2. Alternative Risk Transfer and Insurance-Linked Securities.
Alexander Braun (University of St. Gallen).
The increasing convergence of insurance and capital markets throughout the last two decades has led to the development of substitutes for traditional reinsurance contracts which are known under the term alternative risk transfer (ART). Insurance-linked securities (ILS), the most successful of these new tools, are financial instruments that allow sponsoring (re)insurance companies to hedge systematic insurance risks such as natural disaster or mortality/longevity exposures and offer investors an attractive source of yield paired with a low return volatility and very little correlation to other asset classes such as stocks, government and corporate debt, real estate, or commodities. In 2017, the amount of alternative insurance and reinsurance capital has reached new highs and clearly became a permanent force to be reckoned with. This session could bring together the leading researchers in this modern and fast-growing area of the insurance industry.
Main objectives: Discuss and present the latest insights on the impact of alternative capital in the insurance industry. This may include the pricing and risk premiums of systematic insurance risks, the application of ART instruments beyond the traditional insurance industry, the perspective of ILS investors, the future of the insurance industry in a world with strong alternative capital capacity and other related research questions.
Optimisation of Insurance Capital and Reserve Using Insurance Linked Securities.
Dr. Madhu Acharyya (Glasgow Caledonian University),
Ahmed Abdullah Abdelkarrim & Abdullah Zarroug (University of Southampton) Madhusudan.Acharyya@gcu.ac.uk
The study proposed a model to optimise the size of catastrophe bonds within firms’ capital structure and minimize the cost of capital within the scope of Insurative Model proposed by Shimpi (2001, 2002). To do so, a linear optimisation model has been developed, considering the Solvency 2 ratio as a constraint. The linear optimisation model suggests two mixes of the capital structures, one with a size of CAT-BOND 1.24% and the other 55.34% of the capital. In addition, the study concluded to the optimum allocation of CAT-BOND adds value to insurance companies.
The Diffusion of Complex Securities: The Case of CAT bonds.
José Afonso Faias and José Guedes (Universidade Católica Portuguesa)
Complex securities generally do not diffuse smoothly but by fits and starts in response to sudden shifts in demand, occurring as investors learn about the intrinsic value of the securities from their noisy performance. We use CAT bonds, a capital market-based alternative to CAT risk reinsurance, to illustrate the diffusion of a complex security that competes against a legacy financial product offered by financial intermediaries. We find that the diffusion of the security is highly path-dependent with the capricious ups and downs of its actual performance plus the competitive response of CAT reinsurers jointly determining its ultimate success or failure.
Valuation of contingent convertible catastrophe bonds - the case for equity conversion.
Krzysztof Burneckia, Mario Nicolo Giuricich (University of Cape Town) and Zbigniew Palmowskia (Wroclaw University)
Within the context of the banking-related literature on contingent convertible bonds, we comprehensively formalise the design and features of a relatively new type of insurance-linked security, called a contingent convertible catastrophe bond (CocoCat). We begin with a discussion of its design and compare its relative merits to catastrophe bonds and catastrophe-equity puts. Subsequently, we derive analytical valuation formulae for index-linked CocoCats under the assumption of independence between natural catastrophe and financial markets risks. We model natural catastrophe losses by a time-inhomogeneous compound Poisson process, with the interest-rate process governed by the extended Vasicek model. By using an exponential change of measure on the loss process, as well as a Girsanov-like transformation to synthetically remove the correlation between the share and interest-rate processes, we obtain these analytical formulae. Using selected parameter values in line with earlier research, we empirically analyse our valuation formulae for index-linked CocoCats. An analysis of the results reveals that the CocoCat prices are most sensitive to changing interest-rates, conversion fractions and the threshold levels ultimately defining the trigger times.
Beyond semi-permeable financialisation: on ILS-based (re)insurance practices and knowledge production in capital markets environments.
Julius Kob (University of Edinburgh)
This paper investigates what prevented the space of insurance-linked securities (ILSs) to become a fully-fletched financial market-dominated arena by analysing the intertwined socio-technical development of the underlying knowledge production of natural catastrophe risk within and beyond this market, and the development of communities of producers and users of such risk knowledge practices over the last 25 years. Catastrophe modelling, the most significant ‘market device’ (c.f. Callon et al., 2007) in this space, did and does not only serve as one of the underlying risk assessment tools in (re)insurance pricing, capital requirements and solvency ratings, but also sociotechnically enabled the very ILS market itself in the first place. In the early days of ILSs, it seemed a path had opened for traditional financial markets instruments, such as futures and bonds; this would enable straightforwardly applicable financial market practice to catastrophe risk in particular, elevating it out of the traditional (re)insurance realm and onto capital markets. More specifically, Jarzabkowski et al. (Jarzabkowski, et al., 2015a, 2015b) present this development, otherwise known as the ‘convergence’ of (re)insurance and capital markets (e.g. Cummins and Weiss, 2009), as a threat, that partly replaces traditional (re)insurance practices and is accompanied by an overreliance on abstract, technocratic financial risk assessment and management. This paper, however, argues that instead of being a straightforward story of financialisation (e.g. Chiapello, 2015) and a formalisation of an economic space towards a capital markets trajectory of liquidity and exchange-mediated risk trading, the ILSs market developed in a rather different direction since its inception in the 1990s. (Re)insurance practice and culture have seeped into capital market practices just as much as financial practices have seeped into the alternative risk transfer (ART) space – a move that does not necessarily support the view of the classical ‘semi-permeable’ nature of ever-increasing influence of financial markets practices. The paper will demonstrate how communities of practice of catastrophe modelling integrated securitising catastrophe risk into (re)insurance practice largely without abandoning an understanding of ‘traditional’ risk-bearing. Also critically analysed will be the positions of incumbent interests, epistemic power struggles, and how practice and sociotechnical devices as well as prevalent regulation in a simultaneously geographically narrow (75% of securitised catastrophe risk are US perils) and geographically spread (the global nature especially of reinsurance) world of catastrophe risk are skewed towards an artificially normalised understanding of risk distribution and what consequences this bears for the catastrophe risk markets, wider capital markets, and societal catastrophe risk management. This study is based on original qualitative data obtained between 2014 and 2018. In specific, the paper draws on 49 semi-structured and in-depth interviews with (re)insurers, catastrophe modellers, ILS financial managers and engineers, SPV providers, ILS investors, brokers, and specialist consultants, as well as five site visits to firms and industry conferences, totalling about 100 hours of recordings in Edinburgh, London, New York, Boston, Chicago, 2 San Francisco, Zürich, and Bermuda.
This project is funded by the UK Economic and Social Research Council and the School of Social and Political Science of Edinburgh University.
- Callon M, Millo Y and Muniesa F (2007) Market devices. Sociological review monograph. Oxford: Blackwell.
- Chiapello E (2015) Financialisation of Valuation. Human Studies 38(1): 13–35.
- Cummins JD and Weiss MA (2009) Convergence of Insurance and Financial Markets: Hybrid and Securitized Risk-Transfer Solutions. Journal of Risk and Insurance 76(3): 493–545.
- Jarzabkowski P, Bednarek R and Spee P (2015a) Making a market for acts of God the practice of risk trading in the global reinsurance industry. First edition. Oxford: Oxford University Press.
- Jarzabkowski P, Smets M and Spee P (2015b) The value of relationships, models and face-to-face interaction in a global reinsurance market. London: Aston Business School.
Generalization of Parametric Insurance as a Solution to the Problems of Protection Gaps.
W. Jean Kwon and Joyce Lin (St. John’s University)
The claims payment obligations in nonlife insurance contracts are generally based on the indemnity principle. Under this principle, the insurer’s obligation is to restore the claimant to the actual or potential financial position prior to the covered loss event, less any applicable deductibles, coinsurance and policy limits. Accordingly, the coverage is triggered when there is an actual loss – generally of pure risk – of the subject matter of insurance. This principle fits well in most property, MAT and liability lines where the loss exposure can be objectively measured based on statistical analysis of historical or simulation data as well as in reimbursement-based health insurance plans. Traditional reinsurance contracts – proportional and excess-of-loss alike – are commonly based on the indemnity principle.
Where the risk is unique or the loss exposure cannot be objectively estimated, an insurance contract can be written on a valued policy basis under which the insurer compensates the policyholder an agreed amount upon the occurrence of a covered event. Life insurance, by design, is a valued policy for the death protection part. Travel insurance, personal accident insurance and selected health insurance plans are on a per-diem/event basis under which the insurer is obligated to pay a specified amount per day or event, regardless the actual loss sustained, as long as the coverage has been triggered.
The indemnity-principle based insurance (“indemnity insurance”) benefits the policyholder because the insurer’s obligation is up to the amount of loss. However, the policyholder bears the basis risk because the indemnification is subject to the terms, conditions and exclusions of the policy. This basis risk is known to cause non-consumption or sub-optimal consumption of (traditional) insurance. Companies offering indemnity insurance faces its own basis risk – the difference between the incurred losses and the premiums in the pool the policyholder belongs. As an attempt to reduce this basis risk, insurers attempt to best estimate their loss exposures via, among others, comprehensive underwriting and claims investigation practices. As a result, indemnity insurance coverages come with a heavy loading for underwriting and loss adjustment expenses. Holding all other factors constant (thus in theory), insurance consumption remains optimal only when the premium is actuarially fair and carries no loading. Besides, the comparatively weak transparency in risk underwriting/pricing schemes, delays claims settlements and occasional differences in loss valuation may induce lower-than-expected consumption of indemnity insurance.
The expense ratios in the U.S. nonlife insurance market have risen to around 28~30%. The operating cost of Lloyd’s also remains high at or above 40%. The insurance markets in other developed economies may exhibit similar traits. Due to these and other reasons (for example, social (religion), economic and political environmental factors), insurance consumption remains sub-optimal in many lines of insurance, especially in non-compulsory lines and in personal lines, where the consumers are exposed to a wide array of natural catastrophe and other losses. The presence of this so-called protection gap – the difference between economic losses and insured losses – calls for alternative insurance solutions. The rise in competition within the market and from alternative financing arrangements also induces insurers to find ways to effectively control their expense loadings.
Parametric insurance can be a solution. Although whether it can be fully developed as a substitute or complement to traditional indemnity insurance warrants examination, parametric insurance has been available in selected markets. (In this study, we limit the discussion about the application to ILS and other alternative risk transfer markets. Neither do we examine the impact of technology, artificial intelligence or blockchain on insurance coverage development.)
Under parametric insurance, the insurer becomes responsible for the payment of an ex-ante agreed amount – as compared to compensation of the pure loss as in indemnity insurance –once a triggering event exceeds a parametric threshold. The triggering event is often a natural disaster or man-made catastrophe but can also be a typical risk as long as we can establish a correlation between the loss data and the parameter/index of the risk. It is important to note that the policyholder must also demonstrate that he or she has actually suffered a covered economic loss.
The underwriting scheme is comparatively transparent and the claims settlement can be made fast. Because the measurement of measure the correlation is with respect to the loss frequency, insurers can reduce their expense loadings and the coverages can be affordable. The insurers also benefit from an immediate release of capital holding, thus the more efficient use of capital, as compared to the case of capital and reserve requirements under the indemnity insurance regime. Further, a number of risks under parametric insurance are regarded as uninsurable (Smith, 2017) or what have become unavailable due to the strong presence of information asymmetry or underwriting cost issues.
Applications and discussions of parametric insurance and its derivatives are found in agricultural insurance (for example, (Molini, Keyzer, Boom, & Zant, 2007), (Lu, Ramirez, Rejesus, Knight, & Sherrick, 2008), (Ozakia, Goodwinb, & Shirota, 2008)), catastrophes (for example, (Margulescu & Margulescu, 2013), (Green, 2015)) and health insurance (for example, (Prieger, 2002)). Selected insurance companies are also in this business. For example, the World Bank designs parametric insurance for earthquake risk. Swiss Re has a multiyear partnership with HWind Scientific for the analysis of real-time hurricane wind analysis for its parametric insurance solutions (c. 2014), and partners with Ping An Insurance mobile-enabled typhoon property parametric insurance (c. 2017). AXA launches parametric insurance business (c. 2017).
This study investigates parametric insurance in depth, examines the factors affecting a suitable development of market infrastructure and offers good practice guidelines for efficiency in insurer operations. With the collaboration of the Lloyd’s team, the study will attempt to cover the following areas of research requirement set by Lloyd’s:
- Examination of the current market structure and characteristics of parametric products and their derivatives, including, where feasible, case studies;
- Discussion of the pros and cons of indemnity insurance, parametric insurance and other derivative forms of insurance;
- Recommendation for parameter trigger setting and data sources/analysis;
- Recommendation for pricing, risk assessment and measurement, capital and suitability of capital structure, and other key functions of insurance operations;
- Identification of the lines of insurance that are suitable or desirable for parametric insurance, which include the areas under-served by traditional insurance operations as well as the potential for expansion of business into broader retail/personal lines;
 This is an area to be verified with additional analysis using the data available from Lloyd’s, the OECD and other sources.
HOME LIST OF ACCEPTED SESSIONS