Dwelling Fire, HomeownersOwner-Occupied, andHomeowners Tenant andCondominium/CooperativeUnit Owners Insurance:Data for 2013Market Distribution andAverage Cost by Policy Formand Amount of Insurance

2013 Dwelling Fire, HomeownersOwner-Occupied, and HomeownersTenant and Condominium/CooperativeUnit Owner’s InsuranceJanuary 2016

The NAIC is the authoritative source for insurance industry information. Our expert solutions support the effortsof regulators, insurers and researchers by providing detailed and comprehensive insurance information. TheNAIC offers a wide range of publications in the following categories:Accounting & ReportingInformation about statutory accounting principles and theprocedures necessary for filing financial annual statementsand conducting risk-based capital calculations.Special StudiesStudies, reports, handbooks and regulatory researchconducted by NAIC members on a variety of insurancerelated topics.Consumer InformationImportant answers to common questions about auto,home, health and life insurance — as well as buyer’sguides on annuities, long-term care insurance andMedicare supplement plans.Statistical ReportsValuable and in-demand insurance industry-wide statisticaldata for various lines of business, including auto, home,health and life insurance.Financial RegulationUseful handbooks, compliance guides and reports onfinancial analysis, company licensing, state auditrequirements and receiverships.LegalComprehensive collection of NAIC model laws, regulationsand guidelines; state laws on insurance topics; and otherregulatory guidance on antifraud and consumer privacy.Market RegulationRegulatory and industry guidance on market-relatedissues, including antifraud, product filing requirements,producer licensing and market analysis.Supplementary ProductsGuidance manuals, handbooks, surveys and researchon a wide variety of issues.Capital Markets & Investment AnalysisInformation regarding portfolio values and procedures forcomplying with NAIC reporting requirements.White PapersRelevant studies, guidance and NAIC policy positions ona variety of insurance topics.For more information about NAICpublications, view our online catalog at:NAIC ActivitiesNAIC member directories, in-depth reporting of stateregulatory activities and official historical records ofNAIC national meetings and other activities. 2016 National Association of Insurance Commissioners. All rights reserved.ISBN: 978-1-59917-886-8Printed in the United States of AmericaNo part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic ormechanical, including photocopying, recording, or any storage or retrieval system, without written permission from the NAIC.NAIC Executive Office444 North Capitol Street, NWSuite 700Washington, DC 20001202.471.3990NAIC Central Office1100 Walnut StreetSuite 1500Kansas City, MO 64106816.842.3600NAIC Capital Markets& Investment Analysis OfficeOne New York Plaza, Suite 4210New York, NY 10004212.398.9000

Note: Information which is copyrighted by and proprietary to Insurance Services Office, Inc. (ISOMaterial) is included in this publication. Use of the ISO Material is limited to ISO Participating Insurersand their Authorized Representatives. Use by ISO Participating Insurers is limited to use in thosejurisdictions for which the insurer has an appropriate participation with ISO. Use of the ISO Material byAuthorized Representatives is limited to use solely on behalf of one or more ISO Participating Insurers.

Dwelling Fire, Homeowners Owner-Occupied, and Homeowners Tenant andCondominium/Cooperative Unit Owners Insurance: Data for 2013Table of ContentsPurpose of Report . 1Data. 1Limitations on the Data . 2Residual Market Data . 3Policy Forms/Types . 3Analysis of the Data . 4Factors Affecting the Cost of Insurance . 8Geographic Area, Real Estate and Construction Costs . 8Catastrophe Exposure . 8Defective Drywall . 12Mold Damage . 12Terrorism . 13Other Variables . 13Summary. 14Table 1: 2013 House-Years by State and Countrywide by Policy Type . 15Table 2: 2013 House-Years by State and Countrywide by Policy Form . 19Table 3: 2013 House-Years by State and Countrywide by Amount of Insurance . 23Table 4: 2013 Average Premium by Policy Form by Amount of InsuranceDwelling Fire and Homeowners Owner-Occupied Policy Forms . 31Table 5: 2013 Average Premium by Policy Form by Amount of InsuranceHomeowners Tenants and Condominium/Cooperative Unit Policy Forms . 137 2016 National Association of Insurance Commissioners

2016 National Association of Insurance Commissioners

Dwelling Fire, Homeowners Owner-Occupied, and Homeowners Tenant andCondominium/Cooperative Unit Owners Insurance: Data for 2013Market Distribution and Average Cost byPolicy Form and Amount of InsurancePurpose of ReportThis report provides countrywide and state-specific premium and exposure information for noncommercial dwelling fire insurance and for homeowners insurance package policies.Homeowners package policy data are for the homeowners owner-occupied policy forms (HO-1,HO-2, HO-3, HO-5 and HO-8), the tenant policy (HO-4) and the condominium/cooperative unitowner’s policy (HO-6). This narrative describes the data and discusses the way economic,demographic and natural phenomena impact the price of homeowners insurance.DataData consist of written exposures, expressed as house-years, 1 and aggregate written premiums bystate and countrywide for the 2013 data year. Premium and exposure information was collectedfor all states and the District of Columbia. The data are displayed in five tables. Three tablesshow individual state and countrywide exposures grouped by 1) policy type; 2) individual policyform; and 3) amount of insurance coverage, divided into ranges, with percentages of totalexposures provided. The last two tables display by-state and countrywide average premiums.Average premiums are calculated by dividing premiums by exposures for each policy form andrange of insurance coverage, and represent the cost of a year of coverage. Percentages of totalsare provided.Policy forms included in the report are described in detail in the following section. The ranges ofinsurance amounts extend to higher levels of coverage for the dwelling fire and homeownersowner-occupied policy forms than those for the tenant and condominium insurance, becausepremiums for the latter two policy forms do not include coverage for the residential structure.To the extent that data are reported to statistical agents, data for statutorily established FAIRplans are included.This report is not comprehensive of all statistical agents as there may be smaller statistical agentsthat do not report data to the NAIC. The data is limited to what was provided by the followingcompanies, whose cooperation and assistance in compiling this report is greatly appreciated: Data for all states, except Texas and California, were provided by the AmericanAssociation of Insurance Services; ISO Data, Inc.; the National Independent StatisticalService; and Independent Statistical Service, Inc. (ISS).1One house-year represents policy coverage on a dwelling for 12 months. Most often, it consists of coverage for onepolicy for an entire 12-month period, but it may also represent a number of policies for which the combined lengthsof coverage total 12 months (e.g., four policies, each with three months of coverage). 2016 National Association of Insurance Commissioners1

Texas data were obtained from the Texas Department of Insurance. Historically, theTexas department developed its own home insurance policy forms that are similar, butnot identical, to homeowners policy forms countrywide. Although, starting in 2002,insurers were permitted to file their own independent forms, some companies continue touse the forms previously promulgated by the department. California data were provided by the California Department of Insurance. The state begancollecting homeowners data in 1998, starting with the 1996 data year.Limitations on the DataAverage premium is an imperfect measure of the relative “price” of insurance due to widevariations in hazards, economic conditions and real estate values from state to state. Even whencomparing identical policy forms and amounts of insurance, premiums for homeowners coveragecan differ dramatically across the country. These market differences are explored in more detailin a later section of this narrative. Premium for a homeowners policy is determined by theamount of insurance purchased (generally based on the value of the insured property), the typesof property covered, the types of perils covered, and the specific limits and deductibles apolicyholder chooses.Averages developed for this report reflect all of these variables and more. For each state, somegeneral assumptions can be made about the types of insurance policies sold, the value of propertyinsured, and policyholders’ cost for loss protection for residential property and personal belongings.The premium and exposure data Texas reports to the NAIC include the premiums but not theunderlying exposures for the Texas Windstorm Insurance Association (TWIA), Texas’ coastalwind insurer of last resort. TWIA writes exclusively “wind-only” policies, and only collectsbasic information about policy type of the underlying “ex-wind” homeowners policy. Theoverwhelming majority of homeowners premium reported by TWIA is coded as “HO-3” (Texas“HO-B”); this is significantly different from the distribution of policies in the coastal region ofTexas. Therefore, it is highly likely that a significant portion of TWIA’s premiums are reportedas “HO-3” when the underlying exposures are reported as “HO-1,” “HO-2,” or “HO-5.” Thus,the average “HO-3” premium for Texas is artificially high as a result.Other Southeastern states have wind pools in operation that similarly may not be included in thisreport. 2016 National Association of Insurance Commissioners2

Residual Market DataBeginning with the 2011 data year, this report has sought to include data directly from residualmarket mechanisms. These mechanisms serve as a state’s insurer of last resort and/or to writehigh-risk policies, such as wind-only supplemental policies. When possible, complete policiesand supplemental policies were either categorized as the form number associated with theunderlying policy or in accordance to the residual market mechanism’s criteria. If the underlyingpolicy was unavailable, the supplemental policies were marked as HO-3 policies, the mostcommon owner-occupied policy form. For the supplemental policies, premium has beenincluded, but exposures were excluded. This was done to avoid double-counting, as it is likelythe underlying primary policy, and its exposure, has already been included in the report.The cooperation and assistance of the residual market data providers in compiling this reportwere greatly appreciated. Data were provided by the Alabama Insurance UnderwritingAssociation; Florida Citizens Property Insurance Corporation; Louisiana Citizens PropertyInsurance Corporation; Massachusetts Property Insurance Underwriting Association; MichiganBasic Property Insurance Association; Mississippi Windstorm Underwriting Association;Mississippi Residential Property Insurance Underwriting Association; New Jersey InsuranceUnderwriting Association; North Carolina Joint Underwriting Association; Ohio FAIR PlanUnderwriting Association; Rhode Island Joint Reinsurance Association; and South CarolinaWind and Hail Underwriting Association.Policy Forms/TypesData for this report were collected for eight policy forms that are grouped into three broadcategories (policy types) for comparison purposes.Dwelling fire policy (one family, owner-occupied, non-seasonal buildings)Under a dwelling fire policy, an insured purchases individual coverages on an a la carte basis;i.e., fire perils separately from extended coverage perils, and coverage for buildings separatelyfrom outbuildings and contents. Only the data for fire coverage for single-family owner-occupieddwellings are included in this report. Thus, the dwelling fire data (indicated by “DW” in thereport) are not directly comparable to the homeowners data, but are presented to provide anestimate of the cost for insurance purchased under the dwelling fire program.Homeowners package policies for owner-occupied dwellings (1-4 family units) HO-1: Basic “named-perils” 2 coverage on buildings and personal property.HO-2: Broad “named-perils” coverage on buildings and personal property; providescoverage for more perils than HO-1 package.HO-3: Provides “all-risks” 3 coverage on buildings, broad named-peril coverage onpersonal property; most common package written.HO-5: Provides “all-risks” coverage on buildings and personal property.2Insures against any loss incurred by the insured due to a peril named in the policy (e.g., fire, lightning, hail, etc.).Insures against risks of direct loss, except losses specifically stated in the policy as excluded from coverage(e.g., flood).3 2016 National Association of Insurance Commissioners3

HO-8: Repair cost coverage for a dwelling whose replacement cost greatly exceeds its marketvalue. Personal property, theft and additional coverages provided are similar to coveragesprovided under an HO-1 policy.Homeowners package policies for tenants, condominium and cooperative unit owners HO-4 (Renter’s Insurance): Broad “named-perils” coverage for the personal property oftenants.HO-6 (Condo/Co-op Insurance): Broad “named-perils” coverage for personal property ofcondominium or cooperative unit owners, as well as certain building items in which the unitowner may have an insurable interest.Homeowners owner-occupied policies represent a “package” of coverages for buildings, contents andliability. Accordingly, in each coverage range, the average premium for the dwelling fire policy representsless coverage than the corresponding homeowners policies. Homeowners tenants and condominiumpolicies are similar to homeowners owner-occupied policies with respect to covered perils, contentscoverage and liability. However, there is no building coverage other than the condo/co-op owner’sinsurable interest.Analysis of the DataTable 1 provides exposure data in house-years by policy type. The table shows a countrywide total of81,250,810.7 house-years. In 2013, homeowners owner-occupied policy exposures accounted for 75.9percent of overall exposures countrywide. Tenant and condominium policy exposures accounted for 21.9percent of the total, while dwelling fire exposures made up the remaining 2.2 percent.Exposure data for the eight individual policy forms is provided in Table 2. The HO-3 accounted for 61.8percent of all policy exposures and remains the most common policy sold by far. Figure 1 (next page)shows the percentage breakdown of exposures for the homeowners owner-occupied policy forms only.Countrywide, 81.5 percent of these exposures were written on the HO-3 form.Figure 2 shows the percentage breakdown of countrywide exposures for the tenant and condo/co-op policyforms. Of these, 69.8 percent were written on the HO-4 form. 2016 National Association of Insurance Commissioners4

Figure 1 - 2013 Percent of Homeowners Owner-OccupiedWritten ExposuresCountrywide By Policy %Figure 2 - 2013 Percent of Tenant and Condominium/Co-op Written ExposuresCountrywide By Policy FormHO-6(Condo/Co-op)30.16%HO-4(Tenants)69.84% 2016 National Association of Insurance Commissioners5

Tables 3A and 3B present countrywide and by-state exposure data divided between each of the ranges ofinsurance coverage amount. Dwelling fire policy data and data for the homeowners owner-occupiedpolicy forms are grouped together in Table 3A, and data for the HO-4 and HO-6 forms are groupedtogether in Table 3B. Countrywide, in 2013, 69.7 percent of dwelling fire and homeowners owneroccupied policies were written for insurance coverage amounts between 50,000 and 300,000.Tenant and condominium policies do not provide coverage for the building; therefore, the distribution ofexposures for these types of policies is concentrated at significantly lower insurance amounts. Table 3Bshows that 59.8 percent of the exposures for the HO-4 and HO-6 forms are concentrated at amounts below 32,000, and 88.5 percent of these policies provide less than 75,000 in coverage.Table 4, 4 which provides premium and exposure percentages by form, also groups data for the dwellingfire and the homeowners owner-occupied forms. Note that for the homeowners owner-occupied forms, thenumber of exposures in the “ 24,999 and under” range was determined to be too low to producemeaningful average premium results. Data for this range was, therefore, combined with data in the“ 25,000– 49,999” range, and an “ 49,999 and under” range was created. An average premium for eachpolicy form was then calculated for the combined range.Figure 3 provides a comparison of dwelling fire and the five homeowners owner-occupied policyexposures by amounts of insurance coverage. Dwelling fire exposures represent 2.8 percent of totalexposures and are most prevalent at insurance coverage amounts less than 50,000. In the less than 50,000 range, dwelling fire exposures account for 43.4 percent of the total, then drop to 22.6 percent atinsurance amounts of 50,000– 74,999. At coverage amounts above 75,000, dwelling fire exposuresaccount for no more than 7.9 percent of the total for each range.Figure 4 compares HO-4 and HO-6 policy forms by coverage amounts. Countrywide, the HO-4 (tenants)form represents more policies written at lower coverage amounts. At coverage amounts above 44,000,the majority of exposures are written on the HO-6 (condo/co-op) form.Tables 4 and 5 display state average premiums for each policy form. Examining the countrywide averagepremium data for dwelling fire and homeowners owner-occupied policies reveals some expected results.In general, the average premium increases as the amount of coverage increases for all policy types.Dwelling fire premiums are generally lower when compared to the five homeowners premiums, reflectingthe more limited coverage offered by dwelling fire policies compared to homeowners packages. Anincrease from prior years in HO-8 average premium has been noted. All data related to this increase hasbeen confirmed as accurate.4Negative and zero exposures and premiums are denoted by an asterisk in Table 4. All exposure and premium amounts,including those that are negative, are included within the Table 4 totals. 2016 National Association of Insurance Commissioners6

Figure 3 - 2013 Percent Comparison of Dwelling Fire andHomeowners Owner-Occupied Written 20.0%10.0%0.0%HO Owner-OccupiedDwelling FireAmount of InsuranceFigure 4 - 2013 Percentages of Tenant and CondominiumWritten Exposures by Amount of Insurance100%90%80%70%60%50%40%30%20%10%0%Amount of Insurance 2016 National Association of Insurance Commissioners7HO-6 (Condo/Co-op)HO-4 (Tenants)

Factors Affecting the Cost of InsuranceGeographic Area, Real Estate and Construction CostsMany factors impact the cost of home insurance, resulting in large differences in average premiumsthroughout the United States. In general, real estate values and construction costs tend to be higher inareas of greater population density. Because the amount of home insurance needed is based on the valueof the home, premiums are often higher in more heavily populated places. Vacation and retirement areas,as well as areas experiencing rapid economic growth, also tend to have relatively higher real estate values.Construction costs vary based on the type of residence, availability of building materials and factors suchas local climate and building regulations. Higher expected repair costs for value-added designs to reducedamages to the structure from earthquakes or hurricanes will impact the price of insurance. As shown inthe following maps, these variations in costs are reflected in the range of median amounts of insurancepurchased throughout the United States.Catastrophe ExposureDegree of exposure to catastrophe affects the cost of insurance to homeowners. Brush and forest fires,tornadoes, high winds, hail, freezing rain, snow storms, hurricanes, earthquakes, riots and even terroristattacks are all types of catastrophes that can occur in the United States. Every place in the world has anexposure to some type of catastrophe, but some areas are more prone to certain types. Brush and forestfires are more common in the West. Hurricane exposure is greater in areas near the Gulf of Mexico andthe Atlantic Ocean. Exposure to tornado damage is greatest in the central and southwestern United States,even though tornadoes can and do occur in nearly every state. Earthquake exposure also exists throughoutthe country because seismic faults are located in all regions. 5 Terrorist attacks also are not specific to anygeographic area, but have typically occurred in larger urban areas.5Although earthquake coverage is commonly endorsed onto a homeowners insurance policy, premiums for earthquakecoverage are not included in the data. 2016 National Association of Insurance Commissioners8

2013 Homeowners Median Amount of Insurance2013 Dwelling Fire Median Amount of Insurance 2016 National Association of Insurance Commissioners9

Since the late 1980s, catastrophes have been occurring with greater frequency and severity, andare a significant consideration in the pricing of home insurance. Until 1996, the Property ClaimsServices (PCS) unit of the Insurance Services Office considered an event a catastrophe wheninsured losses totaled 5 million or more. Beginning in 1997, the PCS no longer considered anevent a “catastrophe” unless it resulted in insured losses that totaled 25 million or more. For theperiod 2005–2014, the total insured losses for U.S. catastrophes (in 2014 dollars) were more than 247.4 billion.6The following table shows, in descending order of loss, the 10 most costly insured property U.S.catastrophes through 2014. 7 Four of these were hurricanes that occurred during 2004 and 2005,including Hurricane Katrina. Most of the events listed below occurred after 2003.RankDatePeril12345678910Aug. 2005Sep. 2001Aug. 1992Oct. 2012Jan. 1994Sep. 2008Oct. 2005Aug. 2004Sep. 2004Apr. 2011Hurricane KatrinaWorld Trade Center, Pentagon Terrorist AttacksHurricane AndrewSuperstorm SandyNorthridge, CA EarthquakeHurricane IkeHurricane WilmaHurricane CharleyHurricane IvanTuscaloosa, AL TornadoDollars whenOccurred (millions) 107,300In 2014 Dollars(millions)* 397,652Property coverage only. Does not include flood damage covered by the federally administered National Flood Insurance Program.*Adjusted for inflation through 2014 by ISO using the GDP implicit price deflator.The Federal Emergency Management Agency (FEMA) definition of disaster includes chemicalemergencies, dam failure, earthquake, fire, wildfire, flood, hazardous material, heat, hurricane,landslide, nuclear power plant emergency, terrorism, thunderstorm, tornado, tsunami, volcano,wildfire, and winter storm. The following table lists the number of declared disasters, by stateand year. 86Insurance Information Institute, statistics/catastrophes-us.html; includes copyrighted material ofInsurance Services Office, Inc., with its permission.7Ibid.8Federal Emergency Management Agency (FEMA),,as of 11/25/2015. 2016 National Association of Insurance Commissioners10

2011-2015 Disaster orniaColoradoConnecticutDelawareDistrict of sMichiganMinnesotaMississippiMissouriGrand TotalDisaster 1422222303232015 201440 45201362201247201199StateMontanaNebraskaNevadaNew HampshireNew JerseyNew MexicoNew YorkNorth CarolinaNorth DakotaOhioOklahomaOregonPennsylvaniaRhode IslandSouth CarolinaSouth est 201112035242214231015224310113*Other - American Samoa, Federated States of Micronesia, Guam, Northern Mariana Islands, Puerto Rico, US Virgin Islands, & Tribal GovernmentsThe number of tornadoes averaged 1,202 per year during the 2005-2014 time period, but thenumber in individual years ranged from a low of 888 in 2014 to a high of 1,692 in 2008. 9 Thesefigures serve to emphasize the variability and unpredictability of catastrophe losses. The impactthat various catastrophes have on rates from state to state must be considered in any evaluation ofaverage premiums.9Insurance Information Institute, statistics/tornadoes-and-thunderstorms.html; U.S. Departmentof Commerce; Storm Prediction Center; National Weather Service. 2016 National Association of Insurance Commissioners11

Insurers use computer models to estimate the potential cost of catastrophic events, particularly inthe absence of a sufficient amount of relevant loss experience. For example, the potential insuredloss in the New Madrid region due to an earthquake is predicted to be significant, but the factthat the last major earthquake in that area occurred in 1812—when there were considerablyfewer people and buildings—makes it difficult to rely on previous experience to accurately priceearthquake coverage in the area.Computer models use insurer exposure and loss experience; geological, meteorological andseismic data; structural engineering and construction data; and other applicable information tosimulate catastrophes in a specific region to more accurately estimate the cost of paying forlosses that could occur. The methodology used by a catastrophe-modeling firm is typicallyconsidered proprietary information, and insurance regulators have no direct authority over themodelers. However, when a rate filing relies in part on a computer model, an insurancedepartment may require an insurer or rating organization to provide supplemental informationpertaining to the model’s input data to determine whether the filing meets the requirements of thestate insurance laws.Defective DrywallDue to the extensive damage to homes and buildings by the unusual number of hurricanes 10 thatstruck the United States between 2004 and 2008, the amount of drywall needed to repairdamages was not readily available in the United States; therefore, drywall was imported into theUnited States from China. Indicators of the existence of defective drywall normally appearroughly 24 months after the installation of the product. Complaints regarding defective drywallbegan to emerge in the summer of 2008. The initial complaints came primarily from individualsliving in and around the southern region of Florida. The problems with the defective drywallinclude the smell of rotten eggs; failure of air conditioning equipment; and corrosion of pipes,wiring, furniture, fixtures and jewelry. It also should be noted that the corrosion of electricalwiring can hamper the effectiveness of smoke detection and can create a risk of fire. Manysamples of the defective drywall have been tested. The testing of these samples detected threevolatile sulfur-containing compounds: hydrogen sulfide, carbonyl sulfide and carbon disulfide.Health problems—including irritated eyes and skin, asthma attacks, sinus infections, bloodynoses, headaches and persistent cough—also have been reported. 11Although evidence suggests that there are problems associated with certain imported drywallused in the building of homes, it is too early to tell whether the cost of homeowners coveragewill be affected.Mold DamageHigh insurance claim payments for mold damage garnered the attention of media outlets severalyears ago. This can be attributed to higher numbers of claims filed, as well as some extraordinaryamoun

Texas data were obtained from the Texas Department of Insurance.Historically, the Texas department developed its own home insurance policy forms that are similar, but not identical, to homeowners policy forms countrywide. Although, starting in 2002, insurers were permitted to file their own independent forms, some companies continue to