TOP
Property insurance is characterized by a high degree of uncertainty regarding the frequency and severity of claims. Exogenous factors such as climate change, technology, and regulatory requirements increase the volatility of claims patterns. Against this backdrop, underwriting plays a central role, as it forms the interface between risk acceptance, pricing, and capital management.
Underwriting serves as the central management tool for insurance companies in assessing and accepting risks. In property insurance, which protects tangible assets such as buildings, machinery, or inventory – as well as the consequences of business interruptions – against insured perils, the underwriting decision is a key determinant of the economic viability of the insurance portfolio.
The aim of this article is to systematically present the core elements of underwriting and to analyze the functional relationships in the context of product and risk management.
In the insurance industry, underwriting refers to the process of systematic risk analysis, risk assessment, and underwriting decisions (acceptance, rejection, or modification of terms) while determining appropriate premiums and policy terms. It aims to systematically quantify and price risks in order to ensure financial stability, risk equity, and a balanced relationship between risk, return, and capital commitment, while at the same time avoiding adverse selection.
Underwriting is often confused with pricing or used in the same context. In property insurance, pricing refers to the actuarial determination of insurance premiums for property risks (e.g., building, household, industrial, motor property insurance). It should be risk-adequate, competitive, and profitable in the long term.
In addition, pricing serves portfolio management purposes, for example, to withdraw from loss-prone segments.
Unlike pricing, underwriting encompasses not only premium calculation but also qualitative decisions regarding coverage limits, insurance terms, and risk criteria. While pricing primarily uses economic-mathematical models, underwriting integrates qualitative assessments and policy-oriented decisions.
Underwriting in property insurance is structured as a multi-stage process comprising several core elements that together enable efficient risk management at the individual and portfolio levels, including:
Risk identification involves recognizing potential hazards that could lead to insurable losses. It forms the basis of every underwriting decision and encompasses the systematic collection and analysis of all risk-relevant characteristics of an insured object. In property insurance, these include the type of construction, use (operational processes), location, technical equipment, safety measures, and loss history, in addition to other characteristics.
The goal of risk assessment is to estimate the probability of loss and the potential amount of loss for a risk to be insured, based on property characteristics, location, and environmental factors as well as existing safety and prevention measures.
Property-related characteristics describe what specifically is insured and are central to the basic risk assessment, as they primarily determine the amount of loss and susceptibility to loss. Examples of such characteristics include
Site and environmental characteristics are particularly relevant for the identification and assessment of accumulation risks and major losses. These include, for example,
Existing safety and prevention measures reduce both the probability of damage and the potential extent of damage, and are therefore significant for qualitative risk assessment. They also serve to systematically assess the probability and extent of damage. Particularly relevant measures include
Risk selection is based on the data collected during risk identification and risk assessment; that is, the underwriter decides whether a risk to be insured is accepted, accepted with modifications, or rejected. This decision serves to manage the insurance portfolio in a targeted manner and to avoid risks with a systematically negative claims trend (antiselection).
Risk assessment is based on considerations regarding the probability of a loss as well as on loss severity characteristics; that is, risk assessment considers the question of how high a loss might be in the event of a claim. The assessment is based on historical loss statistics, insured values exposed to loss, and maximum loss scenarios and simulations.
Key factors influencing frequency risk include
Loss severity characteristics determine the potential extent of damage to an insured property. Key assessment criteria include
Another key factor in risk assessment is the analysis of risk concentrations and accumulation characteristics to determine whether multiple individual risks could be affected simultaneously in the event of a loss, thereby resulting in major or catastrophic losses. Particular attention is paid to geographic concentrations, industry-specific clusters, interdependencies among underwritten risks, natural hazard accumulation risks, and reinsurance exposure.
When risks are viewed over time, it becomes clear that they are not static but subject to change. Accordingly, factors such as seasonality (e.g., storm or frost periods), the asset’s lifecycle, and technological and climatic developments must be taken into account.
Since not all risks are fully quantifiable, qualitative assessment aspects are incorporated, such as management and organizational quality, risk awareness, safety culture, and maintenance and control processes.
To support the risk assessment, statistical and mathematical metrics are used to determine the insurance premium (pricing), reinsurance, and the necessary capital requirements. These include
In addition, underwriters take into account risk premiums that consist of uncertainty and model risk characteristics such as data uncertainty, model assumptions, insufficient sample size, misjudgments of past trends without considering structural and developmental breaks (e.g., the development of rare catastrophic losses (extrapolation risk) as well as regulatory and internal company requirements (e.g., Solvency II, internal risk models, underwriting guidelines).
In a subsequent step, the results of risk identification and risk assessment are used to determine the terms and price at which an insurance contract is offered.
The basis for pricing consists of actuarial models, historical loss statistics, and individual risk characteristics, supplemented by administrative, safety, and profit margins. In addition to determining the premium, policy parameters such as exclusions, deductibles, sublimits, and coverage limits are established to tailor the scope of coverage specifically to the respective risk. Key features include:
The risk assessment result is based on the classification of the property into a risk class as well as expected loss, expected loss ratio, loss volatility, and maximum-loss measures, including Probable Maximum Loss (PML), Maximum Possible Loss (MPL), and Maximum Foreseeable Loss (MFL).
Factors influencing premium calculation to determine a risk-appropriate, economically viable premium include sum insured, deductibles, maximum indemnities, business types and hazard groups, rate zones, and minimum premiums.
These determine the scope of coverage and the insurer’s liability and limit the frequency and severity of losses; key factors include insured perils, coverage extensions, risk exclusions, sublimits, indemnity limits, and liability periods.
Incentive and control mechanisms are designed to promote the policyholder’s own precautions and responsibility for loss prevention. Examples may include obligations, safety regulations, maintenance and inspection duties, contractual penalties, and requirements for risk mitigation.
Claims and policy histories form a basis for individual adjustments to rates and terms, particularly regarding maximum loss limits, deductibles, exclusions, and premium surcharges; they are dependent on the number, amount, and causes of past claims, as well as preventive and remedial measures taken. Contract duration, contract retention, and prior contract terminations are also relevant.
Consideration is given to the target customer segments, market environment, price levels, and distribution channels in the design of prices and terms.
These influence pricing and coverage concepts, particularly through reinsurance structure, retention, coverage limits, accumulation restrictions, and reinsurance terms, as well as the resulting underwriting capacity.
These set the institutional framework for underwriting, including Solvency II requirements, underwriting rules and authorizations, and the company’s internal risk strategy.
The final step in the underwriting process is the contractual acceptance decision. It is not only an operational step but also a decision-making act with legal, regulatory, and risk-management implications. It directly influences an insurer’s liability and loss potential. A risk may be accepted as is or subject to certain contractual conditions, such as requirements for risk mitigation, the exclusion of specific risk elements, or adjustments to the premium, deductible, or maximum indemnity, provided that the underwriter does not fundamentally reject the insurance of the risk in question.
All decisions must be documented in a transparent manner to withstand subsequent reviews (e.g., audits). To this end, all risk characteristics relevant to the underwriting decision must be described, such as the overall risk assessment based on frequency, maximum loss, and accumulation risk. The central questions to consider are to what extent the risk under consideration deviates from a typical or standard risk in the respective risk class, how high is the remaining uncertainty, and whether the risk appears acceptable overall under the selected conditions.
In addition, economic efficiency and profitability characteristics play a significant role. The benchmark is the sustainable profitability of the individual contract, which results from the calculated gross premium, the cost and commission burden (e.g., operating and reinsurance costs), the necessary capital commitment, and the expected contribution margin. In addition to the individual risk decision, the contribution of the respective risk to the overall portfolio must ultimately be taken into account, particularly with regard to portfolio diversification and expected cumulative losses.
The underwriting decision must comply with the insurance company’s existing underwriting guidelines. This includes underwriting mandates and prohibitions, the respective underwriter’s authority to underwrite, and compliance with internal underwriting guidelines. Deviations require clear documentation of the reasons for the decision, including any necessary approval and authorization processes.
Legal and regulatory requirements must also be observed, such as compliance with pre-contractual disclosure obligations, transparency of contract terms, regulatory compliance, and data protection regulations. Errors in this area entail significant liability risks.
Even after a risk has been accepted, comprehensive documentation supports ongoing contract management, including tracking conditions and deadlines, managing follow‑up inspections, or implementing necessary remediation measures.
Underwriting does not end with the conclusion of the contract. Risk management and loss prevention measures, such as safety requirements, risk inspections, or the implementation of prevention measures, contribute significantly to reducing the frequency and severity of losses.
While the underwriting process is primarily focused on the selection and pricing of risks, risk management and loss prevention aim to systematically avoid or limit losses and stabilize the portfolio in the long term at both the individual risk and portfolio levels to ensure the insurer’s future risk-bearing and underwriting capacity.
Risk management is thus an integral part of corporate governance, for example, within the framework of Solvency II requirements and enterprise risk management, supported by internal control, audit, and compliance structures. In this context, a regular reassessment of existing risks – for example, as part of contract renewals – is just as necessary as continuous portfolio analysis, achieved through protocols such as site inspection reports, loss cause and frequency analyses, or audit and review reports. The goal is to identify new risk characteristics and loss trends at an early stage and to be able to counteract them both through internal measures (e.g., adjustment of underwriting guidelines) and through external prevention recommendations to the policyholder.
Prevention tools can generally be distinguished based on whether they aim to prevent or limit losses. Complete loss prevention is often only possible to a limited extent and, for the insurer, essentially boils down to not underwriting particularly loss-prone risks or not renewing contracts upon expiration.
Of greater practical importance, therefore, is the mitigation of loss impacts through technical, organizational, and contractual measures. Among the tools available for this purpose are technical protective measures (e.g., fire alarm and suppression systems, burglar alarm systems), organizational measures to raise risk awareness, defining roles and responsibilities, improving order and cleanliness, regularly instructing employees on existing hazards, and emergency and crisis plans to minimize potential business interruption periods.
In addition, contractual requirements or economic incentives can contribute to the implementation of risk mitigation measures. Typical contractual requirements include the establishment of obligations, the stipulation of necessary safety regulations, and limitations on insurance coverage. Economic incentives include, for example, reductions in deductibles, discounts on insurance premiums, and the extension of insurance coverage to additional risks.
At the portfolio level, analytical tools such as catastrophe models and simulations support the identification of systematic risks and trend developments. Opportunities for action arise both from improved portfolio diversification (e.g., business lines, regional diversification, insured risks) and from adjustments to the reinsurance strategy (e.g., contract structure, facultative solutions, design of the retention and excess of loss structure).
Capacity and accumulation control is the link between individual risk underwriting and the viability of the overall portfolio. It is particularly essential in property insurance to keep large-scale and catastrophic losses manageable. Available capital and Solvency II capital requirements determine an insurer’s risk-bearing capacity and risk appetite, thereby setting the absolute framework for underwriting capacity at both the individual risk and portfolio levels.
Of particular concern are natural hazards, which can lead to significant accumulations of losses. Underwriters must therefore ensure that aggregated risks remain within the risk-bearing capacity. Accumulation control is heavily scenario-based, for example through extreme, stress, and sensitivity analyses. In this context, data quality and level of detail, geocoding of risks, model assumptions and uncertainties, and the validation of the catastrophe models used are crucial for enabling reliable conclusions.
It should be noted that accumulations build up in the portfolio only over time. Accordingly, in addition to the current portfolio structure, accumulation expectations like seasonal effects, changes in exposure, and climatic and socioeconomic trends must also be considered. These accumulation expectations are also decisive for reinsurance needs, as they influence available reinsurance capacity, retentions, coverage limits, and reinsurance costs.
Reinsurance is one of the key instruments for risk transfer, capital preservation, and profit protection. It is an integral part of underwriting in property insurance and is closely linked to underwriting, capacity and cumulative risk management and regulatory requirements. Reinsurance is an ongoing process, not a one-time transaction. It must therefore be integrated into the underwriting and pricing process. This allows large and extreme risks to be addressed and volatility in earnings and claims to be specifically reduced (see the objectives below).
Reinsurance serves as both a risk management tool and a capital management tool. The basis for the structure, limits, and retention of reinsurance is the nature of the insured risks, Maximum Possible Loss (MPL), aggregated exposure / cumulation risks, and potential exposure to natural hazards.
In addition, there are other forms, among them captive and pool solutions or other financial instruments of the capital market (Alternative Risk Transfer (ART)) such as catastrophe bonds, Industry Loss Warranties (ILW), and parametric coverage.
Key features of reinsurance coverage depend on loss frequency, severity, and volatility. It also depends on which risk the primary insurer bears itself (retention / priority) and what coverage limits / liability limits are desired.
The legal structure is crucial for the effectiveness of reinsurance coverage; drafting the contract therefore requires considerable expertise to avoid errors. Particular attention must be paid, among other things, to follow-the-fortunes / follow-the-settlements clauses, coverage clauses, exclusions, maximum loss / aggregate definitions, and claims control clauses.
Equally important is the selection of suitable reinsurance partners to minimize default risks; financial strength, credit rating, and diversification are key criteria here for ensuring solvency in the event of a claim.
Reinsurance must make economic sense. That is, the reinsurance premium, stabilization of results, and the effects of capital relief for the primary insurer must be in an acceptable cost-benefit ratio. To ensure a smooth exchange between primary insurers and reinsurers, high-quality data on the portfolio to be reinsured and consistent monitoring (e.g., renewal cycles, contract terms) are required.
Reinsurance impacts a primary insurer’s regulatory compliance (eligibility under Solvency II) and financial reporting and is a prerequisite for capital relief effects. In this regard, it is imperative for a primary insurer to demonstrate compliance with legal and accounting requirements through risk transfer testing as well as reporting and disclosure obligations.
Portfolio management, as part of corporate governance, serves as the overarching framework within which underwriting, risk, capital, and reinsurance decisions converge; it forms the basis for optimizing the risk-return profile, stability, and growth of the overall portfolio. It expands underwriting from a focus on individual risks to the overall management of the entire insurance portfolio. The goal is to achieve balanced diversification across lines of business, regions, and risk classes, thereby simultaneously meeting regulatory requirements.
Digitalization is fundamentally transforming underwriting, not just in isolated instances, but structurally in terms of processes, decision-making logic, data infrastructure, and role models.
Fully automated, rule-based underwriting, also known as straight-through processing (STP), is increasingly being adopted for standard risks (mostly smaller commercial and industrial risks). These risks are largely assessed automatically based on rules, automatically assigned to rates and conditions, and underwritten, with the aims of increasing efficiency (e.g., processing speed and costs) and improving the consistency of decisions. In this process, large volumes of internal and external data are used for risk assessment, including claims and portfolio data, geodata, weather and climate data, and economic and market data. The process requires both significant effort in analysis and integration into existing IT systems and sufficient data quality.
Thanks to technological advances in the field of artificial intelligence (AI), insurers are increasingly engaging with automated processes and analyses in order to identify more complex risk patterns and incorporate them into risk assessment.
Machine learning models, for example, identify nonlinear risk patterns and improve forecast accuracy, with special attention to claims probability and severity forecasts, fraud detection, underwriting scoring, and risk segmentation. These methods enable proactive risk assessment (predictive underwriting) based on predictive models. However, critical issues with such models remain, among them the explainability of the algorithms used for underwriting decisions (black box), forecast uncertainties in the event of structural breaks, and regulatory challenges.
AI’s ability to link complex data at high processing speeds supports the assessment of potential risk and catastrophe scenarios, enabling the early identification and evaluation of major risks and potential accumulations. It also facilitates their consideration in both underwriting and portfolio management and appropriate adjustments to reinsurance requirements. The reliability of automated methods depends substantially on the quality of the models and assumptions used.
Property insurance is expected to increase its use of automated underwriting systems, provided that real-time data from sensors can be collected directly from policyholders (keywords: Internet of Things (IoT), telematics), which would enable dynamic underwriting. Through such sensor-based real-time data (e.g., building sensors, condition monitoring of machinery), existing risk situations could be continuously monitored and assessed, which would enable pricing tailored to the respective risk situation.
In addition to the need to process large volumes of real-time data, data protection concerns and policyholder acceptance constrain widespread implementation. Furthermore, significant investment costs would be required on the part of both insurers and policyholders.
Efforts are being made to supplement expert assessments with automated preliminary evaluations and analyses based on available information (e.g., inspection reports, internet research, geolocated data) to reduce the time required of experts and increase the efficiency of expert involvement.
For several years, the use of so‑called digital underwriting platforms has grown, enabling end-to-end digitization of the underwriting process. These platforms are typically based on a unified data platform with interfaces to connected insurers and reinsurers. All data necessary for underwriting and writing a risk is made centrally available to platform participants, enabling uniform transparency and significant time and cost savings through workflow and documentation automation. However, dependencies on the respective platform still exist, as does a considerable implementation effort on the part of the participating market players.
It is expected that digital underwriting approaches will lead to a paradigm shift from experience-based individual risk underwriting to data‑, model‑, and portfolio-driven decision-making processes with the aim of increasing efficiency and risk transparency. At the same time, new challenges are expected in the areas of explainability, governance, and data ethics.
Underwriting property and casualty insurance involves significant risks and challenges. These include, among others, increasing natural hazards, non-stationary risk structures resulting from climate change, model and data risks, regulatory requirements, and conflicting objectives between growth and risk discipline. These factors significantly increase the complexity of underwriting decisions and make clear the limitations of purely history-based risk assessment.
Underwriting increasingly goes beyond the mere assessment of individual risks and should be understood as an integrated, data- and portfolio-driven decision-making process.
An analysis of the individual process stages – from risk identification and assessment to rate and condition selection, underwriting decisions, risk management, concentration control, and reinsurance to portfolio management – makes it clear that the quality of underwriting depends significantly on the consistency and integration of these elements. In particular, the use of statistical methods, scoring models, and geographic information systems (GIS) contributes significantly to improving risk transparency and decision-making quality.
However, technological innovations alone will not be sufficient to achieve sustainable underwriting success. Rather, a balanced interplay of data quality, robust model governance, appropriate organizational frameworks, and human expertise is required. Hybrid approaches, in which automated models are supplemented and monitored by qualified underwriters, prove to be particularly effective.
Successful underwriting in property and casualty insurance today is characterized less by individual tools and more by a holistic, strategically embedded understanding of risk. The core elements range from systematic risk identification and statistical-mathematical assessment to regulatory-compliant underwriting decisions. Underwriting is not merely an operational function, but a key driver of stability, competitiveness, and long-term profitability for insurance companies.
Underwriting in property and casualty insurance will continue to evolve dynamically. Advancing digitalization, the use of artificial intelligence, IoT technologies, and real-time data will further expand the possibilities for risk assessment and management. In particular, forward-looking and preventive underwriting approaches make possible identifying risks early and actively influencing them, rather than merely pricing them.
At the same time, the importance of explainability, fairness, and governance will continue to grow. Regulatory requirements and societal expectations will place greater limits on the use of data-driven models and demand transparent decision-making logic. Explainable AI approaches (methods and techniques that make the decision-making processes of artificial intelligence – particularly black-box models such as deep learning – comprehensible, transparent, and interpretable for humans) as well as structured model validation processes are likely to gain importance in the future.
Climate change, as a structural risk driver, represents one of the greatest challenges. The increasing shift away from stationary risk models requires new methodological approaches, such as scenario analyses, adaptive pricing, and closer integration of underwriting, reinsurance, and capital management. There is significant need for further development in this area, particularly in the core fields of integrating decision-making psychology into underwriting, the effectiveness of preventive measures, and the economic evaluation of hybrid human-machine decision-making models.
Underwriting is moving toward intelligent decision support combined with machine-learning systems and qualified underwriters, rather than toward full automation. Integrating technological innovations into a consistent risk strategy will be a factor in the performance of property insurance.
