Introducing Exposure Management in CoMeta!
- CoMeta's new exposure management feature puts Praedicat's industry-leading emerging risk and liability catastrophe analytics under one roof
- Explore casualty aggregations of deterministic and probabilistic liability catastrophe events by litigation scenario, named peril, company, industry, and policy year in a redesigned intuitive and easy to use interface
- Generate detailed exposure management data reports by casualty portfolio for further analysis with the click of a button
CoMeta's new exposure management feature assesses the exposure of casualty portfolios to both deterministic liability catastrophe scenarios and Praedicat's probabilistic liability catastrophe event set. We first introduced liability catastrophe scenario analysis capabilities in CoMeta in March 2021 and now, with this release, users can access Praedicat's probabilistic loss model in CoMeta as well. [Deterministic and probabilistic liability catastrophe portfolio modeling is still available in Oortfolio, but we encourage you to employ CoMeta instead as Oortfolio is slated for retirement at the end of 2022.]
Portfolios uploaded via CoMeta or Oortfolio (see Portfolio Upload below) are automatically available for review in Exposure Management. Click on Exposure Management (1) and you'll be prompted to select one of the portfolio's you've already uploaded. Selecting a portfolio, like the N.A. Casualty portfolio shown here, will take you to the main Exposure Management page. If you want, you can then set that portfolio to be your default portfolio so that you can skip the portfolio selection step the next time around.
Next to the portfolio name you’ll see a drop down for most recent policy date (2). It defaults to the portfolio with the most recent policy date, but you can select another date if you want to review estimates for earlier versions of your portfolio. You’ll also see a selector for policy years (3). You can select "In-force" to view only your in-force book, "Legacy" to view only policies that have expired, and "All" to view in-force and legacy policies together. Underneath the policy year selector is a button for policy statistics (4). There you’ll see basic summary statistics (e.g., policy count, sum of limits) for the portfolio broken out by line of business.
Liability catastrophe aggregations can be viewed employing either the scenario loss or probabilistic loss sub-features (5). Scenario loss, which evaluates losses under general liability, directors and officers, and other casualty lines of business, is similar to the previously released scenario analysis feature. Please refer to the release notes of March 31, 2021 and June 23, 2021 for a description of that feature.
The probabilistic loss sub-feature, which evaluates general liability policies only, reports probabilistic exposure and loss estimates employing Praedicat's probabilistic liability catastrophe model. The headline statistics area shows the number of Litagion agents that generate ground-up loss for the policies in the portfolio, the expected insured loss in the portfolio, tail value at risk (TVaR) at the 5th percentile of the insured loss distribution, and the probable maximum loss (PML) at the 5th and 1st percentiles of the insured loss distribution. For example, in the N.A. Casualty portfolio shown here, the portfolio is exposed to 57 Litagion agents generating $19.2 M in expected insured loss. There is a 5% chance that insured losses could total at least $53.9 M. The loss latency numbers (6) indicate how many years will pass before 50% of the losses will be paid out. In the present example, 50% of losses pay out within the first 9 years in the events contributing to the PML(5). The full exceedence probability and loss latency curves can be viewed by clicking on the graph icon (7).
The table below the headline statistics provides an indication of how portfolio losses break down by Litagion agent, company, industry (defined in terms of NAICS), policy, and policy year. In the case of Litagion agents, users can view this break down in terms of contributory, standalone, and marginal losses (8):
- Contributory losses. Litagion agent contributory losses measure losses at the Litagion agent level within the same simulated events that contribute to the corresponding portfolio loss statistic. Contributory losses are calculated such that the sum of contributory losses equal the corresponding portfolio loss statistics. For example, the sum of Litagion agent contributory TVaR(5) losses will equal the portfolio TVaR(5) loss of $72.6 M in the N.A. Casualty portfolio.
- Standalone losses. Litagion agent standalone losses are calculated Litagion agent-by-Litagion agent as if a single Litagion agent were the only peril to which the portfolio is exposed.
- Marginal losses. Litagion agent marginal losses are calculated by taking the difference between the portfolio loss statistics with and without the Litagion agent in question. The resulting numbers indicate how much portfolio losses will fall if a given Litagion agent is excluded from the portfolio.
The table also indicates the number of policies with positive ground-up loss for each Litagion agent and the sum of the corresponding limits for those exposed policies. Policies triggered indicates the number of policies in the portfolio expected to generate insured loss given policy terms and conditions. In the case of Litagion agents, policies triggered is not defined for contributory and marginal loss statistics since, in those cases, a policy is not necessarily triggered by a single Litagion agent; losses from multiple Litagion agents might be necessary to generate sufficient loss to breach a policy's aggregate attachment.
Contributory and standalone losses have a similar interpretation when viewing portfolio losses by company, industry, policy, and policy year. Marginal loss statistics are available only for Litagion agents.
Clicking on a Litagion agent name in the table opens a modal that displays the contribution of companies, industries, policies, and policy years to the Litagion agent's contributory or standalone losses. For example, in the modal below, Mitsui Chemicals, Inc. contributes $5.7 M to the N.A. Casualty portfolio's perfluorooctanoic acid TVaR(5) losses. The contributory losses displayed in the modal sum to their corresponding Litagion agent values, regardless of whether the user begins with contributory or standalone Litagion agent losses.
You can download detailed scenario loss and probabilistic loss data reports by clicking the "Download All Results" (9) button just above the headline statistics area. If the report has already been generated, it will download to your browser immediately. If the report has not yet been generated, you will be asked to confirm that you want to generate the report and whether you want to be notified by email once the report is ready for download. You must request the scenario and probabilistic loss data reports separately. Once generated, reports are available for download by clicking "Download All Results" or from CoMeta's new reports feature (10). The report zip files contain the data you need to generate all of the results shown in the scenario loss and probabilist loss sub-features. Please contact your account manager to learn more about how to work with these data reports.
CoMeta's portfolio upload utility was first released in March 2021. Please refer to the release notes published then to learn more about the portfolio upload process. The current release introduces a slight modification to that process. Portfolio validation now occurs in two steps. In the first step, the portfolio template is validated to ensure that it is in the proper format. Once you submit your portfolio template, a message will appear indicating that the template is undergoing validation. If there are errors in the template, a second message in red will appear (11). You can click on the error message to determine the nature of the error that you must correct before resubmitting the portfolio template.
Portfolios that pass the initial template validation stage will appear in the portfolio status table as before. If additional errors are detected in the portfolio data (e.g., invalid industry codes, missing limit and attachment data), an error message will appear that you can then interrogate to determine whether corrections are needed before joining the portfolio to the liability catastrophe event set and computing the portfolio loss statistics viewable in Exposure Management.