Search Presentations
STAY TUNED! If you are anticipating additional search filters by attribute and level to align with the CAS Capability Model, it is coming later this Summer. As the CAS begins to code recorded sessions by specific attributes and levels (starting with the 2023 Annual Meeting), these will be tagged in the CAS database of presentations going forward and should be searchable.
But you may use the Capability Model now to help you identify topics. For example, if you want to move up one level under the content area “Functional Expertise,” you may search topics in the particular functional area to expand your knowledge.
Recorded content is searchable by Capability Model attribute and level in the CAS Online Library.
Refine Your Search
The presentation materials are offered in connection with CAS professional education offerings. © 2022 Casualty Actuarial Society. All Rights Reserved. The presentation materials may contain copyrighted content the use of which has not been specifically authorized by the copyright owner. You are permitted to view and print the materials for personal/professional noncommercial research purposes. Except for the foregoing, you agree not to reproduce, distribute, modify, create derivative works, or commercially exploit the presentation materials without prior written permission from CAS. Please direct any copyright permission inquiries regarding use of the presentation materials to acs@casact.org.
STAY TUNED! If you are anticipating additional search filters by attribute and level to align with the CAS Capability Model, it is coming later this Summer. As the CAS begins to code recorded sessions by specific attributes and levels (starting with the 2023 Annual Meeting), these will be tagged in the CAS database of presentations going forward and should be searchable.
But you may use the Capability Model now to help you identify topics. For example, if you want to move up one level under the content area “Functional Expertise,” you may search topics in the particular functional area to expand your knowledge.
Recorded content is searchable by Capability Model attribute and level in the CAS Online Library.
Predictive Modeling at the Individual Policy Level/Principle Component Analysis and Partial Least Square - Two Dimension Reduction Techniques for Regression
This session will discuss how to predict the loss cost for a single policy. The presenter will argue that separately modeling frequency and severity will produce better results than modeling the pure premium in a single step. The session will discuss: (1) what kind of GLM’s are effective for frequency and severity modeling; (2) special problems like partial year exposures; (3) partial residual plots to diagnose the effectiveness of individual variables, or groups of variables; (4) graphical plots that indicate the effectiveness of the overall model and (5) statistical measures of lift such as the Gini index and the Value of Lift. The session will illustrate its methodology with programs written in R, and will be based on a realistic dataset. Both the code and the data can be made available for users to download.
Dimension reduction is one of the major tasks for multivariate analysis, it is especially critical for multivariate regressions in many P&C insurance related applications. In this paper, we’ll present two methodologies, Principle Component Analysis (PCA) and Partial Least Squares (PLC), for dimension reduction in a case that the independent variables used in a regression are highly correlated. PCA, as a dimension reduction methodology, is applied without the consideration of the correlation between the dependent variable and the independent variables, while PLS is applied based on the correlation. Therefore, we call PCA as an unsupervised dimension reduction methodology, and call PLS as a supervised dimension reduction methodology. We’ll describe the algorithms of PCA and PLS, and compare their performances in multivariate regressions using simulated data.
Source:
2008 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Glenn Meyers, Jun Yan, Saikat Maitra
Commercial Lines Predictive Modeling - BOP
In an attempt to replicate the successful applications in the personal lines industry, the commercial lines industry is speeding up its adoption of predictive modeling. This session will review the current predictive modeling development for small commercial risks. The session will also discuss the underwriting and pricing challenges for small commercial risks and how predictive modeling, such as scoring models, can address these needs. The efforts and challenges involved in building predictive models will be described, including data issues and analysis of models.
Source:
2008 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Robert Walling, Mark Florenz
Predictive Model Development & Implementation - A Commercial Business Perspective
Predictive modeling has generated a great deal of discussion and support in actuarial circles as an improvement over historical pricing and underwriting processes. While model development techniques continue to expand and accelerate across the P&C industry, one area often overlooked is the challenges encountered by the business users of these powerful models.
This presentation will focus on the challenges and opportunities that the business side encounters when building and implementing predictive models. With both underwriting and actuarial points of view represented, the session will provide a balanced perspective on the decisions that need to be considered to most effectively roll-out and leverage the full value of predictive models.
Source:
2008 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Dana Frantz, Raymond Stukel, Mike Nelson
GLM I: Introduction to GLMs
Do terms such as "link function", "exponential family", and "deviance" intimidate you? If so, this session will help demystify Generalized Linear Models. This session will provide a basic introduction to linear and generalized linear models. It is targeted at those who have modest experience with statistics or modeling.
The session will start with a brief review of traditional linear models, particularly regression, that has been taught and widely applied for decades. The session leaders will explain how generalized linear models naturally arise as some of the restrictive assumptions of linear regression are relaxed. GLMs can model a wide range of phenomena including frequencies, severities and the probability that a claim is fraudulent or abusive, to name just a few. The session will emphasize intuition and insight rather than mathematical calculations. Illustrations will be presented using actual insurance data.
Source:
2008 Fall SIS- Predictive Modeling
Type:
concurrent
Moderators:
Abbe Bensimon
Panelists:
Louise Francis, Richard Derrig
Predictive Modeling Applications for Claims
There are numerous applications of predictive modeling to the claims function. Emphasis will be placed understanding how different techniques can be applied to claims functions of the personal and commercial lines of insurance. Some applications to be presented include:
* estimating claim settlement values,
* estimating the impact of law changes on claim values,
* developing claims early warning systems,
* identifying potentially fraudulent claims and
* managing the claims process
We will discuss the applications of techniques to claims beyond Generalized Linear Modeling, including neural networks, decision trees, and ensemble models.
Source:
2008 Fall SIS- Predictive Modeling
Type:
concurrent
Moderators:
Abbe Bensimon
Panelists:
Roosevelt Mosley
Communicating Predictive Modeling Results
Predictive Modeling is highly technical work. The successful implementation of a predictive modeling project often relies on communicating project results to a less technical audience. Graphical presentation of results is thus a key communication tool for predictive modeling work. In this session, the presenters will draw on their experience from a variety of predictive modeling projects in order to demonstrate a number of graphical presentation methodologies that they have found critical for proper presentation of model results. This will include techniques to understand key aspects of the data, identify and analyze predictor variables and summarize key model results to senior management. Selected elements of the presentation will be in case study format.
Source:
2008 Fall SIS- Predictive Modeling
Type:
concurrent
Moderators:
Abbe Bensimon
Panelists:
Louis Mak
Homeowners Predictive Modeling
Many GLM ratemaking applications have focused on private passenger auto examples. This session will discuss how the nature of some homeowners' variables affects a predictive modeling analysis. These include both traditional rating variables (such as amount of insurance, deductible and policy form) as well as external variables related to demographics or weather. The typical indivisible premium approach for analyzing homeowners' data does not lend itself well to proper investigation of these explanatory variables; therefore the presentation will outline a case for modeling homeowners separately by peril. Also, we will survey the myriad ways various companies have incorporated by peril modeling results into their rating plans, and discuss the advantages and disadvantages of various approaches.
Source:
2008 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
James Tanser, Drew Woods
General Session: How Will Predictive Modeling Change the P&C Industry over the Next 5-10 Years?
How Will Predictive Modeling Change the P/C Industry over the Next 5-10 years?
* Will there be any Property Casualty lines of business that will not be touched by Predictive Modeling?
* What areas of a company's operation (i.e. claims, reserving) besides ratemaking will be impacted by Predictive Modeling?
* Will advances come from new statistical methods, new data sources, both or other?
* Most P/C predictive modeling applications have been implemented during good economic times. Do you see any implications for these applications from the economic downturn that we are now experiencing?
* As consumers become more educated about how insurers are using Predictive Modeling to uncover new rating variables what are the risks of them being able to game the system?
* Is the ultimate scoring model, which had a unique rate for every risk, possible? Is it even desirable?
Source:
2008 Fall SIS- Predictive Modeling
Type:
general
Moderators:
Abbe Bensimon
Panelists:
Keith Holler, Glenn Meyers, Robin Harbage
Survey of Risk Factors
The panelists will address the issue of Risk Appetite. Presentations will include a Consultant survey of reinsurer and global re/insurer risk tolerance measures, a Rating Agency perspective on risk statements and risk management culture, a Broker consideration of measuring risk and compliance with corporate risk tolerance constraints, and a Reinsurer discussion of basis risk appetite with respect to corporate risk constraints.
Source:
2008 ERM for Reinsurers Limited Attendance Seminar
Type:
concurrent
Panelists:
Bruce Fell, Mark Puccia, Tim Aman, Steve White
ERM for Invested Assets
Two speakers, one from the investment community and one from the actuarial community will discuss risk management for invested assets from an ERM perspective. After a general discussion, the speakers will discuss the current credit crunch as an example of the sort of risk that can "unexpectedly show up" in an investment portfolio.
Source:
2008 ERM for Reinsurers Limited Attendance Seminar
Type:
concurrent
Panelists:
James Maher, Joshua Zwick
Emerging Risk
This session will deal with how reinsurers handle emerging risks. The session will explore considerations made both on the reserving and underwriting side, at various stages of an emerging risk's development. People often ask about the "next asbestos". This session will look at the lessons that can be learned from asbestos, discuss current issues being faced now, and explore as possible candidates potential emerging risks associated with climate change.
Source:
2008 ERM for Reinsurers Limited Attendance Seminar
Type:
concurrent
Panelists:
Jason Russ, Tanya Havlicek, Fred Gindraux
Implementing an ERM Program
This panel will focus on the implementation and structure of an ERM program within the company. Personal experiences will be shared focusing on challenges to building an ERM framework, successes to highlight and pitfalls to avoid. There will also be discussion of the unique attributes of a Reinsurer ERM program versus a Primary company ERM program.
Source:
2008 ERM for Reinsurers Limited Attendance Seminar
Type:
concurrent
Panelists:
Kevin Lehman, David Guillet
Paper Session 6
"A Model to Test for and Accommodate Reserving Cycles"
In recent years several commentators have noted evidence for a "reserving cycle" linked to the underwriting cycle. It seems that in many classes of non-life insurance, when premium rates are relatively low, claim development patterns tend to be longer-tailed than when premium rates are high. If this is the case, then traditional reserving methods based on an assumption that the development pattern is the same for all origin years will tend to overstate reserves for periods where premium rates were high, and understate reserves for periods where premium rates were low.
The present paper reviews the evidence for a reserving cycle and discusses possible causes. A mathematical model is then proposed that accommodates the main possible causes. The purpose of this model is three-fold: (a) to test for the existence of reserving cycle effects, (b) to help identify the causes, and (c) to produce improved reserve estimates. An example analysis is presented using the proposed model. The evidence for the existence of reserving cycles is now sufficiently strong that, in the author's opinion, it is important for reserving actuaries to be aware of the possibility of cyclical effects, to investigate evidence for such effects in any reserving exercise, and (where there is strong evidence) to adjust reserve estimates accordingly.
The model proposed in the present paper can be implemented in Excel and will often be a useful tool for these purposes.
"The Prediction Error of Bornhuetter/Ferguson"
Together with the Chain Ladder (CL) method, the Bornhuetter/Ferguson (BF) method is one of the most popular claims reserving methods. Whereas a formula for the prediction error of the CL method has been published already in 1993, there is still nothing equivalent available for the BF method. On the basis of the BF reserve formula, this paper develops a stochastic model for the BF method. From this model, a formula for the prediction error of the BF reserve estimate is derived.
Moreover, the model gives important advice on how to estimate the parameters for the BF reserve formula. E.g. it turns out that the appropriate BF development pattern is different from the CL pattern. This is a nice add-on as it makes BF a standalone reserving method which is fully independent from CL. The other parameter required for the BF reserve is the well-known initial estimate for the ultimate claims amount. Here the stochastic model clearly shows what has to be meant with ‘initial'.
In order to apply the formula for the prediction error, the actuary must assess his uncertainty about both sets of parameters, about the development pattern and about the initial ultimate claims estimates. But for both, much guidance can be drawn from the estimates themselves and from the run-off data given. Finally, a numerical example shows how the resulting prediction error compares to the one of the CL method.
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
Paper
Keywords:
Reserving Cycles
Paper Session 5
"Loss Development in Workers Compensation in the Presence of Legislative Reform"
Motivation. Legislative reforms affect loss development patterns in various ways. Some legislative innovations may affect new policy (or accident) years only, while others have diagonal effects as they affect both new and existing claims. Modeling these effects is critical for adequacy in ratemaking and reserving.
Method. Using a Bayesian state space model, workers compensation triangles are developed subject to the applicable legislative stipulations. Most importantly, this model is capable of accommodating the legislative environment as it evolves over time.
Results. The model is applied to an unidentified state, which experienced a reform cluster in the period 1990/92. The model shows how this reform cluster affects the ultimate loss and the 19th to ultimate tail factors.
Conclusions. Ultimate losses are not only dependent on the legislative environment at time of loss, but are also affected by how this legislative environment evolves over time. The statistical model is capable of quantifying the effects of such legislative changes on the loss development pattern.
Availability. The model runs in OpenBUGS 2.2.0 (http://mathstat.helsinki.fi/openbugs/) within the R (www.r project.org) package BRugs 0.3 3 (http://cran.r-project.org). OpenBUGS is administered by the Department of Mathematics and Statistics of the University of Helsinki, Finland; R is administered by the Technical University of Vienna, Austria. OpenBUGS and R are GNU projects of the Free Software Foundation and, hence, available free of charge.
"Property-Liability Insurance Loss Reserve Ranges Based on Economic Value"
A variety of methods to measure the variability of property-liability loss reserves have been developed to meet the requirements of regulators, rating agencies and management. These methods focus on nominal, undiscounted reserves, in line with statutory reserve requirements. Recently, though, there has been a trend to consider the fair value, or economic value, of loss reserves. Insurance regulators worldwide are starting to consider the economic value of loss reserves, which reflects how much needs to be set aside today to settle these claims, instead of focusing on nominal values. If insurers switch to economic values for loss reserves, then reserve variability would need to be calculated on this basis as well. This approach will add considerable complexity to reserve variability calculations. This paper combines loss reserve variability and economic valuation. Loss reserve ranges are calculated on a nominal and economic basis for a simplified insurer to illustrate the key variables that impact loss reserve variability. Nominal interest rate and inflation volatility, interest rate-inflation correlation, and the relationship between claim cost and general inflation are key factors that affect economic loss reserve variability. Actuaries will need to focus on measuring these values accurately if insurers adopt economic valuation of loss reserves.
"Estimating the Ultimate Liability for a non-Insurance Company's Revised Warranty Product"
Motivation. Non-insurance companies are offering ever greater enhancements to their warranty programs, many times as a competitive tool to strengthen market position. Yet, oftentimes very little analysis is performed to understand the cost of these changes. This paper discusses how warranties are accrued for on a manufacturer's balance sheet and offers examples of methods to estimate these costs.
Method. Most of the paper's discussion centers around projecting actual payments over time using an approach similar to an incremental loss development triangle approach, properly adjusted for exposure and inflation changes. Other methods discussed include Bornhuetter-Ferguson, Average Age of Warranty Claim Times Annual Spend, Active Life, and Calendar Year Payments to Revenue Approaches.
Results. The most appropriate projection method may depend on factors such as available data or the nature of the company's product.
Conclusions. Actuarial projections of warranty costs rooted in common actuarial reserving and pricing techniques are appropriate for estimating the future liabilities for the warranty liabilities.
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
Paper
Keywords:
Loss Development in Workers Compensation
Testing Loss Reserving Methods and Assumptions
The Loss Simulation Model Working Party (LSMWP) has been charged to create a simulation model of the processes of loss emergence and settlement, commonly known as loss development, that underlie the loss "triangles" and other statistics used to estimate loss reserves. The goal is to create a tool that researchers could use to generate claims that can be summarized into loss development triangles and complete rectangles which would then be used to test loss reserving methods and models. The panelists will present the prototype model which is now available to CAS members along with components of an eventual full working party report.
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
concurrent
Moderators:
Abbe Bensimon
Panelists:
Glenn Meyers, Joseph Marker, Robert Bear, Richard Vaughan
Keywords:
Testing Loss Reserving
Two Approaches to Calculating Correlated Reserve Indications across Multiple Lines of Business
Understanding and reflecting correlations is essential when looking at reserve risk at the enterprise level. The panel will discuss correlations across lines of business and within the individual accident year and payment year results within a line of business. The presentations will include a practical (spreadsheet-based) methodology for simulating correlated reserve outcomes across lines of business, an empirical study of internal correlations in over 150 runoff triangles, and demonstrations of models that identify common factors that cause correlations both within and across lines of business.
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
concurrent
Moderators:
Abbe Bensimon
Panelists:
Gerald Kirschner, Glenn Meyers
Keywords:
Multiple Lines of Business, Calculating Correlated Reserve Indications
Solvency II and UK Internal Capital Assessments
The European Union's ("EU") Solvency II project seeks to reform regulation of insurance by introducing new rigor and risk sensitivity to the quantification of regulatory capital and incentives to enhance risk management and market discipline. The impact of Solvency II will go beyond the EU member states. This session will explain the concepts and implications of Solvency II on insurance and reinsurance companies, discuss the technical specifications in the newly released Quantitative Impact Study, QIS4, and the internal models developed in the UK in response to the Internal Capital Assessment Studies.
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
concurrent
Moderators:
Abbe Bensimon
Panelists:
Robb Luck, Gavin Hill
Keywords:
Solvency II, UK Internal Capital
Six Conversations That Matter 2 of 2
Creating accountability for the well-being of the institution is a fundamental task of management. Too often, we reduce accountability through our efforts to increase it. We need to stop holding people accountable and start believing in people's capacity to choose accountability. This can be accomplished by changing our conversations. The workshop explores the ideas and tools in Peter Block's recent best-selling book, "The Answer to How is Yes". The workshop will provide participants with the tools to change each and every conversation they have to one where accountability is chosen - not enforced.
Learning outcomes:
* Understand the impact of a personal choice for accountability
* Develop skills to change conversations from patriarchy to chosen accountability
* Learn how to deal with cynics, victims and bystanders
* Learning the power of embracing doubt and reservations
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
concurrent
Moderators:
Abbe Bensimon
Panelists:
Bill Brewer
Keywords:
accountability
Raising your Actuarial IQ (Improving Information Quality)
Sarbanes-Oxley, predictive modeling, and other recent developments have renewed the focus on the quality of information. In this session, we approach data quality from the perspective of the cost of poor information quality. We then define information quality and give tips and examples on how to pursue it including how actuaries can be pro-active in improving data quality. The emphasis will be on:
1. Techniques that should be easy for most actuaries and analysts to apply right away
2. Aspects of data quality that actuaries are best able to fulfill
This session is drawn from the work of the CAS Data Management Educational Materials Working Party (Research Working Party 5).
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
concurrent
Moderators:
Ken Tan
Panelists:
Aleksey Popelyukhin, Keith Allen
Keywords:
Actuarial IQ
Paper Session 4
"A Stochastic Framework for Incremental Average Reserve Models"
Motivation: Chain ladder forecasts are notoriously volatile for immature exposure periods. The Bornhuetter-Ferguson method is one commonly-used alternative but needs a-priori estimates of ultimate losses. Berquist & Sherman presented another alternative that used claim counts as an exposure base and used trended incremental severities to "square the triangle." A significant advantage of the Berquist & Sherman method is the simultaneous estimate of underlying inflation. Though not the first to do so, this paper looks to extend the incremental severity method to a stochastic environment. Rather than using logarithmic transforms or (generalized) linear models, used in many other approaches, we use maximum likelihood estimators, bringing to bear the strength of that approach avoiding limiting assumptions necessitated when taking logarithms.
Method: Given that incremental severities can be looked at as averages over a number of claims, the law of large numbers would suggest those averages follow an approximately normal distribution. We then assume the variance of the incremental payments in a cell are proportional to a power of the mean (with the constant of proportionality and power constant over the triangle). We then use maximum likelihood estimators (MLEs) to estimate the incremental severities, along with the inherent claims inflation to "square the triangle." We also use properties of MLEs to estimate the variance-covariance matrix of the parameters, giving not only estimates of process but also of parameter uncertainty for this method. Not only do we consider the model described by Berquist & Sherman, but we also set the presentation in a more general framework that can be applied to a wide range of potential underlying models.
Results: A reasonably common, and powerful method now presented in a stochastic framework allowing for assessment of variability in the forecasts of the method. Availability. The R script for these estimates appear on the CAS web site.
"Robustifying Reserving"
Robust statistical procedures have a growing body of literature and in actuarial applications have been applied in loss severity fitting. Here an introduction of robust methods is made for loss reserving. In particular, following Tampubolon [1], reserve models for a development triangle are compared based on the sensitivity of the reserve estimates to changes in individual data points. This is then related to the generalized degrees of freedom used by the model at each point.
"Distribution and Value of Reserves Using Paid and Incurred Triangles"
Many loss reserving models are over-parameterized yet ignore calendar-year (diagonal) effects. Venter [1] illustrates techniques to deal with these problems in a regression environment. Venter [2] explores distributional approaches for the residuals. Gluck [3] shows that systematic effects can increase the reserve runoff ranges by more than would be suggested by models fitted to the triangle data alone. Quarg and Mack [4] show how to get more information into the reserve estimates by jointly using paid and incurred data.
This paper uses the basic idea and data from [4] and the methods of [1] to build simultaneous regression models of the paid and incurred data, including diagonal effects and eliminating non-significant parameters. Then alternative distributions of the residuals are compared in order to find an appropriate residual distribution. To get a runoff distribution, parameter and process uncertainty are simulated from the fitted model. The methods of Gluck [3] are then applied to recognize further effects of systematic risk.
Once the final runoff distribution is available, a possible application is estimating the market value pricing of the reserves. Here this is illustrated using probability transforms, as in Wang [5].
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
Paper
Keywords:
Stochastic Framework for Incremental Average Reserve Models
Intermediate Track II - Investigating and Detecting Change 2 of 2
This session will explore a variety of techniques to detect and address changes in mix of business, claim closing patterns, and case reserve adequacy. When changes in history are verified through discussion with claim, underwriting, reinsurance, and field staff, the actuary can pick the right tool for the job. Adjustments of loss reserve methodologies to account for each situation will also be discussed.
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
concurrent
Reserving for Mortgage Insurance and Title Insurance in a Stressed Real Estate Market
Mortgage insurance and Title insurance operating results are sensitive to economic conditions, particularly when the real estate marketplace has been stressed for several years. Carriers issuing Mortgage insurance or Title insurance policies must contend with risks that are correlated with the general direction of the economy, an imploding subprime mortgage marketplace, illiquidity in the lending community, rising delinquency rates on mortgages and trillions of dollars of adjustable rate mortgages that are scheduled to be re-set.
The speakers will summarize the coverage provisions associated with these specialty products, identify unique reserving issues and discuss the facets of the current real estate marketplace that could impact the risk of material adverse deviation.
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
concurrent
Moderators:
Mark Chaplin
Panelists:
Paul Struzzieri, Michael Schmitz
Keywords:
Mortgage Insurance, Title Insurance
Paper Session 3
"Reserving with Incomplete Exposure Information"
This paper outlines a reserving method that allows the actuary to use exposure information, such as onlevel premium, even if that information is only available for a limited number of years. The method is a simple blending of methods already in wide use, but can be shown to be based on a common underlying statistical model. The paper provides an overview of the Over-Dispersed Poisson model, and how it relates to Multiplicative LDF, Cape Cod, and Bornhuetter-Ferguson methods.
Motivation: The reserving actuary may have reliable exposure information (e.g., onlevel premium) for only a few recent years of data, rather than for the full historical period for which reserves need to be set.
Method: This incomplete exposure information can still be used, by implementing a hybrid reserving method equivalent to the Cape Cod method for the recent years and the Multiplicative LDF method for older years.
Results: We show how common reserving methods can be derived from a single statistical model, and then show how these methods are best combined when partial information is available.
Conclusions: This is a practical solution to the problem of stabilizing loss projections for recent accident years, incorporating available rate change information, and being responsive to actual loss emergence.
Corporate Governance and the Loss Reserving Process
Since the implementation at year-end 2004 of requirements under the Sarbanes-Oxley Act of 2002, many publicly traded property/casualty insurance companies have benefited from improved corporate governance surrounding the loss reserving process. However, the degree of improvement and resultant benefit has varied widely by company. While some have embraced the value of having stronger controls, others have viewed these requirements as resulting in significant additional process with only minimal benefit. The authors believe there are significant benefits to having strong corporate governance surrounding the loss reserving process. This paper defines key principles surrounding a well-controlled loss reserving process, and provides an evaluation framework to identify and prioritize opportunities for improvement. The areas addressed in this paper go beyond reserving approaches and data quality to consider the role of management, oversight by the board of directors and audit committee, documentation surrounding the reserve setting process, and financial statement disclosures.
Combined Analysis of Paid and Incurred Losses
Motivation. The new solvency regimes now emerging, insist that capital requirements align with the underlying (insurance) risks. This paper explains how a stochastic model built on basic assumptions is used to monitor insurance risk in order to get a clear insight in the aligned economic capital including prudence margins for loss reserves.
Method. The incurred loss of an insurer consists of payments on claims and reserves for claims that have been reported. As all claims are settled eventually, the cumulative paid and incurred losses for a given loss period become equal. Therefore, a joint model for the paid and incurred loss arrays is constructed, following a multivariate normal distribution, conditioned on equality of the total paid and incurred losses for a given loss period. A new class of functions is designed specifically to model development curves.
Results. A simulation experiment proved that a joint model for both paid and incurred loss arrays as described under Method, leads to a more accurate prediction of loss reserves. While the standard way of estimating percentiles for the reserve is biased, the alternative method of bootstrapping will lead to more accurate outcomes.
Conclusions. Modeling paid and incurred losses jointly leads to a considerable improvement in loss reserving in terms of accuracy of predictions, as well as specification of percentiles. Availability. This method is incorporated in software available from the authors.
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
Paper
Keywords:
Reserving with Incomplete Exposure
International Reserving Issues
As market conditions in the U.S. change, many companies have placed a growing importance on evaluating their liabilities from policies covering foreign exposures. Additionally, companies increasingly consider entering new countries for the first time as opportunities present themselves either as start-up operations or through Mergers and Acquisitions.
Reserving for international business presents unique considerations for U.S. companies who have international exposures. In particular, the last 4 to 5 years have seen the value of the U.S. dollar plummet in relation to several currencies. Actuaries who do not properly consider the impact of changing exchange rates in their operational and reserving processes will likely see considerable distortions to their data as a result.
This session will focus on several considerations related to international exposures. The panel will discuss unique reserving, accounting and cultural issues, the regulatory environment and states of the market, as well as reporting issues including requirements for Statements of Opinion. Unique foreign exposures and coverages will also be discussed.
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
concurrent
Moderators:
Mark Chaplin
Panelists:
Chandrakant Patel, Stewart Mitchell
Keywords:
Reserving
Impact of Reinsurance and Reinsurers on Your Financials
The appropriate level and amount of reinsurance can significantly impact the financial well being of an insurer. One of the main areas of focus of an insurer in evaluating its reinsurance program involves an analysis of its historical losses at various retention levels. Panelists for this session will focus on approaches taken by insurers to determine their optimal levels of reinsurance by addressing questions such as: How do you determine how much reinsurance should be purchased? What should the retention level(s) be? How often should retention levels be reviewed?
Source:
2008 Casualty Loss Reserve Seminar (CLRS)
Type:
concurrent
Moderators:
Sylvie Hulin
Panelists:
Christopher Suchar, Glenn Hiltpold
Keywords:
Reinsurance and Reinsurers
