Findings Of Fact The Department of Insurance (Respondent) is charged with regulating the business of insurance in the State of Florida. As part of this responsibility, Respondent examines and analyzes rate increases. The scope of proposed Rule 4-166.051, Florida Administrative Code, entitled "Examination of Significant Rate Increases," provides: (2) Scope. This rule applies to residential and habitational, personal and commercial property insurance in the State of Florida . . . . The proposed rule requires Respondent to conduct public hearings on rate increases of insurers under certain specific conditions: (3) Public Hearings. Significant Rate Increases. The Department will hold a public hearing on any rate filing where the percentage of rate increase is 25 percent or more and the aggregate amount of such rate increase is $2,000,000 or more, or a rate increase of 50 percent or more. The Department will hold a public hearing as to any rate filing which appears to have a signi- ficant or disproportionate impact on any group or geographic area. Respondent contends that insurers employing the present methods for rate filings would be able to calculate the increases and would, therefore, have no difficulty in determining whether they meet the threshold in the proposed rule. The proposed rule provides in Section (3) a procedure for the public hearings: Procedure. The time and place of the public hearing will be noticed by order of the Department. The public hearing shall be for the purpose of gathering information and evidence, and is not subject to the procedures of Chapter 120, Florida Statutes. Each insurer shall bear its own costs, including any attorney's fees, which may be associated with this examination and with its attendance at the public hearing. Specifically, the public hearing will provide the Department with, and the insurer shall be prepared to present, information necessary to determine whether: The proposed activity will have a hazardous or detrimental effect upon the residential property insurance market in this State and a specific adverse, hazardous, or detrimental effect upon its policyholders located in this State. The proposed activity violates the terms and conditions of residential property insurance policies and constitutes material misrepresentation, or results in the insurer having unlawfully misrepresented the benefits and promises which induced its policyholders to purchase policies from the insurer. The proposed rating structure, elimination of current policyholders, and overall marketing strategies of the insurer, in relation to current market conditions in this State, render the insurer's rates excessive, inadequate, or unfairly discriminatory. The proposed activity constitutes an arbitrary or capricious act of unfair discrimination against policyholders, and constitutes a practice detrimental to the insurer's policyholders or the insurance buying public. The proposed activity will adversely contribute to a further reduction in the availability of residential property insurance in this State and the ability of the current admitted market to absorb further losses or liabilities. The proposed activity will adversely impact the Residential Property and Casualty Joint Underwriting Association's (RPCJUA) ability to provide coverage and/or service to present or potential insureds. Other relevant impact. Respondent is granted general authority by the Legislature, through Section 624.324, Florida Statutes, to hold public hearings within the scope of the insurance code 1/ whenever it deems such action necessary. Respondent contends that this authorization includes holding public hearings on rate increases which it has deemed necessary to be in the public eye. Section 627.062, Florida Statutes, entitled "Rate standards," provides that no rate shall be "excessive, inadequate, or unfairly discriminatory" and mandates certain specific factors and standards to be considered by Respondent when making a determination whether a rate is excessive, inadequate, or unfairly discriminatory. Respondent contends that the factors and standards in Section 627.062 are not all inclusive but are only some of the factors and standards to be considered; the others are located throughout the insurance code. Respondent contends that interpreting Section 627.062, Florida Statutes, requires a reading of the insurance code as a whole. For example, some terms in Section 627.062 are explained in other parts of the code, so the applicable parts of the code defining and explaining the terms would have to be read in para materia with Section 627.062. The proposed rule is designed, as contended by Respondent, to affect only those insurers which significantly increase their residential property insurance rates; it is not designed to affect all insurers which increase their rates. Also, as part of its design, the proposed rule reflects Respondent's experience with the impact of Hurricane Andrew on the consumer and the insurance industry and with an emergency rule which addressed the same subject of rate increases. However, the emergency rule included lower thresholds in which Respondent attempted to affect only those insurers with significant rate increases. The proposed rule serves, as contended by Respondent, four purposes: to assist Respondent in its statutory duty to report annually to the Legislature; (2) to reveal subtle, but important, factors which affect Respondent's decision in determining whether a proposed rate is excessive; (3) to assist Respondent in uncovering unfair trade practices, if any, in proposed rate increases; and (4) to ameliorate the impact of large rate increases on consumers. As to Respondent's reporting duty, part of Respondent's statutory responsibility is to report the ramifications and implications of large rate hikes to the Legislature. Because of the devastation caused by Hurricane Andrew, an unprecedented number of rate filings for large increases have occurred and a severe availability crisis exists with residential property insurance; if insurance is unaffordable, it is unavailable. Respondent contends that in order to make a determination on the rate filings, it must know the ramifications of the rate hikes on insureds and that public hearings provide an avenue to obtain such information. Regarding the subtle, but important, factors affecting Respondent's determination of whether a proposed rate is excessive, a public hearing, contends Respondent, would allow Respondent to include the effects of the proposed rate increase on the lives of consumers, removing the effect from only the mathematical or academic arena. Respondent contends that obtaining information of such an effect would include an examination of market conditions, which it is authorized to do by Section 627.062, 2/ and which examination includes reviewing marketing techniques, such as advertising, for misrepresentations by insurers. As to assisting Respondent to uncover unfair trade practices, Respondent points to "redlining" as an example which in the context of rate increases would encompass the denial of insurance by an insurer to a certain group of people in a given territory because of the increase, i.e., the increase would cause the insurance to be unaffordable and, therefore, unavailable to a certain group of people within a given territory. The insurance code provides a redlining statute 3/ which addresses redlining as the refusal to insure or to continue to insure a risk solely because of certain enumerated factors. Regarding the amelioration of the impact of large rate increases on consumers, Respondent contends that public hearings would provide a forum for the insurer and consumer or insured to freely exchange information and to educate one another on their respective positions and on the effects of a rate increase on both of them. According to Respondent, consumers have a belief that insurers increase rates unreasonably. Additionally, Respondent contends that, through the public hearing, it could direct consumers to less expensive alternatives. The term "proposed activity" is used in the proposed rule but is not defined by it. Respondent contends that no limits would be placed on the subject matter addressed in the public hearings but that any matter which may have a bearing on any of the items in the proposed rule would be addressed in the public hearings. Pursuant to Section 627.0613, Florida Statutes, the insurance code provides for an Insurance Consumer Advocate who is authorized to examine rate filings and to make a recommendation to Respondent that the Consumer Advocate deems to be in the public interest. There are approximately 1,200 rate filings a year, and the Consumer Advocate is unable to examine each filing, so he has developed a formula for selecting the rate filings for review. A public hearing on a rate filing would be useful to him in executing his statutory function regarding rate filings; however, such a public hearing is not authorized by Section 627.0613. American Insurance Association's standing is not at issue in this proceeding.
The Issue The issue addressed in this proceeding is whether Respondent's life, health, life and health and general lines insurance agent's licenses should be suspended, revoked, or otherwise disciplined for violations of Chapter 626, Florida Statutes.
Findings Of Fact Respondent is currently licensed in this state as a life agent, as a health agent, as a life and health agent, and as a general lines agent, Respondent was so licensed in 1993. In 1993, Respondent was doing business as John Drew Insurance Group (Group). Group was and remains a general lines insurance agency located in Panama City, Florida. Around October 4, 1993, Trilby L. Williams of Panama City, Florida phoned Respondent and sought a quotation for homeowners insurance on her mobile home. Respondent specifically requested a quotation for coverage on the mobile home. She did not seek any other insurance or non-insurance product or service from Respondent. Respondent provided a phone quotation of $490.00. During the conversation Respondent described only coverage under the mobile home policy. He did not describe any other insurance coverage or membership in a travel or auto club. Ms. Williams advised Respondent she would accept the quotation and made arrangements to go to Respondent's agency to complete the transaction. Around October 6, 1993, Ms. Williams and her husband, Robert L. Williams went to the office of the Group. The Williams' met with Dana Baxter Watkins, an unlicensed employee of the agency working under Respondent's direct supervision and control. Respondent was not present. However, Ms. Watkins was expecting the Williams. Respondent had left her the Williams' file and paperwork for the mobile home insurance plus an ancillary product. Mr. Williams signed an application for a mobile home insurance policy to be issued by American International Insurance Company through the Florida Residential Property and Casualty Joint Underwriting Association (FRPCJUA). The premium for the mobile home policy was $400.00 which was tendered to Group by Robert and Trilby Williams. Since Respondent had binding authority with the FRPCJUA, he bound the mobile home coverage effective October 6, 1993, at 9:45 A.M. Ms. Watkins also had Robert L. Williams execute a contract for and purchase an American Travelers Association (ATA) ancillary product without the informed consent of Robert or Trilby Williams. Mr. and Ms. Williams were led to believe through Ms. Watkins' statements that they were required to purchase the ATA ancillary product in order to be eligible to purchase the mobile home coverage. Ms. Watkins advised the Williams that "they didn't make any money on the homeowners' policies and that they had to sell this policy." Ms. Watkins explained the ATA ancillary product as an accident benefit policy. The Williams understood the product to be some sort of accident insurance policy designed to provide Ms. Williams with benefits in the event Mr. Williams were to have an accident around their mobile home. The fee for the ATA ancillary product was $90.00, which was paid to Group by Robert and Trilby Williams on October 6, 1993. Robert and Trilby Williams believed that the entire $490.00 which they paid to Group on October 6, 1993, was premium monies required for the purchase of the mobile home insurance policy. The ATA ancillary product is in fact a "motor club" membership as defined in Section 624.124, Florida Statutes, The motor club provides certain benefits of membership such as rewards to witnesses in the event of the theft of the member's private passenger vehicle, travel agency services, rental car discounts, and very circumscribed accidental death and dismemberment coverage for accidental death and qualifying dismemberments while the insured is a passenger in a private passenger automobile. An example of a qualifying dismemberment is the loss of a finger, but only if the digit is a thumb or index finger and only if it is severed through or above the joint closest to the wrist. The cost of the ATA motor club varies from $20.00 to $100.00 depending on the level of benefits selected. Commissions on such products are approximately eighty (80) to ninety (90) percent of the price charged to the consumer. Shortly after departing Respondent's agency Robert and Trilby Williams became concerned about having been required to purchase the ATA "policy". Ms. Williams telephoned Respondent to request a refund of the $90.00 fee they had paid. Respondent informed Robert and Trilby Williams that they could not purchase the mobile home insurance policy without purchasing the ATA ancillary product and again reiterated that "they make very little or no profit selling homeowners or mobile home . . . insurance and that they needed this in order to make overhead." When Ms. Williams, in a somewhat nasty manner, persisted in her request for a refund of the $90.00 ATA fee and threatened to contact the Department of Insurance, Respondent decided he did not wish to do business with the Williams and informed Ms. Williams that he would refund the entire $490.00. Respondent cancelled the American mobile home policy and refunded the entire $490.00 paid by the Williams. However, there was no question that Respondent would not have sold the mobile home insurance without the ATA club membership. At no time did either Williams' authorize the cancellation of the mobile home policy. As a result of the cancellation, Mr. and Ms. Williams were without an insurance policy on their mobile home for approximately a week. Additionally, because the policy had been bound, the FRPCJUA was exposed to a risk of loss for that period of time for which they received no premium. Around September 13, 1993, Doris Steen of Panama City, Florida phoned Respondent and sought a quotation for homeowners insurance on her mobile home. Respondent specifically requested a quotation for coverage on her mobile home. She did not seek any other insurance or non-insurance product or service from Respondent. Ms. Steen and her husband were members of another auto club and did not need a second membership. Around September 15, 1993, Ms. Steen went to the office of the Group. She was met by Dana Baxter Watkins. Ms. Watkins had Ms. Steen's file and paperwork for mobile home coverage and membership in ATA. Ms. Steen signed an application for a mobile home insurance policy to be issued by American through FRPCJUA. The premium for said policy was $277.00, which was tendered to Group by Doris Steen. The coverages applied for were bound effective September 15, 1993. Ms. Watkins also had Doris Steen purchase an ATA ancillary product without Doris Steen's informed consent. Ms. Steen was led to believe through Ms. Watkins' statements that she was required to purchase the ATA ancillary product as part of a package which included the mobile home policy and the one was not available without the other. Ms. Watkins explained the ATA ancillary product as an accident and life insurance policy and Ms. Steen understood the product to be a life insurance policy to cover her if she were killed at her mobile home. The fee for the ATA ancillary product was $100.00, which was paid to Group by Doris Steen on September 15, 1993. The ATA ancillary product sold to Ms. Steen was the same motor club product sold to the Williams. Dana Baxter Watkins was under the direct supervision of Respondent and was trained by Respondent to sell ATA motor club memberships in accordance with a routine business practice implemented by Respondent to increase his agency revenues. In furtherance of the business practice Respondent developed a chart that indicated what price to charge for the ATA motor club given a particular premium level for the mobile home insurance being purchased. The higher the premium for the insurance the less the charge for the motor club. From the charge for the motor club, Respondent would back into the level of benefits provided rather than choosing a level of benefits and then determining the cost. In short, the cost of the ATA policy had nothing to do with the insured's needs. Respondent instructed Ms. Watkins to give mobile home insurance quotations over the phone which included both the mobile home insurance premium and the cost of the motor club as determined by using the chart. No disclosure of the motor club was made at the time of the phone quotation. When consumers came into the office to purchase the coverage they were required to purchase the motor club in conjunction with the insurance and were lead to believe that the two items were a package. The motor club was routinely misrepresented to be some sort of insurance product. In fact, Ms. Watkins was unaware that the ATA product was in fact a motor club rather than an insurance policy. Respondent focused on the accidental death and dismemberment benefit included with the memberships when describing the product to Ms. Watkins and to any consumer that questioned the paperwork. Respondent maintained his inaccurate description of the ATA product at the hearing. Respondent justified the requirement to purchase the ATA motor club on the basis that his commission for the sale of mobile home insurance was too small to cover agency expenses and the sale of the motor club membership made up the small commission on mobile home insurance. In the instant case, Respondent's acts and those undertaken by Ms. Watkins at his direction constituted routine, deceptive, fraudulent and unfair business practices in violation of Chapter 626, Florida Statutes. The deceptive, fraudulent aspect of Respondent's practice makes the violations particularly serious. Respondent offered no credible evidence of mitigation for his business practice. Respondent was disciplined by the Department in 1976 for charging a cancellation fee in violation of the Florida Insurance Code. There he justified the charge on the basis that he was not adequately compensated by commission when policies were cancelled mid-term. Respondent was again disciplined by the Department in 1992 for having collected a "consulting fee" in violation of the Florida Insurance Code. Given the deceptive, fraudulent nature of Respondent's business practice and the previous discipline of Respondent's license, Respondent's license should be revoked.
Recommendation Based upon the foregoing Findings of Fact and the Conclusions of Law, it is accordingly, RECOMMENDED that Respondent, John Walter Drew, be found guilty of the violations set forth in the Conclusions of Law portion of this Order and that the Respondent's license as an insurance agent in this State be revoked and he be ordered to pay a fine of $5,000 within thirty (30) days of entry of the Final Order in this matter. DONE and ENTERED this 5th day of April, 1995, in Tallahassee, Florida. DIANE CLEAVINGER Hearing Officer Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-1550 (904) 488-9675 Filed with the Clerk of the Division of Administrative Hearings this 5th day of April, 1995. APPENDIX The facts contained in paragraphs 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37 and 38 of Petitioner's proposed findings of fact are adopted in substance, insofar as material. The facts contained in paragraphs 3 and 11, are adopted in substance, insofar as material. Paragraphs 1, 2, 6, 7, 8 and 9 of Respondent's proposed findings of fact pertain to either procedural matters or are legal argument. The facts contained in paragraphs 4, 5, 10, 12, 13, 14, and 18 of Respondent's proposed findings of fact are subordinate. The facts contained in paragraphs 15, 16, and 17 of Respondent's proposed findings of fact were not shown by the evidence. COPIES FURNISHED: John R. Dunphy, Esq. Michael McCormick, Esq. Division of Legal Services 612 Larson Bldg. Tallahassee, FL 32399-0333 Charles P. Hoskins, Esq. Wells, Brown & Brady, P.A. P. O. Box 12584 Pensacola, FL 32573-2584 Bill Nelson State Treasurer and Insurance Commissioner The Capitol, Plaza Level Tallahassee, FL 32399-0300 Dan Sumner Acting General Counsel Dept. of Insurance The Capitol, PL-11 Tallahassee, FL 32399-0300
The Issue Whether the Office of Insurance Regulation (the Office) correctly calculated the New Business Rate in accordance with statutory authority provided by Section 627.9407(7)(c), Florida Statutes, with regard to National States Insurance Company’s (National States or Company or Insurer) request for a rate increase.
Findings Of Fact National States is an insurance company licensed in the State of Florida to engage in the sale of health insurance. National States has been in business since 1964, and currently sells life and health insurance products in 33 states. National States currently has four HHC policy forms in force in Florida: HNF-1, HNF-3, HHC-1 and HNC-1 (collectively referred to as the AHome Health@ policies). The Home Health policies pay benefits for home nursing care on an expense incurred basis up to the daily maximum specified for periods of 12, 24 or 36 months for the HNF-1 and HNF-3 policies; 12,24,36,48 or 60 months for the HHC-1; and 12 or 24 months for the HNC-1 policy. All policies under HNF-1, HNF-3 and HNC-1 policy forms are in renewal only. The Home Health policies are guaranteed renewable and cannot be canceled due to poor financial performance of the product. Rates can, however, be increased with the approval of the Office. The Home Health policies are known as "stand alone home health care policies" because they provide benefits for care in the policyholder's home as opposed to care provided in an institution such as a nursing home. The policy forms identified in the rate filing are not currently sold by National States, and are defined as a “closed block” of business, meaning that no new policies are being issued. Under the policy forms in issue in this proceeding, each time a renewal premium is received, another contract term begins which precludes any impairment of a prior contract, per the following language: This policy may be renewed for another term by the payment, . . . of the renewal premium for such term at the rate in effect at the time of such renewal. We reserve only the right to change the table of premiums for this policy and all the policies in this state. No change in the premium or in this policy may be made solely by us because of a change in your health or job, nor solely because of claims under this policy. On August 16, 2006 National States submitted a rate filing requesting a rate revision for its Home Health policies. The requested rate increase in the 2006 filing was 48.1 percent. The Office denied the requested 48.1 percent increase by National States by a notice of intent to disapprove ("NOI") issued September 19, 2006. Section 627.9407(7)(c), Florida Statutes, was enacted on June 20, 2006, and applies to all long term care policies issued or renewed on or after July 1, 2006. The National States’ rate filing in this case, FLR 06- 10794, is subject to Section 627.9407(7)(c), Florida Statutes. The policy forms at issue in this case were issued prior to the enactment of that statute. A guaranteed renewable form cannot be canceled by the insurer and must be renewed by the insurer as long as the policy holder continues to pay the requested premium. National States does, however, have the option under Section 627.6425, Florida Statutes, to request that the State of Florida close its entire block of business if its solvency is in jeopardy. Florida Administrative Code Rule 69O-157.108 was enacted in 2003, when the Florida Legislature adopted the NAIC (National Association of Insurance Commissioners) Model Rule of 2000 which states: An insurer shall provide the information listed in this subsection for approval pursuant to Section 627.410, Florida Statutes, prior to making a long- term care insurance form available for sale. * * * (c) An actuarial certification consisting of at least the following: A statement that the initial premium rate schedule is sufficient to cover anticipated costs under moderately adverse experience and that the premium rate schedule is reasonably expected to be sustainable over the life of the form with no future premium increases anticipated; A statement that the policy design and coverage provided have been reviewed and taken into consideration; A statement that the underwriting and claims adjudication processes have been reviewed and taken into consideration; A complete description of the basis for contract reserves that are anticipated to be held under the form, to include: Sufficient detail or sample calculations provided so as to have a complete depiction of the reserve amounts to be held; A statement that the assumptions used for reserves contains reasonable margins for adverse experience; A statement that the net valuation premium for renewal years does not increase; and A statement that the difference between the gross premium and the net valuation premium for renewal years is sufficient to cover expected renewal expenses; or if such a statement cannot be made, a complete description of the situations where this does not occur; Section 627.9407(7)(c), Florida Statutes, applies universally to all carriers selling long term care insurance in the state of Florida. For carriers currently issuing coverage (i.e. “open blocks” of business), the new business rate is determined by that insurer's book of business so that the premium charged to existing insureds will not exceed the premium charged for a newly issued insurance policy except to reflect benefit differences. For insurers not currently issuing new coverage (i.e. “closed blocks” of business), the new business rate shall be as published by the Office at a rate representing the new business rate of insurers representing 80 percent of the carriers currently issuing policies with similar coverage as determined by the prior calendar year earned premium. § 627.9407(7)(c), Fla. Stat. Dan Keating, acting Chief Actuary for the Office, authored the NOI at issue in this proceeding. He is a Fellow of the Society of Actuaries with over 36 years of experience and has reviewed between 250 to 300 rate filings in the state of Florida. His testimony establishes that market share is a percentage that represents how much a particular carrier’s sales are represented in the market. Each carrier’s percentage of the market is based on earned premium of the total volume of that particular share of the market. Market share theory should be used to determine which carriers represented 80 percent of the market share. Any other type of average would give too much weight to one company who might only sell one percent of the policies in the market. Florida premiums should be used to determine which companies represent the 80 percent market share. The Office instigated a data call to all carriers doing business in Florida to respond with confirmation that they were selling long term care business and to provide their premium information. The request was separated according to the definition for similar benefits which was identified as “facility-only,” “non-facility- only,” and “comprehensive.” Upon receipt, company data was verified and compared to the annual reports filed through the NAIC. The steps of the data call commenced with the publishing of the new business rate on September 29, 2006. Delays in publishing the new business rate were caused by the time allotted from enactment (June 20, 2006) to effective date (July 1, 2006), the type and amount of data requested, difficulty in getting companies to respond, review of the data once received, and the action of calculating the market share. Calculation to determine which companies represent the 80 percent market share, a necessity pursuant to compliance with Section 627.9407(7)(c), Florida Statutes, also required a review in this case of each companies’ first-year earned premium by personnel of the Office. Such a review of first-year earned premium is the proper basis to begin the calculation. Three companies were used to comprise the 80 percent market share: Banker’s Life & Casualty, Penn Treaty and Colonial American were chosen. Banker’s Life & Casualty, however, alone comprised 80 percent of the market share and would have been sufficient used alone. Nevertheless, in the interest of diversity and variance, and so that the new business rate would not rely solely on one company’s rate book, Penn Treaty and Colonial American rates were added to the market share. By adding the two additional companies, the new business rate was increased to some degree because both of the other companies were charging more. Banker's Life remained the major shareholder. A weighted average was then applied to the rates of each company to calculate a new business rate. Banker’s Life originally submitted data for the size of their premium (not the premium rates) that were based on its nationwide numbers. This error was not discovered until January 2007, after the new business rates were already published and affected the percentage of weight each company's rates were given. When the error was corrected, Banker’s Life remained above 80 percent of the market share as required by the statutory language of Section 627.9407(7)(c), Florida Statutes. The Office recalculated the new business rate based on the corrected Florida data which increased the new business rate minimally, but not significant enough to warrant a change to the published rates. The Office disapproved National States’ rate filing because approving the filing would have resulted in a premium charged that would have exceeded the new business rate allowed in accordance with Section 627.9407(7)(c), Florida Statutes. Although the percentages differ from one issue age to another, National States current rates without the increase, in the best case scenario, are at least 106 percent above the new business rate, and in general are on average two and one half times the new business rate. Each rate is above 100 percent of the new business rate, indicating that in every situation, for every issue age on all four policy forms, National States’ current rates, before any increase, are already above the new business rate. The Bankers Life and Casualty nationwide data was used to calculate the weighted average because it was the data provided to the Office in response to the data call. Experts for both parties concede that access to this data could only be had via submission by the carriers, as there is no central depository where this type of data is maintained. The market share calculation itself was accurate. A conscious decision was made by personnel of the Office to normalize the new business rate to reflect a 90-day elimination period because that policy form is the most commonly sold by the carriers. Banker’s Life does not have a corresponding rate for a 90-day elimination period; however, normalizing to Banker’s 42-day elimination period produced a higher new business rate because a shorter elimination period raises the cost of the policy since the policy holder can claim benefits sooner. As with the elimination period, there was a conscious decision to normalize the new business rate to a tax-qualified plan because it was the most commonly sold plan. Normalizing the benefits to calculate the new business rate was done by the use of factors gathered from the carriers making up the 80 percent market share and then weighted as required. The factors were available at the time the office received the data from each company, and prior to the disapproval of National States’ rate filing. National States' personnel were aware of the new statute prior to submitting the rate filing to the Office. Additionally, checks of the Office's website were made in July and again in August of 2006, in order to identify the new business rate applicable to this filing. New business rates were not published on the Office website at that time. National States' actuary concedes that he did not make any attempt to contact the Office to determine if the new business rate for stand alone home health care was available prior to submitting the rate filing to the office, although doing so would have been relatively easy. National States' actuary offered testimony that Florida home health care policies have been performing poorly not just for National States, but for the industry as a whole. This poor performance occurs when the actual claims experience that had emerged is much worse than had been expected when initially pricing the product. He described the pricing process in terms of the durational loss ratio curve, and how that curve impacts subsequent filings under Florida law. National States did not anticipate increasing the premiums at the time the policies were sold. National States' strategy as outlined by its actuary is at variance with requirements of Florida Administrative Code Rule, 69O-149.006(3)(b)23b(IV). Under that Rule's provision, the actuary is required to project the experience that he actually expects to occur. For a plan that was developed more than 15 years ago, it is highly unlikely that expectations today would match those in the original pricing product. 34. Rule 69O-149.006(3)(b)23b(IV) reads: (IV) The projected values shall represent the experience that the actuary fully expects to occur. In order for the proposed premium schedule or rate change to be reasonable, the underlying experience used as the basis of a projection must be reflective of the experience anticipated over the rating period. The Office will consider how the following items are considered in evaluating the reasonableness of the projections and ultimate rates. In order to expedite the review process, the actuary is encouraged to provide information on how each of the following have or have not been addressed in the experience period data used as the basis for determining projected values, or otherwise addressed in the ratemaking process. Large nonrecurring claims; Seasonality of claims; Prior rate changes not fully realized; Rate limits, rate guarantees, and other rates not charged at the full manual rate level; Experience rating, if any; Reinsurance costs and recoveries for excess claims subject to non-proportional reinsurance; Coordination of benefits and subrogation; Benefit changes during the experience period or anticipated for the rating period; Operational changes during the experience period or anticipated for the rating period that will affect claim costs; Punitive damages, lobbying, or other costs that are not policy benefits; Claim costs paid which exceed contract terms or provisions; Benefit payments triggered by the death of an insured, such as waiver of premium or spousal benefits; Risk charges for excess group conversion costs or other similar costs for transferring risk; The extent and justification of any claim administration expenses included in claim costs; and Other actuarial considerations that affect the determination of projected values. Testimony of National States' actuary is not credited.
Recommendation Based on the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that a final order be entered denying National States' requested rate increase. DONE AND ENTERED this 15th day of June, 2007, in Tallahassee, Leon County, Florida. S DON W. DAVIS 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 15th day of June, 2007. COPIES FURNISHED: Cynthia S. Tunnicliff, Esquire Brian A. Newman, Esquire Pennington, Moore, Wilkinson, Bell and Dunbar, P.A. 215 South Monroe Street, Second Floor Post Office Box 10095 Tallahassee, Florida 32302-2095 Charlyne Khai Patterson, Esquire Assistant General Counsel Office of Insurance Regulation 200 East Gaines Street 612 Larson Building Tallahassee, Florida 32399-4206 Kevin M. McCarty, Commissioner Office of Insurance Regulation 200 East Gaines Street Tallahassee, Florida 32399-0305 Steve Parton, General Counsel Office of Insurance Regulation 200 East Gaines Street Tallahassee, Florida 32399-0305
The Issue The issue is whether Respondent discriminated against Petitioner based on her sex or an alleged handicap contrary to Sections 760.20 through 760.37, Florida Statutes (2008), Fair Housing Act, as amended (the Act).
The Issue The issue is whether Petitioner's application for a 25.75 percent increase for its individual long term care policy form, number 93710(FL), filed on February 7, 2006, meets the applicable tests of Section 627.410, Florida Statutes,1 and Florida Administrative Code Rule 69O-149.005, and should be approved. Also at issue is whether Subsection 627.9407(7)(c), Florida Statutes (as amended by Section 9, Ch. 2006-254, Laws of Florida, effective July 1, 2007), applies in this case; and, if so, the propriety of Respondent's intended implementation of that amended statute to Petitioner's rate filing, and whether Petitioner's rate increase filing should be disapproved due to Respondent's implementation of that statute.
Findings Of Fact Petitioner's rate increase application concerns policy form number 93710(FL), a stand-alone individual home health care policy form, which is a sub-line of long term care health insurance. The policy provides reimbursement for certain medical care delivered outside the in-patient hospital or nursing home setting. The policy form and initial rates for the form were first approved in Florida in 1994. The policy has not been actively sold since 1998, and the existing policies constitute a closed block of business. Periodic rate increases for this policy form have been approved in the past: 30 percent in 1998, 20 percent in 1999, 20 percent in 2000, 50 percent in 2001, 30 percent in 2002, and 25 percent in 2003. OIR has prior approval authority over the rate increase applied for by Petitioner. Petitioner timely sought a formal administrative hearing in this matter. Petitioner's rate filing seeks approval of a 25.75 percent rate increase. Christine Jung, the actuary who submitted the rate filing, is a qualified actuary who meets the Qualification Standards for Prescribed Statements of Actuarial Opinion published by the American Academy of Actuaries ("AAA"). Mr. Yee's opinion that Ms. Jung is not a qualified actuary is not persuasive. He could point to no provision in the AAA's Qualification Standards for Prescribed Statements of Actuarial Opinion that did not meet Ms. Jung's education, training, and experience when she submitted the rate filing at issue in this case. Any objections to Ms. Jung's qualifications go to the weight to be given her testimony when compared with the testimony given by the other actuarial experts in this matter. The greater weight of the evidence supports the fact that a rate increase of 25.75 percent meets the tests prescribed for approval in Florida Administrative Code Rule 69O- 149.005(2)(b). Petitioner's evidence, which is the more persuasive evidence, establishes that the requested rate increase meets the tests of that rule when anticipated future experience is projected properly, in accordance with sound actuarial techniques and Actuarial Standards of Practice ("ASOPs"). Ms. Jung concluded that a rate increase of 25.75 percent was actuarially appropriate based on her actuarial analysis. Ms. Dawn Helwig is a highly qualified health insurance actuary who specializes in rate analysis and rating of long term care insurance, including home health care insurance. Her testimony is persuasive. She evaluated Petitioner's rate filing and its component data, and performed an actuarial analysis using generally accepted actuarial techniques, and in accordance with the ASOPs and actuarial principles. Ms. Helwig concluded, and it is so found, that Petitioner's rate increase request of 25.75 percent is actuarially justified, will yield premiums that are not excessive in relation to the benefits offered under policy form number 93710(FL), and satisfies the tests for approval set forth in Florida Administrative Code Rule 69O- 149.005(2)(b)1.a. and b. Ms. Helwig concluded that a rate increase of 37.4 percent to 39.9 percent (depending upon whether industry medical trend is used in the first year of the rate projection) is actuarially justified based upon Petitioner's claims and premium experience reflected in the rate filing. By comparison, Ms. Jung's rate indication of 25.75 percent was conservative. Ms. Jung used an un-trended average of historical loss ratios to arrive at her starting loss ratio value (or current loss ratio) from which to project forward to determine Petitioner's rate need. Had Ms. Jung trended the claims costs in the three years of Petitioner's historical experience that she used to determine her starting loss ratio, her rate indication would have been higher than the 25.75 percent rate increase for which Petitioner applied. At the hearing, OIR offered three independent grounds for disapproving Petitioner's rate filing: 1) that the proposed rate increase was not based upon a "credible body of past data," which must have occurred over the lesser period of the past five years or at least 1,000 claims; 2) that the use of a medical trend has not been justified; and 3) that the rate increase would result in a rate higher than the OIR's published new business rate pursuant to Subsection 627.9407(7)(c), Florida Statutes. OIR did not assert, as a basis for denial, that Petitioner's requested rate increase would result in inadequate premium rates under Florida Administrative Code Rule 69O- 149.005(1), and no evidence was offered at the hearing that the requested rate increase would result in inadequate rates. Accordingly, it is found that Petitioner's requested rate increase would not result in inadequate rates. OIR did not assert, as a basis for denial, that Petitioner's requested rate increase would result in unfair discriminatory premium rates under Florida Administrative Code Rule 69O-149.005(1), and no evidence was offered at the hearing that Petitioner's requested rate increase would result in unfair discriminatory rates. Accordingly, it is found that Petitioner's requested rate increase would not result in unfair discriminatory rates. Mr. Keating interprets Florida Administrative Code Rule 69O-149.0025(6) to mean that, in developing the Florida and nationwide rate indications called for in Florida Administrative Code Rule 69O-149.0025(6), the actuary must use the five years of historical data that the Rule prescribes for weighing the separate Florida and nationwide rate indications for credibility. Ms. Jung derived her separate Florida and nationwide rate indications from the three most recent full years of Petitioner's data. Mr. Keating, therefore, asserted that Petitioner's requested rate increase is not supported by sufficiently credible experience data. The more persuasive evidence does not support Mr. Keating's opinion that the experience data Petitioner used to develop its rate request is not sufficiently credible. Mr. Keating's interpretation of Florida Administrative Code Rule 69O-149.006(3)(b)23.b.(II) is not borne out by the Rule's language. With reference to projecting rate need based on in-force experience, the portion of the Rule states as follows: The experience period shall reflect the most current data available, generally the most recent 12 months for coverage subject to medical inflation, or the period of time to determine the credible data pursuant to subsection 69O-149.0025(6), F.A.C. (emphasis supplied) That provision is disjunctive, and its use of the word "generally" connotes that its disjunctive indicators for experience periods are examples, not requirements. As Mr. Keating conceded at the hearing, neither that rule passage nor other related rules equate "the most current data available" with "the period of time to determine credible data pursuant to subsection 69O-149.0025(6), F.A.C." The rules thus leave to actuarial judgment the decision concerning "the most current data available" on which to develop the separate Florida and nationwide rate indications called for in the rules. The rules do not mandate that the actuary use five full years of data to develop the separate Florida and nationwide rate indications. The rules do require that, once those separate Florida and nationwide rate indications are developed, they are to be credibility-weighted using five years of data under Florida Administrative Code Rule 69O-149.0025(6) to arrive at the ultimate rate indication. This is what Ms. Jung did. OIR has not consistently required use of the five years of data as the experience period for making the separate Florida and nationwide rate indications. One of Respondent's exhibits, a Guarantee Trust Life rate filing for this same policy form, was approved by consent order, dated December 23, 2003. It used only the most recent calendar year's loss ratio as the starting value for projection, though there were fewer than 1,000 claims in that calendar year. The more persuasive evidence shows that Mr. Keating's interpretation would double-count the credibility factor in the rules. Moreover, the more persuasive evidence shows that applying Mr. Keating's interpretation of the credibility weighting method set forth in Florida Administrative Code Rule 69O-149.0025(6)(e)3. as "equivalent to" the alternative methods described in Florida Administrative Code Rule 69O-149.0025(6)(e)2. would, in fact, not yield equivalent results for the same data set, as a matter of mathematics. Ms. Jung developed a rate indication based on Petitioner's Florida experience under policy form 93710(FL), and a separate rate indication based on Petitioner's nationwide data on that policy form. She used the most recent three full years of experience on the policy form (2002-2004) which she found to be appropriate current data for developing Florida and nationwide rate indications. After separately deriving an indicated rate increase from Florida data and a rate increase indicated by total nationwide data, consistent with Florida Administrative Code Rule 69O-149.0025(6)(e)3., she then weighed the two separate rate indications for credibility by weighting the resulting rate changes from each district analysis by the credibility of each distinct component assigned by that rule. Her analysis resulted in a 25.79 percent rate increase need, which she rounded down to 25.75 percent. Florida Administrative Code Rule 69O-149.0025(6) establishes the acceptable range for credible data (the number of claims needed for making rate projections) and the procedures for weighting the data for credibility when the claims frequency used in rate projections falls between the Rule's upper and lower bounds of acceptability. A frequency of at least 1,000 claims over the five years preceding the rate filing is given 100 percent, or full, credibility. Two hundred claims or fewer are given no (zero percent) credibility. Claim frequency counts falling between 200 and 1,000 receive a proportionate credibility weight under the formulae in Florida Administrative Code Rule 69O-149.0025(6)(e), which provides, in pertinent part: (b)1. For policy forms . . . such as accident and long term care, at least 1,000 claims, over a period not to exceed the most recent 5-year period, shall be assigned 100 percent credibility; 200 claims shall be assigned 0 percent credibility. * * * (e)1. Florida only experience shall be used if it is 100 percent credible. 2.a. If Florida experience is not 100 percent credible, a combination of Florida and nationwide experience shall be used. The Florida data shall be given the weight of the ratio of the Florida credibility to the nationwide credibility. For example, if Florida data is 10 percent credible and nationwide is 40 percent credible, the Florida data will be given the weight of [10%/40%] 25 percent. The nationwide data shall be given the weight of the ratio of the nationwide credibility less the Florida credibility to the nationwide credibility. In the above example, the nationwide data will be given the weight of [(40%-10%)/40%] 75 percent. The data is combined using the indicated weights (in the example above, the experience data would be weighted 25%/75%). The combination of the two weights will always equal 100 percent. A rate change is determined from the blended data. 3. The analysis in subparagraph 2. above is equivalent to determining the indicated rate increase from the Florida only data and the total nationwide data separately, and then weighting the resulting rate changes from each distinct analysis by the credibility of each distinct component. Ms. Jung's separate Florida and nationwide rate indications were based on claim counts that fall within the credibility range posted by Florida Administrative Code Rule 69O-149.0025(6)b.1. The number of claims in the last three years of data from which Ms. Jung derived her Florida and nationwide rate indications exceeded 200 claims, in each case. Ms. Jung applied five full years of company claims data to combine her Florida and nationwide rate projections, that is, to separate rate indications for credibility in arriving at the ultimate indicated rate need of 25.75 percent. She did so properly, in accordance with the provisions of Florida Administrative Code Rule 69O-149.0025(6)(e)3. Moreover, in her actuarial projections of Florida and nationwide rate indications, Ms. Helwig, Petitioner's consulting actuary, used the five years of Petitioner's historical experience data that Mr. Keating believes should be used. Ms. Helwig concluded that a rate increase of at least 25.75 percent was actuarially justified, and that the full five years of data would support an increase of up to 39.9 percent in her opinion. Mr. Yee, Respondent's actuarial expert, testified that in his opinion neither Ms. Jung's analysis nor Ms. Helwig's analysis is based on a sufficient number of claims to be regarded as adequately credible. The premise for Mr. Yee's opinion is his personal, subjective standard for "fully credible" long term care ("LTC") data. Mr. Yee's full-credibility standard contradicts the full- credibility standards for LTC ratemaking prescribed by Florida in Florida Administrative Code Rule 69O-149.0025(6). Florida Administrative Code Rule 69O-149.0026(6)(b)1. provides that for LTC forms, "at least 1,000 claims, over a period not to exceed the most recent 5-year period, shall be assigned 100 percent credibility; 200 claims shall be assigned 0 percent credibility." The Rule thus prescribes that 1,000 LTC claims over five years are to be considered fully credible; that 200 claims is the credibility floor; and it establishes methods for weighting the credibility value of claim counts falling between 200 and 1,000. Contrary to the Florida rule, Mr. Yee holds the view that at least 3,246 LTC claims are required for "full credibility." Based upon his "full credibility" premise, he believes that at least 649 claims are needed to consider LTC data credible enough for making a rate projection. He arrives at his 649-claim credibility minimum by noting that since 200 claims in the Rule (the credibility floor) is 20 percent of the 1,000 claims assigned full credibility by that Rule, 649 is the minimum number of claims required for making a LTC rate projection (20 percent of 3,246). Since Petitioner's data contains 621 claims for the most recent five-year period, less than the 649 claim floor, Mr. Yee concludes that Petitioner's data is not credible, and, therefore, should not be used for making a rate projection. He, therefore, concludes that the rate increase is not justified by either Ms. Jung's or Ms. Helwig's actuarial analysis. However, Mr. Yee's data credibility opinion directly conflicts with Florida Administrative Code Rule 69O-149.0025(6). The Rule is clear that 1,000 claims or more shall be accorded full credibility for LTC ratemaking and that claim counts falling between 200 and 1,000 are to be credibility-weighted as set forth in the Rule. Mr. Yee concedes that his "full credibility" standard, the foundation of his no-credibility opinion, is a subjective standard that he and four other actuaries recommended in a report to the National Association of Insurance Commissioners on the subject of LTC data credibility. He further concedes that his "full credibility" standard has not been adopted by any regulatory body or embraced by AAA. It has not even been circulated for comment to the AAA membership. Mr. Keating testified that a claims frequency of 1,000 claims should be regarded as fully credible for LTC ratemaking under the Florida rules. Mr. Yee's opinion, while interesting and apparently well-intended, is, therefore, not credited. The more persuasive evidence shows that Petitioner's data is adequately credible under Florida's adopted standards, and was properly used by Petitioner's actuaries in developing the rate indications supporting Petitioner's requested rate increase. Although in its interrogatory responses OIR stated that Petitioner failed to adjust earned premium on a current rate basis, Mr. Keating testified at hearing that Petitioner had, in fact, provided earned premium on a current rate basis. Therefore, this asserted reason for denial of Petitioner's rate increase is not supported by the evidence. The actuaries who testified agreed that putting both historical premiums and losses (claim costs) on a "current basis" or a "current rate basis" is actuarially appropriate in order to arrive at the correct starting point for a rate projection (whether that starting point is expressed as a loss ratio on a current basis or claim costs on a current basis), and that an accurate current starting point is centrally important in making and evaluating rates. However, neither the phrase "earned premium on a current rate basis" nor the terms "current rate basis" or "current basis" is defined in Section 617.410, Florida Statutes, or in Florida Administrative Code Rule 69O-149, Part I. Moreover, the statutes and rules do not prescribe how to arrive at a "current rate basis" starting value for rate projection. The propriety of a "current rate basis" technique, therefore, must be judged by whether it comports with the ASOPs and accepted actuarial principles. The major point of contention among the actuaries who testified was on the question of the appropriate "current rate basis" methodology for determining the starting value for future projection (the correct starting loss ratio on a current rate basis or the correct starting claims cost on a current basis). Mr. Keating made what he refers to as a "current rate basis" adjustment to Petitioner's historical data. He testified that, based on his current rate basis adjustment, he estimated the "current rate basis" Actual-to-Expected (A/E) ratio for Petitioner's historic Florida experience to be 95.41 percent, which he testified results in a current-rate-basis starting loss ratio value of 74.34 percent for making a rate projection. Mr. Keating applied that 74.34 percent starting-point value, and restructured Petitioner's rate projection, culminating in his estimate of a future A/E ratio for Florida of 96.66 percent. Since, according to his restructuring, the future A/E ratio is less than 1.0, he concluded that Petitioner failed to meet the "Anticipated Loss Ratio" test of Florida Administrative Code Rule 69O-149.005(2)(b)1.a. He performed the same analysis, using the same "current rate basis" method on Petitioner's nationwide data, and reached the same conclusion. The more persuasive evidence, however, shows that the "current rate basis" adjustment Mr. Keating made to Petitioner's data does not comport with the ASOPs and the Florida rules. His "current rate basis" analysis yields starting values for projecting Petitioner's need (his starting loss ratio of 74.34 percent for Florida data and 76.66 percent for nationwide data), which do not match Petitioner's recent experience, and which, in fact, Petitioner has not experienced since 2002, even when Petitioner's historical experience is adjusted by earned premium on a current rate basis. Under Mr. Keating's starting loss ratio of 74.34 percent, and the corresponding number of claims implied by that loss ratio, Ms. Helwig demonstrated that if Mr. Keating's starting loss ratio were correct (if it were the true mean loss ratio), then the probability of Petitioner experiencing the actual number of claims it has in fact experienced in the two- to three-year period before 2005 is less than one percent. ASOP 25, which instructs that the methods an actuary uses should produce a starting value with less than a one percent probability of occurrence, is not found to be reasonable. ASOP 8 instructs that the actuary is to adjust past experience, specifically historical loss ratios, for trends in morbidity in a way that reasonably matches claim experience to exposure. Mr. Keating's current rate basis method did not do so. The more persuasive evidence shows that Mr. Keating's "current rate basis" analysis does not comport with ASOP 8, which provides that when past experience is used to project future results, both past premium rates and morbidity (claims experience) should be adjusted to reflect changes and trends. Mr. Keating did not recognize or take into account the claims trend in the company's historical experience when coming to his "current rate basis" A/E ratios and his starting values for the restated projections he made from Petitioner's data. To derive the "current rate basis" A/E ratios of 95.41 percent (Florida) and 98.53 percent (nationwide), which he then used to develop starting loss ratios of 74.34 percent based on Florida data and 76.66 percent based on nationwide data, Mr. Keating simply took a weighted average of 4.75 years of Petitioner's historical A/E ratios (each year's A/E ratio, as adjusted by earned premium on a current rate basis). Mr. Keating did not trend Petitioner's historical data to arrive at his current basis A/E ratios and his starting-point loss ratio values. To derive the "current rate basis" starting point A/E ratios, Mr. Keating used an average of Petitioner's historic yearly A/E ratios, and did not trend the yearly A/E ratios, based upon this rationale: In his view, only medical (utilization) trend should be used to trend historical data, since he believes that "aging" trend should be pre-funded in the initially approved premium for any long term care product. Since he believed that Petitioner had not sufficiently identified medical "utilization" trend separately from the "aging" trend in the historical data, Mr. Keating, therefore, averaged Petitioner's historical experience and did not apply any trends in the company's claims experience when arriving at his starting loss ratio. The actuaries, four of whom testified at hearing, disagreed about whether, in the context of making projections of future anticipated experience from a current-basis starting- point value, Petitioner's actuaries properly isolated medical (utilization) trend from "aging" trend, and whether, therefore, utilization trends should be used in projecting forward. The more persuasive trend, as discussed more fully below, shows that utilization trend over and above aging trend is adequately identified in the company's data, and was appropriately used for projecting forward. That debate, however, is separate from the issue of whether an actuary may ignore trend in a company's historical data to develop the current-basis starting-point for projecting into the future. The more persuasive evidence is that the ASOPs require the actuary to take trend in the company's historical data into account in coming to a starting-point value for future projection (the current basis loss ratio or current basis claims cost), regardless of whether the historical trend is categorized as "medical," "aging," or otherwise. The more persuasive evidence is that, if the actuary fails to recognize and take into account historical trend in evaluating the starting point for a future projection, the starting-point value will be actuarially incorrect, thus making the future projection incorrect, regardless of whether, for purposes of the future projection, the actuary includes a medical trend assumption or excludes it. The persuasive evidence on this point is reinforced by the language of Florida Administrative Code Rule 69O- 149.006(3)(b)18. That Rule passage calls for differentiating between medical (utilization) trend and insurance (aging) trend in making "trend assumptions." "Assumptions" about medical trend or aging trend denotes something the actuary assumes to be the case for future periods, not something that is an observable past fact. In contrast, a company's historical claims experience, which must be taken into account in arriving at the starting value for a rate projection, is a directly observable fact, not an assumption. Likewise, the trend in a company's past experience is a directly observable fact, not an assumption. According to the more persuasive evidence and the applicable ASOPs, an actuary may not disregard such observed historical claims trend in arriving at a "current basis" starting point to make future projections. Mr. Keating arrived at his starting-point values without considering and accounting for trend in Petitioner's historical data. His methodology was, therefore, actuarially flawed. If Mr. Keating had properly accounted for historical trend in his evaluation, his starting-point loss ratio values would have been approximately 113 percent, which compares closely with Ms. Helwig's starting value of 107 percent. His projection would, therefore, have resulted in a future A/E ration of 148 percent, which calls for approval of Petitioner's rate request under the "Anticipated Loss Ratio" test of Florida Administrative Code Rule 69O-149.005(2)(b)1.a. Ms. Helwig properly took historical trend into account in reaching the starting value for her future projections, and concluded that the appropriate starting value was a loss ratio of 107 percent. Her projections show that Petitioner's rate increase application meets the approval tests of Florida Administrative Code Rule 69O-149.005(2)(b)1. Ms. Jung used a 10 percent medical trend assumption, for one year only, in her projection forward of anticipated experience, appropriately weighted under Florida Administrative Code Rule 69O-149.0025(6). Ms. Helwig also assumed medical trend, likewise for one year only, in her projection of anticipated experience, appropriately weighted under Florida Administrative Code Rule 69O-149.0025(6). The more persuasive evidence shows that using medical (utilization) trend in projecting forward in this case is appropriate, and the medical trend values used by Petitioner's actuaries for projection are reasonable and appropriate. Florida Administrative Code Rule 69O-149.006(3)(b)18. provides that, when making future projections, medical trend (which includes "utilization" trend) may be used, but "aging" trend may not be included. The Rule directs the actuary to "make appropriate adjustments to claims data to isolate the effects of medical trend." The evidence demonstrates that Ms. Helwig appropriately adjusted Petitioner's claims data to isolate the effects of the medical (utilization) trend, and properly included utilization trend in her projections. "Aging" trend refers to the expected increase in frequency of claims as policyholders' ages increase after policies are originally issued. As Ms. Helwig persuasively testified, companies filing original rates for long term care policies make an actuarial assumption about what the expected increase in claim frequency due to aging will be over the life of the policies, which is commonly understood by health actuaries to be aging trend. The aging trend assumption is incorporated into the durational loss ratio curve (or table) for the policy form, approved by Respondent when the policy form and its original rates are approved. The originally approved premium rate is intended to cover increases in claims frequency assumed in the durational loss ratio curve, that is, increase over time in claim frequency from aging, assumed in the durational loss ratio curve, is pre-funded by the originally- approved premium rates. As Ms. Helwig persuasively testified, increases in claims frequency, over and above the aging trend originally assumed in the durational loss ratio curve, is utilization trend, which is properly included in medical trend. It is an experienced trend in claims frequency that exceeds the trend which was originally assumed, and that exceeds what was assumed to be pre-funded in the originally approved premium rate. Ms. Helwig testified, without contradiction that, in Petitioner's five years of experience, she evaluated that a clear, observable trend exists in the frequency of claims that exceeds the frequency trend assumed in Petitioner's approved durational loss ratio curve, and that this excess claims frequency trend was not pre-funded in the original approved premium rate. She, therefore, subtracted the aging trend assumed in the durational loss ratio curve from the average observed total trend, and properly included the excess utilization trend as medical trend in evaluating the propriety of Petitioner's requested rate increase. She properly isolated the effects of medical trend, and properly included the medical trend, so derived, in her experience projections. Mr. Keating's and Mr. Yee's opinions on the use or non-use of medical trend to make projections in this case, under Florida Administrative Code Rule 69O-149.006(3)(b)18., are not persuasive. Both Mr. Keating and Mr. Yee testified that Petitioner had not "separately identified" (isolated) medical trend, in their opinion, and thus use of medical trend in projecting future experience to derive rates for Petitioner's policy is not justified. However, both of Respondent's actuaries based their opinions on the assumption that all utilization increases were intended to be pre-funded in the rates originally submitted and approved for this particular policy form (i.e., that all claims frequency increases were intended to be subsumed in the aging trend originally included in the durational loss ratio curve when the initial premium rate for this policy form was proposed and approved). They, therefore, assumed that all frequency increases under this policy form must be aging trend, and, therefore, that no utilization increase in the company's experience can be treated as part of medical trend for this product. However, Mr. Keating's and Mr. Yee's assumption on this point is contradicted by the evidence. The actuarial memorandum for the original 1994 rate filing (which Respondent approved) clearly noted that the initial premium rate for this particular policy form was not intended to fully pre-fund all expected utilization increases, and that utilization increases in excess of what is pre-funded in the original premium rate would be funded by periodic rate increases. Mr. Keating and Mr. Yee did not consult the original rate filing and the original actuarial memorandum in forming their opinions. As Ms. Helwig testified, and it is concluded here, it is consistent with how the premium rate for this particular policy form was initially filed and approved to classify, as medical trend, the utilization trend in the company's experience that exceeds the utilization frequency assumed in the durational loss ratio curve. Mr. Yee and Mr. Keating additionally offered the opinion that medical trend was not justified because Petitioner's data was insufficiently credible. These opinions are not credited. Petitioner's actuaries used adequately credible data in their analyses under the standards set forth in the Florida rules, as discussed above. Respondent disagreed at hearing with Ms. Helwig's use of industry medical trend in her projections. However, the more persuasive evidence shows that Ms. Helwig properly used industry medical trend. The industry medical trend value she used was derived from the Milliman Claim Cost Guidelines, adjusted to Petitioner's benefit structure and in-force business. The Milliman Claim Cost Guidelines are the type of data health actuaries reasonably rely upon in reaching professional opinions in their field. Ms. Helwig gave appropriate credibility weight to the industry medical trend she used in her projections, in accordance with Florida Administrative Code Rule 69O- 149.0025(6). Moreover, she testified without contradiction that, even if industry medical trend were excluded from her projections, the indicated rate need from her analysis would still meet or exceed the 25.75 percent rate increase for which Petitioner has applied. Mr. Keating and Mr. Yee testified that Petitioner's rate increase application should not be approved because Petitioner should have filed a new durational loss ratio curve for approval, and phased-in the resultant rate increase over time, rather than making a periodic rate increase filing, as Petitioner chose to do. Respondent did not assert this as a basis for denial in its denial letter, in its interrogatory responses, or otherwise before hearing. Petitioner chose to present evidence on this issue, and on the more persuasive evidence, Mr. Keating's and Mr. Yee's opinions that Petitioner was required to file a new durational loss ratio curve are not found to be persuasive. Both of Respondent's actuaries rest their opinion upon their interpretation of two provisions in Florida Administrative Code Chapter 69O-149, Part I. Their interpretation does not comport with the plain language of the rules they rely upon. Respondent's actuaries testified that Florida Administrative Code Rule 69O-149.006(3)(b)20. requires Petitioner to file a new durational loss ratio curve (or table), rather than seek a periodic rate increase, when company experience shows utilization exceeding what was assumed in the existing loss ratio curve. That subsection reads, in pertinent part, as follows: [The actuarial memorandum for a rate increase] shall also include the current approved durational loss ratio table for the form. If a revised durational loss ratio table is being proposed, the proposed table, together with a justification for the new table, shall be provided. The proposed new table shall be consistent with the claim projections contained in the new filing. If approved, the new table will be used in filings made subsequent to the one in which it is being proposed. (V)(A) When the slope of the shape of the durational loss ratio table is changed . . . from those used in the last approved rate filing, any rate increase due to the change shall be uniformly implemented over a 3 year period. Contrary to Mr. Keating's and Mr. Yee's views, the plain language of this Rule does not require Petitioner to propose a revised curve. The Rule prescribes a particular course of conduct if a revised curve is proposed, but that does not require that the rate filer propose a new curve. Mr. Keating and Mr. Yee also testified that Florida Administrative Code Rule 69O-149.0025(7)(a) required Petitioner to file a new durational loss ratio curve. These opinions are inconsistent with the plain language of that Rule, as well. That Rule provides, in pertinent part: (a)1.a. The company shall adjust the durational loss ratio table when the average annual premium at the time of filing results in a loss ratio standard pursuant to the provisions of subsection 69O-149.005(4), F.A.C., that is changed by at least .5 percent from the current lifetime loss ratio standard for the form. b. Each loss ratio in the durational loss ratio table shall be increased by the ratio of the loss ratio standard determined from the current average annual premium divided by the prior lifetime loss ratio standard . . . . This Rule refers to Florida Administrative Code Rule 69O- 149.005(4). Florida Administrative Code Rule 69O-149.005(4), in turn, contains tables setting out minimum loss ratio standards (and formulas for revising minimum loss ratio standards in those tables) for some types of health insurance policies: policy forms whose loss ratio standards are subject to change based on the average annual premium at the time of the rate filing. Florida Administrative Code Rule 69O-149.005(4), however, explicitly excludes long term care policies. It provides: These tables are not applicable to Medicare Supplement or Long-Term Care Policy Forms. The minimum loss ratios for those policy forms are found in Rule Chapters 69O-156 and 69O-157, respectively. Florida Administrative Code Rule 69O-149.005(4) excludes long term care policies because, unlike the types of policies governed by Florida Administrative Code Rule 69O-149.005(4), the minimum loss ratio standard for long term care policies does not change when the average annual premium changes. The minimum loss ratio standard for long term care policies is a constant value, as stated in Florida Administrative Code Rule 69O- 157.022. Florida Administrative Code Rule 69O-149.0025(7) only requires a revised durational loss ratio curve to be submitted for policy forms whose minimum loss ratio standards are subject to change under Florida Administrative Code Rule 69O-149.005(4). As noted, that is not true of long term care policies. Accordingly, Mr. Keating's and Mr. Yee's opinions that this rule required Petitioner to submit a new durational loss ratio curve are not persuasive. Both Mr. Keating and Mr. Yee further suggested that Petitioner should file a new durational loss ratio curve, rather than seek a periodic rate increase, because of their assumption that all utilization (frequency) increases for long term care policies should have been pre-funded in the rates originally submitted and approved for the product. In other words, they opine, all claim frequency increases should be included in the aging assumption of the original durational loss ratio curve for a long term care policy. Whatever the accuracy of that assumption may be as to the rate structure of long term care products in general, the evidence shows it is factually incorrect as to this particular product. The actuarial memorandum for the initial, approved rate filing for this particular policy form clearly noted that the initial premium rates were not intended to fully pre-fund all expected utilization increases. The rate structure approved by Respondent for this product expressly contemplated that utilization increases, in excess of the utilization trend assumed in the initial durational loss ratio curve, would be funded by future periodic rate increases. Consistent with that approved rate structure, Respondent has approved several periodic rate increases for this product (most recently by consent order in 2003), and did not previously take the position that Petitioner should file a revised durational loss ratio curve. Respondent's actuaries' opinions on this point are contradicted by the evidence and are not persuasive. Mr. Yee opined that annual variation or volatility in Petitioner's claim reserves may mean that the claim reserves in Petitioner's data may not actually reflect the company's true experience. He thus concluded that a rate increase was not justified. Respondent did not assert this as a basis for denial in its denial letter, responses to interrogatories, or otherwise prior to the hearing. Nevertheless, Petitioner presented evidence on the issue, and on the more persuasive evidence Mr. Yee's opinion is not persuasive. Mr. Keating did not testify that annual variation in Petitioner's claim reserves was cause to deny the rate increase, and did not mention any concern over annual claim reserve variation as a basis to deny the rate increase. Mr. Yee admitted that trending and averaging of the data smoothes claim reserve year-to-year volatility. Ms. Helwig persuasively testified that the observed claim reserve variations correspond closely with observed annual fluctuations in claim frequency, and that averaging Petitioner's claims history (as both she and Ms. Jung did) smoothes year-to-year data variability. The more persuasive evidence does not support Mr. Yee's opinion on this point. Mr. Yee testified that he believed there was a discrepancy between current premiums on this book of policies and "active life reserves" displayed in Ms. Jung's actuarial memorandum supporting the filing. Mr. Yee characterized the discrepancy as "alarming," but did not offer an objective basis for that characterization. He testified that, because Ms. Jung did not explain the active life reserves in her actuarial memorandum, he concluded that she was not qualified to make the filing, and it should, therefore, be disapproved. Respondent did not assert this as a basis for denial in its denial letter, its interrogatory responses, or otherwise before hearing. Nevertheless, Petitioner presented evidence on the issue, and on the more persuasive evidence Mr. Yee's opinion is not persuasive. On cross-examination, Mr. Yee testified that a rate filing actuarial memorandum is not intended to be a reserve opinion by the actuary, and that annual reserve opinions are separately made. Mr. Yee also testified that companies often pool active life reserves across policy forms, and he does not know whether Petitioner's reported active life reserves are the result of such multi-policy-form pooling, or pertain exclusively to policy form 93710(FL). He admitted he had no personal knowledge concerning the accuracy of Petitioner's life reserves displayed in Ms. Jung's actuarial memorandum. He, therefore, has no factual basis to question them. He further conceded that neither Florida Administrative Code Rule 69O- 149.006, nor applicable ASOPs place any obligation on Petitioner or its actuaries to provide explanation in the rate filing actuarial memorandum concerning the reported life reserves. Mr. Keating did not request any further explanation from Petitioner concerning its reported life reserves, did not list a deficiency of explanation concerning life reserves as a basis for denial, and did not testify that he asked for more explanation concerning life reserves, or found the absence of explanation a reason for disapproving Petitioner's rate increase filing. Ms. Helwig testified that active life reserves play no role in evaluating the appropriateness of a proposed rate for a home health care product. Consistent with her testimony, Florida Administrative Code Rule 69O-149.006(3)(b)17. expressly provides: "Because these [active life] reserves do not represent claim payments, but provide for timing differences, they shall not be included in any benefit and loss ratio calculations." The Rule expressly acknowledges that active life reserves are not material to evaluating whether the benefit/premium relationship, often expressed in terms of loss ratios, meets the rules' tests for approval of the requested rate. Mr. Keating testified that certain data inconsistencies in Ms. Jung's actuarial memorandum supporting the rate increase filing were not significant, and did not affect his review or his conclusions. Respondent did not cite these minor data discrepancies as a basis for denial. Ms. Jung testified that the data discrepancies were immaterial, and did not affect her rate analysis. It is found that the data inconsistencies are immaterial to the issues to be decided in this case, and are not a basis to deny Petitioner's requested rate increase. As noted in the Preliminary Statement above, before the final hearing, Petitioner made a Motion to Strike and Motion in Limine, directed to Respondent's assertion that amended Subsection 627.9407(7)(c), Florida Statutes, would serve as an additional basis for the disapproval of Petitioner's rate increase. Respondent asserted this additional basis for the first time in its statement of position in the Pre-Hearing Stipulation filed in this matter. Petitioner's Motion was denied at the time it was made in order to hear evidence as to how Respondent intended to implement and apply the amended statute to Petitioner's rate filing. Numerous facts were elicited at hearing that bear on the applicability of Subsection 627.9407(7)(c), Florida Statutes. Section 9, Chapter 2006-254, Laws of Florida, amended Subsection 627.9407(7), Florida Statutes, in June 2006, after this rate filing was made and disapproved. Section 9 (creating Subsection 627.9407(7)(c), Florida Statutes) provides, in part, and Section 11, Chapter 2006-254, Laws of Florida, provides in pertinent part: (c) Any premium increase for existing insureds shall not result in a premium charged to the insureds that would exceed the premium charged on a newly issued insurance policy, except to reflect benefit differences. If the insurer is not currently issuing new coverage, the new business rate shall be as published by the office at the rate representing the new business rate of insurers representing 80 percent of the carriers currently issuing policies with similar coverage as determined by the prior calendar year earned premium. Section 11. This act shall apply to long- term care insurance policies issued or renewed on or after July 1, 2006. Petitioner is not issuing new coverage under this policy form. Amended Subsection 627.9407(7)(c), Florida Statutes, is not self-executing in its application to insurers, such as Petitioner, that are not issuing new coverage. As to Petitioner (and similarly situated insurers), Respondent implemented amended Subsection 627.9407(7)(c), Florida Statutes, by publishing benchmark charges, developed by Respondent, to which Petitioner's rates were compared. In developing its published benchmarks, Respondent made a number of policy decisions, which are discussed below. Respondent posted its benchmark charges on its website in late September of 2006. Well before hearing, while a meaningful opportunity for discovery still was available to Petitioner, Respondent had ample opportunity to notify Petitioner that it would be asserting its benchmarks as additional grounds for the disapproval of Petitioner's rate increase filing. Respondent could have done so by amending the denial letter or by amending its interrogatory responses. Respondent failed to do so. Respondent asked to continue the final hearing scheduled for November 21, 2006, which was granted, and never asserted as a basis for the continuance that it intended to assert a new basis for denial of Petitioner's rate increase filing, namely, the implementation of the amended statute. Respondent did not even assert the amended statute as a further basis for denial at a deposition taken by Petitioner of its actuary and designated agency representative, Mr. Keating, on November 30, 2006. Mr. Keating was specifically asked at that deposition whether Respondent intended to assert any additional bases for disapproval of Petitioner's rate filing. Mr. Keating responded that Respondent had no such intention. Had Respondent given timely advance notice, Petitioner would have had a fair opportunity to take discovery on the issue, to amend its Petition, or both. Prior approval of Petitioner's rate increase is an authorization required by law before it can implement the rate pursuant to Subsection 120.52(15) and Section 627.410, Florida Statutes. Respondent did not assert this additional basis for disapproval until December 22, 2006, after discovery had been completed. In light of this late date and the intervening holidays, Respondent's delay deprived Petitioner of a fair and meaningful opportunity to take discovery and prepare to defend Respondent's assertion at the final hearing beginning January 3, 2007. Since Petitioner could not implement the new rate unless and until the proceedings were completed and it received approval, additional continuances and delays would only have served to further prejudice Petitioner's position. Additionally, Respondent's proposed implementation and application of amended Subsection 627.9407(7)(c), Florida Statutes, constitutes agency action that determines Petitioner's substantial interests on the basis of non-rule policy. Numerous facts support this finding. Amended Subsection 627.9407(7)(c), Florida Statutes, is not self-executing to insurers like Petitioner that are not issuing new coverage. Respondent made policy decisions on several matters to implement the statute and to apply it to Petitioner, including: 1) What constitutes "similar coverage;" 2) Which insurers comprise 80 percent of carriers currently issuing policies "with similar coverage" for purposes of developing the benchmarks; 3) Within that 80 percent of carriers, whether the charges of each carrier are to be given equal weight in developing the benchmark charges, or instead, will be weighted according to each carrier's percentage of the previous year's earned long term care premium; 4) Since the carriers whose charges are used in developing benchmarks offer multiple home health care policy forms and various benefit levels and correspondingly varying premium charges, which particular charges by those carriers will be used in developing the benchmark charges; 5) What information, if any, other than the published benchmark charges will be used in making the comparison called for by the statute; and 6) What claims experience adjustments, if any, to the published benchmark charges will be used to make the comparison. When Respondent first gave notice, on December 22, 2006, that it would assert benchmarks under Subsection 627.9407(7)(c), Florida Statutes, Petitioner promptly put Respondent on notice that it believed the benchmarks to be non- rule policy assertions by the agency, by filing its Motion to Strike and Motion in Limine on December 27, 2006. Respondent did not in any manner publish the benchmarks it actually used to make the Subsection 627.9407(7)(c) comparison to Petitioner. The benchmarks Respondent published were developed for benefit configurations that are materially different from the benefit configurations in Petitioner's policy. Mr. Keating made a number of adjustments to the published benchmarks in making his comparison with Petitioner. However, none of these adjusted benchmarks were published in any manner, and Mr. Keating could not recall details of the methods by which the adjustments were made. Respondent compared Petitioner's rate filing to unpublished information, contrary to the plain language of Subsection 627.9407(7)(c), Florida Statutes. Respondent was aware of 20-25 insurers who would be affected by the published benchmarks, yet did not serve any affected insurer, including Petitioner, with a copy of the benchmarks published; and did not offer any of them a point of entry to proceedings in which to explore and test the adequacy or validity of Respondent's benchmarks. The published benchmarks developed by Mr. Keating for Respondent gave more than 92 percent weight to just one company- -Bankers Life. Two other companies' charges were also used, but Mr. Keating gave less than five percent combined weight to them. Mr. Keating did not give equal weight to the selected charges of the three insurers chosen as benchmark companies. Instead, he gave greater weight based upon the companies' prior year earned premium. He offered no actuarial basis for this decision. He also offered no testimony about the other companies' charges and what the posted benchmark values would be had he given equal weight to all three companies' charges. Respondent offered no point of entry for proceedings to question the decision. Mr. Keating also selected the lowest Bankers Life charges to develop the published benchmarks. He gave no actuarial basis for that decision, and no point of entry was given for affected insurers to question the decision. The benefit configurations in Petitioner's policy differ significantly from those in the Bankers Life policy from which Mr. Keating developed the published benchmarks. The differences in benefit configurations are material to the Subsection 627.9407(7)(c) comparison that Mr. Keating undertook. Mr. Keating used unpublished information in performing the comparison, and he could not recall the underlying methods and assumptions in making the comparisons. Therefore, some doubt was cast as to their validity. Further, no affected insurers were given a point of entry to question the comparisons or their underlying assumptions. The greater weight of the evidence shows that the policy decisions made by Respondent in implementing and applying Subsection 627.9407(7)(c), Florida Statutes, in many respects are not supported by logic or critical facts. Petitioner offered facts to demonstrate that Respondent's intended implementation and application of amended Subsection 627.9407(7)(c), Florida Statutes, would impair Petitioner's existing contracts, and would violate due process of law. The factual findings in this regard are set forth below. Petitioner's existing contracts were entered into at least nine years before amended Subsection 627.9407(7)(c), Florida Statutes, was proposed and became law, and before Respondent proposed to apply its benchmark charges to Petitioner. Petitioner's contracts are guaranteed-renewable contracts. Petitioner must continue to renew them indefinitely. When Petitioner entered into these contracts, the Laws of Florida, which were incorporated into the contracts, provided that Petitioner had the right to receive premiums for such guaranteed-renewable policies at rates actuarially justified under the provisions of Section 627.410, Florida Statutes, and the tests in Florida Administrative Code Rule 69O- 149.005(2)(b)1. Petitioner has demonstrated that the proposed rates in question are actuarially justified and meet those tests.
Recommendation Based upon the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that Respondent issue a final order approving Petitioner's rate increase request of 25.75 percent. DONE AND ENTERED this 8th day of June, 2007, in Tallahassee, Leon County, Florida. S ROBERT S. COHEN 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 8th day of June, 2007.
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.
The Issue Whether the Department has the authority pursuant to Section 627.062(2)(g), Florida Statutes, to disapprove Petitioner's Custom Homeowners' rates currently in effect throughout the State of Florida as "inadequate" in Zones I, II and III, and as "excessive" in Zone IV. If so, are Petitioner's Custom Homeowners' rates currently in effect throughout the State of Florida "inadequate" in Zones I, II and III, and "excessive" in Zone IV.
Findings Of Fact The Parties. Petitioner, Fortune Insurance Company (hereinafter referred to as "Fortune"), possesses a certificate of authority to conduct insurance business in Florida. Fortune engages in the business of writing homeowners insurance throughout Florida. Fortune's business offices are located in Jacksonville, Florida. Respondent, the Florida Department of Insurance (hereinafter referred to as the "Department"), is an agency of the State of Florida charged with the responsibility for, among other things, regulating rates charged for homeowners insurance in Florida. Fortune's Current Rates. In 1985 Fortune sought approval of homeowners insurance rates to be charged by it in four geographic zones it has divided Florida into. See Petitioner's exhibit 7. Fortune did not make any material misrepresentation in its filing. (Stipulated Fact). Nor was there any material error in Fortune's filing. (Stipulated Fact). The Department reviewed and approved Fortune's Custom Homeowners Program, "HO-8", insurance rates in January of 1986. (Stipulated Fact). The Department approved the following rates: a. Zone I: $387.00 Zone II: $389.00 Zone III: $332.00 Zone IV: $655.00 Zone IV includes, among other counties, all of the counties in the northern portion of Florida, including the Panhandle. No modification of Fortune's HO-8 rates was requested by Fortune between January of 1986 and June of 1993. Fortune's 1993 Request for a Rate Increase. On or about June 11, 1993 Fortune filed Petitioner's exhibit 1 with the Department requesting approval on an increase in rates pursuant to Section 627.062(2)(a)1., Florida Statutes. (Stipulated Fact). Fortune sought approval from the Department for an HO-8 rate increase for Zone I (to $443.00), Zone II (to $445.00) and Zone III (to $380.00). No HO-8 rate increase was requested by Fortune for Zone IV. Fortune suggested in the June 11, 1993 rate increase request (hereinafter referred to as the "1993 Request"), that it was requesting a statewide increase of 14.5 percent. The statewide increase sought actually amounted to a 12.9 percent increase. The parties communicated about the 1993 Request through the remainder of 1993. (Stipulated Fact). On January 25, 1994, the Department provided Petitioners exhibit 2 to Fortune proposing base rates for all four zones of Fortune's Custom Homeowners Program. (Stipulated Fact). The Department proposed to accept a statewide increase of 8.9 percent. This weighted average for the four zones included increases for Zones I, II and III and a decrease of approximately 49.9 percent for Zone IV. Fortune disagreed with the Department's determination. Therefore, Fortune withdrew the 1993 Request by letter dated February 18, 1994. (Stipulated Fact). Petitioner's exhibit 3. By letter dated March 7, 1994 the Department accepted the withdrawal of the 1993 Request. (Stipulated Fact). The Department informed Fortune that the withdrawal was "equivalent to the filing never having been submitted." Petitioner's exhibit 4. The Department's Notice to Fortune that Fortune's Rates are Excessive and/or Inadequate and Fortune's Response. Based upon the Department's review of the 1993 Request, the Department concluded that Fortune's HO-8 rates for Zones I, II and III were inadequate and that Fortune's HO-8 rate for Zone IV was excessive. By letter dated March 8, 1994, the Department informed Fortune, in part, of the following: You are hereby notified that pursuant to the provision of Section [627.062(2)(g)] the Department has reviewed the current rates, rating schedule, and rating manual for your Custom Homeowners Program and finds on a preliminary basis that certain rates are excessive and certain rates are inadequate. (Stipulated Fact). Petitioner's exhibit 5. The Department also informed Fortune in the March 8, 1994 letter that it would be given the opportunity to prove to the Department that "your current rates are not excessive, inadequate or unfairly discriminatory." The Department also informed Fortune in the March 8, 1994 letter that Section 627.062, Florida Statutes, provides that Fortune "shall, within sixty days of the date of this Notice, file with the Department all information which you believe proves the reasonableness, adequacy and fairness of your current rates." The Department also informed Fortune in the March 8, 1994 letter that it had the right to request a hearing pursuant to Chapter 120, Florida Statutes. The Department closed the March 8, 1994 letter with the following: If you request a hearing but intend to submit additional information within the allotted 60-day period you may request that the transmittal of your hearing request be delayed until the Department has had an opportunity to review the additional information submitted. Fortune did not provide additional information to the Department within sixty days of the Department's March 8, 1994 letter. On or about March 21, 1994 Fortune mailed a petition to the Department requesting a formal administrative hearing pursuant to Section 120.57, Florida Statutes. Prior to the March 8, 1994 letter from the Department, Fortune had provided all relevant information it had concerning its rates to the Department. Fortune, through Mr. Scourtis, verbally informed the Department that Fortune had no further information to support its rates. The evidence failed to prove that, between March 8, 1994 and March 21, 1994 when Fortune mailed its request for formal administrative hearing, Fortune had any other information concerning its rates which it had not provided the Department. The evidence also failed to prove that after March 21, 1994, when Fortune's request for formal hearing was mailed, the Department unsuccessfully attempted to obtain any information from Fortune through discovery. The Department's Review of Fortune's Rates. In determining whether Fortune's rates were inadequate or excessive, the Department first determined that a "credibility ballast" equal to the annual trend factor, or 4 percent, was generally accepted and reasonable actuarial technique. Fortune failed to prove that the use of a "credibility ballast" of 4 percent was not reasonable. The Department next determined that a "credibility factor" of 24.94 percent, the credibility factor Fortune had used in its 1993 Request, was generally accepted and reasonable actuarial technique. Fortune failed to prove that the use of a "credibility factor" of 24.94 percent was not reasonable. Based upon the foregoing, and using a catastrophic load factor of 1.141, the Department determined that the overall statewide indication for Fortune was the need to increase it's rates by 6.7 percent. Fortune had used the 1.141 catastrophic load factor in the 1993 Request. The Department agreed, and the evidence proved, that a catastrophic load factor of 1.357 is actuarially acceptable and reasonable. Using a 1.357 catastrophic load factor results in an increase of 9.1 percent. The evidence failed to prove that this conclusion is unreasonable. The Department next determined the relativity between the rates sought for the zones by Fortune to determine how to allocate the overall statewide rate increase indicator. The base rate sought for Zone I ($443.00) was used as the base. The base rate sought for Zone III was .86 of the base rate in Zone I ($380.00/$443.00) and it was concluded that this relativity was reasonable. The evidence failed to prove that this conclusion was unreasonable. The relativity between Zone I and Zone II was determined to be unreasonable because the rate sought for Zone II ($445.00) was greater than that of Zone I ($445.00/$443.00 or 1.01). The Department concluded that the relativity of Zone II should have been between Zones I and III. Therefore, the relativity of Zone II was changed to .94 by reducing the rate sought by Fortune in its 1993 Request from $445.00 to $400.00 ($400.00/$443.00 or .94). The evidence failed to prove that this conclusion was unreasonable. The relativity between Zone I and Zone IV was also determined to be unreasonable: $655.00 rate for Zone IV divided by the rate sought for Zone I of $443.00 or a relativity of 1.48. This relativity was determined to be unreasonable and was reduced to .77, which placed it below Zones I, II and III. The evidence failed to prove that this conclusion was unreasonable. Having determined the relativity of the Zones, the Department determined the appropriate rate for Zone I based upon its determination of the overall statewide rate increase indicator. Except for the fact that the Department should have based this final calculation on a catastrophic load factor of 1.357 and not 1.141, resulting in an overall statewide rate increase indicator of 9.1 instead of the 8.9 indicator utilized by the Department, the evidence failed to prove that the Department's calculation of the rates for the four zones based upon the relativities determined by the Department was unreasonable. Fortune's Zone I, II and III Rates. The parties agreed that the rates for Zones I, II and III are inadequate. The only disagreement of the parties involved the extent of the inadequacy. Fortune argued that the inadequacy of its rates for Zones I, II and III is greater than that determined by the Department. Fortune failed to prove that the Department's determination of the extent of the inadequacy of the rates for Zones I, II and III was incorrect except to the extent that the Department used a catastrophic load factor of 1.141 instead of 1.357. The correct overall statewide rate increase indicator to be utilized in determining the extent of inadequacy of the Zones I, II and III rates should be 9.1 instead of 8.9. Fortune's Zone IV Rate. The Department determined that Fortune's Zone IV rate was excessive. Fortune failed to prove that the Department's determination that the Zone IV rate is excessive is incorrect except to the extent that the Department used a catastrophic load factor of 1.141 instead of 1.357. The correct overall statewide rate increase indicator to be utilized in determining the extent of excessiveness of the Zone IV rate should be 9.1 instead of 8.9.
