The IRS Losses: An Overview

On Wednesday, June 15th, we're hosting a webinar titled "An Introduction to Captive Insurance."  It starts at 1PM CST and will last for about half an hour.  You can sign up at this link.

We formed and operate the first series LLC in Montana (named Aegis) for captive insurers.  Several other firms provide key services such as accounting, audit and actuarial work.  Please contact us at 832.330.4101 if you'd like to discuss forming a captive for your company. ​

   Had the service continued to focus on smaller companies, it is entirely likely courts would have continued to rule in their favor.  However, beginning with Humana and continuing through the UPS case, the IRS sought bigger game.  At the time of trial in 1978, Humana was a publicly traded company that operated 92 hospitals over 23 states.[1]  Amerco -- more popularly known a U-Haul -- had 250 subsidiary corporations by 1984.[2]  Sears Roebuck was one of the largest retailers in the US and its “captive” Allstate was one of the country’s largest insurers.[3]  Harper was an international transportation company with subsidiaries throughout the world, including Taiwan, Singapore, Thailand, U.K., South Africa, Canada and Australia.[4]   
 
   In addition to the larger companies forming captives, three important structural developments emerged in response to the IRS victories.  First, most captives in these transactions had significant third party risk: the Harper captive contained 30% non-parent risk;[5] the Sears captive contained over 99% non-parent risk;[6] the captive in Amerco contained between 52%-74% non-parent risk[7] while the Ocean Drilling captive underwrote 56% and 57% non-parent business in the years in question.[8]  These large amounts of non-corporate group insureds gave the captive a sufficiently large pool of clients over which to distribute risk, allowing each to comply with the second prong of the Helvering standard.  Second, the captive insurance operations for most captives were elevated to legitimate corporate divisions with an adequate level of staff, separate books and records and their own offices.[9] 

   But most importantly, IRS victories were premised on a parent-captive relationship so inter-twined that the captive was conceptually transformed from a separate company to a financial account on the parent’s balance sheet -- a concept based on the parent owning all the captive’s stock.[10]  In direct contrast to the IRS victories, the captives in the IRS losses were insuring a larger group of “brother-sister” companies, meaning the companies purchasing insurance had no captive ownership interest.  This issue has its conceptual origins in the Clougherty case where the parent owned 100% of its insurance subsidiary.  There the court ruled for the IRS, noting

   When petitioner sustains losses covered by its workers' compensation insurance, 92 percent is sustained by Lombardy. Accordingly, because petitioner, through its wholly owned Arizona corporation, owns all of Lombardy, it has not shifted the risk of sustaining such losses to unrelated parties in exchange for insurance premiums because the premiums were  paid to the wholly owned subsidiary of its wholly owned subsidiary. … We hasten to point out, nevertheless, that parents and wholly owned subsidiaries are involved in Carnation and here. We express no opinion, however, as to situations where the taxpayer does not completely own the subsidiary [italics added].[11]
           
   This issue came to the forefront in the Humana case, where the holding company owned all the stock of various subsidiaries, one of which was captive.  While the lower and appellate courts determined Humana’s parent’s insurance premiums were non-deductible (because it owned 75% of the captive’s stock and therefor directly analogous to the fact pattern of the IRS victories), the appellate court ruled the premiums paid by the “brother-sister” companies were deductible because the brother sister companies had no ownership interest in the captive.  Central to this reasoning was the analysis provided by Dr. Irving Plotkin who noted,

“True insurance relieves the firm’s balance sheet of any potential impact of the financial consequences of the insured peril…[However] as long  as the firm deals with its captive, its balance sheet cannot be protected from the financial vicissitudes of the insured peril.”[12]  As the brother-sister companies did not own the captive’s stock, an indemnification payment would have no negative impact on their respective balance sheets, thereby removing the reasoning underlying the premium non-deductibility.[13]    

    And finally, one can’t help but conclude the sheer weight of the historical tide forced the courts to develop some type of compromise regarding the legality of captive insurance companies.  Perhaps the strongest historical thread was the continual implementation of captives for legitimate risk management purposes.  Insurance advisors also continued to promote and recommend this structure as a legitimate risk management tool while several US states passed captive insurance statutes with the obvious intent to attract business.  All these events occurred over the clearly stated objections of the service in its two Revenue Rulings and their continued filing of cases challenging the substance of a captive insurance company.  Unlike the numerous tax evasion cases and schemes of the 1990s, there was a clear business need for this concept.


[1] Humana Inc. v. Comm’r of Internal Revenue, 88 T.C. 197, 199 (1987)

[2] Amerco v. Comm’r of Internal Revenue 96 T.C. 18, 21 (1991)

[3] Sears, Roebuck and Co. v. Comm’r of Internal Revenue 96 T.C. 61, 63 (1991)

[4] The Harper Group v. Comm’r of Internal Revenue 96 T.C. 45, 49 (1991)

[5] Ocean Drilling and Exploration Corp. v. United States, 24 Cl. Ct. 714, 726 (1991); see also Harper at 51 (see chart showing that unrelated parties premium comprised between 29% to 72% of the captives premium income from 1974-1983)

[6] Id; see also Sears at 63 (“The total premiums for policies issued to Sears by Allstate represented approximately .25% to 1% percent o the total premiums earned by Allstate from all insureds on all lines of business for the years in issue.)”

[7] Id; see also Amerco at 26 (“Non-U-Haul policies provided a wide variety of insurance and reinsurance coverages.”)

[8] Id

[9] See Harper at 49 (“Rampart maintained its own books of account, business records, bank accounts and investments…Rampart conducted its business in a manner consistent with its status as a separate corporation.”); Amerco at 21 (“In 1979 it [the captive] was licensed in 12 states; by the end of 1985 it was licensed in 45 states and the District of Columbia.  Republic Western held A.E. Best Rating of “A” for 1980 through 1984.”; Allstate was one of the largest insurance companies in the United States;

[10] Mobil Oil Co.v United States, 8 Cl. Ct. 555, 566 (1985)(“Any losses suffered by the insurance affiliate would be reflected on Mobil’s financial statement.” See also Beech Aircraft Corp. v. United States, 1984 U.S. Dist. LEXIS 15251, page 11 (1984)(“…if and when a loss occurred and was paid by Travel Air, the net worth of Beech was reduced to approximately the same extent as though Beech had paid the loss itself.”); see discussion above

[11] Clougherty at 958-959

[12] Humana 88 T.C 197 at 211

[13] Humana 881 F.2d 247 at 250