Yesterday we updated you on the PERILS AG industry loss index usage within the insurance-linked securities and ILW markets and our article shows how use of the index and underlying data has grown to include 40 insurance risk transfer transactions with $3.18 billion worth of capacity. PERILS AG published their latest newsletter yesterday and it included a really interesting look at ‘Bespoke Industry Loss Triggers’ which they have kindly agreed to allow us to republish here for our readers.
Weighted Industry Loss triggers can significantly reduce basis risk. The below article outlines how this can be achieved.
Re/Insurance is the business of risk transfer. The risk ceding party pays the risk accepting party a premium for assuming a certain risk and in return the re/insured is compensated in should a loss event occur. The re/insurance contract thereby stipulates which loss events are covered and what will trigger the loss payment. The triggers are commonly based on the actual loss of the re/insured and are known as indemnity triggers or ultimate net loss (UNL) triggers.
UNL triggers require the re/insured to disclose proprietary company information which is used by the re/insurer to assess the risk and determine the premium necessary to cover it. Such disclosure can be hugely complex for the risk ceding party. Likewise, the assessment of a large amount of complex data, in particular in the case of retrocession, can pose a major challenge for the risk assessment by the risk assuming party.
To overcome these challenges, the markets have established simpler, non-indemnity triggers. As early as the 1970s, industry-loss-triggered covers were structured in aviation re/insurance. The concept of using a market loss as a protection trigger was adopted by other sectors such as property and marine insurance. In the 1990s, with the onset of catastrophe risk transfer to capital markets, physical parameters and modelled loss were added to the trigger palette. Today, these three non-indemnity trigger types, i.e. industry loss, physical parameters and modelled loss, and combinations thereof, dominate the non-indemnity triggered risk transfer market.
Industry loss is the most common non-indemnity trigger type. It is a simple concept and readily understood, hence its appeal. The main prerequisite is an objective and independent reporting agency (a sort of referee) to determine the industry loss. For property Cat insurance in the US and Europe, the Property Claim Services (PCS) and PERILS, respectively, act as such reporting agencies. For the rest of the world, widely cited market loss publications by Munich Re (MR NatCat SERVICE) and Swiss Re (SR sigma) are often used.
The latter two sources are partly viewed as problematic in terms of the independence requirement for a reporting agency. Both Munich Re and Swiss Re act as risk takers and hence there can be an inherent conflict of interest, at least in theory (compare it to a football player acting as a referee at the same time). In public ILS transactions, also known as 144A ILS transactions after Rule 144A of the US Securities Act of 1933, industry loss estimates by Munich Re or Swiss Re are therefore not applied as triggers. The ILS market rather relies on dedicated and independent specialists such as PCS and PERILS. On the other hand, in the private over-the-counter (OTC) market with a less formalised framework, the use of Munich Re and Swiss Re industry loss estimates as triggers is nevertheless common.
All non-indemnity triggers must overcome one big drawback which is called “basis risk”. Basis risk is the term used to describe imperfect hedging. In terms of insurance risk transfer, it describes the non-perfect correlation between coverage triggered by the actual loss and coverage triggered by an index value such as a modelled loss, physical parameters or an industry loss.
Much work has been done in recent years to minimize the basis risk of non-indemnity covers. These efforts have been accentuated by the risk-based capital adequacy assessments conducted by rating agencies and new regulation such as Solvency II. Minimizing basis risk means to align the non-indemnity trigger value as closely as possible to the actual loss. As a consequence, non-indemnity covers will then perform just like the corresponding indemnity covers. In covers triggered by physical parameters or modelled loss this alignment is hard to achieve largely because in any big event, many unforeseen (or un-modelled) factors contribute to a loss. In industry loss triggered covers, such “unknown unknowns” are implicitly included and hence basis risk is lower. Moreover, if the market share of a ceded risk portfolio is known it becomes possible to weight the industry loss so that a close alignment of weighted industry loss and the actual portfolio loss can be achieved. Weighting factors can be defined by geography, such as countries, counties or CRESTA zones, or by lines of business, such a personal lines and commercial lines. This further reduces the basis risk.
Weighted industry loss triggers are now a common feature in the alternative risk transfer market. They are used in 144A ILS transactions as well as in collateralized reinsurance and Industry Loss Warranty (ILW) risk transfer in the OTC market. They mimic the performance of indemnity-based covers but at much lower disclosure requirements and at the same time higher risk transparency. In this context, it is telling that more than half of the currently placed PERILS-based capacity is using bespoke triggers, and the trend is increasing. It seems therefore that weighted industry loss covers are the non-indemnity cover of choice.