Post July 2013 Renewal Update

Transcription

1 Catastrophe Reinsurance Post July 213 Renewal Update 1 July 213 Australian and New Zealand Catastrophe reinsurance renewals saw an additional AUD1.2 billion of vertical catastrophe reinsurance purchased by Aon Benfield clients with a market risk adjusted rate reduction of 7-8%. In contrast to the previous two years, 212/13 was a relatively benign catastrophe year for Australian insurers. Damage caused by ex-tropical Cyclone Oswald was of limited consequence to catastrophe renewals as abundant capacity worked to decrease reinsurance rates. In the twelve months to July 213: an additional AUD1.2 billion of vertical catastrophe reinsurance was purchased by Aon Benfield clients the average catastrophe reinsurance programme limit increased by 5% on an expected return period of exhaustion point basis the average catastrophe retention reduced by 7.4% on a likelihood of attachment basis, and remained stable in real dollar terms, and risk adjusted catastrophe programme rate decreases were between 5% and 15% with the market average being 7-8% In addition, many insurers looked to consolidate the changes made to their property reinsurance structures over the previous 12 month to accommodate APRA s prudential standard GPS 116 for exposure concentration. Global Market Background Global reinsurance loss activity in 212 and early 213 was mainly confined to US and, to a lesser extent, European reinsurance programmes. Even then, this loss activity was essentially only an earnings event for the reinsurance industry and had little impact on the eventual year-end total reinsurance capital pool. The lack of significant reinsurance losses from catastrophes in 212 was a reasonable departure from the building view of a new normal of higher level global catastrophes and led to a restoration of confidence in both insurers and reinsurers. Figure 1: Annual Catastrophe Insurance Losses by Region USD Billion (213) Africa Asia Europe North America (Non-U.S.) Oceania South America United States The lighter loss burden also saw the year end with a record level of reinsurer capital (as measured by the Aon Benfield Aggregate) topping USD5 million for the first time. Global reinsurance supply was therefore at an all-time high, whilst demand (measured by capacity placed) was stable in catastrophe lines and declined in nearly all non-catastrophe lines creating what is likely the widest gap between reinsurance supply and demand to date. Figure 2: Change in Global Reinsurer Capital -17% 18% $411B $342B 11% 17% -3% $47B $455B $42B 2% $55B $515B Q1 213

2 Aon Benfield Locally, loss activity in the Australasian region was minimal from a reinsurance perspective, with natural peril events less extreme following two years of above average catastrophe loss activity. For the events experienced by the reinsurance industry, it was observed that: ex-tropical Cyclone Oswald was the most significant event with losses reaching into the lower layers of a number of insurers catastrophe reinsurance programmes and was also a litmus test for insurers who were then able to demonstrate: minimal exposure in North Queensland, and/or the de-risking of their portfolios in Cyclone prone regions, and/or that losses were in line with expected exposures as reported to reinsurers whilst the Tasmanian bushfires were expansive in area and caused considerable damage, they were confined to a single-state and impacted few reinsurance programmes adverse loss development with respect to the reserving for the New Zealand earthquakes of 21 and 211 was more restrained than in previous years This positive regional loss experience combined with a recognition of: global reinsurance capital increases the strong returns experienced by the reinsurance industry, and the increasingly short-term focus on results meant that insurers expectations leading into the 1 July 213 Australasian renewals was for a moderate renewal with flat premium rate changes and flexible negotiation positioning. As the renewal progressed, however, the additional available capacity and appetite saw a better pricing outcome achieved than these original expectations would have suggested. Regulatory Changes With the full LAGIC changes in respect of the new Insurance Concentration Risk Charge (ICRC) framework being introduced as at 1 January 214, 1 July 213 was the first renewal to be directly influenced by the APRA changes to capital calculations for natural perils scenarios. Specifically, the ICRC replaced the previous Maximum Event Retention (MER) approach which centered on a 1 in 25 year single peril occurrence scenario and tended to be the anchor for determination of the amount of vertical limit purchased in the Australian market. The full implementation of these capital charge changes resulted in reinsurers re-evaluating their reinsurance structures and their approach to limit and retention setting. The implementation of the vertical requirement of the ICRC alone at 1 January 213 did not result in a significant change in the amount of vertical limit purchased at the time with the majority of insurers already having adopted a form of 1 in 25 year whole of portfolio approach to their modelling within their MER calculations. Limit purchases at 1 July 213, however, did increase above risk-adjusted levels by 5% or so as a result of increased focus on modelled output. The implementation of the horizontal requirement at 1 January 214 resulted in an increased level of interest and appetite for sideways reinsurance purchases within and below the 1 year return period level of exhaustion more particularly with respect to 3 rd and 4 th event covers at this level. It is noted, however, that many companies already had elements of protection at these levels for earnings volatility (rather than capital supplementation) purposes and so the renewal objective was to refine, enhance and optimise these protections rather than start from scratch. Programme Comparison Long Term Pricing Trends The chart below (figure 3) shows the Aon Benfield rate index comprising an analysis of Australian and New Zealand market catastrophe excess of loss placements since 23. Following significant catastrophe rate uplift in 211 and 212, 213 saw a market risk adjusted rate reduction of 7-8%. Other lines of business saw similar or larger reductions - given the additional capacity available and the corresponding pressure on rates for non-catastrophe programmes from reinsurers looking to diversify their catastrophe writings in Australia and New Zealand. Most noticeable was the additional capacity available and corresponding pressure on expiring rates for casualty programmes for reinsurers who were looking to diversify their property writings. Figure 3: Composite Australia/New Zealand Catastrophe Reinsurance Rate Index 2% 15% 1% 5% Rate Index % Under GPS 116, the capital charge (ICRC) is now determined as the greater of: vertical requirement of a 1 in 2 year scenario horizontal requirement which is the greater of three 1 in 1 year events or four 1 in 6 year events other accumulations vertical requirement

3 Core Programme Pricing 125 xs 25 years With the varied retentions and limits of insurers reinsurance programmes, comparative pricing on a total programme basis can be misinterpreted. We therefore extract the core of each programme and compare pricing at a level of 125 years in excess of 25 years. We find that core pricing has decreased from an average of 4.25% ROL in 212 to 4.16% ROL in 213. This is a decrease of approximately 2.1% and reflects the benign catastrophe experience on mid-tier catastrophe layers in Australia and New Zealand. This contrasts with the corresponding index for July 211 renewals, which saw an increase of approximately 25%. Variation around this average increase was significant, depending on each company s own loss experience, portfolio mix and geographical exposure. Figure 4: Catastrophe Programme 125 xs 25 years Pricing ROL 125 XS 25 Years 8.% 7.% 6.% 5.% 4.% 3.% 2.% 1.%.% 2.42% 2.87% 3.% Average In-force 1 January 213 (4.25%) Average In-force 1 July 213 (4.16%) 3.49% 3.58% 3.58% A B C D E F G H I J K L Source: Aon Benfield Australia Limit as a Return Period Vertical reinsurance purchase reached a new peak at the conclusion of 1 July 213 catastrophe renewals, with an additional AUD1.2 billion of vertical catastrophe reinsurance purchased by Aon Benfield clients. This increase in catastrophe cover was in excess of property exposure growth, with the average programme return period of exhaustion growing by 5% to 341 years. This increase was largely a function of heightened scrutiny by companies and boards on the non-modelled components of peak perils (for example, fire following quake, demand surge and loss adjustment expenses) within their limit setting deliberations and we expect further dialogue and consideration on these aspects in coming renewals as companies expand their understanding and interrogation of assumptions in this area. 3.99% 4.12% 4.32% 5.71% 6.2% 6.77% Figure 5: Catastrophe Programme Return Period of Exhaustion Return Period (Years) A B C D E F G H I J K L M N Source: Aon Benfield Australia 213 Average (345) 212 Average (317) Retention as a Return Period Benign loss activity leading into 1 July 213 renewals saw little pressure on programme retentions with the result that catastrophe programme retentions typically remained stable on an absolute dollar value basis. There is no shortage of market appetite for lower attritional layers and competitive pricing from these markets is leading insurers to leave retentions at expiring levels. Stability of retention levels was also assisted by an increased market appetite for equal shares across all layers. Risk adjusted, however, retention levels are returning to 21 levels with the average across the market being 3.38 years on a likelihood of attachment basis, which is a decrease of 7.4% from 212. This downward trend in risk adjusted retention rates is expected to continue for reinsurance programmes renewing in the 6 months to 1 January 214. Figure 6: Catastrophe Programme Return Period of Attachment Return Period (Years) A B C D E F G H I J K L M Source: Aon Benfield Australia 213 Average (3.38) 212 Average (3.65) In comparing catastrophe programme limits, we have chosen the annual aggregate, whole-of-portfolio losses. This method implicitly takes into account risk metrics to provide a meaningful risk adjusted comparison of the sufficiency of reinsurance.

4 Catastrophe Reinsurance Comparison to World Figure 7: Comparison of Australian Insurers Attachment & Limit Points Compared to Nominal Programmes Internationally 6 12 Average Return Period of Exhaust (Years) Average Return Period of Attach (Years) Australia/ ANZ US P&SC US CC UK All Perils Germany All Perils Italy EQ Greece EQ CEE Mixed Perils/Models Australia/ ANZ US P&SC US CC UK All Perils Germany All Perils Italy EQ Greece EQ CEE Mixed Perils/Models For contextual purposes given that the measures of risk/modelling and the complexity of perils exposing reinsurance programmes in various territories differ quite markedly we highlight a nominal comparison of limit and retention return periods for other major markets. Of note, the graphics demonstrate what has always been anecdotally understood that Australian and New Zealand programmes purchase more vertical cover above lower retentions (in terms of return period) than most other major markets. Premium to Expected Loss Multiple The number of loss affected layers impacted by 212/13 was limited. Figure 8 compares a measure of reinsurance margin - the ratio of premium to expected layer loss cost - and distributes these according to the probability of attachment of the layer. As expected, we see higher premium to expected loss multiples for higher layers (those with a lower probability of attachment). In this figure, a layer movement downward and to the right indicates a decrease in reinsurer margin with corresponding increase in the probability that the layer will experience a loss. Layers moving upward and to the right indicates a margin increase and correspond to layers that experienced a loss in the previous treaty year. Figure 8: Layer Premium to Expected Loss Multiple as a Function of Layer Attachment Probability Premium/ Expected Loss Market 213 Market % 1% 2% 3% 4% 5% 6% 7%. % 5% 1% 15% 2% 25% 3% 35% Probability of Attachment (%) Top Layers Core Working Layers

5 Catastrophe Reinsurance Comparison Basis The data in this report is taken from Aon Benfield clients whose treaty reinsurance placements incept between 1 October 212 and 1 July 213. No state government reinsurance programs or New Zealand domiciled insurers are included. Programs contained herein are a mixture of single state, niche and mid-size Australia wide markets, as well as international, residential and commercial insurers. In total these insurers: purchase over AUD15 billion in vertical catastrophe cover manage over AUD3.5 trillion of property assets in Australia and New Zealand, and cede in excess of AUD1.1 billion of traditional catastrophe excess of loss premium (this excludes proportional treaties with event limits and facultative excess of loss). Methodology and Assumptions While the reinsurance coverage provided to insurers under their catastrophe reinsurance protections is largely similar, there are at times some subtle differences. These include: Model Mix When determining catastrophe limits, the first point of reference most insurance companies rely on is output from catastrophe models. All programs in this study are benchmarked on a whole-ofportfolio multi-model approach basis. The models used are: RMS v11 for Australian Earthquake and Cylone Impact Forecasting s Other Perils model for hail, storm, bushfire and flood This choice of model mix reflects that used by insurers in determining the sufficiency of their catastrophe reinsurance purchase. Other measures that are used by insurers to determine include: Limit as a percentage of in-force aggregate or GNPI Capital at risk Loss experience and realistic disaster scenarios geographic scope global versus Australia and/or New Zealand only, as well as regional the benefit of inuring reinsurances such as proportional covers with catastrophe event limits modelling assumptions with prominent examples being underinsurance, business interruption exposure and construction More importantly, the methodology insurers undertake to measure the sufficiency of their limit can vary significantly. Notable differences include allowances for: data quality and integrity non-modelled perils demand surge It is important to note that many companies make post modeling adjustments to their target return period modeled loss to account for these and other factors.... About Aon Benfield Aon Benfield, a division of Aon plc (NYSE: AON), is the world s leading reinsurance intermediary and full-service capital advisor. We empower our clients to better understand, manage and transfer risk through innovative solutions and personalized access to all forms of global reinsurance capital across treaty, facultative and capital markets. As a trusted advocate, we deliver local reach to the world s markets, an unparalleled investment in innovative analytics, including catastrophe management, actuarial and rating agency advisory. Through our professionals expertise and experience, we advise clients in making optimal capital choices that will empower results and improve operational effectiveness for their business. With more than 8 offices in 5 countries, our worldwide client base has access to the broadest portfolio of integrated capital solutions and services. To learn how Aon Benfield helps empower results, please visit aonbenfield.com.

6 Definitions Return Period - determined on an all perils basis, based on each cedant s choice of model mix Expected Loss - determined on an all perils basis, based on each cedant s choice of model mix Premium to Expected Loss Multiple - reinsurance premium for a layer divided by the modelled expected loss to that layer Loss affected layers - any excess of loss layer with an estimated reinsurance recovery greater than zero... Aon Benfield Contacts Australia John Carroll john.carroll@aonbenfield.com Matt Botfield matt.botfield@aonbenfield.com Ryan Springall ryan.springall@aonbenfield.com Aon Tower Level Kent Street Sydney NSW 2 t f aonbenfield.com Aon Benfield, 213. All rights reserved. This document is intended for general information purposes only and should not be construed as advice or opinions on any specific facts or circumstances. This analysis is based upon information from sources we consider to be reliable, however Aon Benfield Inc. does not warrant the accuracy of the data or calculations herein. The content of this document is made available on an as is basis, without warranty of any kind. Aon Benfield Inc. disclaims any legal liability to any person or organization for loss or damage caused by or resulting from any reliance placed on that content. Members of Aon Benfield Analytics will be pleased to consult on any specific situations and to provide further information regarding the matters. 9/213