Benyamin and Orna Avahami came to the United States from Israel in 1974 and 1980 respectively. They live in Phoenix, have three children, and own three successful jewelry stores in Arizona which have about 35 employees. Their business, called American Findings Corp., elected to be taxed as an S-Corporation.
Additionally, the Avrahamis own other commercial real estate companies which own commercial properties in Arizona. More on these later. Suffice it to say now that the relevant Avrahami entities in 2006 paid about $150,000 in insurance premiums.
Desiring greater tax efficiency, the Avrahamis first talked to their CPA, Craig McEntee, who then put them on to Phoenix estate planner Neil Hiller. In turn, Hiller passed the Avrahamis on to Ms. Celia Clark, a New York lawyer who focused on tax and estate planning. Clark had also been very active with captive insurance companies since 2002, and who helped draft the captive insurance legislation for St. Kitts, a Caribbean island-nation.
This seemed to be a good fit, since the Avrahamis were already looking to forming a captive insurance company. The Avrahamis cut a check to Clark for $75,000 to form what became the Avrahamis' captive, called Feedback Insurance Company.
Feedback was formed at the end of 2007 and obtained its insurance license from St. Kitts that year. Feedback was managed by a St. Kitts captive manager called Heritor Management, Ltd., which was itself owned by Robin Trevors. In 2008, Feedback made two tax elections with the IRS: First, because it was a foreign company, under 953(d) to be taxed as a domestic company for federal income tax purposes and, second, under 831(b) to be taxed as a small insurance company. The latter election, which was then available only to insurance companies which received less than $1.2 million in premiums, allowed Feedback to avoid paying taxes on its premium income received.
Feedback sold insurance to the Avrahami entities, and also into a terrorism risk pool (more on this later) which Clark had set up. Curiously, although the Avrahami entities were purchasing insurance from Feedback, they continue to purchase insurance as they had been from their third-party commercial insurance carriers.
To price the insurance policies which Feedback sold to the Avrahami entities, Clark hired an actuary by the name of Allen Rosenbach. Clark gave Rosenbach a variety of information, including Feedback's business plan and the insurance policies that the Avrahamis wanted Feedback to underwrite for their other businesses. Rosenbach used this information to develop a "base premium" for each policy, which then was modified using various factors.
In general, Rosenbach would find a third-party commercial insurance policy that was similar to that which Feedback issued to the Avrahami entities, and then used the pricing model charged for the commercial insurance policy to determine the base rate. First, Rosenbach modified the rate upwards because the Feedback policies were "claims made" policies, which would ostensibly cover even claims which arose before the policies were issue so long as the claim was made while the policy was in place. Second, Rosenbach modified the rate upward because Feedback's policies had no deductible. Third, Rosenbach modified the rate upward because Feedback's policies had a higher aggregate limit than the commercial policies. Fourth, Rosenbach modified the rate upwards because the scope of coverage of the Feedback policies were wider than that of the commercial insurance policies. Fifth, and finally, Rosenbach modified the rate upwards because the Feedback policies covered fines and penalties, which were not covered in the commercial policies.
If some of these modifications seem somewhat duplicative, they were, and Rosenbach would later have significant difficulty in justifying some of his modifications to the U.S. Tax Court. Applying these modifications to the "base rate" of commercial policies meant that Feedback would charge many multiples of what the Avrahamis would have paid on the street for commercially-available insurance -- for instance, Feedback charged 26.5 times ($71,000) the "base premium" ($2,675) for its employee fidelity policy.
To litigators, the term "e-mail" stands for "evidence mail" which often tells the true story of what happened in a given case. Here, the story told by e-mail was that Clark was essentially giving Rosenbach what amounted to a "target premium", and then Rosenbach would back in his numbers so that they would come out fairly close to the desired premium.
The IRS provides for two "safe harbors" for captive insurance companies attempting to meet the requirement of risk distribution (an element of "insurance" for tax law purposes). The first safe harbor is the captive underwrite at least 12 insureds and it is fine that the 12 insureds are subsidiaries of the parent company that owns the captive. This is how the vast bulk of large corporate captives meet risk distribution, since they often have dozens if not hundreds of subsidiary companies to insure.
But smaller businesses, such as the Avrahamis', don't have 12 insureds, and thus must look to the other means of providing risk distribution, which is the captive receiving at least 50% (the case law goes down as low as 30%) of its premiums from a third-party. Thus, manager of captives often will set up their own insurance companies, and sell "pooled" insurance to the operating businesses of their own clients. The manager's captive, or "pool", will then buy reinsurance from their clients' captive in about the same amount, and -- Voila! -- each captive then theoretically has taken in 50% of its premiums from a third-party source (the pool).
Which is how it worked here. The Avrahami entities in 2007 paid $360,000 to a company called Pan American Reinsurance Company, Ltd., in St. Kitts ... and the Avrahami's Feedback was paid $360,000 by Pan American for reinsurance. As Private Pyle might say, "Coincidence, coincidence!"
Pan American was owed 47.5% by Sheila Trevors, who was the wife of Robin Trevors, mentioned above as running the captive management company hired by Feedback. Another 47.5% of Pan American was owned by Laurence Mohn, who testified that he was a "courtesy director" because he had some insurance experience in the past but didn't really do anything for Pan American. The remaining 5% of Pan American was split between Diana Gentry and Carl Gentry, who were attorney Clark's children.
The Pan American risk-pooling scheme worked this way: Pan American would sell terrorism risk insurance to the operating business of Clark's clients. Concurrently, Pan American would purchase reinsurance (or cede in insurance parlance) in like amounts from the individual captives of Clark's clients. The money was held briefly in a trust account; ostensibly, to provide the risk pooling and mixing, but more realistically just for show. Clark charged clients an annual, fixed "all-inclusive" fee to participate in the program.
Although the IRS safe harbor for such "third-party risk distribution" is 50% of premiums paid to the captive, Clark thought that she could get away with only 30% of premiums paid to the captive, as this was validated in older case law involving much bigger captives. In actual dollars, this meant that the Avrahamis' jewelry stores ended up paying Pan American $360,000 for $5.25 million in terrorism insurance for the period 2009-2010, and Pan American later paid the $360,000 back over to Feedback to the "reinsurance".
Through this program, Pan American sold terrorism policies to 103 insureds for $20 million in premiums in 2009, and this money was then reinsured back into 85 captives owned by Clark's clients. The following year, 2010, Pan American sold terrorism policies to 139 insureds for $23 million, and reinsured that money back into 101 captives owned by Clark's clients. The money didn't pass all at once, but went in tranches: 50% after 90 days, 47.5% after 180 days, and then a 2.5% "loss reserve" required by Nevis was finally paid to the captives on December 15.
The terrorism risk policies issued by Pan American were similar in appearance to commercial terrorism policies, and like those policies required the certification by the U.S. Secretaries of Treasury and State, plus the Attorney General, that an act of terrorism had occurred which resulted in at least $100 million in losses in a city of less than 1.5 million residents. The Pan American policies offered somewhat broader coverage than the standard commercially-available policies, however. This was offset by some peculiar provisions, such that if Pan American suffered an event of "impaired solvency", instead of paying cash to insureds, it could issue them three-year promissory notes. Also, no individual captive which was reinsuring Pan American could be liable for the share of another captive that either could not, or would not, make up its own share.
The idea behind a risk pool is, of course, that the losses of any one insured would be made up by the larger pool. Despite his companies paying $360,000 per year for this very expensive coverage, Mr. Avrahami would later testify that he thought that Feedback's "reinsurance" liability was limited to $360,000 (i.e., $360,000 premium it received) and that he would "freak out" if Feedback lost more than the $360,000. To put this in perspective, it would be analogous to State Farm receiving a $500 premium for auto insurance, and then "freak out" if it paid more than $500 on a claim.
So how did Pan American set the pricing for its terrorism risk policies? And how did Clark's clients' captives price their reinsurance policies for that same insurance to Pan American? Re-enter actuary Rosenbach who again basically started with the commercial rates for similar policies, and then started modifying the rate steeply upwards without regard to where the company was located (a jewelry store in Phoenix would thus be priced the same as a hotel near ground zero in New York City) or the type of company being insured.
The confidence that the Avrahamis placed in the Pan American policies was amply illustrated by the fact that the Avrahamis continued to buy federally-backed terrorism policies from a third-party commercial insurance company called Jewelers Mutual. Pan American charged $360,000 per year for $5.25 million in terrorism risk cover which was not federally backed; Jewelers Mutual charged about $1,500 for $2 million in terrorism risk cover. A difference is that Pan American also covered chemical and biological hazards, whereas Jewelers Mutual did not -- always a top concern for a jewelry store in Phoenix, apparently.
So, $360,000 of the Avrahami entities' premium dollars went into Pan American for each of 2009 and 2010, and then were reinsured into Feedback. This was more than 30% of the total premiums paid by the Avrahami entities to Feedback, so that risk distribution was ostensibly met. The other 70% or less of the Avrahami entities' premiums were paid directly to Feedback for other policies, being $730,000 in 2009, and $810,000 in 2010. Thus, the Avrahami entities paid -- and deducted -- a total of $1.09 million in 2009 and $1.17 million in 2010. This was against a lost history of those two years of all policies of, well, zero; no claims were made against Feedback during those years.
Taking in big premiums, but paying out no claims, will make an insurance company like Feedback fat and happy in short order, and that is exactly the situation that transpired. But what to do with the dough? Enter Belly Button.
Belly Button was a tax partnership owned equally by the Avrahamis' three adult children, and was created in 2007. Remarkably, the Avrahamis' children, though being the record owners, had no knowledge of what it did or what it owned. In fact, Belly Button owned 27 acres of land in Snowflake, Arizona, which is about 100 miles as the crows flies northeast of Phoenix. The 27 acres were purchased for approximately $2 million, with $1.2 million in cash coming from Mr. Avrahami and the balance in a note carried by the seller.
In 2008, Feedback loaned $830,000 to Belly Button as an investment in "real estate loans", and that money was used to pay the note due to the seller and some interest due. In March of 2010, Feedback loaned another $1.5 million to Belly Button; two days after this loan, the Avrahamis (not their children) transferred $1.5 million from the Belly Button bank account to their personal one.
Later that same year, December of 2010, Feedback transferred $200,000 directly to Mrs. Avrahami, but the transaction was papered as if the money had first gone to Belly Button as a loan. Although related-party loans are usually subject to scrutiny by insurance regulators, the Feedback-to-Belly Button loans were not disclosed to the St. Kitts' regulators until 2014, which was well after the IRS started snooping around Feedback.
Speaking of the IRS, the Service sent Feedback a notice of deficiency in 2013 that challenged Feedback as a valid insurance company for tax purposes, and claimed that Feedback's premium income was simply ordinary income dressed up as insurance payments. Feedback petitioned to the U.S. Tax Court, which then set up the following opinion as to whether the moneys earned by Feedback for 2009 ($1.09 million) and 2010 ($1.17 million) were insurance premiums that were not taxable under IRC § 831(b), or were just ordinary income. It also set up whether the Avrahami entities could deduct their payments to Feedback and Pan American for like amounts. Finally, the case also set up for examination the payments to Belly Button, including the $200,000 paid by Feedback directly to Mrs. Avrahami which had been booked as if it had circulated through Belly Button. I'll get to the Opinion in a moment.
But first, some more Kabuki theatre.
In July 2013, the IRS sent the Avrahamis a document which explained their changes to the returns for the Avrahami entities, including a statement to the effect that the Avrahami entities paid Feedback some $3.9 million in premiums, but Feedback paid no claims, and that the failure to pay claims was one of the nonexclusive factors for determining that a captive insurance company was a sham. As Judge Holmes of the U.S. Tax Court was later to write:
This looks like it triggered something: By March 2013, the claims started rolling in from entities owned by the Avrahamis.
Indeed, after having gone years without paying a single claim, suddenly in 2013 the Avrahami entities suddenly started making claims, which totaled about $150,000 and included such things as $58,248 for "roof repair" and $9,800 for a "ring dispute".
It is here that we find that, despite claiming to be an insurance company, Feedback didn't actually have anything like a formal procedure for dealing with any claims that might be made. Instead, claims were resolved on an ad hoc basis, with attorney Clark herself determining whether it was covered, and then sending the claim and supporting documentation to the captive manager. The latter would then approve the claims, apparently whether they were valid or not (the IRS raised questions about claims still being paid although they were out of time under the "claims made" language of the policies).
These were the facts before Judge Holmes of the U.S. Tax Court in the case of Avrahami v. Commissioner. It is now time to see how he ruled.
Judge Holmes first noted that tax breaks for small insurance companies had been in the Internal Revenue Code since the Revenue Act of 1924. At first, only small mutual-type insurance companies were benefitted, but through the years the exemption for premium income as did the insurance companies to which it applied. By the time the Avrahamis started Feedback, section 831(b) exempted the premium income (but not the investment income) of insurance companies which earned $1.2 million or less annually. This is on the captive side; on the insured side, those paying money for what is properly "insurance" (a term defined by case law and by the Code) may deduct their payments under § 162(c) as ordinary and necessary expenses paid or incurred with a trade or business.
Insurance purchased from a true third-party is usually deductible; self-insurance usually is not. The problem is in the middle, as Judge Holmes noted:
Insurance taxation gets slightly more complicated when the insurer and the insureds are related because the line between insurance and self-insurance begins to blur.
For a captive arrangement to qualify as "insurance", the case law has evolved such that four factors must be met:
(1) Risk Shifting -- The risk is passing in a legally binding fashion from one party to the next;
(2) Risk Distribution -- The risk is distributed among other risks such that it is not a single, or predominantly single, bet against an event;
(3) Insurance Risk -- The risk is of an insurance type (loss or liability, etc.), as opposed to, say, an economic option against rising interest rates; and
(4) The Risk Meets "Commonly Accepted Notions Of Insurance" -- Basically, the risk would be considered "insurance" in the commercial insurance marketplace.
Judge Holmes noted the various landmark court opinions involving large corporations, which opinions legitimized the concept of captive insurance for tax purposes. But did the same rules apply to a captive electing under § 831(b)?
Combining these two concepts--captive insurance companies and the section 831(b) tax advantages for small insurance companies--yields the "microcaptive". In theory, a microcaptive could be run in the manner of the captives in [the large corporate captive cases], albeit on a much smaller scale. But it might also be run so that related parties pay the captive deductible insurance premiums of just under $1.2 million a year. In turn the captive might pay out few if any claims, might make a section 831(b) election so it pays tax only on its investment income, and might quickly build up a large surplus. Then, if the captive were to be licensed and regulated in a jurisdiction with extremely low reserve requirements and loose rules on related-party transactions, it might lend its surplus back to its affiliates. This might generate nearly $1.2 million in tax deductions while arguably only moving money from one pocket to another. Or perhaps the captive could be owned by a Roth IRA, which might mean it could make large dividend payments to its stockholder, creating a form of deductible, yet tax-free, retirement savings. Or perhaps the captive could be owned by its business owner's children or an irrevocable family trust, which might enable the avoidance of future gift and estate taxes.
There may be other variations on this theme; the Commissioner, however, finds none of them pleasing.
In other words, there was a concern that 831(b) captives were not being appropriately used as bona fide insurance vehicles, but were instead were being used as abusive tax shelters. This concern was to permeate Judge Holmes' opinion, though not in so many words.
The IRS argued that the arrangement between the Avrahami entities and Feedback lacked all the criteria to be considered "insurance" for tax purposes, and thus were neither deductible to the Avrahami entities as business expenses, nor constituted non-taxable premiums under § 831(b) to Feedback. In particular, the IRS argued that some of the policies included risks that were not insurable, that Feedback did not have adequate risk distribution, and that the Feedback-Pan American arrangement not really insurance either, because they didn't interact in anything like a true arms' length relationship. This mean, sayeth the IRS, that Feedback was not an insurance company for tax purposes, and thus the money that circulated through Feedback and back to the Avrahamis via Belly Button was simply ordinary income to the Avrahamis individually.
The Avrahamis took issue with all of this, of course, and urged that all the policies met the tax qualifications for insurance, and that Feedback was indeed an insurance company for tax purposes because it distributed its risk through the Pan American deal. As for the money passing through Belly Button, those were simply loans made by Feedback to Belly Button, except for the money which passed directly to Mrs. Avrahami which was simply an honest accounting error by their CPA.
So, Issue No. 1: Were the Feedback policies issued to the Avrahami entities contracts of "insurance" for tax purposes? Judge Holmes here applied the four factors of insurance (risk shifting, risk distribution, insurance risk, and meets commonly accepted notions of insurance) to the Feedback policies.
To start with risk distribution, Judge Holmes noted that this
occurs when the insurer pools a large enough collection of unrelated risks. * * * The idea is based on the law of large numbers--a statistical concept that theorizes that the average of a large number of independent losses will be close to the expected loss.
But how much pooling of risk is enough? The Avrahamis claimed that Feedback issued seven different types of insurance to the Avrahami entities, and then also distributed risk by insuring over 100 diverse third parties that were geographically spread throughout the United States. The IRS retorted that these numbers weren't big compared to the opinion on which the Avrahamis relied, Harper Group v. Commissioner, insofar as the pool in that case involved 7,500 unrelated insured. Moreover, claimed the IRS, the Pan American program involved relatively small and independent risks in its terrorism policies.
It was clear that if Feedback simply insured the Avrahami entities (three in 2009 and four in 2010) that risk distribution was not present. But even beyond that, wrote Judge Holmes:
We also want to emphasize that it isn't just the number of brother-sister entities that one should look at in deciding whether an arrangement is distributing risk. It's even more important to figure out the number of independent risk exposures.
(Italics in original)
Thus, R.V.I. v. Commissioner, risk distribution was present because the captive not only insured 714 different unrelated parties, but through 951 policies covered 750,000 vehicles and 2,000 parcels of land, plus another 1.3 million in various equipment in seven geographic regions. In Rent-A-Center v. Commissioner, risk distribution was present where the captive insured workers compensation for 14,000 employees, and provided other insurance for 7,100 vehicles and 2,600 stores nationwide. And, in Securitas v. Commissioner, risk distribution was present where 25 entities were insured and more than 300,000 employees were covered by workers compensation policies, with other insurance provided to 2,250 vehicles and sundry other coverages to 25 separate entities.
These three cases were examples of adequate risk distribution. Feedback's covering of three jewelry stores, 2 key employees, 35 other employees, and three commercial real estate properties in the Phoenix area was not. The bottom line was that Feedback insurance activities did not cover a sufficient number of risk exposures to rise to the level of risk distribution for tax purposes.
This left the Pan American deal as the Avrahamis' last hope to meet risk distribution. But was Pan American a bona fide insurance company?
First, the Feedback-Pan American arrangement reeked of a circular flow of funds. Every year, $360,000 went into Pan American, and every year, $360,000 came out of Pan American and went into Feedback. In the end, $720,000 left the Avrahamis left pocket and ended up in the Avrahamis right pocket -- the very essence of a circular flow of funds.
Second, the Pan American premiums were unreasonably high. Actuary Rosenback modified the commercial rates upwards in a way that made little sense. Although federally-backed terrorism insurance can be as much as 10 times more expensive in terrorism-target cities like New York or Washington, Pan American was charging the same rates to its policyholders everywhere. Further, the Pan American insureds were being charged the same rate even if they already had federally-back terrorism insurance, which would have paid off first, before Pan American's coverage kicked in. Worse, even though Pan American's policies were somewhat more inclusive than that charged by the federally-backed terrorism policies, the rates charged by Pan American for its policies were not in the same ballpark, zip code, area code, or even time zone as those charged for the federally-backed insurance policies. The IRS's expert pointed out that the commercial rates for terrorism insurance worked out to $24 per million of total insurable value, but the Pan American rates were $45,000 per million of total insurance value. Thus, Pan American's premiums were found to be not only excessive, but "grossly excessive".
Third, there was substantial evidence that the Pan American relationship with the Avrahami entities was something other than arm's length. Although Pan American was charging exorbitant premiums for its terrorism policies, they were not federally-backed, and if the insured lived in a city of less than $1.5 million population, the Secretaries of Treasury and State plus the Attorney general all had to certify that an act of terrorism had occurred which resulted in over $100 million in losses. (In other words, had Al Qaeda committed a biological-weapons attack only against the Avrahamis' jewerly stores, the Pan American policies still would not have paid off.) Even actuary Rosenbach admitted that he could not identify a single event in history that would have met with these requirements. Further, it was highly questionable whether Pan American could even have paid these losses, since it paid back all of its reinsurers 100% of their premiums and only kept $500,000 on hand to pay claims.
Fourth, Pan American didn't act like a bona fide insurance company. Because insureds would have looked first to Pan American for payment of claims, Pan American had to assume the credit risk that its own reinsurers (like Feedback) would not fork over when the time came. Real insurance companies charge a "ceding commission" (5% is a common minimum) for taking on this credit risk; Pan American did not. Instead, attorney Clark simply charged all the participants in the pool a flat fee of $5,000 per year, which didn't even begin to cover the credit risk that Pan American was undertaking.
For these reasons, Pan American was not a bona fide insurance company for tax purposes. Thus, the Pan American arrangement did not satisfy risk distribution, and so therefore Feedback did not itself satisfy the risk distribution requirement either.
Judge Holmes also considered an alternative ground for risk distribution, which is that the arrangement looks like insurance in the commonly accepted sense, i.e., whether Feedback was operating like most other insurance companies.
Here, the first question was whether Feedback was organized, operated, and regulated as an insurance company. Oh sure, Feedback met the minimum requirements to be organized and licensed as an insurance company under St. Kitts law -- all the basic paperwork check out. The problem was in Feedback's operations. Feedback didn't have a policy or procedure for handling any claims that might appear, but that was handled only on an ad hoc basis. Plus, said Judge Holmes:
And we will not overlook the fact that the Avrahami entities made no claims whatsoever against Feedback from its inception in 2007 until March 2013--two months after the IRS sent the Avrahamis documents about the audits of the returns of [the Avrahami entities] that suggested Feedback was a sham.
(Italics in original)
Moreover, the claims that were belatedly paid were made outside the time limits of the policies, and were approved even though some of the circumstances were highly questionable. For instance, Feedback paid the Avrahamis' claim for hail damage even though the Feedback policy was an excess policy that would pay only after the Avrahamis made a claim against their existing commercial property damage policy, and apparently they didn't.
Further, Feedback's investments were in long-term unsecured loans to a related party, being Belly Button, and Feedback didn't bother to get regulatory approval for those loans, even though St. Kitts law required that approval. With 65% of its assets being in the unsecured loans to Belly Button, if Belly Button would have gone bust, then Feedback would have had trouble paying any significant claims that materialized. The bottom line was that Feedback wasn't operated like an insurance company, despite all the corporate formalities, and that it did meet the almost de minimis capital requirements of St. Kitts law.
Another factor was that the insurance policies issued by Feedback be "valid and binding". While the policies contained the basic language needed for an insurance policy, some of the policies were shoddily drafted. Some policies were stated to be claims-made policies on their face, but language in the policy indicated that there were instead occurrence policies.
This now brings the Court to revisit the issue of how the policies were priced; namely, were the premiums reasonable and the result of an arm's-length transaction. Here, the Avrahamis relied on attorney Clark and actuary Rosenbach, but:
Clark did hire Rosenbach to help price the policies, and he testified extensively about his process; but because he priced captives only for Clark and his explanations were often incomprehensible, we don't find them persuasive. We find instead that the premiums were utterly unreasonable.
Another problem was that before the Avrahamis started Feedback in 2006, their businesses were spending in the neighborhood of $150,000 per year for insurance. After Feedback was formed, the Avrahami entities went to paying $1.1 million in premiums in 2009 and $1.3 million in 2010 -- all while maintaining their commercial insurance coverage in place for about $90,000 per year.
Judge Holmes found that Rosenbach's work largely amounted to the actuarial version of GIGO a/k/a "garbage in, garbage out". To give one example, in deriving the base rate from Chubb's (a commerical carrier) pricing, Rosenbach chose a higher hazard group for two of the Avrahami entities than they should have been place, which unnecessarily made Feedback's premiums more pricey. Rosenbach then assumed that all of the Feedback policies were retroactive policies (that is, they would cover old claims which happened before the policy inception), but in fact several of the policies were claims-made policies which limited coverage to events occurring during the policy period. This jacked the premiums higher still, quite unnecessarily. Further, in commenting upon Rosenbach's work for one of the Avrahami entities, Chandler One, Judge Holmes noted:
Apart from the factors suggested in the Chubb filing, Rosenbach also made adjustments based on his professional judgment--most without a coherent explanation. For example, he increased Chandler One's 2009 Business Risk Indemnity premium by $2,500--$1,000 in 2010--for providing excess general-liability coverage but offered no rationale for the amount chosen. In the same calculation, Rosenbach also increased the premium by more than $26,000--and more than $62,000 in 2010--for "construction defects" coverage. But Chandler One wasn't constructing anything; it just owned and leased commercial real estate. Rosenbach also struggled to explain the premium calculation for the Litigation Expense policy, which included a subjective "base exposure" and several subjective factors that Rosenbach testified were multiplied together to reach the premium amount. As the Avrahamis had to explain in their reply brief, however, to reach the premium that Rosenbach calculated the base exposure has to be multiplied by all of the factors except the expense ratio and then divided by one minus the expense ratio. In 2010 Rosenbach came up with a nearly 60% higher premium for the Litigation Expense policy because he "appl[ied] the model in a little different light." Similarly, Rosenbach was unable to explain the 7.5% event rate used in the Tax Indemnity calculation or even what it represented. Without a comprehensible explanation we can't find these premium amounts justified.
(Italics in original)
Another example was Rosenbach's premium calculations for the Loss of Key Employee policy, which would pay in the event of a reduction of income of another Avrahami entity, called American Findings, in the event that a key employee who was not a shareholder left the business. Rosenbach made his calculations based on what would happened if Mr. and Mrs. Avrahami left the business, but they were the only shareholders, i.e., they were the very persons who were excluded by the policy! Thus, not only was Rosenbach's method for pricing premiums horribly and fatally flawed, but the whole process was basically shambolic.
Next was whether Feedback was acting as an insurance company in administering claims, and the answer here was a very cynical "yes", but only once the IRS had noted the absence of claims.
Thus, Judge Holmes was to conclude:
Does this add up to "insurance in the commonly accepted sense?" We find that the answer is "no". Although Feedback was organized and regulated as an insurance company, paid the claims filed against it, and met the minimal capitalization requirements of St. Kitts, these insurance-like traits cannot overcome its other failings. It was not operated like an insurance company, it issued policies with unclear and contradictory terms, and it charged wholly unreasonable premiums.
Because what Feedback was selling, and what the Avrahami entities were buying, was not "insurance" for tax purposes, it was unnecessary for the Court to consider whether these transactions had the additional requirements of risk shifting and involving insurance risk. For the same reason, the Court did not need to consider other challenges made by the Service, such as economic-substance, substance-over-form, and step-transaction doctrines.
The holding that Feedback was not an insurance company for tax purposes had two immediate ramifications. The first was that the 831(b) election was not available to Feedback, since that election is only available to valid insurance companies. The second was that the 953(d) election -- for a foreign insurance company to be taxed as a domestic one -- was similarly not available because Feedback was not an insurance company. The latter ruling meant that Feedback was essentially a Foreign Controlled Corporation (CFC), and would have had utterly disastrous consequences for the Avrahamis, but prior to trial the Avrahamis and the Service stipulated to the effect that in the event the Avrahamis lost, Feedback would not be taxed as a foreign corporation.
There were also ramifications for the Avrahami entities; namely, that since they weren't buying insurance, their payments to Feedback were not ordinary and necessary business expenses under § 162(a) and were thus not deductible by those companies.
We then arrive once again at the Belly Button deal, recalling that Feedback transferred $1.5 million to Belly Button, and then two days later Belly Button transferred the $1.5 million to the Avrahamis' personal accounts. The Avrahamis claimed these were loans, and at any rate $1.2 million was used by Belly Button to repay Mr. Avrahami for his loan of that amount to Belly Button. The Service claimed they were taxable distributions. Commented Judge Holmes:
This is a common chew toy in tax litigation. How do we distinguish bona fide loans from distributions?
The Court looked at whether the parties intended to establish a creditor-debtor relationship, and other factors such as whether the borrower had the money to repay the loans. While the Court was skeptical of the loans, in the final analysis it decided to recognize them as valid loans. There was, however, a difference of $300,000 between the $1.5 million loaned by Feedback to Belly Button, and the $1.2 million which went to repay Mr. Avrahami for his original loan for that amount to Belly Button. The Court treated half of that $300,000 as interest, and the other half as a taxable distribution to Mrs. Avrahami.
The last issue was penalties. With regard to the premiums paid to Feedback, the Avrahamis could avoid penalties if they could show that they reasonably and in good faithrelied on the advice of a competent tax professional who wasn't a "promoter". The Avrahamis urged that his included attorney Clark, accountant McEntee, and estate planning attorney Hiller.
The problem with attorney Clark is that she was clearly a promoter of the transaction, and profited by both her original $75,000 fee and subsequent fees. But McEntee and Hiller were not, even though Hiller received a portion of Clark's $75,000 fee, and also assisted the Avrahamis with the Belly Button loans.
But was the Avrahamis' reliance in good faith? Here, the Avrahamis got a pass simply because they were first case to address so-called "microcaptive" transactions:
This is a case of first impression--we have discovered no other case addressing microcaptives and the interplay among sections 162, 831(b), and 953(d). And we have previously declined to impose accuracy-related penalties when there is no clear authority to guide taxpayers.
The Court then turned to the Belly Button loans. Although the Court found those loans to be legitimate, there was the $300,000 difference between the $1.5 million which Feedback loaned, and the $1.2 million which was used to repay Mr. Avrahamis' original loan to Belly Button. The Court found that this $300,000 should have been reported but was not, and thus imposed penalties under § 6662(a).
Finally, there was the $200,000 which passed directly from Feedback to Mrs. Avrahamis, that the Avrahamis tried to book through Belly Button. Judge Holmes had little sympathy for the Avrahamis not reporting this transaction, and so assessed § 6662(a) penalties on it as well.
And that was the end of a very long but very useful opinion.
ANALYSIS
Since this Opinion came out, the Avrahamis moved for reconsideration, urging the Court to reverse its determination that Feedback was not an insurance company and that the policies were claims made policies and not occurrence policies. I wouldn't hold my breath, and at least to this litigator, the Motion for Reconsideration looks less like an attempt by the Avrahamis to actually change the result than it does to get a supplementary opinion out of Judge Holmes that somebody could use in a professional negligence case.
Speaking of which, since this Opinion came out, a memo has floated about purporting to be from attorney Clark to her clients stating that she was getting out of the captive business. Draw your own conclusions from this one.
Also since this Opinion came out, the 831(b) risk-pooled captive sector (i.e., captive managers and associated professionals who run their own risk pools for 831(b) captives), have attempted to portray the facts of this case as extreme and an outlier. The problem with this claim is: It is just flat wrong. I've reviewed a number of risk pools that were every bit as dubious as Pan American, policies that were absurd as to both their terms and their pricing, and had big related-party loans such as were made to Belly Button.
The truth is that the standard-of-practice for 831(b) risk-pooled captives has been very bad. It is not very different than in the "acronym tax shelter" days of the late 1990s (when CARDS, BLIPS, Son of BOSS, and other tax shelters were being sold by major CPA firms like so much fast food), where transactions are being poorly papered and the real faith of the participants was in low audit rates. There are risk pools out there which have never had a year of significant losses, policies that seem to have been drafted as a third-grade class project, and actuarial studies that have only the most existential connection to Planet Earth. The good news is that most of the purveyors of this sort of captive junk are already under promoter audits, or have numerous of their clients lined up in the U.S. Tax Court, and presumably someday they will all just move on to their next tax shelter and let 831(b) captives be.
Which is to say that not all 831(b) captives are bad -- the truth is quite otherwise. There are a lot of completely legitimate captives out there which have made the 831(b) election that are not under audit and have utterly no reason to suspect that the IRS will ever challenge them. These are 831(b) captives that were actually formed for insurance, not tax, reasons and include captives as diverse as farm mutuals in the Midwest, captives that are used to support warranties, workers compensation captives, captives used to cover construction defects for real estate builders, and many captives set up to cover the extraordinary but very real insurance needs of businesses across the U.S.
The fundamental problem with the Avrahamis case was that they attempted to meet risk distribution through the risk pool (Pan American), but that risk pool in reality was simply an account where money was temporarily parked between the Clark's clients' businesses and those clients' captives. This flaw with risk pools is not unique to Clark or Pan American. Again, I've seen other pools that were just as bad, if not worse.
Heck, I've seen deals where there was no intermediary insurance company like Pan American, but the captive manager (being cheap) simply circulated around hinky agreements between clients whereby they promise to cross-insure each other, and each other's captives. Good luck defending that. Even worse, I engaged an expert witness in one case where the captive manager didn't even realize that its clients needed risk distribution to qualify as insurance companies for tax purposes -- this is the sort of junk that is out there, and the IRS should quite rightly be cleaning up.
But just as not 831(b) captives are bad, not all risk pools are bad either. Some risk pools actually provide insurance, have claims, and are administered in every sense as insurance companies. These risk pools are certainly much more defensible than the bad ones. It is easy enough to spot the bad pools, and easy enough to spot the good pools. Where it gets difficult is where everything gets difficult, meaning in the middle. Smarter folks than me are going to have to sort those gray-area pools out.
One thing is clearly evident from this Opinion, whether the industry likes it or not: The pricing of premiums for pooled insurance must over time correlate in some reasonable fashion to the pool's loss history. Time-after-time, I have reviewed pools where there was an expected frequency of claims of let's say 10 for a given risk in a given year, and premiums were based upon 10 claims per year. In the first year, there are no claims (though 10 were expected) and yet the premiums remained the same. Even after five years, when by that time 50 claims should have materialized, there might be only one or two claims, yet the premiums remain the same.
What is really going on is that the captive managers are thinking, "If I reduce the price the premiums from the pool, then how will my clients get enough premiums into their captives to satisfy risk distribution?" Whatever else that last thought is, it is not a legal justification for keeping premium levels artificially high because the real-world loss history is being ignored. That practice has to stop; it is a loser.
For the bad risk pools, being those like Pan American, the Avrahami opinion very likely signals an extinction-level event. Folks looking for a tax shelter don't want to actually take on significant amounts of risk, and thereby the potential for significant losses, with third-parties that they don't know. Instead, they want an arrangement where their money runs around the barn without much possibility of economic loss, and then they get a deduction for running the money around the barn. That's what most if not all of the captive managers now under a promoter audit were doing, for a sizeable fee of course, but the Avrahami opinion now tells us that doesn't work. If you want a risk pool that works, you actually have to risk all your chips that go into the pool to a significant possibility of loss because of other insureds' claims, but folks that just want a tax shelter aren't going to go for that. They don't want that risk, and in fact they don't want any real risk. For an example, look at Mr. Avrahami, who testified that he would "freak out" if he suffered losses through Pan American.
The other thing to note is that -- as I have warned for over a decade -- if any one participant in the pool blows up because the pool was not deemed to be insurance, then every participant in the pool blows up. Look what happened here: The U.S. Tax Court deemed Pan American not to be an insurance company for tax purposes. But this doesn't affect just the Avrahamis, but it also takes down the other 100 or so participants in the Pan American risk pool. I've also long said that "pools are for fools". It is 100 folks standing in a tank of gasoline during a thunderstorm, each holding a lighting rod, and each praying that nobody gets hit. Today, there are about 100 more fools as a result of the Pan American debacle, all from just one successful IRS challenge against a single participant in that pool
Also remember that the IRS will not challenge the best-looking captive, but instead can cherry-pick the absolute ugliest, most egregious case to drag before the U.S. Tax Court. The problem is that via the pooling arrangement, the bad apple can take everybody else in the barrel down.
EXODUS 831(b)
This things brings us to the end game for a lot of existing 831(b) captive clients, particularly those whose captive managers are under promoter audits: How to exit a bad 831(b) captive?
In ordinary times, exiting an 831(b) captive merely meant surrendering the company's license to regulators, distributing the captive's assets to its shareholders, winding up the company, and filing a final return. But these are not ordinary times.
The problem is that once the IRS shows up, a fundamental conflict-of-interest arises between the captive owners and the captive manager: namely, the captive manager wants to cut off liability in the event that the IRS is successful in its challenges, and the captive owners want to preserve their own rights to sue the captive manager and affiliated professionals (the actuaries, auditors, and tax advisers) in the event they get slammed with penalties. In fact, if the captive arrangement is deemed not to be insurance, and the captive owner loses the tax benefits pitched by the captive manager, the captive owner will probably want all their fees returned too. "Hey, isn't that what I was paying you for all along, to make sure that my captive qualified as an insurance company for tax purposes?"
Thus, it is now common for captive managers to demand releases of liability for their negligence (and the negligence of their affiliated professionals) as a precondition to exiting their captive program. This demand is usually also accompanied by the captive manager basically holding their pooled money hostage until the release is signed. Note that this is different from a release of liability given by the client's operating business to the pool for any policies that are being cancelled before their term expires, and which are relatively benign and would be expected. Instead, the egregious releases of liability are those which attempt to exculpate the captive manager and associated professionals from being sued for malpractice for putting together a captive arrangement that failed to constitute insurance from a tax perspective.
Suffice it to say that the captive owners will probably require legal assistance by counsel who are experienced in this area to get them out of their captives while still preserving all their rights. But getting an independent review of the entire arrangement and how it was operated was never the worst idea anyway. The bottom line: Don't sign any waivers of liability unless and until you have consulted with independent counsel who are familiar with this issue first.
Avrahami is just the first of a veritable tidal wave of 831(b) captive cases before the U.S. Tax Court, there are others that have been tried and are awaiting decision (most notably Wilson v. Commissioner and Caylor Land v. Commissioner), and there are literally dozens of other cases awaiting trial.
Stay tuned.
CITE AS
Avrahami v. Comm'r, 149 T.C. No. 7 (Aug. 21, 2017). Full Opinion at https://captiveinsurancecompanies.com/2017-avrahami-tax-court-captive-insurance-celia-clark.html
This article at https://goo.gl/G5PZ5n
