Excess Casualty Loss Data

Transcription

1 Excess Casualty Loss Data MSCAD METHODOLOGY What are the factors and protocols that Advisen uses to capture, collect, and curate $1,000,000+ Excess Casualty loss data? April 2016

2 TABLE OF CONTENTS I. Introduction 3 II. Overview 3 III. Casualty Case Data Sources 6 IV. Casualty Loss Data Organization 7 V. Casualty Loss Data Taxonomy 8 VI. Large Casualty Loss Data Analysis, Applications, and Use Cases 8 VII. Frequently Asked Questions (FAQs) 10 2 Excess Casualty Loss Data

3 INTRODUCTION Advisen s casualty loss data focuses on low frequency and high severity cases over $1,000,000 in value. Some entries are clash situations, some are catastrophes, and others are simply good examples of big, bad things happening. Most of these cases represent unpredictable clash incidences which can cause significant financial distress to companies and their insurers/reinsurers. Casualty loss data represents almost 25% of Advisen s current overall collection of loss data. With over 325,000 total entries comprising $7 trillion in loss value, Advisen s MSCAd contains loss data stemming from cyber events, factory fires, product liabilities, securities class actions, and floods. We currently track 10 LOBs within Casualty: GL, Int l / Foreign Liability, Auto Liability, Products Liability, Liquor Liability, Nuclear Liability, Owner / Contractor Protective, Environmental/Pollution Liability, Railroad Protective, and Aircraft Products Liability. All of this loss data resides in a structured database which is related back to the underlying company entity and linked to that company s demographic and exposure data. Determining the financial impact of loss events is key to our evolution of this database. Beyond actual judgments, settlements, insurance payouts, events, and other determinations of liability, MSCAd also includes estimates of potential losses as well as dismissed cases. Insurance entities use this loss data for a variety of purposes, such as predictive modeling and pricing strategy as well as underwriting, client advising, and reviewing loss experience of peers. This combination of casualty events, cases, and company information provides Advisen clients with unparalleled database capabilities, which can be accessed online, through reports, data analysis systems, and via direct data feeds. OVERVIEW CASUALTY LOSSES REPRESENT ALMOST 25% OF ADVISEN S CURRENT OVERALL COLLECTION OF LOSS DATA This loss data is particularly strong in products liability and general liability 3 Excess Casualty Loss Data

4 This loss data is particularly strong in products liability and general liability Proportionally, this loss data is well-distributed across most industries. All minor SIC codes are represented, as well. Several large insurance companies routinely relate their loss experience above $5m or above $10m to Advisen s loss data in order to compare loss experiences, strategize, and evaluate underwriting trends. 4 Excess Casualty Loss Data

5 Because this loss data represents low frequency and high severity loss events, it is particularly relevant for middle-market and large companies. About 160,000 unique companies are represented by at least one entry in Advisen s loss data. Within Advisen s casualty loss database of over 75,000 loss events are 7,000 Related Cases. These Related Cases are clash events, catastrophes, and other incidents where two or more Loss Events or potential Loss Events are linked to a single underlying cause. While we can debate the terms of clash vs. catastrophe, if macro situations like Enron, Madoff, Deepwater Horizon, PCB contamination, Vioxx, Subprime, etc., torment or threaten your book of business and risk management planning, you should find our fully-indexed and detailed listing of Related Cases extremely useful. TOP TEN RELATED CASUALTY CASES Related Case Name Case Count Approximate Aggregate Value Deepwater Horizon 152 $52,000,000,000 Fen-Phen Drug Litigation 55 $21,000,000,000 World Trade Center 72 $17,000,000,000 Takada Pharmaceutical 2 $9,000,000,000 Vioxx Recall and Litigation (Avandia) 82 $7,000,000,000 Hip & Knee Replacement Products 56 $7,000,000,000 Paxil Recall and Litigation 11 $6,000,000,000 Exxon Valdez Oil Spill 10 $5,000,000,000 Astrolabe Reef Oil Spill 6 $5,000,000,000 Takata Defective Airbags (ongoing) 53 $4,000,000,000 5 Excess Casualty Loss Data

6 We refer to our largest, most complicated clash incidents as Scenarios. Scenarios provide underwriters, actuaries, and data science experts with highly researched, structured, and actionable Related Cases which are model-ready. Our library of Scenario Cases includes noteworthy names of: CASUALTY CASE DATA SOURCES All of our loss data is sourced from publicly available information and subscription legal and news sites. All of our data is curated by hand; no entries occur solely as a result of text-mining or are automatically uploaded by computer program without human quality control and supervision. All loss research, categorization, and updates by our data research teams is overseen and directed by a Knowledge Manager and a Content Manager. This organization, encompassing over 70 staff members, enables us to add approximately 1,000 total cases per week. Our team of dedicated researchers and subject matter experts actively monitor all major legal and regulatory jurisdictions in the United States and internationally. We also regularly review hundreds of publications and other sources to spot emerging issues or trends. This approach ensures that we capture legal cases, actions, or events early for inclusion in the database. Specifically, our casualty research team relies on dozens of sources that can be generalized into three primary categories of source data: Information on individual events from a variety of public and subscription sources Bulk information on events from governmental and regulatory sources Licensed data sourced from other reputable third parties Government Websites Securities & Exchange Commission (SEC) Federal Trade Commission (FTC) Food and Drug Administration (FDA) US Department of Agriculture (USDA) National Highway & Transportation Administration (NHTSA) Consumer Product Safety Commission (CPSC) Other Sources Keyword-based alerts Official court and litigation sources Internet sites focusing on large losses Legal sites Lawyer sites International sites, both private and governmental Vendor partners 6 Excess Casualty Loss Data

7 For all new cases that are curated, we research for litigation on the Federal and State level. We update litigation with one of the following statuses: Settled, Pending, Dismissed, or Stayed. All litigation that has not settled is repeatedly checked every 6 to 12 months to update the status if appropriate. All Events/Incidences are matched with our Company database (20 million global public and private companies) that we obtain from D&B, S&P/CapIQ, and Thomson Reuters. Matching loss data to this company data benefits our clients tremendously by allowing for extensive exposure analysis by geography, industry, revenue, employee count, etc. Advisen also offers an additional file called StatMaster to provide Time Series data for these analysis and modeling efforts. CASUALTY LOSS DATA ORGANIZATION Advisen s casualty data consists of individual event records that are linked to the principal entity impacted by and/or likely to be responsible for any costs involved. The following are definitions as they relate specifically to the casualty risk: Event - An instance of a particular situation, incident, or event. The word Event indicates a casualty instance which is the original occurence when no money can be attached. Estimate An Event for which we have estimated costs to associate. Related Case A grouping of events and losses which have resulted from the same occurrence, including the foundational event and any resulting costs, fines and penalties, litigation, regulatory actions, etc. This not only groups the entire event and its impact together, but also illustrates clash potentials involving common casualty, single occurrences resulting in multiple losses, multiple defendants on individual lawsuits, and potentially multiple insurance policies and coverage parts responding to a single event or events. Litigation - Litigation filed as a result of an incident, which can have one of seven statuses: Award, Settled, Tentative Settlement, Proposed Settlement, Dismissed, Stayed, or Pending. Abstract A summary of the descriptions for the individual cases. Provides an overview of how the individual cases and companies relate to each other, reviews how the insurance may be applied, and lists those areas where Advisen found no details which would normally be expected in the type of loss. Affected Count The number of products recalled for a specific recall. Indeterminate When circumstances of the loss suggest an insurance coverage would have been triggered, but we have no ability to curate the information. An example would be when there was an employed fatality but we have no information on the workers compensation loss. Complex Litigation This category of cases requires more intensive judicial management. Complexity may be determined by multiple parties, multiple attorneys, geographically dispersed plaintiffs and defendants, numerous expert witnesses, complex subject matter, complicated testimony concerning causation, procedural complexity, complex substantive law, extensive discovery, choice of law, requisites of a class-certification order, complex damage determinations, and diversity. 7 Excess Casualty Loss Data

8 Class Actions - A class action is a procedural legal device that permits one or more plaintiffs to file and prosecute a lawsuit on behalf of a larger group, or class. Multidistrict Litigation - In the United States, multidistrict litigation (MDL) refers to a special federal legal procedure designed to speed the process of handling complex cases, such as air disaster litigation or complex product liability suits. Advisen Company Database - Advisen s casualty database is also linked to our company information that is sourced from industry-leading providers (including D&B, S&P/CapIQ, and Thomson Reuters) and supplemented by Advisen s proprietary research. Company information includes demographic information about the main organization linked to a case, such as company demographics and exposures, industry, revenue, employee count, market cap, etc., which facilitates expanded informational, modeling, and analytic capabilities on the casualty event and related MSCAd case data. CASUALTY LOSS DATA TAXONOMY Taxonomy is the orderly classification of items into distinct groups. Advisen has developed a taxonomy for casualty losses that follow the expected insurance coverages available at the time of the loss. While we have no specific coverage information for individual companies, the categorization of losses is performed by experienced insurance professionals. We break down casualty losses into the following categorizations: Services & Operations (third-party liability) Environment Products Transport & Shipping Workplace General Litigation These categories are further broken down into Types and Likely Coverage Lines. LARGE CASUALTY LOSS DATA ANALYSIS, APPLICATIONS, AND USE CASES It can be difficult to quantify the potential exposure to a future large casualty loss. Experts face questions such as: What is the likelihood of such an event? How much would the event cost if it occurred? Which policies would be exposed? What is the likely financial impact on our policies if the event occurred? Our large casualty loss data supports a number of use cases: 1. Limit Evaluation: The data provides anecdotal evidence for brokers to illustrate large losses. The value is that they help buyers to understand their potential financial exposure and brokers to assure that their clients are purchasing appropriate limits. 8 Excess Casualty Loss Data

9 2. Coverage Review: The large casualty losses describe situations which can be used to test the coverage in primary and excess forms. 3. Increased Limits Factors: Insurers can apply the large loss data to evaluate the pricing of increased limits factors for excess liability. 4. Industry Relativities: Help to fine-tune the relativities among different industries (one industry is more prone to large losses than another). 5. Parameterizing the frequency and severity of historical casualty catastrophes by line of business. 6. Simulating future casualty catastrophes by line of business using a frequency-severity approach. 7. Design Underwriting Strategy: Prospectively measure the impact of different underwriting or risk management strategies for casualty business, such as entry into a different industry or writing different coverage layers. 8. Insurers and reinsurers many times do not know the total amount of the loss beyond their participation in the liability tower. This Advisen data is available in two formats: 1. Online access through our website, advisenltd.com. Here, through filters, the data can be sliced and diced into a number of interesting formats. The full detail including descriptions and source documentation of each individual case can be reviewed as well. 2. Offline through a data feed from our ftp site. Here we give you the raw casualty data along with the static company data, and related case tables in a.csv format. 9 Excess Casualty Loss Data

10 FREQUENTLY ASKED QUESTIONS (FAQS) How frequently do we update the database? This casualty loss database, as well as the related MSCAd losses available to our clients, is updated 5 out of 7 days per week. How many people maintain and curate the casualty database? Over 70 people maintain Advisen s entire arsenal of loss data, and 20% of these resources are expressly dedicated to research and review of casualty loss data. Why is the Advisen casualty database better and more complete than other available databases? The Advisen loss data is composed of 140 data elements or attributes, including the detailed company data that is integrated into our offering. Similar offerings available in the marketplace contain a maximum of 10 to 15 data elements with no integrated company data or resulting financial loss information. Besides the basic case data on which Advisen includes information, we also have research on litigation, fines, and penalties, first party property damage, and cost estimates of various kinds of damages. What is the importance of Related Cases? Related Case IDs tie together the various results of a casualty event such as litigation, fines and penalties, estimates, and first party costs. This allows for more detailed analysis by actuaries and modelers. Are there actual claims in the Advisen casualty database? All of Advisen s loss data is sourced from actual referenceable material which is publicly accessible. None of the entries are expressly from actual claims or arrive in this database directly from a claims department of an insurance company. Advisen pursues the financial impact of the entire loss event which transcends any singular claim. However, we do provide our best estimate as to what coverage and line of business to which the case would be associated. Why is the MSCAd casualty data this superior to insurer data? Insurer data is generally truncated by policy limits. Because the data describes the entire settlement amount, it is more applicable to pricing analysis. What other ways can Advisen help with casualty risk? Advisen s casualty data is available online, in reports, and as data feeds. We also have other proprietary content related to casualty, including: Policy transactions and insurance program benchmarking Policy language wording comparisons Tagged and organized news feeds Proprietary white papers, studies, and surveys 10 Excess Casualty Loss Data