Terence J. Keating & Janet D. Keating

Opinion

United States Tax Court

T.C. Memo. 2024-2

TERENCE J. KEATING AND JANET D. KEATING, ET AL., 1 Petitioners

v.

COMMISSIONER OF INTERNAL REVENUE, Respondent

[*2] A. Contractual Insurance Obligations .......................................... 8

B. Commercial Insurance Policies ................................................ 8

1. Background ........................................................................ 8

2. Workers’ Compensation Policies ....................................... 9

3. Mr. Hill’s Brokering Efforts .............................................. 9

4. Commercial Policies in Effect During the Years at Issue ................................................................................. 10

III. Captive Insurance Program ........................................................... 13

A. Background ............................................................................. 13

B. Formation of Captive Insurance Program ............................. 14

C. Captive Owner Operations Manual ....................................... 15

D. Structure of Captive Insurance Program ............................... 17

E. Captive Policies ....................................................................... 19

1. Coverages, Policy Limits, and Premium Amounts......... 19

2. Coverage Selection and Policy Terms ............................. 26

F. Operations and Practices ........................................................ 27

1. Transaction Documentation Practices............................ 27

2. Underwriting Process, Premium Determination, and Premium Payments ......................................................... 29

3. Claims Handling .............................................................. 32

4. Related-Party Loans and Payments ............................... 37

5. Risk Retention ................................................................. 41

6. Capitalization of Provincial and Provincial Pool............ 42

IV. Dividends ........................................................................................ 43

3

[*3] V. Sale of RMS..................................................................................... 43

VI. Tax Reporting ................................................................................. 43

A. RMS ......................................................................................... 43

1. 2012 .................................................................................. 44

2. 2013 .................................................................................. 44

3. 2014 .................................................................................. 44

B. Risk Retention ......................................................................... 44

C. Petitioners ............................................................................... 45

OPINION ................................................................................................ 45

I. Evidentiary Matters ....................................................................... 45

II. Jurisdiction and Burden of Proof ................................................... 46

III. Credibility and Fact-Finding ......................................................... 48

IV. Microcaptive Arrangement ............................................................ 50

A. Whether the Arrangement Is Insurance ................................ 51

1. Commonly Accepted Notions of Insurance ..................... 53

2. Conclusion ........................................................................ 64

B. Effect on Petitioners................................................................ 64

1. Section 162 ....................................................................... 64

2. Section 165 ....................................................................... 66

3. Dividends ......................................................................... 70

V. Accuracy-Related Penalties............................................................ 73

4

[*4] MEMORANDUM FINDINGS OF FACT AND OPINION

MARVEL, Judge: During the years 2012–14 (years at issue), petitioners Terence J. Keating, Cheryl L. Doss, and Arthur D. Candland 2 were shareholders of Risk Management Strategies, Inc. (RMS), an S corporation in the business of acting as a sole employer for its clients, which were primarily banks administering special needs trusts. 3 RMS assumed the employer liability resulting from the employment of caregivers who worked for special needs trusts, handled payroll, and generally carried out the responsibilities of being an employer to caregivers and other employees that would have otherwise fallen on its clients. For each year at issue RMS reported incurring approximately $1.2 million of expenses for purported insurance coverage provided through an arrangement among its affiliated captive insurance company, Risk Retention, Ltd. (Risk Retention), and other entities.

Respondent contends, among other things, that this arrangement did not actually provide insurance and that petitioners cannot deduct the amounts that RMS paid for the purported insurance and related fees nor take advantage of a preferential rate for dividends paid by Risk Retention. Respondent also contends that petitioners are liable for accuracy-related penalties. Petitioners disagree, arguing that the deductions and preferential dividend rate were proper because the arrangement provided insurance. They also assert a reasonable-causeand -good-faith defense to the accuracy-related penalties. We agree with respondent that the challenged deductions and preferential dividend rate were improper and that accuracy-related penalties are appropriate.

On May 11, 2018, respondent determined deficiencies in petitioners’ federal income tax and accuracy-related penalties under section 6662(a) 4 as follows:

2 Petitioners Janet D. Keating and Michelle M. Candland had no involvement

in the transactions at issue in these cases, and we do not discuss them further.

3 In addition, some of its employees provided services to grantor trusts, family

offices, and other entities.

4 Unless otherwise indicated, statutory references are to the Internal Revenue

Code, Title 26 U.S.C. (Code), in effect at all relevant times, regulation references are to the Code of Federal Regulations, Title 26 (Treas. Reg.), in effect at all relevant times, and Rule references are to the Tax Court Rules of Practice and Procedure. Some monetary amounts have been rounded to the nearest dollar.

5

[*5] Docket No. 15066-18—Terence J. Keating and Janet D. Keating

Year Deficiency § 6662(a) Penalty

2012 $274,039 $54,808

2013 244,578 48,916

2014 317,682 63,536

Docket No. 15067-18—Cheryl L. Doss

Year Deficiency § 6662(a) Penalty

2012 $18,039 $3,608

2013 21,299 4,260

2014 21,510 4,302

Docket No. 15068-18—Arthur D. Candland and Michelle M. Candland

Year Deficiency § 6662(a) Penalty

2012 $287,535 $57,507

2013 244,578 48,916

2014 312,275 62,455

Petitioners timely filed Petitions in these cases on August 2, 2018, contesting respondent’s determinations. These cases were consolidated pursuant to Rule 141 for purposes of trial, briefing, and opinion.

The issues for decision are (1) whether transactions conducted through a purported microcaptive insurance arrangement among RMS, Risk Retention, and other entities during the years at issue constitute insurance for federal income tax purposes; (2) whether expenses RMS incurred during the years at issue (a) through the purported

6

[*6] microcaptive insurance arrangement or (b) to Artex Risk Solutions, Inc. (Artex), or PRS Insurance (PRS) for services rendered in connection with the arrangement constitute ordinary and necessary business expenses deductible under section 162; (3) if not, whether any of those expenses are deductible as losses under section 165; (4) whether dividends paid by Risk Retention to Mr. Keating and Mr. Candland in their 2012 and 2014 taxable years are qualified dividends or ordinary dividends; and (5) whether petitioners are liable for accuracy-related penalties imposed under section 6662(a) for the years at issue. We also address deferred evidentiary rulings.

FINDINGS OF FACT

Some of the facts have been stipulated and are so found. The First, Second, Third, and Fourth Stipulations of Facts and the accompanying Exhibits are incorporated herein by this reference. Petitioners resided in California when they filed their Petitions. 5 Use of the terms “insurance,” “insurer,” “insured,” “policy,” “premium,” “claim,” “reinsurance,” “reinsurer,” and other insurance-related terms in this Opinion replicate the terminology used by the parties throughout the litigation and do not imply that we have determined that any financial arrangement constitutes insurance, or that any company is an insurance company, as a matter of fact or law for federal income tax purposes.

I. RMS

A. Background

Mr. Keating and Mr. Candland incorporated RMS, also known as Trust Employee Administration & Management or TEAM, in 2003. At the time of incorporation, they split RMS’s ownership evenly, with each owning 50% of RMS’s stock.

RMS’s primary business was the employment, administration, and management of service providers for the benefit of trusts. Specifically, RMS acted as the sole employer for caregivers, guardians, case managers, household staff, and others who provided services for special needs trusts, grantor trusts, family offices, and other entities. Most of RMS’s employees provided services to special needs trust beneficiaries. RMS contracted its services primarily to national banks,

5 Unless otherwise agreed by the parties in writing, venue for an appeal is the

U.S. Court of Appeals for the Ninth Circuit. See § 7482(b)(1)(A).

7

[*7] and specifically to their wealth management and private banking departments. RMS also provided payroll and human resources services, benefits administration, and legal consultation.

RMS had contracts with its bank clients in their capacities as trustees. Pursuant to these contracts, the parties agreed that it was their “mutual intention” that “RMS shall do all acts necessary to employ individuals who will be the employees of RMS . . . and that Trustee shall not in any manner be deemed to be the employer of such persons, whether in its corporate or fiduciary capacity.” The contracts also obligated RMS to secure and maintain “workers’ compensation benefits, unemployment insurance and the like,” as well as commercial general liability insurance. An exhibit to the contracts disclosed applicable service fees, including amounts for payroll taxes and insurance, amounts for benefits chosen and paid for by the trustee, and a monthly administrative fee.

After the enactment of the Affordable Care Act, RMS had to offer health insurance coverage to its employees. RMS could not obtain a guaranteed-cost group health plan because many of its employees were parents of disabled trust beneficiaries. Instead, RMS formed a voluntary employees’ beneficiary association (VEBA) to provide health coverage to its employees and contracted with a stop-loss insurance carrier and a claims administrator.

B. Petitioners’ Roles at RMS

In 2003 Mr. Candland and Mr. Keating recruited Ms. Doss to work at RMS. In 2005 or 2006 Mr. Candland and Mr. Keating gave Ms. Doss a 5% stake in RMS in the form of nonvoting stock in the corporation. She eventually became director of client services for RMS, a role she held during the years at issue.

During the years at issue RMS was a California corporation and had a valid S corporation election in effect with the Internal Revenue Service (IRS) pursuant to section 1362. RMS was owned 47.5% by Mr. Keating, 47.5% by Mr. Candland, and 5% by Ms. Doss during the years at issue. Mr. Keating was president of RMS and oversaw operations, including managing payroll, human resources, and customer relations. Mr. Candland was the chief financial officer of RMS. Ms. Doss served as director of client services and assisted with onboarding new clients, paperwork completion, and dealing with the state agencies that licensed RMS.

8

[*8] RMS contracted with Charter Management Services, Inc. (Charter), a California corporation formed in November 2003 and owned by Mr. Keating, Mr. Candland, and Ms. Doss, during the years at issue to provide administration and management services to RMS and to serve as the employer of the staff handling the day-to-day operations of RMS. Charter handled payroll, human resources, and benefits and administration functions for RMS’s employees. Charter also had a legal department. Ms. Doss oversaw Charter’s employees, who included payroll specialists, human resources employees, and accountants.

II. Commercial Insurance Coverage

A. Contractual Insurance Obligations

In its service contracts with its clients during the years at issue, RMS agreed that it would secure and maintain “workers’ compensation benefits, unemployment insurance and the like” and “commercial general liability insurance.” Specifically, RMS agreed that it would maintain at least the following coverages (with specified minimum policy limits): commercial general and professional liability, including personal injury; nonowned automobile liability; workers’ compensation and employer’s liability; employment practices liability insurance; and third-party fidelity coverage. No other insurance coverage was specifically required by the contracts.

B. Commercial Insurance Policies

1. Background

During the years at issue RMS worked with BB&T, an insurance brokerage, to purchase insurance policies in the commercial marketplace. Specifically, Mr. Keating and Mr. Candland worked with John Hill and Geoff Shelton, insurance brokers at BB&T. Mr. Hill, a commercial property and casualty insurance broker at BB&T, is a certified insurance counselor and an accredited advisor of insurance. 6 Mr. Hill acts as an intermediary between policyholders and insurance companies and assists with negotiating insurance coverages and managing insurance programs.

Mr. Hill first met Mr. Candland and Mr. Keating when they asked Mr. Shelton to help with their insurance program sometime in the mid-2000s. Mr. Shelton then asked Mr. Hill to assist with RMS’s

6 Mr. Hill testified only as a fact witness, however, not as an expert witness.

9

[*9] account. When Mr. Hill first started brokering insurance for RMS, it had a series of workers’ compensation policies through individual state insurance funds. 7 Mr. Hill was able to find one workers’ compensation policy for RMS that replaced approximately 17 independent state fund policies.

2. Workers’ Compensation Policies

From at least July 1, 2006, to July 1, 2008, RMS had a workers’

compensation policy with the Employers Insurance Co. of Wausau (Wausau). The Wausau workers’ compensation policy was a retrospectively rated policy. 8 Beginning in July 2008 and continuing through the years at issue, RMS had workers’ compensation insurance with Crum & Forster through United States Fire Insurance Co. Crum & Forster applied discounts to RMS’s workers’ compensation premiums for large deductibles, schedule modifications, and loss experience, among other items.

3. Mr. Hill’s Brokering Efforts

During the years at issue Mr. Hill sought insurance coverage for RMS in both the retail market and the wholesale market. 9 Mr. Hill is a retail broker. Retail brokers have a direct relationship with a policyholder and standard insurance companies but must engage with wholesale brokers to access the wholesale market. Mr. Hill provided RMS with marketing summaries, which summarized each insurance company approached for certain insurance coverages and the results of those efforts, annually throughout the years at issue. We discuss RMS’s commercial insurance policies immediately below.

7 State insurance funds are insurance carriers of last resort when the private

marketplace is unable or unwilling to provide the coverage.

8 With a retrospectively rated policy, the insurance company collects a deposit

premium, but at the end of the policy period it performs a calculation based on the number and cost of claims that occurred during the policy period. If claims are lower than anticipated, then the policyholder receives money back. If claims are higher than anticipated, then the policyholder shares the cost of those claims by paying additional sums to the insurance company. The deposit premium for the July 2006–July 2007 Wausau policy was $728,303, and the deposit premium for the July 2007–July 2008 Wausau policy was $729,307.

9 We sometimes refer to policies obtained in either of these markets as

commercial insurance policies.

10

[*10] 4. Commercial Policies in Effect During the Years at Issue

RMS purchased the following commercial insurance policies that were in effect during the years at issue:

Coverage Type Insurer Coverage Period Premiums and Fees

Crime Chubb “2011-2012” 10 Unknown

Crime Chubb June 1, 2012, to $3,644 June 1, 2013

Crime Chubb “2013-2014” 3,910

Crime Chubb “2014-2015” 4,447

Cyber liability Chubb January 20, 2011, 39,308 to June 1, 2012

Cyber liability Lloyd’s of June 1, 2012, to 25,614 London June 1, 2013

Cyber liability Lloyd’s of June 1, 2013, to 18,628 London June 1, 2014

Cyber liability Lloyd’s of June 1, 2014, to 21,113 London June 1, 2015

Employment Lloyd’s of June 1, 2011, to 43,750 practices London through June 1, 2012 liability Beazley Insurance Co.,

Inc.

10 Only one crime policy for the years at issue (with a coverage period of June 1,

2012, to June 1, 2013) is in the record. For the other three policies, the parties have stipulated that they had coverage periods of “2011-2012”, “2013-2014”, and “2014- 2015”, respectively. The record does not provide any additional detail about what that means.

11

[*11] Coverage Type Insurer Coverage Period Premiums and Fees

Employment Lloyd’s of June 1, 2012, to 63,882 practices London through June 1, 2013 liability Beazley Insurance Co.,

Inc.

Employment Lloyd’s of June 1, 2013, to 73,500 practices London through June 1, 2014 liability Beazley Insurance Co.,

Inc.

Employment Lloyd’s of June 1, 2014, to 75,000 practices London through June 1, 2015 liability Beazley Insurance Co.,

Inc.

Excess Nautilus June 1, 2011, to 31,617 liability Insurance Co. June 1, 2012

Excess Nautilus June 1, 2012, to 36,824 liability Insurance Co. June 1, 2013

Excess Nautilus June 1, 2013, to 37,475 liability Insurance Co. June 1, 2014

Excess Nautilus June 1, 2014, to 37,611 liability Insurance Co. June 1, 2015

General Nautilus June 1, 2011, to 12,047 liability Insurance Co. June 1, 2012

General Nautilus June 1, 2012, to 14,217 liability Insurance Co. June 1, 2013

General Nautilus June 1, 2013, to 14,601 liability Insurance Co. June 1, 2014

General Nautilus June 1, 2014, to 14,641 liability Insurance Co. June 1, 2015

12

[*12] Coverage Type Insurer Coverage Period Premiums and Fees

Professional Nautilus June 1, 2011, to 51,812 liability Insurance Co. June 1, 2012

Professional Nautilus June 1, 2012, to 60,143 liability Insurance Co. June 1, 2013

Professional Nautilus June 1, 2013, to 61,471 liability Insurance Co. June 1, 2014

Professional Nautilus June 1, 2014, to 61,635 liability Insurance Co. June 1, 2015

Property Greenwich June 1, 2011, to 2,577 Insurance Co. June 1, 2012

Property Greenwich June 1, 2012, to 3,387 Insurance Co. June 1, 2013

Property Greenwich June 1, 2013, to 3,451 Insurance Co. June 1, 2014

Property Greenwich June 1, 2014, to 3,435 Insurance Co. June 1, 2015.

Workers’ Crum & Forster July 1, 2011, to 536,590 11 compensation through United July 1, 2012 States Fire Insurance Co.

Workers’ Crum & Forster July 1, 2012, to 424,242 12 compensation through United July 1, 2013 States Fire Insurance Co.

11 This figure is after the policy’s annual audit. Before the annual audit, the total cost for the policy was $487,216.

12 This figure is after the policy’s annual audit. Before the annual audit, the

estimated annual cost for the policy was $378,734.

13

[*13] Coverage Type Insurer Coverage Period Premiums and Fees

Workers’ Crum & Forster July 1, 2013, to 500,615 13 compensation through United July 1, 2014 States Fire Insurance Co.

Workers’ Crum & Forster July 1, 2014, to 563,334 14 compensation through United July 1, 2015 States Fire Insurance Co.

III. Captive Insurance Program

A. Background

In addition to its commercial insurance, RMS was the insured on several purported insurance policies maintained through a captive insurance program 15 during the years at issue. RMS’s captive insurance program began in 2008. In an email exchange with an external auditor on June 18, 2012, Mr. Candland described the captive insurance program as “RMS self-insur[ing]” (emphasis added) workers’ compensation claims although the scope of the captive program was not limited to policies relating to workers’ compensation. In a later email exchange with a potential buyer of RMS on October 24, 2012, Mr. Candland also described the captive insurance program as a vehicle for funding workers’ compensation and liability insurance deductibles and for covering esoteric risks that either could not be covered commercially or had such low risks that it made no sense to purchase them

13 This figure is before the policy’s annual audit, unlike the figures for the

2011–12 and 2012–13 policies. An October 16, 2014, email from a Crum & Forster representative to Mr. Candland referencing an attached audit statement for this policy is in the record, but the record does not disclose the results of the audit.

14 This figure represents the estimated annual cost for the policy before the

annual audit, unlike the figures for the 2011–12 and 2012–13 policies. The results of the annual audit are not in the record.

15 We sometimes also refer to the captive insurance program interchangeably

as the captive program, the captive arrangement, the microcaptive arrangement, the captive insurance arrangement, or the microcaptive insurance arrangement.

14

[*14] commercially. He explained that “[t]ypically we pay premium of just under $1[,]200,000 per year.”

B. Formation of Captive Insurance Program

Mr. Candland had at least three discussions with Ken Kotch (Mr.

Kotch), a vice president at Tribeca Strategic Advisors, LLC (Tribeca), from May to October 2008. Mr. Candland did not believe Mr. Kotch was qualified to discuss or underwrite insurance risks. Mr. Candland’s handwritten notes from these discussions focus on the topics of federal income taxation, fees, and formation of the captive insurer (including Anguillan 16 regulatory requirements), but none of the notes contain any description of insurance needed by RMS. Mr. Candland wrote down that “upon termination of the captive the funds return as capital gains” and referred to “[section] 831(b) captives” (i.e., microcaptive insurers), as well as to the fact that Mr. Kotch was “an . . . [attorney with] emphasis in taxation[.]” The notes also refer to IRS Revenue Rulings 2002-89, 2002-90, and 2002-91 (concerning risk distribution for insurance companies) and to an IRS “safe harbor” for risk distribution. In addition, the notes reflect Mr. Candland’s understanding that $1.2 million was “the max[imum] we can put into [the] captive” insurer each year and that “of the funds deposited to the captive, 51% will go into [a risk pool] for 366 days [and] then be transferred to [the] captive. 49% will stay in [the] captive [and] we can invest [those funds].”

Contemporaneously with these discussions, Tribeca prepared a feasibility study for RMS dated August 27, 2008. 17 According to the feasibility study, RMS was motivated to create the captive in part because it wanted “platinum-level coverage” and was willing to pay “platinum-level premiums” for that coverage. However, the feasibility study also stated that one of the advantages of forming a captive insurer was the elimination or reduction of certain costs that commercial insurers face and predicted that a captive insurer could generate expense savings of up to 35% of the costs of conventional insurance. The feasibility study contained financial forecast models for RMS that assumed RMS would have no direct insured claim losses, nor any claims against the risk pool (described below), for the first six years of

Anguilla is an island of the British West Indies. 16 See Monahan v.

Commissioner, 109 T.C. 235, 236 (1997).

17 Two different versions of the feasibility study are in the record although the

differences are immaterial for purposes of this discussion, and we refer to them as a single study.

15

[*15] operations. Instead, the models assumed that RMS would have annual pretax income of $2 million and pay annual captive insurance premiums of $1.2 million each year for six years and that the captive would provide RMS with a total net benefit of $3,279,823 over six years. This net benefit was derived from (1) savings on the amount of income taxes paid and (2) having greater assets available for investment in each year beginning with the second year because of the decrease in income taxes paid and a lack of claims. The models included no estimate of savings from commercial insurance expenses despite the study’s statement that such savings were potentially a significant advantage from using a captive insurer. The feasibility study identified policies that RMS never purchased, such as ones covering goodwill or identity protection, as among “the most likely to be incorporated into a new captive insurance program.” Conversely, the study omitted any mention of policies relating to workers’ compensation, which RMS did purchase.

On October 31, 2008, Mr. Candland and Mr. Keating signed an engagement letter on behalf of RMS agreeing to pay Tribeca $40,000 to form a captive insurer, Risk Retention. 18 The feasibility study was submitted to Anguilla regulators as part of Risk Retention’s license application along with, inter alia, a business plan. The business plan states that Risk Retention would “underwrite highly customized policies carefully tailored to the specific needs of its insured,” but it also repeatedly refers to Risk Retention’s intended insured erroneously as “GTI” rather than RMS.

Risk Retention was formed in November 2008. 19 On November 25, 2008, the Anguilla Financial Services Commission issued Risk Retention a Class B Insurance License following an application by Risk Retention. The Anguilla Financial Services Commission renewed the license for each of the years at issue.

During the years at issue Risk Retention had no employees.

C. Captive Owner Operations Manual

Tribeca provided a Captive Owner Operations Manual (Owners’

Manual) to Mr. Candland and Mr. Keating on May 12, 2009. The

18 Risk Retention is not a party to these cases.

19 Risk Retention was initially capitalized with $100,000, and it maintained its

$100,000 paid-in capital during the years at issue. Risk Retention had signed bylaws in effect as of November 24, 2008.

16

[*16] Owners’ Manual set out the responsibilities of the owner of a captive insurer and noted that while Tribeca managed a captive insurer, the owners of the captive insurer had ultimate decision-making authority.

The Owners’ Manual stated that Tribeca could not properly underwrite captive insurance policies without a completed underwriting application for each insured. It also stated that captive owners should determine the amount of coverage and premiums for the next policy year before the end of the current policy year. Tribeca stated in the Owners’ Manual that it required owners of captive insurers to notify it of any changes to the amount of premiums paid to the captive insurer for the current policy year by November 15 and that it “strongly recommends that premiums be paid during the policy period on a regular schedule, and not after the end of the policy period.” In particular, Tribeca stated that “[i]n traditional insurance companies, premiums are usually paid in monthly, quarterly or annual payments. Premiums are usually considered due either before the policy period begins or in equal installments during the policy period.” Tribeca stated that any premium payments due must be sent by December 31 of the coverage year and received by January 8 of the following year.

Tribeca strongly discouraged owners of captive insurers from using their captive insurers to make loans, especially loans from the captive insurer to an insured or affiliated party. The Owners’ Manual stated that loans to related parties could increase the likelihood that the IRS would find that a captive insurer was a sham or that a circular flow of cash existed. For captive insurers that decided to make loans despite Tribeca’s advice, the Owners’ Manual advised that loans must be evidenced by a promissory note or other written document; must be enforceable; must contain commercially reasonable repayment terms and interest rates; should be secured; and must be repaid timely. Tribeca also stated that “it is critical that we receive full documentation on all transactions involving the Captive and that you notify us in advance regarding any proposed transaction . . . or movement of funds involving the Captive.” It advised owners of captive insurers to “strictly follow the policies of this manual” in view of legal authority taking into account all of the facts and circumstances in determining what constitutes insurance or an insurance company.

Regarding claims handling, the Owners’ Manual stated that if an insured incurred a claim, it should notify Tribeca of the claim in writing. The Owners’ Manual also stated that Tribeca would provide the insured

17

[*17] with a claim form, that the insured would need to submit supporting documentation to substantiate the loss, and that the insured “should make a claim for any loss covered by insurance.”

D. Structure of Captive Insurance Program

The purported captive insurance arrangement between Risk Retention and RMS did not primarily involve Risk Retention simply issuing insurance policies to RMS. Instead, two different general structures were used, one from 2008 to 2010 and the other from 2011 to 2014. Both structures shared commonalities, including that Risk Retention participated in a risk pool with other captive insurers managed by Tribeca or, later, Artex. 20 However, the structures varied in other respects. We describe the structure used from 2008 to 2010 first because it forms the basis for our discussion of the structure later used during the years at issue. Except as otherwise indicated, our findings of fact in this subsection concern the structures set forth by the transaction documentation and do not address other relevant practices by the parties to the arrangement.

From 2008 to 2010 RMS and other insureds of Tribeca-managed captive insurers participating in the risk pool purchased (1) a primary (or direct) layer of purported insurance coverage for each insured risk directly from their respective captive insurer and (2) an excess (or quota-share) layer of purported insurance coverage for each same risk from Procedant Insurance Co., Inc., a Nevada insurer that we do not discuss further, or Provincial Insurance, PCC (Provincial), a fronting insurer organized, licensed, and domiciled in Anguilla as of December 2009. 21 Tribeca allocated the total net premiums received from each insured approximately 49% to the primary layer and approximately 51% to the excess layer. An agent, PRS, collected payment from the insured,

20 Arthur J. Gallagher & Co., Inc. (Gallagher), an insurance brokerage and risk

management services firm based in Illinois, acquired the assets of Tribeca in 2010. The acquisition was announced on December 21, 2010. While the record is unclear as to whether the acquisition had also closed by December 21, 2010, we generally refer to Artex rather than Tribeca with respect to events occurring after this date. After the acquisition, the operations formerly conducted by Tribeca continued out of its Mesa, Arizona, office under the direction of Gallagher’s wholly owned subsidiary, Artex.

21 The record discloses that Provincial was organized in the British Virgin

Islands as Provincial Insurance, Ltd, before its organization in Anguilla in December 2009.

18

[*18] retained a 2.5% administrative fee, and transmitted to the captive insurer and Provincial the net amounts owed to them.

Considering the primary and excess layers of coverage and their policy limits together, the primary layer insured any covered loss up to 25% of the combined policy limits, and the excess layer insured the portion of any loss exceeding the primary coverage, subject to a cap equal to 75% of the combined policy limits (sometimes described as “75% x/s 25%” or 75%-in-excess-of-25% coverage). Therefore, a smaller covered loss might be completely covered by the primary layer, while the excess layer applied to relatively larger covered losses and would pay out the lesser of (1) the portion of a loss exceeding the policy limits of the primary layer or (2) its own policy limit (which was triple the amount of the primary layer’s policy limit and therefore constituted 75% of the combined policy limits).

Regarding the excess layer, Provincial ceded the risks and premiums from the excess layer (also known as quota-share risks and quota-share premiums, respectively) to each of the captive insurers participating in a risk pool. The risk pool was a purported reinsurance arrangement conducted pursuant to Master Reinsurance Contracts or Master Reinsurance Agreements (each also known as quota-share agreements). The risk pool is known as the Provincial Pool.

Each captive insurer participating in the Provincial Pool, including Risk Retention, bore a fixed quota-share percentage of any loss covered by the excess coverage and was allotted the same quota-share percentage of the premiums allocated to the Provincial Pool. 22 A captive insurer later received from Provincial its quota share of the premiums remaining after reduction by its quota share of any claims allowed against the excess layer coverage. From at least 2009 to 2014 Risk Retention’s quota share of pool premiums was equal to the net premiums Provincial received from RMS for excess coverage. Therefore, if there were no allowed claims or other withheld amounts, 23 Risk Retention would receive the same amount from the Provincial Pool as RMS had paid Provincial for excess coverage (net of the 2.5% administrative fee).

22 The quota-share percentage was calculated by computing the ratio of the

premiums paid by a captive insurer’s related insured (here, RMS) for excess coverage to the total premiums received from all insureds by the Provincial Pool.

23 As discussed below, there were eventually some allowed claims and withheld

amounts.

19

[*19] The structure used from 2011 to 2014 retained this basic model with some modifications. First, Artex replaced Tribeca as manager of the captive insurers after the latter’s assets were acquired by the former’s parent company in December 2010.

Second, insureds no longer purchased a primary layer of coverage directly from their affiliated captive insurer but instead purchased a socalled facultative layer of coverage from Provincial. Provincial then ceded the associated risks and premiums (sometimes called facultative risks and facultative premiums, respectively) to the affiliated captive insurer under Reinsurance Contracts (also known as Facultative Reinsurance Contracts or Facultative Reinsurance Agreements). Risk Retention remained responsible for all of the losses allowed under the facultative layer of coverage, albeit through a Facultative Reinsurance Contract or Agreement rather than by directly issuing insurance policies to RMS. Artex still allocated approximately 49% of total net premiums to Risk Retention for this coverage, and Provincial wired these premiums to Risk Retention within two weeks of receipt. Approximately 51% of total net premiums remained allocable to the Provincial Pool. Finally, Artex made changes in its practices that we discuss further below.

E. Captive Policies

1. Coverages, Policy Limits, and Premium Amounts

We summarize RMS’s captive coverages, policy limits, and premium amounts for the years at issue here. The captive insurance policies for the years at issue were all claims-made policies, meaning that they applied only to claims reported during the coverage period or extended reporting period. The general terms and conditions common to all Provincial policies during the years at issue included a 45-day extended reporting period, among other terms. The policies in effect during the years at issue all had coverage periods running from January 1 to the following January 1.

RMS paid premiums approximating $1.2 million for each year at issue. Below, we set forth charts outlining (1) RMS’s captive coverages and policy limits for the years at issue and (2) the premium amounts applicable to each coverage for the years at issue. Both in these charts and throughout the rest of this Opinion, we refer to the concept of a self-insured retention (SIR), which is a dollar amount specified in an insurance policy that must be paid by the insured before the insurance

20

[*20] policy will respond to a loss. SIRs generally operate slightly differently from deductibles, such as with respect to how they erode the policy limit or whether the insurer has an obligation for indemnity and defense costs before the deductible or SIR is paid, but they serve a similar overall function and purpose.

a. 2012

The following chart shows the coverages and policy limits in the Risk Retention captive insurance program for 2012:

Coverage Self-Insured Total Policy Facultative Pool Limit Retention Limit Policy Limit

Administrative $250,000 $750,000 — $750,000 actions

Employment — 100,000 $100,000 — practices deductible / SIR reimbursement

Legal expense — 1,000,000 250,000 750,000

Loss of key 250,000 750,000 — 750,000 contract

Loss of key — 1,000,000 250,000 750,000 customer

Professional — 1,000,000 250,000 750,000 liability difference in conditions

Worker’s [sic] — 100,000 100,000 — compensation deductible / SIR reimbursement

Total $500,000 $4,700,000 $950,000 $3,750,000

21

[*21] The following chart shows the premium amounts applicable to these coverages for 2012:

Coverage Gross Administrative Net Facultative Pool Premium Fee Premium Premium Premium

Administrative $157,281 $3,932 $153,349 — $153,349 actions

Employment 145,500 3,638 141,862 $141,862 — practices deductible / SIR reimbursement

Legal expense 175,051 4,376 170,675 83,631 87,044

Loss of key 167,340 4,184 163,156 — 163,156 contract

Loss of key 194,294 4,857 189,437 92,824 96,613 customer

Professional 183,723 4,593 179,130 87,774 91,356 liability difference in conditions

Worker’s 167,500 4,188 163,312 163,312 — compensation deductible / SIR reimbursement

Total $1,190,689 $29,768 $1,160,921 $569,403 $591,518

22

[*22] b. 2013

The following chart shows the coverages and policy limits in the Risk Retention captive insurance program for 2013:

Coverage Self-Insured Total Policy Facultative Pool Limit Retention Limit Policy Limit

Administrative $250,000 $750,000 — $750,000 actions

Employment — 100,000 $100,000 — practices deductible / SIR reimbursement

General liability — 50,000 50,000 — deductible / SIR reimbursement

Legal expense — 1,000,000 250,000 750,000

Loss of key 250,000 750,000 — 750,000 contract

Loss of key 250,000 750,000 — 750,000 customer

Professional — 1,000,000 250,000 750,000 liability difference in conditions

Worker’s — 100,000 100,000 — compensation deductible / SIR reimbursement

Total $750,000 $4,500,000 $750,000 $3,750,000

23

[*23] The following chart shows the premium amounts applicable to these coverages for 2013:

Coverage Gross Administrative Net Facultative Pool Premium Fee Premium Premium Premium

Administrative $137,699 $3,442 $134,257 — $134,257 actions

Employment 159,500 3,988 155,512 $155,512 — practices deductible / SIR reimbursement

General liability 89,000 2,225 86,775 86,775 — deductible / SIR reimbursement

Legal expense 171,287 4,282 167,005 81,832 85,173

Loss of key 146,506 3,663 142,843 — 142,843 contract

Loss of key 158,715 3,968 154,747 — 154,747 customer

Professional 179,773 4,494 175,279 85,887 89,392 liability difference in conditions

Worker’s 181,500 4,538 176,962 176,962 — compensation deductible / SIR reimbursement

Total $1,223,980 $30,600 $1,193,380 $586,968 $606,412

24

[*24] c. 2014

The following chart shows the coverages and policy limits in the Risk Retention captive insurance program for 2014:

Coverage Self-Insured Total Policy Facultative Pool Limit Retention Limit Policy Limit

Administrative $250,000 $750,000 — $750,000 actions

Employment — 400,000 400,000 — practices deductible / SIR reimbursement

General liability — 100,000 100,000 — deductible / SIR reimbursement

Legal expense 250,000 750,000 — 750,000

Loss of key 250,000 750,000 — 750,000 contract

Professional — 1,000,000 250,000 750,000 liability difference in conditions

Regulatory 250,000 750,000 — 750,000 change

Worker’s 100,000 1,000,000 1,000,000 — compensation deductible / SIR reimbursement

Total $1,100,000 $5,500,000 $1,750,000 $3,750,000

25

[*25] The following chart shows the premium amounts applicable to these coverages for 2014:

Coverage Gross Administrative Net Facultative Pool Premium Fee Premium Premium Premium

Administrative $118,923 $2,973 $115,950 — $115,950 actions

Employment 159,500 3,988 155,512 155,512 — practices deductible / SIR reimbursement

General liability 89,000 2,225 86,775 86,775 — deductible / SIR reimbursement

Legal expense 134,851 3,371 131,480 — 131,480

Loss of key 140,329 3,508 136,821 — 136,821 contract

Professional 157,290 3,932 153,358 75,145 78,213 liability difference in conditions

Regulatory 135,275 3,382 131,893 — 131,893 change

Worker’s 255,543 6,389 249,154 249,154 — compensation deductible / SIR reimbursement

Total $1,190,711 $29,768 $1,160,943 $566,586 $594,357

As shown in the tables above, throughout the years at issue Artex increasingly allocated risks and premiums from individual RMS policies solely either (1) to the Provincial Pool (i.e., to pool premium and the pool

26

[*26] limit) or (2) to Risk Retention (i.e., to facultative premium and the facultative policy limit), and it increasingly used SIRs. 24

2. Coverage Selection and Policy Terms

Mr. Hill, RMS’s commercial insurance broker, did not shop for insurance policies covering administrative actions, loss of a key customer, or regulatory change because Mr. Candland never asked him to shop for these policies in the commercial marketplace. Mr. Hill did not know what administrative action or loss of key customer policies were. Mr. Candland did not direct Mr. Hill to seek out a zero-dollar deductible workers’ compensation policy in the commercial marketplace. 25

During the years at issue the general terms and conditions common to all Provincial policies contained a number of coverage exclusions, including for claims that were the subject of any notice given under other insurance before the inception date of the policies; for claims based upon circumstances or events that any insured knew about before the policy period; for criminal, dishonest, or deliberately fraudulent acts, including sexual abuse or molestation or fraud of any insured; and for personal profit, remuneration, or advantage gained by any insured to which it was not legally entitled.

We pause to discuss some terms and context regarding RMS’s Worker’s Compensation Deductible / SIR Reimbursement policy given

24 In an email dated August 2, 2011, an Artex underwriter explained that deductible reimbursement policies were “no longer being insured through the pool.” However, without adjustment, reinsuring deductible reimbursement policies entirely with Risk Retention would have caused over 49% of Risk Retention’s premium volume to come from RMS and under 51% to come from reinsuring its quota-share percentage of the Provincial Pool. To address this perceived problem, Artex “needed to put 2 other policies 100% in the pool to achieve [the] 49%/51% split that the IRS likes to see for risk distribution.” For 2011 and 2012 the policies whose risks and premiums were allocated completely to the Provincial Pool were the Administrative Actions and Loss of Key Contract policies. This, however, could have exposed the Provincial Pool to small claims on those policies that previously would have been retained in the primary coverage layer. Therefore, in order “[t]o protect the pool in these instances,” Artex added a $250,000 SIR to each policy that “mimics the 25%/75% limit split the standard structure would have between the captive and the pool.”

25 In fact, Mr. Candland emailed an Artex employee during the years at issue

to inform him that RMS had raised its deductible from $100,000 to $250,000 on the Crum & Forster policy and stated: “This should make it easier to justify our $1,200,000 captive contribution.”

27

[*27] its importance to the issues in these cases. This policy covered losses within the deductible or SIR of RMS’s commercial workers’ compensation policy.

Artex did not separately adjust underlying workers’

compensation claims because RMS’s commercial carrier was responsible for the settlement of claims and then billed RMS for amounts within the deductible. Instead, RMS filed claims under its workers’ compensation policy with Crum & Forster, its commercial workers’ compensation insurance carrier. Crum & Forster adjusted each claim and invoiced RMS monthly for the deductible portion of any approved claims. When RMS received a monthly deductible billing invoice from Crum & Forster, Risk Retention paid it on RMS’s behalf by issuing a check to United States Fire Insurance Co., which received it on behalf of Crum & Forster. On one occasion during the years at issue, Mr. Candland raised the deductible on RMS’s commercial workers’ compensation policies to “make it easier to justify our $1,200,000 captive contribution.”

F. Operations and Practices

The operations and practices of RMS, Risk Retention, Provincial, Artex, and petitioners provide additional context to the transactions among them beyond what is evident from the transaction structure or captive policies alone. We describe those operations and practices that are relevant in this subsection.

1. Transaction Documentation Practices

The Master Reinsurance Contracts or Agreements, under which Provincial ceded risks and premiums to the captive insurers participating in the Provincial Pool, were not executed by the captive insurers participating in the Provincial Pool. Instead, Artex employees executed these contracts. In some cases, Karl Huish, a co-founder of Tribeca who remained involved with the business after the sale of Tribeca’s assets to Artex’s parent company, executed both sides of the same contract, including during the years at issue. In addition, both before and during the years at issue, policy documents were sometimes irregularly dated, and policy or coverage periods often began retroactively relative to policy issuance. We describe some of these occurrences during the years at issue below.

28

[*28] a. 2012

RMS’s 2012 policy documents were not actually issued until May 24, 2012, 26 more than four months into the 2012 coverage period. 27 Furthermore, on January 28, 2013, Artex prepared a Change Endorsement for RMS’s 2012 Worker’s Compensation Deductible / SIR Reimbursement policy with an effective date of January 1, 2013.

The Change Endorsement stated that losses under the Worker’s Compensation Deductible / SIR Reimbursement policy would be paid either directly to RMS’s commercial insurer or to a collateral account held by the insurer unless otherwise directed by RMS. Nonetheless, Risk Retention had already begun paying RMS’s commercial insurer directly (rather than paying Provincial) in 2011. The Change Endorsement is thus anomalous not only in its effectiveness on January 1, 2013, a date both before its execution and coinciding with the end of the applicable coverage period, but also in its late documentation of a payment practice that had already begun much earlier. 28

b. 2013 and 2014

Beginning in 2013, Artex’s practice was to have Provincial issue policy documents only once a captive had paid at least 10% of its annual premiums. Artex finalized RMS’s 2013 and 2014 captive insurance policies only on June 19, 2013, and July 3, 2014, respectively, well into the applicable coverage periods. Even the essential terms of the policies were not always agreed upon before the beginning of each applicable coverage period. For example, Artex prepared a renewal policy summary for RMS’s 2013 captive insurance policies dated

26 On February 1, 2012, Artex provided RMS with a renewal policy summary,

describing the coverage period, coverage type, limits, SIR, premium, and policy number. Although an Artex underwriter testified that a policy summary is “like a binder,” we find that the renewal policy summary was on its face simply a summary of the intended policy issuance; that the record contains no credible contemporaneous evidence that it was intended to have any binding effect; and that the renewal of the policies was actually completed no earlier than May 24, 2012.

27 RMS’s renewal endorsements during the years at issue refer to a renewal

period rather than a coverage period. We refer to renewal periods as coverage periods throughout in order to avoid undue confusion.

28 An Artex employee raised the issue that RMS’s Worker’s Compensation

Deductible / SIR Reimbursement policy did not permit Risk Retention to pay RMS’s commercial insurer directly on January 17, 2013, 11 days before the Change Endorsement was executed on January 28, 2013.

29

[*29] January 23, 2013, and a revised renewal policy summary dated March 19, 2013, both after the 2013 coverage period was underway. Comparing the revised renewal policy summary to the original one, Artex (1) increased RMS’s 2013 gross premiums from $1,217,018 to $1,223,980, (2) added a General Liability Deductible / SIR Reimbursement policy, and (3) either increased or decreased the premium amounts applicable to each of RMS’s other 2013 policies.

2. Underwriting Process, Premium Determination, and Premium Payments

Mr. Candland provided Tribeca or Artex with the amount that he was willing to pay, and provided a target premium for all policies purchased by RMS, both before and during the years at issue. 29 Regarding premium payments, Artex required only that RMS (1) pay its pool premiums and 2.5% administrative fee by December 31 of the applicable policy year and (2) pay its facultative premiums by the end of the first quarter of the following year.

We now describe some additional practices of petitioners, RMS, Provincial, and Artex pertaining to the underwriting process, premium determination, and premium payments during the years at issue. We discuss these topics together because RMS generally decided how and when to pay its premiums, and Artex adapted its purported underwriting after the fact to accommodate its preferred payment amounts and schedule. We specifically find that petitioners, RMS, Provincial, or Artex (as applicable) engaged in the following practices related to underwriting, premium determination, and premium payments during the years at issue:

• The Provincial policies were not objectively rated by evaluating the risk and magnitude of loss on a prospective basis informed by detailed underwriting. The premiums that RMS paid for its captive coverages were inappropriately inflated by subjective, judgment-driven factors that made little sense under the circumstances here. The premiums were not supported by

29 Mr. Candland claimed the opposite in a sworn interview with respondent in

September 2015, stating that he told Tribeca he was interested in particular types of coverage rather than in paying a certain dollar amount of premiums. Mr. Candland changed his answers in this regard at trial.

30

[*30] actuarial analysis, 30 nor was Artex’s allocation of 49% of premiums to individual captives and 51% of premiums to the Provincial Pool.

• Insurance transactions, including premium pricing and premium payments, were completed after the fact even though in a typical insurance program they would be completed prospectively. Artex backdated policy changes and permitted the late issuance of insurance contracts and late premium payments.

• Artex did not obtain sufficient information from RMS to support the underwriting process.

• Artex placed undue weight in its purported underwriting on target premium figures provided by RMS without regard to whether the target premiums were supported by objective exposure information.

• Artex permitted its clients, including RMS, to alter their coverages or total premiums well into coverage periods in a manner that rendered clients’ decisions of whether to fund the policies and in what amount as essentially optional and retrospective, not binding and prospective.

• RMS and Mr. Candland sometimes requested premium increases to $1.2 million. 31

30 The primary actuarial report Provincial relied on for pricing was prepared

by James Rech (Mr. Rech), an actuary, in 2008. In it, Mr. Rech stated it was his opinion that “the rating methodology, pricing models, rating factors and rate parameters are reasonable.” Nonetheless, Mr. Rech did not opine on the ratings for any individual policies, and his report therefore does not constitute an actuarial endorsement of those premiums. Mr. Rech did not testify at trial, and Artex’s underwriters never documented how they derived rating factors.

Mr. Rech’s analysis attempted to provide support for a “Captive Risk Factor,”

which is not a typical rating factor used in the insurance industry. In his definition of this factor Mr. Rech implied that the adjustment was necessary because an additional premium is necessary for the first five or more years of a captive’s existence to be viable in the event of unusual losses. This is not a typical or industry standard adjustment made by actuaries and would not be a viable business methodology in the commercial market due to the competitive disadvantage created by excessive premiums.

31 Mr. Candland also requested a premium decrease on one occasion if Artex

did not permit RMS to pay a portion of its premiums after the coverage period. Artex ultimately relented and permitted the late payment, however.

31

[*31] • RMS paid a disproportionate share of its captive premiums during the years at issue toward the end of, or after, each coverage period, and Artex acquiesced in this practice.

• The premiums RMS paid for coverage from Provincial were not reasonable compared to typical industry pricing. 32 RMS’s total premiums were always remarkably close to the $1.2 million limit for nontaxable premium income under section 831(b), regardless of any variation in coverage.

• Artex generally relied on existing information in its purported underwriting of RMS’s policies instead of requesting up-to-date information. Artex’s relatively small underwriting staff was ill-prepared to underwrite the many different types of policies that Artex provided.

• Artex caused Provincial to issue RMS’s policy documents well into the coverage periods that the policies purported to cover without binders in place.

• Provincial wired facultative premiums to Risk Retention within about two weeks of receipt.

• Provincial often released pool premiums to Risk Retention within a few weeks after RMS paid them.

• RMS and Artex did not consistently recognize RMS’s premium payments for the insurance written by Provincial as constituting separate amounts from the amounts that Provincial ostensibly paid to Risk Retention for (1) providing reinsurance to unrelated members of the Provincial Pool pursuant to Master Reinsurance

32 For example, the average rate-on-line for RMS’s captive policies during the

years at issue was more than ten times greater than the average rate-on-line for comparable commercial insurance policies, even though RMS did not have major issues with its existing commercial insurance coverage, or in obtaining the insurance required by its client contracts. A higher rate-on-line means that insurance coverage is more expensive per dollar of coverage and could therefore lead to a greater deduction for premiums. See Syzygy Ins. Co. v. Commissioner, T.C. Memo. 2019-34, at *31.

The pricing for some individual policies did not make sense on its face. For example, the Employment Practices Deductible / SIR Reimbursement policy had premiums of $145,500 for 2011 and 2012 and $159,500 for 2013 but had a peroccurrence limit of $100,000. This cost does not make sense unless RMS anticipated multiple high-dollar claims per year (or a very large volume of small-dollar claims). In fact, however, RMS filed only one claim against the policy, for $3,452.

32

[*32] Contracts or Agreements or (2) providing reinsurance to Provincial pursuant to Facultative Reinsurance Contracts or Agreements. 33

3. Claims Handling

Artex’s director of underwriting, Deborah Inman, was involved with the claims process at Artex during the years at issue. Ms. Inman supervised the claims function at Artex until 2018. In March 2014 Artex hired a licensed claims adjuster. 34

The general terms and conditions for all Provincial policies during the years at issue provided that if an insured incurred a claim, it was required to give Artex prompt notice of the claim. The general terms and conditions further stated that an insured was required to give Artex a description of the events and circumstances that led to the claim as soon as possible.

33 For example, on November 4, 2013, Mr. Candland asked an Artex employee:

“If I wire a captive premium, how long before you can turn the funds around and deposit [them] in [Risk Retention’s] bank account?” The employee told Mr. Candland that direct premiums were “returned” three to five business days after payment and that pool premiums were “returned” within ten business days. On December 3, 2013, Mr. Candland told that Artex employee that he would be wiring $300,000 for a captive insurance premium and instructed him to deposit the funds “back into” Risk Retention’s bank account. The employee responded that he would “make sure” that those funds ended up “back at” Risk Retention’s bank. On January 7, 2014, Mr. Candland asked the same employee for help in getting other premium payments “moved through the system and back” because “[t]he previous premium took one week to turn around and I had anticipated the same for this last payment.” Provincial transferred the direct premiums to Risk Retention that day.

34 Before the hiring of a licensed claims adjuster, an underwriting assistant

and Artex’s risk pool administrator assisted Ms. Inman with handling claims.

33

[*33] a. RMS’s Claims

Risk Retention paid claims filed by RMS in the following amounts:

Year Amount

2008 —

2009 $2,450

2010 34,354

2011 323,379

2012 231,455

2013 400,868

2014 81,094

All paid claims filed by RMS under its 2012–14 captive policies were filed against its Worker’s Compensation Deductible / SIR Reimbursement policy. We make the following findings regarding the handling of RMS’s claims during the years at issue:

• RMS generally did not submit the deductible billing invoices or other claim documents it received from Crum & Forster to Artex before Risk Retention paid the deductibles billed, and it did not otherwise await approval from Artex. 35 Artex performed little

35 Petitioners’ expert witness Michael Angelina opined that “since the

[Worker’s Compensation Deductible / SIR Reimbursement] policy is a deductible reimbursement policy, there is no real need for Artex to ‘re-adjust’ a claim that has already been handled by the claims team of the commercial insurer (Crum & Forster). . . . While the approach to pay the claims in a ‘batch mode’ . . . was ‘not the norm’ for Artex, it was an approved process by Artex for these claims.” We reject as unsupported by the record any suggestion that Artex had no obligation to adjust claims for deductible or SIR amounts under RMS’s Worker’s Compensation Deductible / SIR Reimbursement policy simply because Crum & Forster had adjusted the underlying

34

[*34] timely review of these claims. It is not a typical practice in the insurance industry to approve a claim after it has already been paid.

• An objective coverage assessment could have resulted in a denial of most of RMS’s workers’ compensation deductible claims because the underlying losses had been previously reported before the inception of the applicable captive insurance policies.

• RMS sometimes notified Artex of claims after both the coverage period and the extended reporting period for a policy had lapsed. Risk Retention nonetheless issued payments for such claims.

• RMS used, and Artex acquiesced in the use of, board resolutions to authorize the payment of claims that should have been denied. RMS’s use of a board resolution to permit a settlement payment to Wausau, RMS’s former workers’ compensation carrier, under the Worker’s Compensation Deductible / SIR Reimbursement policy was intentionally misleading. 36

workers’ compensation claims under RMS’s separate commercial workers’ compensation policy.

Even if Risk Retention had been permitted to pay RMS’s workers’

compensation deductibles directly without any approval from Artex, this would be a major process deficiency because it allowed Risk Retention to pay claims that should not have been covered and to escape independent claims adjustment. There was a need for such claims adjustment to determine whether the deductibles charged by Crum & Forster were covered under the terms of each captive policy during the years at issue. For example, an objective coverage assessment could have resulted in a denial of most of RMS’s workers’ compensation deductible claims because the underlying losses had been previously reported before the inception of the applicable captive insurance policies.

36 On May 10, 2012, Mr. Candland notified Artex that RMS had settled a dispute regarding 2007 workers’ compensation claims with its former workers’ compensation carrier, Wausau. The Wausau claim arose from RMS’s nonpayment of disputed retrospective premium adjustments on its 2007–08 policy with Wausau; the retrospective adjustments were calculated on February 9, 2010, and January 31, 2011.

Ms. Inman accurately enumerated several issues with the claim on May 11, 2012. First, according to Ms. Inman, Mr. Candland “wants to make payment from the captive for claims that occurred in 2006 & 2007 which is before the captive was formed so the captive didn’t have any policies in force during that time.” Second, he “had knowledge of these claims when he started his captive and any . . . [workers’ compensation] policies that were written for him when the captive started. Claims of which the insured has prior knowledge are excluded.” Third, “[t]he claims would be

35

[*35] • Risk Retention, a purported reinsurer, inappropriately paid certain workers’ compensation deductible claims directly to the commercial carrier instead of paying Provincial. The January 28, 2013, Change Endorsement that Artex prepared for RMS’s 2012 Worker’s Compensation Deductible / SIR Reimbursement policy provided belated approval at best for this practice.

b. Provincial Pool Claims

The Provincial Pool paid a single claim in 2011 of $8,274, representing about 0.016% of the $51,702,549 in pool premiums for that year. The Provincial Pool paid $210,615 on account of three claims in 2012, which amounts to 0.324% of total pool premiums for 2012. In 2013 the Provincial Pool paid $2,631,536 on account of nine claims, which amounts to 3.322% of total pool premiums for 2013. The Provincial Pool had paid $2,507,682 in pool claims for the 2014 policy year as of February 17, 2021. This amounts to 3.019% of total pool premiums for 2014.

Risk Retention’s quota share of pool claims and loss adjustment expenses from 2012 to 2014, in dollars and as a percentage of pool premiums paid by RMS, was as follows:

Year Quota Share Quota Share as Approximate Percentage of Pool Premiums Paid by

RMS

2012 $1,921 0.325%

2013 20,209 3.333%

2014 18,732 3.152%

considered as late reported even if they were covered by the 2008 policy.” Fourth, “[b]ecause Risk Retention is now and was in 2011 a reinsurer of Provincial instead of a direct insurer, claims should be authorized by and paid through Provincial instead of directly from the captive.”

Nonetheless, Artex informed RMS that Risk Retention could pay Wausau if Risk Retention executed a board resolution authorizing the payment. On May 15, 2012, Risk Retention passed a board resolution authorizing Risk Retention to pay Wausau (and a related insurer, Liberty Mutual) $235,000. On the same day, Ms. Inman signed a proof of claim form approving the claim, and Risk Retention wired $235,000 to Wausau.

36

[*36] We make the following findings regarding Artex and Provincial’s handling of claims in the Provincial Pool:

• The low ratio of losses to premiums in the Provincial Pool compared to the insurance industry as a whole contributed to nearly a full round trip of pool premiums paid by RMS to Risk Retention, through various entities managed by Artex.

• Artex added or altered policies for its clients retroactively in order to permit them to file claims against the Provincial Pool or to reduce their premiums if they were unable to pay in full.

• Artex did not consistently enforce the prior-knowledge limitation 37 when adjusting claims, or treat claims as uncovered because no coverage was in effect at the time of a loss, even though it should have. It also did not consistently enforce the requirement that a claim be promptly submitted after an insured learned about it.

• There was inappropriate overlap between the claims and underwriting functions at Artex. On one occasion, Ms. Inman backdated a policy document to a date that preceded her employment at Artex in order to facilitate a client’s filing of claims under a retroactively added policy. 38 The claims were ultimately paid.

• Artex permitted its clients to use board resolutions to obtain claims payment for claims that should have been denied.

• Artex encouraged the submission of pool claims during the years at issue in order to improve the public perception of the legitimacy of the Provincial Pool, regardless of whether those claims should have been denied.

• Artex required only slight documentation in support of some pool claims.

37 The general terms and conditions to the captive policies excluded claims for

which an insured had knowledge of a covered cause of loss before the coverage period.

38 The client was Lanter Delivery Systems, Inc.

37

[*37] 4. Related-Party Loans and Payments

Before and during the years at issue, Risk Retention made loans to RMS to fund various business expenses of RMS and made other related-party payments. We describe those loans and payments here.

a. Premium Finance Agreements

Beginning in 2009 and continuing through the years at issue RMS executed eight premium finance agreements (PFAs) with Risk Retention in order to finance premiums on certain of RMS’s commercial insurance policies. 39 On August 19, 2009, Mr. Candland sent an email to an Artex employee explaining that “[i]n the past we have chosen to finance these premiums in an effort to cash flow the payment, rather than take an annual hit to our cash. Is there any legal reason why we can’t use some of our captive money and run the financing through our captive?” Mr. Candland also stated that “we would feel comfortable paying to Risk Retention” an interest rate for premium financing that “is much higher than we have paid in the past.” Mr. Candland prepared the PFAs on behalf of Risk Retention and signed each one on behalf of RMS.

Under the terms of the PFAs, Risk Retention paid commercial insurers directly for the full amount of certain of RMS’s annual commercial insurance policy premiums. RMS was then required to repay the total premiums plus an interest or finance charge to Risk Retention over the course of 12 months, except that the term for the fourth PFA, which was executed on January 24, 2011, was 16 months. All of the PFAs carried an annual interest rate of 10% with 12 or 16 equal monthly payments, as applicable. RMS sometimes notified Artex of the execution of PFAs only after the fact. 40

39 The parties stipulated that RMS and Risk Retention executed seven PFAs

during these years, but we note that eight PFAs for these years have been received into the record. We find that RMS and Risk Retention executed eight PFAs during these years. See Cal-Maine Foods, Inc. v. Commissioner, 93 T.C. 181, 195 (1989) (holding that we are not obliged to accept a stipulation between the parties when it is clearly contrary to facts disclosed by the record).

40 For example, Mr. Candland notified an Artex employee on June 15, 2011, of

the June 1, 2011, execution of the fifth PFA. Likewise, he informed Artex of the execution of the fourth PFA only on the same day that he executed it, which was January 24, 2011.

38

[*38] We do not discuss the first three PFAs further. RMS executed the fourth through eighth PFAs as follows:

PFA Execution Date Amount of Type of Premiums Premiums Financed Financed

Fourth January 24, 2011 $39,308 Cyber liability policy

Fifth June 1, 2011 144,642 Crime, general liability, professional liability, excess liability, and employment practices

liability policies

Sixth June 19, 2012 210,162 Crime, employment practices liability, commercial property,

general liability, professional liability, excess liability, and cyber liability policies

Seventh June 6, 2013 217,763 Crime, employment practices liability, property, general liability, professional liability, excess liability, and cyber liability

policies

Eighth June 14, 2014 224,033 Crime, employment practices liability, property, general liability, professional liability, excess liability, and cyber liability

policies

RMS timely repaid the principal and interest on each PFA in accordance with its terms.

b. Life Insurance Policy Payments

During the years at issue, RMS also purchased various life insurance policies for Mr. Keating, Mr. Candland, and Ms. Doss. Risk

39

[*39] Retention financed the premiums on these life insurance policies by paying the commercial life insurance carriers directly and in full. Artex and RMS considered this arrangement a loan. Risk Retention accounted for the payments to commercial insurance carriers for life insurance as notes receivable in its books and records. RMS repaid Risk Retention periodically for the life insurance premiums with interest. Nonetheless, petitioners have not produced a promissory note or any other writing evidencing a loan or another financing arrangement permitting Risk Retention to finance petitioners’ personal life insurance premiums nor one permitting RMS to repay Risk Retention with interest.

Artex characterized amounts paid in excess of principal repayment as captive insurance premiums rather than solely as interest. 41 Furthermore, in an email dated August 26, 2011, an Artex employee also described the “extra” money as being “circled back out pretty quickly” because it was used for RMS to pay further life insurance premiums. In 2012 Risk Retention financed life insurance premiums on behalf of Mr. Keating, Mr. Candland, and Ms. Doss of approximately $72,000.

Throughout most of 2012 Artex recorded the life insurance policies themselves as “Other Assets” in Risk Retention’s books and records. However, on December 1, 2012, Mr. Candland informed Artex that the policies were not owned by Risk Retention; instead they were used to fund a buy-sell agreement between Mr. Keating and himself, and each personally owned the policy taken out on the other. He further stated: “We have never intended for the life insurance policies to be owned by the captive, unless there is a significant tax advantage.”

41 In an email dated August 3, 2011, an Artex employee described the process

used by Risk Retention to finance the life insurance premiums by stating that RMS “make[s] monthly payments directly to the captive each month for loans that they took. When they make these payments, they pay ‘extra money’ directly into the captive that the captive then turns around and pays to buy life insurance policies. This extra money is considered premiums paid so at the end of the year when I calculate what they still owe(direct premiums, risk pool premiums, and 2.5% fee) everything worked out right to the very last zero.” (Emphasis added.)

40

[*40] c. Miscellaneous Loans

Risk Retention also made loans to RMS to finance software, hardware, and excess self-funded group health plan claims. We describe those loans here.

i. Software Note

On July 16, 2012, Risk Retention lent $126,000 to RMS to finance new computer software. The loan was evidenced by a promissory note although the promissory note stated that RMS’s obligation to make 24 monthly payments began on August 1, 2012, and ended on a maturity date of July 31, 2012, a patent error. The loan carried an annual interest rate of 10% and was secured through a security agreement. RMS made monthly payments of principal and interest to Risk Retention and repaid the loan in full on July 21, 2014.

ii. Hardware Note

On August 20, 2012, Risk Retention lent $71,000 to RMS for the purchase of computer hardware. The loan was evidenced by a promissory note and required 24 monthly payments. The loan carried an interest rate of 10% and was secured through a security agreement. RMS made monthly payments of principal and interest to Risk Retention and repaid the loan in full on August 21, 2014.

iii. Group Health Plan Notes

On November 1, 2013, Risk Retention lent $300,000 to RMS to finance the portion of RMS’s self-funded group health plan claims that exceeded premiums received in 2013. The November 1, 2013, loan was evidenced by a promissory note (First Stop Loss Bridge Note), and a security agreement was also executed on the same date. The loan carried an interest rate of 10%. Mr. Candland informed Artex of this loan through an email dated December 18, 2013. The First Stop Loss Bridge Note required repayment of all outstanding principal, interest, and other amounts on its maturity date, April 1, 2014, but RMS had not made any payments toward it as of that date.

Risk Retention made a second $300,000 loan to RMS in respect of its excess self-funded group health plan claims on January 8, 2014, which was also evidenced by a promissory note (Second Stop Loss Bridge Note). The loan carried an annual interest rate of 10%. Although the Second Stop Loss Bridge Note states that it was secured by a

41

[*41] contemporaneous security agreement, petitioners did not produce a copy of the security agreement. The Second Stop Loss Bridge Note required repayment of all outstanding principal, interest, and other amounts on its maturity date, May 1, 2014, but RMS had not made any payments toward it as of that date. Risk Retention did not take any action to enforce repayment of either the First or Second Stop Loss Bridge Note following default. On June 10 and September 29, 2014, RMS made payments to Risk Retention of $500,000 and $132,822, respectively, for its liabilities on both notes.

Risk Retention made a third $300,000 loan to RMS on December 10, 2014, which was also evidenced by a promissory note (VEBA Stop Loss Note). The loan carried an annual interest rate of 10%. Although the VEBA Stop Loss Note states that it was secured by a contemporaneous security agreement, petitioners did not produce a copy of the security agreement. The VEBA Stop Loss Note came due on March 1, 2015. The VEBA Stop Loss Note was repaid in full on March 2, 2015.

d. Deductible Agreements

RMS and United States Fire Insurance Co., the company through which RMS obtained workers’ compensation insurance from Crum & Forster, executed a deductible agreement in 2009. RMS and Risk Retention together executed various deductible agreements with United States Fire Insurance Co. beginning in July 2011 and continuing throughout the years at issue. Under its agreements with RMS and Risk Retention, United States Fire Insurance Co. was responsible for making initial payment of any deductibles owed under the applicable workers’ compensation policies, and RMS and Risk Retention were responsible for reimbursing it. The agreements required RMS and Risk Retention to ensure that a collateral fund contained a minimum amount of cash. Risk Retention paid substantial sums into the collateral fund before and during the years at issue.

5. Risk Retention

Mr. Keating and Mr. Candland owned Risk Retention equally during the years at issue, and Risk Retention’s board of directors consisted of Mr. Keating and Mr. Candland. Risk Retention held annual board meetings during the years at issue. More than 60% of Risk Retention’s assets were highly liquid assets during the years at issue.

42

[*42] Risk Retention’s books and records included a general ledger, a balance sheet, a profit and loss statement, and an adjusted trial balance.

6. Capitalization of Provincial and Provincial Pool

We first discuss the capitalization of Provincial generally because Provincial was a fronting insurer under both the facultative and quota-share reinsurance agreements. We then specifically discuss the capitalization of the Provincial Pool, which is relevant only to the quota-share reinsurance portion of the captive arrangement.

a. Provincial

Although Provincial reported substantial cash on hand throughout the years at issue, Provincial held minimal capital by other measures. Provincial’s reported current liabilities either exceeded, or were only marginally exceeded by, its current assets on each of its balance sheets for the years at issue. The disparity was most marked at yearend 2014, when Provincial had only $21,861,284 in current assets, compared to $34,982,548 in current liabilities. 42

b. Provincial Pool

Beginning on June 1, 2013, Artex withheld 2% of the pool premiums paid by each captive insurer as a risk pool claim reserve. 43 Artex was unable to pay pool claims quickly before that date because it generally collected funds from each pool member as each claim arose. 44

42 Provincial also had little equity to draw upon to meet its liabilities.

Provincial reported $1,193,735 in equity at yearend 2012 compared to $89,037,865 in current liabilities; $1,181,908 in equity at yearend 2013 compared to $132,933,824 in current liabilities; and $368,427 in equity at yearend 2014 compared to $34,982,548 in current liabilities.

43 Artex held the claim reserves in a non-interest-bearing “Reserve Account”

that was recorded as an asset on each captive insurer’s balance sheet and other financial documents.

44 Jeremy Huish at Artex described the change in an email on May 1, 2013,

stating: “[W]e are starting a reserve account to pay pool claims in the future. Under our current system, we can’t pay out a pool claim until the middle of the next year because of the time it takes to gather funds from everyone. While a slow insurance payment has been fine in the fast [sic], there may be claims in the future where a quick payment is needed.”

43

[*43] In 2013 and 2014 45 the Master Reinsurance Agreements also required all captive insurers participating in the Provincial Pool to provide funds as collateral to support potential pool claims, either by allowing Provincial to hold the funds or by holding the funds in a collateral account. Risk Retention had provided pledged accounts to the Provincial Pool pursuant to a separate pledge agreement before this time although the record is not clear with respect to participating reinsurers other than Risk Retention.

IV. Dividends

In 2012 Risk Retention paid Mr. Candland and Mr. Keating dividends of $500,000 each. In April 2014 Risk Retention paid Mr. Candland and Mr. Keating dividends of $200,000 each. In October 2014 Risk Retention paid Mr. Candland and Mr. Keating additional dividends of $300,000 each. In both 2012 and 2014 Risk Retention issued copies of Form 1099–DIV, Dividends and Distributions, reporting the dividends paid to Mr. Candland and Mr. Keating.

V. Sale of RMS

In 2015 petitioners sold a controlling interest in RMS to a third party. Before that sale, in 2012, Mr. Candland discussed a sale of RMS with a potential buyer. Mr. Candland emailed a calculation of EBITDA (i.e., earnings before interest, taxes, depreciation, and amortization) to the potential buyer. The EBITDA calculation provided by Mr. Candland added back into earnings the amounts paid to Risk Retention as insurance premiums (less claims), as well as other amounts not typically understood as interest, taxes, depreciation, or amortization, such as $200,000 in “[p]erks” for petitioners.

VI. Tax Reporting

A. RMS

RMS was an accrual basis taxpayer during the years at issue.

RMS timely filed its Form 1120S, U.S. Income Tax Return for an S Corporation, for each year at issue.

45 The 2012 Master Reinsurance Contract did not address the subject of collateral.

44

[*44] 1. 2012

RMS reported gross receipts of $65,750,508 and net ordinary business income of $59,397 on its Form 1120S for its 2012 taxable year. RMS deducted $1,229,089 that is in dispute here, comprising $1,160,921 in captive insurance premiums, $38,400 in fees paid to Artex, 46 and $29,768 for the 2.5% administrative fee.

2. 2013

RMS reported gross receipts of $78,078,987 and net ordinary business income of $270,269 on its Form 1120S for its 2013 taxable year. RMS deducted $1,262,380 that is in dispute here, comprising $1,193,380 in captive insurance premiums, $38,400 in fees paid to Artex, and $30,600 for the 2.5% administrative fee.

3. 2014

RMS reported gross receipts of $84,464,179 and a net ordinary business loss of $226,863 on its Form 1120S for its 2014 taxable year. RMS deducted $1,229,111 that is in dispute here, comprising $1,160,943 in captive insurance premiums, $38,400 in fees paid to Artex, and $29,768 for the 2.5% administrative fee.

B. Risk Retention

Risk Retention filed an election as a foreign insurance company to be treated as a domestic corporation under section 953(d) on February 18, 2009, which the IRS accepted. Risk Retention filed a Form 1120–PC, U.S. Property and Casualty Insurance Company Income Tax Return, for each year at issue. Risk Retention attached its Foreign Insurance Company Election under section 953(d) and a Small

46 Regarding the fees paid to Artex, Artex would periodically issue invoices for

its captive management services to RMS. Artex typically invoiced RMS $3,200 monthly for these services during the years at issue. An October 6, 2008, engagement letter between Tribeca and RMS states that the management fee paid for the following fees and services: insurance management fees; costs for annual reviews regarding policies and premiums; reviewing and determining insurable risks; underwriting and drafting policies; consultation regarding qualification of the captive insurance company; preparations of financial statements; auditing fees; consultation regarding business operations; preparation and filing of tax returns; annual insurance license fees; and annual corporate fees for the captive.

45

[*45] Insurance Company Election under section 831(b) to its income tax return for each of the years at issue. 47

C. Petitioners

Petitioners timely filed Forms 1040, U.S. Individual Income Tax Return, for each year at issue. Mr. Candland and Mr. Keating reported the dividends they received from Risk Retention in both 2012 and 2014 as qualified dividends on their respective individual income tax returns for each year and paid tax on the dividends at the qualified dividend rate. See § 1(h)(11).

OPINION

I. Evidentiary Matters

As a preliminary matter, the Court must address the admissibility of certain documentary or other nontestimonial evidence introduced at trial but for which we reserved ruling. Our evidentiary rulings are determined under the Federal Rules of Evidence. See § 7453; Rule 143(a).

Under the Federal Rules of Evidence, irrelevant evidence is not admissible. See Fed. R. Evid. 402. An item of evidence is relevant to the extent it tends to make a fact more or less probable and the fact is consequential to determining the action. See Fed. R. Evid. 401. When the relevance of evidence depends on a fact, proof must be introduced sufficient to support a finding that the fact does exist. See Fed. R. Evid. 104(b); see also David S. Schwartz, A Foundation Theory of Evidence, 100 Geo. L.J. 95, 140 (2011) (“[C]onditional relevance is a requirement that foundations be complete rather than relying on generalizations to do the work of case-specific, evidenced facts.”).

Most of the outstanding evidentiary determinations involve instances where (1) we advised the offering party that the proposed item of evidence required a foundation to be established at trial or (2) the offering party advised us that the proposed item of evidence could be introduced during the course of trial or explored further in conjunction with witness testimony. Cf. Jerden v. Amstutz, 430 F.3d 1231, 1237 (9th Cir. 2005) (stating that a trial court “may not exclude evidence before trial [on the ground of lack of foundation] without allowing the parties

47 Risk Retention did not request that the Secretary of the Treasury revoke

either of these elections for its 2008–14 taxable years.

46

[*46] to lay a foundation for its admission”). The offering party never examined any witness about, or else failed to establish an adequate foundation for, the following Exhibits: 17-R, 18-R, 19-R, 21-R, 22-R, 23-R, 29-R, 30-R, 31-R, 32-R, 33-R, 34-R, 35-R, 57-P, 58-P, 60-R, 61-R, 100-R, 523-R, 524-R, 525-R, 1503-R, 1714-R, 1715-R, 1716-R, and 1717-R. The relevance of these Exhibits is entirely speculative without an adequate foundation established through witness testimony or other means at trial. We therefore exclude them from evidence.

Other evidentiary determinations involve instances where we excluded an Exhibit at trial but did not expressly rule on the admissibility of a related Exhibit. Our review of the following Exhibits shows that they lack an adequate foundation, and their relevance is entirely speculative, in view of our exclusion of related Exhibits: 1-R, 526-R, and 528-R. We therefore exclude them from evidence.

Finally, Exhibit 50-P is a spreadsheet in Excel format that shows statistics regarding certain captives as of March 3, 2021. Exhibit 49-P is a copy of the same spreadsheet in another format that we excluded at trial because petitioners did not adequately establish that it was an accurate summary of voluminous records pursuant to Rule 1006 of the Federal Rules of Evidence. Cf. United States v. Lynch, 735 F. App’x 780, 785 (3d Cir. 2018) (“Rule 1006 summaries . . . must be supported by a foundation showing that the exhibit is an accurate summary of the underlying materials . . . .”); United States v. Scales, 594 F.2d 558, 563 (6th Cir. 1979) (“[E]ven under Rule 1006, the summary or chart must be accurate, authentic and properly introduced before it may be admitted in evidence.”). Exhibit 50-P is the same document, albeit in a different format, and we exclude it for the same reason.

II. Jurisdiction and Burden of Proof

Where a notice of deficiency issued to an S corporation shareholder includes adjustments to both S corporation items and other items unrelated to the S corporation, 48 we have jurisdiction to determine the correctness of all adjustments in the shareholder-level deficiency

48 An S corporation is governed under the rules in subchapter S of chapter 1 of

subtitle A of the Code. S corporations are not generally themselves subject to federal income tax but, like partnerships, are conduits through which income flows to their shareholders. See § 1366; Gitlitz v. Commissioner, 531 U.S. 206, 209 (2001) (“Subchapter S allows shareholders of qualified corporations to elect a ‘pass-through’ taxation system under which income is subjected to only one level of taxation.”).

47

[*47] proceeding. 49 See Johnson v. Commissioner, 160 T.C. 18, 28 (2023) (citing Winter v. Commissioner, 135 T.C. 238, 245–46 (2010)). We thus have jurisdiction to redetermine the correctness of respondent’s adjustments to petitioners’ flowthrough share of RMS’s income and any other determinations in the notice of deficiency.

The Commissioner’s determinations in a notice of deficiency are generally presumed correct, and the taxpayer bears the burden of proving that the determinations are incorrect. See Rule 142(a)(1); see also Welch v. Helvering, 290 U.S. 111, 115 (1933); Rockwell v. Commissioner, 512 F.2d 882, 885–87 (9th Cir. 1975), aff’g T.C. Memo. 1972-133. However, if the Commissioner raises a new matter, seeks an increase in deficiency, or asserts an affirmative defense, the Commissioner has the burden of proof as to the new matter, increased deficiency, or affirmative defense. Rule 142(a)(1).

Gross income generally includes all income from whatever source derived, including dividends. See § 61(a); Commissioner v. Glenshaw Glass Co., 348 U.S. 426, 429–30 (1955); Wilcox v. Commissioner, 848 F.2d 1007, 1008 (9th Cir. 1988), aff’g T.C. Memo. 1987-225; Treas. Reg. §§ 1.61-1(a), 1.61-9(a). Deductions are a matter of legislative grace, and taxpayers bear the burden of proving that they are entitled to any deduction claimed. See INDOPCO, Inc. v. Commissioner, 503 U.S. 79, 84 (1992); New Colonial Ice Co. v. Helvering, 292 U.S. 435, 440 (1934). A taxpayer claiming a deduction on a federal income tax return must demonstrate that the deduction is provided for by statute and must maintain records sufficient to enable the Commissioner to determine the correct tax liability. See § 6001; Hradesky v. Commissioner, 65 T.C. 87, 89–90 (1975), aff’d per curiam, 540 F.2d 821 (5th Cir. 1976); Treas. Reg. § 1.6001-1(a).

Under section 7491(a), if the taxpayer provides credible evidence concerning any factual issue relevant to ascertaining the taxpayer’s liability and complies with certain other requirements, the burden of proof shifts to the Commissioner as to the factual issue. Petitioners do not contend that the burden of proof shifts to respondent under

49 RMS is not a party to these cases. The unified subchapter S corporation

audit and litigation procedures formerly set forth in subchapter D of chapter 63 of subtitle F of the Code were repealed for taxable years beginning after December 31, 1996. See Allen Family Foods, Inc. v. Commissioner, T.C. Memo. 2000-327, slip op. at 5 & n.3. Neither RMS nor respondent has revoked or terminated RMS’s S corporation election.

48

[*48] section 7491(a) as to an issue of fact. 50 Therefore, petitioners bear the burden of proof on all issues. 51 We discuss the burden of proof applicable to the accuracy-related penalties that respondent has determined against petitioners separately in connection with our discussion of those penalties.

III. Credibility and Fact-Finding

“The most important and most crucial action the courts take in [a trial] is to resolve facts.” United States v. Gainey, 380 U.S. 63, 88 (1965) (Black, J., dissenting); see Diaz v. Commissioner, 58 T.C. 560, 564 (1972) (“[T]he distillation of truth from falsehood . . . is the daily grist of judicial life.”). The fact-finding process often requires the Court as the finder of fact to evaluate the credibility of witness testimony before making findings on the basis of that testimony. This Court has stated that in determining credibility,

[w]e observe the candor, sincerity, and demeanor of each witness in order to evaluate his or her testimony and assign it weight for the primary purpose of finding disputed facts. We determine the credibility of each witness, weigh each piece of evidence, draw appropriate inferences, and choose between conflicting inferences in finding the facts of a case. The mere fact that one party presents unopposed testimony on his or her behalf does not necessarily mean that the elicited testimony will result in a finding of fact in that party’s favor. We will not accept the testimony of witnesses at face value if we find that the outward

50 Petitioners asserted in their respective Petitions that “the Commissioner has

the burden of proof with respect to all issues raised in his Notice of Deficiency,” but they have not raised this issue on brief. We therefore deem any argument by petitioners that section 7491(a) is applicable to have been waived or conceded. See Estate of Atkinson v. Commissioner, 115 T.C. 26, 35 (2000) (deeming issue not addressed in posttrial brief to be waived or conceded), aff’d, 309 F.3d 1290 (11th Cir. 2002); Estate of Blount v. Commissioner, T.C. Memo. 2004-116, slip op. at 55 n.29 (deeming burden of proof shift under section 7491 waived when the taxpayer failed to raise it), aff’d in part, rev’d and remanded in part, 428 F.3d 1338 (11th Cir. 2005); see also Mendes v. Commissioner, 121 T.C. 308, 312–13 (2003) (“If an argument is not pursued on brief, we may conclude that it has been abandoned.”).

51 This statement does not apply to respondent’s invocation of the duty of

consistency, which is an affirmative defense as to which respondent bears the burden of proof. See Estate of Ashman v. Commissioner, T.C. Memo. 1998-145, slip op. at 5, aff’d, 231 F.3d 541 (9th Cir. 2000). Nonetheless, as explained below, we find it unnecessary to reach respondent’s duty of consistency argument.

49

[*49] appearance of the facts in their totality conveys an impression contrary to the spoken word.

Neonatology Assocs., P.A. v. Commissioner, 115 T.C. 43, 84 (2000), aff’d, 299 F.3d 221 (3d Cir. 2002). As the trier of fact we may credit evidence in full, in part, or not at all. We may credit the part of a witness’s testimony that is not self-serving, while requiring some form of corroboration before crediting the portion that is. See Factor v. Commissioner, 281 F.2d 100, 114 n.27 (9th Cir. 1960) (“The Tax Court may accept parts and reject other parts of a witness’s testimony.”), aff’g T.C. Memo. 1958-94; Baumgardner v. Commissioner, 251 F.2d 311, 321 (9th Cir. 1957) (“The Tax Court was willing to accept in part the taxpayer’s claim of alleged profits from buying and selling improvement bonds. It was not required to accept it in full.”), aff’g T.C. Memo. 1956-112.

It is “the exclusive province of the fact finder to determine the credibility of witnesses, resolve evidentiary conflicts, and draw reasonable inferences from proven facts.” United States v. Hubbard, 96 F.3d 1223, 1226 (9th Cir. 1996); see Anderson v. City of Bessemer City, N.C., 470 U.S. 564, 573–74 (1985) (stating that if the trial court’s view of the evidence is plausible in light of the record, a reviewing court may not disturb it absent clear error, even when the trial court’s findings “do not rest on credibility determinations, but are based instead on physical or documentary evidence or inferences from other facts”); United States v. Yellow Cab Co., 338 U.S. 338, 342 (1949) (stating that where there are two permissible views of the evidence, the factfinder’s choice between them is not clearly erroneous); United States v. U.S. Gypsum Co., 333 U.S. 364, 395 (1948) (stating that a finding is clearly erroneous when “the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed”); Estate of Rau v. Commissioner, 301 F.2d 51, 54 (9th Cir. 1962) (“The Tax Court personally observed the witnesses . . . and from that vantage point was in a position to evaluate their testimony in the light of their attitude and demeanor while being interrogated.”), aff’g T.C. Memo. 1959-117. As the Ninth Circuit has stated in relation to the allowability of deductions in particular, “[t]he question of whether a taxpayer is allowed a deduction for particular expenses is a question of fact to be established by the taxpayer’s evidence, the credibility of the taxpayer, and the credibility of supporting witnesses. . . . [T]he Tax Court determines the credibility of the proffered testimony.” Schachter v. Commissioner, 255 F.3d 1031, 1034 (9th Cir. 2001), aff’g T.C. Memo. 1998-260, supplemented by 113 T.C. 192 (1999); see Norgaard v. Commissioner,

50

[*50] 939 F.2d 874, 878 (9th Cir. 1991), aff’g in part, rev’g in part T.C. Memo. 1989-390; see also McKay v. Commissioner, 886 F.2d 1237, 1238 (9th Cir. 1989), aff’g 89 T.C. 1063 (1987). We determine the credibility of witnesses, resolve evidentiary conflicts, and draw inferences from the voluminous record developed by the parties with this framework in mind.

IV. Microcaptive Arrangement

We begin by briefly explaining the taxation of microcaptive insurance companies and the deductibility of payments to them. We have recently considered other purported microcaptive insurance arrangements. See Avrahami v. Commissioner, 149 T.C. 144 (2017); Caylor Land & Dev., Inc. v. Commissioner, T.C. Memo. 2021-30; Syzygy Ins. Co., T.C. Memo. 2019-34; Reserve Mech. Corp. v. Commissioner, T.C. Memo. 2018-86, aff’d, 34 F.4th 881 (10th Cir. 2022); cf. Patel v. Commissioner, T.C. Memo. 2020-133 (deciding the issue of timely supervisory approval for penalties pursuant to section 6751(b) in the context of a purported microcaptive insurance arrangement).

Insurance companies (other than life insurance companies) are generally taxed on their income in the same manner as other corporations. See §§ 11, 831(a). However, section 831(b) provides an alternative taxing structure for certain small insurance companies. During the years at issue, an insurance company with net written premiums (or, if greater, direct written premiums) that did not exceed $1.2 million for the year could elect to be taxed under section 831(b). 52 § 831(b)(2). A small insurance company that makes a valid section 831(b) election is subject to tax only on its investment income, § 831(b)(1), and is not subject to tax on its earned premiums, see id. When a captive insurance company 53 makes a section 831(b) election, it is commonly referred to as a microcaptive insurance company.

Typically, amounts paid for insurance are deductible under section 162(a) as ordinary and necessary expenses paid or incurred in

52 Amendments to section 831(b) in 2015 increased the premium ceiling to $2.2 million (adjusted for inflation) and added new diversification requirements that an insurance company must meet to be eligible to make a section 831(b) election. See Syzygy Ins. Co., T.C. Memo. 2019-34, at *27 n.25.

53 A captive insurance company is typically a corporation whose stock is owned

by one or a small number of shareholders and which handles all or a part of the insurance needs of its shareholders or affiliates. See Harper Grp. v. Commissioner, 96 T.C. 45, 46 n.3 (1991), aff’d, 979 F.2d 1341 (9th Cir. 1992).

51

[*51] connection with a trade or business. See Treas. Reg. § 1.162-1(a). Section 162(a) does not prohibit deductions for microcaptive insurance premiums. When such a deduction is available, an insured may be able to deduct a premium payment to its affiliated microcaptive insurance company without a corresponding inclusion of the premium in income by the microcaptive insurance company. See Syzygy Ins. Co., T.C. Memo. 2019-34, at *28.

Nonetheless, the deductibility of insurance premiums depends on whether they were truly payments for insurance. See Avrahami, 149 T.C. at 174, 199; Syzygy Ins. Co., T.C. Memo. 2019-34, at *28; see also Clougherty Packing Co. v. Commissioner, 811 F.2d 1297, 1300 (9th Cir. 1987) (“In lieu of purchasing insurance, one may elect to self-insure, paying off claims as they arise or setting aside fixed sums into a reserve account to pay off intermittent losses. While insurance premiums are deductible, amounts placed into self-insurance reserves are not. . . . The appropriate starting point of our analysis is the meaning of ‘insurance.’”), aff’g 84 T.C. 948 (1985); Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *31. In addition, as explained below, the characterization of the dividends paid by Risk Retention to Mr. Candland and Mr. Keating as ordinary or qualified dividends depends on whether Risk Retention transacted in insurance. Thus, these cases hinge on whether the captive insurance arrangement meets the definition of insurance. 54

A. Whether the Arrangement Is Insurance

Neither the Code nor the Treasury Regulations define insurance, and we are guided by caselaw in determining whether a transaction constitutes insurance. See Avrahami, 149 T.C. at 174; Syzygy Ins. Co., T.C. Memo. 2019-34, at *28–29. Courts have looked to four criteria in deciding whether an arrangement constitutes insurance: (1) the arrangement involves an insurance risk; (2) the arrangement shifts the

54 In his Simultaneous Opening Brief respondent argues that we should apply

various substance-over-form doctrines in order to disregard the transactions at issue. However, because we consider the transactions at issue in accordance with their actual form and particular facts (i.e., without resort to recharacterizing their form) and conclude that they do not constitute insurance for federal income tax purposes, we need not decide whether any substance-over-form doctrine would apply in these cases. See Avrahami, 149 T.C. at 197 (“In light of our holding [that the transactions at issue are not insurance for federal tax purposes] we need not address the Commissioner’s other arguments—i.e., that the amounts deducted as insurance expenses should be disallowed under the economic-substance, substance-over-form, and step-transaction doctrines.”).

52

[*52] risk of loss to the insurer; (3) the insurer distributes its risk among its policyholders; and (4) the arrangement is insurance in the commonly accepted sense. See Avrahami, 149 T.C. at 177; Rent-A-Center, Inc. v. Commissioner, 142 T.C. 1, 13 (2014); Black Hills Corp. v. Commissioner, 101 T.C. 173, 182 (1993), supplemented by 102 T.C. 505 (1994), aff’d, 73 F.3d 799 (8th Cir. 1996); Harper Grp., 96 T.C. at 58; AMERCO & Subs. v. Commissioner, 96 T.C. 18, 38 (1991), aff’d, 979 F.2d 162 (9th Cir. 1992); Syzygy Ins. Co., T.C. Memo. 2019-34, at *29. Each part of the test must be satisfied. Harper Grp., 96 T.C. at 58. These four criteria are “nonexclusive,” Avrahami, 149 T.C. at 177, although we have noted they are “rarely supplemented,” Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *32. Respondent concedes on brief that we “can assume without deciding that the transaction involved risk shifting,” so we assume that point. Respondent has also made a similar concession regarding insurance risk, so we assume that the arrangement involved insurance risks.

We find for the reasons stated below that petitioners have not met their burden of proof to show that the microcaptive arrangement is insurance in the commonly accepted sense, and we therefore determine that it is not insurance for federal income tax purposes. See Avrahami, 149 T.C. at 190–91 (“[T]he cases tell us that in deciding whether an arrangement is insurance we can also look at whether it looks like insurance in the commonly accepted sense. This is an alternative ground.”); Rent-A-Center, Inc., 142 T.C. at 13 (“[T]he arrangement must . . . meet commonly accepted notions of insurance.”); Harper Grp., 96 T.C. at 58 (stating that “each part” of our test for “determining the propriety of claimed insurance deductions by a parent or affiliated company to a captive insurance company . . . must be satisfied,” including “whether the arrangement was for ‘insurance’ in its commonly accepted sense”); Syzygy Ins. Co., T.C. Memo. 2019-34, at *37 & n.26; see also Reserve Mech. Corp. v. Commissioner, 34 F.4th at 913–16 (upholding finding that microcaptive insurance policies “did not satisfy the requirement that they be insurance in the commonly accepted sense”); AMERCO v. Commissioner, 979 F.2d at 165; Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *39. It is unnecessary for us to address whether the arrangement involves risk distribution. After outlining the reasons for our conclusion that the microcaptive arrangement is not insurance in the commonly accepted sense, and therefore does not constitute insurance for federal income tax purposes, we discuss the legal effect of that conclusion in the next subsection.

53

[*53] 1. Commonly Accepted Notions of Insurance

To determine whether an arrangement constitutes insurance in the commonly accepted sense, we look at numerous factors including: (1) whether the insuring company was organized, operated, and regulated as an insurance company; (2) whether it was adequately capitalized; (3) whether the policies were valid and binding; (4) whether premiums were reasonable and the result of arm’s-length transactions; and (5) whether claims were paid. See Avrahami, 149 T.C. at 191 (first citing R.V.I. Guar. Co. v. Commissioner, 145 T.C. 209, 231 (2015); then citing Rent-A-Center, Inc., 142 T.C. at 24–25; then citing Harper Grp., 96 T.C. at 60; and then citing Securitas Holdings, Inc. & Subs. v. Commissioner, T.C. Memo. 2014-225, at *27); see also Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *39–40; Syzygy Ins. Co., T.C. Memo. 2019-34, at *37–38; Reserve Mech. Corp., T.C. Memo. 2018-86, at *48; cf. Syzygy Ins. Co., T.C. Memo. 2019-34, at *41 (noting that “whether the fronting carriers operated in a bona fide fashion” is also relevant). We will address each of these factors in turn.

a. Organization, Operation, and Regulation

Risk Retention and Provincial were organized as insurance companies in Anguilla, and they were regulated by the Anguilla Financial Services Commission. Generally, they complied with the requirements of Anguillan law. They obtained insurance licenses, satisfied Anguilla’s low capitalization requirements, and filed required documents with regulators. The record also shows that Risk Retention held annual board meetings, kept organizational books and records, and maintained separate bank accounts. Apart from generally observing the requisite formalities, however, the facts demonstrate that Risk Retention and Provincial were not operated as insurance companies. Cf. Reserve Mech. Corp., T.C. Memo. 2018-86, at *50.

Under the management of Artex, and with some significant input by Mr. Candland, Risk Retention and Provincial operated during the years at issue in a manner in which only unthinking insurance companies would operate. Insurance transactions, including premium pricing, premium payments, and claims approval, were completed after the fact, even though in a typical insurance program they would be completed prospectively. Cf. Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *41–42. Underwriting for policies in the Risk Retention captive program often occurred well into the coverage period or after the coverage period had expired. In any case, underwriting was based on

54

[*54] woefully inadequate information and methods, and it was disproportionately influenced by meeting target premiums near the $1.2 million section 831(b) limit, regardless of the coverage being provided. Artex and Provincial backdated documents, approved of retroactive policy changes, and permitted the late issuance of insurance contracts 55 and even later premium payments. RMS paid a disproportionate share of its captive premiums during the years at issue toward the end of, or after, each coverage period. RMS never paid premiums on a regular schedule of any kind, as opposed to making payments whenever it decided to do so. Artex’s own invoice to one of its other clients that is in the record makes the point best: “One typical attribute of an insurance transaction[] is that premium is paid up front, monthly, or quarterly. It is not commonly paid in one lump sum at the end of the policy term.”

Mr. Candland, Mr. Keating, and RMS also treated Risk Retention as if it were a tax-free savings account rather than a bona fide insurance company with which they were dealing at arm’s length. Risk Retention posted collateral with United States Fire Insurance Co. to fund deductibles under RMS’s commercial workers’ compensation policy without any clear obligation for it to do so, other than its self-imposed one under the deductible agreements. Risk Retention never documented its purported loan to finance Mr. Keating’s and Mr. Candland’s buy-sell life insurance policy premiums on each other, and it is unclear whether these loans were enforceable or secured. Artex also characterized the purported loan repayments from RMS to Risk Retention in excess of principal repayment not solely as interest but also as (1) premiums paid and (2) as “extra” money that is “circled back out pretty quickly” to pay further life insurance premiums. 56

While RMS and Risk Retention documented miscellaneous loans for hardware, software, and excess group health plan claims with promissory notes and repaid some of them in accordance with their terms, other aspects of these loans are concerning. RMS failed to repay

55 Despite petitioners’ assertion to the contrary, we do not see any credible

evidence in the record that binders were in place during the coverage period until final insurance policies were issued.

56 These characterizations are consistent with petitioners’ use of Risk

Retention as a de facto tax-free savings account because a bona fide insurance company would require interest on a loan to compensate it for its impairment to its capital base, its ability to pay claims, and its ability to generate investment income. It would not likely treat interest as “extra” money that it could “circle[] back out pretty quickly” to fund further loans or related-party expenses unless it expected few claims.

55

[*55] the First and Second Stop Loss Bridge Notes timely and indeed had not made any payments by their maturity dates. Risk Retention did not take any action to enforce either note following default. The Second Stop Loss Bridge Note and the VEBA Stop Loss Note were not accompanied by any security agreement that is in the record. The July 16, 2012, promissory note documenting the software loan contained a patent error on its face that calls its enforceability into question.

While RMS repaid the PFAs timely in accordance with their terms, Mr. Candland notified Artex of the PFAs only after the fact, instead of obtaining advance approval from Artex for the related-party dealings he was organizing. His commitment to paying a “much higher” interest rate to Risk Retention than RMS had paid in the past to finance its commercial insurance policy premiums is not supported by any legitimate business purpose discernible from the record. In any case, it casts doubt on the reasonableness of the interest rate charged in the PFAs.

We also have significant concerns about the reinsurance aspects of the microcaptive arrangement. We agree with respondent’s expert James MacDonald that Artex’s failure to disclose that it reserved the right to cede 100% of the premium for some coverages to either the captive or the Provincial Pool was a significant departure from applicable reinsurance customs and practices. Neither is there any credible evidence in the record that Risk Retention performed adequate due diligence on the quota-share risks that it assumed through the Provincial Pool. Overall, in numerous facets of their operations, Risk Retention and Provincial did not operate as bona fide insurers or reinsurers would.

We also accord some weight to the nontax characterizations of the microcaptive arrangement by the parties to it. See Sears, Roebuck & Co. v. Commissioner, 96 T.C. 61, 101–02 (1991) (considering whether “the arrangements . . . are characterized as insurance for essentially all nontax purposes”), supplemented by 96 T.C. 671 (1991), aff’d in part, rev’d and remanded in part, 972 F.2d 858 (7th Cir. 1992). The parties to the arrangement did not characterize it as insurance for essentially all nontax purposes. RMS and Artex did not consistently recognize RMS’s premium payments as separate from the amounts that Risk Retention ostensibly received for providing facultative or quota-share

56

[*56] reinsurance to Provincial. 57 This characterization is consistent with a near-circular flow of funds or, when considering it together with the various loans, dividends, and other disbursements that Risk Retention made, a circular one. 58 Mr. Candland also described the arrangement to an external auditor as a means by which RMS self- insured workers’ compensation claims. Furthermore, Mr. Candland emailed an EBITDA calculation to a potential buyer of RMS that added back into earnings the amounts paid to Risk Retention as insurance premiums (less claims), as well as other amounts not typically understood as interest, taxes, depreciation, or amortization, such as $200,000 in “[p]erks” for petitioners. We take this to mean that a potential buyer of RMS did not need to subtract the captive insurance expenses from this metric of RMS’s profitability because they did not detract from RMS’s profitability in an economic sense. Overall, the characterizations of the arrangement by the parties to it reflect their understanding that Artex’s approach gave RMS the benefit of an upfront, tax-deductible premium charge without a loss of control over its disposition of the funds that had proverbially been moved from one pocket to another (i.e., to Risk Retention).

b. Capitalization

We have consistently held that an insurer is adequately capitalized if it meets the relevant jurisdiction’s minimum capitalization requirements. See Avrahami, 149 T.C. at 193; Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *43; Syzygy Ins. Co., T.C. Memo. 2019-34, at *41; Reserve Mech. Corp., T.C. Memo. 2018-86, at *53. Risk Retention and Provincial met Anguilla’s minimum capitalization requirements during the years at issue. We do not upset the consensus here.

57 For example, during the years at issue, Mr. Candland asked for Artex’s help

in getting a premium deposited “back into” Risk Retention’s bank account and “moved through the system and back,” and he also asked how long it would take to “turn the funds around and deposit [them] in [Risk Retention’s] bank account.”

58 This is further supported by the facts that (1) Risk Retention’s quota share

of pool premiums was equal to the net premiums Provincial received from RMS for excess coverage from at least 2009 to 2014 and (2) Risk Retention’s quota share of pool claims and loss adjustment expenses was relatively low as a percentage of the pool premiums paid by RMS for each year at issue (0.325% for 2012, 3.333% for 2013, and 3.152% for 2014). We also think that intent and absence of mistake are demonstrated by the feasibility study that Tribeca prepared for RMS, which assumed no claim losses and payment of a $1.2 million premium each year.

57

[*57] c. Valid and Binding Policies

We have held that policies were valid and binding when “[e]ach insurance policy identified the insured, contained an effective period for the policy, specified what was covered by the policy, stated the premium amount, and was signed by an authorized representative of the company.” Securitas Holdings, Inc., T.C. Memo. 2014-225, at *28; see also R.V.I. Guar. Co., 145 T.C. at 231 (finding that policies were valid and binding when the insured filed claims for covered losses and the captive insurance company paid them). We have also examined factors beyond whether the policies are simply binding such as conflicting or cookie-cutter policy terms or the delivery of claims-made policies after the end of the claims period. See Avrahami, 149 T.C. at 194 (examining conflicting policy terms); Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *44 (“Writing and delivering ‘claims made’ insurance policies after the claim period is, we find, abnormal and is to any reasonable observer just plain silly.”); Reserve Mech. Corp., T.C. Memo. 2018-86, at *54 (describing policies as cookie-cutter and not necessarily appropriate); see also Syzygy Ins. Co., T.C. Memo. 2019-34, at *42 (“Here the dispute surrounding valid and binding policies centers on whether the policies were timely issued, identified the insured, and specified what was covered by the policies.”). Overly restrictive provisions generally indicate that the parties to an arrangement intended their arrangement to look like insurance without actually providing it. Cf. Syzygy Ins. Co., T.C. Memo. 2019-34, at *32.

We find that the policies were not valid and binding. Our first concern is the delivery of claims-made policies well into the coverage period without binders in place in the interim. 59 See Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *44. While RMS’s captive policies during the years at issue were issued midway through their coverage periods, rather than after, the late issuances still create substantial doubt about the validity and binding effect of the policies. In the absence of a binder, an insurer might choose to increase premiums or change the policy terms before issuing the policies, or simply not issue the policies at all, if, for example, a covered loss occurred between the coverage period inception date (i.e., January 1) and the policy issuance

59 Similarly, we are also concerned by the January 28, 2013, Change

Endorsement that materially changed RMS’s Worker’s Compensation Deductible / SIR Reimbursement policy after the end of the coverage period.

58

[*58] date. 60 It is also unclear whether Provincial or Risk Retention would have been obligated to pay a claim made between those dates under a policy that had not yet been issued.

The only apparent purpose for issuing restrictive claims-made policies was to accommodate the desire of pool participants to receive their funds back relatively quickly after they were paid. The policies also contained ambiguous wording. For example, an independent adjuster could have concluded that the Workers’ Compensation Deductible / SIR Reimbursement policies applied only to accidents occurring during the policy year that resulted in a deductible invoice received during the policy period. The failure of the policies to make clear whether this was the case is a significant failing given that all paid claims that RMS filed under its 2012–14 captive policies were made under this policy. Cf. Avrahami, 149 T.C. at 194 (discussing a policy with terms indicative of both a claims-made policy and an occurrence policy).

The 2013 and 2014 general terms and conditions also conditioned claims payment on a requirement that the insured be in compliance with all terms of its engagement letter with Artex, as well as all terms of the Master Reinsurance Agreement between Provincial and any applicable reinsurer, and remain an ongoing client of Artex. The binding effect of the policies, if any, therefore depended in substantial part on considerations extraneous to the policies themselves.

The parties to the arrangement did not themselves treat the policies as valid and binding. They used board resolutions to pay claims when the terms of the policies did not support the claims, or to document claims that had already been paid. Alternatively, RMS simply took and repaid loans from the captive on its own terms if a policy did not cover a desired use of the funds. Mr. Candland sometimes decided the amount

60 While, in addition to the annual coverage periods, evergreen policy periods

nominally allowed each policy to remain in force until canceled, the record is clear that the essential terms of each policy were set forth in annual renewal endorsements that used an annual coverage period. The policies are devoid of any indication, for example, of whether any additional premiums were due and owing during an evergreen period for which no renewal endorsement was in place; what coverage obtained during the interregnum (e.g., the coverage for the last renewal endorsement or the coverage for a later-issued renewal endorsement that purported to have a retroactive coverage period); how claims made during it were to be handled; or whether procedures differed depending on whether a later renewal endorsement was or was not issued. We regard it as nothing more than an attempt by Artex and Provincial to imbue their practice of belatedly issuing insurance contracts with a legitimacy on paper that it lacked in fact.

59

[*59] of premiums that he wished to pay at the end of the year, and Artex facilitated this practice.

Other discrepancies underscore our lack of confidence that the policies were valid and binding. The provision in the general terms and conditions that all premiums were earned at inception exclusively benefits the insurer at the expense of the insured and is at odds with the typical insurance industry practice of providing refunds (less early cancellation penalties). Cf. Syzygy Ins. Co., T.C. Memo. 2019-34, at *32. While such a provision is not necessarily fatal, we view its inclusion as unusual under the circumstances here. The feasibility study’s failure to mention RMS’s workers’ compensation needs, which featured prominently in the captive program, while mentioning goodwill and identity protection policies, which RMS never purchased, undermines petitioners’ contention that the policies were intended to provide valid and binding coverage for actual insurance needs. In sum, the policies were designed only to resemble insurance policies superficially while in reality giving the parties to the arrangement the option to proceed, or not to proceed, with funding the policies until well into the coverage period.

d. Reasonableness of Premiums

The next question is whether Provincial’s premiums were reasonable and the result of an arm’s-length transaction. See Avrahami, 149 T.C. at 194–96. We find that they were not. Mr. Candland provided Artex with an amount he was willing to pay or a target premium for all policies purchased regardless of coverage. Mr. Candland sometimes requested increases in RMS’s premiums. Cf. Syzygy Ins. Co., T.C. Memo. 2019-34, at *33–34 (“In an arm’s-length negotiation, an insurance purchaser would want to negotiate lower premiums instead of higher premiums.”). The target premiums Mr. Candland provided played an outsized role in Artex’s purported underwriting.

Before discussing premium determination and underwriting in detail, we pause to consider RMS’s coverage needs. RMS had a comprehensive program of insurance obtained in the commercial marketplace, some of which it negotiated at arm’s length with its clients. While some carriers that Mr. Hill approached declined coverage, there is no credible evidence that RMS was unable to obtain any type of insurance coverage that it sought or that it did not maintain a robust program of commercial insurance during the years at issue.

60

[*60] Given that RMS agreed to insurance requirements in its client contracts and passed on insurance costs to its clients, it is odd that there is no evidence that RMS consulted in any detail with its clients about the massive insurance costs that it incurred through the captive program. In any case, the reasons Mr. Candland offered at trial for obtaining each policy were largely pretextual. While we grant that some of his testimony may explain why he chose a given coverage over other captive coverages that Tribeca or Artex offered, they do not explain why he would have paid such exorbitant sums for them in the context of RMS’s business. Taking RMS’s tax returns literally, the amounts paid for insurance reduced RMS from a profitable enterprise to one that was approximately breaking even. Most of the captive coverages were not required by RMS’s contracts with its clients, and there is no credible evidence indicating that RMS replaced any of its commercial coverages with any of the captive coverages. 61 A much more detailed explanation of the need for such expensive policies was warranted than the ones provided by Mr. Candland. This is especially true given RMS’s specialization in dealing with insurance issues on behalf of its clients.

Moving on to premium determination, the Provincial policies were not objectively rated by evaluating the risk and magnitude of loss on a prospective basis informed by detailed underwriting. The premiums were also inflated by numerous subjective, judgment-driven factors, each of which could modify the premiums significantly; and there is very little documentation to support how Artex applied these factors. The captive risk factor was especially inappropriate because its stated object should have been addressed by making capital

61 Regarding the Worker’s Compensation Deductible / SIR Reimbursement policies, petitioners argue that the premiums RMS paid for this coverage were “significantly less than the discount provided by Crum & Forster for the large deductible.” Petitioners thus appear to argue that these policies replaced the deductible amount on the corresponding Crum & Forster policies in a cost-effective manner.

We are not convinced. Mr. Candland emailed an Artex employee during the years at issue to inform him that RMS had raised its deductible from $100,000 to $250,000 on the Crum & Forster policy and stated: “This should make it easier to justify our $1,200,000 captive contribution.” It thus appears that petitioners used the deductible amount on the Crum & Forster policies to justify the captive contribution, not that the captive coverage replaced the Crum & Forster deductible amount in a demonstrably cost-effective manner. Furthermore, petitioners did not present credible evidence to prove how Crum & Forster calculated the discount listed on its billing statements for large deductibles, how the amount of the discount varied with the amount of the deductible, or that Artex incorporated the amount of the deductible or the large deductible discount in its purported underwriting of the captive policies.

61

[*61] contributions or obtaining aggregate stop-loss reinsurance, not charging the insured additional premiums. See Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *47. This is not merely an academic proposition: RMS’s own VEBA had stop-loss insurance coverage. Furthermore, total annual premiums on RMS’s captive coverages always hovered around $1.2 million, the section 831(b) limit, even when coverage types or limits varied or RMS’s revenue or payroll changed.

The amounts of premiums charged were also patently unreasonable. The average rate-on-line for RMS’s captive policies during the years at issue was more than ten times greater than the average rate-on-line for comparable commercial insurance policies, even though there is no credible evidence indicating that RMS had major issues with its existing commercial insurance coverage, or in obtaining the insurance required by its client contracts. A higher rate-on-line means that insurance coverage is more expensive per dollar of coverage and could therefore lead to a greater deduction for premiums. See Syzygy Ins. Co., T.C. Memo. 2019-34, at *31. There is no credible evidence in the record that these charges were justified by a substantial loss history from RMS or the pool. 62

We are also unconvinced that the Artex underwriting staff had sufficient expertise to exercise the judgment required by these subjective factors and the numerous types of policies they underwrote. Moreover, Artex did not take into account the fact that it was often underwriting claims-made policies toward the end of the applicable claims period when it priced premiums. While captive insurance companies may legitimately be more profitable than large commercial insurance companies in some cases, the substantial profits during the years at issue here appear to be derived mostly from a failure to determine the premiums actuarially.

The premium determination process was not adequately supported by detailed underwriting. While Artex appeared familiar with the practice of obtaining detailed applications in insurance underwriting, it obtained virtually no information from RMS that would have informed the underwriting process. Nor did Artex underwriters adequately account for RMS’s loss experience over time. The Provincial policies could hardly have had reasonable premiums without adequate

62 Neither is there any credible evidence in the record indicating that Artex,

the Provincial Pool, or Risk Retention were burdened by unusually high overhead expense.

62

[*62] information and expertise to price the policies. While the proverbial broken clock may be right twice a day, this is an inadequate method for pricing insurance policies.

Neither is there any credible evidence in the record that RMS achieved cost savings through the captive program. Risk Retention actually had a significant enough surplus during the years at issue that it financed some of RMS’s commercial insurance premiums through the use of the captive insurance premiums that it received. Risk Retention also financed group health plan claims for RMS and buy-sell life insurance premiums for Mr. Keating and Mr. Candland, and it extended miscellaneous loans to RMS to finance its operations and posted collateral for its workers’ compensation deductibles. Repeated execution of these agreements shows confidence on RMS’s and petitioners’ part that Risk Retention’s surplus would not be needed to pay substantial claims on the captive policies and that Provincial was overcharging for the coverage provided. Likewise, petitioners’ failure to consult Mr. Hill or another qualified insurance broker about whether the captive coverages were available on a more cost-effective basis in the commercial marketplace shows their intent not to use the captive arrangement to provide actual insurance.

We also briefly discuss some issues related to premium determination in the Provincial Pool specifically. There is no apparent reason for allocating each captive’s premiums approximately 51% to the pool policies and 49% to the facultative policies other than to come within a perceived IRS safe harbor. In a typical captive arrangement involving quota-share reinsurance, one would expect members to pay individually and actuarially determined premiums based on the expectation of each member’s losses. A one-size-fits-all approach to allocating premiums between layers of reinsurance, like the one used here, suggests that the allocation is inconsistent with an actual actuarial determination. See Avrahami, 149 T.C. at 186; Syzygy Ins. Co., T.C. Memo. 2019-34, at *36; Reserve Mech. Corp., T.C. Memo. 2018-86, at *43. The one-size-fits-all approach was particularly strange here because the Provincial Pool reinsured dozens of lines of coverage during the years at issue, not a homogenous pool of risks, and because other insureds operated businesses that were highly dissimilar to RMS’s. Cf. Avrahami, 149 T.C. at 186–88 (finding a one-size-fits-all approach to risk pool premium pricing objectionable even when only one form of coverage was at issue).

63

[*63] e. Payment of Claims

Risk Retention and Provincial paid claims. Nonetheless, the process by which those claims were handled was abnormal. See Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *42–43, *48 (holding that the abnormal payment of claims supports a conclusion that an arrangement is not insurance in the commonly accepted sense).

RMS used a board resolution to pay the legal settlement with Wausau from Risk Retention’s funds despite numerous defects with that claim. Throughout the years at issue RMS also frequently provided Artex with deductible billing invoices on its workers’ compensation policy with Crum & Forster only after Risk Retention had already paid the invoices. Not only is it highly unusual for claims approval to occur after claims payment, but it also shows that Artex gave little timely review to these claims.

Artex effectively allowed RMS to manage its own claims under the Worker’s Compensation Deductible / SIR Reimbursement policy. Artex also failed to place proper controls on RMS’s insistence that it be allowed to directly manage the claims process as though no formal captive insurance program were in place. Risk Retention, a purported reinsurer, played an inappropriate role in the direct payment of claims.

For 2012, 2013, and 2014 the Provincial Pool paid claims amounting to 0.324%, 3.322%, and 3.019% of total pool premiums, respectively, which resulted in $1,921, $20,209, and $18,732 quota-share payments by Risk Retention for the respective years at issue. These amounts are relatively small compared to the approximately $1.2 million that RMS paid in captive premiums each year, or even compared only to the 51% or so of those premiums that Artex allocated to pool premiums. The Provincial Pool had a very low ratio of losses to premiums compared to the insurance industry as a whole, which resulted in nearly a full round trip of premiums, considering that the captives participating in it were affiliated with their insureds.

Perhaps more concerning, however, is the manner in which Artex and Provincial handled the claims. Artex added or altered policies for its clients retroactively in order to permit them to file claims against the Provincial Pool or to reduce their premiums if they were unable to pay in full. It did not consistently enforce the prior-knowledge limitation when adjusting claims or treat claims as uncovered because no coverage

64

[*64] was in effect at the time of the loss, even though it should have. It also approved some claims on the basis of only slight documentation. Furthermore, the contractual linkage of consulting, insurance, and reinsurance agreements had an inappropriate influence on claims management, as did the staffing overlap between Artex’s underwriting and claims functions.

Artex facilitated the use of board resolutions to provide an end-run around the claims process. Routine use of these ex gratia payments is counter to standard claims procedures. While a bona fide insurance company may settle a claim with an insured because of a reasonable expectation of coverage, its relationship with a client, or an acknowledgment that the insurance company could have done something better, there is no credible evidence indicating that these reasons motivated Artex’s decision-making. Instead, the claims process was largely illusory, and Artex used board resolutions precisely to address situations where insureds wanted to access funds held by their captive insurer but had no reasonable expectation of coverage.

2. Conclusion

Petitioners have not proven that RMS’s payments that they seek to deduct as insurance expenses were for insurance in the commonly accepted sense. Petitioners have therefore failed to prove that the payments were for insurance for federal income tax purposes.

B. Effect on Petitioners

Having determined that the microcaptive arrangement among petitioners, RMS, Risk Retention, Provincial, and Artex was not insurance, we proceed to discuss the legal effect of that conclusion on petitioners for the years at issue.

1. Section 162

Section 162(a) allows taxpayers a deduction for all ordinary and necessary expenses paid or incurred during the taxable year in carrying on any trade or business. To be deductible under section 162(a), an expense must be both ordinary and necessary. Welch v. Helvering, 290 U.S. at 113. An expense is necessary if it is appropriate and helpful to the development of the taxpayer’s business. Commissioner v. Tellier, 383 U.S. 687, 689 (1966); Welch v. Helvering, 290 U.S. at 113. An ordinary expense is one that is “normal, usual, or customary. . . . [T]he transaction which gives rise to it must be of common or frequent

65

[*65] occurrence in the type of business involved.” Deputy v. Du Pont, 308 U.S. 488, 495 (1940). It is “the kind of transaction out of which the obligation [to pay] arose and its normalcy in the particular business which are crucial and controlling.” Id. at 496. In addition to being ordinary and necessary, as well as paid or incurred during the taxable year, a deductible business expense must be reasonable in amount. See United States v. Haskel Eng’g & Supply Co., 380 F.2d 786, 788–89 (9th Cir. 1967) (“An expenditure may be, by its nature, ordinary and necessary, but at the same time it may be unreasonable in amount. In such a case only the portion which was reasonable would qualify for a deduction under § 162(a).”); Hopkins v. Commissioner, T.C. Memo. 2005-49, slip op. at 16–17. Whether an expense is deductible under section 162 is a question of fact to be decided on the basis of all relevant facts and circumstances. See Cloud v. Commissioner, 97 T.C. 613, 618 (1991).

Premium payments to a captive insurance company that are not for insurance are generally not ordinary and necessary business expenses and cannot be deducted under section 162(a). See Avrahami, 149 T.C. at 174, 199. We have recognized, however, that “[i]n the context of captive insurance there may be instances where noninsurance payments for indemnification protection might be appropriate and helpful to the development of the insured.” Syzygy Ins. Co., T.C. Memo. 2019-34, at *46; cf. id. at *47–48. Nonetheless, “[t]he cases tell us to be more skeptical about expenses between related parties. . . . The reason is that ‘expenses’ from one related party to another are more likely to be distributions of profits, which are not deductible.” Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *29.

Petitioners have not established that the captive premium payments were ordinary. The payments were not for insurance. RMS’s clients did not require RMS to obtain the captive coverages, even though they required RMS to maintain certain insurance coverage. Mr. Hill had never heard of some of the captive coverages, and petitioners never directed him or any other insurance broker to seek out many of the coverages in the commercial marketplace before implementing the captive program. Petitioners have not attempted to establish that businesses similar to RMS typically relied on the types of coverages provided by Artex, on the terms provided by Artex, for their coverage needs. While petitioners made some claims for deductible reimbursements, we see no credible evidence in the record indicating that businesses like petitioners’ typically purchase deductible or SIR reimbursement policies rather than simply paying their deductibles

66

[*66] directly. Even if they do, we have not been directed to any evidence that they purchase policies with the restrictive or ambiguous terms found in the Artex and Provincial policies. Likewise, we have seen no evidence that similar businesses purchase policies from insurance companies using the irregular pricing and claims handling practices that Artex and Provincial used.

Regarding the fees paid to Artex and PRS in particular, petitioners have not proven that captive management fees, or fees for a paying agent controlled by a captive management company, are normal, usual, or customary in RMS’s line of business. Cf. Reserve Mech. Corp., T.C. Memo. 2018-86, at *50 (finding a captive’s management entirely by a captive management company to be a factor weighing against a determination that the captive operates as an insurance company).

Petitioners argue that “insurance is normal, usual and customary for many businesses, as risk shifting has been around since groups gathered in Lloyds coffee house in London to indemnify ship owners for cargo they might lose at sea.” Nonetheless, our concern is not with the ordinariness of insurance or indemnification payments in general, but with the ordinariness of the particular “kind of transaction out of which the obligation [to pay] arose and its normalcy in the particular business” here. Deputy v. Du Pont, 308 U.S. at 496. Given that petitioners have failed to establish that the expenses were ordinary, we need not decide whether the expenses met the other requirements for deductibility under section 162. We also need not address petitioners’ argument that the reasonable portions of the premiums should be allowed.

2. Section 165

Petitioners argue that “[i]f the Court determines the premiums paid are not deductible under I.R.C. § 162 or that the transaction is not insurance or otherwise lacks economic substance, the losses paid are deductible by RMS in the year they were sustained and paid by Risk Retention.” Specifically, petitioners argue,

The Court must decide whether the premiums paid by RMS are deductible insurance expenses or reserves set aside for self-insurance. . . . If the Court determines that the transaction is not insurance for federal income tax purposes, the transaction should be treated as a self-insured reserve. . . . [I]f the taxpayer utilizes a self- insured reserve fund, the allowable deduction is limited to

67

[*67] the losses actually incurred and paid out of the reserve. . . .

Deductions are allowed for losses sustained during a taxable year, for which a taxpayer is not compensated by insurance, or otherwise.

Petitioners thus appear to argue that we should characterize the arrangement as a self-insurance reserve and permit deductions as “claims” (generally, workers’ compensation deductible payments63) were made. Respondent disputes this argument on the merits and has also affirmatively invoked the duty of consistency.

Petitioners’ proposed characterization of Risk Retention, a corporation and a separate taxpayer from both RMS and petitioners, as a mere reserve or account of RMS is not borne out by the record. Even if it was RMS’s pocketbook, it was an incorporated one and therefore a separate entity. 64 See Moline Props., Inc. v. Commissioner, 319 U.S. 436, 438–39 (1943) (“Whether the purpose be to gain an advantage under the law of the state of incorporation or to avoid or to comply with the demands of creditors or to serve the creator’s personal or undisclosed convenience, so long as that purpose is the equivalent of business activity or is followed by the carrying on of business by the corporation, the corporation remains a separate taxable entity.” (Footnotes omitted.)).

A taxpayer generally may not deduct another person’s expense or loss. See Deputy v. Du Pont, 308 U.S. at 493–94. The Ninth Circuit has stated that

if a taxpayer chooses to conduct business through a corporation, he will not subsequently be permitted to deny the existence of the corporation if it suits him for tax purposes. . . . In particular corporate shareholders will not be permitted to claim deductions for ordinary and

63 In 2012 Risk Retention also paid $3,452 for a claim filed under RMS’s 2011

Employment Practices Deductible / SIR Reimbursement policy.

64 Notwithstanding our conclusion below that Risk Retention’s section 953(d)

elections were invalid for the years at issue by reason of its failure to satisfy section 953(d)(1)(B), we regard the section 953(d) elections that Risk Retention filed as prima facie evidence that it was a foreign corporation because only foreign corporations are eligible to make a section 953(d) election. See § 953(d)(1). We also deem petitioners’ argument that Risk Retention’s section 953(d) elections were valid to be a concession that Risk Retention was a foreign corporation if the section 953(d) elections were not valid.

68

[*68] necessary expenses incurred by the corporation even though paid by the shareholders.

Betson v. Commissioner, 802 F.2d 365, 368 (9th Cir. 1986), aff’g in part, rev’g in part T.C. Memo. 1984-264. Neither respondent nor petitioners dispute that RMS shifted risks, including liability for payment of certain commercial insurance policy deductibles, to Risk Retention, a separate taxable entity, in exchange for making premium payments. The parties dispute whether the arrangement by which it did so was insurance or was otherwise a deductible expense or loss, but there is no factual basis for a finding that RMS retained the liabilities it shifted to Risk Retention or incurred the losses when they came due. Risk Retention’s assumption of RMS’s liability for workers’ compensation deductibles is also evidenced by Risk Retention’s repeated execution of deductible agreements with United States Fire Insurance Co. and the substantial sums that Risk Retention paid into the collateral fund before and during the years at issue pursuant to the deductible agreements. Petitioners could have set up an unincorporated self-reserve fund or account and deducted the losses as they occurred, but they did not do so. Petitioners’ argument also ignores the requirement that a corporation affirmatively make an S corporation election in order for it and its shareholders to receive passthrough entity treatment. See generally § 1362; Treas. Reg. § 1.1362-6. Risk Retention never did so.

Petitioners read our and the Ninth Circuit’s caselaw as requiring a binary choice between a finding that the arrangement involved either payments for insurance or a finding that petitioners set funds aside in a reserve for self-insurance. Although we are skeptical of this reading, we need not decide whether it is correct because petitioners have not suggested any other characterizations of the arrangement; and even if Risk Retention was a reserve for self-insurance, it was an incorporated one. Petitioners chose to transact business through the corporate form rather than on RMS’s or their own account; it follows that deductions arising from the liabilities they took pains to shift to Risk Retention belong to Risk Retention. The general rule that losses from a self-insurance reserve are deductible as they are incurred does not conflict with a finding that such deductions belong to a taxpayer other than petitioners. Petitioners have failed to meet their burden of proving that the amounts paid by Risk Retention as claims were losses incurred by RMS.

Petitioners argue that Spring Canyon Coal Co. v. Commissioner, 43 F.2d 78 (10th Cir. 1930), aff’g 13 B.T.A. 189 (1928), provides support

69

[*69] for their position. In that case a taxpayer self-insured its workers’ compensation obligations by setting up a separate fund into which it paid premiums. Id. at 78–79. The court held that the taxpayer was not entitled to deduct the self-insurance premiums but noted that “its right to deduct payments made out of the fund,” id. at 79, was not in dispute. This case is inapposite because the taxpayer “carried the fund on its books as an asset,” id., whereas here Risk Retention was a separate entity. Expenses or losses paid out of the taxpayer’s fund in Spring Canyon Coal Co. were the taxpayer’s own expenses or losses; but the expenses or losses arising under the deductible reimbursement policies that RMS purchased were Risk Retention’s.

Petitioners also invoke Anesthesia Service Medical Group, Inc. v.

Commissioner, 825 F.2d 241 (9th Cir. 1987), aff’g 85 T.C. 1031 (1985), for support. In that case, the court held that a taxpayer’s contributions to a grantor trust it established to pay potential malpractice claims against its employees were not deductible. The court stated that “[a]mounts placed into self-insurance reserves are not deductible business expenses under I.R.C. § 162(a). . . . Rather, the taxpayer must wait until a loss recognizable under I.R.C. § 165 occurs.” Id. at 242. We agree with this general statement of the law, but the Ninth Circuit also held that the taxpayer’s “ability to use Trust funds to discharge its potential vicarious liability requires taxing the Trust’s income” to the taxpayer. Id. at 243. The cited case is therefore distinguishable on the ground that it did not involve a separate entity to which the deductions were attributable. Instead, it involved only a grantor trust. Furthermore, the cited case is distinguishable because the taxpayer never shifted its risk of loss to the trust; on the contrary, it was obligated to reimburse the trust for any shortfall caused by claims. See Anesthesia Serv. Med. Grp., Inc., 85 T.C. at 1039–41 (holding that a contributory agreement between the taxpayer and its trust alone indicated that the risk of loss did not shift from the taxpayer). In the cases at bar, neither petitioners nor respondent disputes that RMS shifted risks to Risk Retention through the captive arrangement. Risk Retention, a separate entity, in fact retained the risks that RMS shifted to it, and the tax treatment follows from that fact.

Finally, petitioners argue that generally accepted accounting principles (GAAP) support a finding that transactions not qualifying as insurance should be treated as reserves or deposit arrangements. Regardless of whether petitioners’ application of the accounting rules it cites to the circumstances here is correct, nontax rules of accounting do not control, or even necessarily inform, the determination of a taxpayer’s

70

[*70] tax liability. See AMERCO, 96 T.C. at 35–36 (rejecting Commissioner’s expert’s reliance on GAAP in a captive insurance case as “simply irrelevant to the tax law considerations before this Court” and stating that “[i]t is clear that the Federal income tax does frequently perceive related corporate entities as separate enterprises and taxpayers”); see also Foster v. United States, 303 U.S. 118, 120–22 (1938); Old Colony R. Co. v. Commissioner, 284 U.S. 552, 562 (1932). Given our holding that the deductions at issue belonged to Risk Retention, not RMS, we do not reach respondent’s alternative argument that the duty of consistency applies to bar the deductions.

3. Dividends

We must decide the tax characterization of the distributions that Risk Retention made to Mr. Keating and Mr. Candland in 2012 and 2014. Under section 301(c), a corporation’s distribution of property to a shareholder generally may, in whole or in part, (1) constitute a dividend, (2) reduce the adjusted basis of the shareholder’s stock to the extent it is not a dividend, or (3) be treated as gain from the sale or exchange of property to the extent it both exceeds the adjusted basis of the stock and is not a dividend. Section 316(a) generally defines a dividend as a distribution of property made by a corporation to its shareholders out of its earnings and profits accumulated after February 28, 1913, or out of its earnings and profits of the taxable year without regard to the amount of earnings and profits at the time the distribution was made. Petitioners conceded in both their Simultaneous Opening Brief and their Errata to Petitioners’ Simultaneous Opening Brief that “the dividends were paid from earnings and profits.” We therefore deem petitioners to have conceded that the distributions are dividends for purposes of sections 301(c)(1) and 316(a).

Petitioners attempt to walk back their concession in their Simultaneous Answering Brief, in which they argue:

Respondent has failed to calculate Risk Retention’s earnings and profits if the transaction is not insurance for federal income tax purposes, establishing that the payments petitioners’ [sic] Keating and Candland received were still paid out of earnings and profits, [and] therefore, are dividends. Changing the character of the payments to Risk Retention to something other than premiums would change the earnings and profit calculation for Risk Retention. . . . [Section 964(a)] provides the earnings and

71

[*71] profits of a foreign controlled corporation are calculated in the same manner as a domestic corporation. Funds that are paid out of a C Corporation that are not paid out of earnings and profits are taxed as return of capital. I.R.C. § 301.

Petitioners’ late attempt to withdraw their earlier concession subverts our briefing schedule and takes respondent by surprise by not permitting him to respond to this new argument. We decline to allow the withdrawal of the concession. See Estate of DeMuth v. Commissioner, T.C. Memo. 2022-72, at *8–9 (enforcing concession that opposing party relied on in drafting its simultaneous answering brief), aff’d, No. 22-3032, 2023 WL 4486739 (3d Cir. July 12, 2023).

Because Risk Retention’s distributions to Mr. Keating and Mr.

Candland in 2012 and 2014 were dividends, the only issue is whether they were ordinary or qualified dividends. Section 1(h)(11) provides preferential tax rates for “qualified dividend income” if the dividend is received from a domestic corporation or a qualified foreign corporation. See § 1(h)(11)(B)(i); Avrahami, 149 T.C. at 199. We start with the latter category. A qualified foreign corporation is generally any foreign corporation that is either (1) incorporated in a possession of the United States or (2) eligible for the benefits of a comprehensive income tax treaty with the United States which the Secretary determines is satisfactory and which includes an exchange of information program. See § 1(h)(11)(C). Anguilla is not a possession of the United States. The IRS has published a list of income tax treaties satisfying the statutory requirements, see I.R.S. Notice 2011-64, 2011-37 I.R.B. 231, and Anguilla is not on it, cf. Smith v. Commissioner, 151 T.C. 41, 57 (2018). Risk Retention was not a qualified foreign corporation during the years at issue.

This leaves us to decide whether Risk Retention’s election under section 953(d) to be treated as a domestic corporation is valid. To make a valid section 953(d) election, a controlled foreign corporation, as defined in section 957(a), must qualify under part I (life insurance companies) or II (other insurance companies) of subchapter L. See § 953(d)(1)(B); Avrahami, 149 T.C. at 198. To qualify for either part, a company must meet the definition of “insurance company” in section 816(a). See §§ 816(a) (flush language), 831(c), 953(d)(1)(B); Avrahami, 149 T.C. at 198. This means that more than half of its business during the taxable year must be the issuing of insurance or annuity contracts or the reinsuring of risks underwritten by insurance companies. We

72

[*72] have already held that the captive arrangement did not constitute insurance, and petitioners have not proven that Risk Retention had other business that constituted insurance. Risk Retention was not an insurance company for the years at issue, and its section 953(d) elections were invalid. We therefore hold that the distributions Risk Retention made to petitioners in 2012 and 2014 should be taxed at ordinary income rates.

Petitioners argue that the dividends are qualified because section 953(d)(2)(B) provides that if a corporation which made a section 953(d) election fails to meet the requirements of section 953(d)(1)(B) “for any subsequent taxable year, such election shall not apply to any taxable year beginning after such subsequent taxable year.” (Emphasis added.) Therefore, petitioners argue, “if respondent determined that during 2012 Risk Retention first failed to meet the requirements for its [section] 953(d) election, then the election would become inapplicable beginning in 2013.” As an initial matter, this argument, taken on its own terms, does not aid petitioners with respect to the characterization of the 2014 dividends. In addition, this argument is wrong on the merits.

We consider the validity of a section 953(d) election at the time it was made before deciding whether it was terminated under section 953(d)(2)(B). See Chapman Glen Ltd. v. Commissioner, 140 T.C. 294, 318–20 (2013). Risk Retention’s section 953(d) election states that it “shall be effective as of the first day of the corporation’s taxable year (including a short taxable year) commencing” on November 24, 2008. Petitioners have failed to meet their burden of proving that Risk Retention was an insurance company within the meaning of section 953(d)(1)(B) in 2008. During that year Mr. Candland met with Mr. Kotch; and while his handwritten notes show that they discussed the taxation of a captive insurer, they do not reflect any discussion of RMS’s coverage needs. After Risk Retention was formed on November 25, 2008, RMS paid it only $500,000 in 2008, despite initially setting a premium budget of $800,000 for this period of little over a month. Artex did not even finalize the policies until 2009. After another payment in March 2009, RMS ultimately paid over $670,000 for purported insurance policies that, viewed charitably, offered little more than a month of coverage. Cf. Reserve Mech. Corp., T.C. Memo. 2018-86, at *57 (noting a large amount paid for only one month of insurance coverage). The declaration pages in the record are backdated to December 10, 2008, and the coverage was made retroactive to November 25, 2008, despite the policies’ preparation in 2009. This is not an insurance arrangement.

73

[*73] Petitioners have failed to prove that Risk Retention was a life or nonlife insurance company in 2008, and Risk Retention’s section 953(d) election was accordingly invalid when made. Furthermore, petitioners have failed to prove that even if Risk Retention’s section 953(d) election was valid when made, Risk Retention continued to meet the requirements of section 953(d)(1)(B) for 2009, 2010, and 2011 as would be necessary to prevent a termination under section 953(d)(2)(B) before the years at issue. 65 Our review of the record shows that, if anything, virtually all aspects of the purported insurance arrangement were even more deficient in 2009–11 than they were during the years at issue.

Petitioners argue that section 953(d)(2)(A) provides that the section 953(d) election, once made, applies for subsequent taxable years unless revoked with the consent of the Secretary. This is a misstatement of section 953(d)(2)(A), which provides that its rule applies “[e]xcept as provided in subparagraph (B).” Petitioners also argue that Revenue Procedure 2003-47 provides support for their position. It does not, because it expressly states that “[o]nce approved, the election generally remains effective for each subsequent taxable year in which the requirements of this revenue procedure and section 953(d) are satisfied unless revoked by the electing corporation with the consent of the Commissioner.” Rev. Proc. 2003-47, § 4.02(1), 2003-2 C.B. 55, 55 (emphasis added). In sum, Risk Retention was a foreign corporation, not a domestic corporation, and its dividends in 2012 and 2014 were not qualified because Anguilla did not have a comprehensive income tax treaty with the United States during the years at issue.

V. Accuracy-Related Penalties

The last issue is whether accuracy-related penalties under section 6662(a) are justified. Respondent determined accuracy-related penalties on grounds of negligence or disregard of rules or regulations, see § 6662(b)(1), (c), or in the alternative, on grounds of substantial understatements of income tax, see § 6662(b)(2), (d). Respondent bears the burden of production with respect to the accuracy-related penalties. See § 7491(c). Once respondent comes forward with sufficient evidence

65 The evidentiary objections that petitioners made to some items of evidence

on the grounds that they concerned taxable years prior to the years at issue are somewhat confusing in view of the necessity of such evidence for petitioners to prove that Risk Retention’s section 953(d) election did not terminate before the years at issue. At a minimum, even without considering other grounds for their relevance, petitioners opened the door to such evidence by placing the qualified dividend rate at issue.

74

[*74] to show that it is appropriate to impose a particular penalty, petitioners have the burden of proof to show that respondent’s penalty determination is incorrect, including the burden of proving that penalties are inappropriate because of reasonable cause. See Higbee v. Commissioner, 116 T.C. 438, 446–47 (2001). The parties have stipulated that respondent complied with the written supervisory approval requirements of section 6751(b) for the accuracy-related penalties determined against petitioners for each year at issue.

Section 6662(a) imposes a 20% penalty on the portion of an underpayment of tax attributable to any substantial understatement of income tax, see § 6662(b)(2), or negligence or disregard of rules or regulations, see § 6662(b)(1). An understatement is substantial if it exceeds the greater of (1) 10% of the tax required to be shown on the return for the taxable year, or (2) $5,000. See § 6662(d)(1)(A). Negligence includes any failure to make a reasonable attempt to comply with the provisions of the Code, and disregard includes any careless, reckless, or intentional disregard. See § 6662(c). The understatements in these cases are substantial as an arithmetic matter for all petitioners and for all years at issue. It is therefore unnecessary for us to determine whether the underpayments are attributable to negligence or disregard of rules or regulations. See Avrahami, 149 T.C. at 204–05; see also Treas. Reg. § 1.6662-2(c) (providing that only one accuracy-related penalty for a given year may be applied with respect to any given portion of an underpayment, even if that portion is subject to the penalty on more than one ground). Respondent has met his burden of production with regard to the accuracy-related penalties, and petitioners have the burden of proof to show that respondent’s penalty determinations are incorrect.

Petitioners assert that they had reasonable cause for, and acted in good faith with respect to, the underpayments. Section 6664(c)(1) provides that the penalty under section 6662(a) shall not apply to any portion of an underpayment if it is shown that there was reasonable cause for the taxpayer’s position and the taxpayer acted in good faith. See Higbee, 116 T.C. at 448. This determination is made on a case-by-case basis, taking into account all of the pertinent facts and circumstances. See Treas. Reg. § 1.6664-4(b)(1). Generally, the most important factor is the extent of the taxpayer’s effort to assess the taxpayer’s proper tax liability. See id. For underpayments related to passthrough items, we look at all pertinent facts and circumstances, including the taxpayer’s own actions, as well as the actions of the passthrough entity. See id. para. (e). Reliance on professional advice

75

[*75] may constitute reasonable cause and good faith, but only if, considering all the circumstances, such reliance was reasonable. See id. paras. (b)(1), (c)(1); see also Freytag v. Commissioner, 89 T.C. 849, 888 (1987), aff’d, 904 F.2d 1011 (5th Cir. 1990), aff’d, 501 U.S. 868 (1991). Advice is “any communication . . . setting forth the analysis or conclusion of a person, other than the taxpayer, provided to (or for the benefit of) the taxpayer and on which the taxpayer relies, directly or indirectly, with respect to the imposition of the section 6662 accuracy- related penalty.” Treas. Reg. § 1.6664-4(c)(2). Advice does not have to be in any particular form. Id.

Reasonable cause exists if a taxpayer relies in good faith on the advice of a qualified tax adviser where the following three elements are present: (1) the adviser was a competent professional who had sufficient expertise to justify the reliance; (2) the taxpayer provided necessary and accurate information to the adviser; and (3) the taxpayer actually relied in good faith on the adviser’s judgment. See Neonatology Assocs., P.A., 115 T.C. at 99. Reliance may be unreasonable if the adviser is a promoter of the transaction. Id. at 98. A promoter is “an adviser who participated in structuring the transaction or is otherwise related to, has an interest in, or profits from the transaction.” 106 Ltd. v. Commissioner, 136 T.C. 67, 79 (2011) (quoting Tigers Eye Trading, LLC v. Commissioner, T.C. Memo. 2009-121, slip op. at 48), aff’d, 684 F.3d 84 (D.C. Cir. 2012).

There is no credible evidence in the record that petitioners or RMS took any substantial steps to ascertain their proper tax liability. Petitioners rely upon advice purportedly given by Tom Goddard, an accountant who testified at trial, to establish reasonable cause and good faith. Petitioners argue that they “relied upon Mr. Goddard’s advice in reporting the captive insurance premiums paid as deductible business expenses” during the years at issue. Although Mr. Goddard was not a promoter, petitioners’ argument lacks merit.

Neither any written tax opinion nor other contemporaneous documentary evidence concerning any advice Mr. Goddard may have given to Mr. Candland or RMS is in the record. Mr. Goddard did not even testify that he provided any express advice to petitioners regarding the tax treatment of the captive insurance arrangement. Instead, he testified he would receive RMS’s books for the year, which included a captive insurance deduction, and that “the communication was . . . we weren’t objecting to their deduction in that year for the insurance captive. To me, it was an ordinary expense, ordinary insurance expense,

76

[*76] and I felt it was necessary and reasonable.” We specifically find that Mr. Goddard did not provide any advice to petitioners about the microcaptive arrangement. Mr. Goddard’s lack of objection to captive insurance deductions that petitioners had already been taking for years before they hired him does not itself constitute advice on which petitioners may rely in good faith. See Neonatology Assocs., P.A., 115 T.C. at 100 (“The mere fact that a certified public accountant has prepared a tax return does not mean that he or she has opined on any or all of the items reported therein.”); Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *52 (stating that taxpayers in microcaptive cases could not “rely on advice that was not given”); Flume v. Commissioner, T.C. Memo. 2020-80, at *37 (“Simply employing a tax return preparer for the years at issue does not permit [the taxpayers] to avoid accuracy-related penalties.”).

We also specifically find that Mr. Goddard did not review, nor did petitioners provide him with, some of the information that would have been necessary to form an opinion on the deductibility of the captive expenses. 66 Furthermore, the evidence of petitioners’ actual reliance, let alone good-faith reliance, on any judgment that Mr. Goddard may have reached is underwhelming. Cf. Avrahami, 149 T.C. at 207 (stating, as part of a finding of reasonable reliance on professional advice, that the adviser credibly testified that specific advice was given and that the taxpayer credibly testified that he proceeded with a microcaptive arrangement because of the adviser’s blessing).

Petitioners also argue that the issues are novel and complex and were essentially ones of first impression at the time their returns were filed and that they should be excused from accuracy-related penalties. While the issues were somewhat novel at the time, this does not excuse petitioners from penalties in the absence of any efforts on their part to ascertain their correct tax liabilities or apply well-settled principles of taxation to their situation. See Neonatology Assocs., P.A. v. Commissioner, 299 F.3d at 234–35 (stating that taxpayers could not avoid accuracy-related penalties, even though they were without direct precedent to guide them, because their case “does not involve novel questions of law but rather is concerned with the application of

66 Mr. Goddard did not review the actual policies issued by Risk Retention.

Neither did he review Risk Retention’s formation documents, engagement letter with Tribeca, business plan, or various reinsurance contracts and agreements. In addition, he did not review the Owners’ Manual, any of the general terms and conditions in force for the captive insurance arrangement, or Artex’s claims handling practices.

77

[*77] well-settled principles of taxation to determine whether certain expenditures made by close corporations are deductible as ordinary and necessary business expenses”). While we observed as part of our analysis of good faith in Avrahami that “[t]his is a case of first impression,” 149 T.C. at 207, we did so only after finding that the taxpayers actually and reasonably relied on advice from a competent professional, see id. at 206–07. These cases are more like Caylor in that petitioners did not actually “get advice or a professional’s judgment that they could have reasonably relied upon.” Caylor Land & Dev., Inc., T.C. Memo. 2021-30, at *53. Petitioners are liable for accuracy-related penalties across the board.

We have considered the parties’ other arguments and, to the extent they are not discussed herein, find them to be irrelevant, moot, or without merit.

To reflect the foregoing,

Decisions will be entered under Rule 155.

Footnotes

1. Docket Nos. 15066-18, 15067-18, Filed January 4, 2024. 15068-18. ————— Kacie N.C. Dillon, Jonathan A. Halmi, and Tim Alan Tarter, for petitioners. Andrea M. Faldermeyer, Estevan D. Fernandez, Marco Franco, John Robert Gordon, Erin Kathleen Salel, and Emerald G. Smith, for respondent. TABLE OF CONTENTS MEMORANDUM FINDINGS OF FACT AND OPINION ..................... 4 FINDINGS OF FACT .............................................................................. 6 I. RMS ................................................................................................... 6 A. Background ............................................................................... 6 B. Petitioners’ Roles at RMS ......................................................... 7 II. Commercial Insurance Coverage ..................................................... 8 1 Cases of the following petitioners are consolidated herewith: Cheryl L. Doss, Docket No. 15067-18; and Arthur D. Candland and Michelle M. Candland, Docket No. 15068-18. Served 01/04/24