Elawyers Elawyers
Ohio| Change
Find Similar Cases by Filters
You can browse Case Laws by Courts, or by your need.
Find 49 similar cases
HARTFORD ACCIDENT AND INDEMNITY COMPANY, HARTFORD INSURANCE COMPANY OF THE SOUTHEAST, HARTFORD FIRE INSURANCE COMPANY, TWIN CITY INSURANCE COMPANY, AND HARTFORD CASUALTY INSURANCE COMPANY vs OFFICE OF INSURANCE REGULATION, 07-005188 (2007)
Division of Administrative Hearings, Florida Filed:Tallahassee, Florida Nov. 09, 2007 Number: 07-005188 Latest Update: Jun. 03, 2008

The Issue Whether Petitioners' proposed rates are justified pursuant to the requirements of Section 627.062, Florida Statutes, or whether the Department of Financial Services, Office of Insurance Regulation (OIR) was correct in denying the requested rate increases.

Findings Of Fact The Hartford companies are property and casualty insurers transacting insurance in the State of Florida pursuant to valid certificates of authority and the Florida Insurance Code. Two types of personal lines insurance filings submitted by Hartford to the OIR are at issue in this proceeding: two filings for homeowners insurance (Case Nos. 07-5185 and 07-5186) and two filings for dwelling fire insurance (Case Nos. 07-5187 and 07- 5188). Hartford's substantial interests are affected by the notices disapproving the filings in this case. Homeowners insurance includes coverage for a variety of perils in and around a home, is usually purchased by a homeowner, and covers both the structure and the contents of a home. Dwelling/fire insurance is usually purchased by the owners of properties that are leased or rented to others, and provides coverage for the structure only. Both types of insurance cover damage caused by hurricanes. The New Legislation and its Requirements In a special session held in January 2007, the Florida Legislature enacted changes to the Florida Hurricane Catastrophe Fund (CAT Fund), as reflected in Chapter 2007-1, Laws of Florida. The special session was precipitated by a perceived crisis regarding the cost and availability of homeowners insurance after the 2004 and 2005 hurricane seasons. As a result of the substantial number of claims incurred after multiple severe hurricanes each of these years, changes in the insurance marketplace resulted in some insurance companies withdrawing from the Florida market, others non-renewing policies, one company becoming insolvent, and the cost for reinsurance available to all insurers rising dramatically. One of the primary features of the legislation was an expansion of the CAT Fund. The CAT Fund was established in 1993 after Hurricane Andrew to provide reinsurance to insurers for property insurance written in Florida at a price significantly less than the private market. The CAT Fund is a non-profit entity and is tax exempt. Prior to the enactment of Chapter 2007-1, the CAT Fund had an industry-wide capacity of approximately $16 million. The purpose of the changes enacted by the Legislature was to reduce the cost of reinsurance and thereby reduce the cost of property insurance in the state. As a result of Chapter 2007-1, the industry-wide capacity of the CAT Fund was increased to $28 billion, and insurers were given an opportunity to purchase an additional layer of reinsurance, referred to as the TICL layer (temporary increase in coverage limit), from the CAT Fund. Section 3 of Chapter 2007-1 required insurers to submit a filing to the OIR for policies written after June 1, 2007, that took into account a "presumed factor" calculated by OIR and that purported to reflect savings created by the law. The new law delegated to the OIR the duty to specify by Order the date such filings, referred to as "presumed factor filings" had to be made. On February 19, 2007, the OIR issued Order No. 89321-07. The Order required insurers to make a filing by March 15, 2007, which either adopted presumed factors published by the OIR or used the presumed factors and reflected a rate decrease taking the presumed factors into account. The presumed factors were the amounts the OIR calculated as the average savings created by Chapter 2007-1, and insurers were required to reduce their rates by an amount equal to the impact of the presumed factors. The OIR published the presumed factors on March 1, 2007. In its March 15, 2007, filings, Hartford adopted the presumed factors published by OIR. As a result, Hartford reduced its rates, effective June 1, 2007, on the products at issue in these filings by the following percentages: Case No. 07-5185 homeowners product: 17.7% Case No. 07-5186 homeowners product: 21.9% Case No. 07-5187 dwelling/fire product: 8.7% Case No. 07-5188 dwelling/fire product: 6.2% The Order also required that insurers submit a "True-Up Filing" pursuant to Section 627.026(2)(a)1., Florida Statutes. The filing was to be a complete rate filing that included the company's actual reinsurance costs and programs. Hartford's filings at issue in these proceedings are its True-Up Filings. The True-Up Filings Hartford submitted its True-Up filings June 15, 2007. The rate filings were certified as required by Section 627.062(9), Florida Statutes. The filings were amended August 8, 2007. Hartford's True Up Filings, as amended, request the following increases in rates over those reflected in the March 15, 2007, presumed factor filings: Case No. 07-5185 homeowners product: 22.0% Case No. 07-5186 homeowners product: 31.6% Case No. 07-5187 dwelling and fire product: 69.0% Case No. 07-5188 dwelling and fire product: 35.9% The net effects of Hartford's proposed rate filings result in the following increases over the rates in place before the Presumed Factor Filings: Case No. 07-5185 homeowners product: .4% Case No. 07-5186 homeowners product: 2.8% Case No. 07-5187 dwelling/fire product: 54.3% Case No. 07-5188 dwelling/fire product: 27.5% Case Nos. 07-5185 and 07-5186 (homeowners) affect approximately 92,000 insurance policies. Case Nos. 07-5187 and 07-5188 (dwelling/fire) affect approximately 2,550 policies. A public hearing was conducted on the filings August 16, 2007. Representatives from Hartford were not notified prior to the public hearing what concerns the OIR might have with the filings. Following the hearing, on August 20, 2007, Petitioners provided by letter and supporting documentation additional information related to the filings in an effort to address questions raised at the public hearing. The OIR did not issue clarification letters to Hartford concerning any of the information provided or any deficiencies in the filings before issuing its Notices of Intent to Disapprove the True-Up Filings. All four filings were reviewed on behalf of the OIR by Allan Schwartz. Mr. Schwartz reviewed only the True-Up Filings and did not review any previous filings submitted by Hartford with respect to the four product lines. On September 10, 2007, the OIR issued Notices of Intent to Disapprove each of the filings at issue in this case. The reasons give for disapproving the two homeowners filings are identical and are as follows: Having reviewed the information submitted, the Office finds that this filing does not provide sufficient documentation or justification to demonstrate that the proposed rate(s) comply with the standards of the appropriate statute(s) and rules(s) including demonstrating that the proposed rates are not excessive, inadequate, or unfairly discriminatory. The deficiencies include but are not limited to: The premium trends are too low and are not reflective of the historical pattern of premium trends. The loss trends are too high and are not reflective of the historical pattern of loss trends. The loss trends are based on an unexplained and undocumented method using "modeled" frequency and severity as opposed to actual frequency and severity. The loss trends are excessive and inconsistent compared to other sources of loss trends such as Fast Track data. The catastrophe hurricane losses, ALAE and ULAE amounts are excessive and not supported. The catastrophe non-hurricane losses, ALAE and ULAE amounts are excessive and not supported. The particular time period from 1992 to 2006 used to calculate these values has not been justified. There has been no explanation of why the extraordinarily high reported losses for 1992 and 1993 should be expected to occur in the future. The underwriting profit and contingency factors are excessive and not supported. Various components underlying the calculation of the underwriting profit and contingency factors, including but not limited to the return on surplus, premium to surplus ratio, investment income and tax rate are not supported or justified. The underwriting expenses and other expenses are excessive and not supported. The non-FHCF reinsurance costs are excessive and not supported. The FHCF reinsurance costs are excessive and not supported. The fact that no new business is being written has not been taken into account. No explanation has been provided as too [sic] Hartford believes it is reasonable to return such a low percentage of premium in the form of loss payments to policyholders. For example, for the building policy forms, only about 40% of the premium requested by Hartford is expected to be returned to policyholders in the form of loss payments. As a result of the deficiencies set forth above, the Office finds that the proposed rate(s) are not justified, and must be deemed excessive and therefore, the Office intends to disapprove the above-referenced filing. The Notices of Intent to Disapprove the two dwelling/fire filings each list nine deficiencies. Seven of the nine (numbers 1-6 and 8) are the same as deficiencies listed for the homeowners filings. The remaining deficiencies named for Case No. 07-5187 are as follows: 7. The credibility standard and credibility value are not supported. 9. No explanation has been provided as too (sic) why Hartford believes it needs such a large rate increase currently, when the cumulative rate change implemented by Hartford for this program from 2001 to 2006 was an increase of only about 10%. The deficiencies listed for Case No. 07-5188 are the same as those listed for Case No. 07-5187, with the exception that with respect to deficiency number 9, the rate change implemented for the program in Case No. 07-5188 from 2001 to 2006 was a decrease of about -3%. Documentation Required for the Filings Florida's regulatory framework, consistent with most states, requires that insurance rates not be inadequate, excessive, or unfairly discriminatory. In making a determination concerning whether a proposed rate complies with this standard, the OIR is charged with considering certain enumerated factors in accordance with generally accepted and reasonable actuarial techniques. Chapter 2007-1 also amended Section 627.062, Florida Statutes, to add a certification requirement. The amendment requires the chief executive officer or chief financial officer and chief actuary of a property insurer to certify under oath that they have reviewed the rate filing; that to their knowledge, the rate filing does not contain any untrue statement of a material fact or omit to state a material fact necessary to make the statements made, in light of the circumstances under which the statements were made, not misleading; that based on their knowledge, the information in the filing fairly presents the basis of the rate filing for the period presented; and that the rate filing reflects all premium savings reasonably expected to result from legislative enactments and are in accordance with generally accepted and reasonable actuarial techniques. § 627.062(9)(a), Fla. Stat. (2007). Actuarial Standards of Practice 9 and 41 govern documentation by an actuary. Relevant sections of Standard of Practice 9 provide: Extent of documentation - . . . Appropriate records, worksheets, and other documentation of the actuary's work should be maintained by the actuary and retained for a reasonable length of time. Documentation should be sufficient for another actuary practicing in the same field to evaluate the work. The documentation should describe clearly the sources of data, material assumptions, and methods. Any material changes in sources of data, assumptions, or methods from the last analysis should be documented. The actuary should explain the reason(s) for and describe the impact of the changes. Prevention of misuse - . . . The actuary should take reasonable steps to ensure that an actuarial work product is presented fairly, that the presentation as a whole is clear in its actuarial aspects, and that the actuary is identified as the source of the actuarial aspects, and that the actuary is available to answer questions.. . . . * * * 5.5 Availability of documentation- Documentation should be available to the actuary's client or employer, and it should be made available to other persons when the client or employer so requests, assuming appropriate compensation, and provided such availability is not otherwise improper. . . . In determining the appropriate level of documentation for the proposed rate filings, Petitioner relied on its communications with OIR, as well as its understanding of what has been required in the past. This reliance is reasonable and is consistent with both the statutory and rule provisions governing the filings. Use of the RMS Catastrophic Loss Projection Model In order to estimate future losses in a rate filing, an insurer must estimate catastrophic and non-catastrophic losses. Hartford's projected catastrophic losses in the filings are based upon information provided from the Risk Management Solutions (RMS) catastrophic loss projection model, version 5.1a. Hartford's actuaries rely on this model, consistent with the standards governing actuarial practice, and their reliance is reasonable. Catastrophe loss projection models may be used in the preparation of insurance filings, if they have been considered by and accepted by the Florida Commission on Hurricane Loss Projection Methodology (the Hurricane Commission). The Hurricane Commission determined that the RMS model, version 5.1a was acceptable for projecting hurricane loss costs for personal residential rate filings on May 17, 2006. In addition to approval by the Hurricane Commission, use of the model is appropriate "only if the office and the consumer advocate appointed pursuant to s. 627.0613 have access to all of the assumptions and factors that were used in developing the actuarial methods, principles, standards, models, or output ranges, and are not precluded from disclosing such information in a rate proceeding." §627.0628(3)(c), Fla. Stat. Both the Consumer Advocate and a staff person from the OIR are members of the Hurricane Commission. In that context, both have the ability to make on-site visits to the modeling companies, and to ask any questions they choose regarding the models. Both OIR's representative and the Consumer Advocate participated in the meetings and had the same opportunity as other commissioners to ask any question they wished about RMS 5.1a. The Hurricane Commission members, including the Consumer Advocate, clearly have access to the information identified in Section 627.0628(3)(c). However, there are restrictions on the Hurricane Commission members' ability to share the information received regarding trade secrets disclosed by the modeling companies. For that reason, the Commission's deliberations are not, standing alone, sufficient to determine that the Office of Insurance Regulation has access. In this case, credible evidence was submitted to show that RMS officials met with staff from the Office in July and October 2006 to discuss the model. RMS offered to provide any of its trade secret information to the OIR, subject to a non- disclosure agreement to protect its dissemination to competitors. RMS also opened an office in Tallahassee and invited OIR staff to examine any parts of the model they wished. In addition, both RMS and Hartford have answered extensive questionnaires prepared by OIR regarding the RMS model, and Hartford has offered to assist OIR in gathering any additional information it requires. Most of the questions posed by OIR involve the same areas reviewed by the Commission. RMS' representative also testified at hearing that RMS would not object to disclosure of the assumptions during the hearing itself if necessary. Finally, OIR Exhibit 1 is the Florida Hurricane Catastrophe Fund 2007 Ratemaking Formula Report. The Executive Summary from the report explains how rates were recommended for the Florida Hurricane Catastrophic Fund (CAT Fund) for the 2007- 2008 contract year. The report stated that the RMS model, as well as three other models accepted by the Hurricane Commission, were used for determining expected aggregate losses to the CAT Fund reinsurance layer. Three models, including the RMS model, were also used for analysis of detailed allocation to type of business, territory, construction and deductible, as well as special coverage questions. The models were compared in detail and given equal weight. The report notes that these three models were also used in 1999-2006 ratemaking. The report is prepared by Paragon Strategic Solutions, Inc., an independent consultant selected by the State Board of Administration, in accordance with Section 215.555(5), Florida Statutes. While OIR did not prepare the report, they show no hesitation in accepting and relying on the report and the modeled information it contains in these proceedings. Indeed, one of OIR's criticisms is Hartford's failure to use the report with respect to CAT Fund loss recovery estimates. Based upon the evidence presented at hearing, it is found that the OIR and Consumer Advocate were provided access to the factors and assumptions used in the RMS model, as contemplated by Section 627.0628. The Alleged Deficiencies in the Homeowners Filings1/ A rate is an estimate of the expected value of future costs. It provides for all costs associated with the transfer of risk. A rate is reasonable and not excessive, inadequate or unfairly discriminatory if it is an actuarially sound estimate of the expected value of all future costs associated with an individual risk transfer. In preparing a filing, an actuary identifies the time period that its proposed rates are expected to be in effect. Because ratemaking is prospective, it involves determining the financial value of future contingent events. For the rate filings in question, actuaries for Hartford developed their rate indications by first considering trended premium, which reflects changes in premium revenue based on a variety of factors, including construction costs and the value of the buildings insured. Trended premium is the best estimate of the premium revenue that will be collected if the current rates remain in effect for the time period the filing is expected to be in place. Expenses associated with writing and servicing the business, the reinsurance costs to support the business and an allowance for profit are subtracted from the trended premium. The remainder is what would be available to pay losses. This approach to ratemaking, which is used by Hartford, is a standard actuarial approach to present the information for a rate indication. As part of the process, expected claims and the cost to service and settle those claims is also projected. These calculations show the amount of money that would be available to pay claims if no changes are made in the rates and how much increased premium is necessary to cover claims. The additional amount of premium reflects not only claims payments but also taxes, licenses and fees that are tied to the amount of premium. The first deficiency identified by OIR is that "the premium trends are too low and are not reflective of the historical pattern of premium trends." In determining the premium trend in each filing, Hartford used data from the previous five years and fit an exponential trend to the historical pattern, which is a standard actuarial technique. Hartford also looked at the factors affecting the more recent years, which were higher. For example, the peak in premium trend in 2006 was a result of the cost increases driven by the 2004 and 2005 hurricanes, and the peak in demand for labor and construction supplies not matched by supply. Costs were coming down going into 2007, and Hartford believed that 2006 was out of pattern from what they could anticipate seeing in the future. The premium trends reflected in Hartford's filings are reasonable, reflective of historical patterns, and based on standard actuarial techniques. The second identified deficiency with respect to the homeowner filings was that the loss trends are too high and are not reflective of the historical pattern of loss trends. A loss trend reflects the amount an insurance company expects the cost of claims to change. It consists of a frequency trend, which is the number of claims the insurance company expects to receive, and a severity trend, which is the average cost per claim. The loss trend compares historical data used in the filing with the future time period when the new rates are expected to be in effect. Hartford's loss trends were estimated using a generalized linear model, projecting frequency and severity separately. The model was based on 20 quarters of historical information. The more credible testimony presented indicates that the loss trends were actuarially appropriate. The third identified deficiency is that the loss trends are based on an unexplained and undocumented method using "modeled" frequency and severity as opposed to actual frequency and severity. As noted above, the generalized linear model uses actual, historical data. Sufficient documentation was provided in the filing, coupled with Hartford's August 20, 2007, letter. The method used to determine loss trends is reasonable and is consistent with standard actuarial practice. The fourth identified deficiency is that loss trends are excessive and inconsistent compared to other sources of loss trends, such as Fast Track data. Saying that the loss trends are excessive is a reiteration of the claim that they are too high, already addressed with respect to deficiency number two. Fast Track data is data provided by the Insurance Services Office. It uses unaudited information and is prepared on a "quick turnaround" basis. Fast Track data is based on paid claims rather than incurred claims data, and upon a broad number of companies with different claims settlement practices. Because it relies on paid claims, there is a time lag in the information provided. Hartford did not rely on Fast Track data, but instead relied upon its own data for calculating loss trends. Given the volume of business involved, Hartford had enough data to rely on for projecting future losses. Moreover, Respondents point to no statutory or rule requirement to use Fast Track data. The filings are not deficient on this basis. The fifth identified deficiency in the Notice of Intent to Disapprove is that catastrophe hurricane losses, ALAE and ULAE amounts are excessive and not supported. ALAE stands for "allocated loss adjustment expenses," and represents the costs the company incurs to settle a claim and that can be attributed to that particular claim, such as legal bills, court costs, experts and engineering reports. By contrast, ULAE stands for "unallocated loss adjustment expense" and represents the remainder of claims settlement costs that cannot be linked to a specific claim, such as office space, salaries and general overhead. Part of the OIR's objection with respect to this deficiency relates to the use of the RMS model. As stated above at paragraphs 25-33, the use of the RMS model is reasonable. With respect to ALAE, Hartford analyzed both nationwide data (4.4%) and Florida data (4.8%) and selected an ALAE load between the two (4.6%). This choice benefits Florida policyholders. It is reasonable to select between the national and Florida historical figures, given the amount of actual hurricane data available during the period used. With respect to ULAE, the factors used were based upon directions received from Ken Ritzenthaler, an actuary with OIR, in a previous filing. The prior discussions with Mr. Ritzenthaler are referenced in the exhibits to the filing. The more credible evidence demonstrates that the ALAE and ULAE expenses with respect to catastrophic hurricane losses are sufficiently documented in Hartford's filings and are based on reasonable actuarial judgment. The sixth identified deficiency is that the catastrophe non-hurricane losses, ALAE and ULAE amounts are excessive and not supported. According to OIR, the particular time period from 1992 to 2006 used to calculate these values has not been justified, and there has been no explanation of why the extraordinarily high reported losses for 1992 and 1993 should be expected to occur in the future. OIR's complaint with respect to non-hurricane losses is based upon the number of years of data included. While the RMS model was used for hurricane losses, there is no model for non- hurricane losses, so Hartford used its historical data. This becomes important because in both 1992 and 1993, there were unusual storms that caused significant losses. Hartford's data begins with 1992 and goes through 2006, which means approximately fifteen years worth of data is used. Hartford's explanation for choosing that time period is that hurricane models were first used in 1992, and it was at that time that non-hurricane losses had to be separated from hurricane losses. Thus, it was the first year that Hartford had the data in the right form and sufficient detail to use in a rate filing. Petitioners have submitted rate filings in the past that begin non-hurricane, ALAE and ULAE losses with 1992, increasing the number of years included in the data with each filing. Prior filings using this data have been approved by OIR. It is preferable to use thirty years of experience for this calculation. However, there was no testimony that such a time-frame is actuarially or statutorily required, and OIR's suggestion that these two high-loss years should be ignored is not based upon any identified actuarial standard. Hartford attempted to mitigate the effect of the severe losses in 1992 and 1993 by capping the losses for those years, as opposed to relying on the actual losses.2/ The methodology used by Hartford was reasonable and appropriate. No other basis was identified by the OIR to support this stated deficiency. The seventh identified deficiency is that the underwriting profit and contingency factors are excessive and not supported. The underwriting profit factor is the amount of income, expressed as a percentage of premium, that an insurance company needs from premium in excess of losses, settlement costs and other expenses in order to generate a fair rate of return on its capital necessary to support its Florida exposures for the applicable line of business. Hartford's proposed underwriting profit factor for its largest homeowners filing is 15.3%. Section 627.062(2)(b), Florida Statutes, contemplates the allowance of a reasonable rate of return, commensurate with the risk to which the insurance company exposes its capital and surplus. Section 627.062(2)(b)4., Florida Statutes, authorizes the adoption of rules to specify the manner in which insurers shall calculate investment income attributable to classes of insurance written in Florida, and the manner in which investment income shall be used in the calculation of insurance rates. The subsection specifically indicates that the manner in which investment income shall be used in the calculation of insurance rates shall contemplate allowances for an underwriting profit factor. Florida Administrative Code Rule 69O-170.003 is entitled "Calculation of Investment Income," and the stated purpose of this rule is as follows: (1) The purpose of this rule is to specify the manner in which insurers shall calculate investment income attributable to insurance policies in Florida and the manner in which such investment income is used in the calculation of insurance rates by the development of an underwriting profit and contingency factor compatible with a reasonable rate of return. (Emphasis supplied). Mr. Schwartz relied on the contents of this rule in determining that the underwriting profit factor in Hartford's filings was too high, in that Florida Administrative Code Rule 69O-170.003(6)(a) and (7) specifies that: (6)(a) . . . An underwriting profit and contingency factor greater than the quantity 5% is prima facie evidence of an excessive expected rate of return and unacceptable, unless supporting evidence is presented demonstrating that an underwriting profit and contingency factor included in the filing that is greater than this quantity is necessary for the insurer to earn a reasonable rate of return. In such case, the criteria presented as determined by criteria in subsection (7) shall be used by the Office of Insurance Regulation in evaluating this supporting evidence. * * * An underwriting profit and contingency factor calculated in accordance with this rule is considered to be compatible with a reasonable expected rate of return on net worth. If a determination must be made as to whether an expected rate of return is reasonable, the following criteria shall be used in that determination. An expected rate of return for Florida business is to be considered reasonable if, when sustained by the insurer for its business during the period for which the rates under scrutiny are in effect, it neither threatens the insurer's solvency nor makes the insurer more attractive to policyholders or investors from a corporate financial perspective than the same insurer would be had this rule not been implemented, all other variables being equal; or Alternatively, the expected rate of return for Florida business is to be considered reasonable if it is commensurate with the rate of return anticipated for other industries having corresponding risk and it is sufficient to assure confidence in the financial integrity of the insurer so as to maintain its credit and, if a stock insurer, to attract capital, or if a mutual or reciprocal insurer, to accumulate surplus reasonably necessary to support growth in Florida premium volume reasonably expected during the time the rates under scrutiny are in effect. Mr. Schwartz also testified that the last published underwriting profit and contingency factor published by OIR was 3.7%, well below what is identified in Hartford's filings. Hartford counters that reliance on the rule is a misapplication of the rule (with no explanation why), is inconsistent with OIR's treatment of the profit factors in their previous filings, and ignores the language of Section 627.062(2)(b)11., Florida Statutes. No evidence was presented to show whether the expected rate of return threatens Hartford's solvency or makes them more attractive to policyholders or investors from a corporate financial perspective than they would have been if Rule 69O- 170.003 was not implemented. Likewise, it was not demonstrated that the expected rate of return for Florida business is commensurate with the rate of return for other industries having corresponding risk and is necessary to assure confidence in the financial integrity of the insurer in order to maintain its credit and to attract capital. While the position taken by OIR with respect to Hartford's filings may be inconsistent with the position taken in past filings, that cannot be determined on this record. The prior filings, and the communications Hartford had with OIR with regard to those filings, are not included in the exhibits in this case. There is no way to determine whether Petitioners chose to present evidence in the context of prior filings consistent with the criteria in Rule 69O-170.003, or whether OIR approved the underwriting profit and contingency factor despite Rule 69O- 170.003. Having an underwriting profit factor that is considered excessive will result in a higher rate indication. Therefore, it is found that the seventh identified deficiency in the Notices of Intent to Disapprove for the homeowners filings and the second identified deficiency in the Notices of Intent to Disapprove for the dwelling/fire filings is sustained. The eighth identified deficiency is that various components underlying the calculation of the underwriting profit and contingency factors, including but not limited to the return on surplus, premium to surplus ratio, investment income and tax rate are not supported or justified. Return on surplus is the total net income that would result from the underwriting income and the investment income contributions relative to the amount of capital that is exposed. Surplus is necessary in addition to income expected from premium, to insure that claims will be paid should losses in a particular year exceed premium and income earned on premium. Hartford's expected return on surplus in these filings is 15%. The return on surplus is clearly tied to the underwriting profit factor, although the percentages are not necessarily the same. It follows, however, that if the underwriting income and contingency factor is excessive, then the return on surplus may also be too high. Hartford has not demonstrated that the return on surplus can stand, independent of a finding that the underwriting profit and contingency factor is excessive. Premium-to-surplus ratio is a measure of the number of dollars of premium Hartford writes relative to the amount of surplus that is supporting that exposure. Hartford's premium-to- surplus ratio in the AARP homeowners filing is 1.08, which means that if Hartford wrote $108 of premium, it would allocate $100 of surplus to support that premium.3/ The premium-to-surplus ratio is reasonable, given the amount of risk associated with homeowners insurance in Florida. The OIR's position regarding investment income and tax rates are related. The criticism is that the filing used a low- risk investment rate based on a LIBOR (London Interbank Offering Rate), which is a standard in the investment community for risk- free or low-risk yield calculations. The filing also used a full 35% income tax rate applied to the yield. Evidence was presented to show that, if the actual portfolio numbers and corresponding lower tax rate were used in the filings, the rate after taxes would be the same. The problem, however, is that Section 627.062(2)(b)4., Florida Statutes, requires the OIR to consider investment income reasonably expected by the insurer, "consistent with the insurer's investment practices," which assumes actual practices. While the evidence at hearing regarding Hartford's investments using its actual portfolio yield may result in a similar bottom line, the assumptions used in the filing are not based on Petitioner's actual investment practices. As a result, the tax rate identified in the filing is also not the actual tax rate that has been paid by Hartford. The greater weight of the evidence indicates the data used is not consistent with the requirements of Section 627.062(2)(b)4., Florida Statutes. Therefore, the eighth deficiency is sustained to the extent that the filing does not adequately support the return on surplus, investment income and tax rate. The ninth identified deficiency is that the underwriting expenses and other expenses are excessive and not supported. Hartford used the most recent three years of actual expense data, analyzed them and made expense selections based on actuarial judgment. The use of the three-year time frame was both reasonable and consistent with common ratemaking practices. Likewise, the commission rates reflected in the agency filings are also reasonable. The tenth identified deficiency is that the non-FHCF (or private) reinsurance costs are excessive and not supported. The criticism regarding private reinsurance purchases is three- fold: 1) that Hartford paid too much for their reinsurance coverage; 2) that Hartford purchases their reinsurance coverage on a nationwide basis as opposed to purchasing coverage for Florida only; and 3) that the percentage of the reinsurance coverage allocated to Florida is too high. Hartford buys private reinsurance in order to write business in areas that are exposed to catastrophes. It buys reinsurance from approximately 40 different reinsurers in a competitive, arm's-length process and does not buy reinsurance from corporate affiliates. Hartford used the "net cost" of insurance in its filings, an approach that is appropriate and consistent with standard actuarial practices. Hartford also used the RMS model to estimate the expected reinsurance recoveries, which are subtracted from the premium costs. Hartford buys private catastrophic reinsurance on a nationwide basis to protect against losses from hurricanes, earthquakes and terrorism, and allocates a portion of those costs to Florida. Testimony was presented, and is accepted as credible, that attempting to purchase reinsurance from private vendors for Florida alone would not be cost-effective. The cost of reinsurance, excluding a layer of reinsurance that covers only the Northeast region of the country and is not reflected in calculating costs for Florida, is approximately $113 million. Hartford retains the first $250 million in catastrophe risk for any single event, which means losses from an event must exceed that amount before the company recovers from any reinsurer. In 2006, Hartford raised its retention of losses from $175 million to $250 million in an effort to reduce the cost of reinsurance. Hartford purchases reinsurance in "layers," which cover losses based on the amount of total losses Hartford incurs in various events. Hartford allocates approximately 65% of the private reinsurance costs (excluding the Northeast layer) to Florida in the AARP homeowners filing. Only 6-7% of Hartford's homeowners policies are written in Florida. The amount Hartford paid for reinsurance from private vendors is reasonable, given the market climate in which the insurance was purchased. Hartford has demonstrated that the process by which the reinsurance was purchased resulted in a price that was clearly the result of an arms-length transaction with the aim of securing the best price possible. Likewise, the determination to purchase reinsurance on a nationwide basis as opposed to a state-by-state program allows Hartford to purchase reinsurance at a better rate, and is more cost-effective. Purchasing reinsurance in this manner, and then allocating an appropriate percentage to Florida, is a reasonable approach. With respect to the allocation of a percentage of reinsurance cost to Florida, OIR argues that, given that Florida represents only 6-7% of Hartford's homeowner insurance business, allocation of 65% of the reinsurance costs to Florida is per se unreasonable. However, the more logical approach is to examine what percentage of the overall catastrophic loss is attributable to Florida, and allocate reinsurance costs accordingly. After carefully examining both the testimony of all of the witnesses and the exhibits presented in this case, the undersigned cannot conclude that the allocation of 65% of the private reinsurance costs is reasonable, and will not result in an excessive rate.4/ The eleventh identified deficiency is that the FHCF (or CAT Fund) reinsurance costs are excessive and not supported. Hartford purchases both the traditional layer of CAT Fund coverage, which is addressed in a separate filing and not reflected in these filings, and the TICL layer made available pursuant to Chapter 2007-1, Laws of Florida. Hartford removed the costs of its previously purchased private reinsurance that overlapped with the TICL layer and those costs are not reflected in these filings and have not been passed on to Florida policyholders. In estimating the amount of premium Hartford would pay for the TICL coverage, it relied on information provided by Paragon, a consulting firm that calculates the rates for the CAT Fund. As noted in finding of fact number 31, the RMS model, along with three other models accepted by the Hurricane Commission, were used by Paragon for determining expected aggregate losses to the CAT Fund reinsurance layer, clearly a crucial factor in determining the rate for the CAT fund. Hartford did not use the loss recoveries calculated by Paragon, but instead estimated the total amount of premium it would pay for the TICL coverage and subtracted the expected loss recoveries based on the RMS model alone. The expected loss recoveries under the RMS model standing alone were 60% of the loss recovery estimate calculated by Paragon when using all four models. Hartford claimed that its use of the RMS model was necessary for consistency. However, it pointed to no actuarial standard that would support its position with respect to this particular issue. Moreover, given that the premium used as calculated by Paragon used all four models, it is actually inconsistent to use one number which was determined based on all four models (the Paragon-based premium estimate) for one half of this particular calculation and then subtract another number using only one model for the other half (the loss recoveries rate) in order to determine the net premium. To do so fails to take into account the unique nature of the CAT fund, in terms of its low expenses and tax-exempt status. Accordingly, it is found that the CAT-Fund reinsurance costs for the TICL layer are excessive. The twelfth identified deficiency is that Hartford did not consider in the filing that no new business is being written. OIR's explanation of this asserted deficiency is that the costs associated with writing new business are generally higher than that associated with writing renewals. Therefore, according to OIR, failure to make adjustments to their historical experience to reflect the current mix of business, means that the costs included in the filing would be excessive. Hartford began restricting the writing of new business for these filings in 2002. Ultimately, no new business for the AARP program was written after November 2006 and no new business was written for the agency program after June 2006. Credible evidence was presented to demonstrate that a very low percentage of new business has been written over the period of time used for demonstrating Hartford's historical losses. As a result, the effect of no longer writing new business is already reflected in the data used to determine expenses. No additional adjustment in the filing was necessary in this regard. The thirteenth identified deficiency is that no explanation has been provided as to why Hartford believes it is reasonable to return such a low percentage of premium in the form of loss payments to policyholders. For example, for the building policy forms, OIR states that only about 40% of the premium requested by Hartford is expected to be returned to policyholders in the form of loss payments. OIR pointed to no actuarial standard that would require a specific explanation regarding how much of the premium should be returned to policyholders. Nor was any statutory or rule reference supplied to support the contention that such an explanation was required. Finally, the more credible evidence presented indicates that the correct percentage is 44%. In any event, this criticism is not a basis for finding a deficiency in the filing. Alleged Deficiencies in the Dwelling/Fire Filings The seventh deficiency identified in the dwelling/fire filings, not reflected in the homeowner filings, is that the credibility standard and credibility values are not supported. Credibility is the concept of identifying how much weight to put on a particular set of information relative to other potential information. Credibility value is determined by applying the "square root rule" to the credibility value, a commonly used actuarial approach to credibility. Hartford used the credibility standard of 40,000 earned house years in these filings. This credibility standard has been the standard within the industry for personal property filings for over forty years and has been used in prior filings submitted to OIR. Mr. Schwartz testified that his criticism with respect to the credibility standard and credibility values is that Hartford did not explain why they used that particular standard. However, Florida Administrative Code Rule 69O-170.0135 discusses those items that must be included in the Actuarial Memorandum for a filing. With respect to credibility standards and values, Rule 69O-170.0135(2)(e)5., provides that the basis need only be explained when the standard has changed from the previous filing. Given that no change has been made in these filings with respect to the credibility standard, this criticism is not a valid basis for issuing a Notice of Intent to Disapprove. The ninth deficiency in the Notice relating to the dwelling/fire filing in Case No. 07-5187 provides: "No explanation has been provided as too (sic) why Hartford believes it needs such a large rate increase currently, when the cumulative rate change implemented by Hartford for this program from 2001 to 2006 was an increase of only about 10%." With respect to Case No. 07-5188, the deficiency is essentially the same, except the cumulative rate change identified for the same period of time is a decrease of about -3%. Testimony established that the dwelling/fire rate increases were larger than those identified for the homeowners filings because Hartford did not seek rate increases for these lines for several years. The decision not to seek increases was not based on the adequacy of current rates. Rather, the decision was based on an internal determination that, based on the relatively small number of policies involved in these two filings, the amount of increased premium reflected in a rate increase was not sufficient to incur the costs associated with preparing the filings. Mr. Schwartz pointed to no authority, either in statute, rule, or Actuarial Standard, that requires the explanation he desired. He acknowledged that he understood the basis of how Hartford reached the rate increase they are requesting. The failure to provide the explanation Mr. Schwartz was seeking is not a valid basis for a Notice of Intent to Disapprove.

Recommendation Upon consideration of the facts found and conclusions of law reached, it is RECOMMENDED: That a final order be entered that disapproves the rate filings in Case Nos. 07-5185 and 07-5186 based upon the deficiencies numbered 7,8,10 and 11 in the Notices of Intent to Disapprove, and that disapproves the rate filings in Case Nos. 07-5187 and 07-5188 based on the deficiencies numbered 2,3,5 and in the Notices of Intent to Disapprove. DONE AND ENTERED this 28th day of March 2008, in Tallahassee, Leon County, Florida. S LISA SHEARER NELSON Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 28th day of March, 2008.

Florida Laws (6) 120.569120.57215.555627.0613627.062627.0628 Florida Administrative Code (3) 69O-170.00369O-170.01369O-170.0135
# 1
DEPARTMENT OF INSURANCE vs RICHARD EDWARD PANAGOS, 00-000455 (2000)
Division of Administrative Hearings, Florida Filed:West Palm Beach, Florida Jan. 27, 2000 Number: 00-000455 Latest Update: Nov. 30, 2000

The Issue Whether Respondent, a licensed insurance agent, committed the offenses alleged in the Amended Administrative Complaint and, if so, the penalties that should be imposed.

Findings Of Fact Petitioner is a licensing and regulatory agency of the State of Florida charged with the responsibility and duty to enforce the provisions of the Florida Insurance Code, which consists of Chapters 624-632, 634, 635, 641, 642, 648, and 651, Florida Statutes. See Section 624.307(1), Florida Statutes. Respondent has been continuously licensed in the State of Florida as a life insurance agent (a 2-16 license) and a general license agent (a 2-20 license) since March 1974, and continuously as a RPCJUA insurance agent (a 00-17 license) since March 1993. On November 4, 1996, Respondent was charged with possession of cocaine in violation of Section 893.13(6)(a), Florida Statutes. This charge, filed in Palm Beach County Circuit Court and assigned Case Number 96-12206 CFA02, is a third degree felony. On May 14, 1997, Respondent entered a plea of nolo contendere to the charge of possession of cocaine, which was accepted. Adjudication of guilt was withheld and Respondent was placed on probation for a period of 18 months. The terms and conditions of Respondent's probation included working at a lawful occupation, intensive drug and alcohol evaluation, successful completion of any recommended treatment, payment of a fine in the amount of $250.00 and court costs in the amount of $461.00, performance of 100 hours of community service, random testing for the use of alcohol and drugs, six months' suspension of driver's license, and DUI school. Respondent successfully completed his probation on November 13, 1998. Respondent continued to work as an insurance agent during the term of his probation. Respondent voluntarily reported the incident to State Farm shortly after its occurrence. As a result, State Farm placed Respondent on probation and conducted a series of random alcohol and drug tests, which Respondent satisfactorily completed. Section 626.621(11), Florida Statutes, provides that the following constitutes grounds for the discretionary discipline of an agent's licensure: (11) Failure to inform the department in writing within 30 days after pleading guilty or nolo contendere to, or being convicted or found guilty of, any felony or a crime punishable by imprisonment of 1 year or more under the law of the United States or of any state thereof, or under the law of any other country without regard to whether a judgment of conviction has been entered by the court having jurisdiction of the case. Respondent failed to report to Petitioner within 30 days of doing so that he entered a plea of nolo contendere to a third degree felony charge of possession of cocaine in Case Number 96-12206 CFA02 on May 14, 1997. On or about March 18, 1998, Respondent applied for licensure as a Variable Annuity Insurance Agent (a 2-19 license). That application contained Question 18, which provides as follows and to which Respondent answered "yes": Have you ever been convicted, found guilty, or pleaded guilty or nolo contendere (no contest) to a felony under the laws of any municipality, county, state, territory or country, whether or not a judgment of conviction has been entered. As a result of his answer to Question 18, Petitioner started an investigation, with which Respondent fully cooperated. As a result of that investigation, Petitioner learned the details of Respondent's plea in the criminal proceeding. Respondent testified, credibly, that he did not timely report the entry of his plea in the criminal proceeding because he did not know he was required to do so. 1/ Respondent has continuously worked as an insurance agent licensed by Petitioner in the State of Florida since March 1974. Respondent has been continuously appointed by State Farm and has built up a successful insurance business. This proceeding is the first disciplinary proceeding brought against Respondent's insurance licenses. There have been no other complaints filed by anyone in this state against Respondent's insurance licenses. Respondent's insurance licenses have not been previously disciplined in the State of Florida. The testimony of Respondent's witnesses established that he enjoys a good reputation for honesty, trustworthiness, truthfulness, and integrity in his community. He has engaged in charitable works, including work with the food bank, the Guardian Ad Litem Program, and Brazilian Indians. Respondent's witnesses also established that they had been pleased with their business dealings with Respondent, and that he has the ability and trustworthiness to successfully engage in the business of insurance. Respondent testified that State Farm will terminate his appointment as an agent if his license is suspended. Respondent testified that he will lose his business and his employees will lose their employment.

Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that Petitioner enter a final order finding Respondent guilty of violating Section 626.621(8), Florida Statutes, as alleged in Count I of the Amended Administrative Complaint, and guilty of violating Section 626.621(11), Florida Statutes, as alleged in Count II of the Amended Administrative Complaint. It is further recommended that Respondent's licensure as an insurance agent be suspended for two months for the violation of Count I and for three months for the violation of Count II, to run concurrently. DONE AND ENTERED this 30th day of June, 2000, in Tallahassee, Leon County, Florida. CLAUDE B. ARRINGTON Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 30th day of June, 2000.

Florida Laws (5) 120.57624.307626.611626.621893.13
# 2
DANIEL BRUCE CAUGHEY vs DEPARTMENT OF INSURANCE AND TREASURER, 90-004473F (1990)
Division of Administrative Hearings, Florida Filed:Tallahassee, Florida Jul. 19, 1990 Number: 90-004473F Latest Update: Dec. 27, 1990

Findings Of Fact On September 8, 1987, the Department of Insurance received a letter dated September 1, 1987, from Joseph F. Kinman, Jr., which stated: Another insurance agent (Daniel Bruce Caughey) from Pensacola, Florida and his incorporated agency (Caughey Insurance Agency, Inc.) are refusing to forward premium payments on to Jordan Roberts & Company, Inc. despite a final judgment for such amounts here in Hillsborough County Circuit Court. Enclosed is a copy of the Final Judgment entered August 13, 1987, as well as a copy of the Complaint. We represent Jordan Roberts & Company, as well as Poe & Associates, Inc. here in Tampa, Florida. In approximately August of 1982, Daniel Bruce Caughey and Caughey Insurance Agency, Inc. entered into a brokerage agreement with Jordan Roberts & Company, Inc. wherein Mr. Caughey and the Agency were to collect premiums on behalf of Jordan Roberts & Company, Inc. and in turn, Mr. Caughey and the Agency were to receive commissions. Mr. Caughey signed an Individual Guarantee Agreement on October 21, 1983, guaranteeing that Brokerage Agreement with Caughey Insurance Agency, Inc. Mr. Caughey and the Agency failed to forward the insurance premiums collected on behalf of Jordan Roberts & Company, Inc. despite repeated demands and inquiries. Finally, a lawsuit was filed against Mr. Caughey and the Agency in the Circuit Court of the Thirteenth Judicial Circuit of the State of Florida, in and for Hillsborough County in December of 1986. Final judgment for Jordan Roberts & Company, Inc. against Mr. Caughey and the Agency was entered on August 13, 1987, for an amount of $6,595.94. Mr. Caughey and his Agency have unlawfully withheld monies belonging to an insurer, Jordan Roberts & Company, Inc. and, accordingly, appear to be in violation of Florida Statutes 626 et seq. Jordan Roberts & Company, Inc. has a judgment for unpaid insurance premiums against Mr. Caughey and the Agency, however, Mr. Caughey and the Agency refuse or fail to pay over to Jordan Roberts & Company, Inc. premium funds rightfully belonging to Jordan Roberts & Company, Inc. Accordingly, we would respectfully request that your office conduct an investigation of Mr. Caughey and the Caughey Insurance Agency, Inc. Enclosed with this letter were copies of the complaint and final judgment in the circuit court case, Case No. 86-21454. As found in the main administrative case, Case No. 89-2651: In Count 1, JORO's complaint [in Case No. 86-21454] alleges the existence of a brokerage agreement between JORO and Caughey Insurance Agency, Inc., entered into "[o]n or about April 27, 1982"; execution and delivery of respondent's guarantee "[o]n or about October 21, 1983"; and the agency's indebtedness "for premiums on policies underwritten by [JORO] for the sum of $20,975.36." Petitioner's Exhibit No. 3. In Count II, the complaint also alleges execution and delivery of a promissory note "[o]n or about October 21, 1983," without, however, explicitly indicating its relationship (if any) with the guarantee executed the same date. Petitioner's Exhibit No. 3. The final judgment does not specify which count(s) JORO recovered on. Petitioner's Exhibit No. 4. Attached to the complaint are copies of the promissory note, executed by "CAUGHEY INSURANCE AGENCY, INC., By: D B Caughey Vice President"; the guarantee, executed in the same way; and the brokerage agreement, executed on behalf of Caughey Insurance Agency by "William C. Caughey, President." Although the Individual Guarantee Agreement names respondent as guarantor in the opening paragraph, the corporation is shown as guarantor on the signature line. The complaint does not allege and the judgment does not recite that respondent personally failed to remit premiums but says he is responsible as an officer of the agency. Without any further investigation, as far as the record shows, the Department of Insurance filed a complaint amended on April 24, 1989, to allege, inter alia, that "[o]n or about August 19, 1982 Caughey Insurance Agency, Inc. entered into a brokerage agreement with Jordan Roberts and Company, Inc. . . . requir[ing] Caughey Insurance Agency, Inc. to remit premiums, unearned commissions and additional premiums to Jordan Roberts and Company, Inc."; and that respondent "personally guaranteed the [agency's] obligation under this agreement in" writing, but "failed to remit five thousand five dollars and forty-four cents due under th[e] agreement" for which sum Jordan Roberts and Company, Inc. obtained judgment. After a formal administrative hearing, a recommended order was entered on April 2, 1990, recommending dismissal of the administrative complaint, because "ambiguities in the court papers do not clearly and convincingly rule out the possibility that the court's judgment rests on the dishonored promissory note . . . [rather than] a breach of respondent's [here petitioner's] fiduciary responsibilities." In its final order, the Department dismissed the administrative complaint; Daniel Bruce Caughey was the prevailing party in that case. The parties have stipulated that "Daniel B. Caughey qualifies as a small business party as defined in Section 57.111(3)(d), Florida Statutes." The parties also stipulated that the "total value of the reasonable attorney's fees and costs at issue is $2,830."

Florida Laws (3) 120.57120.6857.111
# 3
DEPARTMENT OF FINANCIAL SERVICES, DIVISION OF WORKERS' COMPENSATION vs HERNANDEZ ENTERPRISES, 04-001174 (2004)
Division of Administrative Hearings, Florida Filed:Jacksonville, Florida Apr. 07, 2004 Number: 04-001174 Latest Update: Mar. 23, 2006

The Issue The issue is whether Respondent complied with Sections and 440.38, Florida Statutes, with regard to workers' compensation insurance for his subcontractors, and if not, the appropriate amount of penalty that should be assessed.

Findings Of Fact Hernandez, Inc., is a contractor based in the Jacksonville, Florida area, and is in the business of installing dry wall, among other construction related activities. The Department of Financial Services is the state agency responsible for enforcing the Workers' Compensation Law. This duty is delegated to the Division of Workers' Compensation. On February 5, 2004, Hernandez, Inc., was engaged in installing drywall in the Bennett Federal Building in Jacksonville, Florida. Hernandez, Inc., was a subcontractor for Skanska, Inc., who was the general contractor for the building. Hernandez, Inc., was accomplishing the installation of drywall by using two subcontractors, GIO & Sons (GIO), of Norfolk, Virginia, and U&M Contractors, Inc., (U&M), of Charlotte, North Carolina. Hernandez, Inc., was also using its own personnel, who were leased from Matrix, Inc., an employee leasing company. Prior to contracting with GIO and U&M, Hernandez, Inc., asked for and received ACORD certificates of insurance, which on their face indicated that the subcontractors had both liability coverage and workers' compensation coverage. It is the practice of Hernandez, Inc., to ensure that certificates of insurance are provided by subcontractors and the office staff of Hernandez, Inc., tracks the certificates so that they are kept current. Since the beginning of 2001, Hernandez, Inc., has received approximately 310 certificates of insurance from subcontractors. These certificates listed Hernandez, Inc., as the certificate holder. Though most of the producers and insureds on these certificates are from Florida, a substantial number are from other states. Hernandez, Inc., relied on the certificates as evidence that the subcontractor's workers were covered by workers' compensation insurance. Hernandez, Inc., has relied on certificates of insurance for more than twenty years and, with the exception of this case, has never known an instance where the underlying policy was invalid. On February 5, 2004, Katina Johnson, an investigator with the Division, made a routine visit to the Bennett Federal Building with another investigator. She observed personnel from Hernandez, Inc., and its subcontractors, installing dry wall. On February 5, 2004, Ms. Johnson determined that Hernandez, Inc., also had a contract to install dry wall as a subcontractor participating in the construction of the Mayport BEQ. L. C. Gaskins Company was the general contractor engaged in the construction of the Mayport BEQ. U&M worked at both the Bennett Federal Building site and the Mayport BEQ site as a subcontractor of Hernandez, Inc. Ms. Johnson issued a Stop Work Order on February 26, 2004, to Hernandez, Inc., GIO, and U&M. By the Stop Work Order, Hernandez, Inc., was charged with failure to ensure that workers' compensation meeting the requirements of Chapter 440, Florida Statutes, and the Florida Insurance Code, was in place for GIO and U&M. The Stop Work Order indicated that the penalty amount assessed against Respondent would be subject to amendment based on further information provided by Hernandez, Inc., including the provision of business records. An Amended Order of Penalty Assessment dated March 19, 2004, was served on Hernandez, Inc., which referenced the Stop Work Order of February 26, 2004. The Amended Order of Penalty Assessment was in the amount of $157,794.49. The Amended Order of Penalty Assessment reached back to September 29, 2003. An Amended Order of Penalty Assessment dated March 22, 2004, was served on GIO. This Amended Order of Penalty Assessment was in the amount of $107,885.71. An Amended Order of Penalty Assessment with a March 2004 date (the day is obscured on the document by a "filed" stamp), was served on U&M. This Amended Order of Penalty Assessment was in the amount of $51,779.50. The sum of these numbers is $159,665.21. However, the parties agreed at the hearing that the amount being sought by the Division was $157,794.49, which represented the total for GIO and U&M. Hernandez, Inc.'s, employees leased from Matrix were covered by workers' compensation insurance through a policy held by Matrix. The Matrix policy did not cover the employees of GIO and U&M. Although Skanska, Inc., and L. C. Gaskins Company had workers' compensation insurance in force, their policies did not cover the workers used by Hernandez, Inc., or the employees of GIO or U&M. GIO and U&M employees were considered by the Division to be "statutory employees" of Hernandez, Inc., for purposes of the Workers' Compensation Law. This meant, according to the Division, that Hernandez, Inc., was required to ensure that the employees of GIO and U&M would receive benefits under the Workers' Compensation Law if a qualifying event occurred, unless the subcontractors had workers' compensation insurance policies in force that satisfied the Division. GIO had a policy of workers' compensation insurance evidenced by an ACORD certificate of liability insurance for the period December 3, 2002, until December 3, 2003. The policy was produced by Salzberg Insurance Agency in Norfolk, Virginia. It listed Hernandez as the certificate holder. The policy was issued by Maryland Casualty Company, a subsidiary of the Zurich American Insurance Company. These companies are admitted carriers in Florida. The Classification of Operations page of this policy indicated class code 5022, masonry work. GIO employers were installing drywall during times pertinent. Rates for drywall installation are substantially higher than for masonry work. In the policy section titled "Other States Insurance," Florida is not mentioned. William D. Hager, an expert witness, reviewed the certificate of insurance and the policy supporting the certificate. Mr. Hager is a highly qualified expert in insurance and workers' compensation coverage. Among other qualifications, he is an attorney and a former member of the National Association of Insurance Commissioners by virtue of his position as Insurance Commissioner for the State of Iowa. He concluded that this policy did not conform to the requirements of Chapter 440 because the policy was Virginia based and did not apply Florida rates, rules, and class codes. Mr. Sapourn, testified as an expert witness. Mr. Sapourn has a degree from the University of Virginia in economics with high distinction and a juris doctorate from Georgetown. He is a certified insurance counselor and owned an insurance agency in the District of Columbia area. As an insurance agent he has issued tens of thousands certificates of insurance and written hundreds of workers' compensation policies. Mr. Sapourn, opined that this certificate represented workers' compensation coverage that complied with Chapter 440, Florida Statutes. Upon consideration of the testimony of the experts, and upon an examination of the documents, it is concluded that the policy represented by the certificate of insurance for the period December 3, 2002, to December 3, 2003, did not comply with the requirements of Chapter 440. Subsequently, someone forged an ACORD certificate of liability insurance, which indicated that it was produced by Salzberg Insurance Agency, and that indicated that GIO was covered from December 4, 2003, until December 4, 2004. The forged certificate was presented to Hernandez, Inc., upon the expiration of the policy addressed above. It was accepted by Hernandez, Inc., and considered to be a valid certificate. Both of the experts pointed out that with their practiced eye they could easily determine that the certificate was a forgery. However, there was no evidence that Mr. Hernandez, or his employees, had training in forgery detection. Accordingly, it was reasonable for them to accept the certificate as valid. U&M presented Hernandez, Inc., with an ACORD certificate which indicated insurance coverage from October 24, 2003, until October 24, 2004. The producer was Insur-A-Car Commercial Division of Charlotte, North Carolina. The insurer was The St. Paul, an admitted carrier in Florida. The insured was U &M. The certificate holder was Hernandez Enterprises, Inc. William D. Hager reviewed the certificate of insurance and the policy supporting the certificate. He noted that The St. Paul policy upon which the certificate was based did not apply in Florida because U&M was not working temporarily in Florida and because it included a policy endorsement that stated: "The policy does not cover work conducted at or from 3952 Atlantic BLVD #D-12 Jacksonville, FL 32207." U&M's mailing address in Jacksonville was 3952 Atlantic Boulevard, Suite D-12. The information page of the policy, at Part 3.A. states that Part One applies to North Carolina. Part 3.C., Other States Insurance states that Part 3 of the policy applies to the states listed, and then refers to the "residual market limited other states insurance." Mr. Hager testified that the policy did not indicate compliance with Chapter 440, because the policy is North Carolina based, applies only North Carolina rates, and does not provide Florida coverage. Mr. Sapourn, on the other hand, opined that the policy provided workers' compensation that complied with Chapter 440. Although it is possible that a worker who was injured during times pertinent may have received benefits, it is clear that the policy did not comply with the requirements of Chapter 440. The Division instituted a Stop Work Order against U&M and sought to impose penalties upon it for failure to comply with Chapter 440 for offenses committed at the exact times and places alleged in this case. U&M demanded a hearing and was provided one. In a Recommended Order entered April 7, 2005, an Administrative Law Judge recommended that the Division enter a final order affirming the Stop Work Order and assessing a penalty in the amount of $51,779.50. See Department of Financial Services, Division of Workers' Compensation vs. U and M Contractors, Inc., Case No. 04-3041 (DOAH April 7, 2005). The recommendation was adopted in toto by the Department of Financial Services on April 27, 2005. See In the Matter of: U and M Contractors, Inc., Case No. 75537-05 WC (DFS April 27, 2005). The evidence taken as a whole demonstrates that U&M did not have workers' compensation coverage in Florida that complied with the requirements of Chapter 440, during times pertinent. Mr. Sapourn testified that the theory behind ACORD certificates of insurance is that they provide a uniform document upon which business people may rely. This testimony is accepted as credible. In order to continue working on a project not addressed by the Stop Work Order, Hernandez, Inc., entered into and agreement with the Division which provided for partial payments of the penalty in the amount of $46,694.03. This payment was made with the understanding of both parties that payment was not an admission of liability.

Recommendation Based upon the Findings of Fact and Conclusions of Law, it is

Florida Laws (7) 120.57440.10440.107440.13440.16440.38694.03
# 4
HARTFORD FIRE INSURANCE COMPANY, HARTFORD INSURANCE OF THE SOUTHEAST, HARTFORD CASUALTY INSURANCE COMPANY, TWIN CITY FIRE INSURANCE COMPANY, HARTFORD UNDERWRITERS INSURANCE COMPANY, AND HARTFORD ACCIDENT AND INDEMNITY COMPANY vs OFFICE OF INSURANCE REGULATION, 07-005185 (2007)
Division of Administrative Hearings, Florida Filed:Tallahassee, Florida Nov. 09, 2007 Number: 07-005185 Latest Update: Jun. 03, 2008

The Issue Whether Petitioners' proposed rates are justified pursuant to the requirements of Section 627.062, Florida Statutes, or whether the Department of Financial Services, Office of Insurance Regulation (OIR) was correct in denying the requested rate increases.

Findings Of Fact The Hartford companies are property and casualty insurers transacting insurance in the State of Florida pursuant to valid certificates of authority and the Florida Insurance Code. Two types of personal lines insurance filings submitted by Hartford to the OIR are at issue in this proceeding: two filings for homeowners insurance (Case Nos. 07-5185 and 07-5186) and two filings for dwelling fire insurance (Case Nos. 07-5187 and 07- 5188). Hartford's substantial interests are affected by the notices disapproving the filings in this case. Homeowners insurance includes coverage for a variety of perils in and around a home, is usually purchased by a homeowner, and covers both the structure and the contents of a home. Dwelling/fire insurance is usually purchased by the owners of properties that are leased or rented to others, and provides coverage for the structure only. Both types of insurance cover damage caused by hurricanes. The New Legislation and its Requirements In a special session held in January 2007, the Florida Legislature enacted changes to the Florida Hurricane Catastrophe Fund (CAT Fund), as reflected in Chapter 2007-1, Laws of Florida. The special session was precipitated by a perceived crisis regarding the cost and availability of homeowners insurance after the 2004 and 2005 hurricane seasons. As a result of the substantial number of claims incurred after multiple severe hurricanes each of these years, changes in the insurance marketplace resulted in some insurance companies withdrawing from the Florida market, others non-renewing policies, one company becoming insolvent, and the cost for reinsurance available to all insurers rising dramatically. One of the primary features of the legislation was an expansion of the CAT Fund. The CAT Fund was established in 1993 after Hurricane Andrew to provide reinsurance to insurers for property insurance written in Florida at a price significantly less than the private market. The CAT Fund is a non-profit entity and is tax exempt. Prior to the enactment of Chapter 2007-1, the CAT Fund had an industry-wide capacity of approximately $16 million. The purpose of the changes enacted by the Legislature was to reduce the cost of reinsurance and thereby reduce the cost of property insurance in the state. As a result of Chapter 2007-1, the industry-wide capacity of the CAT Fund was increased to $28 billion, and insurers were given an opportunity to purchase an additional layer of reinsurance, referred to as the TICL layer (temporary increase in coverage limit), from the CAT Fund. Section 3 of Chapter 2007-1 required insurers to submit a filing to the OIR for policies written after June 1, 2007, that took into account a "presumed factor" calculated by OIR and that purported to reflect savings created by the law. The new law delegated to the OIR the duty to specify by Order the date such filings, referred to as "presumed factor filings" had to be made. On February 19, 2007, the OIR issued Order No. 89321-07. The Order required insurers to make a filing by March 15, 2007, which either adopted presumed factors published by the OIR or used the presumed factors and reflected a rate decrease taking the presumed factors into account. The presumed factors were the amounts the OIR calculated as the average savings created by Chapter 2007-1, and insurers were required to reduce their rates by an amount equal to the impact of the presumed factors. The OIR published the presumed factors on March 1, 2007. In its March 15, 2007, filings, Hartford adopted the presumed factors published by OIR. As a result, Hartford reduced its rates, effective June 1, 2007, on the products at issue in these filings by the following percentages: Case No. 07-5185 homeowners product: 17.7% Case No. 07-5186 homeowners product: 21.9% Case No. 07-5187 dwelling/fire product: 8.7% Case No. 07-5188 dwelling/fire product: 6.2% The Order also required that insurers submit a "True-Up Filing" pursuant to Section 627.026(2)(a)1., Florida Statutes. The filing was to be a complete rate filing that included the company's actual reinsurance costs and programs. Hartford's filings at issue in these proceedings are its True-Up Filings. The True-Up Filings Hartford submitted its True-Up filings June 15, 2007. The rate filings were certified as required by Section 627.062(9), Florida Statutes. The filings were amended August 8, 2007. Hartford's True Up Filings, as amended, request the following increases in rates over those reflected in the March 15, 2007, presumed factor filings: Case No. 07-5185 homeowners product: 22.0% Case No. 07-5186 homeowners product: 31.6% Case No. 07-5187 dwelling and fire product: 69.0% Case No. 07-5188 dwelling and fire product: 35.9% The net effects of Hartford's proposed rate filings result in the following increases over the rates in place before the Presumed Factor Filings: Case No. 07-5185 homeowners product: .4% Case No. 07-5186 homeowners product: 2.8% Case No. 07-5187 dwelling/fire product: 54.3% Case No. 07-5188 dwelling/fire product: 27.5% Case Nos. 07-5185 and 07-5186 (homeowners) affect approximately 92,000 insurance policies. Case Nos. 07-5187 and 07-5188 (dwelling/fire) affect approximately 2,550 policies. A public hearing was conducted on the filings August 16, 2007. Representatives from Hartford were not notified prior to the public hearing what concerns the OIR might have with the filings. Following the hearing, on August 20, 2007, Petitioners provided by letter and supporting documentation additional information related to the filings in an effort to address questions raised at the public hearing. The OIR did not issue clarification letters to Hartford concerning any of the information provided or any deficiencies in the filings before issuing its Notices of Intent to Disapprove the True-Up Filings. All four filings were reviewed on behalf of the OIR by Allan Schwartz. Mr. Schwartz reviewed only the True-Up Filings and did not review any previous filings submitted by Hartford with respect to the four product lines. On September 10, 2007, the OIR issued Notices of Intent to Disapprove each of the filings at issue in this case. The reasons give for disapproving the two homeowners filings are identical and are as follows: Having reviewed the information submitted, the Office finds that this filing does not provide sufficient documentation or justification to demonstrate that the proposed rate(s) comply with the standards of the appropriate statute(s) and rules(s) including demonstrating that the proposed rates are not excessive, inadequate, or unfairly discriminatory. The deficiencies include but are not limited to: The premium trends are too low and are not reflective of the historical pattern of premium trends. The loss trends are too high and are not reflective of the historical pattern of loss trends. The loss trends are based on an unexplained and undocumented method using "modeled" frequency and severity as opposed to actual frequency and severity. The loss trends are excessive and inconsistent compared to other sources of loss trends such as Fast Track data. The catastrophe hurricane losses, ALAE and ULAE amounts are excessive and not supported. The catastrophe non-hurricane losses, ALAE and ULAE amounts are excessive and not supported. The particular time period from 1992 to 2006 used to calculate these values has not been justified. There has been no explanation of why the extraordinarily high reported losses for 1992 and 1993 should be expected to occur in the future. The underwriting profit and contingency factors are excessive and not supported. Various components underlying the calculation of the underwriting profit and contingency factors, including but not limited to the return on surplus, premium to surplus ratio, investment income and tax rate are not supported or justified. The underwriting expenses and other expenses are excessive and not supported. The non-FHCF reinsurance costs are excessive and not supported. The FHCF reinsurance costs are excessive and not supported. The fact that no new business is being written has not been taken into account. No explanation has been provided as too [sic] Hartford believes it is reasonable to return such a low percentage of premium in the form of loss payments to policyholders. For example, for the building policy forms, only about 40% of the premium requested by Hartford is expected to be returned to policyholders in the form of loss payments. As a result of the deficiencies set forth above, the Office finds that the proposed rate(s) are not justified, and must be deemed excessive and therefore, the Office intends to disapprove the above-referenced filing. The Notices of Intent to Disapprove the two dwelling/fire filings each list nine deficiencies. Seven of the nine (numbers 1-6 and 8) are the same as deficiencies listed for the homeowners filings. The remaining deficiencies named for Case No. 07-5187 are as follows: 7. The credibility standard and credibility value are not supported. 9. No explanation has been provided as too (sic) why Hartford believes it needs such a large rate increase currently, when the cumulative rate change implemented by Hartford for this program from 2001 to 2006 was an increase of only about 10%. The deficiencies listed for Case No. 07-5188 are the same as those listed for Case No. 07-5187, with the exception that with respect to deficiency number 9, the rate change implemented for the program in Case No. 07-5188 from 2001 to 2006 was a decrease of about -3%. Documentation Required for the Filings Florida's regulatory framework, consistent with most states, requires that insurance rates not be inadequate, excessive, or unfairly discriminatory. In making a determination concerning whether a proposed rate complies with this standard, the OIR is charged with considering certain enumerated factors in accordance with generally accepted and reasonable actuarial techniques. Chapter 2007-1 also amended Section 627.062, Florida Statutes, to add a certification requirement. The amendment requires the chief executive officer or chief financial officer and chief actuary of a property insurer to certify under oath that they have reviewed the rate filing; that to their knowledge, the rate filing does not contain any untrue statement of a material fact or omit to state a material fact necessary to make the statements made, in light of the circumstances under which the statements were made, not misleading; that based on their knowledge, the information in the filing fairly presents the basis of the rate filing for the period presented; and that the rate filing reflects all premium savings reasonably expected to result from legislative enactments and are in accordance with generally accepted and reasonable actuarial techniques. § 627.062(9)(a), Fla. Stat. (2007). Actuarial Standards of Practice 9 and 41 govern documentation by an actuary. Relevant sections of Standard of Practice 9 provide: Extent of documentation - . . . Appropriate records, worksheets, and other documentation of the actuary's work should be maintained by the actuary and retained for a reasonable length of time. Documentation should be sufficient for another actuary practicing in the same field to evaluate the work. The documentation should describe clearly the sources of data, material assumptions, and methods. Any material changes in sources of data, assumptions, or methods from the last analysis should be documented. The actuary should explain the reason(s) for and describe the impact of the changes. Prevention of misuse - . . . The actuary should take reasonable steps to ensure that an actuarial work product is presented fairly, that the presentation as a whole is clear in its actuarial aspects, and that the actuary is identified as the source of the actuarial aspects, and that the actuary is available to answer questions.. . . . * * * 5.5 Availability of documentation- Documentation should be available to the actuary's client or employer, and it should be made available to other persons when the client or employer so requests, assuming appropriate compensation, and provided such availability is not otherwise improper. . . . In determining the appropriate level of documentation for the proposed rate filings, Petitioner relied on its communications with OIR, as well as its understanding of what has been required in the past. This reliance is reasonable and is consistent with both the statutory and rule provisions governing the filings. Use of the RMS Catastrophic Loss Projection Model In order to estimate future losses in a rate filing, an insurer must estimate catastrophic and non-catastrophic losses. Hartford's projected catastrophic losses in the filings are based upon information provided from the Risk Management Solutions (RMS) catastrophic loss projection model, version 5.1a. Hartford's actuaries rely on this model, consistent with the standards governing actuarial practice, and their reliance is reasonable. Catastrophe loss projection models may be used in the preparation of insurance filings, if they have been considered by and accepted by the Florida Commission on Hurricane Loss Projection Methodology (the Hurricane Commission). The Hurricane Commission determined that the RMS model, version 5.1a was acceptable for projecting hurricane loss costs for personal residential rate filings on May 17, 2006. In addition to approval by the Hurricane Commission, use of the model is appropriate "only if the office and the consumer advocate appointed pursuant to s. 627.0613 have access to all of the assumptions and factors that were used in developing the actuarial methods, principles, standards, models, or output ranges, and are not precluded from disclosing such information in a rate proceeding." §627.0628(3)(c), Fla. Stat. Both the Consumer Advocate and a staff person from the OIR are members of the Hurricane Commission. In that context, both have the ability to make on-site visits to the modeling companies, and to ask any questions they choose regarding the models. Both OIR's representative and the Consumer Advocate participated in the meetings and had the same opportunity as other commissioners to ask any question they wished about RMS 5.1a. The Hurricane Commission members, including the Consumer Advocate, clearly have access to the information identified in Section 627.0628(3)(c). However, there are restrictions on the Hurricane Commission members' ability to share the information received regarding trade secrets disclosed by the modeling companies. For that reason, the Commission's deliberations are not, standing alone, sufficient to determine that the Office of Insurance Regulation has access. In this case, credible evidence was submitted to show that RMS officials met with staff from the Office in July and October 2006 to discuss the model. RMS offered to provide any of its trade secret information to the OIR, subject to a non- disclosure agreement to protect its dissemination to competitors. RMS also opened an office in Tallahassee and invited OIR staff to examine any parts of the model they wished. In addition, both RMS and Hartford have answered extensive questionnaires prepared by OIR regarding the RMS model, and Hartford has offered to assist OIR in gathering any additional information it requires. Most of the questions posed by OIR involve the same areas reviewed by the Commission. RMS' representative also testified at hearing that RMS would not object to disclosure of the assumptions during the hearing itself if necessary. Finally, OIR Exhibit 1 is the Florida Hurricane Catastrophe Fund 2007 Ratemaking Formula Report. The Executive Summary from the report explains how rates were recommended for the Florida Hurricane Catastrophic Fund (CAT Fund) for the 2007- 2008 contract year. The report stated that the RMS model, as well as three other models accepted by the Hurricane Commission, were used for determining expected aggregate losses to the CAT Fund reinsurance layer. Three models, including the RMS model, were also used for analysis of detailed allocation to type of business, territory, construction and deductible, as well as special coverage questions. The models were compared in detail and given equal weight. The report notes that these three models were also used in 1999-2006 ratemaking. The report is prepared by Paragon Strategic Solutions, Inc., an independent consultant selected by the State Board of Administration, in accordance with Section 215.555(5), Florida Statutes. While OIR did not prepare the report, they show no hesitation in accepting and relying on the report and the modeled information it contains in these proceedings. Indeed, one of OIR's criticisms is Hartford's failure to use the report with respect to CAT Fund loss recovery estimates. Based upon the evidence presented at hearing, it is found that the OIR and Consumer Advocate were provided access to the factors and assumptions used in the RMS model, as contemplated by Section 627.0628. The Alleged Deficiencies in the Homeowners Filings1/ A rate is an estimate of the expected value of future costs. It provides for all costs associated with the transfer of risk. A rate is reasonable and not excessive, inadequate or unfairly discriminatory if it is an actuarially sound estimate of the expected value of all future costs associated with an individual risk transfer. In preparing a filing, an actuary identifies the time period that its proposed rates are expected to be in effect. Because ratemaking is prospective, it involves determining the financial value of future contingent events. For the rate filings in question, actuaries for Hartford developed their rate indications by first considering trended premium, which reflects changes in premium revenue based on a variety of factors, including construction costs and the value of the buildings insured. Trended premium is the best estimate of the premium revenue that will be collected if the current rates remain in effect for the time period the filing is expected to be in place. Expenses associated with writing and servicing the business, the reinsurance costs to support the business and an allowance for profit are subtracted from the trended premium. The remainder is what would be available to pay losses. This approach to ratemaking, which is used by Hartford, is a standard actuarial approach to present the information for a rate indication. As part of the process, expected claims and the cost to service and settle those claims is also projected. These calculations show the amount of money that would be available to pay claims if no changes are made in the rates and how much increased premium is necessary to cover claims. The additional amount of premium reflects not only claims payments but also taxes, licenses and fees that are tied to the amount of premium. The first deficiency identified by OIR is that "the premium trends are too low and are not reflective of the historical pattern of premium trends." In determining the premium trend in each filing, Hartford used data from the previous five years and fit an exponential trend to the historical pattern, which is a standard actuarial technique. Hartford also looked at the factors affecting the more recent years, which were higher. For example, the peak in premium trend in 2006 was a result of the cost increases driven by the 2004 and 2005 hurricanes, and the peak in demand for labor and construction supplies not matched by supply. Costs were coming down going into 2007, and Hartford believed that 2006 was out of pattern from what they could anticipate seeing in the future. The premium trends reflected in Hartford's filings are reasonable, reflective of historical patterns, and based on standard actuarial techniques. The second identified deficiency with respect to the homeowner filings was that the loss trends are too high and are not reflective of the historical pattern of loss trends. A loss trend reflects the amount an insurance company expects the cost of claims to change. It consists of a frequency trend, which is the number of claims the insurance company expects to receive, and a severity trend, which is the average cost per claim. The loss trend compares historical data used in the filing with the future time period when the new rates are expected to be in effect. Hartford's loss trends were estimated using a generalized linear model, projecting frequency and severity separately. The model was based on 20 quarters of historical information. The more credible testimony presented indicates that the loss trends were actuarially appropriate. The third identified deficiency is that the loss trends are based on an unexplained and undocumented method using "modeled" frequency and severity as opposed to actual frequency and severity. As noted above, the generalized linear model uses actual, historical data. Sufficient documentation was provided in the filing, coupled with Hartford's August 20, 2007, letter. The method used to determine loss trends is reasonable and is consistent with standard actuarial practice. The fourth identified deficiency is that loss trends are excessive and inconsistent compared to other sources of loss trends, such as Fast Track data. Saying that the loss trends are excessive is a reiteration of the claim that they are too high, already addressed with respect to deficiency number two. Fast Track data is data provided by the Insurance Services Office. It uses unaudited information and is prepared on a "quick turnaround" basis. Fast Track data is based on paid claims rather than incurred claims data, and upon a broad number of companies with different claims settlement practices. Because it relies on paid claims, there is a time lag in the information provided. Hartford did not rely on Fast Track data, but instead relied upon its own data for calculating loss trends. Given the volume of business involved, Hartford had enough data to rely on for projecting future losses. Moreover, Respondents point to no statutory or rule requirement to use Fast Track data. The filings are not deficient on this basis. The fifth identified deficiency in the Notice of Intent to Disapprove is that catastrophe hurricane losses, ALAE and ULAE amounts are excessive and not supported. ALAE stands for "allocated loss adjustment expenses," and represents the costs the company incurs to settle a claim and that can be attributed to that particular claim, such as legal bills, court costs, experts and engineering reports. By contrast, ULAE stands for "unallocated loss adjustment expense" and represents the remainder of claims settlement costs that cannot be linked to a specific claim, such as office space, salaries and general overhead. Part of the OIR's objection with respect to this deficiency relates to the use of the RMS model. As stated above at paragraphs 25-33, the use of the RMS model is reasonable. With respect to ALAE, Hartford analyzed both nationwide data (4.4%) and Florida data (4.8%) and selected an ALAE load between the two (4.6%). This choice benefits Florida policyholders. It is reasonable to select between the national and Florida historical figures, given the amount of actual hurricane data available during the period used. With respect to ULAE, the factors used were based upon directions received from Ken Ritzenthaler, an actuary with OIR, in a previous filing. The prior discussions with Mr. Ritzenthaler are referenced in the exhibits to the filing. The more credible evidence demonstrates that the ALAE and ULAE expenses with respect to catastrophic hurricane losses are sufficiently documented in Hartford's filings and are based on reasonable actuarial judgment. The sixth identified deficiency is that the catastrophe non-hurricane losses, ALAE and ULAE amounts are excessive and not supported. According to OIR, the particular time period from 1992 to 2006 used to calculate these values has not been justified, and there has been no explanation of why the extraordinarily high reported losses for 1992 and 1993 should be expected to occur in the future. OIR's complaint with respect to non-hurricane losses is based upon the number of years of data included. While the RMS model was used for hurricane losses, there is no model for non- hurricane losses, so Hartford used its historical data. This becomes important because in both 1992 and 1993, there were unusual storms that caused significant losses. Hartford's data begins with 1992 and goes through 2006, which means approximately fifteen years worth of data is used. Hartford's explanation for choosing that time period is that hurricane models were first used in 1992, and it was at that time that non-hurricane losses had to be separated from hurricane losses. Thus, it was the first year that Hartford had the data in the right form and sufficient detail to use in a rate filing. Petitioners have submitted rate filings in the past that begin non-hurricane, ALAE and ULAE losses with 1992, increasing the number of years included in the data with each filing. Prior filings using this data have been approved by OIR. It is preferable to use thirty years of experience for this calculation. However, there was no testimony that such a time-frame is actuarially or statutorily required, and OIR's suggestion that these two high-loss years should be ignored is not based upon any identified actuarial standard. Hartford attempted to mitigate the effect of the severe losses in 1992 and 1993 by capping the losses for those years, as opposed to relying on the actual losses.2/ The methodology used by Hartford was reasonable and appropriate. No other basis was identified by the OIR to support this stated deficiency. The seventh identified deficiency is that the underwriting profit and contingency factors are excessive and not supported. The underwriting profit factor is the amount of income, expressed as a percentage of premium, that an insurance company needs from premium in excess of losses, settlement costs and other expenses in order to generate a fair rate of return on its capital necessary to support its Florida exposures for the applicable line of business. Hartford's proposed underwriting profit factor for its largest homeowners filing is 15.3%. Section 627.062(2)(b), Florida Statutes, contemplates the allowance of a reasonable rate of return, commensurate with the risk to which the insurance company exposes its capital and surplus. Section 627.062(2)(b)4., Florida Statutes, authorizes the adoption of rules to specify the manner in which insurers shall calculate investment income attributable to classes of insurance written in Florida, and the manner in which investment income shall be used in the calculation of insurance rates. The subsection specifically indicates that the manner in which investment income shall be used in the calculation of insurance rates shall contemplate allowances for an underwriting profit factor. Florida Administrative Code Rule 69O-170.003 is entitled "Calculation of Investment Income," and the stated purpose of this rule is as follows: (1) The purpose of this rule is to specify the manner in which insurers shall calculate investment income attributable to insurance policies in Florida and the manner in which such investment income is used in the calculation of insurance rates by the development of an underwriting profit and contingency factor compatible with a reasonable rate of return. (Emphasis supplied). Mr. Schwartz relied on the contents of this rule in determining that the underwriting profit factor in Hartford's filings was too high, in that Florida Administrative Code Rule 69O-170.003(6)(a) and (7) specifies that: (6)(a) . . . An underwriting profit and contingency factor greater than the quantity 5% is prima facie evidence of an excessive expected rate of return and unacceptable, unless supporting evidence is presented demonstrating that an underwriting profit and contingency factor included in the filing that is greater than this quantity is necessary for the insurer to earn a reasonable rate of return. In such case, the criteria presented as determined by criteria in subsection (7) shall be used by the Office of Insurance Regulation in evaluating this supporting evidence. * * * An underwriting profit and contingency factor calculated in accordance with this rule is considered to be compatible with a reasonable expected rate of return on net worth. If a determination must be made as to whether an expected rate of return is reasonable, the following criteria shall be used in that determination. An expected rate of return for Florida business is to be considered reasonable if, when sustained by the insurer for its business during the period for which the rates under scrutiny are in effect, it neither threatens the insurer's solvency nor makes the insurer more attractive to policyholders or investors from a corporate financial perspective than the same insurer would be had this rule not been implemented, all other variables being equal; or Alternatively, the expected rate of return for Florida business is to be considered reasonable if it is commensurate with the rate of return anticipated for other industries having corresponding risk and it is sufficient to assure confidence in the financial integrity of the insurer so as to maintain its credit and, if a stock insurer, to attract capital, or if a mutual or reciprocal insurer, to accumulate surplus reasonably necessary to support growth in Florida premium volume reasonably expected during the time the rates under scrutiny are in effect. Mr. Schwartz also testified that the last published underwriting profit and contingency factor published by OIR was 3.7%, well below what is identified in Hartford's filings. Hartford counters that reliance on the rule is a misapplication of the rule (with no explanation why), is inconsistent with OIR's treatment of the profit factors in their previous filings, and ignores the language of Section 627.062(2)(b)11., Florida Statutes. No evidence was presented to show whether the expected rate of return threatens Hartford's solvency or makes them more attractive to policyholders or investors from a corporate financial perspective than they would have been if Rule 69O- 170.003 was not implemented. Likewise, it was not demonstrated that the expected rate of return for Florida business is commensurate with the rate of return for other industries having corresponding risk and is necessary to assure confidence in the financial integrity of the insurer in order to maintain its credit and to attract capital. While the position taken by OIR with respect to Hartford's filings may be inconsistent with the position taken in past filings, that cannot be determined on this record. The prior filings, and the communications Hartford had with OIR with regard to those filings, are not included in the exhibits in this case. There is no way to determine whether Petitioners chose to present evidence in the context of prior filings consistent with the criteria in Rule 69O-170.003, or whether OIR approved the underwriting profit and contingency factor despite Rule 69O- 170.003. Having an underwriting profit factor that is considered excessive will result in a higher rate indication. Therefore, it is found that the seventh identified deficiency in the Notices of Intent to Disapprove for the homeowners filings and the second identified deficiency in the Notices of Intent to Disapprove for the dwelling/fire filings is sustained. The eighth identified deficiency is that various components underlying the calculation of the underwriting profit and contingency factors, including but not limited to the return on surplus, premium to surplus ratio, investment income and tax rate are not supported or justified. Return on surplus is the total net income that would result from the underwriting income and the investment income contributions relative to the amount of capital that is exposed. Surplus is necessary in addition to income expected from premium, to insure that claims will be paid should losses in a particular year exceed premium and income earned on premium. Hartford's expected return on surplus in these filings is 15%. The return on surplus is clearly tied to the underwriting profit factor, although the percentages are not necessarily the same. It follows, however, that if the underwriting income and contingency factor is excessive, then the return on surplus may also be too high. Hartford has not demonstrated that the return on surplus can stand, independent of a finding that the underwriting profit and contingency factor is excessive. Premium-to-surplus ratio is a measure of the number of dollars of premium Hartford writes relative to the amount of surplus that is supporting that exposure. Hartford's premium-to- surplus ratio in the AARP homeowners filing is 1.08, which means that if Hartford wrote $108 of premium, it would allocate $100 of surplus to support that premium.3/ The premium-to-surplus ratio is reasonable, given the amount of risk associated with homeowners insurance in Florida. The OIR's position regarding investment income and tax rates are related. The criticism is that the filing used a low- risk investment rate based on a LIBOR (London Interbank Offering Rate), which is a standard in the investment community for risk- free or low-risk yield calculations. The filing also used a full 35% income tax rate applied to the yield. Evidence was presented to show that, if the actual portfolio numbers and corresponding lower tax rate were used in the filings, the rate after taxes would be the same. The problem, however, is that Section 627.062(2)(b)4., Florida Statutes, requires the OIR to consider investment income reasonably expected by the insurer, "consistent with the insurer's investment practices," which assumes actual practices. While the evidence at hearing regarding Hartford's investments using its actual portfolio yield may result in a similar bottom line, the assumptions used in the filing are not based on Petitioner's actual investment practices. As a result, the tax rate identified in the filing is also not the actual tax rate that has been paid by Hartford. The greater weight of the evidence indicates the data used is not consistent with the requirements of Section 627.062(2)(b)4., Florida Statutes. Therefore, the eighth deficiency is sustained to the extent that the filing does not adequately support the return on surplus, investment income and tax rate. The ninth identified deficiency is that the underwriting expenses and other expenses are excessive and not supported. Hartford used the most recent three years of actual expense data, analyzed them and made expense selections based on actuarial judgment. The use of the three-year time frame was both reasonable and consistent with common ratemaking practices. Likewise, the commission rates reflected in the agency filings are also reasonable. The tenth identified deficiency is that the non-FHCF (or private) reinsurance costs are excessive and not supported. The criticism regarding private reinsurance purchases is three- fold: 1) that Hartford paid too much for their reinsurance coverage; 2) that Hartford purchases their reinsurance coverage on a nationwide basis as opposed to purchasing coverage for Florida only; and 3) that the percentage of the reinsurance coverage allocated to Florida is too high. Hartford buys private reinsurance in order to write business in areas that are exposed to catastrophes. It buys reinsurance from approximately 40 different reinsurers in a competitive, arm's-length process and does not buy reinsurance from corporate affiliates. Hartford used the "net cost" of insurance in its filings, an approach that is appropriate and consistent with standard actuarial practices. Hartford also used the RMS model to estimate the expected reinsurance recoveries, which are subtracted from the premium costs. Hartford buys private catastrophic reinsurance on a nationwide basis to protect against losses from hurricanes, earthquakes and terrorism, and allocates a portion of those costs to Florida. Testimony was presented, and is accepted as credible, that attempting to purchase reinsurance from private vendors for Florida alone would not be cost-effective. The cost of reinsurance, excluding a layer of reinsurance that covers only the Northeast region of the country and is not reflected in calculating costs for Florida, is approximately $113 million. Hartford retains the first $250 million in catastrophe risk for any single event, which means losses from an event must exceed that amount before the company recovers from any reinsurer. In 2006, Hartford raised its retention of losses from $175 million to $250 million in an effort to reduce the cost of reinsurance. Hartford purchases reinsurance in "layers," which cover losses based on the amount of total losses Hartford incurs in various events. Hartford allocates approximately 65% of the private reinsurance costs (excluding the Northeast layer) to Florida in the AARP homeowners filing. Only 6-7% of Hartford's homeowners policies are written in Florida. The amount Hartford paid for reinsurance from private vendors is reasonable, given the market climate in which the insurance was purchased. Hartford has demonstrated that the process by which the reinsurance was purchased resulted in a price that was clearly the result of an arms-length transaction with the aim of securing the best price possible. Likewise, the determination to purchase reinsurance on a nationwide basis as opposed to a state-by-state program allows Hartford to purchase reinsurance at a better rate, and is more cost-effective. Purchasing reinsurance in this manner, and then allocating an appropriate percentage to Florida, is a reasonable approach. With respect to the allocation of a percentage of reinsurance cost to Florida, OIR argues that, given that Florida represents only 6-7% of Hartford's homeowner insurance business, allocation of 65% of the reinsurance costs to Florida is per se unreasonable. However, the more logical approach is to examine what percentage of the overall catastrophic loss is attributable to Florida, and allocate reinsurance costs accordingly. After carefully examining both the testimony of all of the witnesses and the exhibits presented in this case, the undersigned cannot conclude that the allocation of 65% of the private reinsurance costs is reasonable, and will not result in an excessive rate.4/ The eleventh identified deficiency is that the FHCF (or CAT Fund) reinsurance costs are excessive and not supported. Hartford purchases both the traditional layer of CAT Fund coverage, which is addressed in a separate filing and not reflected in these filings, and the TICL layer made available pursuant to Chapter 2007-1, Laws of Florida. Hartford removed the costs of its previously purchased private reinsurance that overlapped with the TICL layer and those costs are not reflected in these filings and have not been passed on to Florida policyholders. In estimating the amount of premium Hartford would pay for the TICL coverage, it relied on information provided by Paragon, a consulting firm that calculates the rates for the CAT Fund. As noted in finding of fact number 31, the RMS model, along with three other models accepted by the Hurricane Commission, were used by Paragon for determining expected aggregate losses to the CAT Fund reinsurance layer, clearly a crucial factor in determining the rate for the CAT fund. Hartford did not use the loss recoveries calculated by Paragon, but instead estimated the total amount of premium it would pay for the TICL coverage and subtracted the expected loss recoveries based on the RMS model alone. The expected loss recoveries under the RMS model standing alone were 60% of the loss recovery estimate calculated by Paragon when using all four models. Hartford claimed that its use of the RMS model was necessary for consistency. However, it pointed to no actuarial standard that would support its position with respect to this particular issue. Moreover, given that the premium used as calculated by Paragon used all four models, it is actually inconsistent to use one number which was determined based on all four models (the Paragon-based premium estimate) for one half of this particular calculation and then subtract another number using only one model for the other half (the loss recoveries rate) in order to determine the net premium. To do so fails to take into account the unique nature of the CAT fund, in terms of its low expenses and tax-exempt status. Accordingly, it is found that the CAT-Fund reinsurance costs for the TICL layer are excessive. The twelfth identified deficiency is that Hartford did not consider in the filing that no new business is being written. OIR's explanation of this asserted deficiency is that the costs associated with writing new business are generally higher than that associated with writing renewals. Therefore, according to OIR, failure to make adjustments to their historical experience to reflect the current mix of business, means that the costs included in the filing would be excessive. Hartford began restricting the writing of new business for these filings in 2002. Ultimately, no new business for the AARP program was written after November 2006 and no new business was written for the agency program after June 2006. Credible evidence was presented to demonstrate that a very low percentage of new business has been written over the period of time used for demonstrating Hartford's historical losses. As a result, the effect of no longer writing new business is already reflected in the data used to determine expenses. No additional adjustment in the filing was necessary in this regard. The thirteenth identified deficiency is that no explanation has been provided as to why Hartford believes it is reasonable to return such a low percentage of premium in the form of loss payments to policyholders. For example, for the building policy forms, OIR states that only about 40% of the premium requested by Hartford is expected to be returned to policyholders in the form of loss payments. OIR pointed to no actuarial standard that would require a specific explanation regarding how much of the premium should be returned to policyholders. Nor was any statutory or rule reference supplied to support the contention that such an explanation was required. Finally, the more credible evidence presented indicates that the correct percentage is 44%. In any event, this criticism is not a basis for finding a deficiency in the filing. Alleged Deficiencies in the Dwelling/Fire Filings The seventh deficiency identified in the dwelling/fire filings, not reflected in the homeowner filings, is that the credibility standard and credibility values are not supported. Credibility is the concept of identifying how much weight to put on a particular set of information relative to other potential information. Credibility value is determined by applying the "square root rule" to the credibility value, a commonly used actuarial approach to credibility. Hartford used the credibility standard of 40,000 earned house years in these filings. This credibility standard has been the standard within the industry for personal property filings for over forty years and has been used in prior filings submitted to OIR. Mr. Schwartz testified that his criticism with respect to the credibility standard and credibility values is that Hartford did not explain why they used that particular standard. However, Florida Administrative Code Rule 69O-170.0135 discusses those items that must be included in the Actuarial Memorandum for a filing. With respect to credibility standards and values, Rule 69O-170.0135(2)(e)5., provides that the basis need only be explained when the standard has changed from the previous filing. Given that no change has been made in these filings with respect to the credibility standard, this criticism is not a valid basis for issuing a Notice of Intent to Disapprove. The ninth deficiency in the Notice relating to the dwelling/fire filing in Case No. 07-5187 provides: "No explanation has been provided as too (sic) why Hartford believes it needs such a large rate increase currently, when the cumulative rate change implemented by Hartford for this program from 2001 to 2006 was an increase of only about 10%." With respect to Case No. 07-5188, the deficiency is essentially the same, except the cumulative rate change identified for the same period of time is a decrease of about -3%. Testimony established that the dwelling/fire rate increases were larger than those identified for the homeowners filings because Hartford did not seek rate increases for these lines for several years. The decision not to seek increases was not based on the adequacy of current rates. Rather, the decision was based on an internal determination that, based on the relatively small number of policies involved in these two filings, the amount of increased premium reflected in a rate increase was not sufficient to incur the costs associated with preparing the filings. Mr. Schwartz pointed to no authority, either in statute, rule, or Actuarial Standard, that requires the explanation he desired. He acknowledged that he understood the basis of how Hartford reached the rate increase they are requesting. The failure to provide the explanation Mr. Schwartz was seeking is not a valid basis for a Notice of Intent to Disapprove.

Recommendation Upon consideration of the facts found and conclusions of law reached, it is RECOMMENDED: That a final order be entered that disapproves the rate filings in Case Nos. 07-5185 and 07-5186 based upon the deficiencies numbered 7,8,10 and 11 in the Notices of Intent to Disapprove, and that disapproves the rate filings in Case Nos. 07-5187 and 07-5188 based on the deficiencies numbered 2,3,5 and in the Notices of Intent to Disapprove. DONE AND ENTERED this 28th day of March 2008, in Tallahassee, Leon County, Florida. S LISA SHEARER NELSON Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 28th day of March, 2008.

Florida Laws (6) 120.569120.57215.555627.0613627.062627.0628 Florida Administrative Code (3) 69O-170.00369O-170.01369O-170.0135
# 5
DEPARTMENT OF INSURANCE AND TREASURER vs. CHARLES LEE ARMSTRONG, JR., A/K/A JACK ARMSTRONG, 78-001075 (1978)
Division of Administrative Hearings, Florida Number: 78-001075 Latest Update: Nov. 14, 1978

Findings Of Fact Charles Lee Armstrong, a/k/a Jack Armstrong, (hereinafter referred to as Petitioner or Armstrong) is licensed by the Florida Insurance Department as a general lines agent to represent Foremost Insurance Company and Fortune Insurance Company (Exhibit 1). Prior to 1976 Armstrong was an Aetna agent. From February 10, 1968 through February 10, 1977 Luigi Sesti carried homeowners policy with Aetna with Armstrong Agency. Armstrong's designation as an Aetna agent was terminated by Aetna termination notice (Exhibit 8) dated August 21, 1975 for low volume of business. The company practice is to terminate the agency relationship ninety days after notice of termination. Thereafter Respondent continued as a limited company agent for one year, during which he was authorized to renew Aetna policies. (Exhibit 7). After that one year extension, Respondent had no agency relationship with Aetna and, to renew an Aetna policy, he would have to have an Aetna agent process the renewal. Luigi Sesti had dealt with Armstrong as Sesti's Insurance agent since 1968 and had maintained an Aetna home-owner's policy which had last been renewed through Armstrong for the year ending February 10, 1977. Upon receipt of notice from Armstrong that his policy would expire February 10, 1977, Sesti sent Armstrong his check in the amount of $165 (Exhibit 3) for renewal of his policy. Although Armstrong was no longer authorized to renew Aetna policies, he deposited Sesti's check but thereafter failed to provide Sesti with insurance coverage on his house or contents. Armstrong advised Aetna that Sesti's policy had been replaced with an Eastern insurance policy, and Aetna failed to notify Sesti that the Aetna policy was not renewed. In August 1977 Sesti's home was burglarized. He lost a television set, radio, watch, spotlight and a ring, and Sesti contacted Armstrong to report the loss. Armstrong visited the home and suggested Sesti submit no formal claim because to do so would make it difficult for Sesti to renew his insurance. In his own explanation, Armstrong testified that he intended to pay Sesti for his loss but Sesti could never establish the value of the ring or establish a price for which he would settle. Armstrong offered Sesti $250 to settle the claim. During the discussions between Armstrong and Mrs. Sesti, Armstrong said he had authority to settle claims for Aetna up to $500 and that he was an attorney. Neither of these statements was true. When Armstrong was unable to agree on the amount of the claim, Mrs. Sesti contacted Aetna and learned that the policy on her her had expired 10 February 1977 and had not been renewed. Because no valid policy had been issued to Sesti, Aetna initially denied liability. When advised by Sesti that Aetna would not pay their claim, Armstrong returned the premium he had received from Sesti for the policy not renewed in one check for $155 dated 9/7/77 and in another check for $10 dated 11/23/77 (Exhibit 5) which Sesti received with a letter from the Insurance Commissioner's office dated November 29, 1978 (Exhibit 14). After further investigation by Aetna revealed the facts as noted above, Aetna issued a policy (Exhibit 15) which effectively renewed Sesti's homeowners policy for one year from February 10, 1977. They deducted the premium and the $100 deductible from the amount they paid Sesti for the loss sustained. Aetna's Regional Manager testified that Aetna paid for the loss because Sesti had been insured by them for several years and they felt a moral obligation for their former agent's failure to provide coverage and for their failure to notify Sesti he was no longer insured by Aetna. Aetna allowed Sesti approximately $450 for the loss of the ring and approximately $350 for the other things stolen. Roseland S. Wood had insured her mobile home with Foremost Insurance Company since 1953, and with Jack Armstrong as Agent since 1964. Policy No. 101-8498757 covered the period 11/3/74 to 11/3/75 (Exhibit 13). By check dated November 5, 1975 made payable to Armstrong (Exhibit 9) Wood forwarded the premium for renewal of this policy. Unbeknownst to Wood the policy was not renewed until July 28, 1976 by policy No. 8498643 (Exhibit 12). This is the policy that Armstrong forwarded to Foremost. Armstrong was in Europe on vacation when this policy was issued by the woman he had hired to keep his office open during his vacation and he professed no knowledge of why the policy was issued at this particular time. In October 1976 Wood wanted additional coverage and Armstrong came out to assist in providing the additional coverage. After discussing increasing personal property coverage, plus garage and contents and boats, Respondent advised Wood that the additional coverage would cost $326. Wood gave Respondent a check that day (Exhibit 10). Thereafter Armstrong issued policy No. 8498518 (Exhibit 11) for the period 10/28/76 to 10/28/77 but the personal property coverage was less than Wood had asked for and the garage and contents were not included. Neither Exhibit 11 nor the premium for this coverage was ever received by Foremost from Armstrong. They became aware of Exhibit 11 after Wood suffered a burglary in July 1977 and came to the Foremost office to file a claim. The costs of coverage on Exhibit 11 are not correct and had this policy been received by Foremost it would have been rejected by the computer due to inaccurate premium charges, the inclusion of boats on this policy and incorrect comprehensive liability coverage. By failing to renew Wood's coverage in November 1975, Respondent left Wood without coverage until Exhibit 12 was issued providing coverage from 7/28/76. This renewal was written by Armstrong Agency, who had authority from Foremost to write this renewal. As noted above, this policy was written while Armstrong was on vacation. The $145 premium paid by Wood for the renewal of the policy was not remitted to Foremost until after July 28, 1976. At the time of Wood's loss in July 1977 she was covered by this policy. When the existence of the above facts regarding the two policies and dates they were issued to Wood were uncovered, Armstrong refunded to Wood $181 of the $326 premium he collected, Foremost refunded the additional $145 of this premium to Wood, and Wood's claim was settled by Foremost to Wood's satisfaction. Foremost has a claim against Armstrong for this $145 Foremost refunded to Wood. Respondent acknowledged writing Exhibit 11 and assumed that it was mailed to Foremost. He does not remit payment to the company until he is billed. Foremost sends a monthly statement to each agent showing policy numbers received. The agent can readily check this list against the policies he has issued to ascertain if a policy was not received by the company. The company also maintains a policy register where policy numbers are recorded. A copy of this is sent to their agents to check against policies the agents have issued. Failure of the agencies to submit policies in sequential numbers will be picked up on the computer, but only after quite a few numbers have been skipped. There was insufficient volume from Armstrong's agency to trigger this information from the computer. With respect to Charge III, failure to keep office open and accessible to the public during office hours, an insurance investigator visited the office on some six occasions in December 1977 and February and March 1978. At these visits the office was open but neither Armstrong nor a secretary was present. A lady working in an office down the hall from Respondent's office came to the office when the inspector arrived and offered to contact Armstrong. Several telephone calls made to Armstrong's office during March 1978 resulted in the phone being answered by an answering service. Respondent has operated a one-man office for many years and has an answering service cover all calls while he is out of the office. He wears a radio pager and claims his answering service can always contact him. The lady who covers office visits for Respondent during his absence from the office has had several years experience working in a general insurance agency. She fills out applications for clients coming into the office, gives receipts for payments, signs Armstrong's name to applications and other documents; and has done so for 4 or 5 years. She is not on any type of regular salary or otherwise employed by Armstrong. Respondent has been a licensed insurance agent since 1961 and Respondent's testimony was unrebutted. This is the first complaint filed against him in his capacity as a licensed insurance agent.

Florida Laws (6) 626.561626.611626.621626.731626.9521626.9541
# 6
HARTFORD INSURANCE COMPANY OF THE MIDWEST vs OFFICE OF INSURANCE REGULATION, 07-005187 (2007)
Division of Administrative Hearings, Florida Filed:Tallahassee, Florida Nov. 09, 2007 Number: 07-005187 Latest Update: Jun. 03, 2008

The Issue Whether Petitioners' proposed rates are justified pursuant to the requirements of Section 627.062, Florida Statutes, or whether the Department of Financial Services, Office of Insurance Regulation (OIR) was correct in denying the requested rate increases.

Findings Of Fact The Hartford companies are property and casualty insurers transacting insurance in the State of Florida pursuant to valid certificates of authority and the Florida Insurance Code. Two types of personal lines insurance filings submitted by Hartford to the OIR are at issue in this proceeding: two filings for homeowners insurance (Case Nos. 07-5185 and 07-5186) and two filings for dwelling fire insurance (Case Nos. 07-5187 and 07- 5188). Hartford's substantial interests are affected by the notices disapproving the filings in this case. Homeowners insurance includes coverage for a variety of perils in and around a home, is usually purchased by a homeowner, and covers both the structure and the contents of a home. Dwelling/fire insurance is usually purchased by the owners of properties that are leased or rented to others, and provides coverage for the structure only. Both types of insurance cover damage caused by hurricanes. The New Legislation and its Requirements In a special session held in January 2007, the Florida Legislature enacted changes to the Florida Hurricane Catastrophe Fund (CAT Fund), as reflected in Chapter 2007-1, Laws of Florida. The special session was precipitated by a perceived crisis regarding the cost and availability of homeowners insurance after the 2004 and 2005 hurricane seasons. As a result of the substantial number of claims incurred after multiple severe hurricanes each of these years, changes in the insurance marketplace resulted in some insurance companies withdrawing from the Florida market, others non-renewing policies, one company becoming insolvent, and the cost for reinsurance available to all insurers rising dramatically. One of the primary features of the legislation was an expansion of the CAT Fund. The CAT Fund was established in 1993 after Hurricane Andrew to provide reinsurance to insurers for property insurance written in Florida at a price significantly less than the private market. The CAT Fund is a non-profit entity and is tax exempt. Prior to the enactment of Chapter 2007-1, the CAT Fund had an industry-wide capacity of approximately $16 million. The purpose of the changes enacted by the Legislature was to reduce the cost of reinsurance and thereby reduce the cost of property insurance in the state. As a result of Chapter 2007-1, the industry-wide capacity of the CAT Fund was increased to $28 billion, and insurers were given an opportunity to purchase an additional layer of reinsurance, referred to as the TICL layer (temporary increase in coverage limit), from the CAT Fund. Section 3 of Chapter 2007-1 required insurers to submit a filing to the OIR for policies written after June 1, 2007, that took into account a "presumed factor" calculated by OIR and that purported to reflect savings created by the law. The new law delegated to the OIR the duty to specify by Order the date such filings, referred to as "presumed factor filings" had to be made. On February 19, 2007, the OIR issued Order No. 89321-07. The Order required insurers to make a filing by March 15, 2007, which either adopted presumed factors published by the OIR or used the presumed factors and reflected a rate decrease taking the presumed factors into account. The presumed factors were the amounts the OIR calculated as the average savings created by Chapter 2007-1, and insurers were required to reduce their rates by an amount equal to the impact of the presumed factors. The OIR published the presumed factors on March 1, 2007. In its March 15, 2007, filings, Hartford adopted the presumed factors published by OIR. As a result, Hartford reduced its rates, effective June 1, 2007, on the products at issue in these filings by the following percentages: Case No. 07-5185 homeowners product: 17.7% Case No. 07-5186 homeowners product: 21.9% Case No. 07-5187 dwelling/fire product: 8.7% Case No. 07-5188 dwelling/fire product: 6.2% The Order also required that insurers submit a "True-Up Filing" pursuant to Section 627.026(2)(a)1., Florida Statutes. The filing was to be a complete rate filing that included the company's actual reinsurance costs and programs. Hartford's filings at issue in these proceedings are its True-Up Filings. The True-Up Filings Hartford submitted its True-Up filings June 15, 2007. The rate filings were certified as required by Section 627.062(9), Florida Statutes. The filings were amended August 8, 2007. Hartford's True Up Filings, as amended, request the following increases in rates over those reflected in the March 15, 2007, presumed factor filings: Case No. 07-5185 homeowners product: 22.0% Case No. 07-5186 homeowners product: 31.6% Case No. 07-5187 dwelling and fire product: 69.0% Case No. 07-5188 dwelling and fire product: 35.9% The net effects of Hartford's proposed rate filings result in the following increases over the rates in place before the Presumed Factor Filings: Case No. 07-5185 homeowners product: .4% Case No. 07-5186 homeowners product: 2.8% Case No. 07-5187 dwelling/fire product: 54.3% Case No. 07-5188 dwelling/fire product: 27.5% Case Nos. 07-5185 and 07-5186 (homeowners) affect approximately 92,000 insurance policies. Case Nos. 07-5187 and 07-5188 (dwelling/fire) affect approximately 2,550 policies. A public hearing was conducted on the filings August 16, 2007. Representatives from Hartford were not notified prior to the public hearing what concerns the OIR might have with the filings. Following the hearing, on August 20, 2007, Petitioners provided by letter and supporting documentation additional information related to the filings in an effort to address questions raised at the public hearing. The OIR did not issue clarification letters to Hartford concerning any of the information provided or any deficiencies in the filings before issuing its Notices of Intent to Disapprove the True-Up Filings. All four filings were reviewed on behalf of the OIR by Allan Schwartz. Mr. Schwartz reviewed only the True-Up Filings and did not review any previous filings submitted by Hartford with respect to the four product lines. On September 10, 2007, the OIR issued Notices of Intent to Disapprove each of the filings at issue in this case. The reasons give for disapproving the two homeowners filings are identical and are as follows: Having reviewed the information submitted, the Office finds that this filing does not provide sufficient documentation or justification to demonstrate that the proposed rate(s) comply with the standards of the appropriate statute(s) and rules(s) including demonstrating that the proposed rates are not excessive, inadequate, or unfairly discriminatory. The deficiencies include but are not limited to: The premium trends are too low and are not reflective of the historical pattern of premium trends. The loss trends are too high and are not reflective of the historical pattern of loss trends. The loss trends are based on an unexplained and undocumented method using "modeled" frequency and severity as opposed to actual frequency and severity. The loss trends are excessive and inconsistent compared to other sources of loss trends such as Fast Track data. The catastrophe hurricane losses, ALAE and ULAE amounts are excessive and not supported. The catastrophe non-hurricane losses, ALAE and ULAE amounts are excessive and not supported. The particular time period from 1992 to 2006 used to calculate these values has not been justified. There has been no explanation of why the extraordinarily high reported losses for 1992 and 1993 should be expected to occur in the future. The underwriting profit and contingency factors are excessive and not supported. Various components underlying the calculation of the underwriting profit and contingency factors, including but not limited to the return on surplus, premium to surplus ratio, investment income and tax rate are not supported or justified. The underwriting expenses and other expenses are excessive and not supported. The non-FHCF reinsurance costs are excessive and not supported. The FHCF reinsurance costs are excessive and not supported. The fact that no new business is being written has not been taken into account. No explanation has been provided as too [sic] Hartford believes it is reasonable to return such a low percentage of premium in the form of loss payments to policyholders. For example, for the building policy forms, only about 40% of the premium requested by Hartford is expected to be returned to policyholders in the form of loss payments. As a result of the deficiencies set forth above, the Office finds that the proposed rate(s) are not justified, and must be deemed excessive and therefore, the Office intends to disapprove the above-referenced filing. The Notices of Intent to Disapprove the two dwelling/fire filings each list nine deficiencies. Seven of the nine (numbers 1-6 and 8) are the same as deficiencies listed for the homeowners filings. The remaining deficiencies named for Case No. 07-5187 are as follows: 7. The credibility standard and credibility value are not supported. 9. No explanation has been provided as too (sic) why Hartford believes it needs such a large rate increase currently, when the cumulative rate change implemented by Hartford for this program from 2001 to 2006 was an increase of only about 10%. The deficiencies listed for Case No. 07-5188 are the same as those listed for Case No. 07-5187, with the exception that with respect to deficiency number 9, the rate change implemented for the program in Case No. 07-5188 from 2001 to 2006 was a decrease of about -3%. Documentation Required for the Filings Florida's regulatory framework, consistent with most states, requires that insurance rates not be inadequate, excessive, or unfairly discriminatory. In making a determination concerning whether a proposed rate complies with this standard, the OIR is charged with considering certain enumerated factors in accordance with generally accepted and reasonable actuarial techniques. Chapter 2007-1 also amended Section 627.062, Florida Statutes, to add a certification requirement. The amendment requires the chief executive officer or chief financial officer and chief actuary of a property insurer to certify under oath that they have reviewed the rate filing; that to their knowledge, the rate filing does not contain any untrue statement of a material fact or omit to state a material fact necessary to make the statements made, in light of the circumstances under which the statements were made, not misleading; that based on their knowledge, the information in the filing fairly presents the basis of the rate filing for the period presented; and that the rate filing reflects all premium savings reasonably expected to result from legislative enactments and are in accordance with generally accepted and reasonable actuarial techniques. § 627.062(9)(a), Fla. Stat. (2007). Actuarial Standards of Practice 9 and 41 govern documentation by an actuary. Relevant sections of Standard of Practice 9 provide: Extent of documentation - . . . Appropriate records, worksheets, and other documentation of the actuary's work should be maintained by the actuary and retained for a reasonable length of time. Documentation should be sufficient for another actuary practicing in the same field to evaluate the work. The documentation should describe clearly the sources of data, material assumptions, and methods. Any material changes in sources of data, assumptions, or methods from the last analysis should be documented. The actuary should explain the reason(s) for and describe the impact of the changes. Prevention of misuse - . . . The actuary should take reasonable steps to ensure that an actuarial work product is presented fairly, that the presentation as a whole is clear in its actuarial aspects, and that the actuary is identified as the source of the actuarial aspects, and that the actuary is available to answer questions.. . . . * * * 5.5 Availability of documentation- Documentation should be available to the actuary's client or employer, and it should be made available to other persons when the client or employer so requests, assuming appropriate compensation, and provided such availability is not otherwise improper. . . . In determining the appropriate level of documentation for the proposed rate filings, Petitioner relied on its communications with OIR, as well as its understanding of what has been required in the past. This reliance is reasonable and is consistent with both the statutory and rule provisions governing the filings. Use of the RMS Catastrophic Loss Projection Model In order to estimate future losses in a rate filing, an insurer must estimate catastrophic and non-catastrophic losses. Hartford's projected catastrophic losses in the filings are based upon information provided from the Risk Management Solutions (RMS) catastrophic loss projection model, version 5.1a. Hartford's actuaries rely on this model, consistent with the standards governing actuarial practice, and their reliance is reasonable. Catastrophe loss projection models may be used in the preparation of insurance filings, if they have been considered by and accepted by the Florida Commission on Hurricane Loss Projection Methodology (the Hurricane Commission). The Hurricane Commission determined that the RMS model, version 5.1a was acceptable for projecting hurricane loss costs for personal residential rate filings on May 17, 2006. In addition to approval by the Hurricane Commission, use of the model is appropriate "only if the office and the consumer advocate appointed pursuant to s. 627.0613 have access to all of the assumptions and factors that were used in developing the actuarial methods, principles, standards, models, or output ranges, and are not precluded from disclosing such information in a rate proceeding." §627.0628(3)(c), Fla. Stat. Both the Consumer Advocate and a staff person from the OIR are members of the Hurricane Commission. In that context, both have the ability to make on-site visits to the modeling companies, and to ask any questions they choose regarding the models. Both OIR's representative and the Consumer Advocate participated in the meetings and had the same opportunity as other commissioners to ask any question they wished about RMS 5.1a. The Hurricane Commission members, including the Consumer Advocate, clearly have access to the information identified in Section 627.0628(3)(c). However, there are restrictions on the Hurricane Commission members' ability to share the information received regarding trade secrets disclosed by the modeling companies. For that reason, the Commission's deliberations are not, standing alone, sufficient to determine that the Office of Insurance Regulation has access. In this case, credible evidence was submitted to show that RMS officials met with staff from the Office in July and October 2006 to discuss the model. RMS offered to provide any of its trade secret information to the OIR, subject to a non- disclosure agreement to protect its dissemination to competitors. RMS also opened an office in Tallahassee and invited OIR staff to examine any parts of the model they wished. In addition, both RMS and Hartford have answered extensive questionnaires prepared by OIR regarding the RMS model, and Hartford has offered to assist OIR in gathering any additional information it requires. Most of the questions posed by OIR involve the same areas reviewed by the Commission. RMS' representative also testified at hearing that RMS would not object to disclosure of the assumptions during the hearing itself if necessary. Finally, OIR Exhibit 1 is the Florida Hurricane Catastrophe Fund 2007 Ratemaking Formula Report. The Executive Summary from the report explains how rates were recommended for the Florida Hurricane Catastrophic Fund (CAT Fund) for the 2007- 2008 contract year. The report stated that the RMS model, as well as three other models accepted by the Hurricane Commission, were used for determining expected aggregate losses to the CAT Fund reinsurance layer. Three models, including the RMS model, were also used for analysis of detailed allocation to type of business, territory, construction and deductible, as well as special coverage questions. The models were compared in detail and given equal weight. The report notes that these three models were also used in 1999-2006 ratemaking. The report is prepared by Paragon Strategic Solutions, Inc., an independent consultant selected by the State Board of Administration, in accordance with Section 215.555(5), Florida Statutes. While OIR did not prepare the report, they show no hesitation in accepting and relying on the report and the modeled information it contains in these proceedings. Indeed, one of OIR's criticisms is Hartford's failure to use the report with respect to CAT Fund loss recovery estimates. Based upon the evidence presented at hearing, it is found that the OIR and Consumer Advocate were provided access to the factors and assumptions used in the RMS model, as contemplated by Section 627.0628. The Alleged Deficiencies in the Homeowners Filings1/ A rate is an estimate of the expected value of future costs. It provides for all costs associated with the transfer of risk. A rate is reasonable and not excessive, inadequate or unfairly discriminatory if it is an actuarially sound estimate of the expected value of all future costs associated with an individual risk transfer. In preparing a filing, an actuary identifies the time period that its proposed rates are expected to be in effect. Because ratemaking is prospective, it involves determining the financial value of future contingent events. For the rate filings in question, actuaries for Hartford developed their rate indications by first considering trended premium, which reflects changes in premium revenue based on a variety of factors, including construction costs and the value of the buildings insured. Trended premium is the best estimate of the premium revenue that will be collected if the current rates remain in effect for the time period the filing is expected to be in place. Expenses associated with writing and servicing the business, the reinsurance costs to support the business and an allowance for profit are subtracted from the trended premium. The remainder is what would be available to pay losses. This approach to ratemaking, which is used by Hartford, is a standard actuarial approach to present the information for a rate indication. As part of the process, expected claims and the cost to service and settle those claims is also projected. These calculations show the amount of money that would be available to pay claims if no changes are made in the rates and how much increased premium is necessary to cover claims. The additional amount of premium reflects not only claims payments but also taxes, licenses and fees that are tied to the amount of premium. The first deficiency identified by OIR is that "the premium trends are too low and are not reflective of the historical pattern of premium trends." In determining the premium trend in each filing, Hartford used data from the previous five years and fit an exponential trend to the historical pattern, which is a standard actuarial technique. Hartford also looked at the factors affecting the more recent years, which were higher. For example, the peak in premium trend in 2006 was a result of the cost increases driven by the 2004 and 2005 hurricanes, and the peak in demand for labor and construction supplies not matched by supply. Costs were coming down going into 2007, and Hartford believed that 2006 was out of pattern from what they could anticipate seeing in the future. The premium trends reflected in Hartford's filings are reasonable, reflective of historical patterns, and based on standard actuarial techniques. The second identified deficiency with respect to the homeowner filings was that the loss trends are too high and are not reflective of the historical pattern of loss trends. A loss trend reflects the amount an insurance company expects the cost of claims to change. It consists of a frequency trend, which is the number of claims the insurance company expects to receive, and a severity trend, which is the average cost per claim. The loss trend compares historical data used in the filing with the future time period when the new rates are expected to be in effect. Hartford's loss trends were estimated using a generalized linear model, projecting frequency and severity separately. The model was based on 20 quarters of historical information. The more credible testimony presented indicates that the loss trends were actuarially appropriate. The third identified deficiency is that the loss trends are based on an unexplained and undocumented method using "modeled" frequency and severity as opposed to actual frequency and severity. As noted above, the generalized linear model uses actual, historical data. Sufficient documentation was provided in the filing, coupled with Hartford's August 20, 2007, letter. The method used to determine loss trends is reasonable and is consistent with standard actuarial practice. The fourth identified deficiency is that loss trends are excessive and inconsistent compared to other sources of loss trends, such as Fast Track data. Saying that the loss trends are excessive is a reiteration of the claim that they are too high, already addressed with respect to deficiency number two. Fast Track data is data provided by the Insurance Services Office. It uses unaudited information and is prepared on a "quick turnaround" basis. Fast Track data is based on paid claims rather than incurred claims data, and upon a broad number of companies with different claims settlement practices. Because it relies on paid claims, there is a time lag in the information provided. Hartford did not rely on Fast Track data, but instead relied upon its own data for calculating loss trends. Given the volume of business involved, Hartford had enough data to rely on for projecting future losses. Moreover, Respondents point to no statutory or rule requirement to use Fast Track data. The filings are not deficient on this basis. The fifth identified deficiency in the Notice of Intent to Disapprove is that catastrophe hurricane losses, ALAE and ULAE amounts are excessive and not supported. ALAE stands for "allocated loss adjustment expenses," and represents the costs the company incurs to settle a claim and that can be attributed to that particular claim, such as legal bills, court costs, experts and engineering reports. By contrast, ULAE stands for "unallocated loss adjustment expense" and represents the remainder of claims settlement costs that cannot be linked to a specific claim, such as office space, salaries and general overhead. Part of the OIR's objection with respect to this deficiency relates to the use of the RMS model. As stated above at paragraphs 25-33, the use of the RMS model is reasonable. With respect to ALAE, Hartford analyzed both nationwide data (4.4%) and Florida data (4.8%) and selected an ALAE load between the two (4.6%). This choice benefits Florida policyholders. It is reasonable to select between the national and Florida historical figures, given the amount of actual hurricane data available during the period used. With respect to ULAE, the factors used were based upon directions received from Ken Ritzenthaler, an actuary with OIR, in a previous filing. The prior discussions with Mr. Ritzenthaler are referenced in the exhibits to the filing. The more credible evidence demonstrates that the ALAE and ULAE expenses with respect to catastrophic hurricane losses are sufficiently documented in Hartford's filings and are based on reasonable actuarial judgment. The sixth identified deficiency is that the catastrophe non-hurricane losses, ALAE and ULAE amounts are excessive and not supported. According to OIR, the particular time period from 1992 to 2006 used to calculate these values has not been justified, and there has been no explanation of why the extraordinarily high reported losses for 1992 and 1993 should be expected to occur in the future. OIR's complaint with respect to non-hurricane losses is based upon the number of years of data included. While the RMS model was used for hurricane losses, there is no model for non- hurricane losses, so Hartford used its historical data. This becomes important because in both 1992 and 1993, there were unusual storms that caused significant losses. Hartford's data begins with 1992 and goes through 2006, which means approximately fifteen years worth of data is used. Hartford's explanation for choosing that time period is that hurricane models were first used in 1992, and it was at that time that non-hurricane losses had to be separated from hurricane losses. Thus, it was the first year that Hartford had the data in the right form and sufficient detail to use in a rate filing. Petitioners have submitted rate filings in the past that begin non-hurricane, ALAE and ULAE losses with 1992, increasing the number of years included in the data with each filing. Prior filings using this data have been approved by OIR. It is preferable to use thirty years of experience for this calculation. However, there was no testimony that such a time-frame is actuarially or statutorily required, and OIR's suggestion that these two high-loss years should be ignored is not based upon any identified actuarial standard. Hartford attempted to mitigate the effect of the severe losses in 1992 and 1993 by capping the losses for those years, as opposed to relying on the actual losses.2/ The methodology used by Hartford was reasonable and appropriate. No other basis was identified by the OIR to support this stated deficiency. The seventh identified deficiency is that the underwriting profit and contingency factors are excessive and not supported. The underwriting profit factor is the amount of income, expressed as a percentage of premium, that an insurance company needs from premium in excess of losses, settlement costs and other expenses in order to generate a fair rate of return on its capital necessary to support its Florida exposures for the applicable line of business. Hartford's proposed underwriting profit factor for its largest homeowners filing is 15.3%. Section 627.062(2)(b), Florida Statutes, contemplates the allowance of a reasonable rate of return, commensurate with the risk to which the insurance company exposes its capital and surplus. Section 627.062(2)(b)4., Florida Statutes, authorizes the adoption of rules to specify the manner in which insurers shall calculate investment income attributable to classes of insurance written in Florida, and the manner in which investment income shall be used in the calculation of insurance rates. The subsection specifically indicates that the manner in which investment income shall be used in the calculation of insurance rates shall contemplate allowances for an underwriting profit factor. Florida Administrative Code Rule 69O-170.003 is entitled "Calculation of Investment Income," and the stated purpose of this rule is as follows: (1) The purpose of this rule is to specify the manner in which insurers shall calculate investment income attributable to insurance policies in Florida and the manner in which such investment income is used in the calculation of insurance rates by the development of an underwriting profit and contingency factor compatible with a reasonable rate of return. (Emphasis supplied). Mr. Schwartz relied on the contents of this rule in determining that the underwriting profit factor in Hartford's filings was too high, in that Florida Administrative Code Rule 69O-170.003(6)(a) and (7) specifies that: (6)(a) . . . An underwriting profit and contingency factor greater than the quantity 5% is prima facie evidence of an excessive expected rate of return and unacceptable, unless supporting evidence is presented demonstrating that an underwriting profit and contingency factor included in the filing that is greater than this quantity is necessary for the insurer to earn a reasonable rate of return. In such case, the criteria presented as determined by criteria in subsection (7) shall be used by the Office of Insurance Regulation in evaluating this supporting evidence. * * * An underwriting profit and contingency factor calculated in accordance with this rule is considered to be compatible with a reasonable expected rate of return on net worth. If a determination must be made as to whether an expected rate of return is reasonable, the following criteria shall be used in that determination. An expected rate of return for Florida business is to be considered reasonable if, when sustained by the insurer for its business during the period for which the rates under scrutiny are in effect, it neither threatens the insurer's solvency nor makes the insurer more attractive to policyholders or investors from a corporate financial perspective than the same insurer would be had this rule not been implemented, all other variables being equal; or Alternatively, the expected rate of return for Florida business is to be considered reasonable if it is commensurate with the rate of return anticipated for other industries having corresponding risk and it is sufficient to assure confidence in the financial integrity of the insurer so as to maintain its credit and, if a stock insurer, to attract capital, or if a mutual or reciprocal insurer, to accumulate surplus reasonably necessary to support growth in Florida premium volume reasonably expected during the time the rates under scrutiny are in effect. Mr. Schwartz also testified that the last published underwriting profit and contingency factor published by OIR was 3.7%, well below what is identified in Hartford's filings. Hartford counters that reliance on the rule is a misapplication of the rule (with no explanation why), is inconsistent with OIR's treatment of the profit factors in their previous filings, and ignores the language of Section 627.062(2)(b)11., Florida Statutes. No evidence was presented to show whether the expected rate of return threatens Hartford's solvency or makes them more attractive to policyholders or investors from a corporate financial perspective than they would have been if Rule 69O- 170.003 was not implemented. Likewise, it was not demonstrated that the expected rate of return for Florida business is commensurate with the rate of return for other industries having corresponding risk and is necessary to assure confidence in the financial integrity of the insurer in order to maintain its credit and to attract capital. While the position taken by OIR with respect to Hartford's filings may be inconsistent with the position taken in past filings, that cannot be determined on this record. The prior filings, and the communications Hartford had with OIR with regard to those filings, are not included in the exhibits in this case. There is no way to determine whether Petitioners chose to present evidence in the context of prior filings consistent with the criteria in Rule 69O-170.003, or whether OIR approved the underwriting profit and contingency factor despite Rule 69O- 170.003. Having an underwriting profit factor that is considered excessive will result in a higher rate indication. Therefore, it is found that the seventh identified deficiency in the Notices of Intent to Disapprove for the homeowners filings and the second identified deficiency in the Notices of Intent to Disapprove for the dwelling/fire filings is sustained. The eighth identified deficiency is that various components underlying the calculation of the underwriting profit and contingency factors, including but not limited to the return on surplus, premium to surplus ratio, investment income and tax rate are not supported or justified. Return on surplus is the total net income that would result from the underwriting income and the investment income contributions relative to the amount of capital that is exposed. Surplus is necessary in addition to income expected from premium, to insure that claims will be paid should losses in a particular year exceed premium and income earned on premium. Hartford's expected return on surplus in these filings is 15%. The return on surplus is clearly tied to the underwriting profit factor, although the percentages are not necessarily the same. It follows, however, that if the underwriting income and contingency factor is excessive, then the return on surplus may also be too high. Hartford has not demonstrated that the return on surplus can stand, independent of a finding that the underwriting profit and contingency factor is excessive. Premium-to-surplus ratio is a measure of the number of dollars of premium Hartford writes relative to the amount of surplus that is supporting that exposure. Hartford's premium-to- surplus ratio in the AARP homeowners filing is 1.08, which means that if Hartford wrote $108 of premium, it would allocate $100 of surplus to support that premium.3/ The premium-to-surplus ratio is reasonable, given the amount of risk associated with homeowners insurance in Florida. The OIR's position regarding investment income and tax rates are related. The criticism is that the filing used a low- risk investment rate based on a LIBOR (London Interbank Offering Rate), which is a standard in the investment community for risk- free or low-risk yield calculations. The filing also used a full 35% income tax rate applied to the yield. Evidence was presented to show that, if the actual portfolio numbers and corresponding lower tax rate were used in the filings, the rate after taxes would be the same. The problem, however, is that Section 627.062(2)(b)4., Florida Statutes, requires the OIR to consider investment income reasonably expected by the insurer, "consistent with the insurer's investment practices," which assumes actual practices. While the evidence at hearing regarding Hartford's investments using its actual portfolio yield may result in a similar bottom line, the assumptions used in the filing are not based on Petitioner's actual investment practices. As a result, the tax rate identified in the filing is also not the actual tax rate that has been paid by Hartford. The greater weight of the evidence indicates the data used is not consistent with the requirements of Section 627.062(2)(b)4., Florida Statutes. Therefore, the eighth deficiency is sustained to the extent that the filing does not adequately support the return on surplus, investment income and tax rate. The ninth identified deficiency is that the underwriting expenses and other expenses are excessive and not supported. Hartford used the most recent three years of actual expense data, analyzed them and made expense selections based on actuarial judgment. The use of the three-year time frame was both reasonable and consistent with common ratemaking practices. Likewise, the commission rates reflected in the agency filings are also reasonable. The tenth identified deficiency is that the non-FHCF (or private) reinsurance costs are excessive and not supported. The criticism regarding private reinsurance purchases is three- fold: 1) that Hartford paid too much for their reinsurance coverage; 2) that Hartford purchases their reinsurance coverage on a nationwide basis as opposed to purchasing coverage for Florida only; and 3) that the percentage of the reinsurance coverage allocated to Florida is too high. Hartford buys private reinsurance in order to write business in areas that are exposed to catastrophes. It buys reinsurance from approximately 40 different reinsurers in a competitive, arm's-length process and does not buy reinsurance from corporate affiliates. Hartford used the "net cost" of insurance in its filings, an approach that is appropriate and consistent with standard actuarial practices. Hartford also used the RMS model to estimate the expected reinsurance recoveries, which are subtracted from the premium costs. Hartford buys private catastrophic reinsurance on a nationwide basis to protect against losses from hurricanes, earthquakes and terrorism, and allocates a portion of those costs to Florida. Testimony was presented, and is accepted as credible, that attempting to purchase reinsurance from private vendors for Florida alone would not be cost-effective. The cost of reinsurance, excluding a layer of reinsurance that covers only the Northeast region of the country and is not reflected in calculating costs for Florida, is approximately $113 million. Hartford retains the first $250 million in catastrophe risk for any single event, which means losses from an event must exceed that amount before the company recovers from any reinsurer. In 2006, Hartford raised its retention of losses from $175 million to $250 million in an effort to reduce the cost of reinsurance. Hartford purchases reinsurance in "layers," which cover losses based on the amount of total losses Hartford incurs in various events. Hartford allocates approximately 65% of the private reinsurance costs (excluding the Northeast layer) to Florida in the AARP homeowners filing. Only 6-7% of Hartford's homeowners policies are written in Florida. The amount Hartford paid for reinsurance from private vendors is reasonable, given the market climate in which the insurance was purchased. Hartford has demonstrated that the process by which the reinsurance was purchased resulted in a price that was clearly the result of an arms-length transaction with the aim of securing the best price possible. Likewise, the determination to purchase reinsurance on a nationwide basis as opposed to a state-by-state program allows Hartford to purchase reinsurance at a better rate, and is more cost-effective. Purchasing reinsurance in this manner, and then allocating an appropriate percentage to Florida, is a reasonable approach. With respect to the allocation of a percentage of reinsurance cost to Florida, OIR argues that, given that Florida represents only 6-7% of Hartford's homeowner insurance business, allocation of 65% of the reinsurance costs to Florida is per se unreasonable. However, the more logical approach is to examine what percentage of the overall catastrophic loss is attributable to Florida, and allocate reinsurance costs accordingly. After carefully examining both the testimony of all of the witnesses and the exhibits presented in this case, the undersigned cannot conclude that the allocation of 65% of the private reinsurance costs is reasonable, and will not result in an excessive rate.4/ The eleventh identified deficiency is that the FHCF (or CAT Fund) reinsurance costs are excessive and not supported. Hartford purchases both the traditional layer of CAT Fund coverage, which is addressed in a separate filing and not reflected in these filings, and the TICL layer made available pursuant to Chapter 2007-1, Laws of Florida. Hartford removed the costs of its previously purchased private reinsurance that overlapped with the TICL layer and those costs are not reflected in these filings and have not been passed on to Florida policyholders. In estimating the amount of premium Hartford would pay for the TICL coverage, it relied on information provided by Paragon, a consulting firm that calculates the rates for the CAT Fund. As noted in finding of fact number 31, the RMS model, along with three other models accepted by the Hurricane Commission, were used by Paragon for determining expected aggregate losses to the CAT Fund reinsurance layer, clearly a crucial factor in determining the rate for the CAT fund. Hartford did not use the loss recoveries calculated by Paragon, but instead estimated the total amount of premium it would pay for the TICL coverage and subtracted the expected loss recoveries based on the RMS model alone. The expected loss recoveries under the RMS model standing alone were 60% of the loss recovery estimate calculated by Paragon when using all four models. Hartford claimed that its use of the RMS model was necessary for consistency. However, it pointed to no actuarial standard that would support its position with respect to this particular issue. Moreover, given that the premium used as calculated by Paragon used all four models, it is actually inconsistent to use one number which was determined based on all four models (the Paragon-based premium estimate) for one half of this particular calculation and then subtract another number using only one model for the other half (the loss recoveries rate) in order to determine the net premium. To do so fails to take into account the unique nature of the CAT fund, in terms of its low expenses and tax-exempt status. Accordingly, it is found that the CAT-Fund reinsurance costs for the TICL layer are excessive. The twelfth identified deficiency is that Hartford did not consider in the filing that no new business is being written. OIR's explanation of this asserted deficiency is that the costs associated with writing new business are generally higher than that associated with writing renewals. Therefore, according to OIR, failure to make adjustments to their historical experience to reflect the current mix of business, means that the costs included in the filing would be excessive. Hartford began restricting the writing of new business for these filings in 2002. Ultimately, no new business for the AARP program was written after November 2006 and no new business was written for the agency program after June 2006. Credible evidence was presented to demonstrate that a very low percentage of new business has been written over the period of time used for demonstrating Hartford's historical losses. As a result, the effect of no longer writing new business is already reflected in the data used to determine expenses. No additional adjustment in the filing was necessary in this regard. The thirteenth identified deficiency is that no explanation has been provided as to why Hartford believes it is reasonable to return such a low percentage of premium in the form of loss payments to policyholders. For example, for the building policy forms, OIR states that only about 40% of the premium requested by Hartford is expected to be returned to policyholders in the form of loss payments. OIR pointed to no actuarial standard that would require a specific explanation regarding how much of the premium should be returned to policyholders. Nor was any statutory or rule reference supplied to support the contention that such an explanation was required. Finally, the more credible evidence presented indicates that the correct percentage is 44%. In any event, this criticism is not a basis for finding a deficiency in the filing. Alleged Deficiencies in the Dwelling/Fire Filings The seventh deficiency identified in the dwelling/fire filings, not reflected in the homeowner filings, is that the credibility standard and credibility values are not supported. Credibility is the concept of identifying how much weight to put on a particular set of information relative to other potential information. Credibility value is determined by applying the "square root rule" to the credibility value, a commonly used actuarial approach to credibility. Hartford used the credibility standard of 40,000 earned house years in these filings. This credibility standard has been the standard within the industry for personal property filings for over forty years and has been used in prior filings submitted to OIR. Mr. Schwartz testified that his criticism with respect to the credibility standard and credibility values is that Hartford did not explain why they used that particular standard. However, Florida Administrative Code Rule 69O-170.0135 discusses those items that must be included in the Actuarial Memorandum for a filing. With respect to credibility standards and values, Rule 69O-170.0135(2)(e)5., provides that the basis need only be explained when the standard has changed from the previous filing. Given that no change has been made in these filings with respect to the credibility standard, this criticism is not a valid basis for issuing a Notice of Intent to Disapprove. The ninth deficiency in the Notice relating to the dwelling/fire filing in Case No. 07-5187 provides: "No explanation has been provided as too (sic) why Hartford believes it needs such a large rate increase currently, when the cumulative rate change implemented by Hartford for this program from 2001 to 2006 was an increase of only about 10%." With respect to Case No. 07-5188, the deficiency is essentially the same, except the cumulative rate change identified for the same period of time is a decrease of about -3%. Testimony established that the dwelling/fire rate increases were larger than those identified for the homeowners filings because Hartford did not seek rate increases for these lines for several years. The decision not to seek increases was not based on the adequacy of current rates. Rather, the decision was based on an internal determination that, based on the relatively small number of policies involved in these two filings, the amount of increased premium reflected in a rate increase was not sufficient to incur the costs associated with preparing the filings. Mr. Schwartz pointed to no authority, either in statute, rule, or Actuarial Standard, that requires the explanation he desired. He acknowledged that he understood the basis of how Hartford reached the rate increase they are requesting. The failure to provide the explanation Mr. Schwartz was seeking is not a valid basis for a Notice of Intent to Disapprove.

Recommendation Upon consideration of the facts found and conclusions of law reached, it is RECOMMENDED: That a final order be entered that disapproves the rate filings in Case Nos. 07-5185 and 07-5186 based upon the deficiencies numbered 7,8,10 and 11 in the Notices of Intent to Disapprove, and that disapproves the rate filings in Case Nos. 07-5187 and 07-5188 based on the deficiencies numbered 2,3,5 and in the Notices of Intent to Disapprove. DONE AND ENTERED this 28th day of March 2008, in Tallahassee, Leon County, Florida. S LISA SHEARER NELSON Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 28th day of March, 2008.

Florida Laws (6) 120.569120.57215.555627.0613627.062627.0628 Florida Administrative Code (3) 69O-170.00369O-170.01369O-170.0135
# 7
DEPARTMENT OF INSURANCE vs LARRY RYLAND PARKER, 00-001122 (2000)
Division of Administrative Hearings, Florida Filed:Tallahassee, Florida Mar. 13, 2000 Number: 00-001122 Latest Update: Sep. 22, 2024
# 8
DEPARTMENT OF FINANCIAL SERVICES vs EILEEN P. SUAREZ, 09-005353PL (2009)
Division of Administrative Hearings, Florida Filed:Miami, Florida Oct. 01, 2009 Number: 09-005353PL Latest Update: May 18, 2010

The Issue The issue in this case is whether Respondent committed the offenses alleged by the Department of Financial Services in the Administrative Complaint dated May 27, 2009, and, if so, what penalty should be imposed.

Findings Of Fact Petitioner, the Department of Financial Services ("Petitioner" or "the Department") has regulatory responsibility for Chapter 626, Florida Statutes (2009), the insurance licensing procedures law. Respondent, Eileen P. Suarez ("Respondent" or "Suarez"), is a licensed general lines agent transacting in property and casualty insurance, under license number E129078. She operated and was the agent in charge of the Suarez Insurance Agency, Inc. ("Agency"), in Hialeah, Florida. The Agency held a valid state license from 7/21/2006 to 7/27/2009. The Department filed a three-count Administrative Complaint against Respondent alleging that she violated various provisions of Chapter 626, Florida Statutes. COUNT I John Vila is the president of Vila Home Group, Inc., a trucking company that is in the business of hauling sand, soil, and gravel. In April 2005, he purchased a dump truck and, at the suggestion of the dealer, contacted Suarez for insurance. Suarez sold Vila two insurance policies, for the period April 29, 2005 to April 29, 2006, one with AequiCap Insurance Company ("AequiCap") and the other with the Underwriters at Lloyds, London ("Lloyds"). The AequiCap Policy was a commercial liability insurance policy. The Lloyds Policy was a commercial automobile physical damage insurance policy. In March 2006, Vila gave Suarez a check in the amount of $10,876.41, made payable to the Agency to renew the AequiCap and Lloyds policies, for the period April 29, 2006 to April 29, 2007. The AequiCap policy quote was approximately $5,350.00. The Lloyds policy quote was approximately $5,500.00. The check was deposited in the Agency's trust account, but the Lloyds policy was allowed to expire on April 29, 2006, and was not renewed until October 26, 2006, creating a six-month gap in commercial automobile physical damage insurance coverage for Vila. When it was renewed, the Lloyds Policy cost $5,712.03. Vila's AequiCap policy expired on April 29, 2006, and was not renewed because Suarez failed to pay MAI Risk Management, AequiCap's managing general agent. The funds were not returned to Vila. While the March 2006 quotes were pending, the registered driver of the truck, Andres Vila, was involved in an accident and was at fault for hitting a wire. Rather than risk an increase in the pending insurance quotes, Vila paid Bellsouth $2,390.36 in damages. COUNT II On or about October 26, 2006, Suarez provided Vila a Certificate of Liability showing that the truck was insured with AequiCap, under policy number TC012695, and with Lloyds, under policy number R641440/0251, for the period April 29, 2006 to April 29, 2007. Vila was not insured under AequiCap policy number TC012695 from April 29, 2006 to April 29, 2007. The Certificate of Liability was a false document that Suarez created on her computer, printed, and gave to Vila. COUNT III Shelly, Middlebrooks & O'Leary, Inc. ("Shelly Middlebrooks") is a licensed insurance agency, located in Jacksonville, that acts as a general agent for multiple insurance companies. Suarez collected insufficient funds to include the premiums that were intended to be forwarded to Shelley Middlebrooks for policies to insure the following trucking companies: All Nations Logistics, LLC (Policy Number 486865); Jose Veiga, d/b/a JJ Freightways (Policy Number 486885); Gary Castle/Diamond Mine (Policy Number 74APN338354); and Nics Oil, Inc. (Policy Number 74APN401617). For each of the four companies, she requested and received binders for insurance from Shelly Middlebrooks, followed by invoices for the premiums that were to have been paid within ten days of the date the invoices were received. In each instance, Suarez did not pay Shelly Middlebrooks, which cancelled the policies for non-payment of the premium. It also obtained a default judgment in the Circuit Court in and for Duval County, Florida, that requires Suarez to pay it the outstanding balances due for the four policies and a $25 insufficient funds check fee, for a total of $8,335.60, which she has been unable to pay. Instead of paying for insurance, Suarez used most of the funds she collected to pay for various other corporate expenses for the same trucking companies, including state and federal government filings for intrastate or interstate travel that were prerequisites to their becoming insurable. Suarez expected to collect the additional funds needed for insurance later, but the clients, the owners of the trucking companies, did not pay her. Suarez admits that she failed her clients in 2006, after her father's death in February 2006. She realized the Vila errors and tried to correct them in October. The Agency is now closed. Suarez's husband has been unemployed for over a year, and their home is in foreclosure. She is receiving social security disability payments and has insufficient funds to file for bankruptcy.

Recommendation Based upon the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that a Final Order be entered by the Department of Financial Services: Finding Respondent guilty of violating Subsections 626.611(7), (8) and (10); Subsection 626.561(1); and Subsections 626.621(2) and (6), Florida Statutes, as charged in Count I of the Administrative Complaint; Finding Respondent guilty of violating Subsections 626.611(7) and (8); Subsection 626.621(6); and Subsection 626.9541 (1)(e)1., Florida Statutes, as charged in Count II of the Administrative Complaint; Finding Respondent guilty of violating Subsections 626.611(7), (8) and (10); Subsection 626.561(1); and Subsections 626.621(2) and (6), Florida Statutes, as charged in Count III of the Amended Complaint; Revoking Respondent's licenses and appointments issued or granted under or pursuant to the Florida Insurance Code; Ordering Respondent to make restitution to John Vila in the amount of $5,164.38; and Ordering Respondent to make restitution to Shelly Middlebrooks & O'Leary in the amount of $8,335.60. DONE AND ENTERED this 16th day of February, 2010, in Tallahassee, Leon County, Florida. S ELEANOR M. HUNTER Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 16th day of February, 2010.

Florida Laws (10) 120.569120.57626.561626.611626.621626.692626.753626.9541712.03876.41 Florida Administrative Code (7) 69B-231.04069B-231.08069B-231.09069B-231.10069B-231.11069B-231.12069B-231.160
# 9

Can't find what you're looking for?

Post a free question on our public forum.
Ask a Question
Search for lawyers by practice areas.
Find a Lawyer