The Courts recently made a decision regarding the Reserve case for the IRS. This is the second case of late that has been decided against a captive owner in an effort to crack down on captives the IRS perceives to have a poor fact pattern, and therefore cannot meet insurance tax treatment standards. For example, captives that have been set up to undertake sham transactions or undertaking transactions that do not meet the bona fide insurance company characteristics would fall into this category.

According to the Avarhami v. Commissioner (“Avrahami”) judgement, the court provided four criteria that result in an arrangement constituting insurance.  These four criteria were also addressed in the Reserve Mechanical Corp (“Reserve”) v. Commissioner case, and are as follows:

  • The arrangement must involve insurable risk
  • The arrangement must shift the risk of loss to the insurer
  • The insurer must distribute the risk among its policy holdersIRS captives
  • The arrangement is insurance in the commonly accepted sense

Reserve outlined these four non-exclusive criteria to establish a framework for determining the existence of insurance for federal income tax purposes.  The court’s opinion focused on the idea of risk distribution, which led to investigating PoolRe Insurance Corp. (“PoolRe”), the stop loss insurer for Reserve. The judgement discusses the transaction in detail and stated there was a circularity of funds that invalidated the pooling arrangement.

To determine if a captive insurer has met the risk distribution criteria as a standalone captive without stop-loss or reinsurance protection, the courts looked at the total number of insureds and the total number of independent risk exposures. It has long been believed that the “law of large numbers” allows an insurer to minimize its total risk and reduce the likelihood of a single claim exceeding the premium received. In the Avrahami and Rent-A-Center court cases, risk distribution passing and failing thresholds have been observed as follows:

  • Rent-A-Center ultimately showed distribution of its risk by insuring the risk for 14,000 employees (workers’ compensation), 7,100 vehicles (auto coverage), and 2,600 stores (general liability coverage) in 50 states
  • Avrahami didn’t show distribution of risk by insuring 3 jewelry stores, 2 key employees, and 35 total employees. Further, one of the stores had 5 low frequency coverages and the other 2 stores had 2 low frequency coverages

Reserve, an Anguilla-domiciled captive, wrote 11 to 13 policies over the three tax years in question and had direct policies for 3 insureds.  Peak Mechanical & Components, Inc. (“Peak”), an S Corp for Federal income tax purposes, was owned in equal 50% shares by two individuals and was the primary insured under all policies. The policies were also issued to two other subsidiaries, although the operations were not significant. Peak operated two facilities and had a max of 17 employees. Reserve did not meet risk distribution based on this exposure profile alone; its exposures were similar in scale to the Avrahami’s.  Reserve contended that it still met the risk distribution safe harbor requirements, by having 30% of its gross premium for each of the tax years for unrelated parties via the reinsurance agreement with PoolRe.  A similar argument was made in the Avrahami case with their reinsurance pool.

Before it is determined whether Reserve distributed risk through the agreement with PoolRe, they evaluated whether PoolRe was a bona fide insurance company. In the eyes of the court, a captive should be able to answer “yes” to each of these questions and provide adequate support to:

  1. Is there no circularity to the flow of funds?
  2. Are the policies developed in an arm’s length approach?
  3. Did the captive charge actuarially-determined premiums?
  4. Does the captive face actual exposures and insurance versus business risk?
  5. Is the captive subject to regulatory control, and did it meet minimum statutory requirements?
  6. Was the captive created for non-tax business reason?
  7. Was a comparable coverage in the market place more expensive, or even available?
  8. Was it adequately capitalized?
  9. Were claims paid from a separately maintained account?

The court’s conclusion in Reserve’s case provided details of the concerns with evidence in support of the first six questions listed above, and concluded that the PoolRe quota share arrangement provided the appearance of risk distribution without actual risk distribution.  The court summary also highlighted the following:

  • Circularity of funds was exhibited with PoolRe receiving and distributing the same amount of money to Reserve
  • There was no evidence that the premium payments to PoolRe by Reserve and the other participants were determined by actuaries
  • Contracts were not determined in a like manner, nor using objective criteria

One of the major concerns the IRS addresses with the Avrahami and Reserve cases is that a one-size-fits-all rate for all participants in the pool/reinsurance agreement isn’t valid.  The court also addressed an alternative ground for the case, which would have been to evaluate “Insurance in the Commonly Accepted Sense”. To determine if an insurance arrangement exists, the following factors come into play:

  • Was the insurance company organized, operated and regulated as an insurance company?
  • Was it was adequately capitalized?
  • Were the policies valid and binding?Captive Checklist
  • Were the premiums reasonable and a result of an arm’s-length transaction?

The court summary pointed out several issues in Reserve’s support for answering the above questions, such as:

  • Reserve had no employees of its own performing services and the board members did not know how claims were made or handled.
  • There’s no evidence that activities were performed in its domicile.
  • Claims must have supporting documentation, yet there was no addendum for the program until after the policy date. An employee from the insured signed the checks as opposed to the insurer.
  • Binding and valid policies – policies must, at a minimum, identify the insured.
  • There were peak paid commercial market premiums of $95,828 in 2007 versus $412,089 to Reserve in 2008. This is a 330% increase in insurance premiums. In addition, Peaks premiums vary from year to year with no explanation.
    • Note, the Avrahamis similarly had significant increases in premium spend, with almost an 800% increase over a several year period.

These cases provide us and other captive professionals with guidance and clarity. As the industry grows, cases like these will form the cornerstones of how to properly operate and conduct business as a captive insurance company.

Key Takeaways

The IRS clearly has problems with some of the pooling structures used to qualify captives as meeting the risk distribution safe harbor tests. They are concerned that the premiums ceded to the pool and the transfer of risk into the insurance pool are not commensurate with one another, and that the pools are only being utilized to circle premiums to the captive participants, with each assuming no or minimal losses from the pool.

There should be clear documentation of premium determination by an actuary, illustrating why premiums are reasonable and that sufficient risk transfer exists.  Over time, if the total loss experience and premium ceded to the pool doesn’t produce a long-term average loss ratio consistent with the commercial marker, then the pool’s support in having arm’s-length contracts with each of the participants becomes weak and difficult to defend.  Long term average commercial market loss and loss adjustment expense (“LAE”) ratios for most lines of business generally fall in the range of 50% to 75%, hence the 70% loss and LAE ratio threshold in IRS Notice 2016-66 used to identify captive transactions of interest.

Finally, it is important to show sufficient support in a captive’s business plan, policies, and feasibility studies to address the questions above about an insurer/pool being a bona fide insurance company.

The following two tabs change content below.
Peter Johnson

Peter Johnson

Peter Johnson, FCAS, MAAA is a Senior Actuarial Consultant with Spring Consulting Group, LLC and Spring’s Property & Casualty Practice Lead. He has almost 15 years of actuarial experience in reserving, pricing, alternative risk transfer and reinsurance risk transfer work. This experience includes workers’ compensation, medical professional liability, professional liability, automobile, general liability, cyber liability, and mortgage insurance. Prior to joining Spring, Peter was the President and Consulting Actuary with Bartlett Actuarial Group and a Consulting Actuary with Milliman. Peter has given presentations, published articles and served on various committees in both the Casualty Actuarial Society and captive insurance industry. Peter earned a B.S. in applied math and computer science from the University of Wisconsin - Stout. He is a fellow of the Casualty Actuarial Society and a member of the American Academy of Actuaries.