On August 21, 2017, the Tax Court issued its opinion in two companion cases Benyamin Avrahami and Orna Avrahami v. Commissioner, and Feedback Insurance Company, Ltd. v. Commissioner, 149 T.C. No. 7, holding against the taxpayers in the first section 831(b) micro captive case to be decided. The court concluded that the arrangements in question were not insurance for federal tax purposes, and, accordingly, concluded that amounts purportedly paid as insurance premiums to the micro captive were not deductible under section 162, and that the section 953(d) and section 831(b) elections made by the micro captive were invalid for the years before the court (2009 and 2010). The court also held that:
(i) certain amounts the micro captive transferred to its shareholder were dividends not loans,
(ii) certain amounts the micro captive paid to the petitioners (references to the petitioners are to the individual petitioners) were either interest or dividend payments to the extent they exceeded a loan repayment, and
(iii) the petitioners were not liable for the section 6662(a) accuracy related penalties because they reasonably relied in good faith on an adviser who was not a promoter and the court found the case to be one of first impression.
The petitioners in the case owned three shopping centers, three jewelry stores and some commercial real estate businesses. Before the micro captive (Feedback) was formed with Mrs. Avrahami as its sole shareholder, those businesses paid $150,000 in insurance premiums annually. After Feedback was formed, the businesses were paying between $1.1 million to $1.3 million in “insurance” premiums per year, with the amounts paid to Feedback falling just under the $1.2 million annual premium cap for micro captives.
During 2007, the petitioners sought advice from their CPA who suggested some estate planning advice including the possibility of forming a captive insurance company. The CPA also recommended that the petitioners consult with a lawyer who focused her practice on small captive insurance companies. The lawyer participated in the structuring of the petitioners' micro-captive transactions with Feedback, and she therefore was determined by the court to be a "promoter" of those transactions. The petitioners retained the promoter to begin reviewing information about their businesses and to provide advice about a captive insurance company structure, which led to the incorporation of Feedback in St. Kitts where the promoter had helped draft the legislation governing captive insurance companies.
Feedback was authorized by St. Kitts to “conduct small group captive insurance” by the end of 2007. Feedback made an election under section 953(d) to be characterized as a U.S. company for all federal tax purposes and an election under section 831(b) to pay tax only on its investment income. As a prerequisite for both of these elections, Feedback was required to qualify as an insurance company for federal tax purposes.
Feedback began to issue insurance policies covering the Avrahami’s businesses in 2007 and entered into a terrorism risk distribution pooling insurance program set up by the promoter for her clients. For the years before the court the pooling arrangement was facilitated through a St. Kitts reinsurer (“Pan American”) owned by a courtesy director, by the wife of the owner of Feedback’s insurance management company, and in small part by the promoter’s children. The Avrahamis made no changes to their commercially purchased insurance program. Feedback provided various direct coverages to the Avrahami’s businesses, including: business income, employee fidelity, litigation expense, loss of key employee, tax indemnity, administrative actions, and business risk indemnity.
In its words, the court analyzed the micro captive structure applying the "Supreme Court's definition of insurance in LeGierse and its four nonexclusive criteria" which are risk shifting, risk distribution, insurance risk and commonly accepted notions of insurance. In concluding that Feedback’s insurance coverages were not insurance for federal tax purposes, the court found that Feedback failed to distribute risk and was not selling insurance in the commonly accepted sense, determining that it did not need to decide whether the transactions involved insurance risk or risk shifting. In focusing on the lack of risk distribution and the absence of insurance in the commonly accepted sense, the court found many problems with Feedback’s operations, including those listed below.
Premium Pricing – The actuary who set the pricing for Feedback’s policies, as well as for Pan American, was hired by the promoter and only set pricing for captives for the promoter. The premium for the Pan American program was “grossly excessive” and based on a “one-size-fits-all rate.” The actuary’s pricing methodology for Feedback although in part derived from a commercial carrier’s standard rate filing, was flawed; he consistently made choices that would generate higher premiums; and he made adjustments based on his professional judgement without a coherent explanation, and based on input from the promoter.
Claims – No claims were made under the policies Feedback issued until after the Internal Revenue Service (“IRS”) began its audit of the petitioners’ tax returns, and then they were paid on an “ad hoc basis.”
Investments – Feedback invested in illiquid long term investments to related parties without regulatory approval and Feedback also held significant amounts in cash and in real estate loans which were characterized by the court as “investment choices only an unthinking insurance company would make.”
Policies – The coverage under the Pan American policies had an extremely low probability of ever being triggered and provided for the possibility of the payment of claims with a promissory note. The court also found the policies not to be arm’s length and quoted the IRS’ expert: “no reasonable, profit-seeking business would enter into a contract with the terms of the Pan-American coverage with an insurer absent certain beneficial tax considerations.”
Capitalization – The court indicated that Pan American had no funds with which to pay claims because all of the business it wrote was ceded back out and it “would be hard pressed to enforce the cession agreements against the scores of captive insurers” that reinsured it.
In its risk distribution analysis the court considered both the number of entities that Feedback insured and the number of exposure units it insured. In this analysis, the court determined that all of the coverages assumed from Pan American were not insurance and this business comprised approximately 30 percent of the premium Feedback received. The court concluded without much analysis that three or four insureds and risk units comprised of three jewelry stores, two key employees, 35 employees and three commercial real estate properties were not enough to create risk distribution.
Eversheds Sutherland Observation. Although the case was expected to be a reviewed Tax Court opinion, it was not, and the opinion seems carefully designed not to draw any new lines with respect to the definition of insurance. In addition, the adverse result to the petitioners on the insurance issue and the adverse result to the IRS on the penalties it asserted could signal a basis for settlement of other micro captive cases.