We may experience losses or disruptions from catastrophes and other significant, infrequent loss events. As a company with significant property and casualty insurance underwriting operations, we may experience losses from man-made or natural catastrophes. Catastrophes and other significant, infrequent loss events include, but are not limited to, windstorms, hurricanes, earthquakes, tornadoes, derechos, hail, severe winter weather, floods, and wildfires and may include pandemics and events related to terrorism, broad reaching cyberattacks, riots, and political and civil unrest. While we employ catastrophe modeling tools in our underwriting process, we cannot predict how severe a potential catastrophe will be before it occurs. The extent of losses from catastrophes is a function of the total amount of losses incurred, the number of insureds affected, the frequency and severity of the events, the effectiveness of our catastrophe risk management program, and the adequacy of our reinsurance coverage. Catastrophes can occur over numerous geographic areas; however, some catastrophes may produce significant damage in large, heavily populated areas. We offer coverage against terrorist acts in connection with some of our programs, and in other instances we are legally required to offer terrorism insurance; in both circumstances, we actively manage our exposure, but if there is a covered terrorist attack, we could sustain material losses. Catastrophes also may result in significant disruptions in our insurance and other operations, as well as loss of income and assets. The impacts of climate change may increase the frequency and/or severity of weather-related catastrophes, which may result in elevated catastrophe-related losses or disruptions, which may be material. See "Climate Change" for more information about the potential impacts of climate change.
The failure of any of the methods we employ to manage our loss exposures could have a material adverse effect on us. We seek to manage our loss exposures in a variety of ways, including adhering to maximum limitations on policies written in defined geographical zones, implementing maximum gross limits by coverage for each insured, establishing per risk and per occurrence limitations for each event, employing coverage restrictions, and following prudent underwriting guidelines for each program written. We also seek to manage our loss exposures through geographic and industry diversification. Underwriting is a matter of judgment, involving assumptions about matters that are inherently unpredictable and beyond our control, and for which historical experience and probability analysis may not provide sufficient guidance. For example, as weather patterns evolve, historical data and models may be less predictive, increasing the uncertainty in catastrophe loss estimates and pricing. One or more future events could result in claims that substantially exceed our expectations, which could have a material adverse effect on our results of operations and financial condition. In addition, we seek to manage our loss exposures through policy terms, coverage exclusions, and choice of legal forum. Disputes relating to coverage and choice of legal forum also arise. As a result, various provisions of our policies, such as choice of forum, or coverage limitations or exclusions, may not be enforceable in the manner we intend and some or all of our methods to manage loss exposures may prove ineffective.
The effects of emerging claim and coverage issues on our business are uncertain. As industry practices and legal, judicial, social, and other environmental conditions change, unexpected and unintended issues related to claims and coverage may emerge. These issues could have a material adverse effect on our results of operations or financial condition by either broadening coverage beyond our underwriting intent or increasing the frequency and/or severity of claims. For example, rising costs, litigation funding, social inflation, including new or expanded theories of liability, higher adverse verdicts, and legislative changes, such as extended statutes of limitations, may result in higher and more frequent claims over a longer reporting period than originally expected. In some instances, these changes may not become apparent until after we have issued insurance or reinsurance contracts that are affected by the changes. As a result, the full extent of liability under our insurance or reinsurance contracts may not be known for many years after a contract is issued.
We use analytical models to assist our decision making in key areas such as pricing, reserving, and capital modeling, and actual results may differ materially from the model outputs and related analyses. We use various modeling techniques and data analytics (e.g., scenarios, predictive and stochastic modeling, and forecasting) to analyze and estimate exposures, loss trends, and other risks associated with our insurance businesses. This includes both proprietary and third-party modeled outputs and related analyses to assist us in, among other things, decision-making related to underwriting, pricing, capital allocation, reserving, investing, reinsurance, and catastrophe risk. We incorporate numerous assumptions and forecasts about the future level and variability of policyholder behavior, loss frequency and severity, interest rates, equity markets, inflation, capital requirements, and currency exchange rates, among others. The modeled outputs and related analyses from both proprietary models and third-party models, including statistical, artificial intelligence, and machine-learning models, are subject to various assumptions, uncertainties, model design errors (e.g., bias and inadequate validation or documentation), complexities, and the inherent limitations of any statistical analysis, including those arising from the use of historical internal and industry data and assumptions, which could result in mispricing, misreserving, or capital misallocation.
In addition, the modeled outputs and related analyses may from time to time contain inaccuracies, perhaps in material respects, including as a result of inaccurate inputs or applications thereof (whether due to data error, human error, or otherwise). Consequently, actual results may differ materially from our modeled results. Our profitability and financial condition substantially depend on the extent to which our actual experience is consistent with assumptions we use in our models and ultimate model outputs. If, based upon these models or other factors, we misprice our products or fail to appropriately estimate the risks we are exposed to, our business, results of operations, and financial condition may be materially adversely affected.