Royalty Management Insurance Company, Ltd.

• Court: United States Tax Court
• Decided: September 16, 2024
• Docket: 3823-19
• Status: Unpublished

Opinion

United States Tax Court

T.C. Memo. 2024-87

ROYALTY MANAGEMENT INSURANCE COMPANY, LTD., Petitioner

v.

COMMISSIONER OF INTERNAL REVENUE, Respondent

JOHN B. SHEPERD AND ANDREA SHEPERD, Petitioners

__________

Docket Nos. 3823-19, 4421-19. Filed September 16, 2024.

H. Craig Pitts and Mark A. Weitz, for petitioners.

Ann L. Darnold, Lisa R. Jones, Alex R. Halverson, Alicia H. Eyler, and Vassiliki Economides Farrior, for respondent.

MEMORANDUM FINDINGS OF FACT AND OPINION

LAUBER, Judge: These consolidated cases involve a “microcap- tive insurance” arrangement. 1 The ultimate taxpayers are John

1 “A ‘captive insurance company’ is a corporation whose stock is owned by one

or a small number of companies and which handles all or a part of the insurance needs of its shareholders or their affiliates.” Caylor Land & Dev., Inc. v. Commissioner, T.C. Memo. 2021-30, 121 T.C.M. (CCH) 1205, 1207 n.4 (citing Harper Grp. v. Commissioner, 96 T.C. 45, 46 n.3 (1991), aff’d, 979 F.2d 1341 (9th Cir. 1992)). “A ‘microcaptive’ is a

Served 09/16/24 2

[*2] Sheperd and his wife, Andrea. Together they owned 100% of Shep- erd Royalty, LLC (Sheperd Royalty), an S corporation. Mr. Sheperd also owned 100% of Royalty Management Insurance Co., Ltd. (RMIC), a pu- tative insurance company. For the 2012 taxable year, Sheperd Royalty claimed a business expense deduction of $1,110,206 for “insurance pre- miums,” the bulk of which were routed (through an intermediary) to RMIC for allegedly reinsuring Sheperd Royalty’s risks. Of this total, $1,105,251 relates to the captive insurance arrangement at issue.

During 2012 section 831(b) allowed “small insurance companies” to receive tax-free up to $1.2 million of annual insurance premium in- come (while requiring that tax be paid on their investment income). 2 The principal questions presented are whether the arrangement at issue gave rise to “insurance” for Federal income tax purposes and whether RMIC was an “insurance company” within the meaning of section 831(b). The answers to those questions determine whether the amounts paid by Sheperd Royalty as alleged insurance premiums were deductible by it (and by the Sheperds, to whom the deductions were passed), and whether the amounts received by RMIC as alleged reinsurance premi- ums were exempt from tax under section 831(b).

Answering these questions in the negative, the Internal Revenue Service (IRS or respondent) for 2012 determined deficiencies of $346,389 and $362,802 against RMIC and the Sheperds, respectively. For the Sheperds, the notice determined a 40% accuracy-related penalty for a transaction lacking economic substance, see § 6662(b)(6), (i), and in the alternative a 20% penalty under other provisions of section 6662. For RMIC, the notice determined a 20% penalty only.

To date this Court has decided seven cases involving “microcap- tive insurance” arrangements. 3 All of these cases were decided in favor of the Commissioner. Petitioners fare no better here.

small captive insurance company,” i.e., one that “take[s] in less than $1.2 million in premiums.” Id. (citing Avrahami v. Commissioner, 149 T.C. 144, 179 (2017)). 2 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. We round most monetary amounts to the nearest dollar. 3 See Avrahami, 149 T.C. 144; Patel v. Commissioner, T.C. Memo. 2024-34;

Swift v. Commissioner, T.C. Memo. 2024-13; Keating v. Commissioner, T.C. Memo. 2024-2; Caylor Land & Dev., 121 T.C.M. (CCH) at 1205; Syzygy Ins. Co. v. 3

[*3] FINDINGS OF FACT

The following facts are derived from the pleadings, six Stipula- tions of Facts with attached Exhibits, documents admitted into evidence during trial, and the testimony of fact and expert witnesses. The Shep- erds resided in Oklahoma when their Petition was timely filed, and they have stipulated that venue for appeal of their case is the U.S. Court of Appeals for the Tenth Circuit. See § 7482(b)(1)(A). RMIC, an entity incorporated in the Sac and Fox Nation, has stipulated its agreement “to have its consolidated case heard in the Tenth Circuit as well.” See § 7482(b)(2).

I. Mr. Sheperd’s Background

Mr. Sheperd has lived in Oklahoma since age two. He worked briefly in the oil fields, then toggled between jobs as a car salesman and a commercial bank teller. In both capacities he got to know a lot of peo- ple in western Oklahoma, the focal point for oil and gas (O&G) explora- tion in that State. He had some familiarity early on with O&G leasing because he had inherited mineral rights from his parents.

In late 2011 Mr. Sheperd perceived a business opportunity cre- ated by the huge boom in hydraulic fractioning or “fracking.” This is a process by which water is injected at extremely high pressure into geo- logic formations, enabling recovery of oil and (especially) natural gas that was previously unrecoverable. This was a revolutionary develop- ment in the U.S. natural gas industry. With the availability of fracking, exploration and production (E&P) companies developed a ravenous ap- petite for mineral leases.

Mr. Sheperd had good contacts with farmers and ranchers whose properties were now prime candidates for mineral leasing. He learned that Cordillera Energy Partners (Cordillera) was seeking to acquire leases in the Anadarko Basin area. Mr. Sheperd’s plan was to contact potential mineral lessors, negotiate acquisition of leases from them, and assemble the leases into packages for assignment to Cordillera.

In Oklahoma, as in many States, ownership of land is commonly divided into surface rights and subsurface mineral rights. In a typical mineral lease, the owner of the mineral rights (who may or may not own

Commissioner, T.C. Memo. 2019-34, 117 T.C.M. (CCH) 1165; Rsrv. Mech. Corp. v. Com- missioner, T.C. Memo. 2018-86, 115 T.C.M. (CCH) 1475, aff’d, 34 F.4th 881 (10th Cir. 2022). 4

[*4] the surface rights) leases them to a lessee for a term of years (gen- erally 3 years for the leases involved here). The lessor surrenders to the lessee, for the period of the lease, the right to exploit the subsurface min- erals. In exchange for doing so the lessor receives a premium or “lease bonus.” The bonus is essentially the initial price paid to acquire the lease. If O&G drilling begins during the term of the lease, the lessor is also entitled to a percentage of the value of the production (typically 3/16). If no production occurs during the lease period, the lease termi- nates. The lessor then keeps his bonus, but he receives no royalties.

II. Sheperd Royalty

Mr. Sheperd incorporated Sheperd Royalty in 2011 as the vehicle for conducting his lease-acquisition business. Sheperd Royalty had no formal employees and conducted its operations out of the Sheperds’ home. It elected to be taxed as an S corporation, so that all items of income and expense passed through to Mr. Sheperd and his wife. (Apart from her ownership interest in Sheperd Royalty, Mrs. Sheperd had no involvement in the transactions at issue.)

On its 2012 Form 1120S, U.S. Income Tax Return for an S Corpo- ration, Sheperd Royalty stated that it offered “Landman” services. “Landman” is shorthand for “land manager” or “land management.” A landman serves as the public-facing side of an E&P venture, interacting with landowners and negotiating directly with them to acquire mineral leases. A landman can work in house for an E&P company or operate independently. In either case the landman’s duties are roughly the same: finding landowners with potentially valuable mineral rights, re- searching courthouse records to confirm ownership, preparing reports, negotiating mineral leases and related agreements, obtaining curative documents where necessary, and assembling leases for transfer to the E&P venture.

There was some disagreement at trial as to whether Mr. Sheperd was technically a “landman.” The testimony established that there is a hierarchy of “landmen,” including those who are “certified” or “regis- tered.” To rise to these higher levels, one needs a college degree and field experience. Mr. Sheperd initially had neither, so he was not a cer- tified or registered landman. But he and his associates performed the essential functions of landmen by negotiating for mineral leases and persuading lessors to sign the necessary paperwork to generate leases that could be assigned. 5

[*5] Mr. Sheperd made a profit by selling (assigning) leases to Cordil- lera for amounts in excess of the bonus payments he made to acquire the leases. He had three associates in his business, all of whom were con- tractors or subcontractors. They were not formal partners, but Mr. Sheperd paid them “commissions” representing a percentage of the net proceeds Sheperd Royalty received from lease assignments.

A lawyer named Ryan Cole discharged several back-office func- tions for Sheperd Royalty and also performed landman services. He had previously worked for Devon Energy, a well-known E&P company in Ok- lahoma City, and he was knowledgeable about mineral leases. Once Mr. Sheperd found a prospect, Mr. Cole would prepare the mineral lease, using a form contract that Cordillera supplied. Lease preparation in- volved inserting into the form contract information about the lessor, the acreage being leased, the lease price per acre, and the technical descrip- tion of the property. Mr. Cole took charge of Sheperd Royalty’s record- keeping, handled follow-up communications with lessors, and ensured that bonus payments were properly made.

Mr. Sheperd was the primary field operative. He researched pub- lic databases and courthouse records to find out who owned mineral in- terests in the area in which Cordillera was interested. He contacted the landowners, engaged them in conversation, and tried to negotiate leases of their mineral interests. He had two other “landmen” working with him, Brett Hudson and Christina Sullivan (who later married Mr. Cole). Mr. Sheperd seems to have assigned them work on a territorial basis, but all three performed similar functions.

One aspect of a landman’s job is to confirm that the lessor has good title to the mineral interest being conveyed. This task is called “running title.” Mr. Sheperd acknowledged at trial that he could “run title.” He and his colleagues had “access to the same databases as eve- rybody else,” and they did preliminary research at local courthouses to check the lessor’s title. But some leases presented atypical questions on which Mr. Sheperd lacked expertise.

In-depth title verification for Sheperd Royalty’s leases was per- formed by RK Pinson & Associates (Pinson), a highly regarded “petro- leum landman” firm with operations in Oklahoma, Texas, and Colorado. Pinson represented Cordillera, serving as the broker for leases proposed for assignment to it. Pinson employed numerous technicians specializ- ing in the intricacies of verifying title to mineral interests. Kent Pinson, the head of the firm, testified very credibly at trial. 6

[*6] Whenever Mr. Sheperd proposed a mineral lease for assignment to Cordillera, he presented it to Pinson, accompanied by his title re- search. Pinson’s technicians then did a deep dive into the lease, prepar- ing for each lease a Mineral Ownership Report and a Lease Assignment Purchase Report. Pinson’s team thoroughly checked (among other things) the validity of the lessor’s title to the mineral interest, whether the lessor’s stated acreage and percentage ownership interest were cor- rect, and whether the lessor’s identity was properly reported.

Pinson’s staff identified, and required correction of, several types of errors that appeared occasionally in the leases that Sheperd Royalty presented. These errors included misdescription of the lessor (e.g., im- proper identification of a trust), misdescription of the acreage owned by the lessor, and misidentification of the person required to sign the lease. A typical example of the latter problem arose when mineral interests were owned by minor children. In those situations Pinson’s technicians insisted that the parent be a court-appointed guardian of the minor chil- dren before the parent could sign the lease on their behalf.

Whenever Pinson’s team identified one of these problems, they insisted that Sheperd Royalty go back to the would-be lessor and get the problem fixed before Pinson would accept assignment of the lease. Mr. Pinson testified that there are two general kinds of title problems that can arise with a mineral lease: problems that can be fixed (or “cured”) and problems that cannot be fixed. The latter may result in a “title bust,” i.e., a situation where the lessor does not actually own the under- lying mineral interest.

The vast majority of title problems, however, can be fixed. Mr. Pinson testified that fewer than 0.5% of the title problems his firm dis- covered in mineral leases were incapable of being cured. No Sheperd Royalty lease that Pinson brokered to Cordillera ever produced a “title bust.” Indeed, as far as the trial evidence showed, Cordillera never en- countered a single title problem of any kind in any Sheperd Royalty lease that Pinson had vetted and accepted for assignment to Cordillera.

III. Lease Mechanics

Mr. Sheperd and his colleagues typically negotiated a bonus price with the potential lessor after doing their own title research. Mr. Cole drafted up a lease, and the lease documents were sent to Pinson for vet- ting. After any problems identified by Pinson were fixed, Pinson gave 7

[*7] the green light to consummate the lease. Mr. Sheperd paid the bo- nus to the lessor, and the lessor executed the lease.

Mr. Cole then prepared an “assignment of lease” document, again on Cordillera’s standard form. Sheperd Royalty executed that docu- ment, assigning to Cordillera the right to exploit the mineral resources for the 3-year term of the lease. Mr. Sheperd usually forwarded leases in a package consisting of leases on nearby properties. Pinson, acting as Cordillera’s agent, cut checks to Sheperd Royalty for the leases.

Unless Pinson identified a problem that caused a delay, the leases purchased by Sheperd Royalty were typically assigned to Cordillera within a week. Sheperd Royalty purchased and assigned approximately 511 leases in 2011 and 2012. The vast majority of these leases were recorded in the first half of 2012, with 17 being recorded in 2011 and only 26 being recorded during the second half of 2012. Almost all the leases were recorded in three counties (Beckham, Custer, or Washita).

Sheperd Royalty derived gross receipts of $24,812,500 from its lease-acquisition business during 2012, as shown by the Form 1099– MISC, Miscellaneous Income, issued to it by Pinson. The timing of these receipts is indicated by the deposits into its bank account, which were as follows:

Month Deposit Amount January $907,964 February 1,876,755 March 6,288,037 April 9,489,635 May 4,695,771 June 831,135 July 42,149 August 7,030 September 160 October 622,075 November 51,787 December -0- Total $24,812,500

As this table shows, Sheperd Royalty’s receipts tailed off sharply during the second half of 2012, with only $723,202 (or 3% of the total) 8

[*8] being received after June 30. That was because Cordillera during 2012 was in merger discussions with Apache Oil Co. (Apache), a much larger E&P firm. It was in Cordillera’s premerger interest to maximize the number of leases it held, so it was very eager to purchase leases during the first half of the year. After the merger was consummated in May 2012, Cordillera stopped acquiring leases, with Apache stepping into its shoes on all leases previously acquired. Pinson continued to bro- ker leases to Apache through Fall 2012 (apparently for lease transac- tions that were still in process when the merger occurred). But all such transactions appear to have ceased by the end of November.

IV. Risk Borne by Sheperd Royalty

In each lease assignment contract, Sheperd Royalty warranted good title to the mineral interest being conveyed and warranted that the lease was free of any outstanding mortgages, liens, or other encum- brances. If there was a significant title problem that Sheperd Royalty and Pinson both failed to catch, it was conceivable that Cordillera (or a subsequent assignee) could come back to Sheperd Royalty on its war- ranty and demand repayment of the amount Cordillera had paid for as- signment of the lease. This title warranty risk was the only meaningful risk that Sheperd Royalty bore under the lease assignment contracts.

At trial petitioners asserted that Sheperd Royalty faced risks other than title warranty risk, e.g., liability in damages for drilling too close to a structure, liability for spills or other surface damage caused by drilling, liability for miscalculation of mineral royalties or “improper deductions against the royalty share,” and liability for “shut in royalty payments.” The Court found no evidentiary or logical support for this assertion. Sheperd Royalty’s business consisted solely of acquiring and assigning leases; it never engaged in drilling or other E&P activity and never intended to do so. The only warranty it made in the lease assign- ment contract was a warranty of good title.

Sheperd Royalty could not possibly be liable for the other prob- lems petitioners listed because those events would occur (if ever) at a time when Sheperd Royalty no longer owned the lease. Sheperd Royalty assigned to Cordillera 100% of the mineral exploitation rights granted by each lease. If drilling by Cordillera or a subsequent assignee caused one of the problems petitioners mentioned, or if that E&P company mis- calculated the production royalty due to the lessor, that E&P company would bear liability for any consequent damages. 9

[*9] Because Sheperd Royalty’s risk under the lease assignment con- tracts was limited to title warranty risk, the Court finds as a fact that the risk it bore was quite small. First, in each mineral lease, the lessor himself warranted good title to Sheperd Royalty. If Cordillera ever made a claim against Sheperd Royalty, Sheperd Royalty would have an identical claim against the lessor.

Second, Pinson exhaustively researched the title for each mineral lease and demanded that any identifiable problem be fixed before it would accept assignment of the lease to Cordillera. Pinson was a highly reputable company on which Cordillera primarily relied for verification of title. Pinson’s team was meticulous in scrutinizing every lease for any possible problem, even insisting that parents get court-approved guard- ianship papers before signing a lease for their own children.

Cordillera placed its confidence in Pinson, which served as its bro- ker. The chances that Pinson would fail to spot and cure a title defi- ciency, leading to exposure for itself and Sheperd Royalty, were very small. Indeed, the record contains no evidence of a single title deficiency in any of the 500+ leases that Pinson vetted and approved for assign- ment to Cordillera. Neither Cordillera, Apache, nor anyone else ever made a claim against Mr. Sheperd or Sheperd Royalty on account of any lease assignment contract.

Finally, the evidence at trial suggested that E&P companies did not attribute significant value to title warranties issued by lease assign- ors like Sheperd Royalty. Mr. Pinson credibly testified that the amounts Cordillera paid for lease assignments did not vary much depending on whether the assignor warranted title. One would expect that, if the as- signor were assuming a significant risk in warranting title, he would be compensated for doing so in the form of a higher price for his lease. Con- versely, one would expect that an assignor who declined to warrant title would be paid a lower price. The apparent absence of a meaningful price differential is consistent with the conclusion that Sheperd Royalty did not assume a significant risk by warranting title.

V. Mr. Sheperd’s Alleged Concern About Risk

Mr. Sheperd testified that he was “worried about title errors or flaws” and feared that Pinson could have “messed up” when verifying title. He asserted that he “started losing sleep” because of this concern. We did not find this testimony credible. Although Mr. Sheperd main- tained standard homeowners and automobile insurance policies, 10

[*10] he evinced little interest in insuring against any of his business risks.

Sheperd Royalty maintained no insurance of any kind—not even a general business liability policy—during 2011. And it maintained no insurance of any kind during the first 11 months of 2012, by which time it had assigned $25 million worth of mineral leases to Cordillera or Apache. If Mr. Sheperd were genuinely concerned about risk on the leases being assigned, one might have expected him to secure insurance against that risk before year-end 2012, at which point Sheperd Royalty had stopped acquiring leases. 4

Petitioners contended at trial that there existed in 2012 no com- mercial insurance product that met Sheperd Royalty’s need to secure coverage for title warranty risk. The trial evidence did not provide a conclusive answer to that question. Mr. Sheperd allegedly asked his State Farm agency, which provided his homeowners policy, about insur- ing title to mineral leases, and he was apparently informed that State Farm did not offer such coverage. The American Association of Profes- sional Landmen (AAPL), the trade association for landmen, did offer an “errors and omissions” policy. But Mr. Sheperd insisted that he could not have purchased an AAPL policy because he was not a “landman.” Apart from his testimony, petitioners supplied no evidence on this point.

Evaluating all the evidence, the Court was not convinced that Mr. Sheperd during 2011 and 2012 made a robust search for commercially available insurance. We find that he pursued a microcaptive insurance arrangement at year-end 2012, not because he was genuinely concerned about title risk, but because he desired to reduce petitioners’ tax liabil- ity. Sheperd Royalty had made a lot of money: It had $24 million of gross receipts during the first 10 months of 2012, and Mr. Sheperd was looking at a very substantial tax liability. Seeking advice on how he might reduce his tax bill, he contacted people he knew from his previous work as a car salesman. They recommended that he call Cary Cope.

4 Sheperd Royalty continued to operate during 2013 and 2014, but it paid no

premiums for any sort of insurance coverage—provided by a captive insurer or other- wise—during those years. In 2013 Mr. Sheperd started another business called LDV Resources, LLC, which bought and sold mineral interests. That business likewise maintained no insurance coverage of any kind. All of this suggests that Mr. Sheperd was not overly concerned about risk. 11

[*11] VI. Creditors Captive Formation Co.

Cary Cope graduated from Oklahoma State University with a de- gree in finance. He did not attend graduate or law school and has no insurance-related degrees or endorsements. He has no formal tax edu- cation and is not a certified public accountant (CPA).

By 2010 Mr. Cope had become engaged in the design and market- ing of captive insurance arrangements, performing activities commonly described as being performed by a “promoter.” 5 He operated through an entity called Creditors Captive Formation Co. (CCFC), of which he was the sole shareholder. During 2012 CCFC had no employees apart from Mr. Cope’s 18-year-old stepson. Mr. Cope was assisted by three part- time contractors who rendered accounting, tax, and back-office services.

As its name suggests, CCFC marketed captive insurance prod- ucts, maintaining a website that touted the tax benefits of such arrange- ments. Through the website a customer could download an application form, fill it out, and submit it to CCFC. It would then create for that customer a captive entity purporting to be an “insurance company.” Most of CCFC’s initial customers were in the automobile business, in- cluding owners of car dealerships and automobile financing entities.

CCFC was incorporated in November 2010 under Oklahoma law. At no time during 2010–2012 was it organized, licensed, or regulated as an insurance company under Oklahoma (or any other) law. At trial Mr. Cope testified that CCFC “started getting in the insurance side” of its business during 2012 when it began “insuring bilateral contracts.” The meaning of this assertion was not entirely clear, and no documentary evidence was adduced at trial to support it.

In 2012 CCFC commenced litigation against the Texas Depart- ment of Insurance. On October 31, 2012, it filed a complaint in which it

5 “Promoter” is a loaded term in the tax world because of the penalty imposed

by section 6700(a) for “promoting abusive tax shelters.” In this Opinion we use the term “promoter” in its ordinary sense, making no determination as to whether Mr. Cope’s activities would subject him to the civil penalty under section 6700(a), a ques- tion that is not before us. 12

[*12] made the following representations to the U.S. District Court for the Western District of Texas:

• “Plaintiff CCFC is in the business of forming and managing cap- tive insurance companies throughout the United States.”

• “CCFC receives no consideration for insurance. It charges a fee to creditors to set up a captive insurance company. The captive receives any and all payments for insurance.”

• “CCFC forms captives . . . . CCFC has no part in the insurance transaction . . . . It does not insure risk, sell insurance, take com- missions or premium.”

Petitioners urge that these representations should be interpreted to mean that CCFC did not engage in the business of insurance in Texas, but its representations to the District Court were not so limited. Indeed, its complaint described the nature of its activities “throughout the United States.” There is no plausible evidence in the record that CCFC during 2012 was authorized to engage in the business of insurance any- where in the country. Rather, CCFC was solely a marketing vehicle through which Mr. Cope facilitated the creation and maintenance of other entities as purported insurance companies.

VII. CCFC Insurance Co.

At some point Mr. Cope decided that it would be useful to have his own “insurance company” for use in connection with arrangements in which his client’s captive entity was purporting to act as a reinsurer. To that end he created Creditors Captive Formation Company Insur- ance Co. (CCFC Insurance). Mr. Cope was the sole shareholder of CCFC Insurance.

CCFC Insurance was incorporated on January 3, 2013, in the Sac and Fox Nation, a tribal jurisdiction in Oklahoma. It received its Em- ployer Identification Number on January 31, 2013. It filed its first Fed- eral income tax return for the 2013 calendar year.

CCFC Insurance employed no underwriters, actuaries, or captive managers. It had no employees or other professionals knowledgeable about insurance. Its sole employee during 2013 (if it had any) was Mr. Cope’s 18-year-old stepson. 13

[*13] CCFC Insurance did not exist during calendar year 2012, the tax year at issue. Although it issued purported “insurance policies” to nu- merous “insureds” at various times, petitioners did not establish that it operated as a genuine insurance company during 2013 or at any time thereafter. CCFC Insurance was organized as an ordinary “domestic corporation” under the jurisdiction of the Sac and Fox Nation. The Sac and Fox Nation has no insurance regulatory authority, and there is no evidence that CCFC Insurance was licensed or regulated as an insur- ance company. It functioned as a middleman or facilitator, channeling cash from Mr. Cope’s clients to the purported “reinsurance companies” that he created for them. 6

In conjunction with creation of CCFC Insurance, Mr. Cope cre- ated, also in the Sac and Fox Nation, a new iteration of Creditors Cap- tive Formation Co. (CCFC2). It was incorporated in February 2013. The original CCFC apparently ceased operations near that time; it was ulti- mately dissolved by a certificate of dissolution filed in August 2014. By creating CCFC Insurance and CCFC2 and incorporating them both in the Sac and Fox Nation, Mr. Cope moved all of his microcaptive insur- ance operations outside the jurisdiction of the State of Oklahoma. 7

VIII. Creation of Mr. Sheperd’s Captive

Mr. Sheperd was referred to CCFC by acquaintances who owned car dealerships. Sheperd Royalty was CCFC’s first customer that was not in the automobile sales or automobile lending business. CCFC had never before worked with a company that engaged in O&G activity.

Mr. Sheperd did not produce any emails in response to repeated IRS requests for discovery. He admitted during a deposition that he

6 The record includes an “assignment agreement” dated December 21, 2012, by

which CCFC purports to assign to CCFC Insurance all insurance contracts that CCFC then held. This does not appear to be an authentic document. In any event, there is no evidence that CCFC engaged in the business of insurance at any time, and CCFC Insurance was not in existence during the 2012 calendar year. 7 Petitioners sought admission into evidence of Exhibits 1000-P through

1005-P (spreadsheets listing CCFC’s other clients and transactions during 2012) and Exhibits 1006-P through 1046-P (spreadsheets listing other clients and transactions of CCFC and/or CCFC Insurance in years after 2012). We reserved ruling on respond- ent’s objections to these documents. We will overrule his objections to Exhibits 1001-P through 1005-P and admit those documents into evidence. We will sustain his objec- tions to the other exhibits because they concern facts that postdate the tax year at issue. We will likewise sustain his objection to Exhibit 1000-P, a summary spread- sheet. 14

[*14] might have deleted some older emails, stating that he had a “bad habit” of getting rid of emails after a transaction was finished. However, emails produced by other participants shed some light on his initial com- munications with Mr. Cope. An email from Mr. Cope to Mr. Sheperd dated September 21, 2012, attached five documents that emphasized the tax benefits of CCFC’s arrangements, while saying nothing about title insurance coverage for mineral leases.

Included in those documents was a draft management agreement indicating that Mr. Cope intended to incorporate a captive for Mr. Shep- erd on the Caribbean island of Nevis. That captive, which was ulti- mately named Royalty Management Insurance Co., Ltd. (RMIC), was to act as a purported reinsurer of risks transferred to it. Mr. Cope eventu- ally chose not to organize RMIC in Nevis because captive insurance en- tities created there had become a focus of IRS attention.

In late September Mr. Cope emailed Mr. Sheperd, attaching ma- terials regarding state taxation of insurance companies and stating: “[T]his is why the transaction will actually occur outside of the bounda- ries of Oklahoma.” On September 25, 2012, Mr. Cope prepared articles of incorporation for RMIC in the Delaware Tribal Nation, another tribal jurisdiction in Oklahoma. Although Mr. Cope signed this document and asserted that it had been filed with the Delaware Tribal Nation, we did not find that testimony (or much of his other testimony) credible. The custodian of records for the Delaware Tribal Nation averred, in response to an IRS subpoena, that it had no record of ever having received such a document.

Mr. Cope ultimately incorporated RMIC in the Sac and Fox Na- tion on December 7, 2012. As noted above, the Sac and Fox Nation has no insurance regulatory authority, and RMIC was thus organized as an ordinary domestic corporation. Messrs. Cope and Sheperd signed a management agreement under which CCFC agreed to provide, for a fee equal to 6.9% of the “gross written premiums” it received, all accounting and administrative services needed to maintain RMIC’s existence.

Petitioners assert that RMIC was initially capitalized with a $5,350 cash contribution from Sheperd Royalty or Mr. Sheperd. No ev- idence supports that assertion. The trial established that the $5,350 payment—as explicitly stated on CCFC’s website—was a fee that a cus- tomer was required to pay when submitting an application for creation of a captive. CCFC charged that fee for “preparing and filing” the for- mation documents, and CCFC kept the fee. RMIC’s bank records 15

[*15] confirm that the $5,350 in question was never deposited in its bank account. RMIC’s initial capitalization was thus zero. 8

On November 30, 2012, a week before RMIC was incorporated, Mr. Sheperd opened a bank account in its name at Interbank (Interbank account). In the account agreement with Interbank, Mr. Sheperd stated that RMIC was incorporated in the Delaware Tribal Nation; in fact, RMIC had not yet been incorporated and it was never incorporated in the Delaware Tribal Nation. Mr. Sheperd stated that RMIC was en- gaged in the “oil and gas royalties” business, even though it was suppos- edly going to engage in the insurance business.

IX. The “Master Insurance Policy”

Mr. Cope created RMIC as a captive “reinsurer” that would sup- posedly reinsure Sheperd Royalty risks transferred to it from a primary insurer. Central factual disputes in these cases concern identification of the “primary insurer” and of the documents that governed that alleged insurance relationship.

During the IRS examination, which began in May 2014, the rev- enue agent (RA Currier) repeatedly asked for copies of the “master in- surance policy” pursuant to which Sheperd Royalty had paid roughly $1.1 million of putative insurance premiums. Petitioners asserted that they had no copies of any insurance documents in their files.

Eventually, following an inquiry to CCFC, petitioners produced, on May 20, 2015, an insurance policy that listed CCFC Insurance as the “insurer” and Sheperd Royalty as the “insured.” When furnishing this document to RA Currier, petitioners’ counsel stated: “The Master Policy between CCFC Insurance and Sheperd Royalty LLC is attached.” RA Currier credibly testified that this was a copy of the same master policy she had previously been shown at the office of Attorney Pitts, who

8 During the IRS examination petitioners told the revenue agent that RMIC

when organized had additional paid-in capital (or surplus) of $19,000. The revenue agent found no support for that assertion, and petitioners submitted no evidence at trial to substantiate it. RMIC’s 2012 tax return reports “paid-in capital” of zero. See infra p. 31. In their opening brief petitioners assert that RMIC “had almost $1,000,000 in surplus capital assets to pay claims.” But in so asserting petitioners are referring to the putative “reinsurance premiums” RMIC received from CCFC. Premium income does not constitute “initial capitalization.” 16

[*16] represented petitioners during the examination. We will refer to this document as the Master Insurance Policy or Policy. 9

The Master Insurance Policy states that it was negotiated and issued “in the Sac and Fox Nation,” where CCFC Insurance was eventu- ally incorporated. The Policy and its declaration pages, which specify the lines of coverage, all bear Mr. Cope’s signature as “president” of CCFC Insurance. But none of his signatures is dated.

The term of coverage (policy period) for all lines of coverage was the 2012 calendar year. The Policy was a “claims made” policy, and all claims needed to be submitted during the policy period or within 60 days thereafter, i.e., by March 2, 2013. The Policy could be canceled by CCFC Insurance at any time, for any reason. If CCFC Insurance canceled the Policy for a reason other than nonpayment of premiums, it was required to notify Sheperd Royalty 30 days before “the Effective Date of Cancel- lation.” In the event CCFC Insurance canceled the Policy, it was re- quired to send Sheperd Royalty “any premium refund due.”

The declaration pages of the Policy do not provide any insurance coverage specifically for title warranty risk. The four specified lines of coverage are: (1) Wrongful Acts/Errors & Omissions (E&O), (2) General Liability/Differences in Conditions (DIC), (3) Legal Expense Reimburse- ment, and (4) Product Representation/Service Rework (Warranty) Ex- pense Reimbursement. The declaration pages state for these lines of coverage the following limits of liability, deductibles, and premiums:

Coverage Policy Limit Deductible Premium Wrongful $2,000,000 $5,000/occurrence $52,963 Acts/E&O General Liabil- 2,000,000 25,000/occurrence 264,815 ity/DIC

9 In their opening brief petitioners objected to the admissibility of the Master

Insurance Policy. But in the First Supplemental Stipulation of Facts the parties stip- ulated this document as Exhibit 128-R, with petitioners reserving only a relevancy objection. At the outset of trial, the Court overruled both parties’ relevancy objections and admitted Exhibit 128-R (and other stipulated documents) into evidence. Petition- ers have not moved to be relieved of their stipulation. And they cannot reasonably dispute the document’s relevance: The document bears the signatures of Mr. Cope as president of CCFC Insurance, and petitioners’ counsel produced this document during the IRS examination and explicitly informed RA Currier that it was “[t]he Master Pol- icy between CCFC Insurance Company and Sheperd Royalty LLC.” 17

[*17] Legal Exp. Re- 500,000 5,000/occurrence 132,407 imb. Product Rep/Ser- 2,000,000 50,000/occurrence 622,315 vice Re- work Total $6,500,000 $1,072,500

Included within the Master Insurance Policy was an excerpt from a March 2013 “actuarial report” prepared by an actuary named Marn Rivelle, captioned “Sheperd Royalty LLC/Captive Insurance Program.” It shows the same total premium, $1,072,500, but five rather than four lines of coverage, with different policy limits and deductibles, as follows:

Coverage Policy Limit Deductible Premium E&O $2,000,000 $5,000/occurrence $52,963 General Liabil- 2,000,000 25,000/occurrence 264,815 ity/DIC Legal Exp. Re- 500,000 5,000/occurrence 132,407 imb. Product / Services 2,000,000 25,000/occurrence 92,685 Rework Reps & Warran- 2,000,000 25,000/occurrence 529,630 ties Total $8,500,000 $1,072,500

As a comparison of these tables shows, Mr. Rivelle’s premium numbers are the same as those shown in the Policy declaration pages for the first three lines of coverage. However, Mr. Rivelle split the fourth line into two separate lines of coverage, divided the premium between them, reduced the deductible from $50,000 to $25,000, and increased the overall policy limit from $6.5 million to $8.5 million.

The Master Insurance Policy had at least three odd features. First, the alleged insurer, CCFC Insurance, was not incorporated until January 3, 2013, and it therefore did not exist during the policy period. Second, the Policy included portions of an actuarial report prepared in March 2013; this raised a question as to whether the Policy had been created during 2012. Third, the policy limits and deductibles appearing in the declaration pages of the Policy do not match the policy limits and 18

[*18] deductibles appearing in the actuarial report. Notwithstanding this quandary, petitioners produced no other master insurance policy, either to RA Currier or to respondent’s counsel, during the ensuing seven years.

Respondent scheduled a deposition with Mr. Cope for October 4, 2022. He arrived at his deposition with paper copies of five, allegedly newly discovered, master insurance policies, all of which bear Mr. Cope’s and Mr. Sheperd’s signatures. Petitioners sought admission of these documents into evidence as Exhibits 810-P through 814-P. Respondent reserved authenticity objections to all five documents, and the Court de- ferred ruling on those objections until the conclusion of posttrial brief- ing.

Each policy names the insured party, not as Sheperd Royalty LLC, but as “Sheperd Royalty Management” (Management), an entity that does not exist. The first policy, dated October 1, 2012, lists CCFC as the insurer and Management as the insured. It states that the policy is enforceable only in the Modoc Nation—another tribal jurisdiction in Oklahoma—even though CCFC was an ordinary Oklahoma corporation. It contains a line of purported insurance coverage—for “Mineral Rights- Title Insurance Defects Reimbursement”—that does not appear in the Master Insurance Policy and did not appear on the invoices that CCFC issued to Sheperd Royalty for 2012.

The other four policies are dated December 21, 2012, January 1, 2013, March 9, 2013, and January 1, 2014. They all list CCFC Insurance as the insurer and Management as the insured. CCFC Insurance is shown as the insurer on the December 21, 2012, policy even though it was not incorporated until January 2013. The latter three policies pur- port to provide coverage through the end of calendar year 2014 or 2015, even though Sheperd Royalty paid no “insurance premiums” after 2012.

Neither petitioners nor Mr. Cope supplied any credible evidence to establish when these documents were actually created and signed. All were prepared on a computer. But when asked to produce computer metadata showing the dates on which the documents were created and accessed—often referred to as “timestamps”—petitioners replied that no electronic copies existed. According to Robert Ferguson, who worked in CCFC’s office, this was because of a computer ransomware attack that CCFC suffered during 2013 or 2014, which allegedly made all preexist- ing document files inaccessible. Mr. Cope testified that he found paper copies of the five policies in a client file for Sheperd Royalty, which 19

[*19] CCFC maintained in traditional paper form. He indicated that this file contained numerous other documents relating to Sheperd Roy- alty’s insurance arrangement, including correspondence, premium in- voices, and canceled checks.

For a variety of reasons, we find that petitioners have failed to carry their burden of proving the authenticity of the eleventh-hour “master policies,” and we will therefore exclude them from evidence. First, we find it implausible that five successive insurance policies, pur- portedly executed at various dates over a two-year period, could all fail to name the insured party correctly. Misidentifying the insured party in an insurance contract is not a trivial error. If the policies were actu- ally drafted and signed at the stated intervals, one would expect that someone—either Mr. Sheperd, Mr. Cope, or a CCFC staff person—would eventually have noticed the error. By contrast, if the five documents were created simultaneously years after the fact, the repeated error would be easier to understand. 10

Second, petitioners’ counsel was circumspect when examining Mr. Sheperd about these documents. Counsel showed him each alleged pol- icy and asked him to confirm that, even though Management was erro- neously listed as the insured, the address appearing on the title page was actually Sheperd Royalty’s address. Counsel asked Mr. Sheperd no questions about the circumstances surrounding execution of the docu- ments. In particular, counsel did not ask him to confirm that the signa- tures were actually his, or that he had affixed his signatures on the dates shown. Instead, he simply asked Mr. Sheperd to verify that notations on the upper right-hand corner of each alleged policy were his handwrit- ing.

Third, if these documents were genuine, we find it incredible that Mr. Cope did not find them until the eve of his deposition. The docu- ments were not misplaced; Mr. Cope testified that he found them in CCFC’s client file for Sheperd Royalty—the most obvious place to look. Petitioners had produced other documents taken from this same client file, such as deposit receipts for Sheperd Royalty’s “premium” payments, long before Mr. Cope was deposed. Mr. Cope had every incentive to make an exhaustive search for such policies if they existed: He knew

10 At trial Mr. Sheperd testified that he noticed this error and brought it to Mr.

Cope’s attention, but that CCFC never revised the policies to show the correct insured. Petitioners’ counsel elicited no testimony from Mr. Cope about this, and Mr. Cope did not confirm Mr. Sheperd’s testimony. Assuming that Mr. Sheperd’s recollection was correct, it does not enhance the overall plausibility of the story line. 20

[*20] that the IRS was examining the microcaptive structure he had cre- ated, and an adverse outcome would not be in his interest. It is hard to believe that the five “master policies” were sitting in that same client file but escaped notice for seven years.

X. Reinsurance Agreement

On a date not disclosed by the record, Messrs. Sheperd and Cope signed a purported “Reinsurance Agreement” between RMIC and CCFC Insurance. The cover page of the document shows the date December 6, 2012, but the lines next to the signatures, where dates were supposed to have been affixed by the signatories, were blank. CCFC Insurance, the purported “Ceding Company,” did not exist during the 2012 calendar year.

The Reinsurance Agreement states that CCFC Insurance will cede to RMIC the “net written premiums” paid by Sheperd Royalty. “Net written premiums” were defined as gross premiums paid by Sheperd Royalty, less fees retained by CCFC Insurance. In exchange RMIC agreed to assume 100% of the insured risks.

Article X(E) of the Reinsurance Agreement provided that, “[i]n the event Reinsurer [RMIC] becomes insolvent, the shareholders of such Reinsurer shall be liable for any amounts due to Ceding Company [CCFC Insurance] by Reinsurer, as a result of undercapitalization of Re- insurer.” Mr. Sheperd was RMIC’s sole shareholder. The Reinsurance Agreement thus made him personally liable for, and pro tanto relieved CCFC Insurance of liability for, any approved claims that RMIC was unable to pay. RMIC had no capital and no assets apart from the “net written premiums” that CCFC transferred to it. In substance, therefore, Sheperd Royalty’s “insurance coverage” was limited ab initio to $1,024,000—the dollar amount of “insurance premiums” it paid in 2012 minus the fees retained by CCFC. 11

11 At his deposition Mr. Cope produced, in addition to the five allegedly newly

discovered master policies, two allegedly newly discovered reinsurance agreements (Exhibits 822-P and 825-P). The first agreement, ostensibly dated September 25, 2012, was supposedly executed between RMIC and CCFC; the second agreement, ostensibly dated December 21, 2012, was supposedly executed between RMIC and CCFC Insur- ance. Both agreements, like the five “master policies” we have excluded from evidence, misname the insured party as “Sheperd Royalty Management.” There is no evidence that CCFC was authorized to engage in the insurance business. RMIC, not having been incorporated until December 7, 2012, did not exist on September 25, 2012. And 21

[*21] XI. Premiums Paid by Sheperd Royalty

CCFC’s website stated that its services would begin “with the in- itial feasibility study” for the proposed insurance arrangement. CCFC never performed a “feasibility study” for Sheperd Royalty. Nor did it secure an actuarial report explaining the premium structure until after Sheperd Royalty had already paid the “premiums.”

All “premiums” paid by Sheperd Royalty were remitted, not to CCFC Insurance (the supposed insurer under the Master Insurance Pol- icy), but to CCFC. Sheperd Royalty remitted cash to CCFC beginning in October 2012, with the last payment made on December 3, 2012. The payments were made by checks drawn on Sheperd Royalty’s bank ac- count, and the purpose of each payment was described in its account register. The amounts of these payments, the fees deducted by CCFC, and the “net written premiums” were as follows:

Net Writ- Date Check Description Amount CCFC Fees ten Premi- ums 10/25/2012 Initial Deposit $99,900 $6,893 $93,007 4th Qtr. Premium 232,750 11/20/2012 250,000 17,250 2011 1st Qtr. Premium 232,750 11/20/2012 250,000 17,250 2012 3rd Qtr. Premium 231,819 12/03/2012 249,000 17,181 2012 4th Qtr. Premium 233,681 12/03/2012 251,000 17,319 2012 Total $1,099,900 $75,893 $1,024,007

As shown in the table, CCFC’s fees were calculated as 6.9% of the “gross written premiums,” as provided in the management agreement between Messrs. Sheperd and Cope ($1,099,900 × .069 = $75,893). Rounding the “net written premiums” down to $1,024,000, CCFC remit- ted that sum to RMIC’s Interbank account as a supposed “reinsurance premium.” CCFC paid that amount in two installments, $558,500 in November 2012 and $465,500 the following month.

CCFC Insurance, not having been incorporated until January 3, 2013, did not exist on December 21, 2012. For these reasons and those discussed supra pp. 19–20, we find that Exhibits 822-P and 825-P are not authentic documents. We will accordingly sus- tain respondent’s authenticity objections to these documents and exclude them from evidence. 22

[*22] Petitioners have offered no plausible explanation as to how the “premiums” that Sheperd Royalty paid were determined or how those amounts corresponded either to the period of coverage or the lines of coverage allegedly secured. Sheperd Royalty made cash payments of $1,099,900 during 2012 and showed those payments on its books as se- curing coverage for nine months of 2012 and (incongruously) the fourth quarter of 2011. That total differs from the aggregate annual premium of $1,072,500 shown on the declaration pages of the Master Insurance Policy. The latter premium is identical to, but calculated differently from, the aggregate annual premium shown in the “actuarial report” prepared by Mr. Rivelle in March 2013. But as explained infra p. 24, $1,072,500 was the premium that Mr. Rivelle calculated for the 2013 calendar year, during which Sheperd Royalty paid no premiums for in- surance.

At trial Mr. Cope offered an entirely different explanation. He testified that he telephoned Mr. Sheperd in December 2012 and told him that the premium would be $650,000 per year for the first two years, i.e., that the Policy was providing coverage for 2012 and 2013. Mr. Cope admitted that, on this theory, Sheperd Royalty’s premium payments came up $200,000 short. Mr. Cope offered no coherent explanation as to how the alleged “premium” of $650,000 was determined, suggesting that it was somehow related to the severity and frequency of claims in the automobile industry.

A third explanation appears in an undated CCFC Insurance “Risk Transfer Report.” This document lists the risks that CCFC and/or CCFC Insurance allegedly assumed during 2012. It shows for Sheperd Royalty an annual premium of $536,250—exactly half the $1,072,500 annual premium appearing in the Master Insurance Policy and the March 2013 actuarial report.

The $536,250 premium shown in the Risk Transfer Report was calculated by multiplying the total acreage covered by the mineral leases that Sheperd Royalty assigned to Cordillera (roughly 21,000 acres) by $51.07, then dividing by two. At trial Mr. Cope offered no explanation as to why acreage would be a reasonable index for measuring title risk or how $51.07 was determined to be the correct multiplier. And he could not explain why this annual premium was half the annual premium shown in the Master Insurance Policy or how it correlated to the $1,099,900 gross premium that Sheperd Royalty paid at year-end 2012. 23

[*23] Evaluating all the evidence, the Court finds that Mr. Sheperd at year-end 2012 had no idea what the actual premium under the Master Insurance Policy was supposed to be. What he did know was that RMIC during 2012 could supposedly receive up to $1.2 million of premium in- come tax free. He therefore remitted $1,099,900 of cash to CCFC in round-dollar amounts, not wishing to push the envelope to the absolute limit.

XII. Actuarial Report

The only detailed explanation showing a calculation of premiums for Sheperd Royalty was prepared by Mr. Rivelle in his “actuarial re- port.” He testified as a fact witness and not as an expert in these cases.

Mr. Rivelle appears to have had his first communication with Messrs. Cope and Sheperd about his actuarial assignment on March 7, 2013. Exhibit 789-J is a March 7, 2013, email referencing a conference call among the three of them, and Mr. Rivelle made notes about the call directly on his electronic copy of that email. His notes reflect very basic, introductory information about the transaction—e.g., that Sheperd Roy- alty had purchased mineral leases and made title warranties. Later that day Mr. Cope emailed Mr. Rivelle and gave him the full corporate names of Sheperd Royalty and RMIC. This is the sort of information that one needs for “new client intake” purposes.

At trial Mr. Rivelle admitted that he was not aware, before March 2013, of any information about Sheperd Royalty. Mr. Cope asserted that he had spoken to Mr. Rivelle during 2012 and received a premium quote from him over the phone. Mr. Rivelle could not remember any such con- versation, and his notes and trial testimony show that the March 7, 2013, telephone call was his first exposure to the facts of these cases. The Court does not find credible Mr. Cope’s testimony to the contrary. 12

Mr. Rivelle’s “actuarial report” is dated March 9, 2013, two days after that phone call. Even before one considers the contents of his re- port, two days seems a very short time in which to perform an analysis of title warranty risk incident to the assignment of 500+ mineral leases. Although his report is lengthy, much of it consists of boilerplate material

12 Mr. Rivelle produced numerous emails and other documents in response to

IRS discovery requests. None of those documents indicates any communication with Mr. Sheperd or Mr. Cope before March 7, 2013. 24

[*24] downloaded from a template he regularly used, most of which had no relevance to Sheperd Royalty’s business.

Mr. Rivelle’s report, captioned “Actuarial Study of the Proposed Captive Insurance Program,” states that he was asked to calculate a premium for a proposed captive insurance transaction for “the upcoming policy period.” He apparently was not told that the captive insurance transaction had supposedly been consummated in 2012 and that the pu- tative premium had already been paid. Consistent with his assumption that he was addressing a proposed transaction, he calculated a pre- mium, not for the 2012 calendar year—the policy term stated in the Master Insurance Policy—but for the 2013 calendar year. For calendar year 2013 he calculated an annual premium of $1,072,500—$26,500 less than the premium Sheperd Royalty had already paid.

Mr. Rivelle’s report calculates the $1,072,500 premium based on five lines of insurance coverage. See supra p. 17. He testified that he recommended those forms of coverage in March 2013 as appropriate for Sheperd Royalty. But he had been informed that Mr. Sheperd’s alleged concern was title warranty risk on mineral leases. It is not obvious why an actuary, operating on a clean slate, would have recommended five distinct forms of coverage to address that species of risk.

Several of the recommended lines of coverage, moreover, seem in- apposite. One line of coverage was “general liability—difference in con- ditions.” Because Sheperd Royalty maintained no general liability in- surance policy of any kind, coverage for “difference in conditions” made little sense.

Another line of coverage was for “product/service rework.” The Policy stated that this line of coverage “will reimburse the insured to replace defective products or faulty rework service,” for example, “the manufacture of a product using defective materials.” This would include reimbursement to “a building contractor installing defective drywall, a painter applying defective paint, an assembler using defective parts, etc.” Mr. Rivelle offered no coherent explanation as to why this line of coverage would be appropriate for an insured seeking coverage for title warranty risk on mineral leases.

The lines of coverage addressed in Mr. Rivelle’s report are basi- cally the same as those set forth in the declaration pages of the Master Insurance Policy. See supra p. 17. Mr. Rivelle simply separated out “representations & warranties” as a distinct line of coverage, revised the 25

[*25] premiums and deductibles, and raised the overall policy limit to $8.5 million. In other words, it seems clear that Mr. Rivelle did not rec- ommend these lines of coverage on a clean slate in March 2013, but simply tweaked the lines of coverage he was given by Mr. Cope. 13

Mr. Rivelle admitted that, during the two days he had to prepare his report, he did not review any of Sheperd Royalty’s mineral leases or assignment contracts. Although Mr. Cope promised to send him “addi- tional information” to supplement their phone call, Mr. Cope sent no further information and Mr. Rivelle requested none. Mr. Rivelle admit- ted that he performed no research regarding mineral leases or title war- ranty risk, either with respect to Sheperd Royalty’s business or the O&G business generally.

Sheperd Royalty had been acquiring and assigning mineral leases since the fourth quarter of 2011. During that period, it had acquired and assigned more than 500 leases, all but 26 of which had been con- summated before July 1, 2012. See supra p. 7. As of March 2013, there- fore, Sheperd Royalty had something of a track record: Not a single title warranty claim—or any other type of claim respecting a mineral lease— had been filed against Sheperd Royalty during the 18 months of its ex- istence. Mr. Rivelle admitted that these facts played no role in his pre- mium calculation.

Employing assumptions about Sheperd Royalty’s annual reve- nues and “underlying loss rate,” Mr. Rivelle computed an expected an- nual loss of $1,012,500 under what he called the “realistically possible” worst case scenario. To that he added $60,000 to account for “operating expenses” that a captive insurance company like RMIC might incur “in any given year.” It is curious that the result, $1,072,500, is exactly the same as the aggregate premium shown in the declaration pages for the Master Insurance Policy, but for five lines of coverage rather than four. And his result is exactly double the $536,250 premium that CCFC In- surance’s Risk Transfer Report calculated using an entirely different, acreage-based, method. See supra pp. 22–23.

13 Mr. Rivelle’s actuarial report supplies additional evidence that Exhibit

810-P—the alleged master insurance policy bearing the date October 1, 2012—is not an authentic document. See supra pp. 18–19. That document includes, as a principal line of coverage, “Mineral Interest Title Defect.” If that line of coverage had been in effect during 2012, it is inconceivable that Mr. Rivelle would not have addressed it in an actuarial report prepared in March 2013. 26

[*26] XIII. Circular Flow of Funds

Although Mr. Sheperd showed little curiosity about the details of his insurance coverage, he was very interested in how he might extract cash from RMIC. In a November 27, 2012, email, Mr. Cope explained various options Mr. Sheperd could use to “get money out of the insurance company.” One option was to “have the insurance company issue loans to other parties, affiliates etc.” Mr. Cope explained that CCFC would “generate all loan docs, lines of credit etc. for you to utilize.” CCFC typ- ically recommended that the captive entity distribute cash through a line of credit (LOC), which would enable frequent disbursements with- out the need to create multiple iterations of loan documents.

Mr. Sheperd lost little time implementing Mr. Cope’s suggestion. Beginning in April 2013, Mr. Sheperd began withdrawing cash from RMIC’s bank accounts and depositing the money into bank accounts ti- tled to himself, his wife, Sheperd Royalty, and other entities he con- trolled. He thus ignored Mr. Cope’s advice that loans should not be made “directly to insiders of the insurance company such as [its] officers, directors and stockholders.” Mr. Sheperd was the president and sole shareholder of RMIC.

On April 1, 2013, RMIC extended a $250,000 LOC to Sheperd Royalty, evidenced by a promissory note in that amount. Two months later, on May 30, 2013, RMIC increased the LOC to $750,000. Between April 2013 and May 14, 2014, Sheperd Royalty extracted at least $608,000 in cash from RMIC. Those cash withdrawals were made in 60 or more transactions ranging in size from $1,000 to $100,000.

On April 1, 2013, RMIC extended a $250,000 LOC to La Dolce Vita Farms (LDV Farms), evidenced by a promissory note in that amount. LDV Farms was a horse- and dog-breeding operation, con- ducted by Mr. Sheperd as a sole proprietorship, which had never turned a profit. Two months later, on May 30, 2013, RMIC increased that LOC to $750,000, thus raising to $1,500,000 RMIC’s potential exposure under the two LOCs. RMIC’s potential exposure thus exceeded the value of its assets (consisting solely of “reinsurance premiums” and investment in- come) by roughly $500,000.

The Sheperd Royalty LOC specified that all funds advanced by RMIC were to be used for Sheperd Royalty’s “operating expenses.” That restriction was repeatedly ignored. Using a checkbook linked to the LOC, Mr. Sheperd wrote dozens of checks for personal expenses. These 27

[*27] included payments for private school tuition, a $10,000 diamond ring, a prom dress, wedding presents, birthday gifts, purchase of a jet ski boat, “shopping money” for his children, and transfers into his wife’s investment accounts. Several large checks were written to LDV Re- sources, another entity owned by Mr. Sheperd, and to Big Chief Re- sources, an entity owned by his son. RMIC’s bank records show aggre- gate advances of $847,600 under the two LOCs during 2013–2015.

Besides advancing cash, RMIC made investments that benefited Mr. Sheperd personally. In August 2013 RMIC purchased, for $240,000, a 40-acre parcel located close to petitioners’ residence. That parcel was separated from their residence by another 15-acre tract. When the mid- dle tract became available for sale, Mr. Sheperd purchased it, put RMIC on the deed, and used the 40-acre parcel as collateral.

Mr. Sheperd and his affiliated entities made seven repayments on the LOCs, totaling $151,600, before May 21, 2014. That was the date on which the IRS notified petitioners that it had commenced an exami- nation of their returns. Mr. Sheperd began making more substantial repayments after that date.

XIV. Claims Made Under the Master Insurance Policy

Sheperd Royalty submitted no insurance claims to CCFC during 2012 or 2013. After being notified in May 2014 that the IRS had com- menced an examination, Mr. Sheperd asked Mr. Cole to look for any leases that could be regarded as problematic. Mr. Cole replied in an email dated June 25, 2014, captioned “final payments.” This email listed mineral leases that Sheperd Royalty had acquired, but for which it had not yet remitted payment.

None of these open items involved a title irregularity or similar problem. Rather, these items remained open because Sheperd Royalty had not yet paid the lessors for the leases it had purchased from them. Later in 2014, on a date not disclosed by the record, Mr. Sheperd sub- mitted claims to CCFC referencing the eight items listed in Mr. Cole’s email, as follows: 28

[*28] Subject Alleged Basis for Claim Amount of Claim Kenneth Bennett “Bird dog fee” $57,050 Bethany Bennett “Title bust” 1,600 Zachary Bennett “Title bust” 1,600 Matthew Bennett “Title bust” 1,600 Kevin Bennett “Title bust” 2,400 Susan Olson “Title bust” 5,333 Catherine Finley “Title bust” 5,333 Mary Mileta “Title bust” 40,000 Total $114,917

Mr. Sheperd submitted each claim on a form created by CCFC. On most of these forms a CCFC staff person typed in the following as the basis for the claim: “Title Bust – omission of ownership details/time- liness resulting in title error and warranty transfer not consummated after monies paid.” Below the typewritten description Mr. Sheperd added a handwritten comment, e.g., “Additional paperwork was not filed at the courthouse.” None of the claims is dated, and none specifies the Master Insurance Policy line of coverage under which the claim was be- ing made. Several claims were submitted without any supporting docu- ments, such as copies of the underlying oil and mineral interest leases.

Although the Reinsurance Agreement stated that CCFC Insur- ance would “investigate and decide claims,” CCFC approved each of the eight claims without doing any investigation and without determining whether an insurable loss had actually occurred. Mr. Cope testified that CCFC simply “took Sheperd’s word for it.” Having approved each claim for payment, CCFC did not remit a check to Sheperd Royalty under the Master Insurance Policy, nor did it submit a claim to RMIC under the Reinsurance Agreement. Rather, CCFC applied the claim amount against the LOC that Sheperd Royalty owed RMIC, reducing the bal- ance that the former owed the latter.

The events giving rise to the purported claims occurred in 2012. But all eight claims were submitted to CCFC in late 2014, more than 18 months after the March 2, 2013, deadline for tendering claims under the Master Insurance Policy. Four claims were for amounts less than $5,000; they were thus below the deductible specified in the Policy for all lines of coverage. Two claims were for amounts greater than $5,000 but less than $25,000; they were thus below the deductible specified in Mr. Rivelle’s actuarial report for “representations & warranties” cover- age. CCFC approved all of the claims nonetheless. 29

[*29] Besides displaying these irregularities, the eight claims were de- ficient because none of them involved an insurable risk or an insurance loss under any line of coverage specified in the Master Insurance Policy:

• A “bird dog fee” is essentially a finder’s fee relating to a mineral lease. Kenneth Bennett was a well-known and well-respected fig- ure in a part of Oklahoma with which Mr. Sheperd and his land- men were less familiar. Sheperd Royalty executed a contract with Mr. Bennett, agreeing to pay him a fee calculated with reference to the value of the leases he “brought to the table.” Mr. Bennett’s fee, thus calculated, was $57,050 (leases comprising 2,282 acres at $25 per acre). Sheperd Royalty paid that fee, with Messrs. Hudson and Cole eventually reimbursing Mr. Sheperd for their shares of the amount paid. In paying the “bird dog fee,” Sheperd Royalty was discharging a bona fide contractual obligation. Do- ing so does not give rise to an insurance loss.

• Bethany, Zachary, and Matthew Bennett, all grandchildren of Kenneth Bennett, were minors during 2012. Each child held a fractional share of the family’s mineral interests. Sheperd Roy- alty acquired leases from all three children in mid-2012. Petition- ers contend that Sheperd Royalty suffered a “loss” in the amount of the bonus ($1,600) it paid to acquire each lease.

Sheperd Royalty prepared leases covering the children’s inter- ests, and their father, Kevin Bennett signed the leases. During the title vetting process Pinson’s team insisted that Kevin get ap- pointed as the children’s guardian before he could sign on their behalf. Kevin went promptly to the local courthouse, secured guardianship papers, and properly executed the leases. Sheperd Royalty thereby acquired good title to the mineral interests, so it had no insurance loss.

Petitioners contend that the brief delay occasioned by the need to secure guardianship papers prevented Sheperd Royalty from completing the transactions before Cordillera stopped accepting lease assignments. But that would not give rise to an insurance loss either. Sheperd Royalty had good title to the mineral inter- ests, and it was free to keep the leases or assign them to someone else. If Sheperd Royalty’s delay in consummating the assign- ments were thought to generate a loss, it would be an ordinary 30

[*30] business loss, not a “title bust” qualifying for payment under the Master Insurance Policy.

• Susan Olson (now Susan Birdwell) and Catherine Finley, who are sisters, executed substantially identical mineral leases. Sheperd Royalty paid each a lease bonus of $5,333. Petitioners contend that Sheperd Royalty suffered a “loss” in the amount of the bonus paid to acquire each lease.

Ms. Birdwell testified at trial. She indicated that the lease ini- tially prepared by Sheperd Royalty contained an error, misde- scribing the acreage she owned. Pinson’s team spotted this error and directed Sheperd Royalty to correct the lease. Sheperd Roy- alty did so, and both leases were properly executed. But by the time the leases were perfected, Cordillera had stopped acquiring leases. The same was true for the leases executed by Kevin Ben- nett and Mary Mileta: Their leases were perfected, Sheperd Roy- alty acquired good title to the mineral interests, but Cordillera was no longer accepting assignments.

Petitioners contend that Sheperd Royalty suffered a “loss” in the amount of the bonuses ($53,066 in toto) it paid to acquire these four leases. But Sheperd Royalty acquired good title to the min- eral interests, and it was free to keep the leases or assign them to someone else. There is no evidence that it suffered any economic loss. And if its delay in consummating the assignments were thought to generate a loss, it would be an ordinary business loss, not a “title bust” qualifying for payment under the Master Insur- ance Policy.

In short, contrary to the assertions in each claim form, there was no “title bust” or title irregularity in any of these transactions. The claims that Sheperd Royalty submitted to CCFC had no legitimate bases but were part of a charade concocted to create the appearance of “insur- ance.” Alerted that the IRS had begun an examination of the arrange- ment in May 2014, Mr. Sheperd and his advisers engaged in an after- the-fact effort to generate evidence they believed might be helpful to pe- titioners’ position.

XV. Tax Returns and IRS Examination

Sheperd Royalty filed a return on Form 1120S for 2012. It re- ported gross receipts of $24,686,900 and net business income of $3,510,211. In computing its net income, it claimed a deduction of 31

[*31] $1,110,206 for “insurance” expenses. That sum consisted of $4,956 for the Sheperds’ personal insurance expenses and $1,105,250 for “in- surance premiums” allegedly paid to CCFC. 14 Because Sheperd Royalty was a pass-through entity, the Sheperds reported their distributive shares of its income and deductions on Schedule E, Supplemental In- come and Loss, included in their 2012 joint return.

RMIC filed a return on Form 1120–PC, U.S. Property and Casu- alty Insurance Company Income Tax Return, for 2012. The return showed RMIC’s address as c/o “Creditors Captive Formation Company,” and it was prepared by Renee Woodward, a contractor for CCFC. Ms. Woodward, who is not a CPA, prepared returns for many of the captive entities that CCFC created.

On this return RMIC reported $252 of taxable investment income and indicated that $1,024,008 of “premium insurance income” had been excluded under section 831(b). On Schedule L, Balance Sheets per Books, it showed, as its only year-end asset, cash of $1,024,260, repre- senting the sum of the two figures in the previous sentence. Among its liabilities it listed “common stock” of $5,350 and “paid-in capital” of zero. In fact, the entry for “common stock” should likewise have been zero. As explained supra p. 14, the $5,350 was the “formation fee” that Mr. Shep- erd paid when submitting his application to have RMIC created. That formation fee was paid to CCFC, which kept the fee.

The IRS selected the returns filed by the Sheperds and Sheperd Royalty for examination. It assigned the examination to RA Currier, whose immediate supervisor at all relevant times was Jean M. Thomas. The examination was later expanded to include RMIC.

RA Currier prepared an examination report setting forth her pro- posed adjustments and penalty recommendations. For the Sheperds she recommended a 40% accuracy-related penalty under section 6662(b)(6) and (i), and in the alternative a 20% accuracy-related penalty under sec- tion 6662(a). For RMIC she recommended a 20% accuracy-related pen- alty under section 6662(a).

On January 19, 2017, the IRS sent petitioners Letters 950 (com- monly called “30-day letters”) setting forth these proposed adjustments and penalty recommendations. The letters were signed by Ms. Thomas, RA Currier’s immediate supervisor. RA Currier subsequently prepared

14 The latter number appears to represent the sum of $1,099,900 (the alleged

“premiums”) and $5,350 (the formation fee paid to CCFC for setting up RMIC). 32

[*32] Civil Penalty Approval Forms recommending assertion, against the Sheperds and RMIC, of the penalties set forth in the 30-day letters. Both Forms were signed by Ms. Thomas on August 30, 2017.

On November 20, 2018, the IRS issued the Sheperds a timely No- tice of Deficiency for 2012, determining a deficiency of $362,802, a 40% accuracy-related penalty under section 6662(b)(6) and (i), and (in the al- ternative) a 20% accuracy-related penalty under section 6662(a). The notice determined that the “premiums” paid by Sheperd Royalty “were not paid to an insurance company and . . . were not paid for insurance,” that the purported insurance transactions “lacked economic substance,” and that the transactions were engaged in “for no purpose other than to avoid or evade tax.” The notice adjusted upward the Sheperds’ pass- through income from Sheperd Royalty to reflect disallowance of the de- duction for purported insurance premiums. 15

On November 20, 2018, the IRS issued RMIC a timely Notice of Deficiency for 2012. This notice adjusted RMIC’s income upward by $1,018,650, determined a deficiency of $346,389, and determined a 20% accuracy-related penalty under section 6662(a). The notice determined that RMIC was not an “insurance company” for purposes of section 831, that the transactions generating its alleged premium income “are not insurance transactions,” and that RMIC was ineligible to exclude those amounts from its gross income under section 831(b).

XVI. Trial

A. Petitioners’ Expert

Petitioners offered expert testimony from Brett Sanger, an Okla- homa CPA and attorney who has litigated several cases involving min- eral royalties and O&G leases. We recognized him as an expert in O&G mineral leases in Oklahoma, but with no insurance expertise of any kind. We struck five paragraphs from his opening and rebuttal reports as improperly invading the province of the Court and opining on insur- ance matters outside the scope of his expertise.

15 Although the disallowed deduction for insurance premiums was $1,110,206,

the upward adjustment to the Sheperds’ pass-through income was only $991,575. The difference reflects other adjustments to Sheperd Royalty’s income, most of which were favorable to petitioners. Those other adjustments have been resolved by the parties and are not at issue here. 33

[*33] In his opening report Mr. Sanger offered background information about O&G leasing and the duties performed by a “landman/broker.” He opined that RMIC could face “as many as 1,600 possible and separate ‘risk exposure’ incidents” in its capacity as reinsurer for Sheperd Roy- alty. These risks allegedly included damages from “drilling too close to a structure,” “claims regarding improper surface usage or pollution,” claims regarding “mineral title defect,” and claims for miscalculation of production royalties or “improper deductions from the royalty share.” In his rebuttal report Mr. Sanger critiqued the report submitted by re- spondent’s expert.

B. Respondent’s Expert

Respondent offered expert testimony from Mark Crawshaw. He holds a Ph.D. in mathematics and has 38 years of industry experience as an actuary, including work with state insurance regulators, commer- cial insurers, captive insurers, and reinsurers. He is a member of the American Academy of Actuaries and a fellow of the Casualty Actuarial Society. We recognized him as an expert in actuarial science and found his testimony credible.

In his opening report Dr. Crawshaw identified numerous features of Sheperd Royalty’s insurance arrangement that he had rarely, if ever, observed in traditional insurance relationships. He characterized RMIC as providing a “savings account rather than an insurance arrangement.” He opined that Mr. Rivelle’s premium calculations were inconsistent with sound actuarial principles and had in effect been “reverse engi- neered” to match the premium numbers Mr. Cope had already placed in the Master Insurance Policy. In his rebuttal report Dr. Crawshaw of- fered a critique of Mr. Sanger’s “risk exposure” analysis.

OPINION

I. Burden of Proof

The Commissioner’s determinations in a notice of deficiency are generally presumed correct, and the taxpayer bears the burden of prov- ing them erroneous. Rule 142(a); Welch v. Helvering, 290 U.S. 111, 115 (1933). The taxpayer bears the burden of proving entitlement to any deduction claimed. INDOPCO, Inc. v. Commissioner, 503 U.S. 79, 84 (1992). The taxpayer must establish that the deduction in question is provided for by statute and must maintain records sufficient to enable the Commissioner to determine the correct tax liability. See § 6001; 34

[*34] 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).

Section 7491 provides that the burden of proof on a factual issue may shift to the Commissioner if the taxpayer satisfies specified condi- tions. Among these conditions are that the taxpayer must have “intro- duce[d] credible evidence with respect to [that] factual issue,” § 7491(a)(1), and must have “complied with the requirements under this title to substantiate any item,” § 7491(a)(2)(A). Petitioners have not sat- isfied these conditions with respect to any factual issue that has salience in deciding the questions presented. The burden of proof thus remains on them.

II. Insurance

A. General Principles

Section 162(a) allows the deduction of “all the ordinary and nec- essary expenses paid or incurred during the taxable year in carrying on any trade or business.” Insurance premiums incurred by a business are generally deductible. See Treas. Reg. § 1.162-1(a) (recognizing deducti- bility of “insurance premiums against . . . losses”).

Insurance companies are generally taxed on taxable income, in- cluding premium and investment income, in the same manner as other corporations. See §§ 11, 831(a); see also Syzygy, 117 T.C.M. (CCH) at 1172. Section 831(b), however, provides an alternative taxation regime for certain small insurance companies, commonly called “microcaptive” insurers. See Avrahami, 149 T.C. at 175–76; Syzygy, 117 T.C.M. (CCH) at 1172. During 2012, the tax year at issue, an insurance company with annual written premiums of $1.2 million or less was subject to tax only on its investment income (and not its premium income), provided that it made a valid election under section 831(b). 16 See § 831(b)(1) and (2).

To make a valid section 831(b) election, a captive entity must be an “insurance company.” See § 831(c); Avrahami, 149 T.C. at 180; Syz- ygy, 117 T.C.M. (CCH) at 1172 (“An inherent requirement for a company to make a valid section 831(b) election is that it must transact in

16 For tax years after 2016 Congress raised the premium ceiling to $2.2 million

and added certain diversification requirements as a condition of making a valid section 831(b) election. See Consolidated Appropriations Act, 2016, Pub. L. No. 114-113, div. Q, § 333(b)(1), 129 Stat. 2242, 3108 (2015). These changes have no bearing on the 2012 tax year at issue. 35

[*35] insurance.”). The deductibility of insurance premiums in turn de- pends on whether the payments “were truly payments for insurance.” Syzygy, 117 T.C.M. (CCH) at 1172. Accordingly, the central questions presented here—whether RMIC was entitled to exclude the putative premiums from its income, and whether Sheperd Royalty was entitled to deduct those payments from its income—both hinge on whether the arrangements at issue met the definition of insurance.

Neither the Code nor the Treasury Regulations define “insur- ance.” See R.V.I. Guar. Co., Ltd. & Subs. v. Commissioner, 145 T.C. 209, 224 (2015); Securitas Holdings, Inc. & Subs. v. Commissioner, T.C. Memo. 2014-225, 108 T.C.M. (CCH) 490, 494. But the categorization has profound effects: “[W]hile insurance is deductible, amounts set aside in a loss reserve as a form of self-insurance are not.” Caylor Land & Dev., 121 T.C.M. (CCH) at 1213; see Harper Grp., 96 T.C. at 46. When the alleged insurer and insured are related—as will commonly be true in the case of microcaptive insurers—the line between insurance and self-insurance blurs. See Avrahami, 149 T.C. at 176–77; Caylor Land & Dev., 121 T.C.M. (CCH) at 1213.

Given the lack of a statutory definition, the meaning of insurance “has thus been developed chiefly through a process of common-law ad- judication.” R.V.I., 145 T.C. at 224–25; see Caylor Land & Dev., 121 T.C.M. (CCH) at 1213. The Supreme Court explained long ago that “[h]istorically and commonly insurance involves risk-shifting and risk- distributing.” Helvering v. Le Gierse, 312 U.S. 531, 539 (1941); see also Rsrv. Mech. Corp. v. Commissioner, 34 F.4th at 903 (“At the core of the notion of insurance . . . are risk transfer and distribution.”). Building on this foundation, this Court and other courts have looked “to four nonex- clusive but rarely supplemented criteria: [1] risk-shifting; [2] risk-distri- bution; [3] insurance risk; and [4] whether an arrangement looks like commonly accepted notions of insurance.” Caylor Land & Dev., 121 T.C.M. (CCH) at 1213; accord Rsrv. Mech. Corp. v. Commissioner, 34 F.4th at 904 (citing Rsrv. Mech. Corp., 115 T.C.M. (CCH) at 1483); Av- rahami, 149 T.C. at 181; Rent-A-Center, Inc. v. Commissioner, 142 T.C. 1, 13 (2014).

We will direct our attention to the first, second, and fourth crite- ria listed above, assuming arguendo that Sheperd Royalty’s title war- ranty risk was an “insurance risk.” Before getting to that level of detail, however, we note some “big picture” problems with the arrangements at issue. These problems by themselves are fatal to petitioners’ position. 36

[*36] The Master Insurance Policy pursuant to which Sheperd Royalty paid $1,099,900 of “premiums” during 2012 ran between it and CCFC Insurance. But CCFC Insurance was not incorporated until January 3, 2013. And it was incorporated in 2013 as an ordinary domestic corpora- tion in the Sac and Fox Nation, a tribal jurisdiction that had no law governing insurance companies and no insurance regulatory authority. There is no credible evidence that CCFC Insurance was organized, op- erated, or regulated as an insurance company during the taxable year at issue.

The $1,099,900 in question was paid, not to CCFC Insurance, but to CCFC. But CCFC was not an insurance company either. As it repre- sented to the U.S. District Court for the Western District of Texas, “CCFC form[ed] captives.” See supra pp. 11–12. It had “no part in the insurance transaction” and “d[id] not insure risk, sell insurance, take commissions or premium.” See supra pp. 11–12. Because CCFC Insur- ance did not exist during 2012, and because neither CCFC nor CCFC Insurance was organized or regulated as an insurance company during 2012, the $1,099,900 that Sheperd Royalty paid cannot possibly have constituted “insurance premiums” deductible under section 162.

RMIC, the putative reinsurer, was incorporated in December 2012 as an ordinary domestic corporation in the Sac and Fox Nation. As noted above, the Sac and Fox Nation has no insurance regulatory au- thority. There is no credible evidence that RMIC was organized, oper- ated, or regulated as an “insurance company” during 2012. Because it was not an “insurance company,” it was not eligible to make the election set forth in section 831(b)(2). See § 831(c).

The “Reinsurance Agreement” under which RMIC received al- leged “reinsurance premiums” ran between it and CCFC Insurance. As noted above, CCFC Insurance did not exist during 2012, and it was not regulated as an insurance company after being incorporated as an ordi- nary corporation in 2013. CCFC did exist during 2012, but it was not organized, operated, or regulated as an insurance company during that year. For these reasons, neither CCFC Insurance nor CCFC was capa- ble during 2012 of ceding “insurance premiums” to RMIC that would qualify for tax exemption under section 831(b). 17

17 Recognizing the problem posed by the fact that CCFC Insurance did not exist

during 2012, petitioners assert that “CCFC prepared and executed a binder for insur- ance coverage” during 2012 and then assigned the insurance contracts to CCFC 37

[*37] These reasons suffice by themselves to support our conclusions that the $1,099,900 paid by Sheperd Royalty did not constitute “insur- ance premiums” deductible under section 162 and that RMIC was not an “insurance company” that received “written premiums” qualifying for exemption under section 831(b)(2)(A). For purposes of completeness, however, we will analyze the transactions in question under the criteria we have employed in our prior cases.

B. Risk Shifting

Whether a set of transactions gives rise to “insurance” must be examined from the perspective of both the insurer and the insured. Har- per Grp., 96 T.C. at 57. From the insured’s perspective, insurance is a risk transfer device, that is, a mechanism by which the insured obtains protection from financial loss by paying the insurer a premium. Ibid.; accord Black Hills Corp. v. Commissioner, 101 T.C. 173, 182–83 (1993), aff’d, 73 F.3d 799 (8th Cir. 1996). “By paying a premium, the insured externalizes his risk of loss by shifting that risk to the insurer.” R.V.I., 145 T.C. at 225. For true risk shifting to occur, the insurer must be “a well-capitalized company fully capable of paying claims and absorbing the risks transferred to it.” Id. at 225–26 (citing Harper Grp., 96 T.C. at 59 (finding risk transfer where the insurer “not only was financially ca- pable of satisfying claims made against it, but in fact paid such claims”)).

Under the Reinsurance Agreement, 100% of the risk of loss was supposedly shifted to RMIC, the putative reinsurer. But RMIC had in- itial capitalization of zero and no paid-in capital. See supra pp. 14–15, 31. Its only assets consisted of the “reinsurance premiums” it got from CCFC, plus modest investment income ($252 during 2012). As a result, RMIC was financially capable of paying claims only by returning to Sheperd Royalty the “reinsurance premiums” that Sheperd Royalty had directed to it.

True risk transfer requires that the insured be reimbursed in all realistic loss scenarios, including those in which its claims exceed the premiums paid. As Dr. Crawshaw accurately explained: “Risk is not shifted if claims [paid] can never be significantly more than the

Insurance. There are three fatal problems with this argument. First, petitioners of- fered no credible evidence that CCFC or RMIC ever executed a “binder” for insurance coverage during 2012. Second, CCFC was not an insurance company capable of issuing a binder for insurance. See supra pp. 11–12. Third, the purported “assignment” of policies, allegedly dated December 21, 2012, does not appear to be an authentic docu- ment. See supra note 6. 38

[*38] premium . . . . [I]n that case, the insurance buyer would be better off not buying the insurance, but instead paying the premium into, and claims out of, a savings account.” Sheperd Royalty achieved no transfer of risk because any losses it incurred could be reimbursed only out of the cash it itself had previously supplied.

In a true reinsurance arrangement, the primary insurer remains liable if the reinsurer cannot pay, and risk transfer can arise in that way. But that was not true here. Article X(E) of the Reinsurance Agree- ment provided that, “[i]n the event Reinsurer [RMIC] becomes insolvent, the shareholders of such Reinsurer shall be liable for any amounts due to Ceding Company [CCFC Insurance] by Reinsurer, as a result of un- dercapitalization of Reinsurer.” Mr. Sheperd was RMIC’s sole share- holder, and RMIC was severely undercapitalized. The Reinsurance Agreement thus made him personally liable for, and pro tanto relieved CCFC Insurance of liability for, any approved claims that RMIC was unable to pay. There was thus no transfer of risk either to RMIC or to CCFC Insurance. 18

Because Sheperd Royalty achieved no risk transfer by paying $1,099,900, and because that payment (less CCFC’s fees) was directed to an affiliate wholly owned by Mr. Sheperd, the payment was economi- cally equivalent to establishing a reserve for self-insurance. Indeed, Mr. Sheperd would have been better off, in two respects, if he had simply deposited $1,099,000 in a bank account and saved it for a rainy day. First, he would not have had to pay $75,893 in fees to Mr. Cope for cre- ating the bogus insurance arrangement. Second, by signing the Rein- surance Agreement, he made himself personally liable for any Sheperd Royalty losses that could not be defrayed out of the RMIC self-insurance reserve. He thus forfeited pro tanto the limitation on personal liability that he achieved by organizing Sheperd Royalty as an LLC.

18 Article V(B) of the Reinsurance Agreement likewise operated to obviate risk

transfer to CCFC Insurance. It provided that “Reinsurer [RMIC] agrees that any net amount due to or from Reinsurer [RMIC] under this Agreement may be offset by any amounts due to or from Ceding Company [CCFC Insurance] or any of its affiliates by Reinsurer or Insured(s) [Sheperd Royalty] . . . under any other agreements between the parties.” Article V(B) thus allowed CCFC Insurance to offset amounts it owed Sheperd Royalty by any amounts RMIC or Sheperd Royalty owed it. In the event of an insurance claim by Sheperd Royalty under the Master Policy, CCFC Insurance could invoke this clause—paying nothing to Sheperd Royalty since the amount of its claim would be offset by the amount RMIC owed CCFC Insurance under the Reinsur- ance Agreement. 39

[*39] C. Risk Distribution

“From the insurer’s perspective, insurance is a risk-distribution device, that is, a mechanism by which the insurer pools multiple risks of multiple insureds in order to take advantage of ‘the law of large num- bers.’” R.V.I., 145 T.C. at 228. This statistical phenomenon is reflected in the financial world by the diversification of investment portfolios. “It is embodied in the day-to-day world by the adage, ‘Don’t put all your eggs in one basket.’” Ibid. (quoting Clougherty Packing Co. v. Commis- sioner, 811 F.2d 1297, 1300 (9th Cir. 1987), aff’g 84 T.C. 948 (1985)).

Generally, risk distribution occurs when the insurer pools a suf- ficiently large collection of risks that are completely unrelated or are otherwise independent of each other. See Rent-A-Center, 142 T.C. at 24. As the Tenth Circuit has explained, risks are independent when “the likelihood of a loss under one policy is independent of the likelihood of a loss under a separate policy.” Rsrv. Mech. Corp. v. Commissioner, 34 F.4th at 904 (citing Clougherty Packing Co. v. Commissioner, 811 F.2d at 1300); see Avrahami, 149 T.C. at 181; Rent-A-Center, 142 T.C. at 24 (stating that risks are independent when they “are generally unaffected by the same event or circumstance” (citing Humana Inc. v. Commis- sioner, 881 F.2d 247, 257 (6th Cir. 1989), aff’g in part, rev’g in part and remanding 88 T.C. 197 (1987))); Swift, T.C. Memo. 2024-13, at *28.

The “law of large numbers” posits that “the average of a large number of independent losses will be close to the expected loss.” Av- rahami, 149 T.C. at 181; see Rsrv. Mech. Corp. v. Commissioner, 34 F.4th at 904; R.V.I., 145 T.C. at 228; Securitas Holdings, 108 T.C.M. (CCH) at 496. Thus, “[b]y assuming numerous relatively small, independent risks that occur randomly over time, the insurer smoothes out losses to match more closely its receipt of premiums.” Rent-A-Center, 142 T.C. at 24 (quoting Clougherty Packing Co. v. Commissioner, 811 F.2d at 1300). Distributing risk also “allows the insurer to reduce the possibility that a single costly claim will exceed the amount taken in as a premium.” Securitas Holdings, 108 T.C.M. (CCH) at 496 (quoting Clougherty Pack- ing Co. v. Commissioner, 811 F.2d at 1300).

The existence of risk distribution in the instant cases must be an- alyzed from the perspective of RMIC, the putative “reinsurer,” which allegedly assumed 100% of the risks covered by the Master Insurance Policy. See Rent-A-Center, 142 T.C. at 24; see also Syzygy, 117 T.C.M. (CCH) at 1172 (considering whether putative reinsurer “distributed risk by . . . reinsuring unrelated risks”). RMIC was party to only one 40

[*40] reinsurance agreement, and that agreement covered the risks of only one insured—Sheperd Royalty. Unlike a traditional insurance company, therefore, RMIC did not accomplish risk distribution by “pool[ing] multiple risks of multiple insureds.” See R.V.I., 145 T.C. at 228; see also Avrahami, 149 T.C. at 184 (finding that 7 types of policies issued to 4 related entities did not adequately distribute risk); Rsrv. Mech. Corp., 115 T.C.M. (CCH) at 1479–80 (finding 10+ policies issued to 3 related entities insufficient).

On two prior occasions we have held that a captive insurer achieved risk distribution even though it insured only the risks of a sin- gle affiliated corporate group. See Rent-A-Center, 142 T.C. at 24; Secu- ritas Holdings, 108 T.C.M. (CCH) at 496. Despite the lack of multiple unrelated insureds, we reasoned that the risks insured against were suf- ficiently numerous, diverse, and independent to enable the “law of large numbers” to operate.

In Rent-A-Center, 142 T.C. at 24, the captive insured three differ- ent types of risk: workers’ compensation, automobile liability, and gen- eral liability. The insured parties had operations in all 50 States, the District of Columbia, Puerto Rico, and Canada. Ibid. The insured par- ties had “between 14,300 and 19,740 [insured] employees” and “between 7,143 and 8,027 insured vehicles,” and they conducted their operations at “between 2,623 and 3,081 stores.” Ibid. The risks insured against, in short, arose under three distinct lines of insurance coverage, were ex- tremely numerous, and were geographically disparate. We found these risks to be independent because they were “generally unaffected by the same event or circumstance.” Ibid.

The captive in Securitas Holdings, 108 T.C.M. (CCH) at 496, in- sured five types of risk: workers’ compensation, automobile liability, em- ployment practices, fidelity liability, and general liability. The insured parties included 25–45 separate entities. Ibid. Those entities employed more than 200,000 workers in 20 countries and operated more than 2,250 vehicles. Ibid. As in Rent-A-Center, the risks insured against in Securitas Holdings arose under multiple distinct lines of coverage, were geographically disparate, were largely independent of each other, and were sufficiently numerous to satisfy “the law of large numbers.”

Petitioners contend that RMIC distributed risk by reinsuring against title risk on 511 mineral leases. Dr. Crawshaw characterized this argument as a “red herring that confuses risk associated with an individual lease with the risk [Sheperd Royalty] was exposed to,” 41

[*41] namely, “the risk arising from the aggregate of all the leases.” As Dr. Crawshaw explained, an insurer would regard Sheperd Royalty’s aggregate title warranty risk as a single risk, much as an auto insurer would regard an individual driver’s aggregate risk of accident as a single risk, even though the driver might take hundreds of car trips annually. The insurer would then achieve risk distribution by pooling that risk with the accident risks posed by thousands of other insureds and (possi- bly) with other types of risks (e.g., homeowner’s or worker’s compensa- tion) incurred by that insured and multiple other insureds.

From this perspective, the facts here are vastly different from the facts in Rent-A-Center and Securitas Holdings. In the instant cases there was only one insured entity, Sheperd Royalty. It conducted its operations in one state, and its activities were largely confined to three counties within that state. Although the Master Insurance Policy nom- inally specified four lines of insurance coverage, two were wholly inap- posite for Sheperd Royalty. See supra pp. 17, 24. In reality, Sheperd Royalty sought coverage for only one risk: the risk arising from warrant- ing good title on mineral leases. And the circumstances that might cause a loss were precisely the same for each lease—negligence or over- sight by Mr. Sheperd or his associates that caused them to miscalculate the mineral interest being conveyed. See Rent-A-Center, 142 T.C. at 24 (stating that risks are independent only if they “are generally unaffected by the same event or circumstance”). In short, RMIC reinsured only one entity, and the risks it insured against were not sufficiently numerous, diverse, geographically disparate, or independent to enable the “law of large numbers” to operate. We accordingly conclude that the risk distri- bution necessary for true “insurance” did not exist here. 19

Against this conclusion petitioners advance two principal argu- ments. First, citing Mr. Sanger’s report, they assert that Sheperd Roy- alty bore risks besides title warranty risk. These risks supposedly could include liability for “drilling too close to a structure,” claims regarding “improper surface usage or pollution,” liability arising from “surface

19 This Court has concluded in prior microcaptive cases that the captive insurer

likewise did not achieve risk distribution. See Avrahami, 149 T.C. at 181–90; Patel, T.C. Memo. 2024-34, at *38–42; Swift, T.C. Memo. 2024-13, at *29–31; Caylor Land & Dev., 121 T.C.M. (CCH) at 1213–15; Syzygy, 117 T.C.M. (CCH) at 1172–74; Rsrv. Mech. Corp., 115 T.C.M. (CCH) at 1483–85. In most of these earlier cases, the captive served as the putative “primary insurer” and urged that it achieved risk distribution, not only by insuring multiple risks of its affiliates, but also by participating in a “reinsurance pool” that included unrelated parties. No “reinsurance pool” or similar structure ex- isted here. 42

[*42] damage litigation,” “third-party liability,” liability for miscalcula- tion of the production royalty, and claims regarding “improper deduc- tions from the royalty share,” “shut-in royalty payments,” and “improper unitization.”

We found no evidentiary or logical support for these assertions. Sheperd Royalty’s business consisted solely of acquiring and assigning leases. It never engaged in drilling or other E&P activity and never in- tended to do so. The only warranty it made in the lease assignment contract was a warranty of good title.

Sheperd Royalty could not possibly incur liability for the first seven events listed above because those events would occur (if ever) at a time when Sheperd Royalty no longer owned the lease. Sheperd Royalty assigned to Cordillera 100% of the mineral exploitation rights granted by each lease. If drilling by Cordillera or a subsequent assignee caused one of the problems petitioners mentioned, or if that E&P company mis- calculated the production royalty due to the lessor, that E&P company would bear liability for any resulting damages. Equally unpersuasive is petitioners’ assertion that Sheperd Royalty could be liable for “improper unitization.” As Mr. Pinson credibly testified, “unitization” is generally done by order from the Oklahoma Corporations Commission. Petition- ers did not explain how the assignor of a lease could be liable for the Commission’s actions.

Second, while Sheperd Royalty warranted title on only 511 leases, petitioners cite Mr. Sanger’s opinion that it could face “as many as 1,600 possible and separate ‘risk exposure’ incidents.” He got to that number by treating each successive lease assignment as creating a distinct “risk exposure.” Assuming the assignment of 500 leases to Cordillera, he as- serted that Cordillera’s assignment of those leases to Apache created a second set of 500 “risk exposures.” And he hypothesized that Apache might then assign the leases to somebody else, creating a third set of 500 “risk exposures.”

Mr. Pinson cogently explained the error in Mr. Sanger’s math. Sheperd Royalty purchased and assigned 511 leases, and it had a con- ceivable title warranty risk during the 3-year term of each lease. That was it. When one assignee replaced another, it stepped into the prior assignee’s shoes. This did not create a new “risk exposure.” It simply 43

[*43] changed the identity of the claimant who would be asserting the putative title warranty claim. 20

D. Insurance in the Commonly Accepted Sense

The absence of risk shifting and meaningful risk distribution alone is enough for us to conclude that the arrangements between Shep- erd Royalty and RMIC were not insurance. See AMERCO & Subs. v. Commissioner, 96 T.C. 18, 40 (1991) (holding that risk-shifting and risk- distributing “are necessary to the existence of insurance”), aff’d, 979 F.2d 162 (9th Cir. 1991). We also find that the arrangements did not constitute insurance in the commonly accepted sense. See Rsrv. Mech. Corp. v. Commissioner, 34 F.4th at 913–16; Avrahami, 149 T.C. at 191– 97; Swift, T.C. Memo. 2024-13, at *36–44; Caylor Land & Dev., 121 T.C.M. (CCH) at 1215–17.

In making this evaluation, we consider numerous factors, “includ- ing whether the company was organized, operated, and regulated as an insurance company; whether the insurer was adequately capitalized; whether the policies were valid and binding; whether the premiums were reasonable and the result of an arm’s-length transaction; and whether claims were paid.” Avrahami, 149 T.C. at 191. We have also considered whether the premiums charged were “actuarially deter- mined,” whether “comparable coverage was more expensive or even available,” whether “there was a circular flow of funds,” and whether the putative insurance company “was created for legitimate nontax rea- sons.” Rsrv. Mech. Corp., 115 T.C.M. (CCH) at 1484; accord Caylor Land & Dev., 121 T.C.M. (CCH) at 1215. Petitioners score very poorly under these criteria.

1. Organized, Regulated, and Operated as an Insur- ance Company

RMIC was incorporated in the Sac and Fox Nation, a tribal juris- diction. During 2012 the Sac and Fox Nation had no laws governing insurance and no insurance regulatory authority. RMIC was

20 Mr. Sanger’s math does not work even on his own terms. The Master Insur-

ance Policy was not executed until December 2012. Cordillera had been merged out of existence in May 2012, so it could not possibly assert a title claim. The only party that could assert such a claim would be Apache, which stepped into Cordillera’s shoes. And Mr. Sanger’s assumption that Apache might reassign the 511 leases to somebody else was pure speculation. So we are left where we started—with one conceivable claim on each lease. 44

[*44] accordingly organized as an ordinary domestic corporation. It was neither organized nor regulated as an insurance company.

Nor did RMIC operate as a normal insurance company. The evi- dence supporting that conclusion is overwhelming. As explained in greater detail below, RMIC was inadequately capitalized, reinsured il- logical lines of coverage, received excessive and irrationally calculated premiums, engaged in irregular claims-payment practices, and made risky investments designed to benefit Mr. Sheperd rather than to safe- guard its ability to pay claims.

2. Adequate Capitalization

RMIC had initial capitalization of zero and no paid-in capital. See supra pp. 14–15, 31. Its alleged “capital” consisted solely of the cash that Sheperd Royalty transferred to it (via CCFC) as putative “reinsur- ance premiums.” But “premiums” do not constitute “initial capitaliza- tion.”

Mr. Rivelle, whom Mr. Cope hired to prepare the March 2013 “ac- tuarial report,” acknowledged that adequate capitalization is essential for an insurance company. He testified that, in preparing his report, he “assum[ed] that there’s capital.” He agreed that “all insurance compa- nies need to be capitalized with something other than premiums,” noting that he was “not familiar with any [insurance] company that can be formed without being capitalized.” He agreed that “premium[s] [are] not capital,” while noting that they may become shareholder’s equity if and when “earned.”

Petitioners contend that RMIC’s capitalization of zero is irrele- vant because “initial capitalization is governed by the rules of the juris- diction in which [the captive] was formed.” It is true that adequacy of capitalization is generally determined—at least in the first instance— by the jurisdiction that regulates the insurer. See Avrahami, 149 T.C. at 193; R.V.I., 145 T.C. at 231. But the Sac and Fox Nation in 2012 had no laws governing insurance and no insurance regulatory authority. Of necessity, therefore, it had no capitalization requirements for insurance companies that RMIC could purport to meet. Assuming arguendo that the Sac and Fox Nation allowed ordinary domestic corporations to be capitalized at zero—an assumption petitioners did not prove—that is ir- relevant in deciding whether RMIC had sufficient capital to be recog- nized as an “insurance company” for Federal income tax purposes. 45

[*45] 3. Existence of Valid and Binding Policies

The Master Insurance Policy ran between Sheperd Royalty and CCFC Insurance, the supposed “primary insurer.” The Reinsurance Agreement ran between CCFC Insurance and RMIC, the supposed “re- insurer.” But CCFC Insurance was not incorporated until January 3, 2013, and it therefore did not exist during the 2012 policy period. For that reason alone, no valid and binding policies existed during the tax year at issue.

Recognizing this problem, petitioners sought to have admitted into evidence five “newly discovered” master policies, two of which were allegedly in effect during 2012. See supra pp. 18–19. The first, ostensi- bly dated October 1, 2012, lists CCFC as the insurer and Management (a nonexistent entity) as the insured. The second, ostensibly dated De- cember 21, 2012, lists CCFC Insurance as the insurer and Management as the insured. Finding these documents to be inauthentic, we have ex- cluded them from evidence. See supra pp. 19–20. But even if authentic, they would not help petitioners. CCFC, the supposed “primary insurer” on the October 2012 policy, was not an “insurance company.” See supra pp. 11–12, 36. And CCFC Insurance, the supposed “primary insurer” on the December 2012 policy, did not exist during 2012.

Apart from these fundamental flaws, the Master Insurance Policy had a number of unusual features. The Policy does not mention title warranty risk, allegedly Mr. Sheperd’s main concern. As Dr. Crawshaw explained, the definitions of coverage are extremely brief and vague, “so vague that they do not objectively define the coverage.” Two lines of coverage—for “difference in conditions” and faulty product or rework service—made no sense for the putative insured. See supra p. 24. The lines of coverage, policy limits, and deductibles stated in the Policy did not match the lines of coverage, policy limits, and deductibles appearing in Mr. Rivelle’s “actuarial report.” See supra pp. 16–18.

The Master Insurance Policy also included provisions that appear to negate coverage for much of the risk allegedly insured against. Sec- tion XI(G), captioned “Conditions, Exclusions and Limitations,” states that CCFC Insurance has no liability for “legal fees,” “[d]amages as a consequence of a covered claim,” and “[d]amages for breach of any im- plied or express warranty.” These would seem to be precisely the types of risks about which Mr. Sheperd was supposedly concerned. And CCFC Insurance could cancel the Policy at any time, for any reason, so long as it refunded the residual premium. See supra p. 16. If an “insurer” can 46

[*46] cancel a policy whenever it receives a claim it does not wish to pay, the “insurance” is not worth much.

4. Existence of Actuarially Determined Premiums

At various points in these cases, Messrs. Cope and Rivelle came up with an astounding array of divergent premium calculations, none of which matched the $1,099,900 “premium” that Sheperd Royalty actually paid at year-end 2012. The Master Insurance Policy stated an aggregate annual premium of $1,072,500. That was the bottom-line number shown in Mr. Rivelle’s “actuarial report,” but he determined that pre- mium for 2013. A CCFC “Risk Transfer Report” calculated an annual premium of $536,250. That figure, employing an acreage-based compu- tation method, was exactly half the premium Mr. Rivelle calculated us- ing entirely different data. See supra p. 22. At trial Mr. Cope testified to a two-year premium of $1,300,000. But he did not explain the math justifying that calculation, and it was off-kilter for at least two reasons: Sheperd Royalty evidently sought coverage only for 2012 and (had it sought coverage for two years) its “premium” payment would have been $200,000 short.

Apart from their randomness, these premium numbers had no sound actuarial basis, as Dr. Crawshaw cogently explains in his report. To calculate a reasonable premium, Mr. Rivelle needed to evaluate the actual risk that Sheperd Royalty incurred by warranting good title on the mineral leases. He made absolutely no effort to do that.

Mr. Rivelle conducted no interviews and performed no research regarding the frequency with which “title busts” or other irregularities occur in mineral leases generally. One obvious source of relevant infor- mation would have been landman companies, like Pinson, that “ran ti- tle” on mineral leases for a living. Mr. Rivelle was not informed that Pinson had vetted title on all Sheperd Royalty leases that were assigned to Cordillera.

Instead, Mr. Rivelle relied for his benchmarks on generic data about premiums paid for “representations and warranties” coverage in the context of corporate merger and divestiture transactions (collec- tively, M&A transactions). But he did not explain the basis for his con- clusion that these transactions supplied comparable data. Representa- tions and warranties in M&A transactions—e.g., concerning the acquir- ing company’s intentions and future prospects—cannot be inde- pendently verified, and the premium for insurance coverage will 47

[*47] presumably reflect that risk. But for mineral leases, information about title is readily available in county courthouses and elsewhere. Once a reputable landman firm like Pinson has vetted title after re- searching those publicly available resources, the risk of an error is dif- ferent from the risk that an acquiring company will include a misrepre- sentation of fact in a prospectus.

In any event, Mr. Rivelle’s methodology for computing the pre- mium was based on a false premise. In assessing the title warranty risks that Sheperd Royalty supposedly faced, he keyed his calculation to annual gross revenues. He assumed that Sheperd Royalty would have annual gross revenues of $25 million for 2013 through 2016, as it had had for 2012, then discounted each future year’s revenues to present value.

Mr. Rivelle supplied no basis for assuming that Sheperd Royalty’s revenues would remain constant at $25 million annually through 2016. In fact, its revenues had tapered off to zero by December 2012 and never recovered, because Cordillera and Apache stopped acquiring its leases. That fact was known to Messrs. Sheperd and Cope as of March 2013, but it was apparently not conveyed to Mr. Rivelle. He admitted that, if Sheperd Royalty was not expected to have annual revenues of $25 mil- lion through 2016, his premium calculation would have been dramati- cally different.

The other component in Mr. Rivelle’s premium calculation was his net loss estimate. He assumed that Sheperd Royalty would incur an “underlying loss rate” of $10 for every $1,000 of revenue. Mr. Rivelle admitted that this ratio had nothing to do with the company’s historical loss experience. His report supplied no explanation as to how he arrived at this number, except to say that “the insured’s management and its advisory team provided insight and specific input into the expected losses.” This suggests that he performed no real “actuarial analysis” at all.

5. Payment of Claims

Sheperd Royalty submitted no claims under the Master Insur- ance Policy during 2012 or 2013. The IRS examination began in May 2014, and this turn of events suggested to petitioners that submitting claims might be in their interest. During the second half of 2014 Shep- erd Royalty submitted eight claims totaling $114,917. See supra pp. 27– 28. CCFC paid those claims in full. 48

[*48] As Dr. Crawshaw explains, the circumstances surrounding the claims-payment process did not come close to comporting with insurance norms:

• The Policy covered the 2012 calendar year and set a deadline of March 2, 2013, for submission of claims. The claims were submit- ted roughly 18 months after that deadline.

• Six of the claims did not qualify for payment because they were for amounts below the deductible. See supra p. 28.

• None of the claims was accompanied by meaningful documenta- tion or any proof of loss. Mr. Cope admitted that CCFC did not investigate any of the claims—ignoring the rights CCFC Insur- ance possessed under the Policy—but simply “took Sheperd’s word for it.”

• Contrary to the verbiage on the claim forms, none of the claims involved a “title bust” or title irregularity of any kind. In each case, Sheperd Royalty acquired good title to the mineral interest and was free to retain that interest or assign it. See supra pp. 29– 30. There is no evidence that Sheperd Royalty suffered an eco- nomic loss with respect to any of the leases in question.

• Any loss Sheperd Royalty suffered was attributable to mistakes it made when preparing the leases. Correction of those errors de- layed its ability to assign the leases until after Cordillera had stopped accepting assignments. If those circumstances were thought to generate a loss, it would be an ordinary business loss, not an insurance loss covered by any line of insurance set forth in the Master Insurance Policy.

No genuine insurance company would blithely pay claims that, for four or five distinct reasons, did not qualify for payment. We accord- ingly conclude that CCFC’s payment of these claims should be given no weight in assessing whether the arrangements at issue constituted “in- surance.”

6. Circular Flow of Funds

There was a circular flow of funds from Sheperd Royalty to CCFC to RMIC and then back to Sheperd Royalty (or its shareholders and af- filiates). This circular flow of funds demonstrates (among other things) 49

[*49] that RMIC did not invest its “insurance reserves” as a real insur- ance company would do.

By mid-2014 RMIC had extended LOCs totaling $1.5 million to Sheperd Royalty and LDV Farms, Mr. Sheperd’s sole proprietorship. RMIC’s potential liability on these two LOCs exceeded its assets by al- most $500,000. There is no evidence that Sheperd Royalty was engaged in meaningful business activity during 2013, its lease-acquisition busi- ness having terminated in November 2012. And LDV Farms was a horse- and dog-breeding business that had never turned a profit. These borrowers were not “prudent risks” in objective terms, especially for bor- rowing of the magnitude that RMIC facilitated.

The Sheperd Royalty LOC specified that all funds advanced by RMIC were to be used for Sheperd Royalty’s “operating expenses.” That restriction was repeatedly ignored. Using a checkbook linked to the LOC, Mr. Sheperd wrote dozens of checks for personal expenses. These included payments for private school tuition, a $10,000 diamond ring, a prom dress, wedding presents, birthday gifts, purchase of a jet ski boat, “shopping money” for his children, and transfers into his wife’s invest- ment accounts.

RMIC’s bank records show aggregate cash advances of $847,600 under the two LOCs during 2013–2015. See supra pp. 26–27. Several large checks were written to LDV Resources, another entity owned by Mr. Sheperd, and to Big Chief Resources, an entity owned by his son. RMIC paid $240,676 in August 2013 to purchase a tract of land down the road from petitioners’ residence. True insurance companies invest their reserves in a prudent and diversified manner to safeguard their ability to pay claims. RMIC deployed virtually 100% of its assets, di- rectly or indirectly, to benefit Mr. Sheperd and his family.

7. Conclusion

Apart from the labels attached to the entities and documents dis- cussed above, these cases are bereft of evidence pointing to the existence of true “insurance.” The entities in question either did not exist during 2012 or were not organized or regulated as insurance companies. Shep- erd Royalty achieved no transfer of risk, and RMIC, the putative rein- surer, accomplished no meaningful distribution of risk. And for six dis- tinct reasons, the arrangements at issue did not remotely resemble in- surance in the commonly accepted sense. See Avrahami, 149 T.C. at 191. We accordingly hold that the $1,099,900 paid by Sheperd Royalty 50

[*50] at year-end 2012 did not constitute “insurance premiums” deduct- ible under section 162, and that RMIC was not an “insurance company” that received “written premiums” qualifying for exemption under sec- tion 831(b)(2)(A).

III. Penalties

The IRS issued petitioners timely Notices of Deficiency determin- ing accuracy-related penalties under section 6662. The notice issued to the Sheperds determined a 40% penalty under section 6662(b)(6) and (i) for a transaction lacking economic substance and (in the alternative) a 20% penalty for negligence and/or a substantial understatement of in- come tax under section 6662(a) and (b)(1) and (2). The notice issued to RMIC determined a 20% penalty only. See ibid.

In their opening posttrial brief petitioners did not dispute their liability for any of these penalties. Rather, they challenged the penalties for the first time in their answering brief, in defiance of the Court’s in- structions that answering briefs were to contain no legal argument but were to be confined to rebutting the other side’s proposed findings of fact. For these reasons, we could deem petitioners to have waived any argument against the penalties. See Ashkouri v. Commissioner, T.C. Memo. 2019-95, 118 T.C.M. (CCH) 106, 111 n.9 (“Having conceded an issue by failing to advance a meaningful argument on that issue in their opening brief, [the taxpayers] could not withdraw that concession by be- latedly including a cognizable argument in their reply brief.”). Because we are reluctant to saddle petitioners with their attorneys’ failure to fol- low our instructions, we will nevertheless consider the penalties on the merits.

A. Penalty Approval

Section 6751(b)(1) provides that “[n]o penalty under this title shall be assessed unless the initial determination of such assessment is personally approved (in writing) by the immediate supervisor of the in- dividual making such determination.” In Belair Woods, LLC v. Commis- sioner, 154 T.C. 1, 14–15 (2020), we ruled that the “initial determina- tion” of a penalty assessment is typically embodied in a letter by which the IRS formally notifies the taxpayer that it has made a definite deci- sion to assert penalties. Supervisory approval need not be recorded on any particular form or document; the only requirement is a writing that manifests the immediate supervisor’s intent to approve the penalty. Su- pervisory approval may be shown by the signature of the revenue agent’s 51

[*51] manager on a 30-day letter. See ibid; Tribune Media Co. v. Com- missioner, T.C. Memo. 2020-2, 119 T.C.M. (CCH) 1006, 1010–11.

Petitioners do not dispute that supervisory approval of the penal- ties was timely secured. The IRS first communicated to petitioners its intention to assert the penalties on January 19, 2017, the date on which it mailed them the 30-day letters with enclosed examination reports. Those letters were signed by Ms. Thomas, RA Currier’s immediate su- pervisor. As noted above, supervisory approval may be shown by the signature of a manager on a 30-day letter. The IRS issued the Notices of Deficiency on November 18, 2018. Because RA Currier secured su- pervisory approval before the 30-day letters and Notices of Deficiency were issued to petitioners, the approval was timely. 21

B. Section 6662(b)(1) and (2) Penalties Against the Sheperds and RMIC

Congress has authorized imposition of a 20% accuracy-related penalty on the portion of an underpayment of tax required to be shown on a return that is attributable to “[n]egligence or disregard of rules or regulations” or to “[a]ny substantial understatement of income tax.” § 6662(a) and (b)(1) and (2). Negligence includes any failure to make a reasonable attempt to comply with the provisions of the Code. See Treas. Reg. § 1.6662-3(b)(1). For individual taxpayers, an understate- ment of income tax is “substantial” if it exceeds the greater of 10% of the tax required to be shown on the return or $5,000. § 6662(d)(1)(A). For corporations an understatement is “substantial” if it exceeds the lesser of 10% of the tax required to be shown on the return (or, if greater, $10,000) or $10 million. § 6662(d)(1)(B).

The Sheperds’ corrected tax liability for 2012 is more than $1.3 million. Because they originally reported a tax liability of less than $1 million, their understatement easily exceeds $5,000 and 10% of the tax required to be shown on their return. RMIC’s corrected tax liability for 2012 is more than $346,000. Because it originally reported a tax

21 The instant cases appear to be appealable to the Tenth Circuit. See supra

p. 3. That court has not squarely addressed the question of when supervisory approval must be secured. Other appellate courts have ruled that approval is timely if secured before the tax is assessed or (in some circumstances) before the notice of deficiency is mailed. See, e.g., Kroner v. Commissioner, 48 F.4th 1272, 1278 (11th Cir. 2022), rev’g in part T.C. Memo. 2020-73; Wells Fargo & Co. v. United States, 957 F.3d 840, 854 (8th Cir. 2020) (“By its terms, [section 6751(b)(1)] requires prior written approval to be ob- tained when the government ‘assesses’ a penalty against a taxpayer.”). Supervisory approval in these cases was timely under either standard. 52

[*52] liability of $38, its understatement easily exceeds $10,000 and 10% of the tax required to be shown on its return. The Commissioner having met his burden of production with respect to the substantial un- derstatement penalties, see § 7491(c), it is unnecessary for us to deter- mine whether petitioners’ underpayments were attributable to negli- gence, see Avrahami, 149 T.C. at 204–05; 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 penalty on more than one ground).

The “reasonable cause” defense may be asserted against the sub- stantial understatement penalty. See § 6664(c)(1). “Reasonable cause requires that the taxpayer have exercised ordinary business care and prudence as to the disputed item.” Neonatology Assocs., P.A. v. Commis- sioner, 115 T.C. 43, 98 (2000), aff’d, 299 F.3d 221 (3d Cir. 2002). The determination of reasonable cause is made on a case-by-case basis, tak- ing into account all pertinent facts and circumstances. Treas. Reg. § 1.6664-4(b)(1). For underpayments related to pass-through items we look at all pertinent facts, including the taxpayer’s own actions, as well as the actions of the pass-through entity. See id. para. (e). Petitioners bear the burden of proving that they had reasonable cause and acted in good faith with respect to the underpayment. See Higbee v. Commis- sioner, 116 T.C. 438, 449 (2001).

Petitioners assert reliance on professional advice as the basis for this defense. See Treas. Reg. § 1.6664-4(b)(1). Reliance on professional advice will absolve a taxpayer only if the taxpayer establishes that it fully disclosed all relevant facts to a competent return preparer, that the errors on the return were “a result of the preparer’s mistakes,” and that it actually relied on the preparer’s advice in good faith. Estate of Gold- man v. Commissioner, 112 T.C. 317, 324 (1999), aff’d sub nom. Schutter v. Commissioner, 242 F.3d 390 (10th Cir. 2000) (unpublished table deci- sion); accord Neonatology, 115 T.C. at 99.

There is no credible evidence that the Sheperds or RMIC met these requirements. They make the high-level assertion that they relied on the advice of “a CPA, tax attorneys, financial planners, and other in- dividuals with tax and accounting credentials.” But they failed to iden- tify these advisers or the specific advice the advisers furnished. And they failed to establish that these individuals were “competent profes- sional[s] who had sufficient expertise to justify reliance.” See Neonatol- ogy, 115 T.C. at 99. RMIC’s return was prepared by Ms. Woodward, who was tasked with preparing returns for many of Mr. Cope’s captives. Ms. 53

[*53] Woodward was not a CPA, and Mr. Cope was not a CPA either. Moreover, he was a promoter of the microcaptive insurance scheme, so the Sheperds could not rely on any advice he offered. See Avrahami, 149 T.C. at 206; 106 Ltd. v. Commissioner, 136 T.C. 67, 79 (2011), aff’d, 684 F.3d 84 (D.C. Cir. 2012). Finally, because CCFC Insurance did not exist during 2012, no competent tax adviser could have believed that premi- ums paid to it for insurance coverage during 2012 were tax deductible.

C. 40% Penalty Against the Sheperds

The last issue is whether we should sustain the 40% accuracy- related penalty determined against the Sheperds. In 2010 Congress en- acted the Health Care and Education Reconciliation Act of 2010 (Act), Pub. L. No. 111-152, § 1409, 124 Stat. 1029, 1067–70, adding a new 20% penalty on the portion of an underpayment attributable to “[a]ny disal- lowance of claimed tax benefits by reason of a transaction lacking eco- nomic substance (within the meaning of section 7701(o)) or failing to meet the requirements of any similar rule of law.” § 6662(b)(6). Section 6662(i) increases that penalty to 40% if the underpayment is attributa- ble to a “nondisclosed noneconomic substance transaction.” Section 6662(i)(2) defines a “nondisclosed noneconomic substance transaction” as one with respect to which the relevant facts affecting the tax treat- ment are not adequately disclosed in the return or in a statement at- tached to the return.

The Act also added section 7701(o) to the Code, codifying the “eco- nomic substance” doctrine. That provision provides a conjunctive test whereby a transaction is treated as having economic substance only if: (1) the transaction changes in a meaningful way (apart from Federal income tax effects) the taxpayer’s economic position and (2) the taxpayer has a substantial purpose (apart from Federal income tax effects) for entering into the transaction. § 7701(o)(1). The codified economic sub- stance doctrine applies “[i]n the case of any transaction to which the economic substance doctrine is relevant.” Ibid. And the determination of whether the economic substance doctrine “is relevant” must be made in the same manner as if section 7701(o) had never been enacted. § 7701(o)(5)(C).

To date, there has been minimal caselaw addressing these provi- sions. In none of the seven microcaptive insurance cases decided to date did this Court address whether the transactions lacked economic sub- stance within the meaning of section 7701(o)(1). Nor did those opinions consider what constitutes “adequate[] disclos[ure]” of a microcaptive 54

[*54] transaction under section 6662(i)(2). The Court recently withheld ruling on these questions and ordered additional briefing on the “rele- vancy” question. See Patel, T.C. Memo. 2024-34 at *3 n.5; Order, Patel v. Commissioner, Nos. 24344-17, et al. (July 19, 2024) (No. 366). We will accordingly defer ruling on the applicability of the 40% penalty, which will be addressed in a subsequent opinion.

To reflect the foregoing,

Decisions will be entered in due course upon completion of further proceedings.