2. Catastrophe bonds in Asia and the Pacific: Recent trends and characteristics
The increasing exposure to natural hazards and the widening protection gap in both OECD and Asia and the Pacific countries have led to a growing need for innovative financial solutions to cope with the economic burden of disasters. Catastrophe (CAT) bonds are relatively new disaster risk financing tools for most countries in Asia and the Pacific. This chapter discusses characteristics of CAT bonds, their structure, as well as their advantages and disadvantages. Various trigger mechanisms of CAT bonds such as indemnity and non-indemnity are also discussed. Finally, the chapter presents recent trends in CAT bond markets worldwide, and the development of CAT bond markets in Asia and the Pacific.
Asia and the Pacific faces significant exposure to natural hazards. As this exposure increases, countries in the region are in growing need of risk management strategies to lower the economic burdens they face in the wake of disasters. Large-scale natural hazards often leave countries with insufficient funds to provide emergency relief to victims and to finance recovery efforts.
The costs associated with these outcomes of disasters vary significantly depending on the type, size, severity and frequency of disasters faced by a given country. Floods are the major threat in most of the region’s countries, but Fiji, Myanmar, the Philippines and Viet Nam, are more affected by tropical storms, while Nepal and Indonesia are more exposed to earthquakes (Figure 2.1), meaning there is no “one-size-fits-all” approach.
In general, countries in the region need to seek to broaden their financial options and to adopt innovative financial solutions to cope with increasing exposure to disasters. Among financing options, catastrophe bonds (CAT bonds) are a relatively new solution for countries in the region.
This chapter begins by discussing the characteristics of CAT bonds, including their structure and the advantages and disadvantages of various features, notably CAT bond trigger mechanisms. It then explores recent trends in CAT bond markets worldwide, before narrowing the focus to the development of CAT bond markets in Asia and the Pacific.
There has been a significant increase in the disparity between the financial losses caused by disasters and the level of insurance coverage deployed to mitigate these losses. This disaster protection gap is widening across both OECD countries and the Dynamic Asia and the Pacific1 region since exposure is growing faster than the increase in resilience worldwide. There is thus a pressing need for additional capital to support insurance and reinsurance efforts. Addressing the expanding insurance protection gap necessitates the mobilisation of additional risk capital and the implementation of appropriate financial strategies that can bridge the connection between disaster risks and capital markets.
Insurance-linked securities (ILS) emerged at the beginning of the 1990s as a potential solution for easing financial burdens posed by natural disasters. ILS can offer governments some respite in challenging situations by transferring part of the risk stemming from disasters to investors. However, recourse to ILS may entail some challenges. For instance, it may be difficult to establish an adequate premium to be paid for unexpected natural hazards. In parallel, investors may be wary of investing in this asset class given the difficulties inherent in assessing the associated risks (OECD, 2022[1]). Careful assessment of these challenges is necessary to ensure the effective implementation of ILS as a risk transfer mechanism in addressing the expanding insurance protection gap. There are four main types of ILS: collateralised reinsurance, which has the largest share in the ILS market; CAT bonds; industry-loss warranties; and sidecars. CAT bonds offer a solution for financing disaster risks as they provide countries with funds that can be released more rapidly.
Catastrophe bonds are financial instruments that use a process called securitisation to wrap natural disaster risk into a tradable format. This process is depicted in Figure 2.2. A typical transaction requires the sponsor or cedent (the entity that would like to lay off the risk) to set up a special-purpose vehicle, which acts as a facilitator to transfer the catastrophe risk from the sponsor to the investors between the two parties. The SPV (also called a special-purpose entity or single-purpose company) is a firm with the solitary goal of enabling the transaction. It has neither employees nor property and is capitalised through the CAT bonds. The SPV grants reinsurance coverage or catastrophe swap protection to the sponsor and collects the required risk capital by issuing the CAT bond to investors. During the term of the reinsurance contract between the sponsor and the SPV, the investor’s capital is held in the form of highly liquid and low-risk collateral in a trust account.
Source: Authors’ elaboration.
CAT bonds offer a coupon stream consisting of the floating interest rate (term premium) from the collateral securities and a fixed spread (risk premium) that is determined at issuance. The fixed spread represents the Rate on Line (ROL) paid under the reinsurance contract or catastrophe swap. CAT bonds carry minimal interest rate and credit risk due to their floating rate and the high quality of the collateral, yet investors may lose their principal, because it is paid out to the sponsor if a predefined trigger event occurs during the term of the bond. The payout function can be binary or proportional to an underlying trigger metric.
To determine whether a payout is due under the embedded reinsurance contract or catastrophe swap, CAT bonds utilise different trigger mechanisms. The four primary trigger types – the indemnity trigger, parametric trigger, industry-loss index trigger and modelled-loss trigger – are discussed in detail in the following section.
The simple CAT bond transaction structure depicted in Figure 2.2 could be modified in a variety of ways. A reinsurance company may act as an intermediary between the sponsor and the SPV. Under this arrangement, the reinsurer may absorb basis risk (i.e. the gap between losses to the reinsured portfolio and the recovery amount provided by the CAT bond) before retroceding to the SPV. If a covered catastrophe event occurs, the sponsor would collect reinsurance recoverable amounts based on its own insured losses, while the SPV would pay out on the basis of a particular trigger mechanism (Swiss Re, 2011[2]).
Typically, the default of a CAT bond is measured by a trigger mechanism. CAT bond trigger mechanisms vary and provide investors with different levels of transparency and basis risk (the difference between the actual losses experienced and the expected payout). Trigger mechanisms can be broadly classified into indemnity and non-indemnity. Non-indemnity triggers can be further divided into parametric (index) triggers, industry-loss triggers, modelled-loss triggers and hybrid triggers. Each of these trigger types will be examined in turn.
Indemnity triggers
The most common trigger mechanism is the indemnity trigger, which bases CAT bond payouts on the actual insurance losses experienced by the sponsor. As such, they function similarly to reinsurance. This implies that there is no basis risk for the sponsor; however, investors may face moral hazard due to information asymmetries (Lee and Yu, 2002[3]). For instance, moral hazard may arise when insurers fail to settle catastrophe losses appropriately. Furthermore, the lengthy loss verification process required with indemnity triggers renders rapid settlement impossible. Indeed, given the fact that actual losses must be observed and verified before the bond can be triggered, CAT bonds with indemnity triggers may take two to three years to pay out following a triggering event, compared to three months for CAT bonds with non-indemnity triggers (Polacek, 2018[4]). Furthermore, indemnity CAT bond investors are subject to the operational risk of the ceding company’s underwriting and claims function. Rating agencies tend to require additional stress testing due to this operational risk exposure, which may result in a lower rating (Swiss Re, 2011[2]).
Parametric triggers
Parametric triggers use the measured strength of the covered catastrophe, such as wind speed, earthquake magnitude or barometric pressure, to determine payouts. Data for this parameter can be collected at multiple reporting stations in a given geographical area and entered into a predefined formula, resulting in a parametric index trigger. Information about the event parameters is typically available shortly after the occurrence of an event, allowing transactions to be settled more rapidly than when an indemnity trigger is involved. Furthermore, a parametric trigger can reduce information asymmetry due to the higher level of transparency associated with more objective official measurement. Owing to their higher level of transparency and simplicity of use, parametric triggers have been the preferred type in less-developed countries (Michel-Kerjan et al., 2011[5]). However, imperfect correlation between actual losses and the predefined physical parameters may expose the sponsor to basis risk.
Industry-loss index triggers
Industry-loss index triggers base payouts on the aggregate losses to the entire insurance sector following a large-scale natural disaster. Loss estimates are typically derived by a third-party modeller that can provide an independent estimate of the covered losses, such as Pan-European Risk Insurance Linked Services in Europe and Property Claims Services (PCS) in the United States. With industry-loss indices, payouts are triggered when industry-wide losses from an event reach a specified threshold. If a triggering event occurs, then the CAT bond investors are liable for the percentage of the industry represented by the CAT bond sponsor’s share. This subjects the sponsor to basis risk since the claims that the sponsor must pay may not be exactly equal to its share of the industry loss. At the same time, the estimate of the covered losses provided by the third-party modeller may not be exactly equal to total industry losses (Edesess, 2014[6]). A weighted industry index could be used to customise the industry-loss index further and reduce basis risk. By applying weighted calculation factors to various subregions of the covered area, the sponsor can obtain coverage that more closely aligns with its own risk portfolio.
Modelled-loss trigger mechanisms
Modelled-loss trigger mechanisms are based on loss estimation by a modelling firm that compares the physical parameters of events to entries in the firm’s database of industry exposures. Alternatively, the CAT bond sponsor may opt for using a representative sample of its own portfolio of exposures instead of the modelling firm’s industry database when calculating the modelled losses attributable to a specific catastrophe event (Guy Carpenter and MMC, 2007[7]). The CAT bond is triggered when the modelled losses exceed a specified threshold. This type of trigger is essentially an expansion of the parametric trigger that uses a model instead of an index function. Modelled-loss triggers may be subject to model risk, a condition where the difference between modelled losses and actual losses is large (Cummins, 2008[8]).
Hybrid trigger mechanisms
CAT bonds may feature multiple triggers of different types, known as hybrid trigger mechanisms. These hybrid triggers can be further classified into two main categories. The first category uses different trigger types for different perils within a single tranche (e.g. an index trigger could be used to establish losses due to a windstorm and a parametric trigger to establish losses related to an earthquake). The second category of hybrid triggers has a complex nature and applies different trigger types in a sequential fashion when establishing losses from a covered event. Hybrid trigger mechanisms offer greater flexibility for sponsors, allowing them to use different trigger types to address different perils within a single transaction, but investors may find this approach too complex and transactions difficult to understand (Guy Carpenter and MMC, 2007[7]).
Considerations in the choice of a trigger mechanism
Indemnity triggers remain dominant in the current market, accounting for approximately 75% of total CAT bond and ILS issuance in 2023 (Figure 2.3). They remain favoured by sponsors, since they may offer a better way to fit CAT bond coverage within the sponsor’s overall reinsurance programme. Issuances in 2020 also featured industry-loss index, parametric and hybrid triggers. According to the Artemis Deal Directory, 2020 was the first year since 2017 when none of the CAT bonds issued featured modelled-loss triggers (Artemis, n.d.[9]).
Source: (Artemis, n.d.[9]), Catastrophe Bond & Insurance-linked Securities Deal Directory (database), www.artemis.bm/deal-directory/.
Choosing which trigger mechanism to use involves various elements – a typical one is a trade-off between the costs of moral hazard and potential settlement speeds on the one hand, and the basis risk on the other (see Chapter 3 for a detailed discussion). Moral hazard behaviour occurs when the cost of loss-control efforts by the sponsor (i.e. insurance company) exceeds the benefits from debt forgiveness. The CAT bond’s basis risk refers to the gap between the sponsor’s actual loss and the composite index of losses that prevents the sponsor from receiving full risk hedging. The basis risk may therefore cause insurers to default on their debt in the case of high individual loss but low index of loss (Lee and Yu, 2002[3]).
Sponsors often prefer indemnity triggers as a means of minimising basis risk, although some theoretical literature suggests that CAT bonds with indemnity triggers should incur higher risk premiums because of moral hazard and information asymmetries. Under this trigger mechanism, investors will likely desire information on the risk exposure of the sponsor’s underwriting portfolio, which may be difficult to obtain for complex commercial risks. The need to disclose confidential information on the sponsor’s policy portfolio may influence the choice of this trigger mechanism (Cummins, 2008[8]). CAT bonds with indemnity triggers also tend to have higher transaction costs because more documentation is required regarding the sponsor’s exposures and underwriting standards. Furthermore, they may take longer to settle following an event because the sponsor’s losses must be evaluated (Cummins and Weiss, 2009[10]).
While the potential moral hazard related to the use of CAT bonds with indemnity triggers has been discussed at length in the theoretical literature, empirical evidence on this topic is mixed. Götze and Gürtler (2020[11]) provide some empirical insights by assessing whether the sponsors of CAT bonds with indemnity triggers are susceptible to moral hazard. The study concludes that ex ante moral hazard is present for insurers who use CAT bonds with indemnity triggers, whereas a CAT bond with an indemnity trigger distressed by a catastrophic event does not seem to cause ex post moral hazard on the insurers’ side. Another important conclusion is that vertical loss retention has a positive effect on sponsors’ incentives to contain losses.
Non-indemnity or index triggers tend to be favoured by investors, as they minimise moral hazard. CAT bonds with index triggers may also reduce sponsors’ basis risk, although empirical evidence is ambiguous in this regard. Early work by Harrington and Niehaus (1999[12]), based on a time-series analysis of the correlation between state-specific loss ratios for a sample of US insurers and the CAT loss index compiled by the PCS, indicates that PCS-index derivatives would have provided effective hedges for many non-life insurers. In a similar vein, Cummins, Lalonde and Phillips (2004[13]) suggest that ILS based on index-linked contracts could be used effectively by sponsors in hedging catastrophe risk. In contrast, Major (1999[14]) finds that Florida-based insurers hedging with a state-wide CAT loss index are subject to substantial basis risk.
While the vast majority of the literature dedicated to the trade-off between moral hazard and basis risk has focused on a particular type of CAT bond trigger, some studies have sought to compare alternative options. A study by MacMinn and Richter (2018[15]) compares an index-triggered security and an indemnity-triggered security similar to a CAT bond within a framework in which a corporation (insurer or reinsurer) is subject to insolvency risk, and this insolvency risk creates an additional incentive problem, known as the judgement proof problem. Under the model, the corporate manager is assumed to act in the interest of shareholders, thus the judgement proof problem leads to a conflict of interest between shareholders and other stakeholders. The study analyses the incentive effects associated with securities having different trigger mechanisms. It finds that the index-triggered security dominates the indemnity-triggered one, as it reduces insolvency risk and provides the corporate manager with greater shareholder values. Conversely, the indemnity-triggered security tends to worsen the risk of insolvency, as it introduces an additional incentive problem.
In another study, Doherty and Mahul (2000[16]) investigate the trade-off between moral hazard and basis risk. They find that, due to their correlation with policyholders’ losses, parametric triggers hedge policyholders’ losses well enough to provide a reasonable substitute for an indemnity trigger. Hedging with parametric trigger instruments also preserves the incentive of policyholders to invest in loss mitigation, while hedging with indemnity-triggered instruments does not. The study also highlights that the choice among different index triggers depends on the risk preference of the insurer. For instance, a more risk-averse insurer would prefer an index trigger with more moral hazard and less basis risk. Further, the study remarks that efficiency gains may be possible through a combination of hedging tools, namely an index-linked security with an indemnity insurance policy, which can cover the basis risk of an index-triggered instrument.
Empirical evidence remains scarce on how the different instruments should be compared and combined, particularly with regard to disaster risk transfer strategies for sovereigns. A study by Trottier and Lai (2017[17]) compares hedging strategies using CAT bonds and reinsurance to find the optimal disaster reinsurance programme. Their results suggest that a strategy combining CAT bonds and reinsurance, in which small losses are covered by reinsurance and large losses are hedged through issuing CAT bonds, is optimal. The rationale is that the strategy provides higher shareholder value and lower hedging costs, which may include financial distress costs due to insolvency risk (in the case of hedging using reinsurance) and costs associated with compensating investors for their exposure to moral hazard (in the case of issuing CAT bonds).
Decisions on which trigger to use in risk transfer instruments also depend on the country’s development level and economic structure. The parametric mechanism is often considered preferable in developing countries (Michel-Kerjan et al., 2011[5]; Clyde & Co, 2018[18]). This is due to its key characteristics of speedy disbursement, the certainty of payouts and higher transparency. Instruments using the parametric trigger are increasingly offered in low-income countries as an alternative to traditional indemnity insurance. It is also possible to use the parametric mechanism to cover smaller losses where the overhead costs would be too large to make insurance cost-effective using an indemnity mechanism, and to cover non-property related losses like business interruption (Sengupta and Kousky, 2020[19]). The parametric mechanism can also be useful for microinsurance in developing countries, targeted at low-income households or microenterprises.
Further research should explore how different instruments should be implemented in developing countries, where both insurance markets and CAT bond markets are often underdeveloped. Moreover, greater transparency may be afforded by the adoption of new technologies, such as blockchain technology, both by regulators and within the financial sector. Thus, the technology costs of the various instruments should be explored, especially in developing countries.
Drivers of CAT bond premiums
The specialised literature widely discusses the relationship between the pricing of CAT bonds and bond-specific determinants (e.g. choice of trigger mechanism, insured geographical area, peril type or the available credit rating). Empirical findings are inconclusive, however. For instance, using a data set from 2002 to 2012, Gürtler, Hibbeln and Winkelvos (2014[20]) find no significant influence of the applied trigger mechanism on the premiums. Conversely, Cummins and Weiss (2009[10]) argue that indemnity-triggered CAT bonds have higher premiums compared to non-indemnity triggered CAT bonds. The rationale behind these higher premiums is the moral hazard associated with the product plus the higher transaction costs for indemnity-triggered CAT bonds, given that more documentation is needed for indemnity trigger mechanisms than non-indemnity trigger mechanisms. In addition, Galeotti, Gürtler and Winkelvos (2013[21]) assess the trigger mechanisms of CAT bonds issued between 1999 and 2009 and find that the parametric trigger mechanism has less effect over time on pricing compared to the indemnity trigger. As is the case with standard bonds, better credit ratings are found to be associated with lower CAT bond premiums (Gürtler, Hibbeln and Winkelvos, 2014[20]).
Bond-specific factors, such as the number of perils covered and time to maturity, are additional drivers of CAT bond premiums. CAT bonds can be designed to provide coverage against multiple types of events simultaneously in a single country or in multiple geographic locations. Typically, these multi-peril CAT bonds are not triggered by the first event. For instance, multi-peril CAT bonds provide coverage once a major Japanese earthquake or US windstorm has already occurred during a defined period. Thus, investors are unlikely to lose any principal until at least two major events have occurred. Multi-peril CAT bonds are typically of investment grade and particularly attract institutional investors restricted to purchasing only investment grade bonds (Woo, 2004[22]). From the sponsor’s side, multi-peril CAT bonds reduce transaction costs, as they can insure many peril types in a single bond (Guy Carpenter and MMC, 2007[7]). However, empirical evidence shows that higher deal complexity in terms of the number of insured perils tends to inflate the premiums (Gürtler, Hibbeln and Winkelvos, 2014[20]; Galeotti, Gürtler and Winkelvos, 2013[21]). This may imply that an additional risk load is imposed by the market on multi-peril CAT bonds compared to single-peril ones (Galeotti, Gürtler and Winkelvos, 2013[21]). As regards time to maturity, a recent study by Herrmann and Hibbeln (2021[23]) shows that the seasonality in the probability of CAT bonds being triggered causes strong seasonal fluctuations in spreads (e.g. the spread on a hurricane bond is highest at the start of the hurricane season and declines as time goes by without a hurricane).
Other authors have investigated the impact of financial market variables on CAT bond premiums. Since the bulk of CAT bonds are denominated in US dollars, sponsors whose national currency is not the US dollar are exposed to an exchange rate risk. This risk could be covered by introducing a currency hedging cost in the pricing of these bonds, which would therefore increase their overall premiums. An early contribution to this topic is a study by Poncet and Vaugirard (2001[24]), which shows that the currency exchange risk has a negative effect on the CAT bond price compared to the natural risk. More recently, Lai, Parcollet and Lamond (2014[25]) conclude that CAT bond prices correlate positively with the exchange rate and the foreign interest rate. Conversely, the volatility of the exchange rate and the correlation between the exchange rate and the domestic interest rate have a negative impact on the price.
Prices are also a function of the expected loss. Using a dataset comprising 250 CAT bonds issued on capital markets, Lane and Mahul (2008[26]) find that the market-based catastrophe risk price is estimated to be 2.69 times the expected loss over the long term. The relationship between loss severity and loss frequency has also been discussed quite extensively in the specialised literature. Jaimungal and Chong (2013[27]), assess the role that clustering in activity or severity plays in catastrophe modelling. They propose two marked point processes to account for these features. The first approach assumes that the points are driven by a stochastic hazard rate modulated by a Markov chain, while in the second approach the points are driven by a self-exciting process. These approaches support the hypothesis that the occurrence of a catastrophe event might increase the likelihood of the occurrence of another catastrophe event (Jaimungal and Chong, 2013[27]). Based on available data, the average coupon for CAT bonds issued in 2020 stood at 6.93%, with an average expected loss of 2.16% (Figure 2.4).
Note: Only 144A CAT bonds or similar are included. 144A CAT bonds refers to privately placed CAT bonds under Rule 144A of the US Securities Act.
Source: (Artemis, n.d.[9]), Catastrophe Bond & Insurance-linked Securities Deal Directory (database), www.artemis.bm/deal-directory/.
The CAT bond market has grown steadily since it began developing in the 1990s, with the corporate world as the main issuer over the first decade (Michel-Kerjan et al., 2011[5]). From a single transaction recorded in 1996, cumulative issuance of CAT bonds amounted to more than 560 deals in 2022 (Figure 2.5). In terms of volume, cumulative CAT bond issuance surpassed USD 12 billion in 2021. More importantly, CAT bonds exhibit little return correlation with other asset classes, making them a great portfolio diversifier (Figure 2.6).
Note: Only 144A CAT bonds or similar are included. 144A CAT bonds are privately placed CAT bonds under Rule 144A of the US Securities Act. LHS = left hand scale. RHS = right hand scale.
Source: (Artemis, n.d.[9]), Catastrophe Bond & Insurance-linked Securities Deal Directory (database), www.artemis.bm/deal-directory/.
Source: Authors’ elaboration, based on data from Bloomberg for the stock markets (S&P 500) and bond markets (Barclays Corporate Bond and Treasury Indices).
In 2020, as the COVID-19 pandemic unfolded and uncertainty hit financial markets, a secondary market sell-off hit CAT bond markets. The sell-off was largely caused by entities like multi-strategy asset managers seeking opportunities in other markets. As uncertainty related to the COVID-19 pandemic spread in financial markets, the capital position of the CAT bond market came into question frequently.
In 2022, the prevailing conditions included rising inflation, increasing interest rates and notable political and economic uncertainties. CAT bond investors remain concerned about the potential rise in claims and have demanded higher spreads to compensate for the associated risks. Indeed, the spread levels for CAT bonds have soared to heights not witnessed in more than ten years (Swiss Re, 2023[28]). The widening of spreads has had significant repercussions, including severe markdowns on bonds that were completely unaffected by any catastrophe events (Twelve Capital, 2023[29]).
Despite the challenging market conditions, some sponsors were undeterred and proceeded to issue CAT bonds, disregarding the high spread levels. This may have been in the aim of securing capacity, especially considering the extended January renewals season and the scarcity of capital (Swiss Re, 2023[28]). New issuance in the first half of 2022 stood at USD 8 billion, below the 2021 issuance of USD 12.8 billion (Figure 2.7). Nevertheless, even with a substantial amount of new issuance in the first six months of 2022, the CAT bond market measured by notional outstanding has grown by approximately 6.8% since the end of 2021. At the end of the third quarter of 2022, Hurricane Ian contributed to the increase of uncertainty in terms of trapped capital, losses and the possibility to place new issuance.
Some CAT bonds have recently been given additional features, such as links to environmental, social and governance (ESG) concerns. ESG-themed CAT bonds are an extension of the basic CAT bond model that could attract investors interested in climate change mitigation projects. Under an ESG bond framework, the issuer might agree to invest the proceeds in green projects only. An ESG CAT bond issued by the Italian insurance company Generali to cover multiple perils in North America and Europe raised EUR 200 million (Natixis, 2022[31]). The framework governs what types of projects can have green CAT bonds issued for them and how the freed-up capital can be allocated. The framework has two main approaches. The first is to use the freed-up capital benefit for green assets and underwriting. The second is to invest the proceeds contained in the special-purpose vehicle (SPV) in a portfolio of green investments for growth.
CAT bond markets remain underdeveloped in the countries of Dynamic Asia and the Pacific, with advanced economies dominating most issuance (Artemis, 2020[32]). CAT bonds have mainly been issued to cover certain named perils in Europe, Japan and the United States, while the coverage for developing countries represents a much smaller share (Figure 2.8). Among countries and regions with emerging economies, the Caribbean and Mexico are covered more frequently than others. To date, the Artemis Deal Directory lists a CAT bond issued in November 2019 to provide financial coverage to the Philippines in the event of earthquakes and tropical cyclones as the only CAT bond covering property risks among the member countries of the Association of Southeast Asian Nations (ASEAN) (Artemis, n.d.[9]).
Note: Only 144A CAT bonds or similar issued in 1996-2022 are included. “Others” includes Chile, the People’s Republic of China, Cyprus, Colombia, El Salvador, Guatemala, Israel, New Zealand, Peru, the Philippines and Chinese Taipei. Some percentages involve multiple counting; this applies to a CAT bond transaction that covers multiple countries or regions.
Source: (Artemis, n.d.[9]), Catastrophe Bond & Insurance-linked Securities Deal Directory (database), www.artemis.bm/deal-directory/.
Singapore is the most active market in ASEAN for CAT bond issuance. During the 12-month period ending on 30 June 2020, four CAT bonds were issued out of Singapore. They were issued using the ILS Grant Scheme launched by the Monetary Authority of Singapore (MAS) in February 2018 to help fund upfront ILS bond issuance costs. Three of these were sponsored by ceding insurers based in the United States to cover losses related to hurricanes there (Aon, 2020[33]).
As of April 2023, the ILS Grant Scheme offered by MAS had provided support for the issuance of 23 CAT bonds in Singapore (PMO, 2023[34]). In an effort to support the development of the market, the initiative has been extended for an additional three years and is now available to qualifying CAT bond and other ILS sponsors until the end of 2025. The extension of the scheme specifically aims to assist in covering the costs associated with issuing CAT bonds that focus on risks within the Asian region. This may suggest that the scheme’s focus is increasingly shifting towards supporting CAT bonds and ILS instruments that transfer regional risks, rather than international risks.
Within the region, Hong Kong, China is gaining momentum as an international hub for risk management. The supportive regulatory environment for ILS in Hong Kong is already benefiting (re)insurers seeking to diversify their risk portfolios. Various initiatives have been implemented, including the simplification of authorisation and regulations for special purpose insurers (SPI), as well as the introduction of a pilot ILS grant that offers up to HKD 12 million (Hong Kong dollars), or about USD 1.5 million, to cover issuance costs.
The first CAT bond domiciled in Hong Kong, China was a USD 30 million transaction sponsored by China Re Group, a state-owned reinsurer, in September 2021. However, this particular transaction did not make use of Hong Kong, China’s ILS grant scheme. The scheme was first used by Peak Re, a Hong Kong-based reinsurance company, for its inaugural CAT bond issuance in June 2022. Later in 2022, a Chinese domestic insurer, PICC Property and Casualty Company Limited, sponsored a USD 32.5 million deal to obtain earthquake reinsurance coverage in China, taking advantage of the ILS grant scheme in Hong Kong, China.
The most recent CAT bond based in Hong Kong, China was issued by the World Bank’s International Bank for Reconstruction and Development (IBRD) in March 2023. It provides Chile with USD 350 million of parametric earthquake protection. This landmark transaction marks the first-ever CAT bond to be listed on the Hong Kong Stock Exchange (HKEX) (World Bank, 2023[35]).
The development of Singapore as a hub for CAT bonds and the emergence of Hong Kong, China as an international risk management centre underscore the increasing importance of Asia in the global CAT bond landscape and the potential for further growth and innovation in the region. Countries in Dynamic Asia and the Pacific have the opportunity to take advantage of this progress. Singapore, which is a relatively new but increasingly attractive listing location, can be chosen as the domicile for countries in the region that are looking to sponsor CAT bonds. This choice is likely to be well-received, as it aligns with Singapore’s aspirations to lead the region in green finance. It is particularly relevant for ASEAN.
CAT bonds also have potential in Central Asia, where disasters have taken a significant toll over the last two decades, resulting in losses exceeding USD 1.5 billion and affecting the lives of more than 2.5 million people (World Bank, 2021[36]). The countries of Central Asia – Kazakhstan, Kyrgyz Republic, Tajikistan, Turkmenistan and Uzbekistan – are prone to floods, earthquakes and landslides. As the CAT bond market is primarily exposed to risks in the United States, Europe and Japan, diversifying into risks from Central Asia may offer attractive opportunities for investors. Multilateral development banks can serve as intermediaries between the sponsoring countries and investors, allowing a reduction of investors’ exposure to credit risk.
The projected economic impact of climate change in South Asia makes CAT bonds relevant there as well. A simulation by the Asian Development Bank (ADB) estimates that economic damage in South Asia due to climate change will average 1.8% of GDP by 2050 and could soar to around 8.8% of GDP by 2100 (ADB, 2014[37]). Without proactive efforts to adapt to and mitigate global climate change, average total economic losses could reach 9.4% for Bangladesh, 6.6% for Bhutan, 8.7% for India, 12.6% for the Maldives, 9.9% for Nepal and 6.5% for Sri Lanka. To address this challenge, (re)insurance and ILS, particularly CAT bonds, offer a robust financial tool for mitigating climate-related risks and protecting the well-being of South Asian communities.
The rise of CAT bond markets also represents an opportunity for Pacific Island countries. Many countries in this region have recently embraced contingent credit, a significant step that can allow countries to become familiar with trigger-based financing mechanisms. This may open the door to exploration of more innovative risk transfer instruments using a trigger-based approach, particularly CAT bonds. Moreover, contingent credit arrangements often involve engagement with international financial institutions. As countries navigate these relationships, they establish connections with organisations that are active in global financial markets. This network can be leveraged when considering CAT bonds, which require interaction with a broader range of financial stakeholders. As Pacific Island countries become better prepared for disasters and strengthen their ability to respond, CAT bonds offer them an option for securing financial support in the event of a catastrophe.
Prevalence and consequences of protection gaps
Protection gaps are a major issue for both OECD and emerging economies. The natural catastrophe insurance protection gap is defined as the difference between economic losses and insured losses from natural disasters (Holzheu and Turner, 2017[38]). This gap has now reached a staggering USD 368 billion, with approximately 76% of natural catastrophe exposure remaining uninsured (Evans, 2023[39]). In the United States, protection gaps are relatively high for hurricanes, floods and wildfires. The Federal Emergency Management Agency (FEMA) estimates that fewer than 50% of US homeowners have flood insurance. The situation is far worse for countries with emerging markets. Over the last decade, only 5% of flood losses and 11% of tropical cyclone losses were covered through insurance in these countries, according to estimates by Swiss Re (2022[40]). Figure 2.9 shows that insurance coverage for disasters is limited in Emerging Asia (ASEAN member countries, China, and India). For instance, from 2012 to 2018, just 0.3% of overall losses were insured in Viet Nam, 0.5% in Malaysia and 0.8% in Thailand. In OECD countries the proportion is much higher, for example 51% in Australia and 49% in New Zealand. Emerging Asia exhibits the world’s largest protection gaps: 90-100% of storm, flood and earthquake losses in the region are uninsured (Swiss Re, 2018[41]).
Source: Munich RE’s NatCatSERVICE (database), https://www.munichre.com/en/solutions/for-industry-clients/natcatservice.html.
Another way to look at this issue is to consider insurance penetration, which reflects the development of the national insurance sector. It relates the aggregate volume of insurance premiums in an economy to GDP. Figure 2.10 compares the insurance penetration rates of selected Asian countries in 2021.
Source: (Swiss Re, 2022[40]), World insurance: Inflation risks front and centre.
Dynamic Asia and the Pacific face large exposure to natural hazards, and disasters are increasing in frequency and intensity. As these changes are happening faster than the development of resilience, protection gaps are widening, and the challenge of closing them is growing. While there is no one-size-fits-all approach to disaster risk financing, catastrophe bonds (CAT bonds) are a potentially useful tool that could help sovereigns offload risk onto the private sector, reducing their burden. These instruments offer investors a coupon stream in exchange for taking the risk. The optimal choice of trigger depends on the needs and situations (e.g. speed of payout, basis risk, willingness to pay transaction costs to construct the CAT bond). Factors that affect premiums include the number of perils covered in a CAT bond, time to maturity, the degree of moral hazard in the trigger, the currency in which the bonds are denominated and the expected loss for investors, the estimation of which can be affected by the availability of robust, accurate and current data.
CAT bond markets have developed steadily in advanced economies since 1990s. Financial conditions associated with the COVID-19 pandemic slowed market growth in 2020, but it has since rebounded, and CAT bonds with additional features (such as ESG components) have become more popular. Uptake in countries in Dynamic Asia and the Pacific has been slower but there is significant room to strengthen CAT bond markets in the region.
References
[37] ADB (2014), Assessing the Costs of Climate Change and Adaptation in South Asia, https://www.adb.org/sites/default/files/publication/42811/assessing-costs-climate-change-and-adaptation-south-asia.pdf.
[33] Aon (2020), ILS Annual Report 2020, Aon Securities, http://thoughtleadership.aon.com/documents/280920_aon_securities_ils_annual_2020_update.pdf.
[32] Artemis (2020), Q4 2020 catastrophe bond & ILS market report: Busy fourth-quarter completes record catastrophe bond market year, https://www.artemis.bm/wp-content/uploads/2021/01/q4-2020-catastrophe-bond-ils-report.pdf?utm_source=ReportsPage&utm_medium=Link&utm_content=Report&utm_campaign=Q42020Report.
[9] Artemis (n.d.), Catastrophe Bond & Insurance-linked Securities Deal Directory (database), http://www.artemis.bm/deal-directory/.
[18] Clyde & Co (2018), Parametric insurance: Closing the protection gap, Clyde & Co LLP, https://www.unisdr.org/preventionweb/files/56544_clydecoresilienceparametricinsuranc.pdf.
[8] Cummins, J. (2008), “CAT bonds and other risk-linked securities: State of the market and recent developments”, Risk Management and Insurance Review, Vol. 11/1, pp. 23-47, https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.525.6812&rep=rep1&type=pdf.
[13] Cummins, J., D. Lalonde and R. Phillips (2004), “The basis risk of catastrophic-loss index securities”, Journal of Financial Economics, Vol. 71/1, pp. 77-111, https://doi.org/10.1016/s0304-405x(03)00172-7.
[10] Cummins, J. and M. Weiss (2009), “Convergence of Insurance and Financial Markets: Hybrid and Securitized Risk‐Transfer Solutions”, Journal of Risk and Insurance, Vol. 76/3, pp. 493-545, https://doi.org/10.1111/j.1539-6975.2009.01311.x.
[16] Doherty, N. and O. Mahul (2000), Mickey Mouse and moral hazard: Uninformative but correlated triggers, NBER Working Paper, National Bureau of Economic Research, Cambridge, MA, https://users.nber.org/~confer/2001/insurance01/doherty2.pdf.
[6] Edesess, M. (2014), “Catastrophe Bonds: An Important New Financial Instrument”, EDHEC Business School Working Paper July 2014, https://risk.edhec.edu/sites/risk/files/edhec-working-paper-catastrophe-bonds-an-important_1410341092999.pdf.
[39] Evans, S. (2023), Nat Cat protection gap widens to $368bn. Cat/resilience bonds needed: Swiss Re, news page, 21 June 2023, Artemis, https://www.artemis.bm/news/nat-cat-protection-gap-widens-to-368bn-cat-resilience-bonds-needed-swiss-re/.
[21] Galeotti, M., M. Gürtler and C. Winkelvos (2013), “Accuracy of premium calculation models for CAT bonds: An empirical analysis”, Journal of Risk and Insurance, Vol. 80/2, pp. 401-421, http://www.jstor.org/stable/24548156.
[11] Götze, T. and M. Gürtler (2020), “Hard markets, hard times: On the inefficiency of the CAT bond market”, Journal of Corporate Finance, Vol. 62, p. 101553, https://doi.org/10.1016/j.jcorpfin.2019.101553.
[20] Gürtler, M., M. Hibbeln and C. Winkelvos (2014), “The Impact of the Financial Crisis and Natural Catastrophes on CAT Bonds”, Journal of Risk and Insurance, Vol. 83/3, pp. 579-612, https://doi.org/10.1111/jori.12057.
[7] Guy Carpenter and MMC (2007), The Catastrophe Bond Market at Year-End 2006: Ripples Into Waves, Guy Carpenter & Company LLC and Marsh & McLennan Companies, https://user.iiasa.ac.at/~hochrain/KIT%202017%20Material/Thema%2018%20Catastrophe%20Bonds%20I.pdf.
[12] Harrington, S. and G. Niehaus (1999), “Basis Risk with PCS Catastrophe Insurance Derivative Contracts”, The Journal of Risk and Insurance, Vol. 66/1, p. 49, https://doi.org/10.2307/253877.
[23] Herrmann, M. and M. Hibbeln (2021), “Seasonality in catastrophe bonds and market‐implied catastrophe arrival frequencies”, Journal of Risk and Insurance, Vol. 88/3, pp. 785-818, https://doi.org/10.1111/jori.12335.
[38] Holzheu, T. and G. Turner (2017), “The natural catastrophe protection gap: Measurement, root causes and ways of addressing underinsurance for extreme events”, Geneva Papers on Risk and Insurance: Issues and Practice, Vol. 43/1, pp. 37-71, https://econpapers.repec.org/article/palgpprii/v_3a43_3ay_3a2018_3ai_3a1_3ad_3a10.1057_5fs41288-017-0075-y.htm.
[27] Jaimungal, S. and Y. Chong (2013), “Valuing clustering in catastrophe derivatives”, Quantitative Finance, Vol. 14/2, pp. 259-270, https://doi.org/10.1080/14697688.2013.799775.
[25] Lai, V., M. Parcollet and B. Lamond (2014), “The valuation of catastrophe bonds with exposure to currency exchange risk”, International Review of Financial Analysis, Vol. 33, pp. 243-252, https://doi.org/10.1016/j.irfa.2014.02.014.
[26] Lane, M. and O. Mahul (2008), “Catastrophe risk pricing: An empirical analysis”, Policy Research Working Paper, No. 4765, World Bank, Washington, D.C., https://openknowledge.worldbank.org/bitstream/handle/10986/6900/WPS4765.pdf?sequence=1&isAllowed=y.
[3] Lee, J. and M. Yu (2002), “Pricing Default‐Risky CAT Bonds With Moral Hazard and Basis Risk”, Journal of Risk and Insurance, Vol. 69/1, pp. 25-44, https://doi.org/10.1111/1539-6975.00003.
[15] MacMinn, R. and A. Richter (2018), “The choice of trigger in an insurance linked security: The mortality risk case”, Insurance: Mathematics and Economics, Vol. 78, pp. 174-182, https://doi.org/10.1016/j.insmatheco.2017.09.018.
[14] Major, J. (1999), “Index Hedge Performance: Insurer Market Penetration and Basis Risk”, in The Financing of Catastrophe Risk, University of Chicago Press, Chicago, IL, http://www.nber.org/system/files/chapters/c7956/c7956.pdf.
[5] Michel-Kerjan, E. et al. (2011), “Catastrophe Financing for Governments: Learning from the 2009-2012 MultiCat Program in Mexico”, OECD Working Papers on Finance, Insurance and Private Pensions, No. 9, OECD Publishing, Paris, https://doi.org/10.1787/5kgcjf7wkvhb-en.
[31] Natixis (2022), ESG cat bonds: A highly promising alternative product, Natixis Corporate and Investment Banking, https://home.cib.natixis.com/articles/esg-cat-bonds-a-highly-promising-alternative-product.
[1] OECD (2022), Economic Outlook for Southeast Asia, China and India 2022: Financing Sustainable Recovery from COVID-19, OECD Publishing, Paris, https://doi.org/10.1787/e712f278-en.
[34] PMO (2023), DPM Lawrence Wong at the official launch of Sustainable and Green Finance Institute (SGFIN), Speech, 20 April 2023, Prime Minister’s Office of Singapore, https://www.pmo.gov.sg/Newsroom/DPM-Lawrence-Wong-at-the-Official-Launch-of-the-Sustainable-and-Green-Finance-Institute.
[4] Polacek, A. (2018), Catastrophe bonds: A primer and retrospective, Federal Reserve Bank of Chicago, http://www.chicagofed.org/publications/chicago-fed-letter/2018/405#ftn15.
[24] Poncet, P. and V. Vaugirard (2001), “The valuation of nature-linked bonds with exchange rate risk”, Journal of Economics and Finance, Vol. 25/3, pp. 293-307, https://doi.org/10.1007/bf02745890.
[19] Sengupta, R. and C. Kousky (2020), Parametric insurance for disasters, Wharton Risk Management and Decision Process Center, https://esg.wharton.upenn.edu/wp-content/uploads/2023/07/Parametric-Insurance-for-Disasters_Sep-2020.pdf.
[28] Swiss Re (2023), “Insurance-linked securities market insights”, Vol. 38, March 2023, https://www.swissre.com/dam/jcr:345b8b0e-c2ea-4dad-957b-7ba20658e8e2/2023-03-sr-ils-market-insights-march.pdf.
[30] Swiss Re (2022), “Insurance-linked securities market insights”, Vol. 37, August 2022, https://www.swissre.com/dam/jcr:fb8a641a-f3bc-415f-9264-2335eb004ed7/2022-08-sr-ils-market-insights.pdf.
[40] Swiss Re (2022), World insurance: Inflation risks front and centre, https://www.swissre.com/institute/research/sigma-research/sigma-2022-04.html.
[41] Swiss Re (2018), ILS market update: August 2018, https://www.swissre.com/our-business/alternative-capital-partners/ils-market-update-aug-2018.html.
[2] Swiss Re (2011), The Fundamentals of Insurance-Linked Securities: Transforming Insurance Risk into Transparent and Tradable Capital Market Products, http://www.institutdesactuaires.com/global/gene/link.php?doc_id=871&fg=1.
[17] Trottier, D. and V. Lai (2017), “Reinsurance or CAT Bond? How to Optimally Combine Both”, The Journal of Fixed Income, Vol. 27/2, pp. 65-87, https://doi.org/10.3905/jfi.2017.27.2.065.
[29] Twelve Capital (2023), Annual Cat Bond Review and Market Outlook 2023, https://www.twelvecapital.com/wp-content/uploads/2023/01/202301_Twelve_Cat_Bond_Market_Review.pdf.
[22] Woo, G. (2004), A catastrophe bond niche: Multiple event risk, National Bureau of Economic Research, Cambridge, MA, http://www.ajbuckeconbikesail.net/wkpapers/Homeland/woo.pdf.
[35] World Bank (2023), World Bank and Hong Kong partner on a $350 million cat bond for Chile, World Bank, Washington D.C., https://www.worldbank.org/en/news/press-release/2023/03/28/world-bank-and-hong-kong-partner-on-a-350-million-cat-bond-for-chile.
[36] World Bank (2021), Celebrating risk reduction heroes in Central Asia, World Bank, Washington D.C., https://www.worldbank.org/en/news/feature/2021/10/13/celebrating-risk-reduction-heroes-in-central-asia.
Note
← 1. Dynamic Asia refers to Emerging Asia – the ASEAN-10 countries plus China and India – along with other member countries of the South Asian Association for Regional Co-operation (SAARC), the countries of Central Asia, as well as Mongolia in East Asia.