AmazonScience/document-haystack

1Key Assumptions of our Actuarial Methods by Line of Business2U.S. Workers’3Compensation4We generally use a combination of loss development and expected loss ratio methods for U.S. Workers’ 5Compensation as this is a long-tail line of business. 6The tail factor is typically the most critical assumption, and small changes in the selected tail factor can have a material 7effect on our carried reserves. For example, the tail factors beyond twenty years for guaranteed cost business could 8vary by 1 percentage point below to 2.5 percentage points above those indicated in the 2023 detailed valuation review. 9For excess of deductible business, in our judgment, it is reasonably possible that tail factors beyond twenty years 10could vary by 1.5 percentage points below to 3 percentage points above those indicated in the 2023 detailed valuation 11review.12U.S. Excess Casualty We utilize various loss cost trend assumptions for different segments of the portfolio. In our judgment, after evaluating 13the historical loss cost trends from prior accident years since the early 1990s, it is reasonably possible that actual loss 14cost trends applicable to the year-end 2023 detailed valuation review for U.S. Excess Casualty may range 5 15percentage points lower or higher than this estimated loss trend. The loss cost trend assumption is critical for the U.S. 16Excess Casualty line of business due to the long-tail nature of the losses, and it is applied across many accident 17years. Thus, there is the potential for the loss reserves with respect to a number of accident years (the expected loss 18ratio years) to be significantly affected by changes in loss cost trends that were initially relied upon in setting the loss 19reserves. These changes in loss trends could be attributable to changes in inflation or in the judicial environment, or in 20other social or economic conditions affecting losses.21U.S. Excess Casualty is a long-tail line of business and any deviation in loss development factors might not be 22discernible for an extended period of time subsequent to the recording of the initial loss reserve estimates for any 23accident year. Mass tort claims in particular may develop over a very extended period and impact multiple accident 24years, so we usually select a separate pattern for them. Thus, there is the potential for the loss reserves with respect 25to a number of accident years to be significantly affected by changes in loss development factors that were initially 26relied upon in setting the reserves. 27In our judgment, after evaluating the historical loss development factors from prior accident years since the early 281990s, it is reasonably possible that the actual loss development factors could vary by an amount equivalent to a six 29month shift from those actually utilized in the year-end 2023 detailed valuation review. This would impact projections 30both for accident years where the selections were directly based on loss development methods as well as the a priori 31loss ratio assumptions for accident years with selections based on Bornhuetter Ferguson or Cape Cod methods. 32Similar to loss cost trends, these changes in loss development factors could be attributable to changes in inflation or in 33the judicial environment, or in other social or economic conditions affecting losses.34Given the very long-tail nature of this business, the tail factor selection can also have material impact on our carried 35reserves. The sensitivity around tail selection may also be a proxy for the sensitivity of a calendar year impact of 36monetary inflation on unpaid losses. It is reasonably possible for the tail factors for Excess Casualty could vary by 2 37percentage points below to 3.5 percentage points above those indicated in the 2023 detailed valuation review.38U.S. Other Casualty The key assumptions for other casualty lines are similar to U.S. Excess Casualty, as the underlying business is long-39tailed and can be subject to variability in loss cost trends and changes in loss development factors. These may differ 40significantly by line of business as coverages such as general liability, medical malpractice and environmental may be 41subject to different risk drivers.42U.S. Financial Lines The loss cost trends for U.S. D&O liability business vary by year and subset. After evaluating the historical loss cost 43levels from prior accident years since the early 1990s, including the potential effect of losses relating to the credit 44crisis, in our judgment, it is reasonably possible that the actual variation in loss cost levels for these subsets could vary 45by approximately 10 percentage points lower or higher on a year-over-year basis than the assumptions actually 46utilized in the year-end 2023 reserve review. Because the U.S. D&O business has exhibited highly volatile loss trends 47from one accident year to the next, there is the possibility of an exceptionally high deviation. In our analysis, the effects 48of loss cost trend assumptions affect the results through the a priori loss ratio assumptions used for the Bornhuetter 49Ferguson and Cape Cod methods, which impact the projections for the more recent accident years.50The selected loss development factors are also an important assumption, but are less critical than for U.S. Excess 51Casualty. Because these lines are written on a claims made basis, the loss reporting and development tail is much 52shorter than for U.S. Excess Casualty. However, the high severity nature of the losses does create the potential for 53significant deviations in loss development patterns from one year to the next. Similar to U.S. Excess Casualty, after 54evaluating the historical loss development factors from prior accident years since the early 1990s, in our judgment, it is 55reasonably possible that actual loss development factors could change by an amount equivalent to a shift by six 56months from those actually utilized in the year-end 2023 reserve review.57UK/Europe Casualty and58Financial Lines59Similar to U.S. business, UK/Europe Casualty and Financial Lines can be significantly impacted by loss cost trends 60and changes in loss development factors. The variation in such factors can differ significantly by product and region, 61however the range of potential impacts is much lower than that of other lines of business noted above.62U.S. and UK/Europe63Property and Special64Risks65For shorter-tail lines such as Property and Special Risks, variance in outcomes for individual large claims or events 66typically has a greater impact on results than does changes in actuarial assumptions or methodology. This is because 67a greater proportion of the ultimate loss, at any stage of development, is composed of reported losses than IBNR 68reserves. These outcomes generally relate to unique characteristics of events such as catastrophes or losses with 69significant business interruption claims.70U.S., UK/Europe and71Japan Personal Insurance72Personal Insurance is short-tailed in nature similar to Property and Special Risks but less volatile. Variance in 73estimates can result from unique events such as catastrophes. In addition, some subsets of this business, such as 74auto liability, can be impacted by changes in loss development factors and loss cost trends.75Line of Business or Category Key Assumptions76ITEM 7 | Critical Accounting Estimates7752 AIG | 2023 Form 10-K