In industries the world over, accurate pricing is the key toavoiding financial catastrophe. The property and casualty insurancearena is no exception. During the past decade and a half, the mostsuccessful P&C businesses have gotten stronger by employingever- more sophisticated underwriting strategies across businesslines. And in today's disruptive insurance marketplace, rapidinnovations in analytic technologies continue to revolutionize theconcept of underwriting intelligence and efficiency. As new datasources and categories of risks emerge, P&C insurers face theongoing challenge of effectively implementing analytic tools togain and maintain a competitive edge.

Smart underwriting is necessarily receiving a bigger share offunding for analytics in many insurance enterprises because it isthe core of risk decision-making. Underwriting analytics is amongthe most important of all analytics for an enterprise. If price iswrong, nothing will save a business from creditors in the short runor from its competitors in the long run. This is true not only forP&C risk-based pricing but for every risk management decision.Executives cannot ignore competing on analytics anywhere in theirenterprise — most especially in pricing risks and managingdecisions about risk portfolios.

Multiline carriers are aggressively applying everything theylearn from all segments of their business, be it personal orcommercial lines, marketing, even reported claims. Commercialvehicles have long used vehicle location technologies for assetmanagement, logistics planning, driver management, andaccident/incident event recording. The commercial lines market isnow learning how to benefit from the confluence of predictivemodeling methods and technology honed in personal lines. Propertylines are profiting from the investment in auto lines, mostspecifically because the value of analytics has already been provento line and senior executives. Those decision makers now want tomove aggressively rather than wait for the competition to actfirst. Personal auto has been the prime example for showingexecutives that “catching up” is not the best strategy. Theadvantage of taking the initiative is apparent in companies thatare thriving and those that are no longer contenders.

Environment and InsuredLosses
Indisputable relationships exist betweeninsureds' losses and their immediate environment. For example, inproperty lines, high concentrations of flammable materials with lowwater availability, inability to notify officials of an emergingfire, and poor firefighting capabilities create a recipe for firehazards. The danger is compounded if the area is far from aresponding fire station, hard to access via roadways, surrounded byderelict structures, and subject to high winds and hot, dryconditions with atmospheric lightning strikes. On their own, any ofthose factors would likely prompt a risk manager to increase theestimate of potential for a fire loss to occur and for the fire toburn out of control in ways that would raise the severity of theloss. When the factors are combined, the odds of calamityincrease.

Many insurers might either decline to underwrite such a risk orquote a very high rate, as well as exacting terms and conditionssurrounding the contract. Insurers would likely also demandadditional mitigation measures before initiating coverage.

Many individual attributes with varying degrees of positive ornegative effects contribute to an overall risk equation. Thismultivariate effect manifests itself in a myriad of ingredients,including geographic, physical, social, and political attributes.Improved risk management strategy begins with data being collectedin an intelligent and precise manner by qualified and certifiedprofessionals. The data-collection baseline is then enhancedthrough an analytic approach to deploy the data effectively. Theanalytic paradigm combines specific subject-matter expertise, dataaggregation, interpretation, verification, validation, and modelingsteps. It can also involve the distillation of many companies' lossexperience over a long period of observation. This type ofquantification and analysis is fast becoming the standard of moderninsurance operations. Subsequent steps to validate the riskequation often build from this starting point.

Communities vary significantly regarding factors such asgeography, weather, roadways and traffic patterns, physicalinfrastructure, and investments in firefighting, police, floodmitigation, and building-code enforcement — attributes thatstrongly affect insured loss frequency and severity.

To administer a program in support of loss cost estimation,insurance providers can develop and maintain insured-loss data byworking closely at the grassroots level with local officials acrossthe nation in the assessment of community loss-mitigationcapabilities. Such continual collaboration would keep the datafresh while providing educational avenues to show which communityattributes can be changed to help lower expected losses.

As the relevant data becomes more readily available and asanalytic sophistication grows, insurers and their vendor-partnerscan develop refined statistical models to help both communities andinsurance operations staff better understand complex risk-qualityrelationships, permitting more accurate underwriting and rating ofindividual risks.

Understanding how individual community attributes affect loss isdifficult, but analytics experts are now using time-tested,objective metrics to evaluate communities and to determine how onecommunity measures up against another. A suitable mea-surementprogram can include the following:

• Close working relationship with local fire and buildingofficials
• Clear metrics that can be used to rank/score departmentsnationally
• Identification of attributes that contribute to measurement oflosses and mitigation
• Qualified staff with the necessary certifications and trainingto conduct objective surveys

Municipal fire and building-code authorities and other communityofficials would use such measurement processes to help administerimportant programs that evaluate components of communityinfrastructure, such as fire-hazard risk, compliance withup-to-date building codes and construction guidelines, and thegeospatial threats of flood damage from surrounding bodies ofwater.

Many communities strive to show they are safe places to live andconduct business. When communities actually are safer, citizens,policyholders, local businesses, and public authorities benefitfrom stronger, well-protected communities with reduced risks andlosses. The insurers benefit from more accurate riskassessment.

Predictive Modeling Insight
Experts in thefield of risk assessment make better judgments with betterinformation on both the subject at risk and the surrounding “riskecosystem” (e.g., adjacent community risks, local mitigationcapabilities, topography of land/water, prevailing weatherpatterns, and even catastrophe disposition to events caused bysociety or nature). Such risk-assessment data can deliver thehighest value when it surpasses the level of isolated, silo-basedinformation and reaches the level of actionable insight andknowledge for decisive action.

Today's risk-based modeling methods enhance ratemaking engineswith more and better data. Property underwriters have longrespected this reality and are expanding their awareness bycombining all exposures into their geographic riskconsiderations.

Some carriers have adjusted where they concentrate policies (orare in the process of doing so), while others have already losttheir competitive edge.

In the workers' compensation meltdown of just ten years ago,companies that resolved to forgo a losing market, maintainunderwriting discipline, and get their pricing right, survived tocompete another day.

In the Florida homeowners' marketplace, insurers are competingon risk selection as opposed to pricing. Large insurers areleaving, while the residual market facility has too many policies.In turn, start-up companies are able to cherry pick the best risks.These decisions are largely guided by catastrophe models as opposedto predictive models, but the concept is the same.

The knowledge of future likelihoods can permit more refined riskassessment for individual and aggregated pools of insurancecontracts. This predictive knowledge can allow even more accuraterisk-based pricing and targeted marketing approaches for insurers.Meanwhile, communities can leverage the same information andinsight to improve the quality of life and the safety of theircitizenry.

Advanced statistical methods, cross-industry experts and datasets, text mining, and geospatial information have all contributedto a new breed of data components for modeling loss costs at eachrisk address. In addition to smarter underwriting, new and emerginganalytic methods optimize customer-lifetime-value portfoliomanagement, corroborating that customer-centric approaches can bethe best form of innovation and value generation.

A key driver of decision-making action for insurers and theirpolicyholders is predictive modeling, the process of incorporatingmultiple building blocks of contextual information into a frameworkof statistical analysis. With predictive analytics, compoundelements of risk data can be better observed in relation to lossfrequency and loss severity. Also, more intelligent insights aboutwhat may occur can be generated. Increased accuracy in forecastingaggregate losses on portfolios of risks is what makes predictiveanalytics so useful today.

The improvements under way in obtaining better data andincreasing the quantity of data — in conjunction with moresophisticated modeling technologies — are the factors that willkeep predictive analytics relevant tomorrow.

As the insurance industry continues to make progress in thedeployment of analytics throughout the enterprise, historycontinues to favor the most informed underwriters. For thosecompanies using analytics effectively, their shareholders andcustomers are rewarded with value, and their markets and regulatoryenvironments are endowed with greater stability.

Ranked Risks
A keen understanding of accurate pricing is an obvious competitiveadvantage, but there is a continuum of possible outcomes to everydistribution of ranked risks. Rating plans using predictiveanalytics can assign a unique price to any risk, turning pricinginto an even more potent underwriting tool. Depending on thestrength of a company's underwriting strategies, traditionalinsurance plans with few pricing points can result in long-termprofitability at one end of the continuum or invite adverse resultsand even outright financial ruin at the other end. This isgenerically the same for all lines of business and for everyindustry — proper segmentation of customers and their needs andcosts is essential to a healthy company, and the healthiestcompanies continually analyze data to improve themselves beyondtheir internal benchmarks as well as external competitiveindices.

Carriers caught in the transition from traditional to modernmethods of pricing and ratemaking and in the general adoption ofanalytics as a competitive reality find themselves either gleefulor anxious, based on the successes of their investments inanalytics. Those who waited to see what can happen are now buyingtheir way into the game via experienced consultants. But thelatecomers now need immediate execution of their informationtechnology and business process departmental functions because theyno longer have time to determine those elements by trial and error.Errors in execution are now costlier than ever. The pace ofimprovement and innovation in predictive modeling and theavailability of new data sources continue to accelerate.

Advanced analytics play a critical role in the underwritingpractices of the modern insurance enterprise. New risk-basedmodeling methods enhance ratemaking engines and improve riskmanagement decision-making across all lines of business. Carriersmust take advantage of emerging analytic methods to optimizecustomer service and the lifetime value of policyholders in theirbooks of business. Those carriers who sit and wait will soonexperience the negative effects of the cost of doing nothing. G

Kevin B. Thompson is senior vice president of ISO'sInsurance Services Department. Marty Ellingsworth is president ofISO Innovative Analytics (IIA), a unit of ISO focused on deliveringadvanced predictive analytic tools to the property/casualtyinsurance industry. More information is available at www.iso.com.