance coverage or natural hazard insurance coverage is voluntary. Households or businesses may not be willing to pay for coverage even if that coverage is affordable, which may be driven by the factors related to perceptions of exposure to natural hazard risk or the benefits of acquiring specific insurance coverage for those risks. In one of the surveys of European consumers, uninsured households in France, Greece, Italy and Spain provided an indication of the amount that they would be willing to pay for household insurance coverage (including natural hazard insurance coverage). While the amount varied across the four countries, an overall preference equivalent to approximately 0.95% of gross income appeared to be broadly acceptable (EIOPA, 2024[4]). The estimated average premium cost, for the average home in Indonesia, Malaysia, Pakistan and the Philippines, would be significantly higher than 0.95% of income in most cases, and up to 7 or 8 times this amount among low-income and high-risk households in Indonesia and low-income households in the Philippines (Figure 1.11, right-side).

Figure 1.11. Illustrations of the cost of insurance relative to affordability and willingness-to-pay

Note: The estimates of premiums costs were calculated based on online premium calculators and premium guidance in all four countries. The following rates per 1 000 in coverage (2.28 (Indonesia, low risk), 3.22 (Indonesia, high risk), 2.66 (Malaysia), 1.83 (Pakistan) and 2.98 (Philippines)) were applied to the following house values: IDR 800 million (Indonesia), MYR 370 000 (Malaysia), PKR 11.5 million (Pakistan) and PHP 4.25 million (Philippines). Data on income distribution for Malaysia (Department of Statistics Malaysia, 2022[30]), Pakistan (Pakistan Bureau of Statistics, 2020[31]) and the Philippines (Philippines Statistics Authority, 2023[32]) and data on basic and median salaries in Indonesia (reported in (Mulya, 2023[33])) were used to calculate weekly income and willingness-to-pay (by applying an estimate of willingness-to-pay for coverage of 0.95% of income, derived from (EIOPA, 2024[4])). Estimates of premium costs were not available for Sri Lanka.

Given significant exposure to natural hazards, the cost of natural hazard coverage often accounts for a significant share of the premium costs for household and business property insurance coverage in many Asian countries. For example, under the Revised Fire Tariff in Malaysia, coverage against natural hazards (particularly floods) accounts for approximately 30-60% of the cost of household property insurance depending on the type of construction. In the Philippines, the minimum premiums for earthquake, typhoon and flood coverage account for 60% of the cost of premiums for homes. In Indonesia, the cost of TSFWD and (particularly) EQVET coverage accounts for 60-70% of the cost of property insurance for households in low-risk earthquake zones and more than 80% of premium cost in high-risk zones (for homes with steel or wood construction).

Where coverage for natural hazard risks is an optional add-on, the significant additional cost of this coverage could be an impediment to purchasing this coverage among price-sensitive and lower-income consumers. The additional cost of optional natural hazard coverage was identified as a driver of protection gaps by some insurers in Malaysia, where approximately 50% of households with property insurance choose to acquire the optional coverage for natural hazard risks. This was also identified as a driver of protection gaps in Indonesia where only a small share of households that acquire property insurance choose to acquire the optional coverage for natural hazard risks. For example, one Indonesian insurance company indicated that less than 5% of its household customers had acquired the optional coverage for flood risk.

1.2.2. Factors that limit the willingness of insurance companies to provide natural hazard insurance coverage – supply factors

Given the potential for catastrophic losses, assuming natural hazard risks can create challenges for insurance companies. Natural hazard risks can impact large areas and many households and businesses simultaneously. The potential for large and correlated losses reduces the benefits of diversification upon which the insurance business model is based and requires insurance companies to set aside significant reserves and acquire substantial reinsurance coverage to ensure that they have sufficient resources to respond to those losses. The relatively low frequency of natural hazards and the complexity in assessing potential impacts leads to greater uncertainty about the potential losses that might be incurred. Higher levels of exposure or vulnerability to natural hazard risks, more limited data and tools for risk assessment and costly reinsurance ultimately increase the risk and/or cost in assuming natural hazard risks and can reduce insurers’ willingness to provide natural hazard insurance coverage. Insurance companies, intermediaries and regulators in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka identified some or all of these factors as drivers of protection gaps in insurance for natural hazards.

Highly exposed communities, businesses and households

Insurance companies may not want to provide coverage for households or businesses facing a high risk of severe or frequent losses. Specific communities, or particular households or businesses, may be significantly exposed to either frequent or severe natural hazard risks and may have homes or businesses that are poorly constructed and therefore more vulnerable to damage from natural hazards. Where permitted, insurers could charge higher premiums or impose higher deductibles. However, higher premium rates may be unaffordable or unattractive in highly competitive markets while higher deductibles could have reputational impacts if natural hazard losses are rarely reimbursed because they fall below the policyholder deductible. Insurance supervisors and some intermediaries and insurance companies in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka indicated that the availability of affordable coverage in high-risk areas is a driver of protection gaps.

Some insurance supervisors impose limitations on the premium rates that insurance companies can apply to the insurance coverage of natural hazard risks that they offer to households and businesses (Box 1.1). This includes guidance on minimum and maximum premium rates in Indonesia, a requirement for prior approval of premium rates that deviate substantially below the tariff rate in Malaysia and minimum premium rates in the Philippines. While insurers are authorised to set premium rates for higher-risk businesses and households at higher rates, highly competitive insurance markets have, in general, led insurers to apply minimum (or tariff) rates to all policyholders and make limited use of the flexibility to charge higher rates, particularly in Indonesia and the Philippines (although as noted above, some insurers in Indonesia and Malaysia will impose higher deductibles or limit the sum insured for specific hazards when offering coverage to higher-risk households and businesses).

As a result of this competitive dynamic, insurers may not offer any coverage for natural hazard risks or specific natural hazard risks in areas where the prevailing market rates are likely to be inadequate to cover probable claims payments:

• In Indonesia, insurers may not offer coverage in areas exposed to annual (or regular) river or coastal flooding or in communities in the highest earthquake risk zone or with recent flood or earthquake claims experience. This is likely also a reflection of weaknesses in risk management which hinder the ability of insurance companies to offer affordable coverage.

• In Malaysia, some insurers do not offer coverage for flood or landslides in areas known to be at high risk. One recent survey of business in Malaysia, for example, found that 17% of SMEs with recent flood experience (within three years) were unable to receive a quote for flood insurance coverage (World Bank and Bank Negara Malaysia, 2024[34]).

• In the Philippines, households and businesses may have difficulty accessing insurance coverage in communities near active fault lines or subject to frequent ground-shaking, communities near active volcanoes and in areas exposed to flood or typhoon risks. For example, one insurance website reviewed for this report clearly stated that flood-prone areas could not be insured against flood and typhoon risks.

Flexibility in setting premiums, deductibles and insured limits does not necessarily ensure coverage availability in markets with price-sensitive consumers. Insurers in Pakistan and Sri Lanka are free to charge higher premiums, impose higher deductibles and limit the sum insured for specific hazards in high-risk areas. In Sri Lanka, insurance companies consider a broad range of criteria for determining which areas might be at high risk that capture communities in coastal regions exposed to coastal flooding and tsunamis, valleys and low-lying areas exposed to flooding, and mountainous regions exposed to earthquakes and landslides. In Pakistan, premium pricing that reflects higher risk is more often applied in property insurance for businesses. However, in both countries, insurers can and do decline coverage for specific hazard risks to households and businesses located in risk-prone areas.

Weaknesses in land-use and development planning, deficient or poorly enforced building codes and insufficient investment in risk reduction and adaptation will clearly have an impact on the pervasiveness of high-risk communities, households and businesses.16 Insurance companies, intermediaries and regulators in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka emphasised the criticality of improving risk management, risk reduction and adaptation for addressing protection gaps in natural hazard insurance.

Difficulties in understanding or quantifying natural hazard risk

Insurance companies may be unwilling to provide coverage for natural hazard risks if they face challenges in understanding or quantifying the potential frequency and severity of losses under any coverage provided. Given the relatively low frequency of natural hazards compared to other types of insured risk, such as motor vehicle accidents, natural hazard risks are difficult to quantify using traditional statistical and actuarial methods. The uncertainty of loss frequency and the potential for severe losses may discourage insurers from providing coverage for these risks. The changing frequency and severity of many weather-related risks exacerbates this uncertainty.

The insurance sector uses a number of approaches to assessing natural hazard risk to support underwriting and pricing decisions. These include their own past claims data and loss experience, hazard or risk maps and zones provided by local or national governments or other organisations, risk ratings developed by local or international intermediaries and reinsurance companies, and catastrophe models developed internally or provided by intermediaries, reinsurers or specialised catastrophe modelling companies. These approaches vary in terms of their accuracy and availability:

• Catastrophe models, which are developed using real and simulated natural hazard events and data on buildings and structural vulnerabilities, provide the most sophisticated approach to assessing natural hazard risks and can (to some extent) be calibrated to take into account climate change through “climate-conditioning.”17 However, models are not always available for assessing risks in emerging and developing markets and coverage for some types of natural hazards risks is much more limited (Figure 1.12). They can also be costly to acquire, particularly for smaller insurers.

• Hazard maps and risk zones vary in terms of quality. In some countries, developed and emerging, insurance companies have limited trust in the accuracy of the hazard and risk maps developed by governments.

• The value of past claims data for assessing risk and underwriting coverage is generally more limited for infrequent events such as natural hazard risks. Relying on claims data will also pose particular challenges in countries where insurance coverage of natural hazard risks is limited (thereby limiting the number of claims) and where climate change is having significant impacts on the frequency and/or severity of weather-related hazards.

Figure 1.12. Catastrophe model coverage in Asia-Pacific

Note: Dark blue shading indicates that three or more models are available. Medium blue shading indicates that two models are available. Light blue shading indicates that one model is available.

Source: OECD based on data on model availability from Aon (2024[35]), Moody’s (2025[36]), (2025[37]), (2025[38]), (2025[39]), Verisk Analytics (2025[40]), CATRisk Solutions (n.d.[41]), Risk Frontiers (n.d.[42]), COMBUS (n.d.[43]), Insurance Development Forum (2025[44]).

The development of these different risk assessment tools – and their ultimate accuracy – depends on the availability and quality of data that is normally provided by governments, including hazard maps, data on past hazard events and their impacts, as well as data on exposure (i.e. building stock). Incomplete, inaccessible or low-quality data from governments will limit the accuracy of – and confidence in – risk assessment tools that are developed based on this data. A lack of comprehensive and updated data on property valuations and/or building inventories was identified as challenges in Pakistan and the Philippines. In Malaysia, one recent analysis identified limited scope and granularity, a short historical record and access restrictions as limitations in the flood risk data available to financial institutions (World Bank and Bank Negara Malaysia, 2024[34]). Some insurance companies highlighted efforts (since discontinued) to access more granular data from governments on past inundation levels. In Sri Lanka, limited trust in the accuracy of government-provided flood maps was identified by some insurers as an impediment to their use in underwriting coverage.

Figure 1.13 provides an overview of the main approaches taken by insurers to assessing natural hazard risks in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka. In the Philippines, while some insurers have access to catastrophe modelling tools for earthquake, flood and storm/cyclone risks, insurance companies generally rely on hazard maps and past claims data as the primary basis for underwriting coverage. Similarly, in Indonesia and Malaysia, different insurers use different risk assessment tools. In Indonesia, risk ratings are most commonly used as a basis for underwriting flood, storm and volcano risks, and risk zones are most commonly applied for earthquake risks although insurers also have access to models for flood, earthquake and volcano risks, including from Maipark, a local reinsurer for earthquake risk. Similarly, in Malaysia, some insurers use models, risk ratings and hazard maps for underwriting earthquake, flood and storm risks, although past claims data appear to be the main basis for underwriting coverage among most insurers, including aggregate industry-wide data provided by Insurance Services Malaysia (ISM). In Pakistan and Sri Lanka, past claims and loss data are used as the main basis for underwriting coverage for all natural hazard risks although insurers in Pakistan also access mapping tools made available by international reinsurers and intermediaries. Limited past insured losses and the low quality of some insurers’ claims data, as well as changes in hazard frequency and severity, were identified as challenges to the use of past loss data for underwriting coverage, either by insurance companies or regulators (or both) in Sri Lanka and Pakistan as well as in Malaysia and the Philippines.

Figure 1.13. Primary basis for underwriting natural hazard insurance

Note: As noted above, insurance companies use various risk assessment tools for the different types of hazards. The figure illustrates what appears to be the most common tool used by insurers.

Source: Prepared based on responses to a survey circulated to insurance regulators and insurance companies and discussions with regulatory and insurance market representatives.

There is some evidence that the challenges related to risk assessment, whether due to a lack of data or a lack of risk assessment tools, have had an impact on insurance companies. In Sri Lanka, one public sector agency indicated that a lack of data made insurance companies reluctant to provide coverage for natural hazard risks. Insurance companies in Sri Lanka indicated that a lack of risk assessment tools has led to conservative underwriting, higher deductibles and coverage exclusions. In Malaysia, one analysis found that the inability to accurately assess risks exposures and quantify potential losses has limited insurance companies’ use of risk-based pricing (World Bank and Bank Negara Malaysia, 2024[34]). As discussed in the next section, there have also been impacts on reinsurance access and pricing for some insurers in some countries.

Limited access to affordable reinsurance coverage

Due to the potential for severe and correlated losses, insurance companies that assume natural hazard risks will usually transfer some portion of that risk to reinsurance markets. Global reinsurance and capital markets play a critical role in providing additional financial capacity to absorb natural hazard losses and diversify natural hazard risk internationally (OECD, 2018[45]).

Reinsurers provide two main types of reinsurance coverage: (i) proportional coverage (i.e. where premiums and claims are shared on a proportional basis); and (ii) non-proportional coverage (i.e. where reinsurance coverage is provided for losses above a certain threshold). Both types of reinsurance coverage are accessed to protect against property and natural hazard risks, although non-proportional reinsurance is inherently designed to protect against the severe losses that could result from catastrophes such as natural hazard events. Reinsurance can be provided on a facultative basis (i.e. as a reinsurance coverage for a specific, usually large, policyholder) or treaty basis (i.e. as a reinsurance coverage for a portfolio of policyholders).

Insurers have also increasingly transferred risk to capital markets through insurance-linked securities (e.g. catastrophe bonds18) and other alternative reinsurance structures such as collateralised reinsurance, sidecars and industry loss warranties.19 However, traditional reinsurance remains the main source of coverage for most insurers. Capital backing traditional reinsurance accounted for approximately 82% of all reinsurance capital in 2024 (Guy Carpenter, 2025[46]).

In most countries, insurers cede a significant share of the property insurance premiums that they collect to reinsurance markets. Emerging and developing market insurers tend to cede more to reinsurers than insurers in high-income countries (65%, on average, relative to 34% in high-income countries). Insurers in Indonesia, Pakistan, the Philippines and Sri Lanka cede more to reinsurers than the simple average for emerging and developing markets (Figure 1.14).

Figure 1.14. Share of property insurance premiums ceded to reinsurers (cession ratio)

Note: Calculated as the simple average of premiums ceded as a share of gross written premium for the “Fire and other property insurance” class of insurance for all reporting countries and available years between 2012 and 2023 from the OECD Data Explorer. For Pakistan, data is from SECP reports on Insurance Industry Statistics (2021-23 data years). For the Philippines, data is from the Insurance Commission’s Annual Reports (2013-22 data years). For Sri Lanka, data is from IRCSL Statistical Reviews (2015-23 data years). Countries are classified as high-income based on the World Bank Country and Lending Groups (World Bank, 2025[47])

Source: OECD based on data in premiums written and ceded from OECD (2025[48]), Insurance Commission (2015[49]), (2017[50]), (2019[51]), (2021[52]), (2023[53]), SECP (2022[54]), (2024[55]), IRCSL (2018[56]), (2020[57]), (2022[58]), (2024[59]).

The cost and availability of reinsurance tends to be cyclical. Large reinsurance industry losses tend to lead to an increase in pricing and decrease in availability, referred to as a “hard market”, which tends to stabilise over time as new capacity enters the market and competition leads to lower pricing (OECD, 2018[45]). Figure 1.15 provides an overview of reinsurance pricing trends for property catastrophe risks, based on the Guy Carpenter Global and APAC (Asia-Pacific) Property Catastrophe Rate on Line indices. The Global index shows significant pricing increases after large hurricane losses in the United States in 2005. The APAC index shows significant increases after major flooding in Thailand and a major earthquake in Japan in 2011. Both the Global and APAC indices indicate hard market conditions in recent years with pricing that was 2-25% higher in 2023 and 2024 than in 2004. In the recent hard market, higher attachment points have also been applied which has required primary insurers to retain more of the risk and ultimate claims. For example, prior to 2023, an estimated 20% of global “catastrophe losses” were absorbed by reinsurers – although this declined to an estimated 14% in 2024 (Guy Carpenter, 2024[60]).

Figure 1.15. Reinsurance pricing (Guy Carpenter Property Catastrophe Rate on Line Index)

Note: The Guy Carpenter Global Property Catastrophe Rate on Line Index provides data from 1990 although the Guy Carpenter APAC Property Catastrophe Rate on Line Index only provides data from 2004. To provide a comparison on pricing trends, the global index was adjusted to make 2004 the base year. The indices are calculated based on the change in the USD cost of coverage for excess-of-loss reinsurance placements by brokers for a consistent reinsurance programme.

Source: OECD calculations based on Guy Carpenter (2024[61]), (2024[62]) (using individual figures published in Artemis (2025[63]) and (2025[64])).

In addition to global and regional reinsurance pricing trends, a number of factors specific to national insurance markets and individual insurers also have important impacts on cost and access to reinsurance coverage, including loss experience, data quality and, particularly in the case of proportional reinsurance, pricing adequacy.20 The policy and regulatory framework (Box 1.2) can also have impacts on access to affordable reinsurance. Limited access to reinsurance, higher pricing and/or higher attachment points can limit direct insurer appetite and can have a significant impact on the availability and affordability of property insurance for households and businesses – particularly for insurers and insurance markets that rely heavily on transferring risks to reinsurance markets.

Box 1.2. Policy and regulatory requirements related to risk transfer to reinsurance markets

Mitigating risk to primary insurers

The transfer of risk to reinsurance markets creates risks for the ceding insurer, including counterparty risks (the risk that the reinsurer will not pay valid claims) and execution risks (the risk that reinsurance coverage does not respond as expected by the cedant). Insurance regulators and supervisors oversee and impose requirements on primary insurers’ use of reinsurance to mitigate these risks, including by requiring insurers to submit their annual reinsurance plans and ensuring that insurers implement risk retention and transfer approaches consistent with their financial capacity. For example, in Indonesia, insurers are required to meet quantitative requirements on minimum and maximum retention as a share of equity capital. Many supervisors also establish minimum credit rating requirements for reinsurers to be eligible to assume risk (or substantial risk) from domestic insurers.1

Premium retention and insurance market development

In a number of countries, national reinsurance companies or funds have been established and provide reinsurance to primary insurers. These national reinsurers are often owned by the government (at least partially)2 and have usually been established to retain premiums locally (instead of transferring premiums (and risk) to international reinsurance markets) and support the development of the local (re)insurance market. National reinsurers have been established in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka (along with many other countries in Asia and globally). National reinsurers often benefit from some form of preferential access to the reinsurance placements of domestic insurers, whether through a mandatory offer of a specific share of a reinsurance placement or other measures such as a lower capital charge for risks transferred to the national reinsurer. For example, insurers in Malaysia, Pakistan, the Philippines and Sri Lanka are required to offer a share of their reinsurance placements to the national reinsurer.3 In Malaysia, insurers are also required to maximise local retention while the placements with the national reinsurers in Malaysia and Sri Lanka (as well as other “resident” reinsurers in Malaysia) benefit from a lower capital charge than placements with other reinsurers. In Indonesia, a former system of preferential placements with domestic reinsurers (government-owned and private sector reinsurers) was loosened in 2020 for “non-simple” risks including property-related risks, allowing insurers to transfer risk to reinsurers based in jurisdictions with a bilateral agreement with Indonesia. As discussed below, the national reinsurers play different roles in the markets of each country and therefore have different impacts on the availability and affordability of reinsurance coverage.

Notes:

1. For example, in Pakistan, 80% of ceded risk must be placed with reinsurers rated A (S&P) or above and no risk can be placed with a foreign reinsurer with a rating below BBB (S&P). In Sri Lanka, insurers can only transfer risk to reinsurers with a rating of B+ (A.M. Best) or above. In the Philippines, foreign reinsurers must be represented by a resident agent to be eligible to assume risk which requires that the foreign reinsurer has capital levels equivalent to what is required of domestic counterparts in the Philippines and a minimum credit rating (A- (A.M. Best), Aa (Moody’s), AA (Fitch, S&P)).

2. For example, in Sri Lanka, the national reinsurance company (National Insurance Trust Fund) is wholly-owned by the government. In Pakistan, the national reinsurer (Pakistan Reinsurance Corporation) is partially listed although the government is the major shareholder. In Malaysia and the Philippines, a government-owned agency or fund has a major shareholding in the national reinsurer (Malaysia Re and Nat Re in the Philippines). In Indonesia, there are two reinsurers that are wholly-owned by the government (Indonesia Re and Nasional Re) and a few private sector domestic reinsurers, including a reinsurer established by the insurance sector to assume earthquake risk (Maipark).

3. In Malaysia, 2.5% of all proportional treaty placements must be offered to Malaysia Re. In Pakistan, 35% of all treaty reinsurance placements must be offered to the Pakistan Reinsurance Corporation and all facultative placements must be offered to domestic insurers. In the Philippines, Nat Re must be offered 10% of all outward reinsurance placements, and facultative placements must be offered to at least two domestic insurers. In Sri Lanka, 30% of each treaty and facultative reinsurance placement must be offered to the National Insurance Trust Fund.

Some insurance supervisors, intermediaries and/or insurance companies in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka identified challenges related to the cost and availability of reinsurance as a driver of protection gaps – although with some differences in terms of impact:

• In Malaysia, changes in reinsurance appetite have been generally consistent with changes in the broader reinsurance market although the large floods that affected urban areas in 2021-22 have led reinsurers to impose tighter conditions and higher retentions. The reinsurance market is reportedly satisfied with pricing in Malaysia, including the generally applied tariff rate for flood coverage, which has helped maintain the availability of proportional reinsurance capacity. Most insurers transfer risk to reinsurers on a proportional basis although there are also some limited non-proportional treaty arrangements.

• In the Philippines, insurance companies have been impacted by the increase in pricing as a result of the hard market, although capacity has reportedly remained available for most insurers. Concerns about the adequacy of the minimum pricing as well as a lack of adherence to the minimum pricing requirement by some insurers has led to a reduction in proportional reinsurance capacity and a shift towards non-proportional excess-of-loss coverage with higher retention requirements imposed. The non-proportional coverage has been costly although the higher retentions have reportedly incentivised improved pricing discipline among insurers.

• In Pakistan, there has been a reduction in the number of international reinsurers willing to offer coverage (and act as lead reinsurers) and some concerns about data quality. However, larger insurers have been able to maintain their access to international reinsurance markets, driven in part by a perception that pricing is adequate. Some smaller insurers, who generally acquire proportional reinsurance coverage, have faced some constraints on access to reinsurance coverage. These challenges led the insurance regulator (the Securities and Exchange Commission of Pakistan) to temporarily relax the requirements on foreign reinsurer ratings in order to relieve some of the challenges in accessing capacity.

• In Indonesia, access to affordable reinsurance has been more challenging. Past preferential placements with domestic reinsurers led to international reinsurer concerns about the concentration of risk in domestic insurance and reinsurance markets. These concerns, along with concerns about pricing adequacy due to some insurers’ practice of pricing below the pricing guidance, have led to a reduction in international reinsurance market appetite as insurers transfer most risk on a proportional basis. While the preferences have been largely removed, insurers have still faced challenges in accessing affordable reinsurance, including higher prices and higher retentions. In addition, the maximum retention requirement imposed by the insurance regulator, Otoritas Jasa Keuangan (OJK), means that the capacity of insurers to retain more risk is limited without raising further capital (and, as discussed below, OJK has increased the minimum capital requirements for all insurers which has also led to a need to raise capital).

• In Sri Lanka, challenges in accessing foreign currency during the economic crisis as well as some concerns about the quality of exposure data (i.e. data on insured structures) have had significant impacts on the appetite of reinsurers for assuming risk in Sri Lanka. In addition, premium prices have not increased much in recent years as insurers have focused instead on increased deductibles or coverage withdrawals, which has limited the attractiveness of assuming risk from Sri Lankan insurers (relative to other markets with higher pricing). The resulting reductions in reinsurance coverage have reportedly had an impact on insurer’s willingness to provide coverage to households and businesses.

The contribution of national reinsurers to ensuring the availability of affordable reinsurance capacity has differed across countries. Malaysia Re usually assumes significantly more than the amount that insurers are required to cede and leads many of the reinsurance placements of Malaysian insurers. In the Philippines, Nat Re also assumes more than insurers are required to offer, particularly from smaller insurers, although capacity constraints limit its ability to accept more than 10% for larger placements. The national reinsurers in Indonesia are also significant sources of capacity and lead many placements by domestic insurers. This partly reflects efforts by OJK to enforce strict due diligence on foreign reinsurers, encourage diversification of reinsurance panels and prioritise placements with reinsurers that have strong credit ratios and a proven track record in paying valid claims to ceding insurers. The removal of preferences has led some foreign-owned insurance companies to cede risk to group reinsurance entities although many foreign reinsurers continue to transfer risk to the national reinsurers.

In Pakistan and Sri Lanka, the share of reinsurance placements that need to be offered to the national reinsurers are much higher and some insurers seem to be more reluctant to offer any more than what is required (although small insurers in Pakistan tend to cede more than is required to the national reinsurer). The national reinsurers in Pakistan and Sri Lanka tend to assume the share that insurers are required to offer. However, the National Insurance Trust Fund in Sri Lanka has faced some financial capacity constraints due to the impact of the foreign exchange crisis on its ability to access retrocession from international markets. The national reinsurer in Sri Lanka has also faced challenges due to the lack of an exemption from government procurement rules for placing retrocession, which has reduced retrocession market interest.

The national reinsurers in Indonesia, the Philippines and, in particular, Malaysia, have developed significant technical expertise and are increasingly competitive in the market. Malaysia Re, for example, is one of the largest reinsurers in Asia, is highly rated and assumes significant risk from insurers outside Malaysia. The national (and domestic) reinsurers in Indonesia and the national reinsurer in the Philippines are also increasingly competitive although their ability to compete for business internationally remains limited due to lower credit ratings and concerns about country concentration. The national reinsurers in Pakistan and Sri Lanka are perceived to be more akin to a government agency than an international or professional reinsurer. Some insurance companies in Pakistan and Sri Lanka raised concerns about counterparty risks due to their exposure to the national reinsurers. In Pakistan, some insurers indicated that their counterparty exposure to the (unrated) Pakistan Reinsurance Corporation had a negative impact on their own credit ratings. Neither the Pakistan Reinsurance Corporation nor the National Insurance Trust Fund in Sri Lanka are protected by an explicit government guarantee to assume any losses beyond the reinsurer’s financial capacity.

Pricing by national reinsurers has tended to follow pricing trends in international markets. Therefore, none have been able to make any significant contribution to enhancing the affordability of reinsurance coverage. This results from the fact that most of the national reinsurers are partially listed (with profit-motivated shareholders), need to secure retrocession coverage (in some cases, significant)21 and usually accept reinsurance placements together with other reinsurers.

In countries with significant takaful and other sharia-compliant insurance markets, limited availability of affordable retakaful (i.e. reinsurance based on sharia) could have an impact on the ability of takaful providers to provide financial protection to households and businesses. Retakaful capacity is much more limited than reinsurance capacity and a few stakeholders raised concerns about the limited availability of such capacity. In Malaysia, for example, there are only two domestic retakaful providers (including Malaysia Re) while concerns about the financial soundness of retakaful providers in some other countries limit the option for retakaful placements.

1.3.  Policy, regulatory, supervisory and market approaches to addressing protection gaps in natural hazard insurance

1.3.1. Increasing demand for natural hazard insurance

Relatively more frequent experience with natural hazards across many countries in Asia has likely led to a higher level of risk perception than other parts of the world. However, in many countries in Asia, a higher level of risk perception has not translated into higher levels of insurance coverage for natural hazard losses. For example, in one survey of urban residents in Indonesia that had experienced climate-related extreme events, purchasing insurance was the least common among a set of preparedness actions (10.6% had acquired insurance and 6.8% were in the process of doing so – while 42.6% indicated that they were unlikely to purchase insurance in the future) (Adhi et al., 2019[16]). This divergence is likely driven, in part, by limited awareness of and trust in the value of insurance as a tool for providing financial protection.

Building insurance literacy

Improvements in insurance literacy can make an important contribution to addressing protection gaps in natural hazard insurance. A number of studies have found a positive correlation between the level of insurance literacy and the take-up of insurance. For example, one survey of Sri Lankan consumers found a positive correlation between insurance purchase and higher insurance literacy (along with higher levels of trust and perception of the benefits of insurance) (Weedige et al., 2019[65]). Similarly, a survey of consumers in India found that higher levels of insurance literacy increased the likelihood of acquiring micro-insurance (Uddin, 2017[66]).

Given the growing recognition of the importance of financial literacy, an increasing number of countries have established national financial literacy strategies, consistent with the OECD Recommendation on Financial Literacy (OECD, 2020[67]), or national financial inclusion strategies that include a component related to financial literacy. In some countries, the strategy has been established by a financial sector regulator, such as in Indonesia (OJK, 2021[25]) and Pakistan (State Bank of Pakistan, 2025[68])). The Central Bank of Sri Lanka has led the development of a National Financial Inclusion Strategy and a Financial Literacy Roadmap that includes the Insurance Regulatory Commission of Sri Lanka (and other government departments and agencies) as implementing agencies (Central Bank of Sri Lanka, 2021[69]; Central Bank of Sri Lanka, 2024[70]). In Malaysia and the Philippines, financial inclusion and/or financial literacy strategies have been developed by a committee or network of financial sector regulators and other government departments and agencies, including the agency responsible for insurance regulation (Financial Education Network, 2025[71]; Financial Inclusion Steering Committee, 2024[72]).

High levels of financial literacy do not always translate into higher levels of insurance literacy (Weedige et al., 2019[65]). Achieving insurance literacy requires an understanding of various elements essential to understanding the need for, role and scope of insurance coverage, including: (i) understanding exposure to risk; (ii) awareness of risk mitigation strategies; (iii) understanding of the concept of insurance, including the principles and benefits of insurance; (iv) understanding of the features of insurance products, and the rights and duties of the insured; and (v) awareness of how to find information on available insurance coverage and how to access it (based on (Weedige et al., 2019[65])). The OECD Recommendation on Financial Literacy also outlines some specific elements that need to be incorporated into financial literacy programmes in order to support insurance literacy, including a need to promote an understanding of risk, risk mitigation and the possible role of insurance as well as information and tools to support better use by individuals of insurance products (OECD, 2020[67]).

The national financial literacy and financial inclusion strategies in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka place different levels of emphasis on insurance, relative to other financial services. In Malaysia and the Philippines, the strategies include objectives that are specifically targeted at “risk protection”, including a focus on protection against natural hazard risks (Financial Education Network, 2025[71]; Financial Inclusion Steering Committee, 2024[72]). The objectives of the financial literacy strategy in Indonesia are not specific to a type of financial service although elements related to insurance and insurance literacy are integrated throughout the strategy. This includes specific questions on insurance product features, benefits, risks, costs, consumer rights and obligations and how to access insurance in order to measure insurance literacy through the periodic National Survey on Financial Literacy and Inclusion (SNLIK) (OJK, 2021[25]). OJK has also included a module on insurance within its Financial Education Learning Management System, a platform that provides a mechanism for independent learning and a repository for financial education materials (OJK, 2021[25]).

The national financial inclusion strategies in Pakistan and Sri Lanka, and the financial literacy roadmap in Sri Lanka, place more limited emphasis on insurance. In Pakistan, there is an objective to promote microinsurance and microtakaful in order to help create a conducive environment for the microfinance sector (State Bank of Pakistan, 2025[68]). In Sri Lanka, the national financial inclusion strategy appears to be mostly focused on credit access (Central Bank of Sri Lanka, 2021[69]), while insurance is seen as one of a number of the elements relevant to individuals’ financial management in the financial literacy roadmap (Central Bank of Sri Lanka, 2024[70]).

The insurance regulators in Pakistan and Sri Lanka have developed online guides to support consumers’ understanding of insurance. The Securities and Exchange Commission of Pakistan (SECP), through its investor education initiative, has published a basic guide to insurance that provides an overview of the principles of insurance and potential benefits, describes the different types of insurance and how to access coverage, and outlines the process for complaints, including the role of the SECP in the complaint process (Jama Punji, 2017[73]). The Insurance Regulatory Commission of Sri Lanka has published a booklet outlining the basic principles of insurance, describing the types of insurance and providing practical guidance on what to consider when acquiring insurance and avenues for resolving complaints (Insurance Board of Sri Lanka, 2019[74]). This is complemented by efforts to disseminate information about insurance and insurance concepts through social media channels as well as through print media, radio and television programmes and short messages sent to mobile phones.

Insurance associations and insurance companies also play a key role in supporting insurance literacy. The insurance associations in Indonesia (AAUI, 2018[75]), Malaysia (PIAM, 2025[76]), Pakistan (Insurance Association of Pakistan, n.d.[77]), the Philippines (PIRA, 2021[78]) and Sri Lanka (Ada derana Business, 2019[79]) have all identified increasing public awareness about insurance among their main objectives. The associations support awareness and educational campaigns, conferences and other types of events. In Indonesia, Pakistan and Sri Lanka, national insurance days have been established by – or with the support of – the insurance associations as a means of enhancing public awareness. Some associations have information for consumers on their websites, including information on insurance products and complaints and redress mechanisms. In the Philippines, the insurance association has developed “PIRA TV” which includes short videos in talk-show format on different types of insurance products, including fire insurance (PIRA, n.d.[80]). In Indonesia, financial service providers (including insurance companies) are required to conduct regular activities to improve financial literacy and inclusion and report on those activities to the regulator.

Building trust in insurance

A lack of trust in insurance as an effective form of financial protection is a potential driver of low insurance penetration in some countries and appears to be a challenge in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka. Given the nature of insurance, as a promise of a future payment based on the occurrence of an uncertain event, trust that the provider of that promise will fulfil their contractual obligation is critical. Some studies have identified a direct relationship between levels of trust and the willingness to purchase intangible products such as insurance (Weedige et al., 2019[65]).

Trust in insurance may be impacted by low levels of financial literacy (Weedige et al., 2019[65]), a lack of experience with insurance, preconceived notions about being treated unfairly or about the financial capacity of insurance companies to meet their obligations, concerns about fraud or scams, or negative past experience with insurance (such as disputes or lengthy claims settlement processes (EIOPA, 2024[4])). For example, the OECD’s 2024 Consumer Finance Risk Monitor found that issues related to claims handling and contractual clauses (including exclusions in coverage) were the most frequently identified drivers of complaints received by insurance firms, alternative dispute resolution mechanisms and supervisors in responding jurisdictions (OECD, 2024[81]).

Insurance regulators can make a number of contributions to building trust in insurance by responding to some of the drivers of mistrust. Many of these contributions relate to functions that are generally within the scope of insurance supervision, such as ensuring that the terms and conditions included in policy coverage are clear and easy to understand (including any coverage exclusions), premium pricing is not excessive, claims are settled quickly and that complaint and redress mechanisms are communicated to consumers and are effective in the resolution of disputes. A number of insurance supervisors also implement measures to ensure that consumers can find information on how to avoid potential fraud and scams, such as by publishing lists of licensed insurance companies and intermediaries and, in some cases, public notices about entities that are not authorised by the regulator. For example, the Insurance Regulatory Commission of Sri Lanka published a list of licensed insurance companies and intermediaries on its website and issues public notices to encourage the public to visit the website and learn about entities that are authorised by the Commission. Bank Negara Malaysia maintains a list of companies and websites that may be wrongly perceived as having been licensed or regulated by the regulator (Bank Negara Malaysia, 2025[82]).

Efforts to build public awareness of the role of the insurance regulator in monitoring the financial soundness of insurance companies could also support trust in the insurance sector. Policy makers have a role in supporting public awareness of the functions of insurance supervisors while also ensuring that insurance supervisors have sufficient powers and resources to undertake the functions that support trust in the insurance sector. Several countries have also established policyholder protection schemes which, as long as consumers are aware of – and understand the role of – such schemes, should also support consumer confidence that any valid claims will be paid (Box 1.3).

Box 1.3. Policyholder protection schemes

Policyholder protection schemes, which are sometimes referred to as insurance guarantee schemes, are arrangements to provide protection to policyholders in the event of a failure of an insurance company by either paying valid claims or providing funding for any shortfall in order to support the transfer of the failed insurer’s business to another insurer (IAIS, 2023[83]). Policyholder protection schemes that provide protection for non-life insurance policyholders exist in a number of OECD countries (e.g. Australia, Canada, Denmark, France, Japan, Korea, Norway, the United Kingdom, the United States) (OECD, 2013[84]). A few schemes have also been established in non-OECD Asian countries, including in Malaysia (the Takaful and Insurance Benefits Protection System which provides 100% compensation up to MYR 500 000 for claims due to property insurance policyholders) (PIDM, 2025[85]), Singapore (which provides 100% compensation for claims due to property insurance policyholders up to SGD 300 000) (SDIC, 2018[86]; ASEAN Insurance Council, 2023[87]) and Thailand (which provides compensation for claims due to property insurance policyholders up to THB 1 million) (ASEAN Insurance Council, 2023[87]). In Indonesia, a Policy Guarantee Program is planned for implementation in 2028 although the scope of policies covered has not been announced (Asia Insurance Review, 2023[88]).

The existence of policyholder protection schemes, where disclosed to policyholders, can support confidence in the insurance sector (IAIS, 2023[83]; OECD, 2013[84]). A consultation paper by the European Insurance and Occupational Pensions Authority (EIOPA) on the harmonisation of insurance guarantee schemes suggested that the additional confidence in the insurance sector could lead to increased consumer demand for insurance products (EIOPA, 2019[89]). One rating agency has indicated that Indonesia’s guarantee programme is likely to improve confidence in the insurance sector and support market growth (Pratiwi and Novaryani, 2022[90]).

Increasing take-up of property insurance penetration

Increasing the take-up of property insurance is essential for addressing protection gaps in natural hazard insurance. Coverage for natural hazard risks is usually attached to property insurance and households and businesses without property insurance will not be provided with an opportunity to acquire natural hazard insurance coverage. They would also have more limited access to any insurer efforts to raise awareness of the need for coverage against natural hazard risks. For example, in Sri Lanka, insurer awareness raising campaigns are sometimes limited to existing clients (i.e. households and businesses with an existing insurance coverage). There is a clear link between the level of property insurance penetration and the share of natural hazard losses that are insured. With only a few exceptions, a higher share of natural hazard losses is insured in countries with higher levels of property insurance penetration (Figure 1.16).

Figure 1.16. Property insurance penetration and insurance coverage for natural hazard losses

Source: OECD calculations based on data on insured and economic losses related to natural hazards provided by Swiss Re (Swiss Re, sigma database. All rights reserved.), data on property insurance premiums reported in OECD (2025[48]), SECP (2024[55]), Insurance Commission (2023[53]), IRCSL (2024[59]) and GDP data from IMF (2024[91]).

In addition to insurance literacy and trust, there are a number of other factors that can influence the level of property insurance penetration. For example, one study examined the correlation between insurance density (premiums per capita) in approximately 40 developed and emerging markets and a range of factors related to income, probability of loss, risk awareness, cost of insurance and legal protections for property rights - and found positive links between insurance density, income, probability of loss and property rights protection (Esho et al., 2004[92]). Figure 1.17 illustrates the positive correlation between property insurance penetration and per capita income levels and property rights protection.

Figure 1.17. Property rights, income and property insurance penetration

Source: OECD calculations based on Property Rights Index Scores provided by Property Rights Alliance (2024[93]), data on property insurance premiums reported in OECD (2025[48]), SECP (2024[55]), Insurance Commission (2023[53]), IRCSL (2024[59]) and GDP and GDP per capita data from IMF (2024[91]).

In many countries bank lenders play an important role in distributing and often effectively mandating the purchase of property insurance among their borrowers. This is the case in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka. The insurance supervisors, insurance companies and intermediaries in these five countries all indicated that bank lenders are a major driver of the take-up of property insurance coverage. Bank lenders in all of these countries require their borrowers to acquire property insurance coverage for assets used as collateral for bank loans, particularly in the case of mortgage loans extended to households.

However, mortgage lending is less common in many emerging and developing countries than it is in high-income countries. In 2021, the value of mortgage loans was equivalent to less than 7% of GDP in emerging and developing countries and less than 9% in emerging and developing Asia and Pacific, relative to more than 44% of GDP in high-income countries.22 Mortgage loans accounted for less than 1% of GDP in Pakistan and Sri Lanka and less than 5% of GDP in Indonesia and the Philippines – although it accounted for more than 35% of GDP in Malaysia.23 While mortgage lending is growing, increasing as a share of GDP by 11 percentage points in the Philippines, 23 in Malaysia and 69 in Indonesia, home ownership is already relatively high in many emerging and developing Asia and Pacific countries which may limit the potential for growth in mortgage lending.24 For example, an estimated 84% of households in Indonesia owned their homes in 2019 and 76.9% in Malaysia, relative to approximately 70% of households across high-income OECD countries.25

Encouraging take-up of coverage for natural hazard risks

Insurers take different approaches to offering coverage for natural hazard risks in different countries, ranging from an automatic inclusion of coverage for these risks in property insurance to an optional add-on of coverage for natural hazard risks. The approach taken to offering coverage can have important implications on the magnitude of protection gaps. Where it is optional for insurers to offer coverage for natural hazard risks, some insurers may choose not to offer coverage for these risks, or for specific types of natural hazards, to households and businesses, whether due to:

• Uncertainty related to the exposure that they would be assuming by offering coverage, possibly due to gaps in the availability of quality data on hazards or exposure, or the potential for climate change to lead to increased losses.

• The potential magnitude of natural hazard risk in particular high-risk locations or communities.

• Difficulties in accessing affordable reinsurance coverage for the natural hazard risks that they assume.

• Concerns that offering (or automatically including) coverage for natural hazard risks would lead to a higher premium quote, resulting in the loss of price-sensitive consumers to a competitor.

While insurance companies may, in general, be incentivised to offer coverage for natural hazard risk and collect additional premiums,26 some companies may be more comfortable with limiting their exposure to uncertain natural hazard risks. For example, in Indonesia, some insurers provided estimates of the share of their household and business policyholders that had acquired the optional TSFWD and EQVET natural hazard coverage. One insurer estimated that almost 80% of household policyholders had acquired the TSFWD coverage while almost 30% had acquired (the generally more expensive) EQVET coverage. Another insurer estimated that less than 5% of its household policyholders had acquired either the TSFWD or EQVET coverage.27 While this might result from other differences among the two insurers’ portfolio of policyholders, it could also suggest that one insurer is more willing to offer – or more effective at selling – the optional natural hazard coverage than the other.

In some cases, practices vary across different insurance companies within the same country. For example, in Pakistan, some insurers reportedly include coverage for natural hazards automatically with household property insurance coverage while others only offer it as an optional additional coverage. In Sri Lanka, insurance companies can choose to exclude coverage for specific types of natural hazard risks. Where insurers take different approaches to the inclusion of this coverage, households and businesses may be confused by what is covered in the policies offered by different insurers, although clear policy documents may help to avoid potential confusion if there is an adequate level of insurance literacy. Differences in, and potential complexity of, products can make it more difficult for consumers to compare between the offerings of different insurance companies. It could also exacerbate efforts to improve trust in insurance as an effective form of financial protection should policyholders learn that a loss is unexpectedly not covered.

Offering coverage for natural hazard risks as a standard or automatic inclusion in household and business property insurance, even with an option to opt-out, can ensure that insurance companies are offering coverage and that households and businesses are given an option to acquire that coverage. A mandatory offer of insurance coverage for natural hazard risks could also respond to some of the challenges to enhancing take-up identified above, by:

• Ensuring that consumers are aware of the availability of coverage for natural hazard risks.

• Reducing the possibility that consumers will mistakenly assume that coverage for natural hazard risks is included in basic (property) fire insurance.

• Providing insurers (or intermediaries) with a (mandated) opportunity to build awareness of natural hazard risks and the benefits of - and need for – specific insurance coverage, given that insurers would be required to explain the coverage and its benefits and limits to policyholders.

The automatic inclusion or mandatory offer, with an opt-out option for consumers, will not fully address protection gaps if consumers choose not to take-up coverage for natural hazard risks or acquire any property insurance at all. Some insurance supervisors, intermediaries and/or insurance companies in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka (and elsewhere) have suggested going further and requiring households and/or businesses to acquire property insurance that covers natural hazard risks (i.e. mandatory purchase). In Indonesia, for example, recent legislation on the Development and Strengthening of the Financial Sector authorises the government to establish mandatory insurance programmes for specific segments of the population and lines of business, including household insurance against disaster risks (Government of Indonesia, 2023[94]).

While mandatory purchase, automatic inclusion and mandatory offer of natural hazard coverage have tended to reduce protection gaps (Figure 1.18), these approaches can also create challenges for insurers and policyholders. For insurers, all three approaches create an obligation to offer coverage for risks that they might not have the capacity or appetite to assume and could lead them to not offer any property insurance coverage to households, businesses or entire communities that have high-levels of exposure to one or more natural hazards.28 For policyholders, mandatory purchase could be perceived as a tax and could force them to purchase coverage that they cannot afford.29 Similarly, automatic inclusion of coverage for natural hazard risks could lead to premiums that are unaffordable for some households and businesses and lead them to choose to not acquire any property insurance at all. As discussed below, in countries where coverage is mandatory, natural hazard coverage is automatically included, or where insurers are required to offer coverage, some form of public-private insurance programme is often in place to support insurance availability and affordability.

Figure 1.18. The impact of automatically including flood insurance coverage on protection gaps

Source: OECD calculations based on data provided by Swiss Re (Swiss Re, sigma database. All rights reserved.).

Banks in many countries require borrowers to acquire insurance coverage. The insurance requirements are usually imposed voluntarily by banks as a means to reduce the risk that the value of their collateral will decline as a result of property damage. As a result, there may be inconsistencies in approach as banks may or may not require coverage for natural hazard risks - and may invest more or less effort into ensuring that borrowers maintain their coverage over the life of the loan. Banks in Malaysia, the Philippines and Sri Lanka (for some hazards) tend to require their borrowers to acquire coverage for natural hazard risks, although some insurers in Malaysia indicated that this requirement is not imposed universally by banks. In Sri Lanka, one insurance sector stakeholder noted that banks usually do not monitor whether a borrower renews their coverage in subsequent years. In Indonesia, most banks do not require coverage for natural hazard risks although some banks have recently started to impose this requirement. In Pakistan, banks are required by legislation to ensure that borrowers have property insurance although this requirement does not extend to coverage for natural hazard risks.

While bank-imposed requirements for natural hazard insurance coverage can lead to many of the same challenges for insurers and policyholders as other requirements on insurance purchase or offer, there are likely other benefits in reducing banking sector exposure to natural hazard risks that should be taken into account. Significant protection gaps could create risks to the broader financial system if a major event leads to substantial defaults among uninsured households and businesses. Ensuring that banks require their borrowers to have coverage against natural hazard risks for the properties used as collateral could limit this risk. Reducing the risk that natural hazards lead to broader financial system stress has become an increasing focus for many financial sector policy makers and regulators around the world (Poeschl, 2022[95]; Kauko et al., 2021[96]; Newman, Adams-Kane and Nicholls, 2023[97]; European Central Bank and European Systemic Risk Board, 2023[98]; Allen et al., 2023[99]; Bank of England, 2019[100]; Kahn, Panjwani and Santos, 2024[101]).

In most countries, including in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka, the approach to providing natural hazard insurance coverage is determined by insurance market practice (i.e. insurers decide whether coverage for specific natural hazards is included with – or must be added to – property insurance coverage offered to households and businesses).30 However, insurance supervisors generally have a role in establishing rules or principles to be applied by insurers in their conduct of business, which includes ensuring proper disclosures and documentation on coverage and exclusions. In the context of increasing climate-related risks, the International Association of Insurance Supervisors has proposed that supervisors’ Conduct of Business requirements (Insurance Core Principle 19) should: (i) ensure that information on natural hazard coverage and exclusions is clearly stated in “disclosures, advertising materials, pre-contractual and contractual documentation”; (ii) encourage insurers to develop a common approach to presenting insurance options and different types of coverage to consumers; and (iii) undertake testing to confirm consumer understanding of coverage exclusions (IAIS, 2025[102]). Some insurance supervisors plan to take steps on these issues. For example, the Securities and Exchange Commission of Pakistan plans to introduce standardised retail policy documents (SECP, 2024[103]). Some insurance companies (e.g. Sri Lanka) suggested a need for the insurance regulator to encourage further standardisation of coverage and improved disclosure of exclusions.

Ultimately, the decision on whether to intervene requires an assessment of the potential risks of mandating an approach to offering coverage for insurers and consumers relative to the potential benefits in reducing protection gaps. Some insurers, particularly in Malaysia and the Philippines, indicated some progress in enhancing the take-up of insurance coverage for natural hazard risks as a result of increasing risk awareness and (unfortunately) more frequent experience of the impact of natural hazards.

Expanding reach and offering suitable and affordable coverage

Progress in addressing protection gaps in natural hazard insurance will also require efforts to expand the reach of available and affordable insurance coverage, including by: (i) leveraging available distribution channels; and (ii) developing coverage options that are suitable for – and attractive to – segments of the population that have traditionally been uninsured or financially excluded.

Leveraging distribution channels

Insurance coverage for property and natural hazards may be sold through a variety of distribution channels, including through direct sales by insurance companies or through intermediaries such as agents, brokers and banks (bancassurance). Brokers and agents, including agents representing multiple insurers and those representing a single insurer, account for the largest share of non-life insurance sales in most countries (Figure 1.19). In general, households and smaller businesses tend to access coverage through direct sales, agents and banks, while larger businesses tend to access insurance markets through brokers.

While accounting for a small share of non-life insurance distribution in most countries, sales through banking and digital channels have increased and are expected to account for an increasing share of non-life insurance distribution in the future.

Non-life insurance distribution through banks accounts for a relatively minor share of non-life insurance distribution in the Philippines and Sri Lanka, and particularly in Pakistan, despite the significance of bancassurance in driving take-up among households (given the link to mortgage lending). In other parts of the world, bancassurance has reportedly been one of the fastest-growing distribution channels and now accounts for almost 10% of non-life insurance distribution in Europe and 25% in some Latin American markets (McKinsey & Company (Insurance Practice), 2024[104]).

Figure 1.19. Non-life insurance distribution channels

Note: Exclusive agents are classified as “general agents” in the Philippines and as “corporate agents” in Pakistan. The “digital” category includes “e-commerce, online and digital sales” in the Philippines and telcos, websites and web aggregators in Pakistan (this category was not available in Sri Lanka). For the global data, “Direct sales” includes other channels (other than agents, brokers and bancassurance) while “Agents” includes only tied agents and branches. Global data is for 2018. Data for Malaysia and Indonesia was not available.

Source: OECD based on data on distribution channels from IRCSL (2022[58]), Insurance Commission (2023[53]), SECP (2024[55]), McKinsey & Company (Insurance Practice) (2021[105]).

It is likely that opportunities exist to increase the contribution of bancassurance to distributing property and natural hazard insurance coverage in Indonesia, Malaysia, Pakistan, the Philippines, Sri Lanka and other emerging and developing Asia and Pacific countries. Access to and use of banking services is increasing across Asia, with approximately 50% of individuals in Indonesia and the Philippines reporting having an account (either their own or shared) with a bank or other financial institution in 2021, and close to 90% in Malaysia and Sri Lanka (although only 16% reported having an account in Pakistan) (World Bank, 2023[106]). In addition, increasingly local access to a financial institution means that only about 25% of individuals in Indonesia, Pakistan and the Philippines – and less than 2% of individuals in Malaysia and Sri Lanka – indicated that distance to access a bank or financial institution is a reason for not having an account (World Bank, 2023[106]) (Figure 1.20). Distribution of coverage through banks might also reduce some of the challenges related to mistrust in insurance. For example, in Indonesia, one survey found that over 80% of respondents have confidence in banks, relative to approximately 10% that have confidence in the insurance sector (OJK, 2021[25]). Bancassurance could also provide a means to enhance coverage among SMEs given that many will have an existing relationship with a bank (Ingram, 2025[107]).

Figure 1.20. Access to banking services

Source: World Bank (2023[106]), The Global Findex Database 2021: Financial Inclusion, Digital Payments, and Resilience in the Age of COVID-19, https://www.worldbank.org/en/publication/globalfindex.

Growing access to and use of technology provides new opportunities for digital distribution of insurance coverage. The (vast) majority of the population in emerging and developing Asia and Pacific countries have access to the internet and own mobile phones (Figure 1.21), providing an opportunity to greatly expand the distribution of insurance coverage through digital sales channels. In Europe and Australia, for example, digital sales (including digital bancassurance) account for approximately 30% of non-life premiums (McKinsey & Company (Insurance Practice), 2024[104]).

Figure 1.21. Access to the internet and mobile phones

Source: World Bank (2023[106]), The Global Findex Database 2021: Financial Inclusion, Digital Payments, and Resilience in the Age of COVID-19, https://www.worldbank.org/en/publication/globalfindex.

The distribution of non-life insurance coverage through digital channels, such as via direct internet-based sales or through price comparison websites, remains limited - accounting for less than 1% of all non-life premiums collected in both Pakistan and the Philippines (data for this distribution channel was not available for the other countries). In Pakistan, for example, a recent survey undertaken by the Securities and Exchange Commission of Pakistan found that less than half of insurance companies offered products through digital channels and that the impact of third-party digital platforms (such as price comparison websites) was minimal. Digital products were available for only a few lines of business and only one company offered a property-related digital insurance product (a households contents coverage) (SECP, 2024[108]).

Regulatory frameworks applicable to insurance distribution do not generally impede the use of different distribution channels. While licensing requirements usually apply for different types of intermediaries, few countries restrict the type of insurance products that can be distributed through specific channels (OECD, 2020[109]).31 The regulatory framework applies equally to digital distribution in most countries, including through licensing requirements for price comparison websites. A few countries have established additional requirements applicable to digital distribution channels, particularly requirements related to information security and integrity (OECD, 2020[109]).

Some countries, including Malaysia and Pakistan, have developed tailored regulatory frameworks for digital-only insurers. In Malaysia, a Policy Document on Licensing and Regulatory Framework for Digital Insurers and Takaful Operators (DITOs) was issued in July 2024. The framework provides a tailored licensing and regulatory framework, including a lower minimum capital requirement during a limited foundational phase, to facilitate the entry of insurers and takaful providers that deliver their products digitally and aim to deliver strong and meaningful value propositions related to inclusion, competition and efficiency (Bank Negara Malaysia, 2024[110]). The Securities and Exchange Commission of Pakistan introduced a registration framework for digital-only insurers in August 2022, along with a framework for dedicated microinsurers (see below), with the aim of encouraging innovation, expanding the range of available products and supporting financial inclusion. Similar to the framework in Malaysia, digital-only insurers can benefit from lower minimum capital requirements (SECP, 2024[108]). While these frameworks aim to encourage the establishment of digital insurers rather than promote digital distribution, digital insurers would normally distribute their products through digital channels.

Regulators should monitor the use and impact of price comparison websites given the highly competitive markets, the price-sensitivity of consumers and the ability of insurers to apply exclusions and higher deductibles for natural hazard insurance coverage. Price comparison websites can enhance competition and encourage affordability (Brown and Goolsbee, 2000[111]), although they may incentivise insurers to reduce the scope of coverage and benefits provided or increase deductibles in order to be able to offer the lowest price on the price comparison website (OECD, 2020[109]). Some insurance regulators have imposed requirements on price comparison websites to address this issue by, for example, requiring comparisons that are based on more than price or that compare policies based on a standard coverage (OECD, 2020[109]). The Securities and Exchange Commission of Pakistan is planning to incorporate consumer insurance products into the Emlaak product comparison website, owned by a subsidiary of the Central Depository Company of Pakistan and currently focused on providing access to mutual fund products (ITMinds Limited, 2025[112]).

Offering suitable and affordable insurance products

Offering suitable and affordable natural hazard insurance coverage options that are attractive to the households and small businesses that are excluded from the existing insurance market32 will also be critical in reducing protection gaps for natural hazard risks. This could include products that offer more basic levels of coverage, including microinsurance or a basic natural hazard insurance coverage, innovative products that provide coverage based on parametric triggers as well as products that respond to religious and cultural beliefs and customs, such as takaful and other sharia-compliant financial protection products.

Microinsurance

Microinsurance generally refers to insurance coverage that is offered to low-income households or micro and small enterprises and that provides a small amount of coverage for a low premium cost. Microinsurance products have been developed to protect against a broad range of risks, including for agricultural risks, health, accident, life, credit as well as to protect property.

Insurance companies offer a variety of microinsurance products targeted to low-income households and small businesses, including products that provide coverage for natural hazard risks. In Pakistan and Sri Lanka, some insurers have introduced microinsurance schemes that provide a basic level of coverage against natural hazards to low-income households and small businesses – sometimes in partnership with other organisations such as microfinance providers or rural support programmes. In Indonesia, the insurance associations and some individual insurers have developed microinsurance products for natural hazards (and specific products for volcano and earthquake) as well as for property-related risks more generally (Biese et al., 2023[113]). One reinsurer in Indonesia has also been developing a proposal for a microinsurance coverage for natural hazard risks to be distributed more broadly. In Malaysia, some insurers have introduced microinsurance coverage against natural hazard risks for low-income households, although take-up has been limited.

However, the availability and take-up of microinsurance coverage for property and natural hazard risks is much more limited than for other types of coverage, such as life, credit and health. For example, one global examination of available microinsurance products found that less than 2% of the microinsurance premiums written in 2023 were for non-agriculture property products (Merry and Rozo Calderon, 2025[114]). Microinsurance coverage for property and natural hazard risks is also limited in emerging and developing Asia and Pacific countries. Property-related products reportedly account for only a small share of the microinsurance market in Indonesia (approximately 5%) (Ismail, Sivarajah and Chow, 2020[115]). In Pakistan, microinsurance for property risks accounted for less than 0.1% of the 12.5 million microinsurance policies in force in 2022 (SECP, 2024[116]). In the Philippines, which is among the largest microinsurance markets in the world, the vast majority of coverage (95%) is for life (including credit) and health risks (Insurance Asia, 2025[117]; Ismail, Sivarajah and Chow, 2020[115]).

While the number of individuals covered by microinsurance and the amount of premiums collected have increased33 and are expected to continue to increase,34 the broadening of microinsurance, for property as well as other risks, has faced several challenges. These include challenges in identifying cost-effective distribution channels, a lack of demand and difficulties in achieving financial sustainability, particularly where initial subsidies are withdrawn. Some insurers in Malaysia indicated that microinsurance products have faced high loss ratios, although the average claims ratio for property-related microinsurance is reportedly low on a global basis (generally below 25% in recent years (Merry and Rozo Calderon, 2025[114]; Merry and Rozo Calderon, 2023[118])). In some countries, premium taxes or value-added taxes are applied on an equivalent basis to microinsurance products which could limit the financial viability of these products. For example, in the Philippines, taxes applicable to non-life insurance products can account for more than 20% of the premium charged for microinsurance coverage (Asian Development Bank, 2024[119]). The Securities and Exchange Commission of Pakistan has also raised concerns about the impact of taxes on the affordability of microinsurance products (SECP, 2024[116]).

A greater diversification of distribution channels and broader use of digital distribution channels could help improve the financial viability of microinsurance for property risks. The vast majority (90%) of property-related microinsurance products in Asia-Pacific are distributed through agents and brokers and microfinance institutions (Merry and Rozo Calderon, 2025[114]). Digital distribution accounts for about 8% of microinsurance premiums (for all types of coverage) in Indonesia and only 3% in the Philippines (Ismail, Sivarajah and Chow, 2020[115]). There might also be opportunities to improve distribution and increase take-up by bundling property and natural hazard risk protection with other forms of coverage, particularly with life, credit and/or health coverage, given the more significant microinsurance penetration for these types of products. For example, in Samoa, an innovative coverage has been developed to provide a bundled protection against a number of natural hazard risks (cyclones, earthquakes, excess rainfall and drought) with a parametric trigger and additional benefits paid out on an indemnity basis for funeral, personal accident, hospital stays, house fire and term life (Chanel and Singh, 2025[120]).

Adaptations to the regulatory framework for microinsurance providers, or a specific regulatory framework for microinsurance, can help overcome some of the challenges related to the financial viability of microinsurance products. Tailored regulatory frameworks have been developed in Indonesia, Malaysia, Pakistan and the Philippines (and a framework is under development in Sri Lanka) (Merry and Rozo Calderon, 2023[118]):

• In Indonesia, regulations issued in 2015 and 2017 provide a definition of microinsurance and provide requirements aimed at ensuring that microinsurance products are simple, affordable and provide quick payment for valid claims (Biese et al., 2023[113]).

• In Pakistan, specific rules on the licensing of microinsurance providers and the launch of microinsurance products were implemented by the Securities and Exchange Commission of Pakistan in 2014 (SECP, 2024[116]). A recent report on the microinsurance landscape identified a number of remaining impediments to the growth of microinsurance and proposals for further regulatory adaptations, including the removal of unnecessary product approvals, an easing of agent qualification and a review of rules pertaining to bank distribution (SECP, 2024[116]). The Securities and Exchange Commission of Pakistan also provided a temporary reduction in capital requirements for microinsurance providers (along with digital-only insurers, as noted above) (SECP, 2024[116]).

• In the Philippines, a first circular defining microinsurance and the basic provisions of microinsurance products was issued in 2006 and was enhanced and amended by a set of regulations for the provision of microinsurance products that was issued in 2010. Further guidance on issues such as alternative dispute resolution mechanisms for microinsurance, distribution channels and performance standards has also been developed. In 2013, a legislative amendment to the Insurance Code was made to recognise microinsurance as a product that meets the risk protection needs of poorer segments of society and to establish maximum premiums and sums insured applicable to microinsurance (Insurance Commission, n.d.[121]). In January 2025, the Insurance Commission released a set of guidelines on inclusive microinsurance products aimed at further strengthening the effectiveness of microinsurance in meeting the protection needs of the poor (Insurance Commission, 2025[122]).

• In Malaysia, the Perlindungan Tenang initiative, initiated in 2021, has been established to enhance the availability of microinsurance and microtakaful products to meet the needs of unserved and underserved segments of the population (Bank Negara Malaysia, 2021[123]). It provides a set of requirements that insurance and takaful products must meet to qualify under the initiative, focused on affordability, accessibility, protection value, understandability and ease of use. It also offered some flexibilities in terms of product combinations (e.g. insurance coverage combined with financing) (Bank Negara Malaysia, 2021[123]). Government-provided vouchers that substantially reduced the cost of Perlindungan Tenang products were initially offered to support take-up and were reportedly successful in improving awareness of the value of insurance and takaful products among vulnerable segments of the population.

Basic natural hazard coverage

Coverage for natural hazard risks can account for a substantial portion of the premium for property insurance coverage, particularly in higher-risk areas. To support affordability, insurers could potentially offer a more basic or limited coverage for natural hazard risks as part of property insurance coverage (i.e. where the sum insured in the case of a natural hazard loss is lower than the sum insured for other types of covered risks, such as fire). In Malaysia, to address low take-up of (relatively costly) coverage for flood risks, insurers have proposed the automatic inclusion of a basic (limited) first-loss coverage for flood risk in all property insurance policies. A basic coverage for flood risk could be provided at a lower premium cost and could therefore encourage more households and businesses to acquire coverage for flood risk. Under the proposal, policyholders would be able to acquire full coverage for flood risk for an additional premium and also to opt-out of the basic coverage.

Automatically providing a basic or limited coverage at a more affordable cost could be effective in encouraging greater take-up of coverage for natural hazard risks among lower-income (or price-sensitive) households and businesses and could help insurers establish a broader and more diversified pool of natural hazard risks. However, there is also a risk that it could create confusion among policyholders about their coverage for these risks, particularly if insurers take different approaches when offering basic and comprehensive coverage. As in the case of varying deductibles for natural hazard risks, a basic coverage could also have negative implications for consumer trust if many policyholders face losses above the level of the basic coverage provided.

Parametric insurance

Parametric insurance is a type of coverage that is triggered by the occurrence of an event that meets or exceeds a pre-defined threshold, such as water level, wind speed, precipitation amount or earthquake magnitude – and makes payments of a fixed amount, regardless of the magnitude of losses incurred by the policyholder. Offering insurance coverage based on a parametric trigger can potentially support affordability by reducing underwriting and loss adjustment costs. It can also potentially offer faster post-event claims payouts, which can support quicker recovery and reduce the overall impact of natural hazard events. Microinsurance products often incorporate a parametric trigger (particularly for agricultural risks) although parametric triggers can also be applied as a complement (and possibly an alternative) to traditional indemnity-based property insurance coverage.

The offering of insurance coverage based on parametric triggers is limited in most countries although some studies project significant opportunities for growth.35 Coverage for natural hazard risks (including property and agriculture risks) reportedly accounts for a significant share (70%) of the parametric insurance market (Global Market Insights, 2025[124]). The increasing availability of new data sources, such as earth observation imagery, data from connected devices, as well as enhanced capacities to analyse these new sources of data and automate underwriting, pricing and claims decisions could support the expansion of parametric insurance options (OECD, 2023[125]).

Some insurance companies in Indonesia have developed parametric insurance coverage for earthquake risk as well as products for weather-related risks targeted at agricultural producers. In Pakistan, some parametric products have been launched on a pilot basis to cover agricultural risks. In the Philippines, some insurers are exploring parametric products for natural hazard risks although no products have been broadly introduced to the market. Insurance companies in Sri Lanka noted a limited availability of parametric insurance coverage.

In some countries, insurance legislation or regulation can impede the ability of insurance companies to offer coverage based on parametric triggers. For example, there may be a legislative or regulatory definition of insurance that is tied to the indemnification of specific losses that could place parametric insurance outside the scope of what is considered insurance (Garcia Ocampo and Lopez Moreira, 2024[126]). Legislation and regulation have not impeded the offering of parametric insurance in Indonesia, Malaysia, Pakistan, the Philippines or Sri Lanka.

In a few jurisdictions, specific legal or regulatory frameworks have been developed to support the offering of parametric insurance, including in Argentina, Colombia, Puerto Rico, Uganda and Uruguay (Garcia Ocampo and Lopez Moreira, 2024[126]). In Chile, a legislative amendment in a 2023 FinTech law allowed for the offering of parametric insurance and mandated the development of regulations to establish conditions for offering parametric insurance and identify the types of risks that can be covered by parametric insurance (Comisión para el Mercado Financiero, 2024[127]). In the US state of New York, the state insurance law was amended in 2024 to allow for the offering of parametric insurance to protect against the occurrence of weather-related events and to include requirements for mandatory disclosures when distributing parametric insurance policies (Lema, 2025[128]).

Takaful and other sharia-compliant forms of financial protection

Takaful is a form of insurance based on sharia that operates as a pool funded by the contributions of participants which is used to make payouts to those impacted by a loss. The takaful provider receives a pre-agreed fee to cover the costs of operating the pool although any surplus belongs to the participants in the pool, not the takaful provider. In countries with significant Muslim populations, supporting the availability of takaful and other sharia-compliant financial protection products could potentially support higher take-up given the higher levels of trust in takaful products, relative to insurance (as highlighted above).

Takaful and other sharia-compliant forms of financial protection against property and natural hazard risks are broadly available in Indonesia, Malaysia and Pakistan and increasingly available in the Philippines. In 2023, takaful and sharia-compliant products accounted for 4% of the overall non-life market in Indonesia (OJK, 2024[129]), 11% in Pakistan (SECP, 2024[55]) and just over 20% in Malaysia (Bank Negara Malaysia, 2024[130]; Bank Negara Malaysia, 2024[131]). Specific frameworks have been developed in Indonesia, Malaysia, Pakistan and the Philippines. In Malaysia, for example, there is a specific legislative framework (Islamic Financial Services Act) outlining the requirements for providers of takaful coverage (as well as other Islamic financial services) which is supplemented by a Takaful Operational Framework guideline established by Bank Negara Malaysia in 2019 (Bank Negara Malaysia, 2019[132]). In the Philippines, a regulatory framework was recently issued to provide guidance to insurers wishing to establish a Takaful window (Insurance Commission, 2024[133]). A lack of takaful was identified as a barrier to take-up among some segments of the population in Sri Lanka.

Some insurance and other regulators have implemented measures to support the growth of takaful and sharia-compliant insurance. In Indonesia, sharia insurers have lower minimum capital requirements than conventional insurers (Norton Rose Fulbright, 2024[134]). In Pakistan, the banking sector is required to become fully sharia-compliant by 2028, which is expected to lead to an increase in the take-up of takaful, particularly when distributed by the banking sector. However, as noted above, limited global retakaful capacity could create challenges to the expansion of takaful coverage for natural hazard risks. Bank Negara Malaysia has committed to working with the reinsurance and retakaful industry to build the necessary capacity to meet the risk transfer needs of the domestic insurance and takaful sectors (Bank Negara Malaysia, 2022[135]). The Securities and Exchange Commission of Pakistan is considering the potential role of a pooling arrangement in enhancing retakaful capacity (SECP, 2024[136]).

1.3.2. Enhancing insurance sector appetite and capacity for natural hazard risks

Improving data quality and the availability of risk assessment tools

Quality data on hazards, exposure (i.e. buildings) and vulnerability (i.e. structural resilience), as well as the capacity to analyse that data through risk assessment tools such as catastrophe models, is critical for underwriting and pricing insurance coverage for natural hazard risks. The quality of data and models impacts the willingness of insurers to underwrite coverage, their ability to charge premiums based on risk and their access to affordable reinsurance coverage, particularly proportional reinsurance.

Governments play a key role in providing risk data and assessment tools, such as hazard and risk maps for different natural hazards, data on past flood inundation levels as well as information on existing protective infrastructure such as flood barriers. Some governments have invested significantly in developing quality data and risk assessment tools and make this information broadly available for use by insurance companies and the broader public. For example, in the Philippines, the government has launched a multi-agency initiative to provide a centralised source of information on natural hazard risks, delivered through the Hazard Hunter Philippines software application. The application allows insurers (and the public) to generate hazard assessment reports for specific geographical locations and insurance companies have been encouraged by the Insurance Commission to use this tool in assessing natural hazard risks and underwriting coverage (Insurance Commission, 2020[137]). In Sri Lanka, the Disaster Risk Management Centre publishes risk maps, risk indices as well as inundation levels at district level although, as noted above, insurance companies have concerns about the accuracy of the available data and risk maps. In Malaysia, the Department of Irrigation and Drainage produces flood risk maps and collects data on past damages although data on flash floods is not collected (World Bank and Bank Negara Malaysia, 2024[34]). In addition, granular data and mapping has not been published due to concerns about the potential impact of this information on property prices in flood-prone areas (World Bank and Bank Negara Malaysia, 2024[34]). The Securities and Exchange Commission of Pakistan indicated that insurance companies are not leveraging the data and risk assessment tools available from disaster management and registration authorities (amongst others) and noted a lack of any large-scale data integration initiatives between insurers and public sector data providers (SECP, 2024[108]).

Insurance companies can, to some extent, address gaps in access to or availability of government data by leveraging their own claims data. However, as noted, low insurance penetration as well as evolving risks (e.g. climate change) limit the value of past claims data of individual insurers for underwriting and pricing natural hazard insurance coverage. In some countries, insurance companies (or insurance regulators) have established organisations that aggregate claims data from across the insurance sector and use the aggregated data to provide insurers with insights for use in risk assessment. In India, for example, all insurance companies are required to submit data to the Insurance Information Bureau which leverages that data as well as external sources (such as catastrophe models as well as government databases) to provide data services and analytics to the insurance sector (OECD, 2023[125]). Similarly, Insurance Services Malaysia Berhad (ISM) is an independent organisation that collects and aggregates data from its insurance and takaful member companies and external sources, and uses that data to generate risk assessment insights to support underwriting and pricing of coverage (ISM, 2020[138]). Some insurance companies in Sri Lanka noted the potential benefits of establishing a claims data sharing initiative. The Securities and Exchange Commission of Pakistan has worked to promote data sharing among insurers but has reportedly faced some industry reluctance that has limited the success of these efforts (SECP, 2024[108]).

The availability of catastrophe models is improving as the major model providers expand their offering and as technological developments allow for the entry of new players that leverage analytical tools like artificial intelligence to develop new modelling tools. Increased insurance penetration is usually positively correlated with demand for modelling tools, which drives availability. International organisations, such as the World Bank and Asian Development Bank, and initiatives such as the Global Risk Modelling Alliance have played a key role in supporting the development of models in various countries. With funding from the insurance sector and international development agencies, the Global Risk Modelling Alliance, for example, is leveraging open-source platforms to address critical model and data gaps around the world (Global Risk Modelling Alliance, 2025[139]). At the time of writing, the Alliance had active projects in eight countries across Latin America, Africa and Asia (including in Pakistan) (Global Risk Modelling Alliance, 2025[140]).

Collecting adequate premiums to underwrite risk

In order to ensure the availability of insurance coverage for natural hazard risks over the long-term, insurers must be able to charge an adequate price for the risks that they assume. If insurers are unable to charge an adequate price, whether due to regulatory or supervisory impediments or local market competitive dynamics, they are likely to turn to other approaches to managing their exposure, such as coverage exclusions, limits on sums insured or higher deductibles – or to not offer any coverage at all. As noted, the pricing of coverage for natural hazard risks in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka is driven by a number of regulatory and market factors. Pricing guidance or requirements are imposed in Indonesia, Malaysia and the Philippines although insurers in all three countries are authorised to set premiums based on risk – and charge higher premiums for higher risk. However, in Indonesia and the Philippines, the minimum pricing requirement has become the de facto market price due to high-levels of price-based competition. In Malaysia, the requirement to offer the tariff product limits the ability of insurers to charge premiums above the flood tariff rate. As a result, the flood tariff rate is generally applied to pricing of flood insurance coverage. There is no applicable pricing guidance in Pakistan or Sri Lanka although both the Securities and Exchange Commission of Pakistan and the Insurance Regulatory Commission of Sri Lanka indicated that there was limited price differentiation applied to the natural hazard insurance coverage offered by insurers (although, as noted, some use by insurers of increased deductibles and exclusions to manage exposure). Price-sensitive consumers and intense competition may also limit pricing variation in Pakistan and Sri Lanka. The limited ability of some insurers to assess natural hazard risk could also limit the use of risk-based pricing (in Pakistan and Sri Lanka as well as in Indonesia, Malaysia and the Philippines).

Box 1.4. Pricing adequacy in the Philippines

Pricing for natural hazard insurance coverage is considered to be inadequate by some insurers and reinsurers in the Philippines, particularly for typhoons and floods. The non-life insurance sector, including reinsurers, have faced significant typhoon losses in recent years with loss ratios above 100% in most recent years (and sometimes significantly above 100%). Loss ratios for flood losses have also exceeded 100% in some recent years, although not by as much as typhoon losses (Figure 1.22).

In recognition of the inadequacy of the minimum (and prevailing) rates, the Insurance Commission issued a circular letter in 2022 revising the minimum rates for earthquake and typhoon/flood insurance coverage. The revised minimum rates varied depending on risk zone, occupancy and construction type, and led to rate reductions for buildings with better construction standards in low-risk zones for both earthquake and typhoon/flood coverage (to a lesser extent) along with higher minimum rates overall for typhoon/flood coverage (Insurance Commission, 2022[141]). However, stemming from several considerations, a new circular letter was issued in 2024 reinstating the previous minimum rates for both earthquake and typhoon/flood coverage (Insurance Commission, 2024[142]).

Figure 1.22. Loss ratios in fire and natural hazard insurance in the Philippines

Source: Insurance Commission (2023[53]), Annual Report 2022, https://www.insurance.gov.ph/wp-content/uploads/2025/02/2022-IC-Annual-Report.pdf; Insurance Commission (2021[52]), Annual Report 2020, https://www.insurance.gov.ph/wp-content/uploads/2023/01/2020-Annual-Report-FINAL.pdf; Insurance Commission (2019[51]), Annual Report 2018, https://www.insurance.gov.ph/wp-content/uploads/2022/04/2018-Annual-Report-Final-compressed.pdf.

Limited price differentiation can lead to particular challenges if the prevailing market rate is broadly inadequate as inadequate pricing can lead insurers to limit coverage and result in higher reinsurance pricing. As mentioned, insurance and reinsurance companies in Malaysia and Pakistan indicated that pricing for natural hazard insurance coverage is generally adequate, while in Sri Lanka, pricing is mostly adequate but not high enough to attract sufficient coverage in international reinsurance markets. However, in Indonesia and the Philippines, prevailing market rates are seen by insurers and/or reinsurers as inadequate (Box 1.4).

Limited price differentiation by insurers can have other implications. Rates that do not vary with risk will not provide any risk signals or incentives for policyholder risk reduction (to the extent that policyholders have the ability to implement effective risk reduction measures) as investments to enhance resilience will not be rewarded with lower premiums. Insurance companies and/or insurance regulators in Malaysia, Pakistan, the Philippines and Sri Lanka indicated that policyholders are sometimes rewarded for taking risk reduction measures with lower premium rates, despite the limited application of risk-based pricing. However, this is likely limited to the larger corporate policyholders that are more likely to engage with insurers and intermediaries on risk management. Insurers and intermediaries, in Asia and elsewhere, do not usually provide tailored risk advice or offer substantial premium discounts for risk reduction measures to households and small businesses given the more limited amount of premiums collected from smaller policyholders (OECD, 2023[143]). Bank Negara Malaysia’s Financial Sector Blueprint for 2022-2026 highlighted the potential role of digital tools in supporting climate risk adaptation, including by encouraging lower insurance premiums to reward the implementation of adaptation measures (Bank Negara Malaysia, 2022[135]). The International Association of Insurance Supervisors’ recent application paper on the supervision of climate-related risks also encourages supervisors to promote risk mitigation by ensuring that insurers advertise and offer appropriate premium discounts for risk mitigation measures (IAIS, 2025[102]).

Limited price differentiation in natural hazard insurance coverage could also limit the incentives for insurance companies to make significant investment in developing risk assessment and modelling tools. Insurers will have few incentives to collect new data or apply advanced analytics if better risk assessment has no impact on pricing (OECD, 2023[125]). As a result, the insurance sector contribution to improving the availability of quality data and risk assessment tools is likely to be limited.

Insurance regulators and supervisors have a key role in overseeing the adequacy (as well as the fairness) of insurance pricing. In some cases, including in Indonesia, the Philippines and Sri Lanka, insurance supervisors review insurers’ approach to pricing their coverage before providing approval for the introduction of new products. The International Association of Insurance Supervisors recently highlighted the importance of the role of supervisors in ensuring “that pricing of NatCat products reflects adequate risk-based technical models” in the context of evolving climate risks (IAIS, 2025[102]).

The capital requirements established and supervised by insurance regulators and supervisors can also have an important impact on pricing adequacy. Higher minimum capital requirements can encourage insurance sector consolidation and potentially reduce the competitive pressures that lead to intensive price competition. Minimum capital requirements have recently been increased in Indonesia, Pakistan and the Philippines. In some cases, market fragmentation and excessive competition were identified as a rationale for the capital requirement increase. The application of risk-based supervision and risk-based capital or solvency requirements should also contribute to ensuring adequate pricing. Indonesia, Malaysia, the Philippines and Sri Lanka have implemented a risk-based capital framework (Norton Rose Fulbright, 2024[134]). The Securities and Exchange Commission of Pakistan intends to implement risk-based capital requirements by 2027 (SECP, 2024[103]).

Given the complexity of assessing and quantifying natural hazard risks, supervisors in Asia and elsewhere will likely face difficulties in overseeing pricing adequacy through product approval and risk-based supervision and capital requirements. Assessing pricing adequacy and pricing-related risks requires supervisory resources with sufficient actuarial and modelling expertise – which can be challenging in countries with a limited pool of qualified actuaries and modellers. Insurance regulators and supervisors in a number of the countries that contributed to this report identified a lack of actuarial and/or modelling expertise as a significant challenge.

In recognition of the increasing risk posed by natural hazards, the complexity of risk assessment in a changing climate and the growing use of sophisticated catastrophe models, insurance supervisors in Europe and the United States are investing additional resources to help build supervisory expertise in catastrophe modelling. The European Insurance and Occupational Pensions Authority has established a centre of excellence for catastrophe modelling and data aimed at providing European supervisors and insurers with expertise, data and tools to support risk assessment and the supervision of natural hazard risks (EIOPA, n.d.[144]). In the United States, the National Association of Insurance Commissioners has established a Catastrophe Modeling Center of Excellence to provide regulators with technical training and expertise on catastrophe models and to apply catastrophe models in assessing the risk of losses from natural hazards (NAIC, n.d.[145]).

Sustainable access to affordable reinsurance

Access to affordable reinsurance coverage is critical for ensuring the availability of affordable insurance coverage for natural hazard risks for households and businesses. Some insurers in Malaysia, Pakistan, the Philippines - and particularly in Indonesia and Sri Lanka - have faced challenges in accessing affordable reinsurance coverage due to the quality of data on insured exposures and adequacy of premium pricing.

Reinsurance companies as well as intermediaries often play a critical role in providing access to data and risk assessment tools. International reinsurance companies often develop their own models or acquire licenses to models from other providers and have dedicated modelling teams to support their approach to assuming and pricing risk from insurers. They also play an important role in supporting pricing adequacy. In the case of proportional placements, they can encourage pricing adequacy by setting pricing conditions as a prerequisite to assuming risk. In the case of non-proportional placements, reinsurers may increase insurer retentions (i.e. insurer deductibles) as a means to encourage insurers to improve pricing adequacy.

Regulatory restrictions on access to international markets could have an impact on the contribution that international reinsurers make to risk assessment and pricing adequacy. While preferential placements with national reinsurers and some limitations on cross-border risk transfer exist in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka, international reinsurers are nonetheless significant participants in assuming risk from all of these countries and broadly contribute to risk assessment and pricing adequacy.

The national reinsurers in Indonesia and Malaysia have also supported insurance sector access to data and tools. For example, Malaysia Re developed an in-house flood model and provided access to the model to its insurance consumers (although the model is no longer in use). Indonesia Re operates the National Insurance Data Center Management Agency (BPPDAN) which provides annual statistics on loss experience by occupation based on data collected from across the insurance sector (Indonesia Re, 2025[146]). Maipark, which is an industry-owned reinsurer for earthquake and others natural hazard risks in Indonesia, has developed a probabilistic earthquake risk model and is working on expanding it to incorporate flood and other climate-related risks. Maipark also provides training to insurance companies on how to apply its modelling tools.

The national reinsurers in Indonesia and the Philippines have also contributed to pricing discipline. After facing significant losses in recent years, Indonesia Re and the other Indonesian reinsurers have increased pricing for non-proportional placements and imposed higher retentions on both proportional and non-proportional placements in order to improve the ceding insurers’ incentives to ensure pricing adequacy (given the increased exposure that insurers will need to retain). In the Philippines, Nat Re is imposing compliance with the minimum tariff pricing by excluding non-compliant policies from treaty placements.

National reinsurers play different roles in providing reinsurance capacity to the domestic insurance market. The national reinsurers in Indonesia, Malaysia and the Philippines tend to be significant capacity providers and are often offered a greater share of placements than required by the preferential system. The national reinsurers in Pakistan and Sri Lanka face more capacity constraints and are not generally offered more than required. They also play different roles for different types of insurers. The national reinsurers in the Philippines and Pakistan, for example, tend to be a more important capacity provider for smaller insurers. However, as noted above, none of the national reinsurers are able to play a countercyclical role during hard reinsurance markets given the (syndicated) nature of reinsurance placements and their reliance on retrocession to international markets.

The differences in national reinsurer capacity and market impact are driven by a number of factors. The National Insurance Trust Fund in Sri Lanka is a wholly owned government agency and is perceived by insurers as operating like a government department rather than a private sector reinsurance company. Insurers in Pakistan have a similar perception of the Pakistan Reinsurance Company, even though the company is partially listed. The two wholly owned national reinsurers in Indonesia, on the other hand, are generally perceived as professional reinsurance companies.

The national reinsurers also have different capacities for assuming risk from outside their home jurisdiction, and therefore differing abilities to establish an internationally diversified portfolio of risks. Malaysia Re has assumed significant risk from outside Malaysia (more than 50% of its portfolio) and can compete based on its high credit rating (which is higher than Malaysia’s sovereign rating). Nat Re also assumes business outside of the Philippines although not to the same extent as Malaysia Re. The government-owned national reinsurers in Indonesia have more difficulty competing for international business as they have not been provided with sufficient capital from the Indonesian government to achieve the necessary credit rating. The National Insurance Trust Fund in Sri Lanka and the Pakistan Reinsurance Company do not assume international business and do not have independent credit ratings.36 As noted above, the lack of international diversification creates concentration risk for those national reinsurers and can ultimately lead to counterparty risks for domestic insurers that have ceded significant risk to those national reinsurers.

Ultimately, policy makers should aim to find a sustainable balance that supports long-term access to affordable reinsurance capacity. Improved access to international reinsurance and capital markets (Box 1.5) can help ensure that capacity is available to absorb the expected increase in natural hazard losses. International reinsurance markets allow for losses to be diversified away from the domestic financial system and have likely reduced the economic disruption caused by past significant events (OECD, 2018[45]).37

Box 1.5. Risk transfer to capital markets

Through insurance-linked securities and other alternative reinsurance structures, insurers and reinsurers around the world have increasingly transferred natural hazard risks directly to capital markets. According to one estimate, the annual issuance of catastrophe bonds and other insurance-linked securities has increased from approximately USD 1.4 billion annually between 2000 and 2004 to more than USD 14.1 billion annually between 2020 and 2024 (Artemis, n.d.[147]). The development of the insurance-linked securities market has provided an additional source of reinsurance capacity for natural hazard risks and has likely dampened the impact of cyclical reinsurance pricing (OECD, 2018[45]).

In some countries, regulators may not allow supervised insurers to transfer risk to capital markets or may not allow insurers to recognise such transfers as effective reinsurance arrangements that warrant a corresponding reduction in reserve and capital requirements (OECD, 2020[148]). Insurance companies and/or regulators in Malaysia, Pakistan, the Philippines and Sri Lanka did not identify any limitations on the use or recognition of risk transfer to capital markets, although some insurers in Indonesia indicated that risk transfer to capital markets is not permitted.¹ The Securities and Exchange Commission of Pakistan has indicated an interest in increasing insurance-linked securities market capacity to assume risk from insurers in Pakistan (SECP, 2024[103]).

Figure 1.23. Catastrophe bond and insurance-liked securities outstanding for natural hazard perils

Note: 1. An OECD examination of the regulatory framework for reinsurance in Asian countries in 2020 identified some limitations in the recognition of risk transfer to capital markets in Indonesia and the Philippines (OECD, 2020[148]).

Source: OECD calculations based on data provided by Artemis (n.d.[149])

However, there has been very limited issuance of capital bonds and other insurance-linked securities to provide financial protection against natural hazard risks in Asia (outside of Japan) (Figure 1.23). There have been a number of efforts to expand this market in Asia, including the development of tailored licensing and regulatory frameworks for catastrophe bond issuance in Singapore and Hong Kong (China) (OECD, 2020[148]).

National reinsurers can make an important contribution to supporting that access by addressing gaps in capacity and promoting efforts to improve data quality and pricing adequacy among domestic insurers. National reinsurers receive a benefit through their preferential access to domestic reinsurance placements and should therefore be required to contribute to building sustainable access to affordable reinsurance.

Establishing public private insurance programmes

Public-private insurance programmes (PPIPs), also referred to as catastrophe risk insurance programmes, are arrangements established by the insurance sector and/or government to provide insurance, co-insurance, reinsurance and/or a government guarantee for losses resulting from natural hazards or other types of catastrophe perils to households and businesses (OECD, 2021[150]). These arrangements may be established by the private insurance sector, such as through the creation of a co-insurance pool among insurers or a reinsurer owned by the insurance sector. They could also be established by the public sector as a state-owned insurer or reinsurer.

In the context of increasing losses and challenges in the availability and affordability of private insurance coverage, an increasing number of OECD countries have established public-private insurance programmes in recent years.38 These programmes have taken various forms, including government-owned direct insurers or compensation programmes (e.g. Denmark, Iceland, New Zealand, Spain, Switzerland39, the United States40) and government-owned reinsurers (Australia, France, Japan, the United States⁴¹), as well as private sector co-insurance pools (Belgium, Norway, Switzerland⁴⁰) and special-purpose insurers and reinsurers (Republic of Türkiye (hereafter “Türkiye”), the United Kingdom). Recognising the potential need for these types of arrangements, the G7 Finance Ministers and Central Bank Governors welcomed a High-Level Framework for Public-Private Insurance Programmes against Natural Hazards at their meeting in May 2024, developed by the G7 Climate Change Mitigation Working Group with input from the OECD and the International Association of Insurance Supervisors (G7 Finance Track, 2024[151]). The High-Level Framework provides a set of considerations for the establishment of public-private insurance programmes to address protection gaps for natural hazard risks, including possible objectives and design features.

Public-private insurance programmes can contribute to addressing protection gaps for natural hazard risks by: (i) pooling the risks of many policyholders in order to achieve a more diversified pool of risks that reduces the potential that a large share of the pool’s covered policyholders are affected by losses simultaneously and the corresponding amount of capital and reserves that would need to be set-aside, which can support affordability; (ii) limiting the potential magnitude of losses that individual insurers or reinsurers could face, with similar benefits in terms of capital and reserve needs and affordability; and (iii) improving the availability of data and modelling of natural hazard risks (OECD, 2021[150]). The OECD countries that have established public-private insurance programmes tend to have achieved higher levels of insurance coverage of natural hazard losses and have faced fewer protection gaps (Figure 1.24).

Figure 1.24. The impact of PPIPs on flood insurance protection gaps

Source: OECD calculations based on data provided by Swiss Re (Swiss Re, sigma database. All rights reserved.).

The establishment of public-private insurance programmes to support coverage for natural hazard risks to households and businesses has been less common in Asia, and in emerging and developing countries more generally.41 A few programmes have been established in emerging and developing countries, including in Algeria, Morocco, Romania and (as noted above) Türkiye:

• In Algeria, households and businesses are required to acquire a “CatNat” coverage for natural hazard risks with reinsurance available from a government reinsurer (Compagnie Centrale de Réassurance). The programme covers earthquakes, floods, landslides, mudslides and storm and violent winds.

• In Morocco, property insurance coverage for households and businesses is automatically extended to include coverage for natural hazard risks (floods, mudslides, earthquakes and tsunamis) and assumed by an industry co-insurance pool.

• In Romania, a special-purpose insurance company (Pool-ul de Asigurare Împotriva Dezastrelor Naturale (PAID)) has been established to provide basic coverage for households for losses from earthquakes and floods. The stand-alone coverage is mandatory for all households.

• In Türkiye, a public insurer (Doğal Afet Sigortaları Kurumu or Turkish Catastrophe Insurance Pool) has been established to provide basic coverage for households and some businesses against losses from earthquakes and from fires, tsunamis or landslides caused by earthquakes. The stand-alone coverage is mandatory for households (and some businesses) in urban areas. There are plans to extend the programme to include other natural hazard risks.

A public-private insurance programme has also been established in Indonesia for earthquake and related risks (i.e. the natural hazards included in the EQVET (earthquake, volcanic eruption and tsunami) coverage offered by insurers). Maipark is a reinsurance company owned by all non-life insurance companies that provides reinsurance to Indonesian insurers for the EQVET coverage provided to households and businesses.42 All Indonesian non-life insurers cede a minimum amount of this risk to Maipark (10% for the island of Java and 25% elsewhere) on a compulsory basis with the possibility of ceding a greater amount. As noted above, Maipark also plays a critical role in supporting the availability of data and modelling on earthquake and other natural hazard risks.

There have also been efforts to establish a programme in the Philippines. The insurance industry, the national reinsurer (Nat Re) and the Insurance Commission collaborated to establish the Philippine Catastrophe Insurance Facility (PCIF) as a voluntary43 co-insurance and reinsurance arrangement for natural hazard risks, with the aim of expanding the take-up of coverage for natural hazard risks on a sustainable basis. The pool would assume up to 25% of the natural hazard risk underwritten by participating insurance companies and retrocede a share to member insurers, Nat Re and international markets. Given the need to ensure pricing adequacy within the PCIF, a working group undertook modelling of natural hazard risks with the aim of updating the pricing framework for natural hazard risks, culminating in the new pricing framework proposed in 2022 (Insurance Commission, 2022[141]) which was revoked in 2024 (Insurance Commission, 2024[142]). The reinstatement of the previous minimum tariffs has delayed the implementation of the PCIF although the members have reportedly agreed to proceed with the arrangement for earthquake coverage given that the pricing is deemed to be more adequate for this risk.

Sri Lanka established a programme through the National Insurance Trust Fund, a government agency and national reinsurer, that focused on providing coverage for uninsured households and small businesses, but it is no longer in operation. The National Natural Disaster Insurance Policy provided coverage between 2016 and 2018 for damages from cyclones, storms, floods, landslides, earthquakes, tsunamis and other natural hazard perils. The programme was backed by a specific reinsurance policy and incorporated a ceiling on payments per event and per year, calibrated to the acquired reinsurance policy. However, major flooding in Colombo in 2016 and other factors led to an increase in the cost of reinsurance coverage that eventually led to the termination of the programme. Insurance companies in Sri Lanka have proposed the need for a new programme to provide basic and affordable coverage through some form of pooling arrangement with government backing.

The Securities and Exchange Commission of Pakistan is interested in facilitating the creation of an insurance pool (SECP, 2024[103]) and developed a report examining international approaches and potential models for such a pool in Pakistan (SECP, 2024[136]). The report considers three possible structural approaches to pooling (voluntary co-insurance arrangement, a separate legal entity and a statutory body) and the possibility of including agricultural risks, motor third party liability as well as natural hazard risks (SECP, 2024[136]).

Bank Negara Malaysia’s Financial Sector Blueprint for 2022-2026 similarly includes a commitment to explore the possible establishment of public-private risk pools to deal with high-risk, large-scale perils, including climate-related natural hazards as well as pandemics, cyber risk and environmental risks (Bank Negara Malaysia, 2022[135]). A possible approach identified for consideration would involve a first-loss coverage from the private insurance and takaful sector with potential public sector support for initial capacity development and backstops. As noted above, some Malaysian insurers have proposed the inclusion of basic, first-loss flood insurance coverage on an opt-out basis, which would be consistent with the approach outlined in the Financial Sector Blueprint. The World Bank has also recommended a compulsory basic coverage as a means to address challenges to the availability of flood insurance coverage for high-risk SMEs (World Bank and Bank Negara Malaysia, 2024[34]).

The establishment of public-private insurance programmes in emerging markets and developing countries needs to consider several factors. These include: (i) the government’s policy objectives in terms of who should be protected; (ii) the level of insurance market development (and financial sector development more broadly); and (iii) the financial and institutional capacity of the government to prepare for and respond to disasters (Adam, Hallegate and Mahul, 2025[152]). In the case of programmes that target households and businesses in developing countries with higher levels of insurance and financial market development, the note proposes an approach that leverages domestic insurance market and international reinsurance market capacity, provides premium subsidies to households and micro and small business, and incorporates a contribution from governments as reinsurers of last resort (Adam, Hallegate and Mahul, 2025[152]).

The established public-private insurance programmes take a broad range of approaches to leveraging private (re)insurance market capacity and incorporating public financial support. Some programmes act as residual programmes focused on high-risk households, and possibly businesses, and transfer significant risk to private reinsurance or retrocession markets (e.g. Flood Re in the United Kingdom) while others are the main providers of insurance coverage for natural hazard risks and retain most of the risk that they assume (e.g. Consorcio de Compensación de Seguros in Spain). Very few of the existing models incorporate explicit premium subsidies although many provide implicit subsidies and/or cross-subsidies by incorporating minimal risk variation in pricing or by ensuring that pricing is affordable, even if that leads to more frequent losses for the government as backstop provider.

The existing programmes in emerging and developing countries have tended to be the main providers of coverage for natural hazard risks in the countries where they operate (rather than residual providers), although all of these programmes transfer significant risk to international reinsurance markets. The programmes in Romania and Türkiye provide only basic (first-loss) coverage which could be supplemented by private insurance for residual natural hazard losses although the market for private (supplemental) insurance is limited in both countries and a significant share of losses have remained uninsured.

Premiums for natural hazard insurance coverage in Algeria, Morocco and Romania are not significantly risk-based and therefore entail some cross-subsidies. Premiums for earthquake coverage in Türkiye incorporate some risk factors and the proposed approach to premium pricing for other hazards will reportedly be risk-based. There are no explicit premium subsidies incorporated into any of these programmes. However, cross-subsidisation between low- and high-risk policyholders, as well as the limits on sum insured in Romania and Türkiye, help ensure affordability – which is critical for programmes that impose mandatory purchase requirements. The programme in Morocco incorporates an innovative feature to support more vulnerable segments of the population that are unlikely to be insured. A small premium tax is added to natural hazard coverage to create a solidarity fund that makes compensation payments to uninsured households impacted by natural hazards. As noted above, the discontinued National Natural Disaster Insurance Policy in Sri Lanka similarly focused on providing support to those unable to acquire insurance coverage. The Philippine Catastrophe Insurance Facility is also considering the incorporation of a limited (microinsurance) coverage for households and micro and small businesses.

The level of government reinsurance or backstop support also tends to be more limited for the programmes in emerging and developing countries (relative to the support provided in developed countries). The programme in Türkiye has access to an “Excess-of-Loss” arrangement with the government if it has not been able to secure sufficient financial protection in private reinsurance markets (DASK, 2012[153]) while the programme in Romania has access to repayable government loans if needed. An inability to access government support could have implications for the programme’s ability to provide significant capacity. For example, Maipark in Indonesia does not benefit from a government reinsurance coverage or backstop and may face capacity constraints in the future if it is unable to raise sufficient capital from its shareholders to meet higher capital requirements (as its insurance company shareholders also face an increase in capital requirements).

1.4.  Conclusion

Emerging and developing Asia and Pacific countries face significant protection gaps in natural hazard insurance, which leave households and businesses unprotected against damages and losses from natural hazards and could create burdens on public finances and potential risks to financial stability. Economic losses from natural hazards are increasing across the region, driven in part by the impacts of changes in hazard frequency and intensity. The increasing losses could lead to an increase in the cost of insurance that will make it increasingly difficult to close protection gaps for natural hazard risks.

This chapter examined the drivers of protection gaps in Asia focusing on the experiences of Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka. Protection gaps in natural hazard insurance result from factors that limit the demand for insurance coverage and reduce the willingness of households and businesses to pay for insurance coverage, as well as factors that impact the supply of insurance coverage which can limit the availability and affordability of insurance coverage for these risks.

There appears to be a high level of risk awareness and perception of natural hazards in many parts of Asia relative to other parts of the world given their frequency. However, the relatively high level of natural hazard risk perception does not translate to broad take-up of insurance coverage. This is likely due to a limited understanding of the need for and benefits of insurance coverage to protect against these risks and a lack of trust in insurance as an effective form of financial protection. Differences in insurers’ approach to offering natural hazard insurance coverage may also be leading to misunderstandings about the availability of coverage and the need, in some cases, to acquire specific additional coverage. For lower-income segments of the population and those in high-risk areas, insurance coverage may simply be unaffordable given other more urgent expenditure needs.

Limited access to quality data and risk assessment tools such as catastrophe models can lead insurers to limit the coverage they make available to households and businesses. It can also inhibit their ability to price premiums based on risk, although intensive competition and price-sensitive consumers discourages insurers from charging higher premiums in higher risk areas. As a result, insurers tend to impose higher deductibles, limit the sums insured or exclude coverage for some or all natural hazard risks, rather than charge higher premiums. The challenges in quantifying natural hazard risks, along with concerns about premium pricing adequacy in some countries, have led to some difficulties in accessing affordable reinsurance coverage, particularly proportional coverage. A lack of access to affordable reinsurance reduces the appetite of insurers for providing coverage for natural hazard risks to households and businesses.

Translating the relatively high-level of risk awareness into broader levels of financial protection for natural hazard risks will require further investment across a number of areas, including:

• Improving insurance literacy, with a focus on building understanding of the benefits of, and need for, insurance coverage against natural hazard risks.

• Increasing trust in insurance companies and insurance coverage as an effective form of financial protection, including through efforts to improve awareness of the role of insurance supervisors in protecting consumers and, where applicable, the protection provided through policyholder protection schemes.

• Augmenting property insurance penetration by: (i) enhancing the use of digital distribution channels and bancassurance, leveraging increased access to banks and digital technologies; and (ii) encouraging the development of suitable and affordable insurance coverage products, including microinsurance, parametric and takaful and sharia products, and providing flexibilities that support the attractiveness and reach of these products.

• Expanding the take-up of natural hazard coverage by encouraging insurers to implement consistent approaches in how they offer this coverage and improve clarity on coverage and exclusions, and by considering the potential advantages (and disadvantages) of requiring insurers to offer natural hazard coverage (or basic natural hazard coverage) automatically in property insurance coverage.

Ensuring the availability of insurance coverage for natural hazard risks will require:

• Efforts to improve access to quality data and modelling tools, including by improving the quality of – and access to – government data on hazards and exposures, examining the establishment of claims data sharing initiatives within the insurance sector and leveraging opportunities to access international assistance for model development.

• Improving the ability of insurers to adequately price coverage for natural hazard insurance and provide pricing incentives for risk reduction, as well as the capacity of insurance supervisors to oversee pricing adequacy, including by building expertise in the use of catastrophe models and other sophisticated risk assessment tools.

• Supporting sustainable access to affordable reinsurance capacity, including through capital markets, by addressing the factors that impede access (data quality, pricing adequacy) and enhancing the contribution of national reinsurers to addressing gaps in capacity, including retakaful capacity.

Public-private insurance programmes can make an important contribution to ensuring the availability of affordable insurance coverage for natural hazard risks. The design of such programmes should take into account local factors, such as the level of insurance market development and the capacity of governments to share some of the risk. Responding to the needs of vulnerable segments of the population should be a key objective in the design of such programmes.

The challenges outlined in this chapter to increasing demand for natural hazard insurance coverage and enhancing the insurance sector’s capacity to assume natural hazard risks are shared by countries around the world. As a result, there are significant opportunities to share experience and learn from the efforts of other countries to address these challenges.

References

[75] AAUI (2018), Vision and mission, Asosiasi Asuransi Umum Indonesia.

[157] Abbas, A. et al. (2015), “Non-structural flood risk mitigation under developing country conditions: an analysis on the determinants of willingness to pay for flood insurance in rural Pakistan”, Natural Hazards, Vol. 75/3, pp. 2119-2135, https://doi.org/10.1007/S11069-014-1415-X/TABLES/4.

[79] Ada derana Business (2019), “Boosting the Insurance industry and the role of the MSF”, Ada derana Business.

[152] Adam, B., S. Hallegate and O. Mahul (2025), Mobilizing public-private solutions to manage the financial impacts of natural hazards in emerging markets and developing economies: Challenges and opportunities of operationalizing Public-Private Insurance Programs, World Bank.

[3] ADB (2024), Asia-Pacific Climate Report 2024, Asian Development Bank, Manila, Philippines, https://doi.org/10.22617/SGP240498-2.

[16] Adhi, I. et al. (2019), “The Right to Live Dangerously: Public Perceptions of Extreme Water Events in Urban Areas”, pp. 29-36, https://doi.org/10.2991/ICSGS-18.2019.5.

[99] Allen, T. et al. (2023), “Climate change and residential real estate: what are the risks for the banking sector?”, Bulletin, Banque de France.

[35] Aon (2024), Navigating Catastrophe Risk: Model Coverage, ELEMENTS 18.

[63] Artemis (2025), Guy Carpenter Global Property Catastrophe Rate-On-Line Index, Artemis.

[64] Artemis (2025), Guy Carpenter Regional Property Catastrophe Rate-On-Line Index, Artemis.

[147] Artemis (n.d.), Catastrophe bonds & ILS issued and outstanding by year, Artemis.

[149] Artemis (n.d.), Catastrophe bonds & ILS outstanding by risk or peril, Artemis.

[87] ASEAN Insurance Council (2023), Comparison Among Countries in ASEAN with Existing Insurance Policy Guarantee Scheme.

[88] Asia Insurance Review (2023), “Deposit Insurance Corporation gets ready to implement Policy Guarantee Programme”, Asia Insurance Review.

[119] Asian Development Bank (2024), The Enabling Environment for Disaster Risk Financing in the Philippines, Asian Development Bank, Manila, Philippines, https://doi.org/10.22617/TCS240611-2.

[165] Ayyagari, M. and T. Beck (2015), “Financial Inclusion in Asia: An Overview”, Economics Working Papers, Asian Development Bank.

[82] Bank Negara Malaysia (2025), Financial Consumer Alert List, Bank Negara Malaysia.

[130] Bank Negara Malaysia (2024), General Insurance: Distribution of Gross Direct Premiums, Bank Negara Malaysia.

[131] Bank Negara Malaysia (2024), General Takaful: Distribution of Gross Contributions of Direct Takaful Operators, Bank Negara Malaysia.

[110] Bank Negara Malaysia (2024), Policy Document on Licensing and Regulatory Framework for Digital Insurers and Takaful Operators, Bank Negara Malaysia.

[135] Bank Negara Malaysia (2022), Financial Sector Blueprint: 2022-2026, Bank Negara Malaysia.

[123] Bank Negara Malaysia (2021), Perlindungan Tenang, Bank Negara Malaysia.

[132] Bank Negara Malaysia (2019), Takaful Operational Framework, Bank Negara Malaysia.

[100] Bank of England (2019), Climate change: what are the risks to financial stability?, Bank of England.

[113] Biese, K. et al. (2023), Inclusive insurance and risk financing in Indonesia: Snapshot and way forward 2023, United Nations Development Programme.

[11] Bronfman, N. et al. (2020), “Understanding the Relationship Between Direct Experience and Risk Perception of Natural Hazards”, Risk Analysis, Vol. 40/10, pp. 2057-2070, https://doi.org/10.1111/RISA.13526.

[111] Brown, J. and A. Goolsbee (2000), “Does the Internet Make Markets More Competitive? Evidence from the Life Insurance Industry”, SSRN Electronic Journal, https://doi.org/10.2139/ssrn.253795.

[170] Bündnis Entwicklung Hilft and IFHV (2024), World Risk Report, Bündnis Entwicklung Hilft.

[41] CATRisk Solutions (n.d.), CATRisk Geographic Model Coverage, CATRisk Solutions.

[70] Central Bank of Sri Lanka (2024), Financial Literacy Roadmap of Sri Lanka (2024 - 2028), Central Bank of Sri Lanka.

[69] Central Bank of Sri Lanka (2021), National Financial Inclusion Strategy: 2021-2024, Central Bank of Sri Lanka.

[20] Centre for Research on the Epidemiology of Disasters (CRED) (n.d.), EM-DAT.

[120] Chanel, S. and R. Singh (2025), “UN, Samoa Surety Insurance to unveil “transformative” climate solution”, Samoa Observer.

[171] Chen, C. et al. (2025), Vulnerability (2022), University of Notre Dame Global Adaptation Initiative.

[2] Chen, C. et al. (2025), Exposure (2022), University of Notre Dame Global Adaptation Initiative.

[9] Cisternas, P. et al. (2024), “The influence of risk awareness and government trust on risk perception and preparedness for natural hazards”, Risk Analysis, Vol. 44/2, pp. 333-348, https://doi.org/10.1111/RISA.14151.

[43] COMBUS (n.d.), Models, COMBUS Pty.

[127] Comisión para el Mercado Financiero (2024), CMF publishes for consultation regulation on parametric insurance, Comisión para el Mercado Financiero.

[27] Courbage, C. and C. Nicolas (2019), “Who Trusts Insurance? Empirical Evidence from Seven Industrialised Countries”, SSRN Electronic Journal, https://doi.org/10.2139/SSRN.3404821.

[153] DASK (2012), Law, Doğal Afet Sigortaları Kurumu.

[30] Department of Statistics Malaysia (2022), Household Income by Percentile, Department of Statistics Malaysia.

[4] EIOPA (2024), Measures to address demand side aspects of the NatCat protection gap, European Insurance and Occupational Pensions Authority.

[89] EIOPA (2019), Consultation Paper on Proposals for Solvency II 2020 Review Harmonisation of National Insurance Guarantee Schemes, EIOPA.

[144] EIOPA (n.d.), Centre of excellence for catastrophe modelling and data, European Insurance and Occupational Pensions Authority.

[92] Esho, N. et al. (2004), “Law and the Determinants of Property-Casualty Insurance”, The Journal of Risk and Insurance, Vol. 71/2.

[98] European Central Bank and European Systemic Risk Board (2023), Towards macroprudential frameworks for managing climate risk, European Systemic Risk Board.

[160] Everett-Allen, K. (2022), Which countries have the highest proportion of mortgage-free households?, Knight Frank.

[71] Financial Education Network (2025), Discussion Paper on the Proposed Features of Malaysia’s Second National Strategy for Financial Literacy (2026 – 2030), Bank Negara Malaysia.

[72] Financial Inclusion Steering Committee (2024), 2023 Annual Report: National Strategy for Financial Inclusion, Financial Inclusion Steering Committee.

[151] G7 Finance Track (2024), High-Level Framework for Public-Private Insurance Programmes against Natural Hazards.

[14] Gallagher, J. (2014), “Learning about an Infrequent Event: Evidence from Flood Insurance Take-Up in the United States”, American Economic Journal: Applied Economics, Vol. 6/3, pp. 206-233, https://doi.org/10.1257/app.6.3.206.

[126] Garcia Ocampo, D. and C. Lopez Moreira (2024), “Uncertain waters: can parametric insurance help bridge NatCat protection gaps?”, FSI Insights on policy implementation, No. 62, Bank for International Settlements.

[124] Global Market Insights (2025), Parametric Insurance Market Size - By Coverage, By Distribution Channel, By Application, By End Use, Analysis, Share, Growth Forecast, 2025 - 2034, Global Market Insights.

[139] Global Risk Modelling Alliance (2025), About the Alliance, Global Risk Modelling Alliance.

[140] Global Risk Modelling Alliance (2025), Country map, Global Risk Modelling Alliance.

[94] Government of Indonesia (2023), Undang-undang (UU) Nomor 4 Tahun 2023 tentang Pengembangan dan Penguatan Sektor Keuangan, Database Peraturan.

[46] Guy Carpenter (2025), January 1, 2025 Reinsurance Renwal, Guy Carpenter.

[62] Guy Carpenter (2024), APAC Property Catastrophe Rate on Line Index.

[61] Guy Carpenter (2024), Global Property Catastrophe Rate on Line Index.

[60] Guy Carpenter (2024), Strong reinsurer appetite drives excess property catastrophe capacity at January 1, Guy Carpenter.

[28] Hassan, H. and S. Abbas (2019), “Factors influencing the investors’ intention to adopt Takaful (Islamic insurance) products: A survey of Pakistan”, Journal of Islamic Marketing, Vol. 11/1.

[159] Helgi Library (2025), Mortgage Loans (as % of GDP), Helgi Library.

[162] Helgi Library (2025), Mortgage Loans (As % of GDP) in Indonesia, Helgi Library.

[163] Helgi Library (2025), Mortgage Loans (As % of GDP) in Malaysia, Helgi Library.

[161] Helgi Library (2025), Mortgage Loans (As % of GDP) in Philippines, Helgi Library.

[167] Hoque, M., S. Hossain and M. Siddique (2023), “Financial Inclusion and its Determinants: Empirical Evidence from South Asia”, The Journal of Developing Areas, Vol. 57/3, pp. 199-211, https://doi.org/10.1353/JDA.2023.A907742.

[102] IAIS (2025), Application Paper on the supervision of climate-related risks in the insurance sector, International Association of Insurance Supervisors.

[177] IAIS (2023), Issues Paper on Insurance Sector Operational Resilience, International Association of Insurance Supervisors.

[83] IAIS (2023), Issues Paper on roles and functioning of policyholder protection schemes (PPSs), International Association of Insurance Supervisors.

[91] IMF (2024), World Economic Outlook Database, October 2024, International Monetary Fund.

[146] Indonesia Re (2025), History of BPPDAN, Indonesia Re.

[107] Ingram, A. (2025), Is Bancassurance Poised to Lead the SME Commercial Insurance Market?, Sapiens.

[175] Insurance Asia (2025), “Micro insurance market to grow 6.5% annually until 2032”, Insurance Asia.

[117] Insurance Asia (2025), “Microinsurance expands to cover 344m globally in 2024”, Insurance Asia.

[77] Insurance Association of Pakistan (n.d.), About IAP, Insurance Association of Pakistan.

[174] Insurance Association of Pakistan (n.d.), Insurance & Takaful Day, Insurance Association of Pakistan.

[74] Insurance Board of Sri Lanka (2019), Know more about Insurance, Insurance Regulatory Commission of Sri Lanka.

[122] Insurance Commission (2025), Guidelines on Inclusive Micro Products, Insurance Commission.

[133] Insurance Commission (2024), Circular Letter No. 2024-13 | Consolidated Guidelines for Takaful Window Operation, Insurance Commission.

[142] Insurance Commission (2024), Reinstatement of Circular Letter No. 2016-55 on the “Strict Implementation of the Minimum Rates for Earthquake, Typhoon and Flood Covers and Related Guidelines” and Participation in the Philippine Catastrophe Insurance Facility (PCIF), Insurance Commission.

[53] Insurance Commission (2023), Annual Report 2022, Insurance Commission.

[141] Insurance Commission (2022), Guidelines on the Adoption of the Revised Schedule of Minimum Catastrophe Rates, Insurance Commission.

[52] Insurance Commission (2021), Annual Report 2020, Insurance Commission.

[137] Insurance Commission (2020), Dissemination of Information on the Georisk Philippines Initiative as an Underwriting Tool for Non-LIfe Insurance Companies, Insurance Commission.

[51] Insurance Commission (2019), Annual Report 2018, Insurance Commission.

[50] Insurance Commission (2017), Annual Report 2016, Insurance Commission.

[49] Insurance Commission (2015), Annual Report 2014, Insurance Commission.

[121] Insurance Commission (n.d.), A Glimpse of Microinsurance in the Philippines (Primer).

[44] Insurance Development Forum (2025), CatRiskTools, Insurance Development Forum.

[23] Insurance Journal (2024), “Survey: Most Homeowners Believe Their Policy Covers Flood”, Insurance Journal.

[59] IRCSL (2024), Industry Handbook 2023, Insurance Regulatory Commission of Sri Lanka.

[58] IRCSL (2022), Statistical Review 2021, Insurance Regulatory Commission of Sri Lanka.

[57] IRCSL (2020), Statistical Review 2019, Insurance Regulatory Commission of Sri Lanka.

[56] IRCSL (2018), Statistical Review 2017, Insurance Regulatory Commission of Sri Lanka.

[26] IRCSL and University of Colombo (2024), Public Confidence in Insurance in Sri Lanka.

[138] ISM (2020), About Us, Insurance Services Malaysia Berhad.

[115] Ismail, F., N. Sivarajah and Q. Chow (2020), Asia Microinsurance Supply-side Study, MicroInsurance Centre at Milliman.

[112] ITMinds Limited (2025), Why Emlaak Financials, Emlaak Financials.

[73] Jama Punji (2017), Insurance - Your Protection Against Risk: A Concise Guide, Securities and Exchange Commission of Pakistan.

[101] Kahn, C., A. Panjwani and J. Santos (2024), “Insurance, Weather, and Financial Stability”, Finance and Economics Discussion Series, No. 2024-067, Board of Governors of the Federal Reserve System.

[96] Kauko, K. et al. (2021), “Increased flood risks caused by climate change will also affect banks”, Bulletin Articles on the Economy, Bank of Finland.

[22] Khairunisa, D. et al. (2023), “Public perception of flood risks in Klang Valley, Malaysia: A case study”, Sustainable Social Development, Vol. 1/2, p. 2197, https://doi.org/10.54517/SSD.V1I2.2197.

[156] Khan, M., F. Johar and N. Baba (2017), “Disaster management risk perception of local communities”, Jurnal Kemanusiaan, Vol. 26/1.

[128] Lema, R. (2025), 2025 Development in New York Insurance Law, Phillips Lytle LLP.

[104] McKinsey & Company (Insurance Practice) (2024), Global Insurance Report 2025: The pursuit of growth, McKinsey & Company.

[105] McKinsey & Company (Insurance Practice) (2021), 2020 Global Insurance Pools statistics and trends: Distribution, McKinsey & Company.

[114] Merry, A. and J. Rozo Calderon (2025), The Landscape of Microinsurance: 2024, Microinsurance Network.

[118] Merry, A. and J. Rozo Calderon (2023), The Landscape of Microinsurance: 2022, Microinsurance Network.

[172] Moody’s (2025), Climate-conditioned catastrophe models, Moody’s.

[37] Moody’s (2025), Cyclone, Hurricane & Typhoon Models, Moody’s.

[36] Moody’s (2025), Earthquake Models, Moody’s.

[38] Moody’s (2025), Global Flood Risk Models, Moody’s.

[39] Moody’s (2025), Severe Convective Storm Models, Moody’s.

[33] Mulya, A. (2023), Basic Salary and Average Salary in Indonesia: 2023 Data, Invest in Asia.

[145] NAIC (n.d.), Catastrophe Modeling Center of Excellence, National Association of Insurance Supervisors.

[97] Newman, R., J. Adams-Kane and K. Nicholls (2023), “2022 Flood Risk Assessment for Residential Mortgages”, RBNZ Bulletin, No. 86-2, Reserve Bank of New Zealand.

[134] Norton Rose Fulbright (2024), Insurance regulation in Asia Pacific 2024, Norton Rose Fulbright.

[48] OECD (2025), Insurance business written by class of insurance, OECD Data Explorer, http://data-explorer.oecd.org/s/2ac.

[81] OECD (2024), Consumer Finance Risk Monitor, OECD Publishing, Paris, https://doi.org/10.1787/047b2ea6-en.

[143] OECD (2023), “Enhancing the insurance sector’s contribution to climate adaptation”, Business and Finance Policy Papers, Vol. 26.

[125] OECD (2023), “Leveraging technology in insurance to enhance risk assessment and policyholder risk reduction”, OECD Business and Finance Policy Papers, No. 38, OECD, Paris, https://doi.org/10.1787/2f5c18ac-en.

[150] OECD (2021), Enhancing Financial Protection Against Catastrophe Risks: The Role of Catastrophe Risk Insurance Programmes, OECD Publishing, Paris, https://www.oecd.org/finance/insurance/Enhancing-financial-protection-against-catastrophe-risks.htm (accessed on 27 July 2022).

[148] OECD (2020), Leveraging the Role of Property Catastrophe Reinsurance Markets: The Case of India, Indonesia, Myanmar, and the Philippines, OECD, https://www.oecd.org/finance/insurance/leveraging-the-role-of-property-catastrophe-reinsurance-markets.htm (accessed on 15 July 2021).

[67] OECD (2020), Recommendation of the Council on Financial Literacy, OECD, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0461.

[109] OECD (2020), Regulatory and Supervisory Framework for Insurance Intermediation, OECD, https://www.oecd.org/content/dam/oecd/en/publications/reports/2020/03/regulatory-and-supervisory-framework-for-insurance-intermediation_e78a9ec5/19915a09-en.pdf.

[45] OECD (2018), The Contribution of Reinsurance Markets to Managing Catastrophe Risk, OECD, http://www.oecd.org/finance/the-contribution-of-reinsurance-markets-to-managing-catastrophe-risk.pdf (accessed on 23 January 2019).

[84] OECD (2013), “Policyholder Protection Schemes: Selected Considerations”, OECD Working Papers on Finance, Insurance and Private Pensions, No. 31, OECD Publishing, Paris, https://doi.org/10.1787/5k46l8sz94g0-en.

[129] OJK (2024), Insurance Statistics 2023, Otoritas Jasa Keuangan.

[25] OJK (2021), National Strategy on Indonesian Financial Literacy: 2021-2025, Otoritas Jasa Keuangan.

[158] Paddam, S. et al. (2024), Home Insurance Affordability and Home Loans at Risk, Actuaries Institute.

[31] Pakistan Bureau of Statistics (2020), Household Integrated Economic Surveys 2018-2019, Pakistan Bureau of Statistics.

[154] Petrolia, D., C. Landry and K. Coble (2013), “Risk Preferences, Risk Perceptions, and Flood Insurance”, Land Economics, Vol. 89/2, pp. 227-245, https://doi.org/10.3368/LE.89.2.227.

[32] Philippines Statistics Authority (2023), Family Income and Expenditure Survey, Philippines Statistics Authority.

[76] PIAM (2025), About us, Persatan Insurans Am Malaysia.

[85] PIDM (2025), Coverage For TIPS, Perbadanan Insurans Deposit Malaysia.

[78] PIRA (2021), About us, Philippine Insurers and Reinsurers Association.

[80] PIRA, I. (n.d.), PIRA TV, Philippine Insurers and Reinsurers Association.

[95] Poeschl, J. (2022), “Do banks manage their flood risk exposures? Evidence from the Danish credit register”, Economic Memo, No. 15, Danmarks Nationalbank.

[90] Pratiwi, J. and F. Novaryani (2022), Indonesia’s Policy Guarantee to Restore Consumer Confidence in Insurers.

[93] Property Rights Alliance (2024), 2024 International Property Rights Index: Countries, Property Rights Alliance.

[155] Raman, M., A. Ojo and M. Dorasamy (2015), “A Stakeholder Perspective In Managing Floods In Malaysia”, WIT Transactions on The Built Environment, Vol. 168, pp. 1171-1181, https://doi.org/10.2495/SD151012.

[12] Ren, J. and H. Wang (2016), “Rural Homeowners’ Willingness to Buy Flood Insurance”, Emerging Markets Finance and Trade, Vol. 52/5, pp. 1156-1166, https://doi.org/10.1080/1540496X.2015.1134867.

[6] Reynaud, A., M. Nguyen and C. Aubert (2018), “Is there a demand for flood insurance in Vietnam? Results from a choice experiment”, Environmental Economics and Policy Studies, Vol. 20/3, pp. 593-617, https://doi.org/10.1007/S10018-017-0207-4/TABLES/7.

[42] Risk Frontiers (n.d.), Natural Catastrophe Modelling, Risk Frontiers.

[7] Royal, A. and M. Walls (2019), “Flood Risk Perceptions and Insurance Choice: Do Decisions in the Floodplain Reflect Overoptimism?”, Risk Analysis, Vol. 39/5, pp. 1088-1104, https://doi.org/10.1111/RISA.13240.

[176] Sands, L. (2022), “Pakistan floods: One third of country is under water - minister”, BBC, https://www.bbc.com/news/world-europe-62712301.

[86] SDIC (2018), Scope of PPF Scheme Coverage, Singapore Deposit Insurance Corporation.

[108] SECP (2024), Insured Pakistan through Digitalization: Gaps, Potential & Roadmap, Securities and Exchange Commission of Pakistan.

[136] SECP (2024), Insured Pakistan: Pool Dynamics, Securities and Exchange Commission of Pakistan.

[103] SECP (2024), Journey to an Insured Pakistan: Five-Year Strategic Plan, Securities and Exchange Commission of Pakistan.

[168] SECP (2024), Reinsurance Performance Review 2023, Securities and Exchange Commission of Pakistan.

[55] SECP (2024), Report on Insurance Industry Statistics, 2023, Securities and Exchange Commission of Pakistan.

[116] SECP (2024), Unlocking the Potential of Micro and Inclusive Insurance in Pakistan, Securities and Exchange Commission of Pakistan.

[169] SECP (2022), Concept Paper: Proposed Risk Based Capital Regime for Pakistan’s Insurance Sector, Securities and Exchange Commission of Pakistan.

[54] SECP (2022), Insurance Industry Statistics, 2021, Securities and Exchange Commission of Pakistan.

[8] Seifert, I. et al. (2013), “Influence of flood risk characteristics on flood insurance demand: a comparison between Germany and the Netherlands”, Natural Hazards and Earth System Sciences, Vol. 13/7, pp. 1691-1705, https://doi.org/10.5194/NHESS-13-1691-2013.

[18] Shah, S., I. Rana and A. Ali (2023), “Multi-hazard risk perception: An empirical study of the remote Gilgit city, Hindukush-Karakorum-Himalayan (HKH) region, Pakistan”, International Journal of Disaster Risk Reduction, Vol. 96, p. 103996, https://doi.org/10.1016/J.IJDRR.2023.103996.

[1] Shaw, R. et al. (2023), “Asia”, in Climate Change 2022 – Impacts, Adaptation and Vulnerability, Cambridge University Press, https://doi.org/10.1017/9781009325844.012.

[29] Souiden, N. and Y. Jabeur (2015), “The impact of Islamic beliefs on consumers’ attitudes and purchase intentions of life insurance”, International Journal of Bank Marketing, Vol. 33/4.

[68] State Bank of Pakistan (2025), National Financial Inclusion Strategy: 2024-28.

[24] Swiss Re Institute (2024), World insurance: strengthening global resilience with a new lease of life, Swiss Re.

[173] Tennyson, S. (2011), “Consumers’ Insurance Literacy”, No. 2011-PB-06, Networks Financial Institute.

[66] Uddin, M. (2017), “Microinsurance in India: Insurance literacy and demand”, Business and Economic Horizons, Vol. 13/2, pp. 182-191, https://doi.org/10.15208/beh.2017.14.

[166] Valera, M., Z. Lei and J. Fong (2024), “Determinants of financial inclusion in Southeast Asia”, Applied Economics, https://doi.org/10.1080/00036846.2024.2326529.

[40] Verisk Analytics (2025), Verisk Peril Models.

[19] Vinck, P. et al. (2024), Information Sheet 3: Climate Change – Survey on Disaster Preparedness and Climate Change Perceptions in the Philippines.

[15] Viverita, V. et al. (2014), “Households Perceptions on Factors Affecting Resilience towards Natural Disasters in Indonesia”, The South East Asian Journal of Management, Vol. 8/1, p. 2, https://doi.org/10.21002/seam.v8i1.3099.

[10] Wachinger, G. et al. (2013), “The Risk Perception Paradox—Implications for Governance and Communication of Natural Hazards”, Risk Analysis, Vol. 33/6, pp. 1049-1065, https://doi.org/10.1111/j.1539-6924.2012.01942.x.

[13] Wang, M. et al. (2012), “Are People Willing to Buy Natural Disaster Insurance in China? Risk Awareness, Insurance Acceptance, and Willingness to Pay”, Risk Analysis, Vol. 32/10, pp. 1717-1740, https://doi.org/10.1111/J.1539-6924.2012.01797.X.

[65] Weedige, S. et al. (2019), “Decision Making in Personal Insurance: Impact of Insurance Literacy”, Sustainability, Vol. 11/23, p. 6795, https://doi.org/10.3390/su11236795.

[17] Weerasuriya, W. and U. Rajapaksha (2023), “Effect of risk perception and vulnerability on adaptive architecture in flood resilience: a case from Ratnapura, Sri Lanka”, FARU Journal, Vol. 10/1, pp. 49-58, https://doi.org/10.4038/faruj.v10i1.183.

[47] World Bank (2025), World Bank Country and Lending Groups, World Bank.

[106] World Bank (2023), The Global Findex Database 2021, World Bank.

[21] World Bank (n.d.), Population, total.

[34] World Bank and Bank Negara Malaysia (2024), Managing Flood Risks: Leveraging Finance for Business Resilience in Malaysia, World Bank.

[164] World Population Review (2025), Home Ownership by Country 2025, World Population Review.

[5] Yang, F., J. Tan and L. Peng (2020), “The effect of risk perception on the willingness to purchase hazard insurance—A case study in the Three Gorges Reservoir region, China”, International Journal of Disaster Risk Reduction, Vol. 45, p. 101379, https://doi.org/10.1016/J.IJDRR.2019.101379.

Notes

← 1.  It focuses, in particular, on the drivers of protection gaps and solutions in Indonesia, Malaysia, Pakistan, the Philippines and Sri Lanka. It has been developed with the collaboration of insurance regulators and supervisors in these five jurisdictions and benefitted from consultations with the insurance industry.

← 2.  For example, emerging and developing Asia and Pacific includes 6 of the 10 countries and 8 of the 20 countries deemed to face the highest natural hazard risk (the Philippines, Indonesia, India, Myanmar, Bangladesh, Pakistan, Papua New Guinea and Viet Nam) (Bündnis Entwicklung Hilft and IFHV, 2024[170]) and 3 of the 10 countries and 8 of the 20 counties most vulnerable to significant impacts from climate change (the Solomon Islands, Micronesia, Tonga, the Marshall Islands, Afghanistan, Tuvalu, Kiribati and the Maldives) (Chen et al., 2025[171]).

← 3.  A number of countries in the region, including Indonesia and the Philippines, have developed strategies to leverage the financial capacity of insurance and capital markets and reduce the financial burden on public finances.

← 4.  For the purposes of this study, “emerging and developing Asia and Pacific” includes all countries in Asia and Pacific classified as low income, lower middle income or upper middle income under the World Bank’s Country and Lending Groups classification for 2025 (World Bank, 2025[47]).

← 5.  The collection and reporting of data on economic losses from natural hazards is likely to be less comprehensive in low-income countries, which could lead to an under-reporting of economic losses in those countries.

← 6.  In this ranking, exposure refers to “the nature and degree to which a system is exposed to significant climate change ....independent of socio economic context” (Chen et al., 2025[2]).

← 7.  This approach to measuring protection gaps likely overestimates the size of the gap as some types of losses are unlikely to be insured or intentionally uninsured (depending on the country). For example, in many countries, buildings and infrastructure owned by national or subnational governments are not insured by private insurance markets.

← 8.  The range reflects two approaches to measuring the share of losses insured: (i) including all available data on insured and economic losses; and (ii) including only loss years where both insured and economic loss estimates are provided.

← 9.  For countries classified as Low Income, almost all reported loss years did not include a reported insured loss.

← 10.  Based on (Cisternas et al., 2024[9]), risk awareness refers to the information or knowledge that households or businesses have about natural hazard risks and the potential for natural hazards to affect their community, whereas risk perception can be seen as referring to the outcome of the subjective evaluation by households and businesses of the likelihood that they could be impacted by a natural hazard. Risk perception is generally considered to be the primary basis for implementing preparedness measures (Cisternas et al., 2024[9]).

← 11.  Some regions of Malaysia are subject to frequent monsoon-related flooding with one estimate suggesting that 4.82 million Malaysians are affected by flooding annually (Raman, Ojo and Dorasamy, 2015[155]). A survey of individuals in the highly-populated Klang Valley found that 61% of respondents had experience with a flooding, including 28% that had experience with a flood in recent years and 7% that had experienced flooding more than three times (Khairunisa et al., 2023[22]).

← 12.  A significant portion of the country (75%) is located within the Indus river basin which has faced at least 14 major floods since 1947 (Abbas et al., 2015[157]) including the 2022 floods that reportedly submerged one third of the country (Sands, 2022[176]).

← 13.  The surveys were undertaken in eight European countries. A survey in France, Greece, Italy and Sweden found that an expectation of government compensation for losses was the main reason for not acquiring insurance for 4.8%, 4.9%, 9.6% and 14.7% of uninsured households (the share attributed to each country was not disclosed). Another survey in Belgium, Romania, Spain and Germany found that an expectation of government compensation for losses was the main reason for not acquiring insurance among 6% of respondents overall (EIOPA, 2024[4]).

← 14.  For the purposes of this limited review, a measure of “affordability stress” developed by the Actuaries Institute in Australia has been applied. The Actuaries Institute identifies households that face property insurance premium costs above four weeks of gross household income as affordability-stressed (Paddam et al., 2024[158]). However, discretionary income (i.e. income remaining after accounting for cost-of-living expenses) is likely to be much higher in Australia and therefore the threshold for defining “affordability stress” should likely be lower than four weeks of gross household income in lower income countries.

← 15.  Online offers of coverage are not found for household property insurance in Sri Lanka.

← 16.  An assessment of natural hazard risk management practices is beyond the scope of this report.

← 17.  For example, one of the major commercial providers of catastrophe models advertises the availability of climate-conditioned catastrophe models for a number of perils and regions, including for cyclones in the North Atlantic and Japan, flooding in Europe, Japan and the United States, windstorms in Europe and wildfires in the United States (Moody’s, 2025[172]).

← 18.  Catastrophe bonds are structured debt instruments that are used to fund reinsurance protection provided through a special-purpose entity that issues the bonds to investors (OECD, 2018[45]).

← 19.  Collateralised reinsurance and sidecars are special-purpose reinsurance coverage providers that are fully funded by investors. Industry loss warranties are a type of reinsurance or derivative (option) contract that pays out when aggregate insurance industry losses from a catastrophic event exceed a pre-determined threshold (OECD, 2018[45]).

← 20.  In proportional reinsurance, the premiums (and any claims) are shared proportionally and are therefore based on the pricing applied by the primary insurer. As a result, reinsurers that provide coverage on a proportional basis will seek to ensure that the pricing is adequate in order to avoid facing losses greater than premiums collected. If pricing is deemed to be inadequate, reinsurers are likely to decline coverage or propose coverage on a non-proportional basis.

← 21.  For example, the Pakistan Reinsurance Corporation retroceded over 70% of the risk that they assumed in 2023 (SECP, 2024[168]).

← 22.  Average based on data from (Helgi Library, 2025[159]) for available countries (99), classified based on World Bank’s classification of countries by income (World Bank, 2025[47]). The dataset did not include estimates for Malaysia or Sri Lanka in 2021, so estimates for 2017 were used to calculate the average. Across high-income countries, approximately 23% of households own their home and have a mortgage (Everett-Allen, 2022[160]).

← 23.  Based on (Helgi Library, 2025[159]). The dataset did not include estimates for Malaysia and Sri Lanka in 2021, so estimates for 2017 were used.

← 24.  OECD calculations based on Helgi Library (2025[159]; 2025[161]; 2025[162]; 2025[163]).

← 25.  OECD calculations based on (World Population Review, 2025[164]).

← 26.  For example, in Malaysia and the Philippines, in particular, insurers reported substantial efforts to enhance the take-up of insurance coverage for natural hazard risks, partly driven by a need for increasing revenues to meet increased capital requirements in the case of the Philippines.

← 27.  In the case of business policyholders, the difference was substantially lower although remained significant.

← 28.  For example, requirements to offer earthquake coverage as part of residential property insurance coverage in California (United States) after the Northridge earthquake in 1994 led some insurers to withdraw from the property insurance market altogether.

← 29.  Some policyholders are required to purchase insurance in many countries (including Malaysia and the Philippines), such as households that reside in multi-tenant buildings given the impact that uninsured losses could have on other residents. In the Philippines, commercial establishments in high-risk zones are also required to acquire insurance against natural hazards in order to receive a business permit.

← 30.  Although some countries (e.g. Switzerland) have a legislative requirement to include coverage for natural hazards (or some natural hazards) in property insurance.

← 31.  An OECD review of regulatory and supervisory frameworks for insurance intermediaries found that only a few countries have restrictions on the distribution of some products through specific distribution channels, such as restrictions on the distribution of personal products by independent agents in Chile, Colombia, Israel, Latvia and Portugal and by banks in Colombia, Israel, Latvia and Portugal (OECD, 2020[109]).

← 32.  Financial exclusion is driven by a number of factors, including income and education as well as access to local financial services providers (among other factors) (Ayyagari and Beck, 2015[165]; Valera, Lei and Fong, 2024[166]; Hoque, Hossain and Siddique, 2023[167]).

← 33.  The Microinsurance Network’s annual report on the microinsurance landscape reported a 4% increase in individuals covered by microinsurance between 2023 and 2024 (to 344 million) and a 6.8% increase in premiums collected (Merry and Rozo Calderon, 2025[114]).

← 34.  One recent microinsurance market analysis projected that the microinsurance market will grow by 6.5% annually between 2024 and 2032 (reported in (Insurance Asia, 2025[175])).

← 35.  One recent report estimated that the size of the global parametric insurance market was USD 16.2 billion in 2024 and is projected to grow at an annual rate of over 12% between 2025 and 2034 (Global Market Insights, 2025[124]).

← 36.  Wholly owned government agencies, such as the National Insurance Trust Fund, would not normally have an independent credit rating.

← 37.  In 2018, the OECD published an analysis of the role of international reinsurance markets in supporting post-disaster economic recovery. Using a unique set of data on property reinsurance premiums and claims provided by reinsurance companies, the OECD’s report examined the impact of reinsurance on reducing the economic disruption in the aftermath of 26 major natural catastrophes (or series of natural catastrophes) that occurred between 2010 and 2016. The report found that, in countries where a relatively high share (10% or more) of economic losses related to the specific event(s) were reinsured, the post-event recovery occurred more quickly and these countries had higher than projected GDP growth in the following three years – while those countries with lower levels of reinsurance coverage struggled to recover and faced a cumulative loss in output relative to pre-event projections (OECD, 2018[45]).

← 38.  A number of countries also have risk sharing arrangements between governments and the insurance sector for agricultural risks. These programmes have not been included in the scope of this report.

← 39.  In Switzerland, there are two main approaches to providing property insurance that includes coverage for natural hazard risks. In some cantons, there are Public Insurance Companies for Real Estate that provide all property insurance to households and businesses and collaborate through a collective reinsurance arrangement. In other cantons, property insurance coverage is provided by private insurers who have established a co-insurance pool.

← 40.  In the United States, there is a National Flood Insurance Program that provides insurance for flood risks as well as a government-owned insurance company in the US state of California for earthquake risk. The Florida Hurricane Catastrophe Fund is a government agency that provides a reinsurance coverage to insurers in the US state of Florida. In addition, many US states have residual insurance programmes that provide property insurance for households (and sometimes businesses) that are unable to access insurance from private insurers. Some of the residual programmes (e.g. Florida, Louisiana) have been established as government-owned insurers.

← 41.  There are a number of public-private insurance programmes for agricultural risk (e.g. Philippines Crop Insurance Corporation, Crop Insurance Scheme in Sri Lanka) and for protecting public finances (e.g. Pooling Fund Bencana in Indonesia, parametric insurance for emergency funds and public asset insurance provided by the Government Service Insurance System in the Philippines) although these types of programmes have been excluded from the scope of this report.

← 42.  Maipark recently assumed an additional role in providing reinsurance coverage for government-owned assets.

← 43.  Insurance companies have been encouraged by the Insurance Commission to join the PCIF although membership is voluntary. At the time of writing, less than 50% of the non-life insurance companies in the Philippines have committed to join the PCIF (and some larger insurance companies have not joined the PCIF).