OPINION AND ORDER
OPINION BY
Judge LEAVITT.TABLE OF CONTENTS AND GLOSSARY
Page
I. Introduction ..374
II. Syllabus 375
*371III.Findings of Fact.
A. Business and History of PTNA and ANIC.
i. The Companies.
ii. The Companies’ Products.
iii. OldCo and NewCo Business.
iv. Reserves .
v. Reinsurance.
vi. Financial Condition of the Companies.
vii. Conditions that Led to Rehabilitation.
viii. The Rehabilitation Orders.
ix. April 2009 Preliminary Report and Plan of Rehabilitation ...
x. Rehabilitation Implementation Committee.
xi. Milliman August 2009 PowerPoint Projections.
xii. Termination of New Premium Rate Increase Filings.
B. Companies’ Statutory Surplus as of December 31,2009 .
C. Scope of Expert Evidence.
i. Rehabilitator’s Experts.
ii. Intervenors’ Experts.
iii. Actuarial Projections of the Companies’ Financial Conditions
iv. Actuarial Principles and Concerns.
v. Changes in Milliman’s Projections.
vi. Actuarial Caveats and Limitations.
vii. Credibility of Data Used in Projections.
viii. Actuarial Standard of Practice No. 25.
ix.Actuarial Standard of Practice No. 18.
D. Milliman Expert Report and Testimony.
i. Milliman Morbidity or Claim Costs Assumption.
ii. Milliman Morbidity Improvement Assumption.
iii. Milliman Mortality Assumption.
iv. Milliman Voluntary Lapsation Assumption.
v. Milliman Interest Rate Assumption .
vi. Milliman Claim Adjudication and Wellness Assumption.
vii. Milliman Premium Rate Increase Assumption.
E. United Health Actuarial Services, Inc. Expert Report and Testimony
i. United Health Morbidity or Claim Costs Assumption.
ii. United Health Morbidity Improvement Assumption.
iii. United Health Mortality Assumption.
iv. United Health Voluntary Lapsation Assumption .
v. United Health Interest Rate Assumption.
vi. United Health Claim Adjudication and Wellness Assumption
vii. United Health Premium Rate Increase Assumption .
viii. United Health Agents’ Commission Assumption.
F. Gross Premium Reserve.
G. Ernst & Young Report.
H. State Regulation of Long-Term Care Insurance Rate Making.
I. Impact of Medical Advances on Future Long-Term Care Insurance Claims.
J. Evaluation of Actuarial Projections and Expert Testimony .
IV. Legal Analysis.439
A. Statutory Standard for Conversion of a Rehabilitation to a Liquidation.439
B. Insolvency.441
C. Substantial Increase in Risk of Loss to Policyholders.441
D. Futility of Continued Rehabilitation.446
i. Meaning of “Futility”.446
ii. Rehabilitator’s Case for Futility .448
V. Conclusions of Law.458
*372VI. Conclusion. .458
GLOSSARY
Active Life Reserve — a reserve established for claims to be presented in the future by active lives.
Active Lives — long-term care insurance policyholders not on claim.
Ario — Joel Ario, former Insurance Commissioner of the Commonwealth of Pennsylvania from 2007-2010.
ANIC — American Network Insurance
Company, a subsidiary of PTNA.
AINIC — American Independent Network Insurance Company, a New York domiciled insurance company and subsidiary of ANIC.
ASOP 18 — Actuarial Standard of Practice No. 18: Long-Term Care Insurance.
ASOP 25 — Actuarial Standard of Practice No. 25: Credibility Procedures Applicable to Accident and Health, Group Term Life and Property/Casualty Coverages.
ASOP 41 — Actuarial Standard of Practice No. 41: Communication.
Assumption — an estimate of an uncertain variable input into a financial model often based upon historic experience of a company and/or industry.
Bodnar — Vincent Bodnar, an actuary and principal with DaVinci Consulting, LLC, retained by NOLHGA in 2009 and later by the Rehabilitator as an expert.
Brookfield — Brookfield Investment Management, Inc., the investment advisor to the Companies.
CBER — “Closed But not Expected to Reopen” claims for which liability exists but is not yet known. It is part of the statutory claim reserve.
Claim Reserve — a reserve established for incurred claims, known and unknown.
Cloutier — Mark Cloutier, former Chief Financial Officer of PTNA.
Company or Companies — refers collectively to PTNA and ANIC.
Conrad — Robert Conrad, a claims specialist employed by the Pennsylvania Insurance Department, Office of Rehabilitations, Liquidations and Special Funds.
Consedine — Michael F. Consedine, Insurance Commissioner of the Commonwealth of Pennsylvania from 2011 to the present.
Continuance Curves — graphic depiction of the probability that an individual policyholder will stay on claim, based upon historical data on a number of potential factors (age, sex, type of benefit, etc.)
Dalton — Andrew Dalton, an actuary employed by Milliman.
Department — the Pennsylvania Insurance Department.
DiMemmo — Joseph DiMemmo, Deputy Insurance Commissioner, Pennsylvania Insurance Department, Office of Liquidations, Rehabilitations and Special Funds.
Disabled Life Reserve — reserve established for claims of disabled lives that are known. It is part of the statutory claim reserve.
Disabled Lives — long-term care insurance policyholders on claim.
Ernst & Young (E & Y) — actuarial experts retained by the Rehabilitator.
Guaranty Fund — statutory fund established to pay claims of resident policyholders whose insurer is declared insolvent.
Heitkamp — Carl Heitkamp, an actuary employed by UHAS.
Holland — Dr. Stephen Holland, Chief Medical Officer of Univita, Inc. and a testifying expert for Intervenors.
*373Hunt — William Hunt, former Chief Executive Officer of PTNA and ANIC and a member of the Board of Directors of PTAC.
IBNR — “Incurred But Not Reported” claims for which liability exists but which are not yet known. It is part of the statutory claim reserve.
Intervenors — Eugene Woznicki and PTAC.
Johnson, Stephen — Deputy Insurance Commissioner, Pennsylvania Insurance Department, Office of Corporate & Financial Regulation.
Johnson, Thomas — a principal of Signal Hill and a member of the Rehabilitation Implementation Committee.
Kroll — James Kroll, an actuary employed by PTNA.
LaPierre — Stephen LaPierre, former Executive Vice President of Insurance Operations for the Companies.
Lapse — termination of a policy due to nonpayment of premium, cancellation by the policyholder or death of the policyholder.
Litow- — Mark Litow, an actuary and principal of Milliman. Litow authored a projection report for the Companies dated April 28, 2008, referred to as the Litow Report.
Loss Ratio — the percentage of premium collected that is used to pay claims. Long-term care insurance must have a loss ratio of 60% in most states.
Lucker — Arthur M. Lueker, an actuary with INS Consultants, Inc. and a testifying expert for the Rehabilitator.
Milliman — Milliman, Inc., the actuarial firm retained by the Companies since 2002 to perform actuarial services; subsequently retained by the Rehabilitator to perform actuarial services for both PTNA and ANIC and to provide expert evidence in support of the Rehabilitator in the present litigation.
Minches — David Minches, an actuary and principal of Ernst & Young and testifying expert for the Rehabilitator.
Mohoric — Edward Mohoric, an actuary and principal of Milliman and testifying expert for the Rehabilitator.
Morbidity — a policyholder’s propensity to be on claim.
Morbidity Improvement — a long-term trend in societal changes and improvements in drugs or breakthroughs in technology that improve morbidity over time.
Mortality — the rate of deaths in a specific population.
NAIC — National Association of Insurance Commissioners.
Negative Surplus — see Surplus.
NewCo — long-term care insurance policies sold by the Companies beginning in 2002. NOLHGA — National Organization of Life & Health Insurance Guaranty Associations.
Non-Forfeiture Option — provision in an insurance policy allowing the policyholder to forfeit the policy, cease paying premiums, and receive a refund of some or all of the premiums already paid.
OldCo — long-term care insurance policies sold by the Companies prior to 2002.
Persistency — (1) for claimants, the probability of staying on claim over a certain period of time, or (2) for non-claimants, the propensity for policyholders to stay insured by the Companies.
Petitioner — Insurance Commissioner of the Commonwealth of Pennsylvania, in his capacity as statutory rehabilitator of PTNA and ANIC.
Pfannerstill — Larry Pfannerstill, an actuary and principal of Milliman and testifying expert for the Rehabilitator.
*374Preliminary Plan — the Preliminary Report and Plan of Rehabilitation for the Companies submitted by the Rehabilitator on April 9, 2009, and describing the Reha-bilitator’s plan for rehabilitation and timetable for implementing the plan.
PTNA — Penn Treaty Network America Insurance Company.
PTAC — Penn Treaty American Corporation, the parent company of PTNA and indirectly ANIC.
Rate Increase or Premium Rate Increase — an increase in the premium rates charged for a long-term care insurance product.
Rehabilitator — Insurance Commissioner of the Commonwealth of Pennsylvania, in his capacity as statutory rehabilitator of PTNA and ANIC.
RIC — the Rehabilitation Implementation Committee for PTNA and ANIC.
Robinson — Robert L. Robinson, Chief Rehabilitation Officer for PTNA and ANIC.
Scenario A, Scenario B — alternative surplus projections for PTNA and ANIC used by Milliman in its September 2009 Report and its October 2009 Report.
Scenario A, Scenario B, Scenario C, Scenario D — alternative rate increase assumptions for PTNA and ANIC developed by Volkmar in his expert report.
Severity — the expense of a claim in long-term care insurance caused by the duration of a claim.
Shock Lapse — sudden termination of policies by policyholders in reaction to a certain event, e.g., a premium rate increase.
Signal Hill — a financial advisory and investment banking company retained during the rehabilitation process to evaluate certain rehabilitation alternatives available to PTNA and ANIC.
Surplus — a company’s assets minus its liabilities. A negative surplus is the amount by which a company’s liabilities exceed its assets.
UHAS — United Health Actuarial Services, Inc., the Intervenors’ actuarial expert in the present litigation.
Unearned Premium Reserve — a reserve established to cover the insurer’s liability to refund premium to policyholders whose policies terminate before the policy’s renewal date.
Volkmar — Karl Volkmar, an actuary and principal of UHAS and a testifying expert for Intervenors.
Waite — Cameron Waite, former Chief Financial Officer and Executive Vice President of PTAC, PTNA and ANIC, and consultant for the Rehabilitator.
Woznicki — Eugene Woznicki, Chairman of the Board of PTAC and PTNA and an Intervenor in this matter.
I. Introduction
Before the Court are the consolidated petitions of Michael F. Consedine, Pennsylvania Insurance Commissioner and Statutory Rehabilitator of Penn Treaty Network America Insurance Company and American Network Insurance Company. By these petitions, the Commissioner, in his capacity as Rehabilitator, seeks to convert the rehabilitations of both Companies into liquidations. The standard governing the petitions is a simple one. The Rehabil-itator must prove that continued rehabilitation will substantially increase the risk of loss to policyholders, creditors and the public or is futile.
Combined, the Companies have approximately $1 billion in assets, no debt and are meeting all obligations as they come due. The Companies’ cash flow, in excess of $200 million per year, has been sufficient to pay all policyholder claims in full and on *375a timely basis. On these facts, there is no dispute. Likewise, there is no dispute that the Companies will continue to be able to pay all policyholder obligations, timely, for years to come. Nevertheless, the Companies are insolvent because their existing premium rates are too low to fund all expected future claims, and the Companies cannot non-renew these under-priced policies.
The Insurance Commissioner, wearing his hat as a regulator of the Pennsylvania insurance industry, refused to approve the Companies’ aetuarially justified rate increase filings in the amount requested, both before and after rehabilitation. The Commissioner has even discouraged other state regulators from approving rate increases. Now the Commissioner seeks to liquidate the Companies because their premium rates are inadequate.
Because he has not undertaken a meaningful effort to rehabilitate the Companies and, to the contrary, has acted to frustrate rehabilitation, the Commissioner has not met his burden of proof. For this and other reasons set forth in this Opinion, the petitions are denied.
II. Syllabus
On January 6, 2009, this Court ordered the rehabilitation of Penn Treaty Network America Insurance Company (PTNA) and its subsidiary, American Network Insurance Company (ANIC), which are both Pennsylvania life insurers specializing in long-term care insurance. The Court’s order was issued upon application of the Pennsylvania Insurance Commissioner, Joel Ario, who cited the consent of both insurance companies as the sole grounds for the orders of rehabilitation.1 The orders were issued under Section 515(b) of Article V of the Insurance Department Act of 1921, Act of May 17, 1921, P.L. 789, added by Section 2 of the Act of December 14, 1977, P.L. 280 (Article V), which provides that:
[a]n order of the Commonwealth Court to rehabilitate the business of an insurer shall be issued only after a hearing before the court or pursuant to a written consent of the insurer.
40 P.S. § 221.15(b). The Court’s orders appointed the Insurance Commissioner as Statutory Rehabilitator.
On April 9, 2009, the Rehabilitator submitted a Preliminary Report and Plan of Rehabilitation (Preliminary Plan) that described the history of each company and, in broad terms, described the Rehabili-tator’s plans for rehabilitation and the timetable for implementing the plan. The Preliminary Plan explained that because the premium rates for the Companies’ largest block of business, known as “Old-Co” policies, were inadequate, this business could not be sold to other insurers. For the same reason, capital could not be raised. Accordingly, the Rehabilitator fixed on a course of systematic increases in the OldCo premium rates and reduction of expenses. Attached to the Preliminary Plan were reports from the Companies’ consulting actuarial firm, Milliman, and the Companies’ financial advisor, Signal Hill. These consultants concluded that PTNA could return to a positive surplus by rate increases; by offering policyholders reduced benefit levels; or by offering a non-forfeiture option in lieu of the indicated rate increases. The Rehabilitator reported that he intended to submit a formal rehabilitation plan with more specifics on *376these options by October 4, 2009, for the Court’s review and approval.
Instead, on October 2, 2009, the Rehabil-itator filed petitions with the Court to convert the rehabilitations of PTNA and ANIC into liquidations. On October 23, 2009, the Rehabilitator filed amended petitions for liquidation to correct an error of $200 million. The correction favored the Companies.
On November 2, 2009, Penn Treaty American Corporation (PTAC), the owner of PTNA, and Eugene J. Woznicki, Chairman of the Board of Directors of PTNA and of PTAC, petitioned to intervene to oppose the liquidation petitions. The Court granted the intervention petition.
The year 2010 was devoted to attempted settlement, discovery and preparation of expert testimony and reports. Trial began on January 3, 2011, but paused while, again, the parties sought an interim resolution. The trial resumed in October of 2011. In all, there were 30 days of hearing testimony, and the Court received thousands of pages of exhibits and documentary evidence. Post-hearing briefs, in the aggregate, exceeded 1,000 pages in length. Closing arguments took place on February 21 and 22, 2012.
The Companies’ business written after 2001, consisting of what they call the “NewCo” policies, is priced correctly and, thus, profitable. The challenge lies with the premium rates for the Companies’ Old-Co policies, which are inadequate to cover the expected future claims. For the past decade, the Companies have pursued rate increase filings on OldCo policies, successfully in some states but not in all. This was not for lack of actuarial justification for those filings. The Rehabilitator’s evidence showed that rate regulation is governed by politics, not actuarial evidence or legal principles. The Rehabilitator has even included Pennsylvania in the list of problem states that have refused to approve the Companies’ actuarially justified rate increase filings for the OldCo policies. This case presents a serious indictment of the existing system of rate regulation of long-term care insurance.
The need for rate relief has increased over time as claims experience developed. In turn, the inadequate rates caused the Companies’ actuaries to conclude that the Companies’ reserves, or liabilities, had to be increased. The inadequate premium rates also caused the Companies to lose their reinsurance. Accordingly, PTNA’s surplus fell to negative $141 million at the point it was placed into rehabilitation. By April of 2009, the actuaries projected PTNA’s surplus to have further declined, to negative $223 million.
The Rehabilitator suggests that management “masked” the true financial condition of the Companies and that the truth was only unearthed by the Rehabilitator’s careful examination. Both claims are unfounded.2 First, the rate filings and reserves for each company, both before and during rehabilitation, were prepared by actuaries, not by management. Indeed, it was the same actuarial firm, Milliman, that did this work both before and during rehabilitation, and it used the Companies’ claim data for its actuarial work. The Rehabilitator’s *377own witness, Stephen LaPierre, former Executive Vice President of Insurance Operations for the Companies, offered credible testimony about the efforts of the Companies’ claim department to act promptly on claims and to set responsible claim reserves. Second, it was not the Rehabilitator’s “careful examination” that “unearthed” another financial picture of the Companies. Rather, it was Milliman’s volte-face on its own statutory reserves, or liabilities, that caused the Companies’ indicated surplus to deteriorate.
Nevertheless, there has been a lot of careful and hard work by the rehabilitation team since the takeover. Robert L. Robinson, Chief Rehabilitation Officer, is diligent, knowledgeable and well organized.3 He is not an expert on long-term care insurance, but he has learned fast. He and his team have worked to understand and manage the daily business affairs of the Companies, and they have brainstormed on a number of approaches for returning the Companies to solvency. Robinson was of the opinion that several rehabilitation options could have been supported when the decision to pursue liquidation was made. Respondent’s Exhibit No. 14 (Ex. R-); N.T. 2/14/11 at 139-42, 165. However, Robinson did not direct Milliman or Signal Hill to draw up a formal plan of rehabilitation; it is not clear that he had authority to issue such a directive.
Robinson reports to Joseph DiMemmo, who is the Deputy Insurance Commissioner for the Office of Liquidations, Rehabili-tations and Special Funds of the Pennsylvania Insurance Department. Intervenors suggest that DiMemmo’s decision to convert the proceeding to a liquidation was made in pique, and they fault DiMemmo’s “angry” decision to shut down work on a rehabilitation plan that included benefit modifications to the OldCo policies as well as premium rate increases. Bureaucrats are allowed to emote and why DiMemmo decided to pursue liquidation is simply irrelevant. The only question addressed by the Court was whether the liquidation petition should be granted on the evidence presented and on the governing law. In any case, the Court presumes that DiMem-mo made the decision to pursue a conversion from rehabilitation to liquidation because he believed it to be the appropriate course.
Because Milkman was part of Robinson’s rehabilitation team and also did consulting work for the National Organization of Life and Health Insurance Guaranty Associations (NOLHGA), Intervenors question the firm’s independence. Milli-man’s August 2009 projections caused the Rehabilitator to fix on a kquidation course; the Rehabiktator then offered Milkman’s 2010 expert report in support of Milkman’s 2009 projections. Intervenors ask how Milliman can independently evaluate its own work, and they assert that this purported lack of independence reduces the weight the Court should assign to Milkman’s expert report. The Court understands Intervenors’ point, but it did not factor this point into its consideration of Milkman’s work.
The parties, their consultants and lawyers have been diligent in preparation and presentation of the evidence. The trial was conducted with a commendable level of civility. The post-hearing briefs were *378thorough and careful.4 It is the Court’s responsibility to decide the case on the evidence without regard to the sincere convictions of either the Rehabilitator or In-tervenors.
Insolvency exists when an insurer is unable to meet obligations as they come due or because its assets are insufficient to fund all future liabilities. Here, the evidence proved the second type of insolvency; however, the parties sharply disputed the extent of asset deficiency. The Reha-bilitator’s expert, Milliman, projected PTNA’s statutory surplus to be negative $2.1 billion and ANIC’s surplus to be negative $137 million, as of December 31, 2009. Intervenors’ expert, United Health Actuarial Services, Inc., projected PTNA’s statutory surplus to be negative $833 million and ANIC’s surplus to be positive $600,000, as of December 31, 2009.5
The Court rejects Milliman’s projected surplus. In the space of six months Milli-man increased its claim projections for PTNA by over $1 billion, a revision of a magnitude never before seen by any of the experts who testified. This increase did not result from new claim data received over that period of time but, rather, Milli-man’s negative recast of its own prior projections. Nevertheless, even accepting Milliman’s projected negative surplus, it does not follow that revenues over the next 10 to 20 years cannot be increased in amounts sufficient to pay future claims generated by the OldCo policies over the next 60 years, the majority of which are expected to be presented in the next 10 to 20 years.
In a liquidation, all policyholders will have their policies cancelled in 30 days. At that point the guaranty fund in the state where a particular policyholder resides will offer some kind of replacement coverage.6 From that point forward, policyholders will pay their premium to the appropriate guaranty fund, but the level of replacement coverage they receive will, in most cases, be less than what they have under their policies issued by PTNA or ANIC. State legislatures generally cap the level of guaranty fund coverage provided to their residents.
The Rehabilitator believes, drawing on hoary Pennsylvania precedent that predates the modern statutory insurance insolvency scheme, that policyholders may sue the estates of PTNA and ANIC for breach of contract caused by the termination of their policies in liquidation. This will enable them, the Rehabilitator theorizes, to recover prospective losses that may arise years after their policies have been cancelled in a liquidation. The Rehabili-tator’s novel argument is inconsistent with the position taken by the Department in other liquidations.7 The argument also finds no support in the policy itself, which makes no promise about post-liquidation rights. The potential that the Rehabili-tator’s legal argument may find ultimate success is not evidence, and it does not support immediate liquidation.
*379The Rehabilitator expresses concern that continued rehabilitation may result in some policyholders being treated more favorably than others. This is not very persuasive in light of the fact that a liquidation guarantees widely disparate treatment of policyholders because different states offer different levels of guaranty fund coverage. PTNA policyholders in South Dakota will have their coverage capped at $100,000 even though they may be paying a rate appropriate for a “Cadillac” policy with the highest benefit level.8 ANIC policyholders in New Jersey will have no limits on their post-liquidation coverage.
The risk facing policyholders is that they will not receive the full level of benefits presently provided in their policies. This risk is the same whether the Companies are liquidated today or in ten years. Indeed, that risk may be reduced over time should states increase the amount of coverage provided by their guaranty funds.
The Rehabilitator purports to speak for the policyholders, but there is no empirical evidence on policyholder preference. Robinson conducted meetings with two small groups of policyholders in Philadelphia, after the decision to pursue liquidation had been made. It is impossible to draw any inference about policyholder preference from Robinson’s report of what these policyholders said at the meetings.
Some states, such as Virginia and South Dakota, have acted responsibly on the Old-Co rate filings. As a result, policyholders there are paying close to the “national rate,” i.e., the amount needed to pay claims in any and all states for a particular product.9 Other states, including Pennsylvania, have refused to approve actuarially sound rate filings. (Waite) N.T. 2/1/11 at 104 (describing Pennsylvania as a “difficult state”). Rate inequities are the result. Policyholders in South Dakota, for example, are subsidizing policyholders in Pennsylvania; NewCo policyholders are subsidizing OldCo policyholders.
This history does not make continued rehabilitation futile. There is no reason for presuming that state regulators will respond to rate increase filings presented as part of a rehabilitation plan, which itself will show how a turnaround can be achieved, as they have responded in the past to business-as-usual rate increase filings. A liquidation will shift to the taxpayer the ultimate cost of a state’s refusal to grant actuarially sound rate increases because taxpayers will have to reimburse the guaranty funds. The taxpayer burden should factor into the deliberations of the state regulators who are a necessary part of a workout. A rehabilitation plan need not involve only rate increases; it may also involve benefit modifications to the OldCo policies that would cost the policyholders less and still provide them reasonable coverage.
In mid-2009, in anticipation of a liquidation, the Rehabilitator abandoned its policy of aggressive pursuit of rate increases. Commissioner Ario advised state regulators to follow his example and not approve pending rate filings on OldCo business.
*380In spite of no new rate increases, the 2010 results were more positive than Milli-man had projected. As of December 31, 2010, the Companies have not been forced to liquidate assets to pay claims. In fact, the Companies’ combined assets of approximately $1 billion actually grew in 2010. In 2010, cash flow was approximately $277 million; claims in the amount of $236 million were paid without liquidating assets. This is not to say that the Companies do not have a serious cash flow challenge. They do. Their ability to pay claims, however, is not an immediate problem. Even in the Rehabilitator’s worst case scenario, the Companies will be able to pay existing and future claims for many years.
The Rehabilitator’s evidence did not show that a rehabilitation was tried and failed. Rather, it showed that a rehabilitation plan was abandoned in its nascency. In short, the Rehabilitator did not prove that continued rehabilitation substantially increases the risk to policyholders, creditors and the public or is futile. The quality assets held by the Companies and the existence of guaranty funds provide a safety net for their policyholders during continued rehabilitation. By contrast, liquidation promises immediate harm to the policyholders, creditors and the public, as the Rehabilitator acknowledged to the Court in his Preliminary Plan.
III. Findings of Fact
A. Business and History of PTNA and ANIC through Rehabilitation i. The Companies
PTNA and ANIC are Pennsylvania stock life insurance companies headquartered in Allentown, Pennsylvania. PTAC is the sole shareholder of PTNA, and ANIC is a direct and wholly-owned subsidiary of PTNA.10 The Companies are licensed to issue annuities and life, accident and health insurance policies. PTAC, through its insurance company subsidiaries, has been writing long-term care insurance business since 1972. Historically, the Companies have written a de minimis amount of other insurance coverages, including Medicare supplement policies and life insurance.
As of December 31, 2008, the Companies had approximately 142,000 policyholders generating approximately $290 million of annual in-force premium, 97.7% of which was from long-term care insurance policies, 2.2% from Medicare supplement policies, and 0.1% from other insurance. PTNA and ANIC, until they ceased underwriting in October 2008, wrote new business through a network of over 17,000 independent agents. The Companies are licensed in 44 states and the District of Columbia, and their in-force business is distributed across 49 states and the District of Columbia. As of October 20, 2008, the Companies had 270 employees. The Companies manage claims internally, and as of April 2009, employed 79 individuals in their claims and case management department, including 10 full-time registered nurse case managers, 38 full-time claims examiners and numerous additional support personnel.
ii. The Companies’ Products
The Companies’ principal product is long-term care insurance. Long-term care insurance provides coverage for some of the costs of skilled nursing, intermediate care, custodial care and home health care for a person who needs assistance due to chronic illness or disability. Such care is *381provided to individuals in their home or in an adult day care facility, nursing home or assisted living facility. To receive this coverage, a policyholder must meet certain benefit triggers, which vary depending on whether the policy is tax qualified or non-tax qualified.
Long-term care insurance policies were designed and sold as non-tax qualified policies prior to federal legislation enacted in 1996, which created tax qualified long-term care insurance policies. Under Section 213 of the Internal Revenue Code, premiums paid for a tax qualified long-term care insurance contract are deductible medical care expenses. 26 U.S.C. § 213(d)(1)(D). As rephrased from the United States Code, a tax qualified long-term care insurance contract is one that:
(a) provides insurance coverage only for qualified long-term care services;
(b) does not pay or reimburse expenses that are reimbursable under Medicare;
(c) is guaranteed renewable;
(d) does not provide for a cash surrender value or other money that can be paid, assigned or pledged as collateral for a loan, or borrowed;
(e) provides that all refunds of premiums (other than refunds on the death of the insured or on a complete surrender or cancellation of the contract, which cannot exceed the aggregate premiums paid under the contract) and policyholder dividends, or similar amounts, are to be applied as a reduction of future premiums or to increase future benefits; and
(f) satisfies certain consumer protection requirements as well as disclosure and nonforfeitability requirements.
See 26 U.S.C. § 7702B(b)(l) (emphasis added). “Qualified long-term care services” are defined as
necessary diagnostic, preventive, therapeutic, curing, treating, mitigating, and rehabilitative services, and maintenance or personal care services, which—
(A) are required by a chronically ill individual, and
(B) are provided pursuant to a plan of care prescribed by a licensed health care practitioner.
26 U.S.C. § 7702B(c)(l) (emphasis added). A “chronically ill individual” is one who has been certified within the previous 12 months by a licensed health care provider as
(i) being unable to perform (without substantial assistance from another individual) at least 2 activities of daily living for a period of at least 90 days due to a loss of functional capacity,
(ii) having a level of disability similar ... to the level of disability described in clause (i)[, as determined by the Internal Revenue Service in consultation with the Department of Health and Human Services], or
(iii) requiring substantial supervision to protect the individual from threats to health and safety due to severe cognitive impairment.
26 U.S.C. § 770213(c)(2) (emphasis added). For purposes of federal tax law, “activities of daily living” are “eating, toileting, transferring, bathing, dressing and continence.” 26 U.S.C. § 7702B(c)(2)(B).11
The Companies, like most long-term care insurance companies, began to focus on selling tax qualified products by the early 2000s. In order to be tax qualified, a policy must conform to the above statutory *382requirements, most notably with respect to benefit triggers. For example, a sample policy form for PTNA’s tax qualified Assisted Living Plus II product contains the following conditions of eligibility:
Subject to all other provisions, you become eligible to receive the benefits of this Policy when, due to illness or injury:
1) you require Human Assistance with two or more Activities of Daily Living[12] for a period of at least 90 days; OR
2) you have Severe Cognitive Impairment, (which may be caused by Alzheimer’s disease, Organic Brain Syndrome or senile dementia, etc.)
Ex. R-789, Form No. ALP2-TQ-P(AZ) at 17. Other PTNA and ANIC tax qualified policy forms contain substantially similar language.
By contrast, a non-tax qualified policy contains more benefit triggers, which makes coverage easier to access. For example, one PTNA non-tax qualified Personal Freedom product sets forth the following conditions of eligibility for an Assisted Living Facility:
You become eligible to receive the Assisted Living Facility Benefit when:
1) Your Physician certifies Your confinement to be Medically Necessary. . . . [13]
OR
2) You are unable to perform two (2) or more of the Activities of Daily Living without human assistance or continual supervision....
OR
3) You are afflicted with Cognitive Impairment. ...
Ex. R-789, Form No. PF2600(PA)-P at 9.14 The “medically necessary” trigger makes benefits more accessible for a non-tax qualified policyholder.
A “medically necessary” claimant may be able to perform all activities of daily living without assistance. Eligibility is not based upon objective criteria but a subjective standard; ie., a treating physician’s judgment. Accordingly, non-tax qualified policies with a “medically necessary” trigger are “more challenging ... to manage from a claim perspective.” (LaPierre) N.T. 9/19/11 at 226. LaPierre explained that the claim cost management problem has been exacerbated by looser underwriting standards employed by the Companies before 2002. Id. at 227. The Rehabili-*383tator’s actuary, Edward Mohoric, agreed that “medically necessary” is a lower standard for triggering benefits, requiring the Companies to experience higher claim costs for non-tax qualified policies. N.T. 2/24/11 at 75-76.
The difficulty with managing non-tax qualified policies was set forth in an internal analysis done on October 3, 2010:
While a majority of the long-term care insurance industry began to focus on selling the new [tax qualified] product design in the late 1990s and early 2000s, PTNA continued to sell primarily the [non-tax qualified] product design through 2001. Consequently, 73% of in-force policies are of the [non-tax qualified] design, posing an ongoing challenge to benefit eligibility management.
The 2600 product form, [a non-tax qualified] policy, was marketed between 1996 and 2001. This product form contains the most liberal of all benefit triggers, the Instrumental Activities of Daily Living (“IADL”) benefit trigger. This allows a policyholder to qualify for benefits when they have experienced the least restrictive of physical functioning decline. Benefits currently paid on the 2600 product represented 27% of total claim benefit payments in 2009.
Petitioner’s Exhibit No. 266 (Ex. P — ) at 21-22.
After a predetermined deductible period, a policyholder with a covered claim receives a daily benefit ranging from $60 to $300 per day for a maximum benefit period ranging from one to 10 years. Some policies offer unlimited lifetime benefits. A policyholder will go off claim upon death or if he or she recuperates from the condition that caused the claim. The Companies’ long-term care insurance policies are guaranteed renewable, meaning that policyholders are guaranteed the right to renew their policies irrespective of their age or health as long as they pay their premiums.
Initial premium rates are established at the time the policy is first purchased. The policies provide that premiums will not increase because of the policyholder’s age or medical condition. On the other hand, the policies provide that premiums are subject to increase if such increases become actuarially supported for the entire group of policyholders. The Companies disclosed this fact with a “Premiums Subject to Change” provision on the first page of their policies. Two examples follow:
PTNA — ASSISTED LIVING PLUS II
Premiums Subject To Change — 5 Year Rate Guarantee
The premiums of this Policy can never be changed because your age has changed or because of a change in your individual health. We can change the premiums for this Policy if we change them for everyone that bought this Policy in the same state yours was purchased. We cannot, however, change your premiums during the first five years this Policy is in force. A change in premiums would first have to be filed with the state’s Commissioner of Insurance. Notice of any such change in premiums will be sent at least 45 days in advance of the new premium becoming payable.
Ex. R-789, Form No. ALP2-TQ-P(AZ) at 1.
PTNA — COMPREHENSIVE LONG-TERM CARE POLICY RENEWABILITY GUARANTEED RENEWABLE-PREMIUMS SUBJECT TO CHANGE
This Policy is guaranteed renewable for Your lifetime. It may be kept in force by the timely payment of premiums. We cannot refuse to renew this Policy as long as You pay the premiums. We can *384change the renewal premium rates. We can only change them if they are changed for all policies in Your state on this Policy Form. Renewal premiums due after a change is implemented will be based on the new rate. Notice of any change in rates will be sent at least thirty-one (31) days in advance.
Ex. R-789, Form PF2600(PA)~P at 1.
As noted, the Companies began writing long-term care policies in 1972. By 2001, the Companies, which had “grown very quickly,” found their claims were “somewhat higher than ... expected.” (Mohoric) N.T. 2/16/11 at 85. As was the entire long-term care insurance industry, PTNA and ANIC were affected by the increased availability of home health care services and assisted living facilities in the 1990’s and 2000’s. Prior to that time, there had been an underutilization of policies because home health care and assisted living facility services were not available in many areas of the United States, or the waiting list to obtain such care was very long. As the availability of such care expanded, long-term care claims began to rise to a degree that no long-term care insurer had anticipated.
As a result of these growing pains, the Companies concluded that they needed to revise the assumptions used to set their claim reserves, ie., the amount set aside to meet future claim liabilities. The historic assumptions for policy persistency and claim frequency and severity had resulted in inadequate claim reserves. The required reserve strengthening placed PTNA’s adjusted capital and surplus below the Regulatory Action Level15 as of December 31, 2000. In response, PTNA filed a Corrective Action Plan with the Department, and it approved PTNA’s plan on February 12, 2002.
As part of the Corrective Action Plan, PTNA and ANIC paused writing business for six months while they instituted efforts to strengthen their finances. Accordingly, the Companies entered into a reinsurance treaty with Centre Solutions (Bermuda) Limited, which reinsured 100% of the individual long-term care business in effect on December 31, 2001. This treaty was commuted effective May 24, 2005, and the Companies entered into a new reinsurance agreement covering this business with Imagine International Reinsurance Limited (Imagine) effective June 30, 2005. Other significant components of the Corrective *385Action Plan included making improvements to the Companies’ claims and underwriting departments; writing new and more profitable business; and repricing the existing OldCo business with rate increases.
Milliman recommended and prepared the premium rate increase filings. The first round of rate increases took place in 2001, and this was followed by rate increase filings in 2003, 2005, and 2006. The 2001 rate increase requests averaged 34%, and the Companies received approval for and implemented 92% of the aggregate requested increases. The 2003 rate increase requests averaged 16%, and the Companies received approval for and implemented 80% of the aggregate requested increases. The Companies made rate increase requests in 2005 for those states that had denied the 2003 rate increase requests, in part or in whole. The 2005 rate increase requests averaged 23%, and the Companies received approval for and implemented 66% of the aggregate requested increases. The 2006 rate increase requests averaged 37%, and included the portions of the increases requested in 2005 that were not previously approved. The Companies received approval for 54% of the aggregate requested increases, and continued to pursue the remainder through successive state filings. PTNA and ANIC sought higher rate increases on certain OldCo coverages where the claims experience was worse in comparison to policies without those benefits. These included policy riders that offered inflation protection and policies with an unlimited, ie., lifetime, benefit period. Policies with inflation and lifetime benefits have been subsidized by the other policies.
On each occasion that PTNA and ANIC sought premium rate increases, they offered their policyholders the option of reducing benefits in lieu of a premium rate increase. In September 2008, PTNA and ANIC began offering policyholders a non-forfeiture option whereby policyholders could forfeit their policy, cease paying premiums, and receive a benefit equal to the amount of past premium paid less the amount of any claims paid. By offering the non-forfeiture option, the Companies hoped that state regulators would be “more reasonable in their approach to rate increases.” (Waite) N.T. 1/31/11 at 178.
iii. OldCo and NewCo Business
The Corrective Action Plan called for the development of new and more profitable insurance products. After they resumed writing new business in 2002, PTNA and ANIC began to monitor and manage their business in two main segments: business written prior to January 1, 2002, “OldCo,” and business written after January 1, 2002, “NewCo.” The coverage provided under a particular OldCo or NewCo policy varies according to the consumer’s choice. Certain OldCo riders, such as inflation riders, are not available in NewCo, and NewCo policies are tax qualified. However, OldCo and NewCo products offer the same three basic types of coverages: a nursing home or assisted living facility policy; a home health care policy; and a comprehensive policy, which covers all types of treatment options. Old-Co accounts for approximately 85% of PTNA’s policies and 67% to 70% of ANIC’s policies. (Waite) N.T. 1/31/11 at 133. NewCo accounts for approximately 15% of PTNA’s policies and 30% to 33% of ANIC’s policies. Id. As of December 31, 2008, annualized premium from the New-Co business represented approximately 19% of the Companies’ in-force business, with over 24,000 long-term care insurance policies producing approximately $53 million of annualized premium. Ex. P-335 at 12.
*386A significant portion of the NewCo business has targeted a specific segment of the population — those individuals suffering from chronic non-cognitive health conditions that cannot obtain coverage from any other insurer. The Companies decided to target this market niche because competition for this business was low and because they had the claims and underwriting expertise to underwrite and price the business correctly. This non-standard business proved to be successful. As of April 2009, NewCo policies had a profitable underwriting loss ratio and did not need rate increases. NewCo products were priced according to four underwriting classes, A, B, C, or D, with the best risks in the A class and the worst risks in the D class. The Rehabilitator’s actuary, Edward P. Mohoric, testified that the Companies did a “fairly good job” of classifying applicants in the appropriate A, B, C, and D risk classes for NewCo business. N.T. 2/16/11 at 38.
The average annual OldCo premium in 2011 was approximately $2,200. (Waite) N.T. 1/31/11 at 176. NewCo policyholders are charged significantly higher premium rates than their counterparts in the OldCo book of business. Lapse rates in the New-Co book have performed as expected.
The Companies’ returns on the OldCo business have been much lower than expected. This is attributable to, inter alia, the changes wrought by the creation and proliferation of assisted living facilities in the United States; the increase in policyholder persistency; and the extension of claim duration. Lapse rates in the OldCo book have been “lower than originally anticipated.” (Mohoric) N.T. 2/16/11 at 46. This was likely a result of the very favorable premium pricing in the OldCo business, and the inability of OldCo policyholders to purchase comparable coverage at the same price elsewhere, even taking premium rate increases into account. Due to the foregoing factors, the OldCo business has required substantial increases in reserves and premium rate increases.
iv. Reserves
The Companies’ long-term care insurance policies are long duration policies, and this governs the accounting principles used to evaluate the Companies’ financial condition. Liabilities for future policy benefits are determined by actuaries and accountants in accordance with statutory accounting principles.
There are two components to the Companies’ liabilities. The first liability component combines the unearned premium reserve and the active life reserve. The unearned premium reserve covers the amount of premium that would have to be returned to policyholders who cancel their coverage mid-policy year before their premium is earned. The active life reserve is the amount needed to pay the claims of the active lives, ie., healthy policyholders, that have not yet materialized but will at some point in the future. The second liability component is the claim reserve established for incurred claims of the disabled policyholders, ie., policyholders on claim, and it covers both claims reported and those not yet reported.
In 2001, the Companies, with the approval of the Department, increased their active life reserves by $125 million. In 2002, they increased their claim reserves by $83 million. In 2005, the Companies did a review of claimant mortality, which evaluated the probability of death from year to year of policyholders currently on claim. As a result of the review, the Companies increased their claim reserves by $40 million. In 2007, the Companies further increased their claim reserves by approximately $25 million and again in 2008 *387by $17 million. These adjustments were due to increasing claim development.
v. Reinsurance
Until January 1, 2009, the Companies’ long-term care insurance policies were primarily reinsured by Imagine, an off-shore non-admitted insurance company. In July 2005, Imagine and the Companies entered into a 100% quota share reinsurance treaty covering substantially all of the OldCo business (OldCo Treaty). Effective October 1, 2005, Imagine and the Companies entered into a 75% quota share reinsurance treaty covering NewCo business written after October 1, 2005 (NewCo Treaty). Business written between January 1, 2002, and October 1, 2005, was initially reinsured but was recaptured by the Companies in 2005.
In August 2008, a dispute arose between the Companies and Imagine relating to Imagine’s collateralization of its claim payment obligations under the OldCo Treaty. Because state regulators had failed to approve OldCo premium rate increases requested and justified by the Companies, Imagine asserted that it did not have to collateralize its reinsurance obligations to the Companies. Under Pennsylvania law and applicable statutory accounting principles, the Companies may not reduce their reported claim liabilities by the amounts recoverable from a non-admitted reinsurer unless that reinsurer has collateralized its claim payment obligations to the Companies.
In October 2008, the Companies notified Imagine of their intent to recapture the reinsurance treaty on January 1, 2009. In November 2008, the Companies entered into a settlement agreement with Imagine, by which the Companies were relieved of the obligation to pay Imagine expense and risk charges beyond the first quarter of 2008 for the OldCo Treaty or beyond the second quarter of 2008 for the NewCo Treaty. It was further agreed that the Companies would recapture the policy liabilities under both the OldCo Treaty and the NewCo Treaty effective January 1, 2009, and that the Companies would release approximately $118 million of letters of credit held to secure Imagine’s obligations. The Companies’ December 31, 2008, financial statements reflect substantially all of the recapture’s effects on the Companies’ financial condition.
vi. Financial Condition of the Companies
The Companies have more than $1 billion dollars in assets. As of September 30, 2011, the market value of PTNA’s assets was $889,368,000. Exs. R-1132, 1136.16 As of September 30, 2011, the market value of ANIC’s assets, not including those of its wholly owned subsidiary AINIC, was $144,682,000. Exs. R-1132,1136. Comparable numbers as of December 31, 2008, for PTNA and ANIC were $889,138,876 and $102,849,736, respectively. Ex. R-894 at 26-27. Thus, since rehabilitation, the market value of the Companies’ assets has grown. The Companies have no debt. The Companies are paying all obligations in a timely manner, including policyholder claims; to date, neither company has had to liquidate assets in order to pay claims. (Robinson) N.T. 2/15/11 at 200.
As of December 31, 2009, the Companies’ combined annualized premium reve*388nue was $246.8 million. PTNA’s annualized premium revenue was $228.5 million, and ANIC’s was $23.3 million. Ex. P-961 at 1. In 2009, the Companies paid $214.5 million in claims. The following table shows the combined claims payments by the Companies over the past several years and the percentage increase from the previous year:
[[Image here]]
Ex. R-62. As the chart demonstrates, the average annual increase in claims payments has been less than 4%, without any adjustment for inflation.
Mohoric acknowledged in Milliman’s rebuttal report that the “ultimate solvency position of the companies will be determined by cash flows.” Ex. P-969 at 5. His testimony confirmed that position. N.T. 11/2/11 at 6. Commissioner Ario also acknowledged that the issue of cash flow is relevant to an analysis of whether the Companies should be liquidated. N.T. 2/11/11 at 254-55. Karl Volkmar, the actuarial expert retained by the Intervenors, compared actual cash flows for PTNA and ANIC with the projected cash flows implicit in Milliman’s July 2010 Surplus Report using data he received from Milliman. Milliman projected a negative cash flow of $37.76 million for PTNA in calendar year 2010, but the actual negative cash flow was 50% less, ie., $18.7 million. Ex. R-912. Milliman projected a positive cash flow of $5.2 million for ANIC in 2010, but the actual positive cash flow was $8.1 million. Ex. R-912. As of December 31, 2009, PTNA’s additional paid in capital was $114,049,968 and ANIC’s was $4,861,812. Exs. P-729, P-730 at 6.
vii. Conditions that Led to Rehabilitation
The Rehabilitator continues to employ many who served in senior management prior to the Companies’ receivership. The Companies are operationally sound. The rehabilitations of PTNA and ANIC were not caused by cash flow issues but by a need to strengthen reserves for projected claims that are many years distant. The Companies are saddled with inadequate rates on a subset of policies in a subset of problematic states. This situation arose because regulators in key states such as Arizona, California, Florida, Illinois, and Pennsylvania have denied, delayed, or limited needed premium rate increases for the OldCo policies. Ex. P-335 at 45. Meanwhile, regulators in other states, such as South Dakota and Virginia, instituted the rate increases sought for the same policies in a straightforward manner.
*389The OldCo policies are underpriced based on the current market for long-term care policies, but they were reasonably priced when issued based upon available information. Initial pricing used assumptions generally consistent with industry assumptions at the time. A significant factor that led to the OldCo policies being under-priced, the unanticipated expansion in availability of long-term care facilities and attendant increased utilization of claims, is endemic to the entire long-term care industry. In addition, as a consequence of being underpriced, lapse rates in the Old-Co book have been lower than anticipated.
One OldCo non-tax qualified product, which was sold between 1996 and 2001, provides policyholders with inflation or lifetime benefits, and is referred to as the “Cadillac” policy. This particular product has contributed significantly to the need for rehabilitation.
A secondary reason for the Companies’ financial condition was excess operational capacity. This condition was addressed by reducing overhead and expenses.
viii. The Rehabilitation Orders
The Court’s January 6, 2009, rehabilitation orders “directed” the Commissioner “to rehabilitate the business of Penn Treaty; to take possession of the assets of Penn Treaty; and to administer the Penn Treaty assets in accordance with the orders of this Court.” Ario v. Penn Treaty Network America Insurance Co. (No. 5. M.D. 2009, order filed January 6, 2009) at ¶ B.17 The Rehabilitator was ordered to “take such actions as are necessary to correct the condition that prompted the Board of Directors’ request for and consent to the rehabilitation of Penn Treaty.” Id. at ¶ 5. The Rehabilitator was ordered to prepare a “preliminary plan of rehabilitation,” id. at ¶ 15, and to “prepare a plan of rehabilitation, which may include a consolidation, merger or other transformation of Penn Treaty and to that end may retain accountants, actuaries, attorneys and other consultants at the expense of Penn Treaty.” Id. at ¶7. As this Court has explained, “[o]nce the course of rehabilitation has been chosen by the Insurance Commissioner, it must be pursued unless and until the Commissioner satisfies this Court that the rehabilitation should be terminated under Section 518(a) of Article V and a liquidation order entered.” Koken v. Legion Insurance Co., 831 A.2d 1196, 1229 (Pa.Cmwlth.2003), aff'd sub nom. Koken v. Villanova Insurance Co., 583 Pa. 400, 878 A.2d 51 (2005).18
*390ix. The April 2009 Preliminary Report and Plan of Rehabilitation
On April 2, 2009, the Rehabilitator filed his Preliminary Plan with the Court. The Preliminary Plan described the background and history of the Companies, their products, the differentiation between New-Co and OldCo policies, reserve situation and financial information, sales, claims management, and rate improvement efforts. The Preliminary Plan set forth the areas in need of corrective action. The Preliminary Plan appended a report prepared by Thomas Johnson, an investment banker at Signal Hill, and a report prepared by Mohoric at Milliman. The Signal Hill report analyzed and recommended various rehabilitation approaches, and the Milliman report discussed surplus projections related to the rehabilitation of PTNA and ANIC. The Preliminary Plan concluded, inter alia, that a liquidation would be harmful to policyholders. Ex. P-335 at 47. It also concluded that general creditors would be no worse off in a rehabilitation because they are expected to receive nothing in liquidation. Id. at 47-48.
The Preliminary Plan lauded the Companies for successfully developing since 2002 a “market niche” in underwriting policies for people with chronic non-cognitive health conditions. These NewCo policies were deemed profitable and needed no rate increases. According to the Preliminary Plan, the “primary cause for the Companies’ financial difficulties is the historical and current inadequacy of the insurance rates being charged for the Companies’ OldCo book of business.” Ex. P-335 at 45. Further, the Preliminary Plan stated:
[T]he only feasible rehabilitation alternative at this time is to increase rates on the OldCo business and reduce expenses (the “Rate Increase Alternative”) [and] the Rate Increase Alternative may be in the best interest of the Companies’ policyholders because, if successful, it will allow PTNA to pay more policyholder claims in full, rather than have benefits limited by state guaranty association maximum limits, and may allow the Companies to divest all or portions of their business in the future.
Id. at 47.
The Preliminary Plan identified the failure by some regulators to approve adequate rate increases on the OldCo policies as the root cause of the need for rehabilitation, noting that only 66% and 54% of the Companies’ aggregate requested increases were approved in 2005 and 2006, respectively. Id. at 21. The Preliminary Plan explained that the rate increases would “vary by state based upon each state’s prior rate increase experience.” Id. at 49. With rate increases, the Rehabilitator would offer policyholders the alternative options of (1) maintaining current rates and accepting reduced benefit levels; or (2) selecting a non-forfeiture option. The Preliminary Plan stated that “aggressively” pursuing premium rate increases for the OldCo policies would correct the condition that caused the Companies to be placed in rehabilitation. This conclusion was supported by the Companies’ actuarial firm, Milliman, which stated that “PTNA can in time return to a positive surplus, and that ANIC can maintain and improve its surplus position by taking [the actions listed in the Plan].” (Pfannerstill) N.T. 3/23/11 at 60; Ex. P-335 at 48.
In short, it was identified early in the process that a rehabilitation required the aggressive pursuit of actuarially justified premium rate increases. (Mohoric) N.T. 2/23/11 at 133; (Pfannerstill) N.T. 3/23/11 at 56-57. Each additional dollar generated by actuarially justified premium rate increases will benefit the estate as a whole because it will produce more funds to pay *391claims. (Waite) N.T. 2/1/11 at 46-47. The Preliminary Plan concluded that “[t]he more quickly and aggressively rate increases and expense reductions are implemented, the sooner PTNA’s surplus can become positive and ANIC’s financial position can be improved.” Ex. P-335 at 50; (Pfannerstill) N.T. 3/23/11 at 60-61.
x. Rehabilitation Implementation Committee
Robert L. Robinson was retained by the Rehabilitator to serve as Chief Rehabilitation Officer of the Companies. Robinson was selected on the basis of his extensive insurance experience. This included work on the rehabilitation of Fidelity Mutual Life Insurance Company and work with Summit National Life Insurance Company, Colonial Penn Insurance Company, Leucadia National Corporation and Penn Mutual Life Insurance Company. Robinson educated himself on the long-term care insurance industry and examined the Companies’ functions, management and organization.
Robinson was tasked with organizing the operational aspects of the Companies’ rehabilitation. Robinson created a Rehabilitation Implementation Committee (RIC) to oversee the rehabilitation efforts. Robinson ran the RIC meetings, which were held every Friday beginning in January 2009. Joseph DiMemmo and Preston Buckman, Special Funds Counsel, represented the Department as members of the RIC. Officers and directors of the Companies served on the RIC, including William Hunt, Mark Cloutier, Cameron Waite, Stephen LaPierre and others. Robinson testified that it was important for the RIC to include members of senior management of the Companies because of their familiarity with the business. N.T. 2/3/11 at 230. Two of PTNA’s long-time actuaries at Mil-liman, Edward Mohoric and Larry Pfan-nerstill, were members of the RIC. Other members of the RIC included Thomas Johnson, an investment banker at Signal Hill, and attorneys from the law firm of Cozen O’Connor, the Rehabilitator’s outside counsel.
Robinson assigned the RIC members to subcommittees with the intention that they would collectively frame a rehabilitation plan. Robinson also conducted frequent meetings with members of management, upon whom he relied for their expertise of long-term care insurance issues; he relied on the consultants to analyze and explain the circumstances confronting the Companies.
xi. Milliman August 2009 PowerPoint Projections
On August 18, 2009, Milliman made a PowerPoint presentation entitled “Penn Treaty — Projections and Assumptions Review.” Ex. P-530. The August 2009 presentation contained the following summary comparison of morbidity results:
[[Image here]]
Ex. P-530 at 4. The dramatic increase in the present value of future claims from April 2009 to August 2009 was attributed to Milliman’s (1) discontinuance of its well*392ness assumption (2) reduction in its morbidity improvement assumption and (3) substantial increases to its claim costs assumption. Ex. P-530 at 9.
Commissioner Ario testified that he attended a “pre-meeting” led by Mohoric before the RIC’s September 14, 2009, meeting to discuss the problematic numbers in the August 18, 2009, presentation. N.T. 2/11/11 at 44-45. Ario recalled that “coming out of that we said, okay, well if these numbers don’t change, and we need to look at, you know, what we’re going to do, and I think everybody was thinking, you know, probably towards liquidation. And then we had the September 14th [meeting].Id. at 45.
The August 18, 2009, presentation contained the following caveats and limitations:
• This information is presented in DRAFT format for discussion purposes at this time and is subject to change. We will provide documentation once assumptions are finalized.
• This information is intended to assist the discussion relating to the morbidity assumptions and the resulting financial projections. This information may not be appropriate, and should not be used, for other purposes.
• Penn Treaty’s actual results will likely differ from these estimates. Penn Treaty should monitor its emerging results and take corrective action when necessary.
Ex. P-530 at 2; see also (Volkmar) N.T. 10/26/11 at 100-02.
xii. Termination of New Premium Rate Increase Filings
Premium rate increase filings terminated on August 1, 2009, for PTNA and a month later for ANIC. This decision was made by DiMemmo without the knowledge or approval of the Court or Commissioner Ario, and contrary to Milliman’s recommendation at the time that premium rate increase filings continue. The decision to stop seeking rate increases by way of new filings was made out of a sense of “fairness” or “perceived fairness” to policyholders. N.T. 3/23/11 at 63 (Pfannerstill testified that DiMemmo believed that “it was not appropriate to put the burden of the rehabilitation on the policyholders.”); N.T. 2/14/11 at 220 (Robinson testified that what drove the decision was the perceived unfairness to the policyholders). DiMem-mo testified that he decided to terminate the new premium rate increase filings based on “actuarial noise.” N.T. 2/3/11 at 120-21.19
There is no dispute that the decision to forego new premium rate increase filings cost the Companies hundreds of millions of dollars. (DiMemmo) N.T. 2/3/11 at 174; (Volkmar) N.T. 10/27/11 at 42. It increased the reserve deficiency.
The Rehabilitator elected not to implement some rate increases that had already been approved. For example, Waite testified that “the [R]ehabilitator determined not to implement those particular rate approvals [in Alaska and Nevada].” N.T. 1/31/11 at 224-25; N.T. 2/1/11 at 114-16. Waite explained that “given that those two states had $100,000 guaranty association limits, that in the event [PTNA] and/or [ANIC] ... went to liquidation, we believed it was unfair [to] those policyholders to pay additional premium. That was our rationale.” N.T. 2/1/11 at 115-16.
The Rehabilitator took steps to impede pursuit of actuarially justified premium rate increases. In March of 2010 Commis*393sioner Ario sent letters to the insurance commissioners of four states in which rate increase applications were pending. The letter advised the commissioners that liquidation petitions had been filed and that his Department would not grant any rate increases for PTNA or ANIC in Pennsylvania. Ex. R-58. DiMemmo testified that the purpose of Commissioner Ario’s letters was to tell his fellow commissioners that they did not have to approve the pending rate increases. N.T. 2/3/11 at 121-25. None of the rate increase requests were granted in those states.
The Rehabilitator also relieved Cameron Waite of his principal duty, which was to pursue premium rate increases on behalf of PTNA and ANIC. Waite’s personal visits to state regulators stopped shortly after the rehabilitation began, ostensibly to reduce costs related to Waite’s travel budget.
Finally, the Rehabilitator elected not to appeal the decisions of state regulators to disapprove actuarially justified premium rate increase filings. Waite testified that Florida, one of the more reluctant states to approve rate increases, was “dead wrong” in its disapprovals. N.T. 2/1/11 at 195. Accordingly, the Companies appealed; the Rehabilitator later abandoned the Florida appeal.
B. The Companies’ Statutory Surplus as of December 31, 2009
The centerpiece of the Rehabilitator’s case in support of his liquidation petitions is the statutory surplus report prepared by Milliman in July 2010. Milliman’s July 2010 Surplus Report showed a decline in each company’s surplus as regards policyholders as of December 31, 2009, that far exceeded the impact expected by the loss of the Imagine reinsurance. The surplus decline was attributed to Milliman’s increase to the Companies’ reserves and reduction in anticipated future premium volume.
Statutory accounting principles govern statutory financial statements. Assets held by an insurance company must be “admitted” before they can be placed on the balance sheet and counted toward the statutory surplus of an insurer. Statutory accounting does not assign value to the intangible value of a business, such as goodwill, as generally accepted accounting principles do. The object of a statutory financial statement is to provide regulators with a snapshot of an insurer’s financial strength at a point in time, which is the last day of the calendar year. It is a liquidation statement in that it tells the reader whether an insurer will be able to pay all claims should it shut down on December 31 and never renew or write another policy.20 The surplus as regards policyholders answers that question. It is synonymous with net worth.
Milliman’s July 2010 Surplus Report projected revenues and earnings over the next 60 years; it also projected the Companies’ liabilities over that same period. It restated those 60-year projections at their present value in order to set a surplus as of December 31, 2009, for PTNA and ANIC. The statutory reserves determine surplus.
As noted, the claim reserve represents the amount needed to pay incurred claims until they end, which can be many years. The active life reserve represents a more ephemeral liability, i.e., what is estimated to be paid in the future on claims that may or may not develop. To estimate the reserve for future, but yet to be developed *394claims, the actuaries must conduct an inquiry that is broad in scope. They must estimate the number of policyholders, their mortality, their tendency to morbidity, inflation, deflation, investment earnings, premium volume and future rate increases over a 60-year period. It is a complicated process, and the results are uncertain because the factors that go into the projections keep changing. It is a high wire juggling act to predict what will happen during the period from December 81, 2009, to December 31, 2069, during which time a cure for cancer may be found or an asteroid may hit North America.
Milliman’s July 2010 Surplus Report was rebutted by the report of Intervenors’ expert, Karl Volkmar of United Health Actuarial Services, Inc. The experts reached different conclusions about the surplus of PTNA and ANIC as of December 31, 2009. However, they agree that the Companies’ assets are sufficient to fund all incurred claims, whether or not reported. They also agree that without change to the existing rate structure of the OldCo business, the Companies will not be able to generate the revenue needed to fund the claims that are expected to develop in future years from the presently healthy policyholders.
C. Scope of Expert Evidence
i. Rehabilitator’s Experts
Edward P. Mohoric is a principal with the actuarial consulting firm of Milliman. He joined the firm in 1982 and has served on its board of directors. He is a Fellow of the Society of Actuaries and a Member of the American Academy of Actuaries. He is a consulting life actuary, but approximately 50% of his consulting is in the area of long-term care insurance. He has published six articles relating to life insurance, long-term care riders, and mortality considerations and is active in the Society of Actuaries. The parties stipulated to Mo-horic’s qualifications to testify as an actuarial expert about the matters within the scope of his report.
Larry J. Pfannerstill is a principal and consulting actuary with Milliman, which he joined in 1995. He is a Fellow of the Society of Actuaries and a Member of the American Academy of Actuaries. He is a consulting health actuary who devotes 65% to 70% of his practice to long-term care insurance. He has published seven articles about long-term care insurance and is active in the Society of Actuaries. The parties stipulated to Pfannerstill’s qualifications to testify as an actuarial expert about matters within the scope of his report.
Arthur M. Lucker is employed by INS Consultants, Inc., a regulatory consulting firm, which he joined in 1997. He is a Fellow of the Society of Actuaries and a Member of the American Academy of Actuaries. Lucker has worked for 38 years as an actuary in the insurance industry. For the past seven years, he has focused on the long-term care industry. He does consulting work for state insurance departments on long-term care insurance regulatory matters, including rate filings. Inter-venors objected to Lucker testifying about national rate regulatory matters because his knowledge is limited to a few states and, further, he could not testify about what will happen in the future. The Court overruled Intervenors’ objections, noting that they went to the weight of Lucker’s testimony as opposed to his qualifications as an expert.
David Minches is an Executive Director in the Insurance and Actuarial Services practice of Ernst & Young, LLP. Minches is an Associate of the Society of Actuaries and a Member of the American Academy of Actuaries. His consulting for long-term care insurance carriers concerns audits and reserve work, and this work consti*395tutes approximately 10% of his practice. Minches is a life actuary. Under the terms of an engagement letter dated December 21, 2009, Ernst & Young agreed to review Milliman’s proposed revisions to the methods it used to calculate the statutory claim reserves of PTNA and ANIC; Ernst & Young’s review was expressly “limited in scope and time.” Ex. P-56 at 11. The report was the work of Minches, Bob Hanes and Darrel Knapp.
Vincent Bodnar, an actuary and principal with DaVinci Consulting, LLC, was retained by NOLHGA in 2009. Bodnar authored a report for NOLHGA in June 2010 (NOLHGA Report) that projected the financial conditions of the Companies post-liquidation and advised NOLHGA of the liability of its members in a liquidation scenario. Bodnar was later retained by the Rehabilitator in this matter. He authored a second report dated November 9, 2010 (Rebuttal Report), which purported to rebut the report prepared by Interve-nors’ actuarial expert, Karl Volkmar. The Rehabilitator proposed to offer Bodnar’s two reports and his testimony as part of its rebuttal case.
Intervenors filed a motion in limine on November 12, 2010, to exclude Bodnar’s testimony and his two reports. The Court granted the motion. In doing so, the Court agreed with Intervenors that Bod-nar’s NOLHGA Report, which detailed the Companies’ financial picture post-liquidation, was not relevant to their status in rehabilitation or to whether the Rehabili-tator satisfied the legal standard for converting a rehabilitation to a liquidation. Even if the NOLHGA Report were relevant, it was cumulative of Milliman’s assumptions and projections.
The so-called Bodnar Rebuttal Report was excluded for several reasons. First, it was produced by the Rehabilitator on November 9, 2010, just three weeks before the hearing was scheduled to begin and two weeks after Bodnar testified by deposition that he did not intend to present a written report on Volkmar’s work. The Court agreed with Intervenors that their receipt of this report, which raised many new issues, so close to trial was prejudicial. What the Rehabilitator couched as “rebuttal evidence” to be introduced through Bodnar was in actuality evidence that should have been introduced as part of the Rehabilitator’s case-in-chief. Second, there was no prejudice to the Reha-bilitator because the Rehabilitator had other witnesses, ie., Lucker, Pfannerstill and Mohoric, who were prepared to rebut key aspects of Volkmar’s report, including, for example, the likelihood of obtaining necessary rate increases. Finally, Bod-nar’s Rebuttal Report did not actually rebut Volkmar’s findings and conclusions. It responded to issues raised by Volkmar concerning the NOLHGA Report. Because the NOLHGA Report was excluded for the reasons set forth above, Bodnar’s responses to Volkmar’s concerns about that report became irrelevant.
ii. Intervenors’ Experts
Stephen K. Holland, M.D. is the chief medical officer of Univita, Inc., a long-term care insurance administrator. He serves as an advisor to the California Legislative Advisory Committee on Access to Insurance for Disabled Employees, as a peer reviewer for the Annals of Internal Medicine and as a member of the Health Enhancement Program Advisory Committee in Seattle, Washington. Holland was certified by the American Board of Internal Medicine in 1984. He has over 24 years of experience in long-term care insurance. Holland has published approximately 20 articles and a manual relating to long-term care underwriting, and has lectured at numerous conferences on long-term care insurance risk management. *396He is a member of the International Society to Advance Alzheimer’s Research and Treatment. He was offered as an expert in the field of medicine and, specifically, medical aspects of long-term care insurance claims. This included the following: morbidity compression; the major conditions and diseases that give rise to long-term care insurance claims; past, current and expected medical research and medical advances relating to those conditions; and the impact of these advances on long-term care insurance claims, including those of PTNA and ANIC. The Court admitted Holland as an expert to testify on the matters as listed above.
Karl G. Volkmar is a principal of United Health Actuarial Services, Inc., which he joined in 2003. He is a Fellow of the Society of Actuaries, a Member of the American Academy of Actuaries, and a Fellow of the Conference of Consulting Actuaries. He serves on the American Academy of Actuaries Individual and Small Group Market Task Force and is vice chair and health section representative to the Society of Actuaries. He is a consulting health actuary, and devotes 35% to 40% of his time to long-term care insurance engagements. He has published several articles on health insurance matters and is active in the Society of Actuaries. The parties stipulated to Volkmar’s qualifications to testify as an actuarial expert about matters within the scope of his report.
iii. Actuarial Projections of the Companies’ Financial Condition
Intervenors and their actuary believe that future premium revenue, enhanced by rate increases, together -with earnings on assets and cash flow can cover the future claims that will develop over the next 60 years. The Rehabilitator and his actuaries believe otherwise. Notably, however, the Rehabilitator’s actuaries cited political, not actuarial reasons, for their pessimism. Each actuarial firm prepared projections in support of their active life reserves, which was central to their respective surplus statements for PTNA and ANIC.
To develop their projections, the actuaries used, principally, seven categories of actuarial assumptions. They are (1) Morbidity or Claim Costs; (2) Morbidity Improvement; (3) Mortality; (4) Voluntary Lapsation; (5) Investment or New Money Rate; (6) Claim Adjudication and Wellness; and (7) Premium Rate Increases. Each assumption relates to a particular aspect of the long-term care insurance business of PTNA and ANIC and predicts the development of that business aspect over the course of the next 60 years. Because the Companies are not writing new policies, their business is described as a closed block of business. It consists of approximately 142,000 policyholders, of which approximately 8,000 are on claim. The assumptions try to predict the amount of premium that will be paid by the existing population of policyholders; the extent and timing by which that population of policyholders will decline in number by reason of policyholder death or policy lapse; the investment income that will be earned from assets and from premium revenue that will be collected from those policyholders over the years to come; and the number and amount of claims that will be presented by those policyholders.
The actuaries for both sides agree that neither PTNA nor ANIC satisfies the minimum statutory surplus requirements needed to be in good standing and able to write new business. The actuaries disagree on the amount of that surplus deficiency.
The Rehabilitator’s actuary opined that as of December 31, 2009, PTNA’s statute-*397ry surplus was negative $2.1 billion and ANIC’s statutory surplus was negative $137.0 million. The breakdown of Milli-man’s surplus projections is shown below:
[[Image here]]
Ex. P-961 at 16 (as modified). The gross premium reserve has nothing to do with a return of unearned premium; it is a reserve that tests statutory reserves. It increased the Companies’ apparent deficit.
Intervenors’ actuary reached different surplus conclusions. He opined that as of December 31, 2009, PTNA’s statutory surplus was negative $333 million and ANIC’s statutory surplus was positive $600,000.21
The Rehabilitator’s expert opined that as of December 31, 2009, the present value of the Companies’ future claims to be paid over the next 60 years is $4.1 billion, assuming an earnings rate of 5.74% over that period. Intervenors’ actuary opined that as of the same date, the present value of future claims is $2.7 or $2.9 billion, depending on whether an earnings rate of 6.4% or 5.74% is assumed.
The Companies’ surplus deficiency predicts a future inability to pay future claims but not a present inability to pay incurred claims. As noted, the Companies’ cash flow, approximately $1 billion every 3.4 years, has been sufficient to fund the incurred claims. The challenge is whether the Companies can meet future claim obligations by increasing the premium rates faster than new claims develop from now until the last policy in the closed book of business terminates 60 years from December 31, 2009. The Rehabilitator believes they cannot, and Intervenors believe they can. On this difference the actuarial evidence was extensive.
iv. Actuarial Principles and Concerns
Long-term projections, such as the 60-year projections developed for the Companies, are highly sensitive to the assumptions, which means that a small change in an assumption will have a big impact on the projections. This is because some assumptions are compounded over time. The longer the projections, the greater the *398impact of one assumption change upon those projections.
Sensitivity testing measures the extent to which an assumption change will affect the projections. Milliman’s sensitivity testing was done by changing one assumption at a time. Intervenors’ expert, Volk-mar, testified that changing more than one assumption at a time is more realistic because it measures the aggregate and compounding effect of changes. In any case, Milliman’s sensitivity testing showed that changing even one assumption has a significant impact. For example, increasing the Milliman morbidity improvement assumption by 1% per year produces a $500 million dollar improvement in the Companies’ claim projections.
The actuaries agreed that credible data are needed to develop reliable assumptions and valid projections. Because the long-term care insurance industry is fairly new, credible data are lacking. The Actuarial Standards Board has adopted Actuarial Standard of Practice (ASOP) No. 25, which applies to long-term care insurance. ASOP 25 defines credibility as “[a] measure of the predictive value in a given application that the actuary attaches to a particular body of data.”22 Ex. R-896 at 1, ¶ 2.1. ASOP 25 requires an actuary to select credibility procedures that produce reasonable results, do not bias the results, are practical and balance responsiveness and stability. Giving “consideration to the need to balance responsiveness and stability” means that an actuary’s methodology should incorporate experience as it develops, i.e., be responsive, but balance responsiveness against the need for stability to avoid wide swings in the use of that data that would otherwise be produced by immediate incorporation of new experience. To achieve stability, the actuary incorporates new data in a way that smooths the grading of historical data and the new data; responsiveness is the degree to which an actuary responds to or reflects the new experience as it emerges. (Volk-mar) N.T. 10/25/11 at 152.
Actuaries can assign zero to 100% credibility to a block of data. An assumption based on data with low or partial credibility produces less certain results. The degree of credibility to assign data is a matter of actuarial judgment. Where company data are not fully credible, actuaries augment company data with industry data to improve the credibility of the data used to make projections.
Attachment 5 to Volkmar’s Report consists of a May 28, 2008, “Summary of Discussions” of a Credibility Subgroup of the Long-Term Care Reserving Work Group of the American Academy of Actuaries. The Rehabilitator’s actuary, Mohoric, was a member of the Credibility Subgroup. In its May 2008 summary, the Credibility Subgroup reported that the “rule of thumb” is to require 1,082 claims in a cell for full credibility. Ex. R-911 at 10. The Rehabilitator notes that this “rule of thumb” is not a formal standard that has been adopted by the subgroup or the Academy. Mohoric testified that he treated 13 to 15 claims in a claim cell as credible. N.T. 2/24/11 at 143-45.
The Milliman actuaries used “best estimate assumptions” for their projections. A “best estimate assumption” is one, in the actuary’s judgment, that falls in the 50th percentile because there is an equal probability that actual experience will fall above or below the experience predicted by the assumption. There is no single “best estimate;” different actuaries may each choose different “best estimates.” Not all of Mil-*399liman’s assumptions were “best estimate” or “50/50” assumptions; for example, its rate increase assumption was based upon discussions between Milliman and members of the Companies’ management assigned to rate regulation. (Pfannerstill) N.T. 3/24/11 at 18-19.
Volkmar used what he termed “reasonable assumptions,” not “50/50” assumptions. He testified that “best estimates” were not required for these projections where the goal is to develop a full continuum of results. N.T. 10/25/11 at 176. Volkmar’s goal was to evaluate the Companies’ ability to meet obligations as they come due, as opposed to meeting statutory surplus standards. Volkmar clarified that a “best estimate” describes an actuary’s attempt to choose the middle ground. Id. at 177. Mohoric confirmed that Milliman’s “best estimates” did not represent scientific analysis. N.T. 2/22/11 at 34; N.T. 2/24/H at 15-16. Volkmar testified that the assumptions in his report, which he detailed in his testimony, were “reasonable” given his experience in long-term care insurance actuarial consulting and consistent with the work he has done for other clients.
v. Changes in Milliman’s Projections
Milliman has served as the Companies’ actuary since 2001. In 2009, its assumptions and corresponding projections for the Companies changed dramatically. Milli-man increased the present value of future claims for the Companies from $2.6 billion in April 2009 to $4.1 billion in July 2010. This had a corresponding negative impact on Milliman’s surplus calculations.
Volkmar prepared a Comparison Chart, Ex. R-1057, to show the key projection results, which include (1) the present value of claims to be projected in the next 60 years; (2) the year surplus projections turn positive and the risk-based capital (RBC) ratio23 exceeds 200% for PTNA and ANIC; and (3) the starting statutory surplus for PTNA and ANIC. The Comparison Chart also shows the changes that Milliman made to the seven categories of assumptions between September 2008 and July 2010. The Rehabilitator did not challenge the accuracy of the information in the Comparison Chart.
The year before the Companies’ receivership, Milliman authored two formal projection reports: the April 28, 2008, “Litow Report” authored by senior Milliman actuary, Mark Litow, and a September 20, 2008, “Appraisal Report.” In 2009, Milli-man prepared two more formal projection reports: the report done to support the Rehabilitator’s April 2, 2009, Preliminary Plan and a September 14, 2009, report done to support the Rehabilitator’s liquidation petition. On October 15, 2009, Mil-liman issued an amended report to correct a $200 million error in the September 2009 report. On July 7, 2010, Milliman prepared a surplus projection report as evidence for the Rehabilitator’s use at trial. The 2009 and 2010 reports were prepared under the direction of Mohoric.
The 2008 Appraisal Report projected the present value of projected future claims for PTNA and ANIC at $2.5 billion (as recalculated by Volkmar using an interest rate that was consistent with the interest rates used in the later Milliman reports). The earlier Litow Report had set the present value of projected future claims at $2.2 billion for the OldCo business. The April 2009 report projected the present value of future claims at $2.6 billion. These three Milliman surplus reports offer claim projections within a range of $2.2 to $2.6 billion. This changed. Milliman’s October 2009 Report increased projected future *400claims by more than $1 billion. “Scenario A” in the October 2009 Report set a present value for projected future claims at $3.4 billion, and “Scenario B” set that value at $3.9 billion.24 Ex. P-1021. The July 2010 Surplus Report increased the claim projections even more, setting the value at $4.1 billion. The changes in projected claims from 2008 to 2010 are illustrated in the following table:
[[Image here]]
See Ex. R-1057. The increase of $1.5 billion between April 2009 and July 2010 is a 58% change. The changes between April and October of 2009 are illustrated in the following table:
[[Image here]]
See Ex. R-1057.
The increases in projected claims impacted each company’s projected surplus as illustrated in the following table:
[[Image here]]
See Ex. R-1057. The increase in surplus deficiency adversely affected the Compa*401nies’ ability to return to an RBC ratio of 200%, ie., the level of surplus needed to be solvent and able to be discharged from receivership, as shown in the following table:
[[Image here]]
See Ex. R-1057.
The actuaries for both the Rehabilitator and Intervenors agreed that Milliman’s volte-face on its projected future claims was without precedent in their experience. Between April and October 2009, Milliman increased projected claims by 51%. Volk-mar testified that in over two decades of consulting for long-term care insurance companies, the largest claim projection increase he had seen was 20%, which happened when the insurance company changed its actuaries. In that case, the former and new actuaries worked for two years to reconcile the different results, and the company did not act upon the new results until a new, independent actuary did an independent evaluation using the same data.
The Rehabilitator has not explained the reason for Milliman’s sharp increase in expected future claims, except to note that underwriting for PTNA was “loose.” Re-habilitators’ Reply Brief at 118, ¶284. Underwriting is a screening process that tries to identify applicants who select “against the company, the policyholder knowing they’re going to be on claim very soon.” (Pfannerstill) N.T. 3/22/11 at 137. After a time, underwriting “wears off’ because it is no longer a factor in predicting the likelihood of a claim. Id. In this regard, the Rehabilitator did not address Intervenors’ expert testimony that the effects of underwriting wear off after 5 to 7 years.
Volkmar opined that “[g]iven the significant recent changes, from an actuarial standpoint, I don’t think it is appropriate to rely on [the Milliman projections] in this context.” N.T. 10/26/11 at 33. By “in this context,” he meant a liquidation of PTNA and ANIC. Volkmar also testified that there was no reason to act quickly because the Companies have the assets needed to meet their obligations in the near term. Milliman did not opine that a rehabilitation was futile, as an actuarial matter.
vi. Actuarial Caveats and Limitations
Milliman’s July 2010 Surplus Report was issued with several caveats and limitations. One caveat in that Report states that:
[A]ctual required reserves will only be known once sufficient time has passed such as all claim payments have been made. Actual reserves will vary from estimated values for various reasons, including random fluctuations in claims. Penn Treaty should continue to monitor emerging experience as it develops.
Ex. P-962 at 14. Another caveat states that the “historical experience in the later duration since onset of claim, [ie.,] the ‘tail’ of the continuance curves, was not credible.” Id. at 9. Projections of future payments not based on credible data are not reliable. (Mohoric) N.T. 2/22/11 at 79-80; (Volkmar) N.T. 10/25/11 at 182-83. The caveats in the July 2010 Surplus Re*402port relate not only to projected future claims but also to premium rate increases, the timing of those rate increases, lapsation, expense rates and investment income. The July 2010 Surplus Report states that “new data may or may not change conclusions drawn from the projections and analysis included herein.” Ex. P-961 at 2.
Milliman’s August 2009 PowerPoint illustrating its projections included caveats, but it did not include a specific one on credibility. The October 2009 Report acknowledged the absence of credible claim data for policyholders older than 90 years but otherwise does not present a detailed credibility caveat comparable to that in the July 2010 Surplus Report.25
Volkmar’s Report contained caveats and limitations similar to those presented by Milliman in its July 2010 Surplus Report. Rehabilitator’s Reply Brief at 54-55, ¶ 142.
vii. Credibility of Data Used in Projections
As noted, the projections produced from low credibility data contain a degree of uncertainty. A significant portion of the projected future claims are attributed to policyholders who are older than 90 years and have held their policies for many years. The July 2010 Surplus Report states that: (1) data for policyholders at attained ages 90-plus are not fully credible; (2) credibility is low for policies with a duration longer than 16 years; and (3) claim durations longer than 42 months are not fully credible. The different claim costs in Scenario A and Scenario B of Milliman’s October 2009 Report are largely attributed to differences in projected future claims by those holding policies longer than 16 years. Pfannerstill testified that “[Milliman] will not have actual experience to determine which scenario is occurring for possibly another ten years.” N.T. 3/24/11 at 80-81.
Claims to be presented by policyholders above age 90 represent 29% of the present value of future claims as projected by Mil-liman. Ex. R-903 at 1. Projected claims to be presented by policyholders with policies longer than 16 years represent 71% of the present value of future claims, as projected by Milliman. Id. at 2. To a significant extent, therefore, Milliman’s claim projections are based upon data with low credibility. As part of an internal peer review, a Milliman actuary noted the lack of credible data to support the projections and advised addressing that fact in its reports. Ex. R-350; (Mohoric) N.T. 2/22/11 at 135-37.
viii. Actuarial Standard of Practice No. 25
Actuarial Standard of Practice No. 25 (ASOP 25) provides “guidance to actuaries in the selection of a credibility procedure and the assignment of credibility values to sets of data including subject experience and related experience.” Ex. R-896 at 1. Mohoric testified that he did not apply the credibility mandates of ASOP 25 because he believed it applied only to group experience refunds or rate making. Volkmar testified that ASOP 25 is not so limited and is mandatory. N.T. 10/25/11 at 148; N.T. 10/26/11 at 21-22. Milliman did not balance responsiveness and stability when making credibility judgments. Specifically, Mohoric stated that he did not give any consideration to stability. N.T. 11/1/11 at 25.26
*403Pfannerstill defined credibility as having “enough data to form an assumption that when you apply that assumption and look at it in a retrospective basis, the assumption turns out to be true.” N.T. 3/24/11 at 58.
Volkmar disagreed with Mohoric’s testimony that data were credible if there were 13, 14, or 15 claims in a cell.27 N.T. 10/26/11 at 85. The Credibility Subgroup of the American Academy of Actuaries’ Long-Term Reserving Work Group, in which Mohoric participated, notes that 1,082 claims in a cell is a “rule of thumb” for full credibility. Ex. R-911 at 10. Volkmar did not explain how, or if, using a 1,082 claims per cell “rule of thumb” for full credibility would have changed the way the projections were calculated.
ix. Actuarial Standard of Practice No. 18
Actuarial Standard of Practice No. 18 (ASOP 18) sets forth recommended practices for actuaries involved in, inter alia, designing, pricing and evaluating liabilities for long-term care insurance contracts. ASOP 18 states as follows with respect to recommending premium rate increases:
In developing [recommended rates], the actuary should not use assumptions that are unreasonably optimistic.... In particular, the actuary should not rely on anticipated future premium rate increases to justify the selection of unreasonably optimistic assumptions when recommending premium rates. On the other hand, the actuary should not use assumptions that are unreasonably pessimistic.
Ex. P-2002 at 6, ¶ 3.3. ASOP 18 was introduced on the last day in the last hour of the hearing in the Rehabilitator’s redirect examination of his rebuttal witness, Mohoric. (Mohoric) N.T. 11/2/11 at 64-65. Intervenors objected because ASOP 18 had not been produced or discussed in the Rehabilitator’s direct case, Intervenors’ case, in the Rehabilitator’s rebuttal case or in Intervenors’ cross-examination of the Rehabilitator’s rebuttal witness. The exhibit was admitted because it had been referenced in Milliman’s July 2010 Surplus Report. See Ex. P-961 at 3. Counsel for the Rehabilitator described ASOP 18 as “not a document that’s necessary for an evidentiary point. It’s a standard of practice applicable to the actuaries.” N.T. 11/2/11 at 72. Mohoric testified about Paragraph 3.2, which pertains to assumption setting, and did not testify about Paragraph 3.3 of ASOP 18. Id. at 65.
D. Milliman Expert Report and Testimony
i. Milliman Morbidity or Claim Costs Assumption
The “morbidity” or “claim costs” assumption produces the ultimate claim costs expected from the approximate 142,000 policies in existence as of December 31, 2008, over the course of 60 years. Claim costs vary by age, sex and benefit period; by benefit configuration; by selection factors related to when the policy was underwritten; as well as other factors, such as availability of providers to a potential claimant.
*404The July 2010 Surplus Report sets forth Milliman’s future claim projections. Pfan-nerstill developed the disabled life reserve, which is the amount needed to pay the 8,000 existing claims to completion. James Kroll, in-house actuary for the Companies, developed the reserves for the incurred but not reported claims (IBNR) and the closed but expected to reopen claims (CBER). The statutory claim reserves consist of the disabled life reserve, the IBNR reserve and the CBER reserve.
To develop his disabled life reserve, Pfannerstill abandoned Milliman’s prior methodology and devised a new methodology “to calculate the reserves and to revise several assumptions, including the continuance curves and other adjustment factors used in the calculation process.” Ex. P-962 at 1. The continuance curves measure claim development by age; sex; type of benefit; and type of claim, e.g., cognitive impairment or cancer. The continuance curve estimates how long an existing claim will last. The continuance curve is the same as a continuance table; it determines the duration of claims to the point that they zero out. (Pfannerstill) N.T. 3/22/11 at 23-24.
Using his new methodology, Pfannerstill revised the continuance curves to increase the length of time a policyholder is expected to remain on claim in a nursing home, in an assisted living facility or in his own home but with paid assistance. Pfanners-till used data “for claims incurred from January 1, 2002, through December 31, 2009, with payments through December 31, 2009[,] to develop an initial set of continuance curve tables ... [, which] include 39,413 claims and approximately $1.2 billion in claim payments.” Rehabilitator’s Reply Brief at 56, ¶ 145 (quoting the Milli-man continuance curve report). The new methodology increased the disabled life reserve as of December 31, 2009, by $94.3 million, ie., from $398 million to $494.1 million. This is an increase of 23.6%. Rehabilitator’s Proposed Findings of Fact at 89, n 424, 425.
The continuance curves drive the claim reserves. As the Rehabilitator describes the process, the “continuance curves feed into the claim cost study and the claim cost study flows into the corporate model.” Rehabilitator’s Reply Brief at 56, ¶ 145. “Corporate model” is another term for “projection.” The claim reserves form the basis of the active life reserve, which is the amount needed to be held for “future claims expected in excess of future premium.” Ex. P-961 at 12.
Milliman’s claim costs assumption focused on more recent experience, 2006 to 2009, because it was believed that this experience was “most likely [to] reflect future claims.” Ex. P-963 at 1. During this period, claims were higher than in preceding years. In comparing its claim costs assumption to historical experience, Milliman sought to achieve an actual to expected “ratio [of] 1.00 over the most recent calendar years, keeping in mind the influence of the claim reserve and IBNR on calendar years 2008 and 2009.” Ex. P-963 at 1. The exercise of comparing an assumption to historical experience is known as fitting. Milliman’s fit showed that its claim costs assumption projected a lower incurred claim reserve for the 2006-2009 period than was actually set by the Companies’ claim department.
The Companies experienced a jump in paid claims in 2006, which drove the increase in their claim reserve. Approximately 85% to 90% of the 2009 incurred claims consists of a reserve that estimates how much will be needed to pay the 2009 claims. Sixty percent of the 2008 claims are reserve. These more recent periods of incurred claims, which consist primarily of estimates not actual payments, dramatical*405ly affected Milliman’s claim costs assumption. To illustrate the impact, Volkmar explained that claims payments increased incrementally at approximately $1.5 million per month, or $63 million over 42 months, from mid-2006 to the end of 2009. This $68 million increase in paid claims increased the present value of all projected future claims by $1.7 billion. (Volkmar) N.T. 10/26/11 at 121.
The Litow Report was authored in 2008 by Mark Litow, a Milliman senior actuary, to analyze PTNA OldCo claims experience for use in projections.28 It concluded that the rate of increase in incurred claims, which had slowed down in the period 2002 to 2004, would continue, and produce a flatter slope in the continuance curves. Milliman’s April 2009 Report to the Reha-bilitator updated the Litow Report assumptions by increasing claims on all policies, not just on OldCo business. The April 2009 Report assumed an increase in claims of 5% that would be caused by healthy policyholders seeking coverage from other insurers in response to the rehabilitation order. Milliman’s August 2009 PowerPoint assumed a claims deterioration factor of 25% caused by shock lapses in the first projection year and then grading down to 0% over 10 years. The October 2009 Report further increased claim costs by using what Pfannerstill described as a “quick and dirty” approach. N.T. 3/23/11 at 140-41. Thereafter, Pfan-nerstill undertook the above-referenced continuance curve study that abandoned Litow’s methodology; Pfannerstill’s report is part of the July 2010 Surplus Report.
Milliman’s July 2010 Surplus Report used a claim costs assumption consistent with that used in Scenario B of the October 2009 Report. The report splits the aggregate claim costs into frequency and severity assumptions and then recombines them. It assumed an anti-selection factor of 5% to be caused by cumulative premium rate increases prompting healthy policyholders to look for coverage elsewhere.
Milliman’s “2009 Guidelines” combine data from Milliman’s longterm care clients and are used as a starting point for pricing new products for an insurance company client with no data or actual experience. The 2009 Guidelines indicate trends in the industry, not the experience of a particular company, which will have its own policies with different benefit designs, markets and claims processing systems. Milliman used the curves in the 2009 Guidelines as a benchmark, ie., as a “visual aide” to determine if the slope should be “steep, ... medium steep, [or] should be flat.” (Pfan-nerstill) N.T. 3/24/11 at 47. Pfannerstill’s continuance curve report also states that the Guidelines impact “termination rates for months 43 through 54 as the termination rates for those months are based on a blend of actual experience and the termination rates from the Guidelines.” Ex. P-962 at 9. The Guidelines showed a pattern of steeper claim costs, industry wide.
Attachment 18 to Volkmar’s Report shows the progression of claim costs changes made by Milliman from report to report. Ex. R-813. Claim costs for certain policy benefit designs in the July 2010 Surplus Report were 46% to 130% higher than what was contained in the 2008 Appraisal Report. Attachment 18 to Volk-mar’s Report demonstrates the explosive *406effect of Milliman’s adjustments to its claim costs assumption. For example, the July 2010 Surplus Report adjusted the ultimate claim costs assumption29 for attained age of 95, making it 130% higher than it was in the Appraisal Report. The dramatic increase in claim costs at higher attained ages and longer policy durations affects the entire block of business as it gets projected forward with the 130% increase in claim costs.
Volkmar found the July 2010 Surplus Report to offer more detail on Milliman’s development of its claim costs assumption. However, Volkmar raised three criticisms: (1) lack of appropriate consideration of historical experience; (2) lack of transition from the former model and assumption paradigm developed in 2008 by Litow to the new model and assumption paradigm developed by Pfannerstill after the Reha-bilitator decided to seek a liquidation; and (3) lack of appropriate consideration of data credibility, including its leveraged impact on claim projections. (Volkmar) N.T. 10/26/11 at 103-29.
Volkmar testified that Milliman should have undertaken a detailed reconciliation analysis with respect to the historical experience covered in the Litow Report. N.T. 10/26/11 at 106-07. The Rehabili-tator rejoins that this reconciliation was done by Pfannerstill in his continuance curve report. See Ex. P962. This continuance curve report does not explain, however, how the Companies’ historical experience supported results so different from those in the Litow Report.
According to Volkmar, an appropriate transition from the earlier assumption and modeling paradigms to the ones used in the July 2010 Surplus Report would have merged prior and emerging experience in a systematic way over time, consistent with the mandate of ASOP 25. Instead, there is discontinuity and a huge jump in the claim projections, which nearly doubled between the September 2008 Appraisal Report and the July 2010 Surplus Report. Volkmar testified that, “Mr. Mo-horic and Mr. Pfannerstill ... reacted to data that was not fully credible and extrapolated that data into the future. And that led to a significant increase in projected claims ... as high as 65% within four to five months’ time.” N.T. 10/26/11 at 112-13.
Stephen LaPierre, who formerly headed the Companies’ claims department, now works as a consultant for the Rehabili-tator. LaPierre worked closely with Robert Conrad, a claims specialist employed in the Department’s Office of Rehabilitations, Liquidations and Special Funds.
In early 2010, LaPierre reported to Conrad that the rate of growth in the Companies’ average open claim count had dropped by 0.25% in 2009 as had the rate of increase in paid claims, particularly those relating to cognitive deficit claims. Cognitive claims paid dollars also dropped. In turn, Conrad communicated this information to DiMemmo in a February 22, 2010, memorandum, which presented claims data by gender and age, number of claims, payments by state, premiums by state, and diagnostic code claims for the time period 2003 to 2009. The Conrad Memorandum reported that “the rate of growth for paid claims dropped by approximately 3.5% between 2008 and 2009.” Ex. R-59 at 1. The Conrad Memorandum also reported “cognitive claims paid dollars dropped from an average annual increase *407of 5.4% between 2005 and 2008 to a rate of [0].57% between 2008 and 2009.” Id.
LaPierre provided Conrad a spreadsheet that tracked the claim dollars spent by PTNA and ANIC each year, by each claim condition, for the period 2003 to 2009. The spreadsheet shows that the increase in claims dollars paid annually by the Companies between 2003 and 2009 was in rough parity with the general rate of inflation. The changes in actual claim payments are summarized in the following table:
[[Image here]]
Ex. R-62.
The Rehabilitator does not explain the sudden increase in paid claims in 2006. Rehabilitator’s Reply Brief at 95-96, ¶ 232. Mohoric testified that industry wide, claims increased over time as services, particularly assisted living facilities, became more available. However, this trend had developed long before 2006.
Volkmar described the Milliman process as “leveraging” the incremental claims increase from a short period of time to produce a huge increase in the projected claims over a long period of time. He opined that feeding low-credibility data onto incurred claims data was multiplicative. Volkmar testified that there is no basis for assuming that where claims increase dramatically at a distinct point in time, such as 2006, they will continue to increase at that rate.
ii. Milliman Morbidity Improvement Assumption
The “morbidity improvement” assumption predicts the expected improvement in policyholder morbidity over time. For example, in ten years, a 70-year-old is expected to be healthier and less likely to produce a claim than a 70-year-old of today. This phenomenon impacts expected claims. For example, if an 80-year-old individual has an expected claim of $100 in 2012 and an 81-year-old individual has an expected claim of $110 in the same year, a 1% morbidity improvement posits that by the time the 80-year-old reaches age 81 (in 2013), his claim cost will be $109, not $110. Rehabilitator’s Proposed Findings of Fact at 27, ¶ 125.
Milliman reduced the morbidity improvement assumption between April and August of 2009. Milliman’s changes to the morbidity improvement assumption are illustrated in the following table:
*408[[Image here]]
Ex. R-1057.
Litow developed a morbidity improvement assumption of 2.5% per year, based on a detailed study that showed results as high as 3.9% for some categories of policyholders. The average of these results was 2.5%, which Pfannerstill described as “slightly aggressive.” (Pfannerstill) N.T. 8/28/11 at 154-55. Milliman used a 2.5% morbidity improvement assumption in its April 2009 Report. Id. In its October 2009 Report, Milliman used a morbidity improvement assumption of 1.2% per year for Scenario A, which it arrived at by taking the square root of the prior 2.5% assumption. This calculation had no precedent in the experience of Mohoric, Pfan-nerstill or Volkmar. Nor is there any support in the actuarial literature for “a square root approach.” For Scenario B, Milliman chose a morbidity improvement assumption of 1.5% for 10 years and 1% for the remainder of the projection period. Mohoric testified that morbidity improvement is a matter of judgment that “leans more in the art side than science when you’re projecting [far] into the future.” N.T. 2/24/11 at 81. Mohoric also testified that assuming a 1.5% morbidity improvement “forever” would also be reasonable. Id. at 83. Mohoric referred to a study by Eric Stallard of Duke University, which showed that there is “certainly” a morbidity improvement of anywhere from 1.5% up to 2.5%. N.T. 2/16/11 at 147-48.
Pfannerstill testified that Milliman reduced the morbidity improvement assumption because “paid claims were increasing.” N.T. 3/22/11 at 152-53. He estimated that the reduction in the morbidity improvement assumption accounted for approximately 30% of Milliman’s increase in projected claims from $2.6 bil*409lion to $4.1 billion. N.T. 3/22/11 at 112. Volkmar calculated that Milliman’s reduction in the morbidity improvement assumption between its 2008 Appraisal Report and its July 2010 Surplus Report caused a $530.9 million decline in the projected value of cash flows for the Companies. For PTNA alone, the change was negative $472.4 million, or a 38% change.
iii.Milliman Mortality Assumption
The “mortality” assumption predicts the probability of death of a policyholder, both the policyholder on claim and the one not yet on claim. Milliman’s mortality assumption was based upon an aggregate mortality study done in 2005 using data through September 30, 2003.
iv.Milliman Voluntary Lapsation Assumption
The “voluntary lapsation” assumption predicts the number of policies that are terminated by choice of the policyholder, as opposed to a termination by policyholder death.
Milliman changed its voluntary lapsation assumptions in its various reports and presentations, which are illustrated in the following table:
[[Image here]]
Ex. R-1057. The aggregate impact was positive $277.4 million for the Companies. This is because Milliman excluded the impact of anti-selection in the lapse assumption analysis and, instead, used it in its morbidity analysis.
v.Milliman Interest Rate Assumption
The “reinvestment rate” or “new money rate” assumption predicts the earnings a company expects to make on its assets and cash flow.30 It assumes that asset earn*410ings are reinvested at the assumed new money rate.
Milliman used a reinvestment interest rate assumption of 7.54% in the April 2009 Report; lowered it to 5.92% in the October 2009 Report; and lowered it again to 5.74% in the July 7, 2010 Report. Milli-man’s changes in its interest rate assumptions are illustrated in the following table:
[[Image here]]
Ex. R-1057. Volkmar used two separate interest rate assumptions, a 5.74% rate and an interest rate that begins at 5.34% and then grades to 6.4% over five years. The parties stipulated that the interest rate assumptions of both actuaries are reasonable.
The net impact of Milliman’s revision to its interest rate assumption was negative $320 million, which represents a 23% reduction of the net premium and an 11% reduction in cash flow.
vi. Milliman Claim Adjudication and Wellness Assumption
The “claim adjudication and wellness” assumption predicts reductions in claims that are expected because of improvements in claim adjudication or improvements in morbidity that result from policyholders participating in wellness programs aimed at improving cognitive health and detecting conditions that may lead to claims earlier.
Milliman’s September 2008 Appraisal Report assumed a 9% reduction in future claims by 2011 due to claim adjudication and the wellness program, but that number was reduced to 6% by 2015 in the April 2009 Report. Scenario A in the October 2009 Report assumed up to a 6% reduction over six years; the wellness assumption was eliminated from Scenario B. The changes Milliman made to the claim adjudication and wellness assumption are illustrated in the following table:
[[Image here]]
Ex. R-1057.
Intervenors offered evidence to show that the Companies’ wellness program and improved claims processing were having a favorable impact on their claim costs. During LaPierre’s tenure, the Companies offered three wellness programs: the Brain Fitness Program designed by Posit Science that addressed cognitive health; the life line screening program that provided earlier detection of certain medical conditions; and the emergency response program that gave certain policyholders the means to call for help in the event of a fall or accident. LaPierre testified that he strongly supported the wellness programs *411because they provide a valuable benefit to policyholders and address, positively, the circumstances that can lead to a claim.
Cognitive claims make up approximately one-third of the expected claim costs for the Companies. Approximately 10,000 policyholders participated in Posit Science’s Brain Fitness Program at a cost of $100 per participant, or $1 million per year. Milliman did an analysis that indicated that claim experience for policyholders in the cognitive exercise group was approximately 35% better than a matched sample group. Milliman quoted a contract price of $15,000 to do an analysis of the second year of data from policyholders enrolled in the Brain Fitness program. LaPierre asked Posit Science to pay the cost of the study, and Posit Science offered to split the cost of the study with PTNA. The Rehabilitator declined the offer.
On April 23, 2010, LaPierre wrote a memo supporting the wellness programs based on positive claims data through 2009. LaPierre differentiated the Brain Fitness Program used by the Companies from another program which required fewer hours of participation and was reported to have limited benefit. Ex. R-955 at 3. LaPierre compared 6,200 PTNA policyholders in the program against 6,200 PTNA policyholders who did not use the program. Claim experience for the cognitive exercise group was approximately 35% lower than the experience of the matched sample group. Id. LaPierre also testified at trial that the rate of increased claim payment for cognitive claims between 2005-2008 was over 5% per year; it dropped to 0.6% in 2008-2009, the year that the Brain Fitness Program was in place. N.T. 9/19/11 at 266.
The Department’s claims specialist, Robert Conrad, accepted LaPierre’s study and testified that the drop in cognitive claims was trending favorably. Nevertheless, the Rehabilitator terminated the Posit Science program. No one consulted Conrad about the decision to stop the wellness program, and he did not know how the decision was made. After the wellness program was cancelled, Milliman removed the wellness assumption from the projections. This removal reduced expected cash flow by $190 million.
vii. Milliman Premium Rate Increase Assumption
The “premium rate increase” assumption presents the expected future rate increases that will generate revenue.
Milliman used a variety of aggregate rate increase assumptions in its various reports and presentations, which are illustrated in the following table:
[[Image here]]
*412[[Image here]]
Ex. R-1057.
Milliman’s April 2009 Report assumed a “routine method” of pursuing rate increases and projected positive surplus for PTNA by 2020. Its October 2009 Report assumed slightly higher-than-routine rate increases in Scenario A and lower-than-routine increases in Scenario B. The July 2010 Surplus Report chose the more pessimistic Scenario B rate increase assumption.
Milliman’s rate increase assumptions stop at 10 years even though the claims projections go out 60 years. Mohoric testified that it was the Rehabilitator’s decision not to project any rate increases beyond the 10-year point. He also explained that after 10 years, the block of business remaining should be low enough in population that rate increases would become less relevant. Pfannerstill testified that the rate increase assumptions in the reports are not “best estimates,” i.e., have an equal probability that actual experience will fall above or below the experience predicted by the assumption. Pfannerstill used his judgment, with guidance from Robinson and Waite, to set Milkman’s “estimation” of what could be “reasonably achieved.” N.T. 3/23/11 at 50. Pfannerstill issued a report that the maximum allowable rate increase is one “needed on future premium to solve for a lifetime loss ratio of 60%.” Ex. P-968 at 3. His report also stated that the “maximum allowable rate increase for PTNA is 58% and the maximum allowable increase for ANIC is 121% using the projection assumptions described in [the July 2010 Surplus Report].” Id. at 5.
Milliman projected the amount of premium increases needed to return the Companies to solvency as follows:
[[Image here]]
Ex. P-961 at 14. Milliman did not use these rates in its rate increase assumptions.
The combined impact of Milliman’s reduction to its premium rate increase assumptions was negative $107.4 million, which represents an 8% decline in net premium collection and a 4% decrease in cash flow.
E. United Health Actuarial Services, Inc. Expert Report and Testimony
United Health did separate projections for PTNA and ANIC because they are separate entities; likewise, the question of whether to liquidate PTNA is separate from the question of whether to liquidate ANIC. The purpose of United Health’s projections was to answer the question of whether it is futile from an actuarial perspective to continue the rehabilitation of PTNA and ANIC. Volkmar defined futility as “can I come up with a reasonable assumption set that would allow each ... of the companies to at least meet all their obligations, if not emerge from rehabilitation.” N.T. 10/26/11 at 158. Emergence from rehabilitation would occur when each company’s statutory surplus meets the “200% RBC level.” Id. Volkmar used *413“reasonable assumptions,” not “best estimate” assumptions. Volkmar used the Milliman MG-ALFA model for his projections so that any differences in his projections would result from his different assumptions, not the model. Further, Volkmar reviewed and calibrated the Mil-liman model to ensure this result. Volk-mar testified that his projections are susceptible to the same types of caveats as Milliman’s projections because it is impossible to do a 60-year projection with a high degree of certainty.
i. United Health Morbidity or Claim Costs Assumption
United Health developed its claim costs assumption from the same claims data the Companies supplied to Milliman. United Health’s claim costs assumption used an ultimate claim costs slope consistent with Scenario A from Milliman’s October 2009 Report and the selection factors from Scenario B, but Volkmar removed them after the tenth policy year, when the effects of underwriting will have worn off. Volkmar testified that he used emerging claim experience in a manner consistent with ASOP 25. Volkmar did not use the incurred claims for 2008 and 2009 but, rather, the 2008 and 2009 paid claims data. Paid claims data showed an incremental increase. Incurred claims, which were mostly reserve, i.e., an estimate of the claim cost, increased sharply. Volkmar concluded that Milkman’s decision to use incurred claims and not paid claims was purposefully conservative and pessimistic.
Volkmar did the claim costs analysis by dollar of maximum daily benefit, for the different benefit designs in the Companies’ policies. He then compared his claim costs with those done by Milliman; he also compared his claim costs with those of two other United Health clients with comparable products. N.T. 10/26/11 at 169-72; Ex. R-1058 at 11-13. Attachment 7 of Volkmar’s Report shows that in some cases his claim cost curves were higher, and in other cases they were lower, than the comparables. Ex. R-802. In general, Volkmar’s claim cost curves were consistent with the comparables.
United Health capped selection and marriage wear-off factors at ten years duration. This was supported by: (1) the Society of Actuaries’ 1984-2004 Long-Term Care Intercompany Study selection analysis, which truncates the effect of selection factors at ten years; (2) another long-term care actuarial consulting firm’s standard pricing assumptions, to which Volkmar had access, showing that the effect of selection factors is ten years or less, especially for the average issue-ages and underwriting style associated with the Companies’ OldCo business; and (3) Volk-mar’s experience with long-term care claim costs for various clients. N.T. 10/26/11 at 178-83; Ex. R-1058 at 12.
As noted, the concept of “fitting” refers to the test done to an assumption by looking backwards to see if the assumption, had it been in place in the past, would have produced the actual experience. (Volkmar) N.T. 10/26/11 at 184. Stated otherwise, it compares actual experience to expected experience. To fit his claim costs assumptions, Volkmar employed his normal practices. Volkmar’s fit was illustrated in comparison to Milliman and shown in Attachment 9 to his expert report. Ex. R-804. Volkmar did a number of adjustments to compare actual to expected, incurred and paid claim costs. Volkmar’s Report presents the same level of detail and types of charts that Pfanners-till produced for his continuance curve study and report. Attachment 9 replicates credible historical experience and shows Volkmar’s fit going back to calendar year 1993. Attachment 9 shows that Volkmar’s claim costs assumption tested well; they *414fit actual experience for time periods and policy durations where there was credible data. Volkmar used the information in Attachment 9 to confirm the reasonableness of his claim costs assumption.
United Health’s claim costs assumption is based on more than five times the number of exposures Milliman used. Milliman used data from 2002 through 2009, and Volkmar used the full database worth of exposures.
ii.United Health Morbidity Improvement Assumption
United Health used a morbidity improvement assumption of 2% per year, which Volkmar testified to be within the range of actuarial practice based on (1) his experience, (2) presentations he attended at the Society of Actuaries, including a presentation by Milliman, (3) population data in a Milliman presentation and article, and (4) Milliman’s assumptions through the middle of 2009.
Volkmar has used a 1% morbidity improvement for another client, CalPERS, which is a self-insured, long-term care plan for public employees in California. The CalPERS Board adopted the assumption before Volkmar was engaged as its actuary so he did not recommend changing it. Volkmar testified that the 1% morbidity improvement assumption used for Cal-PERS and the 2% assumption used for PTNA and ANIC are each consistent with the data sources and supported by population experience.
United Health’s 2% assumption is consistent with Milliman’s past practices. Milliman used a 2.5% morbidity improvement assumption in its April 2009 Report, which Pfannerstill characterized as “slightly aggressive.” N.T. 3/23/11 at 154.
iii. United Health Mortality
Assumption
United Health adjusted Milliman’s mortality assumptions to reflect actual experience, which shows an increase in the proportion of disabled lives to total lives. Milliman’s mortality assumption was based on a 2005 study that used data through September 20, 2003. (Volkmar) N.T. 10/26/11 at 202-08. Volkmar testified that the 2005 mortality study used by Milliman was outdated, as Milliman itself recognized, according to its file notes. Id. Using 2003 data could have a significant impact on the projections because there have been a significant number of additional deaths since 2003. Further, the prior study did not separate mortality for active lives versus disabled lives. Even a small number of deaths could swing the mortality rates by as much as 20%. In other words, Milliman did not account for the fact that when a growing proportion of policyholders are on claim, aggregate mortality increases.
Volkmar did an analysis to determine the aggregate mortality rate of the group and then reduced that aggregate mortality figure to reflect voluntary lapsation. He also developed factors that increased mortality over time to reflect the fact that more in-force policies will be on claim as time passes. Volkmar opined that his approach was not optimal but necessary in the absence of an updated study based on data more current than 2003.
iv. United Health Voluntary Lapsation Assumption
Volkmar began with Milliman’s voluntary lapsation assumption for Scenario A in the October 2009 Report and adjusted it to reflect his adjustment to Milliman’s mortality assumption.
*415v.United Health Interest Rate Assumption
United Health used two different interest rate assumptions: (1) the flat 5.74% assumption used by Milliman in the July 2010 Surplus Report and (2) a 5.34% interest rate assumption, described by Milliman as the current rate, which Volkmar graded to 6.4% over five years. This was based upon the 20-year average return for the investment portfolio that was in place for the Companies. The parties stipulated that United Health’s assumption was reasonable.
vi.United Health Claim Adjudication and Wellness Assumption
United Health’s claim adjudication and wellness assumption assumed a 1% increase per year for the next six years and a flat 6% thereafter. It simply retained the assumption developed by Milliman before the Rehabilitator terminated the Companies’ wellness programs. Volkmar explained that the early signs indicated that the results were positive and that it was reasonable to assume that the Companies would experience the same types of gains they had been experiencing should they reinstitute the wellness programs. He acknowledged that he did not use a wellness assumption for CalPERS, explaining that CalPERS does not have a wellness program.
vii.United Health Premium Rate Increase Assumption
Volkmar explained that United Health used rate increase assumptions that would allow the Companies to continue to meet their obligations and potentially emerge from rehabilitation in the future. Attachment 6 of Volkmar’s Report provides a summary of rate increase scenarios he developed, using his two different interest rate assumptions, i.e., 5.34% graded to 6.4% over five years and a flat 5.74%. Ex. R-801. The scenarios relate to and cover OldCo policies and represent aggregate rate increases for all states. The different scenarios are shown in the following table:
*416[[Image here]]
Ex. R-801 at 80.
Volkmar’s rate increases relate only to OldCo policies because NewCo policies do not need rate increases. Scenario A assumes the lower interest rate of 5.34%, and Scenario D assumes the higher rate of 6.4%; the interest rates in Scenarios B and C fell between 5.34% and 6.4%.31 Volk-*417mar testified that these rate increases and assumptions are actuarially justified.32 The table shows that by returning to Milli-man’s April 2009 morbidity improvement assumption of 2.5%, PTNA would achieve 200% ACL between 2027 and 2036, and ANIC would achieve it between 2012 and 2021. The date depends on which of the four scenarios and new money rate assumptions is used. The reference to “200% ACL” in the attachment refers to the minimum RBC level needed to permit PTNA and ANIC to emerge from rehabilitation.33
Using either new money rate assumption and assuming a 2% morbidity improvement assumption, United Health developed two rate increase scenarios for PTNA that required no additional rate increases after 2024. Under a 2% morbidity improvement assumption, PTNA’s statutory surplus remains negative until 2039. Volkmar explained that this made sense because “the point of setting the PTNA assumptions the way we did was in order to make sure that [PTNA] had enough cash to meet obligation[s].... And that would indicate that the surplus was going to remain negative throughout the projection period given that definition.” N.T. 10/27/11 at 88.
The rate scenarios were more favorable for ANIC. Using a 2.5% morbidity improvement assumption, United Health’s rate scenarios demonstrated that ANIC could achieve 200% ACL anywhere from 2012 to 2021.
United Health also analyzed the average rate increases for all OldCo policies should benefit caps be put in place for certain policyholders, who would then be relieved of rate increases. Two scenarios produced by this analysis are set forth in Attachment 6 of Volkmar’s Report:
*418[[Image here]]
Ex. R-801 at 81.
Applying a benefit cap of five years to lifetime benefit policies, or a four-year cap to lifetime and five-year policies, would depress the amount of needed rate increases. Assuming a five-year benefit cap for PTNA, as shown above, PTNA would be able to meet all obligations with much smaller rate increases, which would end in *4192016. A four-year cap would shorten that period by one year.
Assuming a five-year or a four-year benefit cap for ANIC would mean that no rate increases would be required for ANIC to meet all .of its future obligations immediately.
Milliman opined in its Rebuttal Report that United Health’s rate increase assumption was unreasonable. Ex. P-969 at 10-11. It asserted that United Health failed to consider the expense of repeated rate filings; the Companies’ lack of success in securing actuarially justified rate increases in the past decade; or the possibility that states would require a non-forfeiture benefit as a condition of approving substantial rate increases. The Rehabilitator notes that Volkmar did not opine to a reasonable degree of certainty that his rate increases were feasible for PTNA. Rehabilitator’s Proposed Findings of Fact at 4, ¶ 7. Volk-mar testified that his rate increase scenarios were achievable based on his experience; that he ran tests to see if they were reasonable; and that based on those tests and his experience, his projected rate increases were achievable and reasonable. N.T. 10/27/11 at 98-100.
viii. United Health Agents’ Commission Assumption
United Health assumed continued payment of agents’ commissions because they have not currently been canceled, reduced or terminated. Milliman assumed the cessation of agents’ commissions in its projections.
F. Gross Premium Reserve
The first time a gross premium reserve appeared in a formal report was in Milli-man’s October 2009 Report. However, Milliman did not explain its methodology until the July 2010 Surplus Report, at which time it revealed that it used incurred claims, as opposed to paid claims, to establish its gross premium reserve. The use of incurred claims was devised by Mohoric, and it was unique to this engagement. The effect of this methodology was to increase the Companies’ composite reserves by $528 million in Scenario B of the October 2009 Report. After criticism, Mo-horic switched to a “semi-discounted incurred claims” approach to develop his gross premium reserve. (Mohoric) N.T. 2/22/11 at 169. This approach resulted in a $400 million hit to the Companies’ reserves. Id. at 168-69.
As noted, statutory reserves are established to ensure that the insurer will have the funds needed to pay all present claims and those that develop in the future. The Companies’ claim reserves state the amount expected to be paid on open and incurred claims; the active life reserves state the amount expected to be paid on future claims to be developed by the “active lives,” i.e., policyholders not on claim. The gross premium reserve tests whether the statutory reserves meet the minimum statutory reserve requirements and must be done by insurers engaged in long-term care business on an on-going basis. (Volk-mar) N.T. 10/26/11 at 137. The gross premium reserve tests reserve levels, but it does not inform the reader how much money a company will have to pay claims at a point in time. Id. at 138-39.
Intervenors challenged Milliman’s gross premium reserve because it was developed by using “semi-discounted incurred” claims, rather than paid claims. Apart from the fact that the Milliman method was novel, Volkmar testified that it was inappropriate. In a receivership, the object is to test whether assets will be on hand to pay claims that come due; for this purpose, paid claims should be used. Mil-liman’s approach increased the Companies’ apparent deficit, but this apparent deficit *420has nothing to do with its actual ability to pay claims. Volkmar could not follow Mo-horic’s reasoning “at all.” (Volkmar) N.T. 10/26/11 at 143. He opined that Milliman’s gross premium reserve was unnecessarily punitive. Id. at 148-44.
Mohoric concluded that Milliman’s gross premium reserve calculation was done to add an “element of conservatism.” N.T. 11/2/11 at 83. On rebuttal, he explained that it was done to avoid “masking” losses in the future.
Q. When you say, masking the losses in the future, what do you mean?
A. Well, because the minimum gross premium reserve, since we don’t have enough interest, we don’t have assets to back the reserves. Even our minimum gross premium reserve would have losses in the future. And the way we set it also has losses, but it’s got somewhat lower losses, trying to basically put some margin in there for the fact that it’s not fully accomplishing what a gross premium reserve is intended to accomplish.
N.T. 11/1/11 at 60-61. Mohoric believes that his gross premium reserve is still too low. However, Mohoric did not refute Volkmar’s point that apparent deficits are not relevant for determining whether an insurer in rehabilitation will have the cash flow needed to fund claim payments as they come due.
G. Ernst & Young Report
David Minches testified about Ernst & Young’s review of Milliman’s updated anal-yses and studies and about his firm’s report.34 Ernst & Young compared Milli-man’s assumptions to those of two other “LTC carriers we are familiar with.... ” Ex. P-53 at 7. The report offered factual observations of what Milliman did but did not offer any conclusions about what Ernst & Young observed, positive or negative.
Minches explained that Ernst & Young did not peer review Milliman’s work and did not do any projections. N.T. 3/25/11 at 22-23. Further, Ernst & Young’s report was issued to comply with a directive from a principal in the firm that with regard to Milliman’s work, “[n]o matter what, we cannot opine on the reasonableness of anything, including the weather, and cannot have a conclusion on anything.” Ex. R~ 839 at Bates No. EY-PT-RWAI-EM-000062; (Minches) N.T. 3/25/11 at 20, 33-34.
*421H. State Regulation of Long-Term Care Insurance Rate Making
In support of his argument on the futility of seeking “large” rate increases, the Rehabilitator offered the testimony of Arthur Lucker, a life and health actuary for INS Consultants, Inc. INS reviews and analyzes rate filings exclusively for state insurance regulators. The Rehabilitator retained INS in this case to (1) conduct a review of PTNA’s recent rate filing experience; (2) compile historical loss ratio information for PTNA’s long-term care products that could be used in support of future rate increases; (3) compile and analyze information with regard to the long-term care rate increases reviewed by INS for several state insurance departments; (4) document any known legal, procedural or insurance department rule limitations on granting rate increases for long-term care policies; (5) contact several state insurance departments and discuss their outlook on future developments with respect to long-term care rate increases; and (6) review for reasonableness relevant actuarial reports. Ex. R-65 at 2.
Lucker testified that he has 38 years of actuarial experience, the last seven of which he has focused on reviewing long-term care insurance rate filings. Typically, Lucker prepares independent projections to determine whether an insurer’s filing is actuarially justified under the state’s minimum loss ratio requirement, which is 60% in most states. Lucker also checks on the status of the insurer’s requested rate increase in other states, which usually involves discussions with regulators about their decision-making process. Although those discussions do not impact Lucker’s own analysis of the loss ratio for a specific rate increase, he will supply this information to the state insurance department for which he is conducting the review. Lucker testified that he makes “suggestions” to regulators regarding an appropriate rate increase based upon the state’s minimum loss ratio requirement. N.T. 3/25/11 at 121. He does not make recommendations regarding whether a specific rate increase request should be approved or disapproved.
Lucker estimated that he has reviewed more than 500 rate increase filings for INS in the past seven years, and in excess of 250 in the last three years. Lucker acknowledged, however, that at the time he authored his expert report on August 13, 2010, his rate review experience was limited to five states: Arkansas, California, Delaware, Iowa and Wisconsin. Lucker has reviewed rate increase filings by PTNA in three of those states: Delaware, Iowa and Wisconsin. The Rehabilitator offered Lucker as an expert on (1) the regulatory review of long-term care rate increase approval practices, and (2) whether the Companies can procure the aggregate rate increases that Milliman determined are necessary for them to achieve solvency by 2025 (280% for PTNA and 244% for ANIC) in its July 2010 Surplus Report. Ex. P-961 at 14. Lucker opined at trial and in his expert report that such rate increases are unlikely to be approved by a majority of state insurance departments, even if they are actuarially justified. N.T. 3/25/11 at 147; Ex. P-992 at 2.
Lucker did not conduct a formal independent analysis of the specific rate increases that were proposed for PTNA and ANIC in the states where rate increases would be sought in the context of rehabilitation. Instead, Lucker based his expert opinion upon: (1) his prior experience reviewing long-term care premium rate increase filings for five states; (2) his prior general discussions with state regulators; (3) information contained in a survey conducted for the Commonwealth of Massachusetts by another actuarial firm, Gor-*422man Actuarial, LLC (Gorman Survey); and (4) data regarding PTNA’s historical rate increase filing experience that was provided to INS by PTAC. The Court briefly considers the bases of Lucker’s opinion seriatim.
Lucker opined that the rate increases projected by Milliman as necessary to bring PTNA to solvency by 2025 could not be achieved in full in “the majority of states” based upon his experience reviewing long-term care rate increase filings for Arkansas, California, Delaware, Iowa and Wisconsin. N.T. 3/25/11 at 119; N.T. 4/12/11 at 8. Significantly, this experience did not involve rate filings made in the context of rehabilitation. Lucker also did not provide any testimony, evidence or supporting data to show that his experience in those five states could be extrapolated to the majority of the states where PTNA and ANIC have policies.35
Lucker’s opinion that the majority of states will deny actuarially justified rate increase filings by PTNA or ANIC in the future simply because the rate increases needed are “large” is based on conjecture.36 Much of Lucker’s knowledge is admittedly second-hand. For example, his understanding of how Iowa will handle rate increase requests is not based on personal knowledge or experience but on what an actuary working for the state of Iowa told him.
Lucker and INS were specifically engaged by the Rehabilitator to “[cjontact several state insurance departments and discuss their outlook on future developments with respect to [long-term care] rate increases.” Ex. R-65 at 2; N.T. 4/12/11 at 50. Lucker did not provide any examples of his discussions with state regulators about their outlook on future developments with respect to long-term care rate increases. He did not contact state insurance departments specifically for this purpose. He testified that he instead relied on prior general discussions that he had with state regulators in the normal course of business. Those prior general discussions were limited to three states: Delaware, Iowa and Florida. These discussions added only one state, Florida, to the five states where Lucker had actual and direct experience with the state’s review of long-term care rate filings.
Lucker could not identify the person in Florida with whom he had discussions or that person’s position. Lucker could not confirm that anyone he spoke to in any state had policy-making authority.37 Lucker did not testify that any state insurance department personnel actually told him they would not approve large long-term care rate increases in the future. He *423learned that some states have informal rate increase limits, but he did not learn what those specific limits were or how they would impact future rate increase filings. Even the one contact in Delaware identified by Lucker did not tell him anything more than “the Commissioner of Insurance is looking at every long-term care rate increase we do.” N.T. 4/12/11 at 68. One would expect no less of any regulator.
Lucker relied upon the Gorman Survey to augment his five-state experience. The Gorman Survey was provided to Lucker by the Rehabilitator’s legal counsel' specifically for use in his expert report. Lucker acknowledged that the Gorman Survey is actually a compilation of what other states chose to report to an actuary in Massachusetts in response to an informal survey. Lucker testified that the Gorman Survey is not a document that he relies on in the regular course of his profession; he had never used it before this engagement. N.T. 4/12/11 at 75-76. Without evidence verifying the information contained in the Gorman Survey, and none was offered, there is no basis to find that the Gorman Survey represents a reliable source of data. The Rehabilitator never offered the Gorman Survey into evidence.
Finally, Lucker relied upon information provided to INS by the Rehabilitator regarding the status of PTNA’s rate increase requests from 2001 to 2010. Ex. P-59. INS used that information to create Table 1 in Lucker’s report, which is a chart of PTNA’s rate increase filing experience for the five policy forms with the most business. Ex. P-992 at 13. The following table summarizes the data contained in Table 1:
Summary of Recent Nationwide Penn Treaty LTC Rate Increase Filinys
[[Image here]]
Lucker stated in his expert report that INS never verified the accuracy of PTNA’s historical rate increase information supplied by the Rehabilitator. Ex. P-992 at 7.
INS prepared a similar chart summarizing PTNA’s rate increase filing history from 2001 to 2010 in the eight states with 65% of PTNA’s premiums. Ex. P-994. The following table summarizes the data contained in Exhibit P-994:
Summary of Recent Nationwide Penn Treaty LTC Rate Increase Filings — 8 States with 65% of Premium
*424[[Image here]]
Lucker did not cite to or include the chart in Ex. P-994 in his expert report. It is unclear whether or to what extent he relied upon it in formulating his opinions.
The accuracy of the above referenced tables was not established. Nevertheless, they illustrate that PTNA was reasonably successful in obtaining rate increases from 2001-2010. During that time period, PTNA sought rate increases averaging 39%. Approximately 60% were approved in full; 14% were approved for 50%-99% of the requested increase; and 26% were disapproved or approved for an amount less than 49%. PTNA’s experience was even better for the same policy forms in the eight states with 65% of total premium, where PTNA sought rate increases averaging 40%. In those eight states, 66% of the filings were approved in full; 15% were approved for 50%-99% of the requested increase; and 19% were disapproved or approved for an amount less than 49%.
In short, the historical data demonstrate that the majority of PTNA’s requested rate increases from 2001-2010 (as high as 81%) were approved for at least 50% of the requested amount. To what extent PTNA’s rate increase experience from 2001 to 2010 will be a predictor of the success of future rate increase filings in the context of an approved rehabilitation plan is simply unknown.
I. Impact of Medical Advances on Future Long-Term Care Insurance Claims
Intervenors offered the testimony and expert report of Stephen K. Holland, M.D., to address medical advances that are expected to ameliorate the Companies’ future claim payments. Dr. Holland is the Chief Medical Officer of a long-term care insurance administrator, Univita, Inc., and is a board certified internal medicine physician. Approximately five physicians in the United States have long-term care experience comparable to Dr. Holland’s. N.T. 10/24/11 at 27.
Holland testified that mortality and morbidity, i.e., disease, are interrelated. Americans not only live longer lives, they also live more of their longer lives free from disability and dependency. The increase in life expectancy can be traced to advances in treatment of diseases and to improved lifestyle habits, such as better nutrition, exercise and smoking cessation. These same advances shorten the duration of end-of-life disabilities. The consequence for long-term care insurance policyholders is that they will (1) live and pay *425premiums longer than expected and (2) live in a dependent state for a shorter duration. Holland refers to the phenomenon of living longer but spending less of it in a dependent state as “morbidity compression.” Holland opined that morbidity compression will beneficially impact the Companies’ future claim costs, explaining:
Based upon an extensive body of scientific literature and my expert opinion, I conclude that individuals are living longer and that they will pay premium longer than expected when their Penn Treaty [long-term care] policies were priced and purchased. In addition, for those who do claim for [long-term care] benefits, there is a significant likelihood that their durations of dependency at the end of their lives will be much shorter than predicted. In conclusion, I conclude that all these factors will dramatically impact in a positive manner the morbidity trajectory of Penn Treaty’s [long-term care] claims and ultimately improve its profitability over time.
Ex. R-794 at 117. See also N.T. 10/24/11 at 55.
On questioning from the Rehabilitator, Holland acknowledged that James F. Fries, M.D. of Stanford University co-authored an article describing morbidity compression as a phenomenon that was not inevitable. See James F. Fries et al., Compression of Morbidity 1980-2011: A Focused Review of Paradigms and Progress, Journal of Aging Research, 2011, at 1; Ex. R-1121. Holland explained that by “not inevitable,” Fries meant that people will not live longer and healthier lives if they do not make lifestyle changes and if scientists do not continue meaningful research. For example, there are concerns that increasing obesity in younger people may adversely impact morbidity compression because it causes diabetes, which is associated with higher rates of morbidity and mortality. Holland explained that Fries’ article accepts the theory of morbidity compression, which is a widely accepted paradigm in the scientific community and used to measure health gains in the population. N.T. 10/25/11 at 22; Ex. R-1121 at 8. Fries and his group, Healthy People 2020, believe that with comprehensive public health planning, morbidity compression can be improved by lifestyle changes and further medical advances. To date, morbidity compression has been achieved serendipitously.
PTNA’s OldCo, non-tax qualified policies require a policyholder to establish only one of three benefit triggers to go on claim: (1) cognitive impairment, (2) deficits in two or more activities of daily living, or (3) physician certification that coverage is “medically necessary.” (LaPierre) N.T. 9/19/11 at 225-26. Typically, these policies have a 90-day deductible period, or waiting period, during which individuals cannot access policy benefits. If a policyholder recuperates or regains independence, for example, after a successful hip replacement, the policyholder will go “off claim.” (Holland) N.T. 10/24/11 at 160-61. Holland testified that medical advances constantly change the trajectory of disease and disability. These advances mean that long-term care insurance benefits are not triggered or are triggered for shorter durations. Id. at 52-53.
According to an internal Department memo prepared by Bob Conrad on February 22, 2010, the five diseases that are most likely to lead to an OldCo claim are: (1) organic brain disorder;38 (2) stroke or *426cerebrovascular accident; (3) heart disorders and congestive heart failure; (4) osteoarthritis and degenerative joint disease; and (5) musculo-skeletal conditions. Ex. R-552 at 8. Holland’s report focused on the diseases listed in Conrad’s memorandum, along with cancer and multiple sclerosis. Ex. R-794.
Alzheimer’s disease and other forms of dementia account for 20% to 40% of claims for most long-term care insurance carriers, including PTNA and ANIC. Dementia refers, generally, to a progressive impairment in cognitive functions caused by brain cell death. The accumulation of amyloid beta deposits (plaque) and tau protein tangles is the hallmark of Alzheimer’s type dementia. Ex. R-794 at 20. Alzheimer’s disease is now believed to have a 10 to 15-year presymptomatie stage. Id. Most cases of Alzheimer’s disease develop after age 60, and the peak age of onset of the disease is 78. N.T. 10/24/11 at 92-93.
Holland opined that diagnostic advances, such as Positron Emission Tomography (PET) scans, will permit Alzheimer’s to be identified 5 to 10 years before the onset of symptoms and, thus, allow the intervention of preventative strategies. Ex. R-794 at 20, 26-27. Because new interventions will arrest the onset and development of Alzheimer’s disease and dementia, symptoms will be shorter in duration. Id. at 37-38. Stated otherwise, people will die from other causes prior to the onset of Alzheimer’s disease. N.T. 10/24/11 at 92-93; Ex. R-1080. This will reduce the number and duration of dementia related claims.
Holland discussed a large scale 2011 study in the United States that concluded that about 55% of Alzheimer’s disease in the United States is attributable to lifestyle factors such as smoking, depression, midlife obesity, physical and cognitive inactivity, and diabetes mellitus. The research showed that if 10% to 25% of those needing to make a lifestyle change did so, it would reduce new Alzheimer’s cases by 500,000. N.T. 10/24/11 at 179-80. Holland also testified about three 2011 European studies showing that regular exercise of brisk walking three to four miles a day significantly delays the onset of cognitive impairment. Id. at 98.
With respect to the treatment of dementia, Holland testified that there are five FDA-approved medications that forestall nursing home admission by one to two years and improve cognitive testing by 5% to 10%. N.T. 10/24/11 at 109-10; Ex. R-794 at 30. He testified that these first-generation medications have had only a modest impact on the course of Alzheimer’s disease because they were not widely used in the past. Ten years ago, only 10% of claimants with dementia were taking these medications, but his recent study showed that 70% to 80% of patients submitting claims are on these medications. N.T. 10/24/11 at 110.
Holland testified that the development of breakthrough drugs to arrest or delay neuronal cell damage is imminent. As of November 2010, there were 841 studies on Alzheimer’s disease underway. Only one successful study is needed to forestall or cure Alzheimer’s disease. Id. at 120. To forestall a claim by five years will cut initial claims by 35% to 50%. Id. at 121. A number of trials are underway for treatment of Alzheimer’s, and Holland noted that “these drugs [can] come to market very quickly.” Id. at 179. Holland testified to a high degree of medical probability that an advent of effective therapies or preventions will occur within the next five years that will have a positive impact on the Companies’ insured population. If the *427onset of Alzheimer’s disease is pushed back five years, the number of cases would be cut in half. A cure is not necessary to have a significant beneficial effect on long-term care insurance claims.
Cancer represents 4% to 5% of long-term care insurance claims payments, and 10% to 20% of new claims. N.T. 10/24/11 at 125. Cancer does not debilitate its victims until the final stages, and most potential claimants die during the elimination, or waiting period, in their policy. Holland summarized the number of cancer treatments developed in recent years and opined, within a reasonable degree of medical certainty, that continuing advances in treating cancer will have a beneficial impact on future claim costs of PTNA and ANIC. Ex. R-794 at 42-49.
Holland testified about claims produced by stroke, cardiomyopathy, and cardiovascular disease. Prevention by exercise; healthier diet; salt restriction; quitting smoking; surgical intervention, such as cardiac bypass, heart transplants, stents, carotid endarterectomy, and cerebral angioplasty; and medications, such as blood thinners, antiplatelet drugs, and statins will all reduce claims. N.T. 10/24/11 at 139-41. Holland’s report stated that from 1999 to 2007, there has been a dramatic reduction in the death rate from cardiovascular disease and stroke, and that new medical and surgical treatments will reduce the incidence, morbidity, dependency, and mortality from stroke. Ex. R-794 at 60. Holland opined to a reasonable degree of medical certainty that these advances will have a beneficial impact on the future claim costs of PTNA and ANIC. N.T. 10/24/11 at 142.
In the long-term care insurance databases Holland has examined, rheumatoid arthritis and osteoarthritis cause about 4% to 10% of initial claim requests. Ex. R-794 at 64. Osteoarthritis is treated with anti-inflammatory medications and joint replacement surgery. Research in inflammatory pathways will lead to even more improvements in disease prevention. Holland opined to a reasonable degree of medical certainty that improvements in preventing and treating rheumatoid and osteoarthritis have had and will have a beneficial impact on the future claim costs of PTNA and ANIC. N.T. 10/24/11 at 150-51.
In the data set Holland examined, central nervous system diseases, such as Parkinson’s disease and multiple sclerosis, constitute “about 7 percent of initial benefit requests.” Id. at 153. New methods for early diagnosis are being tested, which will mitigate risk factors and claims. As of 2010, four promising clinical trials were using gene therapy to treat Parkinson’s disease. Ex. R-794 at 90. Holland opined to a reasonable degree of medical certainty that improvements in preventing and treating Parkinson’s disease and multiple sclerosis will have a beneficial impact on the future claim costs of PTNA and ANIC. N.T. 10/24/11 at 158-59.
Osteoporosis is a bone condition marked by a decrease in bone mass and density, and it leads to fractures which account for 6% to 12% of new long-term care insurance claims. Ex. R-794 at 98. Preventative interventions, such as intake of calcium and vitamin D; weight-bearing and muscle-strengthening exercise (including walking and jogging); avoidance of tobacco use; and treatment of alcoholism are all effective. Holland opined to a reasonable degree of medical certainty that improvements in preventing and treating osteoporosis will impact future claim costs of PTNA and ANIC positively, decreasing the incidence and duration of claims. N.T. 10/24/11 at 166; Ex. R-794 at 111-12.
Holland provided a set of assumptions to United Health to simulate the impact on *428the PTNA and ANIC policyholder population if there were a 20% decrease in claims starting 5, 10, or 20 years from now. Holland testified that this type of modeling is done in medicine to look at the impact of one therapy over another. A decrease of 20% of claims starting five years from now would reduce the present value of paid claims by $256 million or 10%, and it would increase the present value cash flow by $296 million or 15%. N.T. 10/24/11 at 178.
Volkmar ran projections for the assumptions that Holland provided, which are set forth in Attachment 14 of Volkmar’s Report. Ex. R-809, R-1126.39 See also Ex. R-1058 at 16 (describing Attachment 14). Some medical breakthroughs discussed by Holland could have a combined positive impact on the Companies’ present value cash flow of anywhere from $51 million to $610 million and a combined impact on the Companies’ present value of paid claims of anywhere from $46 million to $556 million. Ex. R-809 at 258-95 (as corrected by Ex. R-1126).
The Rehabilitator did not refute Holland’s testimony about the medical advances in treating and preventing the diseases and conditions that produce claims. However, he is critical of several aspects of Holland’s work and conclusions.
First, the Rehabilitator notes that Holland did not have experience with the Companies’ pool of policyholders but used, instead, data from Univita, Inc., which was drawn from 250,000 insureds. (Holland) N.T. 10/25/11 at 99. Holland testified that all of the risk pools that he looked at behave the same because the same influences are affecting all Americans. He also testified that the “underwriting effect” of application screening wears off. Underwriting identifies those applicants more likely to present a claim in the near term, i.e., 7 to 10 years. N.T. 10/25/11 at 38-89, 94-96. There are no medical questionnaires or tests that can identify those presently healthy applicants who will develop cancer, Parkinson’s or Alzheimer’s disease 7 to 10 years into the future. Thus, the effects of the “loose” underwriting for the OldCo business, last written in 2001, has worn off. Holland explained that all long-term care insurance pools exhibit similar risk for diseases that may lead to claims. N.T. 10/25/11 at 27-28.
Second, the Rehabilitator is skeptical about Holland’s predictions for reduced claims. Rehabilitator’s Reply Brief at 139-40. The Rehabilitator notes that scientific advances in some diseases such as cancer, cardiovascular disease and autoimmune disease will extend lives and, as a consequence, the people who do not die of heart disease or cancer will live long enough to develop Alzheimer’s disease. According to Holland, the incidence of Alzheimer’s increases with age and the peak age is 78 years; for people over age 90, the incidence may be as high as 50%. The Rehabilitator notes that Holland testified that the cost of Alzheimer’s disease is projected to increase from $172 billion to $1.08 trillion over the next 40 years. Further, Holland’s testimony and exhibits demonstrate that Alzheimer’s and other dementias account for the highest percentage of long-term care insurance claims, both in incidence and severity. Rehabili-tator’s Reply Brief at 138, ¶ 334.
As an adjunct to this argument, the Rehabilitator points out that in 2008, Holland predicted that dementia costs were likely to rise, and that the duration of the disability would be longer. Holland testified that he no longer believes what he *429said in 2008. He has revised his opinion because of recent medical advances.
The Rehabilitator argues that the average age of PTNA policyholders, 77, and of ANIC policyholders, 74, is so advanced that medical breakthroughs are unlikely to benefit the Companies’ future claim costs. Intervenors challenge these average age numbers.40
The Rehabilitator offered no evidence to support these arguments, which have some logical appeal. He did not present expert testimony, for example, to show and quantify how improvements in treating heart disease will lead to more Alzheimer’s-related claims. The Rehabilitator did not attempt to refute Holland’s opinion that there will be significant breakthroughs in treating Alzheimer’s in the next five years. Holland could not specify when such an advance will significantly impact the Companies’ claims, but he was unshakable in his conviction.
Holland devotes his professional life to monitoring medical advances that may impact long-term care insurance claims by disease prevention and treatment; his knowledge is both wide and deep. Holland testified to a reasonable degree of medical certainty that the medical advances he identified together with lifestyle changes and new treatments will ameliorate the Companies’ future claim payments. The Court credits Holland’s testimony and proffered opinion that medical advances will mitigate future claims.
J. Evaluation of Actuarial Projections and Expert Testimony
In closing argument, the Rehabilitator challenged United Health’s projections by focusing on the following statement in Volkmar’s report:
Based on a review of the documentation provided, it appears that Milliman’s lead actuary role with respect to the companies transitioned from Mark Litow to Ed Mohoric sometime during 2008. The significant projection revisions discussed throughout this document approximately coincide with this leadership change.
Ex. R-1058 at 6. The Rehabilitator argues that this “fundamental statement” is demonstrably wrong. In truth, the statement is a neutral description of what happened “approximately.” There is no denying the chronology. After Mohoric took responsibility for the Companies’ projections, Milliman’s formal reports underwent stunning revisions. The Reha-bilitator explains these revisions as the result of claim payments that did not conform to Litow’s claim projections. This does not explain the explosion in Milliman’s statutory reserves. Interve-nors concede the increase in claims but contend that Milliman overreacted. They note that the increase in the Companies’ claim payments was incremental, ie., approximately $63 million over 42 months, but the revision to the projected future claims was not incremental. It exploded by $1.7 billion.
Mohoric himself conceded that “[w]e certainly made a major change to morbidity in mid 2009.” N.T. 9/20/11 at 205. Mohoric also testified about his December 5, 2008, phone conversation with DiMem-mo, in which Mohoric made a pitch to have *430Milliman continue to serve as consulting actuary after the Companies were placed into rehabilitation. Mohorie’s notes recorded that DiMemmo “will want conservative assumptions — not continually changing things along the way-critical if to build confidence among regulators.... ” Ex. R-987. Mohoric wrote down these comments “to understand [DiMemmo’s] expectations.” N.T. 9/20/11 at 205.
After the Rehabilitator’s retention of Milliman, the projections changed quickly and drastically. Claim costs projections increased by 65% six months after Milli-man began to report to the Rehabilitator. Milliman projected a steep increase in the tail of the claims curve, ie., the claims to be presented in future years by the older policyholders and the policyholders that have held their policies more than 16 years. The steep increase to the continuance curve resulted from Pfannerstill’s “focus” on two years of incurred claims from 2006-2008, a period when the claims increased. The Litow Report also used the most recent years of incurred claim experience, ie., 2004-2006, to do its continuance curves. However, the period used by Li-tow was favorable and thus led Milliman to project a flattening of the claims curve. This flattening was later disavowed by Mil-liman as too optimistic. Pfannerstill’s “steepening” may also prove too pessimistic for the same reason, ie., the focus on recent negative experience. Focusing on two years of incurred claims to set the continuance curve slope appears to produce unreliable results.
The Rehabilitator desired projections that would not have to be “continually” adjusted. One way to satisfy this directive would be to develop projections for a worst case scenario. Pessimism is seen throughout Milliman’s projections, whether intentional or coincidental.
For example, projected future earnings were set in the middle of the recession. Milliman’s July 2010 Surplus Report acknowledged that its interest rates were chosen in the middle of
exceptional volatility and turbulence in the capital markets. Readers of this report should make their own assessment of the reasonableness of the investment and default assumptions contained in the report.
Ex. P-961 at 19. Milliman lowered its interest rate assumption from 7.54% in April 2009 to 5.74% in July 2010, and this negatively impacted the Companies by $320 million. The change reduced net premium by 23% and cash flow by 11%. This dramatic reduction in earnings was set in the midst of a recessionary economic condition. Milliman acknowledged this fact, but then chose to exercise its judgment in a way that gave more weight to current conditions than to history.
In April 2009, the Rehabilitator filed his Preliminary Plan. This was a point in time where the Rehabilitator believed, apparently, that a rehabilitation was not a futile exercise, or he would not have filed the plan. With the plan, the Rehabilitator filed Milliman’s April 2009 report, which assessed the financial condition of each company and pointed the way to a correction of the Companies’ financial condition by means of a rate increase strategy. It is useful to compare the 2010 assumptions, of both Milliman and United Health, against Milliman’s April 2009 assumptions.
Virtually every one of Milliman’s assumption revisions adversely impacted its projections for the Companies. Projected claims increased by 50%; projected premium rate increases declined; the projected rate of asset earnings declined; the projected morbidity improvement assumption was cut in half; and the wellness assumption declined by 100%, ie., it was eliminat*431ed. Lapsation is usually a positive development for a long-term care insurance company. Here, Milliman assumed a higher rate of lapsation attributed to the rehabilitation, but it was not made clear whether this increased lapsation represented a positive or negative development for the Companies. It would be negative if the lapses represented an exodus of healthy lives. Milliman also assumed the elimination of agents’ commissions, but it was not explained how this would be helpful to the Companies. Because agents’ commissions are a factor in each company’s current approved rate filing, state regulators could demand a reduction in their approved rates to account for the elimination of commissions. This is especially true in those states that have approved adequate rates for OldCo policies. This is also true for NewCo policies, which are properly priced and generating profits. Milliman’s revised assumptions, in the aggregate, moved PTNA’s statutory surplus from negative $233 million to negative $2.2 billion.
United Health’s July 2010 assumptions also negatively impacted the Companies when compared to Milliman’s April 2009 assumptions. Volkmar’s claim cost assumption increased projected claims from $2.4 billion in April 2009 to $2.7 billion. His assumption for earnings from assets declined from 7.54% to 5.3% or 6.4%. He reduced the morbidity improvement assumption from 2.5% to 2.0%. He put the agents’ commissions back into the projections, which increased expenses. On the positive side, Volkmar retained the wellness assumption, and he projected higher future rate increases than were assumed in April 2009.
The differences between the actuaries’ projections reflect differences in the exercise of actuarial judgment, not differences in data.41 Milkman aimed for an assumption that would have a 50/50 likelihood of proving true. This does not advance the Rehabilitator’s case, as he strenuously argues. First, Pfannerstill could not explain how Scenario A and Scenario B in the October 2009 Report could both be “best estimates.” N.T. 3/23/11 at 143-44. Second, every assumption could prove to be wrong without deviating from the 50/50 likelihood. Third, the “best estimate” represents nothing more than the actuaries’ judgment about their own judgment. It is a self-awarded grade. Volkmar judged his assumptions to be “reasonable.” It too is a self-awarded grade. However, he explained that his assumptions were reasonable because they were consistent with his extensive experience in long-term care insurance consulting and with his work for other clients.
None of the actuaries could testify that their projections were presented as having a “reasonable degree of certainty.” This is because there is no certainty when dealing with 60-year actuarial projections. (Volkmar) N.T. 10/26/11 at 160-61; (Mo-horic) N.T. 2/22/11 at 150-51; (Pfanners-till) N.T. 3/24/11 at 78. The validity of a set of projections cannot be determined until actual experience emerges. (Mohoric) N.T. 2/22/11 at 68.
The Court concludes that for purposes of evaluating the respective surplus projections, Milliman’s description of its assumptions as falling into the 50/50 range does not make its work more reliable or persuasive. Accordingly, the difference between “reasonable” and “best estimate” has no bearing in the Court’s evaluation of the actuaries’ work.
*432The Court finds, as fact, that Milliman’s assumptions are too pessimistic. Individually and in the abstract, they may be explainable, such as the reduction in the interest rate assumption that coincided with the deepest economic recession since the Great Depression. However, in the aggregate and considering prior history, Milliman’s assumptions in the July 2010 Surplus Report present too negative a picture of the Companies’ surplus as of December 31, 2009. Milliman made choices that went beyond the course corrections required when paid claims exceeded what was expected under the Litow Report.
For example, Milliman eliminated the wellness assumption for the stated reason that the Rehabilitator eliminated the wellness program in 2010. However, the wellness program was in place for 2009 and it should have continued to have some effect for some period of time into the future. Milliman believed in the wellness assumption when it was made part of its pre-rehabilitation projections and, logically, it would continue to have some effect even after the Rehabilitator ended the program. To the extent the effects of the wellness program could be measured, indications were positive. The wellness assumption should have been eliminated gradually, not suddenly.
The central differences between Milli-man’s projections and United Health’s projections relate to the differences in their claim costs assumptions, their morbidity improvement assumptions and their premium rate increase assumptions. The Court accepts United Health’s morbidity improvement assumption. The Court rejects Milliman’s premium rate increase assumptions and accepts United Health’s rate increase assumption for purposes of determining the futility of continuing rehabilitation. Finally, the Court accepts United Health’s claim costs assumption.
In developing its premium rate increase assumption in the July 2010 Surplus Report, Milliman assumed a tepid, “business as usual” approach to rate increase filings, and it assumed no new rate increases after 10 years. The Rehabilitator requests a factual finding that the maximum allowable premium rate increase for PTNA for the next 60 years is 58%. The Rehabili-tator argues that because regulators do not allow recovery of “past losses,” regulators will reject rate filings at the level needed to pay claims.
The Court begins with the Rehabili-tator’s “past losses” argument. Past losses are not past claims. If that were the case, claims experience could never be used as the basis for a premium rate increase.
The Rehabilitator directs the Court to State Farm Mutual Automobile Insurance Co. v. Insurance Department, 133 Pa. Cmwlth. 644, 577 A.2d 951 (1990). Auto insurance rate making bears little relation to long-term care insurance rate making. Auto insurance rates are structured by policy year, and every year will have a different collection of policyholders. If an auto insurer suffers an unexpected underwriting loss in policy year 2000, for example, the year 2000 claims will be studied and may be used to set premium rates for future policies, should the analysis show that the year 2000 increase in claims represents a trend and not a statistical sport. The auto insurer will not, however, be allowed to add a factor in its rate filing to recover the 2000 underwriting loss. To do so would mean, as this Court noted in State Farm, that the cost of the loss recovery would be imposed upon a group of policyholders who were not the group that generated the loss.
There is no foundation in the record for the Rehabilitator’s past losses argument. His actuaries did not testify that the Com*433panies’ past losses were an impediment to future rate increases. Arthur Lucker, the Rehabilitator’s expert on state restrictions on rate increase filings, said nothing about “past losses” in his report or in his testimony.
During Volkmar’s cross-examination, the Rehabilitator presented a letter from the State of North Carolina denying a rate filing proposed by another long-term care insurance company, Washington National Insurance Company. Although the letter was not admitted, Ex. P-1057, counsel read a statement in the letter that one reason the filing was rejected was because it appeared to have the effect of “recouping from the few remaining policyholders a portion of past losses generated by policies no longer in force....” N.T. 10/27/11 at 116. The letter also stated that “only about five percent of the business originally issued on the affected forms remains in force.” Id. at 118. Volkmar observed that it would take a long time for the PTNA business to reach that 5% level. Id. Volk-mar also stated that he could not offer definitive comments without the “filing materials and everything else associated with this product,” ie., for which the rate filing was made. Id. In short, the Rehabili-tator’s exchange with Volkmar did not lay a foundation for the Rehabilitator’s past losses argument.
The Court rejects the Rehabilitator’s “past losses” argument, which appears to be an afterthought. The policy concern identified in State Farm is not present here. The Companies have a closed block of business. The challenge is to generate future premium revenue from this population to pay the future claims of this population, not to recoup past losses generated by former policyholders.
Next, the Court considers the Rehabili-tator’s argument that PTNA is limited to a maximum allowable rate increase of 58% and ANIC is limited to one of 121%. This proffered conclusion is not supported by the record.
Pfannerstill produced a Rate Increase Report in August of 2010, Ex. P-968, and it was offered by the Rehabilitator to support his belief that states will not approve the rate increases needed to produce a successful rehabilitation of the Companies. As explained in Pfannerstill’s report, states require a minimum expected loss ratio of 60%, which means that $0.60 of every premium dollar collected must be used to pay claims. Ex. P-968 at 3. The 60% loss ratio does not look at a discrete year but is measured over the life of a particular policy form. Id. The 60% ratio is measured by comparing the present value of past and future claims to the present value of past and future premium.
Critical to the 60% loss ratio is establishing the present value of past and future premiums. Pfannerstill identified three approaches for doing so. Approach 1 uses actual historical premium; Approach 2 assumes that current premium levels were in effect when the policy was issued; Approach 3 assumes all past and future rate increases were in place when the policy was issued. Approach 1 produces the lowest value of premium and Approach 3 produces the highest value of premium. Because Approach 3 creates the highest premium value, it yields the lowest maximum allowable rate increase for purposes of calculating the 60% lifetime loss ratio. According to Pfanners-till’s report, the State of Maine has adopted Approach 3. Pfannerstill used Approach 3 to calculate a maximum allowable rate increase, as directed by the Rehabili-tator’s counsel, and it yielded a maximum rate increase of 58% for PTNA and 121% for ANIC. Ex. P-968 at 5.
The Court rejects the Approach 3 method of determining past and future premi-*434urns. First, the Rehabilitator did not produce evidence that any state has followed Maine’s lead by also insisting upon Approach 3. Maine’s standard is of little consequence to the Companies because there are only 138 ANIC policyholders in Maine and zero PTNA policyholders. Ex. P-335 at 30, 31. Second, Pfannerstill used Approaches 1 and 2 to calculate a maximum allowable rate increase in February 2010 and did not mention Approach 3. Ex. R-550 at 1-3. His February analysis stated that under Approach 1, the maximum allowable rate increase for OldCo policies was 364% for PTNA and 334% for ANIC, and under Approach 2, the maximum was 276% for PTNA and 312% for ANIC. Id. at 1. The February analysis was certified by Pfannerstill as actuarially justified. N.T. 3/23/11 at 73. Finally, in September 2009, Milliman prepared projections for the aggregate rate increases needed to reach solvency by 2025 or 2050 under Scenario A in the September 2009 Surplus Report. Milliman projected necessary rate increases of 118% or 101% for ANIC, and 191% or 170% for PTNA. Ex. R-9 at Bates Nos. C002977-2978, C003018-3019.
Accordingly, the Court rejects the Reha-bilitator’s proposed finding that the maximum allowable rate increase for PTNA is 58% and for ANIC 121%. Notably, Pfan-nerstill did not opine that Maine (or any other state) could not be persuaded to use another methodology of calculating the 60% lifetime loss ratio in reviewing the rate increases requested in a plan of rehabilitation. N.T. 3/23/11 at 88.
Finally, the Court rejects Milliman’s premise that future rate increases will be sought on a “business as usual” basis. That will not be the case in a rehabilitation.
Pfannerstill and Lucker detailed the challenges to meaningful rate relief: “desk drawer”42 rules; artificial limits on the implementation of an approved rate increase; and rules limiting rate increases where, as here, a carrier no longer writes new business. These challenges are not insurmountable. No one knows how state insurance departments will react to rate increases that are part of an approved plan of rehabilitation. It may be possible to craft a rehabilitation plan with rate increases that meet some of the concerns of regulators, e.g., by limiting the amount of an annual rate increase in a given year or including a non-forfeiture benefit. Accommodations may be made in a particular state, but this is an unknown. A unified attack on OldCo’s rate structure has yet to be attempted. In some states, where the insurance departments have acted responsibly and have approved actuarially justified rate increases, rate increases may not even be needed as part of a rehabilitation plan.43
For these reasons, Milliman’s future premium rate increase assumption is too low. Future rate increases will be undertaken in conjunction with an approved rehabilitation plan. Energy in implementing the plan will be expected.
United Health’s rate increase assumptions were based upon what would be *435needed to allow the Companies to fund future claims. Volkmar assumed a variety of scenarios for different reinvestment interest rates and for different methods of implementation. Assuming an interest rate of 5.84%, PTNA could begin with a rate increase of 40%, followed by an annual rate increase of 10% for several years, then 9% until 2024. An alternative to reach the same point was a flat annual rate increase of 11% for 14 years. For ANIC, the 40% single increase could be followed by a second 19% increase and then stop. A flat increase for ANIC’s OldCo business would run at 6.8% for 14 years. Volkmar’s rate increase assumptions would require aggregate rate increases of over 300%. Ex. R-801 at 30.
The aggregate rate increase assumptions developed by Volkmar do not reflect the Companies’ experience over the last decade. They were developed to show the indicated rate need and the various ways that rate need could be achieved. They were appropriate for purposes of evaluating whether a rehabilitation is futile. Further, the Companies have never before submitted rate increase filings in the context of a court approved rehabilitation plan.
The Court rejects the Rehabilitator’s proposed finding that Volkmar “refused” to give an opinion that he was “reasonably certain” that his proposed rate increases were feasible. First, he was not asked to give such an opinion. Second, Volkmar testified that the rate increases were achievable based on his experience; he ran tests to see if they were reasonable; and he opined based on his experience and these tests that the actuarially justified rate increases were achievable and reasonable. N.T. 10/27/11 at 98-100. In his deposition, William Hunt, a director of PTAC and former CEO of PTNA and ANIC, testified that multiple, successive rate increases were done for the SHIP business.44 Ex. P-1034 at 104 (Interve-nors’ Counter-Designation).
Milliman’s criticisms of United Health’s rate increase assumption are not, for the most part, actuarial. Milkman targets political considerations that it believes Volk-mar did not consider adequately: the resistance of regulators to any rate increase, the expense of repeated rate filings and the possibility that regulators will require a non-forfeiture benefit. Volkmar’s rate increase scenarios would have to be submitted as part of a single rehabilitation plan. Accordingly, past resistance is not necessarily indicative of what will happen with respect to a rehabilitation plan that includes a paradigm for scheduled rate increases. If the rehabilitation fails, rate increases may be rolled back, where appropriate.
Milliman’s sole actuarial criticism of Volkmar’s rate increase assumption is that he did not consider shock lapses that would follow substantial rate increases. This criticism makes a good point. However, Milkman did not quantify the effect of shock lapsations upon Volkmar’s rate increase assumption. Further, Volkmar did a lapsation assumption, and this assumption may have included a shock lapse factor in response to rate increases. It *436cannot be discerned. Milliman’s criticism is not developed and, thus, not useful.
The Rehabilitator suggests that Volk-mar violated ASOP 18 by projecting “unreasonably optimistic assumptions.” First, ASOP 18 states that it applies to the preparation of rate filings and not to a surplus report used to determine whether a rehabilitation is futile. Second, ASOP 18 cautions equally against unreasonably pessimistic assumptions. The Court assigns no weight to ASOP 18. It cannot be used to fault Volkmar’s work as “optimistic” or, for that matter, to fault Milliman’s work as “pessimistic.”
The Court finds that Volkmar’s rate increase assumption may be high for use in a statutory financial statement, which is intended to be conservative. Volkmar’s rate increase assumption is not too high for its intended purpose, which is to show what could be done in a rehabilitation.
Milliman offered a morbidity improvement assumption of 1.5% for 10 years and 1% thereafter; Volkmar’s assumption was 2.0%. There are several reasons why the Court accepts United Health’s morbidity improvement assumption and rejects Milli-man’s.
The Litow Report did a detailed study using a series of “age-cost curves,” or vectors, to produce a 2.5% morbidity improvement assumption. (Pfannerstill) N.T. 3/22/11 at 111. Some vectors showed a morbidity improvement of 3.9%. By averaging the results, Litow reached the 2.5% assumption, which Pfannerstill described as slightly aggressive. Milliman did not take another look at Litow’s vectors. As Pfannerstill explained, “We changed that ... we just took that concept out. And in this scenario, we used a morbidity improvement factor of one and a half percent for ten years and one percent thereafter.” Id. Mohoric used a square root approach, which he could not explain as anything other than a device used to reach a bottom line number he chose as a matter of judgment. Notably, Mohoric “believes” in morbidity improvement and testified that he would be equally comfortable with a constant, not declining, morbidity improvement of 1.5%. N.T. 2/16/11 at 146.
Based upon his review of various studies, Volkmar exercised his actuarial judgment to develop his morbidity improvement assumption of 2.0%. The Re-habilitator’s criticism of Volkmar’s assumption is focused on the fact that Volkmar uses a 1.0% morbidity improvement assumption for another client, Cal-PERS, a self-insured long-term care program for California public employees. Indeed, the Rehabilitator points to Volk-mar’s Report wherein he states that the 1.0% assumption is “consistent with population experience and relevant actuarial documentation and practice.” N.T. 10/27/11 at 124. First, Volkmar adequately explained why he did not choose a 2.0% morbidity improvement assumption for CalPERS; namely it was chosen by the CalPERS board. Second, Volk-mar’s two different morbidity assumptions fall within a range that makes each consistent with the other.
Volkmar did not rely on Holland’s report to set his morbidity assumption. Nevertheless, Holland’s credited testimony also supports a morbidity improvement of 2.0%.
Milliman’s claim costs assumption increased the present value of future claims from $2.6 billion to $3.4 billion in a period of five months, finally settling on $4.1 billion in July of 2010. The assumption lacks credibility simply by virtue of the speed and size of its revision. These changes caused Robinson to lose confidence in Mil-liman’s numbers and to desire a second opinion. (Robinson) N.T. 2/14/11 at 239. Because he was “very nervous about the *437changes,” Robinson recommended to DiMemmo that another actuarial firm be retained. Id. at 287-38.
The Court agrees with Volkmar’s observation that in order for Milliman’s extreme changes to make any sense there had to have been an event or factor that impacted both claims in process and claims in the future. (Volkmar) N.T. 10/26/11 at 126-127. No such event or factor was identified, and no cogent explanation was offered.
The Court rejects Milliman’s claim costs assumption. First, it was developed by focusing on more recent experience, the period 2006 to 2008, when claims were higher than in the preceding period. Second, Milliman used incurred claims, which were mostly reserve and, thus, subject to change. Third, a significant portion of the claim projections are based upon low credibility data, ie., that generated by policyholders over 90 years of age and by policyholders that have held their policies for longer than 16 years.
As explained by Volkmar, Milliman cranked up the claim reserve based on additional claim experience and then focused on the periods with the cranked up claim costs, fitting the slope to the recent period. This significantly tilted the slope upward into the future. (Volkmar) N.T. 10/26/11 at 117-19. Then, Milliman fit projections onto data that are less credible, ie., for the older policyholders and policies with longer durations. Milliman adjusted the attained age slope to make it steeper by age; extended the selection factors; reduced morbidity improvement; and removed the wellness assumption. Volkmar called Milliman’s process one of leveraging incremental claims experience from a short period of time to project a huge change in projected claims over a long period of time. (Volkmar) N.T. 10/26/11 at 118-26. The Court agrees.
Milliman evaluated data going back to 2000, and Pfannerstill testified that “we took into account the experience ever since the inception of the data tape which is 1993.” N.T. 3/22/11 at 130. It is not clear what is meant by this “accounting.” Milli-man did not state that it used the earlier data to set the continuance curves. Milli-man focused on claims from the 2006-2009 period when claims were higher than in preceding periods, and the Rehabilitator acknowledges this focus. Rehabilitator’s Reply Brief at 95, ¶ 232.
United Health’s claim costs assumption used a claim costs slope consistent with Scenario A in Milliman’s October 2009 Report. Volkmar used selection factors from Scenario B of that report, and he removed them after 10 years because the effects of underwriting wear off. Volkmar used paid claims, not incurred claims, and he used data going back to 1993. Volkmar’s claim costs curves were consistent with those of other United Health clients, and his claim costs assumptions tested well.
The Rehabilitator is critical of Volkmar because he was “uninformed” when preparing his expert report. Rehabilitator’s Proposed Findings of Fact at 103-04, ¶ 506. Specifically, the Rehabilitator asserts that Volkmar was not aware of pre-rehabilitation discussions between the Department and the Companies; the details of the Companies’ 2000 Corrective Action Plan; and the repeated instances of reserve strengthening. The Rehabilitator also challenges Volkmar’s failure to use incurred claims data from 2008 and 2009. The Court rejects these criticisms.
First, Volkmar repeatedly testified that the events cited by the Rehabilitator did not cause him to revisit his actuarial opinions. N.T. 10/27/11 at 154-55, 235-36. Prior history was not relevant because he was working with current data. Second, *438this history was outside the scope of his expert report, which was to review and criticize Milliman’s actuarial work. Third, Volkmar used data available through year-end 2009. N.T. 10/27/11 at 107-08. He did not overlook the incurred claims data for 2008 and 2009 but, as explained above, chose not to use it because the majority of the 2008 and 2009 data represented reserves, which are estimates. Instead, Volkmar used paid claims data for 2008 and 2009.
A speedy return to positive statutory surplus was not the object of United Health’s projections. Volkmar explained that his projections were done for the purpose of ensuring that PTNA “had enough cash to meet obligation[s] ... and that would indicate that the surplus was going to remain negative throughout the projection period given that definition.” N.T. 10/27/11 at 88.
This brings the Court to Milliman’s gross premium reserve, which enlarged the apparent deficit and, thus, increased the surplus deficiency as of December 31, 2009. David Minches, of Ernst & Young, testified that Milliman’s gross premium reserve was greater than required in various states. N.T. 3/24/11 at 151; N.T. 3/25/11 at 59-60. Accordingly, Milliman’s gross premium reserve held the Companies to a higher standard than necessary to comply with minimum statutory reserve standards. Ex. P-53 at 6.
The Court rejects Milkman’s gross premium reserve, characterized by Mohoric as resulting from “conservatism,” because it is excessively conservative and punitive. It has no bearing on the Companies’ ability to pay claims today or in the near future. (Volkmar) N.T. 10/26/11 at 143.
Mohoric knew that the decision had been made to pursue liquidation when he drafted Milliman’s September 2009 Surplus Report. (Mohoric) N.T. 2/23/11 at 114-15, 130-31. In this report’s initial draft, Mohoric wrote that the purpose of the report was “to support the department’s recommendation to liquidate.” Id. at 118-19; Ex. R-91 at 2. On advice of the Rehabilitator’s counsel, Mohoric restated the purpose as helping the Rehabilitator “determine whether rehabilitation of PTNA and ANIC is feasible or whether one or both of the companies should be liquidated.” Ex. R-91 at 2; Ex. R-1125; (Volkmar) N.T. 10/26/11 at 134-36. The purpose of Milliman’s surplus reports is apparent from its contents.
Milliman’s projections were not developed for the purpose of developing a workout plan to address the future claims of the policyholders in this closed book of business. Workout concepts were identified by Pfannerstill, in his article in Long-Term Care News, a publication of the Long-Term Care Insurance Section of the Society of Actuaries. Ex. R-963 at 11-13. Pfannerstill recommended that policyholders consider changing their benefit design to avoid drastic rate increases that were being necessitated as the industry learns more about how long-term care insurance claims develop. The actuarially justified rate increases may be so high that policyholders will be motivated to let their coverage lapse. But the lapsed coverage might provide more than is reasonably necessary for a particular policyholder. Pfannerstill recommended that policyholders be advised about benefit options and discouraged from accepting a non-forfeiture benefit, which would refund all premiums paid to date but deprive the policyholder of the future risk protection provided by a long-term care insurance policy.
Instead of focusing on a workout, Milkman focused on statutory surplus. However, as explained by Volkmar and conceded by Mohoric, a cash flow analysis that shows when, and if, cash flow deficits *439-will occur during the rehabilitation period is the analysis relevant to a rehabilitation. This was not done by Milliman. One of the most troubling aspects of Milliman’s July 2010 Surplus Report lies in its very low premium rate increase assumption. Unlike claims or interest rates, which cannot be controlled, rate increases can be controlled. Milliman’s foot-dragging rate increase assumption appears to have been calculated to support the Rehabilitator’s decision to seek a liquidation.
Intervenors’ actuary opined that a rehabilitation, even one based only on premium rate increases, was not futile from an actuarial standpoint. (Volkmar) 10/25/11 at 135-36. The Rehabilitator’s actuary did not testify that there was an actuarial reason why successive rate increases would not work. Pfannerstill, on questioning of the Court, acknowledged that successive rate increases do not have the tendency to lead to a death spiral and collapse of business as they do in health insurance.45 N.T. 3/24/11 at 119-20. In spite of the bad claims news presented by Milli-man in August 2009, Milliman itself did not opine that a rehabilitation was futile or advise the Rehabilitator that the only choice was a liquidation.
The actuaries who testified before the Court are national experts. The work of all the actuaries has merit in varying degrees. Volkmar’s expert report and testimony were more persuasive, particularly in his responses to Milliman’s critiques. Further, Volkmar focused on the ultimate question in this proceeding, which is to determine whether, with or without benefit design changes, future premium revenue can be increased at the pace needed to pay future claims. The actuaries’ opinions were, in large part, the product of the exercise of actuarial judgment, and the Court finds Volkmar’s explanations for how he exercised his judgment more compelling.
IV. Legal Analysis
A. Statutory Standard for Conversion of a Rehabilitation to a Liquidation
Article V specifies the procedure and grounds for the conversion of an insurance company rehabilitation into a liquidation. It states, in relevant part, as follows:
Whenever he has reasonable cause to believe that further attempts to rehabilitate an insurer would substantially increase the risk of loss to creditors, policy and certificate holders, or the public, or would be futile, the rehabilitator may petition the Commonwealth Court for an order of liquidation. A petition under this subsection shall have the same effect as a petition under section 520. The Commonwealth Court shall permit the directors to take such actions as are reasonably necessary to defend against the petition and may order payment from the estate of the insurer of such costs and other expenses of defense as justice may require.
Section 518(a) of Article V, 40 P.S. § 221.18(a) (emphasis added).46 Section 518(a) governs the instant proceeding.
*440In Koken v. Legion Insurance Co., 831 A.2d 1196 (Pa.Cmwlth.2003), aff'd sub nom. Koken v. Villanova Insurance Co., 583 Pa. 400, 878 A.2d 51 (2005), this Court considered the meaning of Section 518(a) and explained that:
In any petition to liquidate filed under Section 518(a) of Article V, the rehabili-tator has the burden of proof. As the moving party, it is the rehabilitator’s burden to prove insolvency as of the date the petition for liquidation is filed. Commonwealth Ins. Dep’t v. Safeguard Mut. Ins. Co., 18 Pa.Cmwlth. 195, 336 A.2d 674 (1975). The rehabilitator’s next burden is to demonstrate that continued rehabilitation would “substantially increase the risk of loss to creditors, policy and certificate holders, or the public, or would be futile.” Section 518(a), 40 P.S. § 221.18(a).
Id. at 1230.47 In response to the Rehabili-tator’s suggestion that the Court should defer to the statutory rehabilitator with respect to the conversion decision, this Court stated:
Deference is not appropriate where, as here, the Court must apply specific statutory standards to the evidence presented.... To apply deference to the job of factfinding would undermine this Court’s responsibility to act upon the Rehabili-tator’s petition in a fair and neutral manner. Further, to apply the deference standard as proposed by the Reha-bilitator would shift the burden of proof, improperly, to those opposing a petition to liquidate.
Id. at 1232. This holding is consistent with other states. See, e.g., LaVecchia v. HIP of New Jersey, Inc., 324 N.J.Super. 85, 734 A.2d 361, 364 (Ch.Div.1999) (rejecting commissioner’s argument that a liquidation order should be entered absent a showing that commissioner’s determination to seek liquidation is unreasonable or arbitrary).
Liquidation is a remedy of last resort. The rehabilitator may not file a liquidation petition unless he has “reasonable cause to believe” that one of the two elements of the liquidation standard is met. If he files a petition for liquidation, the rehabilitator bears the burden of proof. The rehabilitator first must prove insolvency as of the date the petition for liquidation was filed. Next, the rehabilitator must demonstrate that continued rehabilitation would either (i) “substantially increase the risk of loss to creditors, policy and certificate holders, or the public,” or (ii) “be futile.” Legion, 831 A.2d at 1230.
The Rehabilitator has admitted that he has the burden of establishing that one of the prongs of the liquidation standard has been satisfied before a liquidation order may be entered:
In Legion, this Court held where the Rehabilitator seeks to place an insurer *441which is under rehabilitation into liquidation, the Rehabilitator must prove a) that the insurer is insolvent and b) that there is reasonable cause to believe either (i) that continued rehabilitation efforts would substantially increase the risk of loss to creditors, policy and certificate holders, or the public or (ii) that continued rehabilitation efforts would be futile. The burden is on the Rehabili-tator to prove that the requirements of [Section 518(a)] for conversion from a rehabilitation are met and the decision whether those elements have been shown is for the Court. It is therefore the Rehabilitator’s burden to prove that the insurer is insolvent and that one of the two alternative elements of risk of substantial harm or futility is met.
Rehabilitator’s Memorandum of Law in Opposition to Intervenor’s Motion to Exclude Argument Concerning Liquidation Standard, November 16, 2010, at 3-4 (emphasis added) (internal quotations and citations omitted).
B. Insolvency
Section 503 of Article V defines “insolvency” as follows:
[T]he inability [of an insurer] to pay its obligations when they are due, or whose admitted assets do not exceed its liabilities plus the greater of (i) any capital and surplus required by law for its organization, or (ii) its authorized and issued capital stock.
40 P.S. § 221.3.48 A company’s assets minus its liabilities is referred to as its surplus; a negative surplus is the amount by which its liabilities are greater than its assets and a positive surplus is the amount by which assets exceed liabilities. (Mohoric) N.T. 2/18/11 at 161-62. Here, it has been established that PTNA and ANIC are meeting their obligations as they come due. Neither company has any debt, and both companies will be able to pay claims when presented for many years, even without future rate increases. However, PTNA and ANIC are insolvent by the second test, i.e., liabilities exceed assets.
PTNA and ANIC are required by law to have minimum capital and surplus of at least $1.5 million. Section 206(a) of the Insurance Company Law of 1921, Act of May 17, 1921, P.L. 682, as amended, 40 P.S. § 386(a). The parties disagree by how far the Companies deviate from the minimum surplus of $1.5 million. They agree, however, that the Companies’ “admitted assets do not exceed its liabilities plus the greater of (i) any capital and surplus required by law for its organization, or (ii) its authorized and issued capital stock.” Section 503 of Article V, 40 P.S. § 221.3. Thus, the Companies are insolvent.
Given the insolvency of the Companies, the next question is whether continued rehabilitation will substantially increase the risk of loss to policyholders or is futile. The Rehabilitator’s evidence was focused on the futility of rehabilitation, not upon the claim that continued rehabilitation substantially increases the risk of loss to policyholders and creditors. The Court will consider the standards for conversion in order of their appearance in Section 518(a), not the order in which the Rehabili-tator presented his evidence.
C. Substantial Increase in Risk of Loss to Policyholders
The Rehabilitator argues that a rehabilitation will substantially increase the risk of loss to policyholders because a continued rehabilitation may deplete assets. As a consequence, there may be reduced assets available for distribution to guaran*442ty funds; there may be reduced assets available for policyholder claims that exceed guaranty fund limits; and policyholders presently on claim may receive more than future claimants.
These arguments rest on the factual premise that assets will decline in a rehabilitation. The evidence in support of this premise consists of two paragraphs in Mil-liman’s Rebuttal Report, which states that rehabilitation may dissipate assets, leaving “less — perhaps no — distribution of estate assets for the value of the policyholder’s claim against the estate in excess of the Guaranty Association (GA) limits.” Ex. P-969 at 11. The report directs the reader to Attachment E, which presents a series of tables that show a depletion of the Companies’ $1 billion in assets by 2020. Intervenors challenged the accuracy of Attachment E because it seriously understated the actual value of the assets and assumed zero rate increases in the future. Intervenors’ Brief at 231-37. At trial, Mo-horic acknowledged that he himself no longer had confidence in the numbers in Attachment E projected for 2012. N.T. 11/1/11 at 109. At different points in this proceeding, the Rehabilitator has suggested a different asset depletion date. In the liquidation petition the suggested date of asset depletion was 2025. Amended Petition for Liquidation at 7, ¶ 20. Given Mo-horic’s admission that Attachment E would not be accurate for 2012, it does not support an asset depletion date of 2020. The Rehabilitator’s argument that assets will be depleted in a rehabilitation is tautological because it assumes the rehabilitation will fail.
Even assuming no future rate increases, assets will not be depleted for years and, in any case, later than 2020. On that factual premise, the Court considers the Rehabilitator’s argument that continued rehabilitation will substantially increase the risk to policyholders.
As noted by the Rehabilitator, an order of liquidation is needed before policyholders can become eligible for guaranty fund coverage. However, the Rehabilitator has not shown a need for immediate transfer of policyholders to guaranty funds. In the Legion case, there was a crisis because Legion lacked the cash flow needed to pay claims, which had caused a claims backlog in excess of $162.5 million. Legion, 831 A.2d at 1205. Automobile insurance policyholders in California, for example, could not get their vehicles out of repair shops because Legion was not paying their claims. The crisis grew daily, and a delay in liquidation substantially increased the risk of loss, i.e., unpaid claims, to policyholders.
Here, by contrast, the claims of PTNA and ANIC policyholders are being paid on a timely basis and will continue to be paid for many years. There is no crisis.
Whether a liquidation takes place now or in 10 years matters not to policyholders. The guaranty funds -will still be there. In fact, the policyholders may have greater guaranty fund protection in ten years. Since 2009, some states have increased their guaranty fund limits, and there is no history of any state ever reducing guaranty fund limits.
The Rehabilitator’s argument that rehabilitation, if unsuccessful, may reduce what guaranty funds will receive in a liquidation is irrelevant. Protection of guaranty funds is not a standard in Article V for converting a rehabilitation to a liquidation. They are future creditors and only the present creditors are to be considered in a Section 518(a) proceeding. Assuming, ar-guendo, it were relevant to consider the interests of guaranty funds, they will benefit from a continued rehabilitation to the extent the rehabilitation increases rates *443and eliminates the subsidies between policyholders in different states and between different groups of policyholders. Stated otherwise, properly priced policies are subsidizing underpriced policies. The policyholders, if and when they are transferred to a guaranty fund, will already be paying a more realistic rate for their long-term care insurance, and that more realistic rate will transfer to their replacement insurer provided by the guaranty fund. Rates can be rolled back, where appropriate, if they are too high for the reduced benefit levels that will be provided by a guaranty fund in a particular state.
This leads to the Rehabilitator’s companion argument that continued rehabilitation may diminish the assets available to pay claims in excess of guaranty fund coverage. Many policyholder claims in the estates of Legion Insurance Company and Reliance Insurance Company fall into that category; however, there is a world of difference between those cases and this one. All of the extra-guaranty fund claims in the Legion and Reliance liquidations arose before the policies were cancelled by the liquidation order. Here, the policyholder claims for amounts in excess of guaranty fund limits may not arise for many years after their policies are terminated. There are several problems with the Rehabilitator’s novel argument that claims arising after a policy has been terminated in a liquidation may be presented to the insurer estate.
Under Article V, an order of liquidation cancels insurer’s policies and fixes the policyholders’ rights in accordance with Article V. The Rehabilitator believes that one of the rights so fixed would be the policyholder’s right to bring a common law breach of contract claim caused by the termination of a policy. Contractual damages would be measured by the lost value of the cancelled policies.
In support, the Rehabilitator directs the Court to several cases, beginning with Commonwealth ex rel. Kirkpatrick v. American Life Insurance Co., 162 Pa. 586, 29 A. 660 (1894). In this case, the Pennsylvania Supreme Court held that where a life insurance company became insolvent, its policyholders could pursue a claim against the estate for loss of their life insurance policies, as valued on the date of liquidation.49 The Supreme Court defined this value as
the net value of the policies, without regard to the health of the holder, and calculated as of the date of the dissolution of the corporation, according to the tables of mortality used in the business of life insurance, less the outstanding premium notes.
Id. at 589, 29 A. at 661. The Companies’ long-term care insurance policies are guaranteed renewable non-eancellable policies, which means that the Companies cannot refuse to accept timely tendered renewal premium. 2 Couch on Insurance 3d, Cancellation and Rescission Principles, § 30:15. The Rehabilitator believes that this feature of the Companies’ policies places their policyholders in a position similar to that of the life insurance policyholders in American Life. Notably, American Life was a common law receivership brought by the Attorney General, and it was in that context that our Supreme Court recognized the policyholders’ breach of contract action.
The Rehabilitator also points to a ruling of the New Jersey Supreme Court in In re Liquidation of Integrity Insurance Co., *444147 N.J. 128, 685 A.2d 1286 (1996). There, the court held that a surety bond holder could present a claim for losses that occurred after the date the bonds were terminated by the liquidation. Id. at 1292. It held that the termination of the surety bonds constituted a breach of contract, which entitled the policyholders to more than a premium refund. The court ordered a complicated series of remedies peculiar to surety, one of which allowed bondholders to seek damages from the estate in an amount equal to the cost of replacement coverage. Notably, surety bonds are excluded from coverage by New Jersey’s guaranty funds. See N.J. Stat. Ann. § 17:30A-2(b) (excluding surety bonds from the New Jersey Property-Liability Insurance Guaranty Association Act) and N.J. Stat. Ann. § 17:22-6.71 (excluding surety bonds from the New Jersey Surplus Lines Insurance Guaranty Fund Act).
There was a strong dissent in Integrity Insurance. The dissent argued that “[t]he rights of claimants against an insolvent insurer are governed by statute,” which did not permit claims against the estate for claims that arise after the surety bond was terminated. Integrity Insurance, 685 A.2d at 1295 (Handler, J. dissenting). The dissent concluded that the governing insurer insolvency statute limited claims against the insurer’s estate to claims for unearned premiums and the cash surrender value of policies. And nothing else.
Another case cited by the Rehabilitator, Caminetti v. Manierre, 23 Cal.2d 94, 142 P.2d 741 (1943), involved the termination of non-cancellable disability policies in liquidation. The liquidator argued that the policyholders had no claim other than for return of unearned premium. The liquidator reasoned that
whether or not damages will be suffered depends upon contingencies which may never occur such as that the insured may never become disabled during the term of his policy, that he will keep the premiums paid, and that he may die before the term of his policy expires.
Id. at 745. The California Supreme Court rejected this logic. It concluded that policyholders would be damaged by their loss of coverage even if “the majority of [those] policyholders would in the ordinary course of events pay their premiums but never receive disability payments.” Id. at 746. Accordingly, it held that policyholders not on claim had the right to recover a proportionate share of the insurer’s assets upon liquidation. At the time of this holding, guaranty fund coverage had not yet been established for any line of business in California.
The Rehabilitator also directs the Court to Commissioner of Insurance v. Massachusetts Accident Co., 314 Mass. 558, 50 N.E.2d 801 (1943). This case concerned policyholders with non-cancellable disability policies, who were not on claim when their insurer was liquidated. The Supreme Judicial Court of Massachusetts held that the policyholders had a claim for the lost value of their policy, explaining that the
policyholders have been deprived of their contracts by liquidation of the company. In effect the company has completely broken its contracts of insurance with its policyholders and is no longer ... giving them protection.
Id. at 805. Following the United States Supreme Court’s holding in Carr v. Hamilton, 129 U.S. 252, 9 S.Ct. 295, 32 L.Ed. 669 (1889), the court held that the measure of value for their policy, a level premium type policy, was the difference the policyholders would have received under their policies if their policies had not been can-celled and the premiums they would have had to pay for that coverage.
*445There are several problems with the Rehabilitator’s reliance on this above-reviewed line of precedent. Except for Integrity Insurance, they are ancient. They pre-date any insurer insolvency statute, or at least the modern version developed by the NAIC Model Act, which Pennsylvania has adopted. All concerned policyholders without guaranty fund protection, either because they were exempt from guaranty funds, as in the case of surety bond holders, or because guaranty funds had not yet been invented.
The precedent relied upon by the Reha-bilitator establishes that where there is no guaranty fund protection, it is inequitable to limit a policyholder’s claim against an insolvent insurer’s estate to a refund of premium.50 In Massachusetts Accident, the Supreme Judicial Court of Massachusetts explained the inequity: it would create a windfall for the insurance company in receivership. The court reasoned as follows:
It would be an anomaly if an adjudication of insolvency should itself have the effect of restoring the company to a sort of solvency through the immediate elimination of one of the principal blocks of its liabilities.
Massachusetts Accident, 50 N.E.2d at 805. Accordingly, the court fashioned a way to ensure that estate assets went to policyholders, not to stockholders, by giving them the lost value of their policies. This enabled them to purchase replacement coverage.
In the absence of Article V, that anomaly could occur here. By terminating all policies, a liquidation would erase the Companies’ obligation to renew their non-cancellable policies; this would leave only the Companies’ liability for incurred claims that arose before the policy terminations. Assets are more than sufficient to fund incurred claims. By shedding the liabilities of future, yet to develop, claims, the Companies would be left with substantial assets to use for their own purposes. However, there can be no windfall to the stockholders here because in a liquidation the Companies’ policyholders would be transferred to guaranty funds after their policies terminated. By statute, those guaranty funds then become creditors against the estate of the Companies.
There are other problems with the Re-habilitator’s argument that policyholders may have a breach of contract action to recover extra-guaranty fund claims. Most problematic is the absence of direct language in Article V creating such a right. Intervenors note that because only claims “under 'policies for losses” have the priority of Class B claimants, the extra-guaranty fund claim suggested by the Rehabilitator would be a general creditor claim with a Class E priority. Intervenors also note that guaranty funds are creditors of the estate to the extent of “[ajssets attributable to covered policies.” Section 1712(c) of the Insurance Company Law of 1921, Act of May 17, 1921, P.L. 682, as added by the Act of December 18, 1992, P.L. 1519, 40 P.S. § 991.1712(c). Finally, Interve-nors point to this Court’s holding in Foster v. Colonial Assurance Co., 668 A.2d 174 (Pa.Cmwlth.1995), aff'd, 543 Pa. 626, 673 A.2d 922 (1996).
In Colonial Assurance, a policyholder sought coverage for the full amount of a loss that occurred two years after the liquidation order terminated his policy. Id. at 184. This Court held that a claim for a loss that occurred two years after a casual*446ty policy was terminated in a liquidation was barred by Article V. It held that policyholder rights are fixed as of the date of liquidation and that Section 521 of Article V, 40 P.S. § 221.21, “makes no distinction between cancelable and non-cancelable policies but applies to all risks in effect at the time the liquidation order is entered.” Id. The Department has continued to take this position in other insurance company estates.51 Here, losses that occur more than 30 days after the liquidation would be covered by the replacement guaranty fund coverage; they would not be losses under a PTNA or ANIC policy.
In short, the Rehabilitator’s breach of contract theory faces obstacles. It may or may not be established. That it may succeed does not support the Rehabilitator’s argument that continued rehabilitation substantially increases the risk of loss to policyholders.
Finally, the Court addresses the Rehabilitator’s concern that policyholders presently on claim will fare better than other policyholders unless all policies are immediately terminated and picked up by guaranty funds. Simply, our Supreme Court has held that paying policyholder claims during a rehabilitation does not create a preference. Ario v. Ingram Micro, Inc., 600 Pa. 305, 309, 965 A.2d 1194, 1196 (2009).
The Rehabilitator has not established that a rehabilitation, even one focused solely on rate increases, will not work. Rate increases will only benefit the estate regardless of the ultimate success of the rehabilitation. Although the course of rehabilitation is not certain, there is no evidentiary basis for holding that it will substantially increase the risk to the Companies’ policyholders. They will receive guaranty fund coverage whether liquidation is ordered now or in ten years. It is not a rehabilitation that increases the risk of loss to policyholders but, rather, a liquidation, which will cap policy benefits at the guaranty fund limits and deprive the policyholders of any choice in the matter.
D. Futility of Continued Rehabilitation
i. Meaning of “Futility”
The term “futile” is not defined in Article V, nor has it been defined by the courts of this Commonwealth. The commonly accepted definition of “futile” is
1: Serving no useful purpose: Ineffective, Fruitless ...
2: Occupied with trifles: Frivolous.”
WebsteR’s Third New INTERNATIONAL Dictionary at 925 (2002 ed.).52 In a Section 518(a) conversion petition, the rehabilitator bears the burden of proof. Kolcen v. Legion Insurance Co., 831 A.2d 1196, 1230 (Pa.Cmwlth.2003). Thus, to prove that continued rehabilitation is futile, the Reha-bilitator has to prove that further rehabilitation would serve no useful purpose.
This Court examined the futility standard in Legion, but it did not apply that standard to the rehabilitator’s Section 518(a) petition. Instead, the Court held that the rehabilitator had satisfied her burden of proof on the alternate ground for converting a rehabilitation to a liquidation, i.e., that further attempts to reha*447bilitate Legion would have substantially increased the risk of loss to policy and certificate holders. The Court so held because policyholders’ claims were going unpaid and increasing daily. With respect to futility, the Court cautioned that
[i]f the Insurance Department obtains consent to a rehabilitation, it has a responsibility to move directly on a plan that will correct the insurer’s problems .... A rehabilitation is not to be used as a period of conservatorship while the Insurance Department reviews the options.
Id. at 1244. The Court reasoned that the condition precedent to an application of the futility standard was a plan of rehabilitation, noting that “[o]nly with the filing of a plan, even in the most preliminary form, is it possible to track progress and establish the plan’s viability or its futility.” Id. at 1245.
Here, the Rehabilitator filed a Preliminary Plan. However, a formal rehabilitation plan never moved beyond the discussion stage, so there are no benchmarks by which to track the viability of continued rehabilitation.
Instead, the Rehabilitator did exactly what this Court in Legion stated that a rehabilitator should not do: he has treated the rehabilitation as a conservatorship to give him time to prepare for liquidation. The Rehabilitator stopped making rate increase filings over two years ago. He never drafted a formal rehabilitation plan, let alone present one to the Court. Without a formal plan of rehabilitation, the Rehabilitator cannot make the case that a plan he never proposed or implemented is futile.
The Court recognized in Legion that there are extreme circumstances where futility might be established even without the adoption of a formal plan of rehabilitation. The Court contemplated circumstances where “the insurer’s finances are in total disarray, loss reserves against claim liabilities are drastically understated and investments chimerical.” Id. In other words, the Department must prove, as a matter of fact, that an insolvent insurer’s immediate financial circumstances are in such disarray that they are completely un-salvageable. See, e.g., Sheppard v. Old Heritage Mutual Insurance Co., 492 Pa. 581, 425 A.2d 304 (1980). The Rehabili-tator argues that this is one of those extreme circumstances.53 For the reasons set forth below, the Court disagrees.
*448ii. Rehabilitator’s Case for Futility “The Hole”
In arguing that continued rehabilitation of the Companies is futile, the Rehabili-tator focuses on what he terms is a “drastic understatement” of the Companies’ reserves. Rehabilitator’s Brief at 40. According to the Rehabilitator, at the time Intervenors broached the subject of placing the Companies in rehabilitation in late 2008, they represented that PTNA had a negative surplus in the range of $100 million to $141 million; that ANIC was solvent; and that PTNA’s negative surplus was attributable to the recapture of the Imagine reinsurance treaty and would be curable by increasing aggregate premiums by 10% to 15% and cutting agent commissions. Rehabilitator’s Brief at 1. Even after the Companies were placed in rehabilitation, the Rehabilitator believed that PTNA had a negative surplus of approximately $226 million and that ANIC was solvent. Because of Milli-man’s revised projections, the Rehabili-tator now believes that PTNA has a negative surplus of $2.1 billion and ANIC has a negative surplus of $137 million. The Rehabilitator notes that Intervenors’ actuary agrees that the Companies are under-reserved, albeit to a lesser extent.54
The Rehabilitator argues that the Companies’ reserves are “severely understated and its insolvency so severe that rehabilitation is unlikely to succeed.” Rehabilitator’s Brief at 40 (quoting Koken v. Legion Insurance Co., 831 A.2d 1196, 1245 (Pa.Cmwlth.2003)). This argument is premised on an acceptance of Milliman’s current reserve calculations, which have fluctuated violently in a short period of time and have been rejected by the Court. Nevertheless, even under those rejected projections, there is no threat to the Companies’ ability to pay claims now or for many years into the future.
The Rehabilitator directs the Court to cases where courts have found continued rehabilitation to be futile. In Minor v. Stephens, 898 S.W.2d 71, 81 (Ky.1995), the Supreme Court of Kentucky found that rehabilitation of a life insurer was futile because of its cascading liabilities. The insurer could not pay the surrenders, which had reached $1 million a day, and it could not pay ongoing obligations. The Rehabilitator also points to In re New York Title & Mortgage Co., 156 Mise. 186, 281 N.Y.S. 715 (N.Y.Sup.Ct.1935), a Depression-era case where the insurer was in default on over $2 million in guaranteed interest and had serious cash flow problems. Here, by contrast, there is no “run on the bank” or immediate cash flow problems. The cases do not support the conclusion that the Rehabilitator should be excused from devising a rehabilitation plan.
Accordingly, the instant case is one where a legitimate effort to devise a rehabilitation plan is required. It is too early to declare defeat. Aggressive efforts to *449pursue actuarially justified rate increases and contract modification options, for policyholders and agents, all hold potential.
The Rehabilitator focuses on three methods of rehabilitating the Companies and argues why each is futile: (1) a series of aggregate rate increases; (2) benefit caps; and (3) transfer of PTNA NewCo policies to ANIC with rehabilitation of ANIC and hquidation of PTNA.
Rate Increases
There are cases where a rate increase strategy would be futile in rehabilitation. It would have been futile in Mutual Fire, where the insurance company was not writing or renewing insurance policies. See Foster v. Mutual Fire, Marine and Inland Insurance Co., 531 Pa. 598, 614 A.2d 1086 (1992) (Mutual Fire II), affirming sub nom., remanding in part, Grade v. Mutual Fire, Marine and Inland Insurance Co., 132 Pa.Cmwlth. 196, 572 A.2d 798 (1990) (Mutual Fire I). The company had claim liabilities of $260.1 million and assets of $99.4 million, and there was no way to address the surplus deficiency by rate increases. Mutual Fire I, 614 A.2d at 1089. Once a pohcy has terminated, it is impossible to impose an after-the-fact rate surcharge on policyholders whose rates have been fixed for the pohcy year and already been paid. Accordingly, Mutual Fire was rehabilitated by reducing claims payments.
Here, by contrast, rate increases are not futile because PTNA and ANIC can impose rate increases at pohcy renewal. The policies provide that rate increases cannot be imposed by reason of the pohcyholder’s age or deteriorating health, but rate increases are expressly authorized where the rates on a particular product are inadequate.
According to Milliman’s July 2010 Surplus Report, additional total premiums of 210% and 158%, respectively, are required to bring PTNA and ANIC to solvency by 2025. Ex. P-961 at 14; (Pfannerstill) N.T. 3/23/11 at 5-6. Those percentages drop to 192% and 143%, respectively, if the goal is to achieve solvency by 2050. Id. The foregoing percentages are in addition to those in Milliman’s rate increase assumption. According to Milhman, if all of the premium rate increases were to be obtained at once, on January 1, 2010, and without making adjustments for shock lapses and anti-selection, increases of 280% and 244% would be needed to bring PTNA and ANIC to solvency by 2025. Ex. P-961 at 14. To achieve solvency by 2050, those percentages would be 257% and 221%, respectively. Id.
The Rehabilitator argues that these projected rate increases far exceed that which can reasonably be obtained. In the decade preceding the Companies’ formal rehabilitation, the states did not grant lower and actuarially justified rate increases, so they certainly will not approve higher rate increases now. In other words, the Rehabil-itator’s case for liquidation is based upon the premise that states will refuse to carry out their statutory obligations to approve actuarially justified rate increases. What is stunning about this argument is that it includes Pennsylvania, where the Companies are domiciled. In effect, the Rehabili-tator asks this Court to assume that states will not fulfill their statutory duty to act responsibly on rate filings.55
In support of this contention, the Reha-bilitator offered testimony that the rate increases needed on the OldCo business exceed the level that state insurance regu*450lators have approved in the past. The Rehabilitator also notes that the Companies need rate increases in states that are particularly unwilling to approve substantial rate increases.
For example, PTNA needs large rate increases in Florida, where 13% of OldCo policyholders reside. Florida law, however, limits the amount of obtainable rate increases for any company that is not writing new business. (Pfannerstill) N.T. 3/22/11 at 183-85; (Volkmar) N.T. 10/27/11 at 93-94.56 Neither PTNA or ANIC is writing new business. Pfannerstill testified that other states, including the important states of California, New Jersey and Connecticut, employ “desk drawer” rules57 and other regulatory roadblocks to actuarially justified rate increases, such as a large loss ratio on a historical basis. N.T. 3/22/11 at 182-83,186-90.
The Court has rejected Milliman’s rate increase assumption as too low and has rejected Milliman’s statement that PTNA’s maximum allowable rate increase is 58% and that ANIC’s is 121%. Because the Court has not accepted Milliman’s projected claim costs, it does not accept the premium rate increases Milliman says are needed to cover future claims. Accordingly, the Court rejects the Rehabilitator’s argument that the rate increases needed are too high to be achievable and, thus, futile.
Further, the Court has rejected Luck-er’s proffered opinion that the rate increases needed for the OldCo non-tax qualified policies are beyond reach. The Court has credited Volkmar’s testimony that the rate increases he believes are necessary for the OldCo business, while high, are reasonable, based on his experience.
The Rehabilitator describes the past decade of several rounds of rate filings as “rehabilitation.” What management can do on its own initiative is a “correction,” but it is not a rehabilitation undertaken pursuant to Article V, using all the tools that Article V provides. None of the experts stated what level of rate increases are obtainable in a true rehabilitation of a long-term care insurance company, especially if such rate increases are sought in tandem with other devices, authorized in a formal rehabilitation, such as modification to policy designs.
Benefit Caps
The Rehabilitator argues that a rehabilitation of the Companies that involves revising policy designs and benefit options is a futile exercise. The Rehabili-tator contends that any rehabilitation proposal that would make any policyholder worse off than he or she would be in liquidation cannot be considered or adopted as a feasible alternative to liquidation. In support, the Rehabilitator cites *451our Supreme Court’s holding in Mutual Fire II for the proposition that a rehabilitation must treat policyholders and other creditors at least as well as they would fare in liquidation. The Rehabilitator posits that reducing maximum benefits to $350,000 or to a four or five year benefit period would reduce the coverage of some policyholders below what they would receive in liquidation because their state’s guaranty association limits exceed the reduced benefits. The Rehabilitator also argues a plan of rehabilitation that includes a modified benefit design is not legally permissible and, thus, rehabilitation is futile.
At issue in Mutual Fire II was a plan of rehabilitation filed by the statutory reha-bilitator under Section 516(d) of Article V, 40 P.S. § 221.16(d).58 The plan was challenged by reinsurers, policyholders and creditors on the grounds that the proposed plan was, in reality, a liquidation, not a true rehabilitation. The Court considered numerous objections to the plan by policyholders and other creditors, such as rein-surers, with claims against Mutual Fire. For example, guaranty fund protection was available to a limited number of policyholders, who favored a liquidation so that they could access guaranty fund benefits.
In the present case, the Rehabilitator has not submitted a rehabilitation plan. This is a crucial distinction because a central concern in Mutual Fire II was whether the Insurance Commissioner had abused her discretion in formulating the rehabilitation plan and whether it was inequitable. Here, the Court does not have a rehabilitation plan to consider.
In any case, Mutual Fire II does not stand for the proposition that every single policyholder, or other creditor, must fare as well in rehabilitation as in liquidation. Mutual Fire II stands for the opposite. The Supreme Court held that the goal of rehabilitation is the equitable administration of “the assets of the [insurer] in the interests of investors, the public, the main purpose being the public good.” Mutual Fire II, 531 Pa. at 613, 614 A.2d at 1094. Further, “the goals of rehabilitation necessitate the reality that ‘individual interests may need to be compromised in order to avoid greater harm to a broader spectrum of policyholders and the public.’ ” Id. (quoting Vickodil v. Commonwealth, Insurance Department, 126 Pa.Cmwlth. 390, 559 A.2d 1010, 1013 (1989)).
By the Rehabilitator’s logic, the existence of state guaranty funds makes any rehabilitation a non-option. Because state guaranty fund coverage ranges from $100,000 to unlimited, it would be impossible to administer assets for “the public good” or to compromise the interests of a single policyholder in order to avoid greater harm to a broad spectrum of policyholders. The form of a rehabilitation would be dictated by the state guaranty fund with the highest level of coverage. Guaranty funds do not belong in the driver’s seat of an insolvency proceeding.
Mutual Fire II is instructive on the issue of benefit reductions, which the parties agree is a type of contract impairment. Of course, a liquidation impairs contract. It terminates all policies mid-term and re*452places them with coverage that is limited by whatever is contained in the relevant guaranty fund statute.
A rehabilitation plan is permitted to impair the contractual rights of some policyholders in order to minimize the potential harm to all of the affected parties. Mutual Fire II, 581 Pa. at 613-14, 614 A.2d at 1094; see also Section 501(c) of Article V, 40 P.S. § 221.1(c)59 (explaining that the purpose of the insurance statute is “protection of the interests of insureds, creditors, and the public generally, with minimum interference with the normal prerogatives of the owners and managers of insurers”). The Supreme Court adopted the following three-part test in Mutual Fire II for determining whether a contractual impairment, such as a benefit reduction, is legally permissible in the context of rehabilitation:
The threshold inquiry is to determine whether the state statute in reality has operated to substantially impair a contractual relationship.... Should it be determined that a substantial impairment has occurred, the state must set forth a legitimate and significant public purpose.... Once that purpose is identified, the final inquiry concerns whether the adjustment of contractual rights is reasonable and of a nature appropriate to the public purpose justifying the legislation’s adoption; however, if the state is not a contracting party, deference is given to the state’s enunciated purpose.
Mutual Fire II, 581 Pa. at 615 n. 4, 614 A.2d at 1094 n. 4 (citing Energy Reserves Group, Inc. v. Kansas Power and Light, 459 U.S. 400, 103 S.Ct. 697, 74 L.Ed.2d 569 (1983)).
The Rehabilitator ignores the three-part test adopted in Mutual Fire II and instead cites to different portions of that opinion as well as to this Court’s decisions in Mutual Fire I and Koken v. Fidelity Mutual Life Insurance Co., 803 A.2d 807 (Pa.Cmwlth.2002), to suggest that the only relevant consideration is whether a single policyholder would be better off in liquidation.
In Fidelity Mutual, a policyholders’ committee and several individual policyholders objected to the rehabilitator’s proposed third amended plan of rehabilitation. One of the policyholders objected for the stated reason that policyholders would be better off if Fidelity Mutual were liquidated rather than rehabilitated, and this Court overruled the objection. Fidelity Mutual, which had large asset deficiencies when placed in rehabilitation, had turned around because the value of its assets recovered. The insurer had paid and declared significant dividends, and its common and preferred stock had substantial value. Given those circumstances, a liquidation did not make sense. Accordingly, this Court affirmed the rehabilitator’s exercise of discretion in fashioning the plan. Fidelity Mutual, 803 A.2d at 826.
In Fidelity Mutual, the Court cited Neblett v. Carpenter, 305 U.S. 297, 59 S.Ct. 170, 83 L.Ed. 182 (1938), for the proposition that “[c]reditors and policyholders must fare at least as well under a rehabilitation plan as they would under a liquidation.” Fidelity Mutual, 803 A.2d at 826. At issue in Neblett was a rehabilitation plan offering holders of non-cancellable health and accident policies the option of taking insurance in a newly formed company on less favorable terms, or of proving their claims for breach of their contracts and receiving payment out of the assets of the old company. The United States Supreme Court held, inter alia, that there was no impairment of the contracts be*453cause there was no evidence that the policyholders would not receive as much on their claims under the plan as they would receive in a liquidation sale of assets and distribution of the proceeds. Id. at 805, 59 S.Ct. 170. In short, the objectors in Neb-lett did not prove impairment. Neblett did not establish the broad principle that a rehabilitation plan is per se invalid unless every policyholder will fare as well in rehabilitation as in liquidation.
Further, the cases cited by the Rehabili-tator because they purport to rely on Neblett, ie., the Mutual Fire and Fidelity Mutual decisions, did not involve petitions to convert a rehabilitation to a liquidation under Section 518(a) of Article V. Rather, Mutual Fire and Fidelity Mutual concerned the objections of individual policyholders to a formal rehabilitation plan that they believed should be modified or rejected, lest they receive inequitable treatment.
This Court must be guided by the three-part test adopted in Mutual Fire II. The “threshold inquiry” is whether state action “has operated to substantially impair contractual relationships.” Mutual Fire II, 531 Pa. at 615 n. 4, 614 A.2d at 1094 n. 4 (emphasis added). An impairment of contractual rights is not a per se violation of law. Id. If a particular policyholder is found to be worse off under a rehabilitation plan, the impairment could be considered “substantial,” but the Court still needs to determine whether (1) the reha-bilitator has acted for a legitimate and significant public purpose and (2) the adjustment of contractual rights is reasonable and of a nature appropriate to that public purpose. Id. To that end, the Court must be ever mindful that one of the primary goals of Article V is “the protection of the interests of insureds, creditors, and the public generally....” Section 501(c) of Article V, 40 P.S. § 221.1(c).60
In summary, there is no statutory provision or case law precedent that requires this Court to grant a liquidation petition unless it can be established that every single policyholder will fare at least as well in rehabilitation as in liquidation. The Rehabilitator cannot establish that liquidation is necessary simply by offering hypothetical circumstances where benefit reductions might result in some policyholders receiving less than what they would receive from their state guaranty association in liquidation. The Court must consider the greater good, including the consequences to the larger class of policyholders and the taxpaying public. Vicko-dil, 559 A.2d at 1013 (“individual interests may need to be compromised in order to avoid greater harm to a broader spectrum of policyholders and the public”).
Intervenors did not proffer a specific rehabilitation plan.61 It is worth noting, however, that the hearing evidence suggested several potential viable benefit reduction options.
For example, on June 23, 2009, Mark Cloutier, the former CFO of PTNA, proposed a benefit capping rehabilitation alternative to the RIC that involved benefit caps designed to “one-up” a liquidation plan. Cloutier’s proposal was based on adjustments to the April 2009 projection to be current as of June 23, 2009, and his proposal showed PTNA’s surplus turning *454positive and staying positive by 2017. See (Mohoric) N.T. 9/20/11 at 119-22; (Pfan-nerstill) N.T. 3/23/11 at 136-37; Exs. R-929, R-930. Mohoric advised the RIC that Cloutier’s proposal offered several advantages:
• Keeps entire insured base out of liquidation.
• Is fairly simple.
• Could be applied to entire insured base — PTNA/ANIC, Newco/OldCo, actives/current claimants rather than making distinctions between those who could move to ANIC as in the current plan.
• Clearly provides a better benefit (though possibly small) than liquidation would.
• Could still do an option to not reduce benefits for right priced premium (likely need underwriting to make viable).
Ex. R-930. See also (Mohoric) N.T. 9/20/11 at 126-40. Mohoric requested additional time to analyze the proposal, but DiMem-mo instructed him not to do the analysis. (Mohoric) N.T. 9/20/11 at 139-40. DiMem-mo believed it was not the time to raise “new ideas,” as provided in Ex. R-930, because the Rehabilitator was committed to the rate increase strategy laid out in the Preliminary Plan. (DiMemmo) N.T. 9/21/2011 at 16-20, 25-26, 36-39, 46.
In addition to Cloutier’s “one upping” alternative, there were several different versions of benefit reduction options that were discussed by various RIC members. For example, James Potts, the Rehabili-tator’s counsel from Cozen O’Connor, presented an option that involved giving policyholders a choice between keeping their OldCo policies and staying with PTNA in a liquidation or getting a NewCo policy from ANIC with more appropriate benefit levels. Ex. R-985 at EPM-AHD 01574-01575. This option was considered, initially, but then ruled out for reasons that were never explained. DiMemmo could not explain why this option, which has been used in other receiverships, was not a viable option for the Companies. N.T. 9/21/11 at 13.
There is no evidence that alternative options such as the two discussed above were considered or rejected by the Reha-bilitator because he believed they were legally impermissible. Further, the Reha-bilitator did not establish that there was any meaningful analysis of options such as offering an alternative to lifetime benefits, conversion of non-tax qualified policies to tax qualified, or other benefit modifications, whether done alone or in conjunction with other approaches, such as rate increases. It is undisputed that OldCo policyholders with inflation or lifetime benefits, ie., the “Cadillac” product, are unfairly subsidized by policyholders with more limited benefits. (Waite) N.T. 2/1/11 at 214-16; (Mohoric) N.T. 2/16/11 at 120-21. In response to questioning by the Court regarding whether a product-specific revision regarding the Cadillac product had been considered, DiMemmo referred the Court to Robinson and Mohoric. N.T. 2/3/11 at 190. However, neither Robinson nor Mohoric testified regarding why such an approach would be futile.
The Rehabilitator argues, next, that benefit caps “could place [the Companies] in violation of state insurance laws throughout the country.” Rehabilitator’s Brief at 35. Because the Companies were required to have their policy forms and rates approved by state regulators before issuing them, the Rehabilitator asserts that benefits may not be capped without also adjusting rates. This states the obvious. The object of capping benefits is to offer a more palatable way to reach the actuarially justified rate need. The Reha-bilitator cites four statutory provisions, os*455tensibly to support this position. Rehabili-tator’s Brief at 35 (citing Section 303 of the Accident and Health Filing Reform Act, Act of December 18, 1996, P.L. 1066, as amended, 40 P.S. § 3801.303; Fla. Stat. Ann. § 627.410; Tex. Ins.Code Ann. § 1701.051; VaCode Ann. § 38.2-316).
These state statutes generally require insurers to file their policy forms for approval by the state insurance department before using them. There is nothing in any of the statutory provisions cited by the Rehabilitator that addresses benefit modifications in the context of a court approved rehabilitation plan. Further, the Rehabilitator does not offer any case law that would prevent this Court from approving a rehabilitation plan that modifies policyholder benefits. Indeed, the Reha-bilitator’s counsel advised him that benefit reductions are permissible in rehabilitation. Ex. R-1025; (Mohoric) N.T. 9/20/11 at 171-74; (Waite) N.T. 2/1/11 at 10-12.
In short, there is no support for the Rehabilitator’s position that this Court cannot modify policyholder benefits as part of an approved rehabilitation plan. The modification can involve the use of already approved policy forms. The Rehabili-tator’s position would eliminate an important tool of rehabilitation. A court supervised rehabilitation, however, is intended to provide the flexibility not available in liquidation through “improved methods for rehabilitating insurers.” Mutual Fire II, 531 Pa. at 614, 614 A.2d at 1094 (citing Section 501 of Article V, 40 P.S. § 221.1).
Transfer of PTNA NewCo to ANIC
One option that the RIC considered was transferring PTNA NewCo policyholders to ANIC. Ex. R-82. The option was included in a draft of Milliman’s September 2009 Report but later withdrawn at the request of Robinson. Mohoric supported the transfer option. N.T. 2/23/11 at 51. Mo-horic’s projections showed that if 100% of NewCo policyholders transferred to ANIC, its surplus would immediately recover and exceed the 200% RBC ratio. (Mohoric) N.T. 2/23/11 at 52-54. If 85% transferred, ANIC’s 200% RBC ratio would be reached in 2024. Id.
The Rehabilitator argues that a transfer is futile. He argues that PTNA cannot transfer NewCo policies with all the active life reserves needed to support those policies. If it did transfer the full assets needed, then that would reduce the assets left in PTNA to back the OldCo business. The Rehabilitator argues that PTNA’s assets cover less than 40% of its reserves for policyholder liabilities. (Mohoric) N.T. 2/18/11 at 30-31; N.T. 2/24/2011 at 186.62 A transfer of assets above that percentage would reduce what the guaranty associations would receive in a liquidation of PTNA. (Robinson) N.T. 2/14/11 at 64-65; (Robinson) N.T. 2/15/11 at 208; (Mohoric) N.T. 2/17/11 at 48-49. The Rehabilitator asserts that a rehabilitation plan cannot reduce what policyholder and guaranty funds would receive in liquidation.
The Rehabilitator also argues that a transfer of PTNA NewCo policyholders to ANIC would be detrimental to PTNA NewCo policyholders who might lose guar*456anty fund protection should ANIC become insolvent. The Rehabilitator explains that many guaranty associations limit coverage to policyholders whose policies were issued by a licensed insurer, when issued. The Rehabilitator cites four examples: Cal. Ins.Code § 1067.02(b)(2)(G); Fla. Stat. Ann. § 631.714(6); 215 Ill.C.S. § 5/531.03(2)(b)(xii); Wash. Rev.Code § 48.32A.025(2)(b)(vi). Pennsylvania provides coverage for ANIC policyholders in states where ANIC was never licensed, but this protection would not apply where ANIC had been licensed at one time. See Section 1703(a) (2) (iii) (B) of The Insurance Company Law of 1921, Act of May 17, 1921, P.L. 682, as amended, 40 P.S. § 991.1703(a)(2)(ii)(B)63 (providing guaranty association coverage for nonresident insureds of Pennsylvania insurers but only where, inter alia, “such insurers never held a license or certificate of authority in the states in which such persons reside”). The Rehabilitator reasons that NewCo policyholders in states where ANIC was once licensed, but not still licensed when it assumed a NewCo policy from PTNA, could lose guaranty fund protection by reason of the transfer. This is an unjustifiable risk, according to the Rehabilitator, because ANIC may fail, as may any insurance company. By staying with PTNA, NewCo policyholders will receive guaranty fund protection upon PTNA’s liquidation.
The Rehabilitator’s arguments on the transfer option are flawed from the start because. they are based on the premise that Milliman’s projections are accurate. The Court has rejected those projections. The specific argument that PTNA can only fund 40% of the assets needed to support a transfer of NewCo policyholders is rejected because it is based upon Milliman’s gross premium reserve. Rehabilitator’s Proposed Findings of Fact at 49, ¶240.
The Rehabilitator also offers a grim view of what would happen to ANIC post-transfer. However, Milliman’s short term projections for 2010 missed the mark by more than 50%. PTNA’s cash flow in 2010 was negative $18.7 million compared to negative $37.8 million projected by Milli-man, a difference of 51%. ANIC’s cash flow in 2010 was positive $8.1 million, compared to $5.2 million projected by Milli-man, a difference of 56%. (Volkmar) N.T. 10/25/11 at 197; Ex. R-912.
In short, the Court rejects the Rehabili-tator’s premise that the reserves needed to be transferred to ANIC to back the PTNA NewCo policies are at the level set by Milliman.
The Court also rejects the Rehabili-tator’s argument that licensing issues will frustrate any rehabilitation plan involving the transfer of NewCo policies from PTNA to ANIC. The Rehabilitator cites without fully quoting the operative language from the statutes of California, Florida, Illinois and Washington. By way of example, Washington’s life and health insurance guaranty association statute provides coverage to any residents who are insurance policyholders, subject to the exclusion that “[tjhis chapter does not provide coverage for ... [a] policy or contract issued in this state by a member insurer at a time when it was not licensed or did not have a certificate of authority to issue the policy or contract in this state.” Wash. Rev.Code § 48.32A.025(2)(b)(vi).64
*457The purpose of such a provision is to discourage consumers from purchasing coverage from insurance companies that are not licensed. It is unlawful for insurance companies to issue policies in states where they are not licensed, with exceptions not relevant to this discussion. The Rehabilitator assumes that guaranty funds faced with such exclusionary language would treat the transfer or assumption of a PTNA NewCo policy as a new policy “issued” in the receiving state. It is just as likely that such an exclusion would not be triggered by the transfer of a validly issued policy to ANIC. This would be true whether or not ANIC is licensed in Washington or any “transferee” state with a similar statute. The need to access guaranty association coverage with regard to the transferred policies would still have been as a result of the earlier impairment or insolvency of PTNA (the member insurer that issued the policies or contracts).
Moreover, Pennsylvania provides guaranty fund protection in certain circumstances to non-resident policyholders insured by a Pennsylvania-domiciled insurer. Non-residents must meet the following requirements in Section 170S(a)(2)(ii) of The Insurance Company Law of 1921:
(A) the insurers which issued such policies or contracts are domiciled in this Commonwealth;
(B) such insurers never held a license or certificate of authority in the states in which such persons reside;
(C) these states have associations similar to the association created by this article; and
(D) these persons are not eligible for coverage by those associations.
40 P.S. § 991.1703(a)(2)(ii). The number of non-resident PTNA NewCo policyholders who could avail themselves of Pennsylvania guaranty association coverage is, of course, unknown and will require state-by-state review and analysis. There is certainly no reason, however, to follow the Rehabilitator’s overly pessimistic view that “new” ANIC policyholders coming over from PTNA will forfeit their guaranty association protection.
The Rehabilitator’s stated concerns about the transfer option depends upon an interpretation of state guaranty association statutes that is beyond the scope of this opinion. However, even assuming the Re-habilitator’s strained interpretation is correct, there are possible solutions such as those suggested by Intervenors:
• Entering into a settlement with NOLHGA where the guaranty associations agree that any of PTNA’s New-Co policyholders that are currently eligible for guaranty association coverage retain their eligibility for future protection (if necessary), after they transfer to ANIC as part of a rehabilitation plan;
• A modified transfer option involving only the transfer of PTNA’s NewCo policies in the states where the transfer would not deprive policyholders of guaranty association protections in the future;
• Packaging the transfer option with one or more additional methods of rehabilitation to ensure ANIC’s long-term success; and
• Offering PTNA’s NewCo policyholders the right to opt out and remain in PTNA if they wish, rendering the transfer option completely voluntary from the policyholders’ perspectives.
Intervenor’s Reply Brief at 81.
In summary, the heavy burden of proving that continued rehabilitation of the Companies will be futile rests with, and never shifts from, the Rehabilitator. As *458explained in Legion, the way to prove futility is to adopt a rehabilitation plan and, then, have its implementation fail. Here, the Rehabilitator has not even devised a formal rehabilitation plan.
Essentially, the Rehabilitator offers excuses, arguing that various rehabilitation options, ie., rate increases, benefit reductions and transfer of policies, either will not work or will be exceedingly difficult in today’s regulatory environment. The Re-habilitator’s entire futility argument is premised upon Milliman’s projections, which have fluctuated wildly and which the Court has already deemed unreliable. Using these inherently unreliable numbers, the Rehabilitator shoots down the various proposed rehabilitation options based primarily upon speculation about what state regulators will or will not do in the context of a court ordered rehabilitation. The Re-habilitator’s futility argument is the proverbial house of cards.
The Rehabilitator ignores this Court’s mandate in Legion that futility cannot be established until a plan of rehabilitation has been filed and its progress tracked. Legion, 831 A.2d at 1245 (“Only with the filing of a plan, even in the most preliminary form, is it possible to track progress and establish the plan’s viability or its futility.”). The Rehabilitator filed a Preliminary Plan, but not a formal plan of rehabilitation. In the absence of a plan against which progress can be measured, futility is a matter of speculation. In short, the Court holds that the Rehabili-tator has not satisfied his burden of proving that a continued rehabilitation of the Companies will serve no useful purpose.
Y. Conclusions of Law
1.The Rehabilitator has the burden of proof with regard to a petition to convert a rehabilitation into a liquidation under Section 518(a) of Article V, 40 P.S. § 221.18(a).
2. The Rehabilitator must establish (i) that he had reasonable cause to believe that grounds existed to terminate the re-habilitations before he petitioned the Court; (ii) that the Companies were insolvent on the date the petitions were filed; and (iii) that continued rehabilitation would substantially increase the risk of loss to creditors, policy and certificate holders, or the public, or would be futile. Section 518(a) of Article V, 40 P.S. § 221.18(a); Legion, 881 A.2d at 1229-30.
3. After establishing reasonable cause to believe grounds exist to file the petition, the Rehabilitator must prove that continued rehabilitation would either increase the risk of loss to creditors, policy and certificate holders, or the public, or would be futile.
4. The Rehabilitator has not met his burden of proving that continued rehabilitation of PTNA and ANIC would substantially increase the risk of loss to creditors, policy and certificate holders, or the public.
5. The Rehabilitator has not met his burden of proving that continued rehabilitation is futile with regard to PTNA and ANIC.
6. Intervenors have provided a thorough and careful defense to the petitions and are entitled to an award of reasonable attorneys’ fees and costs pursuant to Section 518(a) of Article V, 40 P.S. 221.18(a), in an amount to be determined at a later date.
VI. Conclusion
The Rehabilitator has not made an earnest effort to correct the condition that caused the Companies’ financial difficulties: the pricing structure for the OldCo non-tax qualified policies, particularly those policies with “Cadillac” coverages. The Rehabilitator never devised a plan of rehabilitation but, rather, looked for reasons to be excused from that duty.
*459The Rehabilitator has approached the Companies’ financial challenges as a regulator, fixated on statutory surplus, and not as a receivership trustee fixated on a turnaround. During a rehabilitation, the impaired insurer operates under the protection and direction of the Court; during this time the payment of claims, not surplus, is paramount. The Rehabilitator’s misplaced fixation on the Companies’ surplus explains, perhaps, his stated concern that a rehabilitation should not be undertaken unless there is confidence that the company can be discharged from receivership in a decade. However, this stated concern is unfounded. Fidelity Mutual was in rehabilitation for 15 years. Ario v. Fidelity Mutual Life Insurance Co., 9B5 A.2d 55, 57 (Pa.Cmwlth.2007). There is no timetable in statute or case law by which an insurer’s correction need be completed.
The Rehabilitator refers, repeatedly, to the Companies’ “severe insolvency.” This was not a term .used by any actuary who testified in this proceeding. The Rehabili-tator has taken it up in the effort to convey a sense of urgency that simply does not exist. The Companies are meeting their obligations as they come due and will be able to do so for many years, notwithstanding their inadequate rate structure and notwithstanding the Rehabilitator’s phlegmatic effort to address that problem before abandoning it entirely.
The financial problem for both Companies is discrete: the OldCo non tax qualified policies are underpriced and overly generous. That is why policyholders do not receive a tax deduction for their premium payment. In a rehabilitation these policies can be reformed and repriced.
This leads to the question of what serves the policyholders’ interest. The Rehabili-tator wants to make the Companies’ policyholders wards of the various state insurance guaranty funds, but he has presented no evidence to explain what this means as a practical matter. In his testimony, Robinson conceded that the guaranty funds have the legal right to raise premium rates, although it was his hope and belief that they would not do so. N.T. 2/14/11 at 65. The Pennsylvania guaranty fund statute expressly authorizes the Pennsylvania Life and Health Insurance Guaranty Association to set premium “in accordance with the amount of insurance provided and the age and class of risk, subject to the approval of the commissioner or by a court of competent jurisdiction.” Section 1706(d)(4) of The Insurance Company Law of 1921, Act of May 17, 1921, P.L. 682, added by the Act of December 18, 1992, P.L. 1519, 40 P.S. § 991.1706(d)(4). Further, Pennsylvania’s guaranty fund may “reissue the terminated coverage or ... issue an alternative policy.” Id. at Section 1706(d)(2), 40 P.S. § 991.1706(d)(2).
In liquidation, therefore, the policies of the Companies will be changed in several ways. The guaranty funds may issue replacement policies without regard to the fact that the Companies’ policies protect against rate increases caused by age or sickness. They can impose the rate increases state regulators have found anathema. The guaranty funds will not, however, be able to honor the amount of coverage for which the policyholders contracted. The guaranty funds can issue replacement coverage only within the limits set forth in their governing statutes. In Pennsylvania, for example, all policies will be capped at $300,000. Section 170S(c)(ii)(A)(IV) of The Insurance Company Law of 1921, added by the Act of December 18, 1992, P.L. 1519, as amended, 40 P.S. § 991.1703(c)(ii)(A)(IV). Further, the guaranty fund coverage is provided per person, not per policy. Section 1703(c)(ii)(A) of The Insurance Company Law of 1921, 40 P.S. § 991.1703(c)(ii)(A). Approximately 2.5% of PTNA’s policyholders have multiple policies. In liqui*460dation, they will be limited to the coverage of a single policy, despite having paid premium on multiple policies. All these changes in coverage and pricing will be effected not by choice of the policyholder but by choice of the guaranty fund.
By contrast, in a rehabilitation, the policyholders will have a voice. They can, and will, be heard in the Court proceeding on the rehabilitation plan. As described above, a rehabilitation plan can be devised to give policyholders choices. They may choose to keep their “Cadillac” non-tax qualified policies, and pay the appropriate rate therefor, or they may choose to move to a tax-qualified product that offers reasonable coverage at a more reasonable price. Waite testified that the average annual premium for an OldCo policy in 2011 was approximately $2,200. Ten years earlier, before successive rate increases, it was $1,900. (Waite) N.T. 1/31/11 at 176. Waite attributed the relatively small increase to the fact that a substantial number of policyholders decided to reduce their coverages rather than pay the indicated rate increase.
There is no downside to continued rehabilitation. Any rate increase will benefit the estate, even if, ultimately, a rehabilitation fails. Restructuring rates can remove the subsidies, unfair to all policyholders, that presently exist. On the other hand, if there were a liquidation, state regulators need to be aware that their state’s inadequate rates will be considered. For example, estate distributions may be made in inverse proportion to the state’s contribution to the insolvency. The farther the state’s approved rate deviates from what is actuarially justified, the less that state’s guaranty fund should expect by way of distribution from the estate in the event of liquidation, absent extenuating circumstances.
Policyholders with dementia constitute the largest category of claimants. In October of last year, Dr. Holland testified that a breakthrough in treatment of Alzheimer’s was imminent. Four months later, a breakthrough received wide coverage. A long-used skin cancer medication, called bexarotene, has been shown to increase brain function dramatically in mice with advanced Alzheimer’s disease. Gau-tam Naik, New Attack on Alzheimer’s: Cancer Drug Reverses Disease’s Symptoms in Mice; Human Tests to Start Soon, Wall St. J., Feb. 10, 2012, at AS.65 The drug helps to clear amyloid fragments from the brain. This report is not evidence of record, and it has not been relied upon by the Court. Indeed, there is a difference between curing mice and curing humans. Nevertheless, the announcement of this breakthrough underscores the self-evident point that the future is uncertain and future developments may benefit the Companies.
The Rehabilitator assumes the opposite, i.e., that future developments will adversely affect the Companies. Given the nature of long-term care insurance, which requires projections 60 years into the future, the Rehabilitator’s desire for “conservative” projections that would not “change along the way” is impossible to satisfy. Ex. R-987. Projections must change as claim experience develops in the coming decades. On this point, the actuaries agreed.
The record does not support the Reha-bilitator’s thesis that the existing pricing structure for the OldCo policies cannot be corrected by any means whatsoever. On the other hand, the evidence has estab*461lished that the Companies are operationally sound, have maintained $1 billion in assets, are paying claims timely and will do so for years, even without rate relief. On this foundation, a plan of rehabilitation can, and will, be built by using one or several of the rehabilitation approaches identified at the hearing. A successful rehabilitation is not certain, but a liquidation, no matter how “successful,” is certain to cause harm to the policyholders, creditors and taxpaying public as the Re-habilitator has acknowledged in his Preliminary Plan.
Intervenors have proposed that this Court order an immediate resumption of the rehabilitation, which has languished since the liquidation petitions were filed. Specifically, they request, inter alia, that they be involved in the preparation of a rehabilitation plan, which must be prepared and filed as soon as possible. The Court agrees. There is no reason to delay the participation of Intervenors until after the plan is filed, given the delays occasioned by this litigation. The Court also agrees with Intervenors that rate increases for OldCo are critical and, accordingly, the Rehabilitator must prepare an action plan for obtaining such relief.
An appropriate order follows.
ORDER
AND NOW, this 3rd day of May, 2012, upon consideration of the Rehabilitator’s Amended Petitions for Liquidation of Penn Treaty Network America Insurance Company (In Rehabilitation) and American Network Insurance Company (In Rehabilitation) (collectively, Companies) and the evidence adduced at the hearings thereon, it is ORDERED as follows:
1. The Amended Petitions are DENIED for the reasons set forth in the accompanying Opinion.
2. The Court’s orders of January 6, 2009, remain in effect. Paragraphs 9 and 10 of the January 6, 2009, orders are amended as follows: The Rehabilitator will not refuse to pay claims or refuse to renew the Companies’ long-term care insurance policies without prior Court approval.
3. The Rehabilitator shall develop a plan of rehabilitation of the Companies, in consultation with Intervenors, and shall submit a plan no later than ninety (90) days following the date of this Order.
4. The plan of rehabilitation must address and eliminate the inadequate and unfairly discriminatory premium rates for the OldCo business.
. At the time the rehabilitation petitions were filed, PTNA was insolvent under statutory accounting principles, but this was not cited in the petitions as a reason for the Companies' rehabilitation.
. The Rehabilitator's own witness, Cameron Waite, former Chief Financial Officer and Executive Vice President of Strategic Operations of the Companies and PTAC, testified that, at the time the Companies entered rehabilitation, he believed that "the projections for the future included sufficient release of reserves so that all claims could be paid, coupled with anticipated premium rate increases that were filed for or expected at that time.” Notes of Testimony (N.T.) 1/31/11 at 214-15. There was no evidence presented that other members of management did not share this belief.
. Under Section 516(a) of Article V, the "commissioner as rehabilitator may appoint a special deputy who shall have all the powers of the rehabilitator granted under this section.” 40 P.S. § 221.16(a). Section 516 was added by the Act of December 14, 1977, P.L. 280. The commissioner has not appointed a special deputy for the rehabilitation of PTNA and ANIC.
. At times, the post-hearing briefs lapsed into invective. Persuasion requires an analysis of the law, the evidence in the record, citations thereto and logic. Nothing more.
. PTNA and ANIC need a surplus of $1.5 million to be solvent.
. The Rehabilitator did not explain how the replacement coverage will be issued, i.e., directly by the funds or by other insurance companies that will receive subsidies from the funds if needed.
.Of course, policyholder claims against an insolvent insurer estate may be filed after the insurer is liquidated. However, the claim must be one for a loss that occurred while the policy was in existence, i.e., before the liquidation.
. Since rehabilitation, South Dakota raised its coverage to $300,000. (Waite) N.T. 2/1/11 at 117.
. The costs of services in a particular state may vary, but it does not affect claims, which are based on the policy’s maximum daily benefit, not actual costs. In rare instances, actual costs may be lower than the maximum daily benefit, in which case the policyholder's claim is determined by the actual cost of the service.
. ANIC has a New York-domiciled wholly owned subsidiary, American Independent Network Insurance Company of New York (AINIC). AINIC is licensed to conduct insurance business only in New York.
. Notably, a tax qualified long-terai care insurance contract does not have to cover all six activities of daily living. It can qualify with five. 26 U.S.C. § 7702B(c)(2)(B).
.The Companies' policy forms contain similar definitions of the "Activities of Daily Living”:
Basic, day-to-day, human functions ... comprised of the following six activities:
1) Eating is feeding oneself by getting food into the body from a receptacle (such as a plate, cup or table).
2) Bathing is washing oneself by sponge bath; or in a tub or shower, including the task of getting into or out of the tub or shower.
3) Dressing is putting on and taking off all items of clothing and any necessary braces, fasteners or artificial limbs.
4) Transferring is moving into and out of a bed, chair or wheelchair.
5) Toileting is getting to and from the toilet, getting on and off the toilet, and performing associated personal hygiene.
6)Continence is the ability to maintain control of bowel and bladder functions; or when unable to maintain control of bowel and bladder functions, the ability to perform associated personal hygiene (including caring for catheter or colostomy bag).
Ex. R-789, Form No. ALP2-TQ-P(AZ) at 17.
. " 'Medically Necessary' means recommended by Your Physician as necessary and in accordance with the usual standards of medical practice for Your injury or sickness.” Ex. R-789, Form No. PF2600(PA)-P at 5.
. The conditions of eligibility are the same for the Home Health Care Benefit and Nursing Facility Benefit.
. The National Association of Insurance Commissioners (NAIC) devised a risk-based capital (RBC) system to measure the adequacy of an insurer's capital given the insurer’s risk. The RBC system has two main components: (1) a formula that establishes a minimum capital level for an insurer and (2) a model law that grants authority to a state’s insurance regulator to take action based upon the insurer's level of impairment. Pennsylvania has adopted the NAIC model law on RBC requirements. See Sections 501-A — 515-A of the Insurance Department Act of 1921, added by the Act of June 25, 1997, P.L. 349, as amended, 40 P.S. §§ 221.1-A-221.15-A. There are five action levels under the RBC system: (1) No Action; (2) Company Action Level, where the insurer must identify what caused its capital deficiency and take steps to correct its financial condition; (3) Regulatory Action Level, where the insurer files an action plan and the state insurance commissioner examines and analyzes the insurer’s business and operations and issues corrective orders; (4) Authorized Control Level, where the state insurance commissioner may take control of the insurer; and (5) Mandatory Control Level, where the state insurance commissioner is required to take control of the insurer. See NAIC, Risk-Based Capital General Overview, July 15, 2009, available at http://www.naic. org/documents/committees_e_capad_RBC overview.pdf. See also Section 501-A of the Insurance Department Act of 1921, 40 P.S. § 221.1-A (noting Pennsylvania's "Regulatory Action Level” is 1.5 times an insurer's authorized control level).
. Exs. R-1132 and 1136 were prepared by Brookfield Investment Management Inc., the investment advisor to the Companies. The Brookfield reports contain information regarding the assets of the Companies as well as AINIC, the wholly owned subsidiary of ANIC. Because Brookfield does not manage all of the assets of the Companies, the Brookfield reports understate the value of the assets.
. An identical order was entered for ANIC at 4 M.D. 2009. The matters were consolidated at 5 M.D. 2009 by order of this Court dated August 13, 2010. Pursuant to this Court's order of July 9, 2012, the consolidated matter was redocketed at 1 PEN 2009.
. The Insurance Commissioner’s appeal of the Commonwealth Court’s order in Legion was consolidated with her appeal of the Commonwealth Court’s order in Koken v. Villano-va Insurance Co., (No. 182 M.D.2002, filed June 26, 2003). The Supreme Court issued one order in the consolidated appeal under the caption Koken v. Villanova Insurance Co., 583 Pa. 400, 878 A.2d 51 (2005). The Supreme Court's per curiam order read:
AND NOW, this 19th day of July, 2005, the order of the Commonwealth Court is hereby AFFIRMED, on the basis of the Commonwealth Court opinion, Koken v. Legion Ins. Co., 831 A.2d 1196 (Pa.Cmwlth.2003).
Id. at 404, 878 A.2d at 53. A per curiam affirmance on the basis of a lower court opinion means that the Supreme Court agrees with the lower court’s rationale employed in reaching its final disposition. Commonwealth v. Tilghman, 543 Pa. 578, 589, 673 A.2d 898, 904 (1996). Otherwise, a per curiam affir-mance has no significance except as law of the case.
. Notably, PTNA requested actuarially justified premium rate increases of 50% in Pennsylvania, but the Insurance Commissioner approved a 15% increase.
. This is not the case for the Companies because they must renew their policies even though they are underpriced.
. The Rehabilitator argues that the true projected surplus as of December 31, 2009, was far more negative for both companies than Intervenors’ expert projected: negative $886 million for PTNA and negative $30 million for ANIC. Rehabilitator’s Proposed Findings of Fact at 2, ¶ 3. The Rehabilitator’s expert manipulated the Intervenors’ surplus projection to remove any future rate increases. Even the Rehabilitator used future rate increases in its projected surplus. Intervenors' Reply Brief at 122. The Court rejects the Rehabilitator’s recast of Intervenors’ projected surplus.
. In ASOP 25 the term "predictive” is used "in the statistical sense and not in the sense of predicting the future.” Ex. P-896 at 1; (Volkmar) N.T. 10/25/11 at 148.
. See note 15 supra.
. In its October 2009 Report, Milliman developed two scenarios, A and B, that used two different sets of assumptions to develop two different surplus projections as of December 31, 2009, for both Companies. Scenario A was more optimistic, and Scenario B was more pessimistic. Scenario A projected a surplus of negative $1.4 billion for PTNA and negative $45 million for ANIC. Scenario B projected a surplus of negative $2.1 billion for PTNA and negative $130 million for ANIC. (Pfannerstill) N.T. 3/23/11 at 17.
For the Scenario A claim costs assumption, Milliman adjusted the claim costs in the Li-tow Report, and Scenario B established a new claim costs assumption, under an abbreviated study by Pfannerstill. Scenario A used a wellness assumption of 6% for six years, and Scenario B did not have a wellness assumption. Scenario A used a 1.2% morbidity improvement assumption, and Scenario B used a 1.5% morbidity improvement assumption for 10 years and 1.0% thereafter. (Pfanners-till) N.T. 3/22/11 at 142-45. Scenario A assumed a higher premium rate increase assumption than Scenario B. (Volkmar) N.T. 10/26/11 at 62-63. Scenario B included a gross premium reserve that used incurred claims to increase reserves by $528 million. (Mohoric) N.T. 11/1/11 at 133. It is not clear whether or not Scenario A included a gross premium reserve.
. The October 2009 Report does not contain a "change caveat,” such as the one Milliman used in its Litow Report, notwithstanding the unprecedented changes between the April 2009 and September 2009 projections.
. In his reply brief, the Rehabilitator claims that Milliman's approach demonstrated its *403"consideration of balancing responsiveness with stability," but this claim is contradicted by Mohoric's testimony. Rehabilitator's Reply Brief at 83, ¶ 202.
. Milliman's "MG-ALFA” model is used to make its projections for a company’s profit for a given year. Rehabilitator’s Proposed Findings of Fact at 73, ¶ 351. It uses 90,000 separate cells of information that break out information on policies by age, sex, type of policy, benefit design and marital status. Id. at ¶ 352.
. Milliman used the assumptions in the Li-tow Report to prepare rate increase filings, actuarial appraisals and other projections that were certified by Milliman as being reasonable and trustworthy. Milliman used the Li-tow Report to reduce the Companies’ active life reserve, which was approved by the Department. Litow still works for Milliman, but he was not involved in the creation of the new methodology used to calculate the Companies' claim reserves or other reserves.
. Pfannerstill testified that an ultimate claim cost assumption is the claim cost that "is no longer impacted by the influences of when the policy was underwritten.” N.T. 3/22/11 at 137.
. The term "reinvestment interest rate” is synonymous with the term "new money rate” and both terms were used interchangeably in the experts’ reports and testimony. (Volk-mar) N.T. 10/26/11 at 60.
. Actuaries must be taught in school to use the word "scenario” to refer to their alternate calculations and to distinguish one calculation from the other by using capitalized letters of the alphabet. Accordingly, Volkmar used "Scenario A,” "B," "C" and "D” to refer to four different rate increase calculations, using four different interest rates. Milliman used *417"Scenario A" and "Scenario B" in its October 2009 Report to refer to two different surplus projections, each using two different sets of assumptions.
"Scenario A" as used in Volkmar's Report has nothing to do with the "Scenario A" used in Milliman’s October 2009 Report.
. Volkmar testified that he had intended to grade the interest rates for the projections on the left side of Attachment 6, Page 30 by five years but accidentally graded them from 5.34% to 6.4% over five months. He communicated the error in a letter to the Reha-bilitator. The impact was insignificant. It produced a 1.8% aggregate increase in the present value of earned premiums for PTNA and a 3% aggregate increase for ANIC.
. See n. 15, supra.
. The Rehabilitator asserts that "Minches expressly stated that he had no reservations about the work performed by Milliman with respect to its updated analyses and studies.” Rehabilitator's Proposed Findings of Fact at 58, ¶ 277 (citing N.T. 3/25/11 at 23, 28-29, 33-34). Minches did not make that statement in the testimony cited by the Rehabilitator.
The Rehabilitator also asserts that Minches testified that "given the numbers, liquidation of Penn Treaty was likely the best decision.” Rehabilitator’s Proposed Findings of Fact at 94, ¶ 449 (citing N.T. 3/24/11 at 164-65). This is not completely accurate. Minches testified about an email he sent to Robinson, after the decision to seek liquidation had been made. Robinson asked Minches what the projections seemed to indicate at that time, and Minches responded in his email that liquidation seemed to be in order based on what was "explained” to him regarding the options available and the fact that none of the projections he had seen showed anything other than a negative surplus. N.T. 3/24/11 at 164-65.
Finally, the Rehabilitator describes Minches as a "life and health actuary,” but Minches stated he was not a health actuary. N.T. 3/24/11 at 205. He is a life actuary. Life actuaries specialize in life insurance and annuity products, for which the data are fixed and, thus, the work of a life actuary involves less judgment. (Volkmar) N.T. 10/25/11 at 106-08, 155. Health actuaries specialize in health insurance products, such as long-term care insurance, which requires the exercise of more judgment to do projections for rate making and reserve setting purposes. Id.
. Interestingly, Lucker is not even told by the state regulators for whom he reviews rate filings whether they are ultimately approved or disapproved. Lucker only knows the outcome of a rate increase filing if he happens to review a filing by the same company in the same state the following year, or if he researches the outcome online. Lucker neither confirmed how often he is actually aware of the outcome of rate increase filings he reviews nor demonstrated that he has ever known the specific reason why any state has denied a rate increase request he reviewed.
. Lucker's direct experience with PTNA rate filings in three states, Delaware, Iowa and Wisconsin, is even less compelling since those three states combined represent 3% or less of PTNA’s liabilities. ANIC has no policies in those states.
.Lucker acknowledged that he did not cite to any specific conversations with state regulators in his expert report to support his opinions and that he only relied on general conversations that he could not specifically identify. Lucker testified that he relied on general conversations because that is "how [he] gain[s] knowledge of the current regulatory environment.” N.T. 4/12/11 at 62.
. Holland testified that “organic brain syndrome” is a catch-all phrase used in medicine where there is no specific diagnosis. Some insurance carriers use the term to cover Par-kinsonian dementia, Alzheimer's disease, brain tumors, alcoholic encephalopathy and *426Creuzfeldt-Jacobs disease. N.T. 10/24/11 at 79.
. Exhibit R-1126 was offered into evidence by Intervenors to correct mathematical errors discovered on page 259 of Attachment 14.
. Intervenors note that the Rehabilitator’s own proposed findings of fact offer two different average ages for the Companies’ policyholder population. Rehabilitator’s Proposed Findings of Fact at 107, ¶ 526 states that the average age of an OldCo policyholder is 78 and the Rehabilitator’s Proposed Findings of Fact at 6, ¶ 13 states that the average age is 75 to 77. Intervenors contend that the evidence cited by the Rehabilitator supports neither finding. Intervenors' Reply Brief at 140-41.
. Some of Milliman’s judgments were directed by the Rehabilitator. For example, the Rehabilitator directed Milliman to assume that there would be no rate increases after ten years.
. “Desk drawer" rules are regulatory rules that have not been codified or formally adopted through regulatory proceedings. Pfannerstill testified that they can present an obstacle to rate filings because individual reviewers often enjoy wide latitude in interpreting their state's regulations and calculating justified increases. N.T. 3/22/11 at 188.
. Lucker’s testimony was offered to support the Rehabilitator’s position that a rehabilitation based on rate increases is futile. The Court also rejects Lucker's testimony as, alternatively, supporting the modest premium rate increase assumption used by Milliman in its July 2010 Surplus Report.
. SHIP, Senior Health Insurance Company of Pennsylvania, is a non-profit insurer owned by a trust created by the Pennsylvania Insurance Department to cover the long-term care insurance business originally written by several Conseco, Inc. companies. The trust was created because Conseco decided to leave the business and was willing to let the Pennsylvania insurance companies that wrote this business go into liquidation. (Ario) N.T. 2/11/11 at 29-31, 201-02. Commissioner Ario "worked hard” to get the rate increases needed to make the SHIP trust work. Id. at 202. Successive annual rate increases posited by Volkmar show, presumably, the manner of implementing a single rate filing.
. In health insurance, when policyholders are faced with significant rate increases, they find coverage elsewhere, leaving behind a pool of risk that gets smaller and less healthy with each rate increase. In turn, the premiums have to get higher and higher to reflect a smaller pool trying to cover the higher costs of insuring a high risk population. This leads to a death spiral where the higher rates kill the insurance mechanism of having a pool of risks large enough to sustain the losses.
. Section 518 was added by the Act of December 14, 1977, P.L. 280.
. In Legion, the rehabilitator asserted the insurer’s alleged "consent” to the conversion. The rehabilitator obtained this “consent” by amending the bylaws to give the rehabilitator all the power of the board of directors. Using this power, the rehabilitator consented to the liquidation on behalf of the insurer. By this sleight of hand, the rehabilitator, not the insurer, was the consenting party to the rehabil-itator’s requested liquidation. The Court rejected this device based on the language of Article V, explaining as follows:
The rehabilitation of these insurers may not be terminated unless and until the Court has evidence of record on which to make the factual findings that continued rehabilitation will substantially increase the risk of loss to policyholders, creditors and the public, or it is futile. These standards, set forth in Section 518(a) of Article V, are mandatory and may not be avoided even where the insurer, as opposed to the rehabilitator, consents to a termination of a rehabilitation.
Legion, 831 A.2d at 1229-30.
. Section 503 was added by the Act of December 14, 1977, P.L. 280.
. A long-term care insurance policy indemnifies against a type of casualty loss, such as entry into a nursing home. It is not a financial instrument such as a life insurance policy in which the policyholder accrues cash value and other benefits.
. See also Davis v. Amra Grotto M. O. v. P. E. R., Inc., 169 Tenn. 564, 89 S.W.2d 754, 758 (1936) (inequitable to limit a policyholder’s claim against an insolvent insurer estate to a premium refund).
. The Insurance Department has refused to allow claims against the estate that arise after the policies were cancelled. See Legion Insurance Company (In Liquidation) at http://www. legioninsurance.com/site/htmi ("Policy Cancellation”).
. Section 1903(a) of the Statutory Construction Act of 1972 states that "[w]ords and phrases shall be construed according to rules of grammar and according to their common and approved usage." 1 Pa.C.S. § 1903(a).
. In spite of this Court’s holding in Legion, 831 A.2d at 1230, that the burden of conversion is upon the statutory rehabilitator, the Rehabilitator argues otherwise. In this argument, the Rehabilitator relies upon a sentence or two from Sheppard.
Sheppard involved an insolvent insurer with a “history of insolvency,” "gross mismanagement” and "illegal business practices.” Id. at 598, 425 A.2d at 312. Accordingly, the commissioner sought a liquidation under Section 520, not a rehabilitation. The insurer defended against liquidation on grounds that it was not insolvent and, alternatively, it could be rehabilitated. In that context, the Supreme Court stated that the insurer "has the burden of showing rehabilitation is feasible.” Id. at 594, 425 A.2d at 310.
Sheppard is simply inapposite. First, the facts are completely different. There is no evidence whatsoever that the Companies’ insolvency resulted from gross mismanagement or illegal accounting devices. Second, the insurer defendant in Sheppard raised rehabilitation as an affirmative defense, on which, not surprisingly, it bore the burden of proof. Third, Sheppard was not a conversion proceeding brought under Section 518 of Article V. The Insurance Department sought a liquidation on grounds of insolvency, on which the Department had the burden of proof, pursuant to Section 520.
Sheppard does not apply to a conversion proceeding and, thus, does not support the Rehabilitator’s contention that Intervenors should prove they have a "feasible” rehabilitation plan. Because the word “futility” does *448not appear in Section 520 of Article V, the language in Sheppard referring to a "feasible” rehabilitation is simply irrelevant to the meaning of "futility.” Notably, the defendant insurer's suggested rehabilitation plan called for (1) writing new business, notwithstanding its insolvency, and (2) masking the insurer’s insolvency by employing illegal accounting devices. Not surprisingly, the defendant's so-called rehabilitation plan was rejected as a viable defense to a Section 520 liquidation petition.
. Intervenors believe that PTNA’s negative surplus is over $333 million, which the Reha-bilitator points out is more than twice what was represented to the Department and approximately 50% higher than was believed to be the case in the early months after rehabilitation was ordered. See (Volkmar) N.T. 10/27/11 at 32-34.
. Although the question has not been decided by this Court, or any court as far as can be determined, the Rehabilitator presumes that rate increases made a part of a rehabilitation plan will require state insurance department approval.
. Pfannerstill testified as follows regarding rate increase filings in Florida:
Florida, which has the most significant impact on Penn Treaty Network America has several statutes — excuse me, one statute which contains several restrictive provisions. First, they require you to aggregate all the company’s [long-term care] experience together to prove the need for a rate increase; in other words, exceeding that 60 percent. The reason that’s important is, in our case, you have to pool the experience of Newco business with Oldco to prove that, overall, the company needs a rate increase. Further, on top of that, since Penn Treaty isn't issuing new business, we’re restricted on the amount of the increase by a set of standard rates that are adopted by the state and set by the state. And those are maximum rates from which you cannot go beyond with your rate increase. And they go down to the individual rating cell level.
N.T. 3/22/11 at 183-84.
. See n. 42, supra.
. Section 516(d) states in relevant part:
The rehabilitator may prepare a plan for the reorganization, consolidation, conversion, reinsurance, merger or other transformation of the insurer. Upon application of the rehabilitator for approval of the plan, and after such notice and hearing as the court may prescribe, the court may either approve or disapprove the plan proposed, or may modify it and approve it as modified. If it is approved, the rehabilitator shall carry out the plan.
40 P.S. § 221.16(d). Section 516(d) was added by the Act of December 14, 1977, P.L. 280.
. Section 501 was added by the Act of December 14, 1977, P.L. 280.
. Section 501(c) was added by the Act of December 14, 1977, P.L. 280.
. One way to defeat any liquidation petition, whether filed under Section 518(a) or Section 520 of Article V, is to present a proposed rehabilitation plan. It is not necessary in a Section 518(a) proceeding where the Rehabil-itator bears the burden of proof. However, Volkmar’s Report proposed a series of rate increase scenarios, which constitute one type of rehabilitation plan.
. The Rehabilitator points to testimony that a transfer of PTNA NewCo policies to ANIC would probably not succeed given the asset challenge. (Robinson) N.T. 2/4/11 at 54-55; (Robinson) N.T. 2/14/11 at 22-25, 162-63; (Robinson) N.T. 2/15/11 at 208; (Mohoric) N.T. 2/18/11 at 35-36; (Mohoric) N.T. 2/24/11 at 185-86; (Pfannerstill) N.T. 3/22/11 at 117-18. The Rehabilitator points out that Interve-nors’ own actuary admitted that he was not offering any opinion that such a transfer could be part of a successful rehabilitation plan. (Volkmar) N.T. 10/27/11 at 185-86.
. Section 1703 was added by the Act of December 18, 1992, P.L. 1519.
. Pennsylvania's statute, which is presumably like its counterpart in other states, provides coverage to any residents who are insur-anee policyholders, see 40 P.S. § 991.1703(a)(2)(i), subject to the exclusion that the article "shall not provide coverage for any of the following ... (vi) any policy or contract issued in this Commonwealth by a member insurer at a time when it was not *457licensed or did not have a certificate of authority to issue such policy or contract in this Commonwealth.” See 40 P.S. § 991.1703(b)(2)(vi).
. The Wall Street Journal article included arresting photographs of a mouse suffering advanced Alzheimer's, in a cage strewn with paper scraps that the mouse could not assemble into a nest. A second photograph showed the same mouse after treatment enjoying a neat paper nest in the corner of the cage.