Tax Notes logo

Reserve Mechanical: Microcaptive Insurance Arrangement Denied on Appeal

Posted on June 27, 2022
Bettina Xue Griffin
Bettina Xue Griffin
Benjamin Alarie
Benjamin Alarie

Benjamin Alarie is the Osler Chair in Business Law at the University of Toronto and the CEO of Blue J Legal Inc. Bettina Xue Griffin is a senior legal research associate at Blue J Legal.

In this article, Alarie and Griffin explore the relative merits and prospective strengths of three of the grounds of appeal advanced by the taxpayer in Reserve Mechanical.

I. Introduction

In our monthly Blue J Predicts column, we use tax research software to analyze pending or recently decided federal income tax cases. This month we analyze the judgment of the Tenth Circuit in Reserve Mechanical,1 which addressed whether a microcaptive insurance arrangement constituted insurance so that the premiums received were exempt from taxation under section 501(c)(15).

Several months ago, we examined the Tax Court’s decision in Reserve Mechanical and predicted with 77 percent confidence that the taxpayer’s appeal would be dismissed and the Tax Court’s decision affirmed.2 That prediction was correct.

In its decision, the Tenth Circuit reiterated that the outcome in Reserve Mechanical should not be regarded as an indictment of all microcaptive insurance arrangements. Instead, the particular facts of the case indicated that the taxpayer was not engaged in the business of insurance. Still, we expect that the result in Reserve Mechanical will have a significant effect on the captive insurance industry. Notably, since the release of the decision, the IRS has issued a statement warning taxpayers to beware of microcaptive insurance arrangements.3 The IRS said, “Taxpayers should be alert to these schemes, normally peddled by promoters, as they will ultimately cost them.” It added that “these transactions will result in serious economic loss to taxpayers, including the loss of deductions, required income inclusion, and penalties.”4 This is not new guidance: Since 2015 the IRS has included abusive microcaptive insurance arrangements on its “Dirty Dozen” list.5

In this analysis, we explore the relative merits and prospective strengths of three of the grounds for appeal advanced by Reserve Mechanical Corp. (Reserve). As it happens, all three were rejected by the Tenth Circuit. First, Reserve argued that the Tax Court erred in classifying the insurance arrangement as one of “excess” insurance only. Second, it claimed that the Tax Court erred in assessing the bona fides of a reinsurer who is not the taxpayer. Third, Reserve argued that the Tax Court erred in determining that the premiums were unreasonably high. We explore the key factors underpinning each of these grounds and use the insurance arrangement predictive module in Blue J Tax to assess how significant the three specific factual characterizations were to the overall decision.6 We then perform a series of distinct analyses, varying the mix of assumptions to identify scenarios that could have in principle led to Reserve succeeding in its appeal.

There are two critical takeaways from our analysis of the facts in Reserve Mechanical. First, the most significant factor driving the result in the government’s favor was whether captive insurance premiums were set using objective methods, such as actuarial data. Second, the factors with a more modest effect were whether the insured maintained complete coverage from other insurers and whether there was a circular flow of funds.

II. Background

A captive insurance company is generally a wholly owned subsidiary that is established to insure the parent and, in many cases, the parent’s other subsidiaries. In the right circumstances and with sound tax planning, these insurance companies stand to obtain favorable tax treatment under section 501(c)(15). This provision creates a tax exemption for insurance companies if the gross receipts for the tax year do not exceed $600,000 and more than 50 percent of the gross receipts consist of premiums.

For the purposes of captive insurance arrangements, the terms “insurance” and “insurance company” are not expressly defined in the IRC.7 Consequently, to characterize insurance activities and whether a particular entity constitutes an insurance company, reference must generally be made to the common law. The common law has established four nonexclusive criteria to determine whether a particular arrangement constitutes insurance for federal income tax purposes: (1) the arrangement involves insurable risks, (2) the arrangement shifts the risk of loss to the insurer, (3) the insurer distributes the risk among its policyholders, and (4) the arrangement is insurance in the commonly accepted sense.8

III. Facts of Reserve Mechanical

From 2008 to 2010, Reserve issued insurance policies to Peak Mechanical Corp. (Peak). Reserve and Peak were owned by the same two individuals, Norman Zumbaum and Cory Weikel. Zumbaum and Weikel engaged the services of Capstone Associated Services Ltd. (Capstone) to form and administer Reserve as a captive insurer to Peak. Capstone consulted for and managed many captive insurance companies besides Reserve. Capstone handled various technical and management issues, such as preparing policies and recommending premiums. Before Reserve issued policies, Peak had commercial insurance policies from third-party insurance companies that cost approximately $100,000 a year. Peak maintained those policies and purchased supplemental insurance policies through Reserve that cost more than $400,000 a year.

Capstone took the position that for Reserve to demonstrate it had adequately diversified its risk and to thereby be able to potentially qualify as a valid insurance company, 30 percent of the premiums it received had to come from companies that are not affiliated with it. This 30 percent threshold had been the lowest judicially approved level of unrelated premiums for captive insurers to demonstrate risk distribution in the case law.9

Figure 1

To achieve this level of diversification, Capstone created a reinsurance pool called PoolRe Insurance Corp. (PoolRe). In a reinsurance relationship, one insurance company will cede some of its liability to another insurance company. The reinsured insurance company pays a premium to the reinsurer, and the reinsurer assumes a portion of the reinsured company’s liabilities. Through PoolRe several dozen captive insurance companies under Capstone management agreed to reinsure policies issued to each other. For a visual representation of the relationship between the relevant parties, see Figure 1.

PoolRe received a fixed percentage (18.5 percent the first year) of the premiums paid on the policies issued by the captive insurance companies. The captive insurance companies in turn reinsured all of PoolRe’s stop-loss coverage, with each captive receiving a premium from PoolRe equal to the premium its insured paid to PoolRe. Through this arrangement, all the captive insurers shared PoolRe’s risk of the stop-loss coverage. If one of the captives incurred liability that triggered the stop-loss coverage, PoolRe would pay its share of the loss but would be fully reimbursed through the reinsurance it obtained from the captives, with each captive paying its proportionate share. The net result was that each captive received the full premium paid by its insureds — the 81.5 percent paid directly to the captive by the insured plus the 18.5 percent paid to PoolRe, which in turn later paid that amount to the captive. See Figure 2 for a visual representation of the reinsurance pool.

Figure 2

IV. Recap of the Tax Court’s Decision

The Tax Court found that Reserve’s transactions with Peak were not insurance transactions and that Reserve was thus not entitled to a federal income tax exemption under section 501(c)(15). Specifically, Reserve failed to satisfy the third and fourth criteria of the common law test for insurance arrangements: It did not distribute risk among its policyholders, and the arrangement was not insurance in the commonly accepted sense.

In concluding that Reserve did not adequately redistribute risk, the Tax Court highlighted the following:

  • Reserve did not pool a sufficiently large number of unrelated risks through its direct insurance policies. While Reserve issued policies to three insureds, the primary insured was Peak. The operations of the other two related insureds were relatively insignificant.

  • The quota-share arrangements did not redistribute risk because they were not bona fide insurance transactions and PoolRe was not a bona fide insurance company.

  • There was a circular flow of premiums between Peak, Reserve, and PoolRe.

In concluding that Reserve did not have the hallmarks of an insurance company, the Tax Court highlighted the following:

  • The premiums charged were unreasonably high and appeared to be priced in ways that were difficult to justify economically. For example, there was no explanation for why a policy for four years with greater coverage cost only approximately $15,000 more than a policy for one year with half the coverage. Occasionally, Peak spent more for one month of insurance coverage than a year of coverage.

  • The premiums did not appear to be determined by actuarial methods or other objective methods.

  • The initial feasibility study prepared by Capstone provided no information on the probability of a loss event.

  • Only one claim was ever made under Reserve’s policies, and it was handled in an irregular manner. There was no investigation, no supporting documentation to substantiate the occurrence, and apparently no significant discussion concerning the amount of the claimed loss.

V. The Tenth Circuit’s Decision

We focus on three arguments the taxpayer advanced on appeal (1) that the Tax Court erred in classifying Reserve’s policies to Peak as “excess” insurance, (2) that the Tax Court erred in assessing the bona fides of PoolRe, and (3) that the Tax Court erred in finding that the premiums paid under the policies were unreasonably high. The Tenth Circuit rejected all three arguments.

A. ‘Excess’ Insurance

Reserve contended that the Tax Court misinterpreted the policies as excess coverage (that is, insurance that applies only after coverage afforded by one or more primary policies is exhausted) rather than primary coverage.10 The Tenth Circuit found no error in the Tax Court’s reasoning that the coverage was excess insurance based on the following clause in the policy:

The coverages afforded by this policy are excess over any other valid and collectible insurance policy issued by any other insurer. The limits and deductibles stated herein only apply after coverage is exhausted from any and all other valid insurance policies issued by any other insurer.

This excess policy does not require the insured to maintain any specific underlying primary insurance policies unless specified by endorsement to this policy. The coverages afforded herein will drop down and provide primary coverage only if there are no other valid and collectible insurance policies in force to which a claim would apply.

Reserve submitted that the second paragraph of the same clause makes it clear that the insurance is primary insurance since some of the policies do not overlap with other policies provided by other insurers. The Tax Court focused on the first paragraph of the clause, and the Tenth Circuit found no reason to alter the Tax Court’s finding on this issue.

B. The Bona Fides of PoolRe

Reserve argued that whether PoolRe was a bona fide insurance company is not relevant for the analysis of whether Reserve properly distributed risk through PoolRe. Reserve argued that the assessment of the bona fides of other companies as an insurance company is not part of the test to determine whether Reserve itself distributed risk to those other companies. In other words, risk can be distributed to non-insurance companies.

However, the Tenth Circuit said that the Tax Court did not base its determination regarding risk distribution on the bona fides of PoolRe as an insurance company. Rather, it was the underlying transactions that were not bona fide insurance agreements. Thus, the appellate court found no error in determining that Reserve did not adequately distribute risk through PoolRe because the transactions through PoolRe were not true insurance arrangements. Ultimately, the Tenth Circuit agreed with the Tax Court’s finding that the quota-share arrangements were a sham.

C. The Reasonableness of Premiums

Reserve argued that the Tax Court erred in finding that the premiums charged were unreasonable. Reserve submitted that they were reasonable when compared with the premiums charged for similar insurance by other captive insurers managed by Capstone. However, the Tenth Circuit dismissed this ground of appeal because there was no evidence that the other captive insurers faced risks similar to those of Peak and there was no evidence that the premiums charged by the other captive insurers were themselves reasonable.

The Tenth Circuit took no issue with the Tax Court’s previous findings of fact on the lack of objective methods and data to determine the premiums. The Tax Court had previously determined that the feasibility study conducted by Capstone provided “no information on the probability of a loss event that the direct written policies covered.”11 Also, it found that the rating worksheet Capstone used was questionable because the record was devoid of evidence of any discussion of the probability or size of the risks covered under the policy. The Tenth Circuit emphasized that insurance companies set their premiums according to predictions of two cost variables: the probability of a particular risk of loss occurring and the magnitude of the loss if it occurs. Evidence that either variable was assessed was lacking in the record.

VI. Analysis

In our August 2021 installment of Blue J Predicts, we analyzed 100 cases involving insurance arrangements.12 We identified the key factual attributes that the courts have examined in applying the legal test for insurance arrangements, and then assembled a structured data set of the factual attributes that were present or not present in each of these cases. Armed with this detailed data set of key factual attributes and the outcomes of previously decided insurance arrangement cases, Blue J’s machine-learning model of insurance arrangement outcomes predicted with 77 percent confidence that Reserve’s appeal to the Tenth Circuit would be dismissed.

In the wake of the Tenth Circuit’s decision in favor of the government, we now go a step further in the analysis and probe the main factors underpinning each of the three arguments Reserve advanced on appeal. Each argument put forward by Reserve consisted of several factors comprising determinations of fact that the court would need to decide in order to accept or reject Reserve’s argument. Further, each factor could only be decided based on the presence or absence of specific evidence in the evidentiary record.

As an example, Reserve argued that the Tax Court erred in finding that the premiums it charged were unreasonably high. Based on the Tax Court’s analysis, the reasonableness of the premiums relied on the following factors: (1) whether the premiums were determined through objective methods such as actuarial data, (2) whether the premiums changed to reflect the actual risk undertaken by the insurer, and (3) whether the premiums were negotiated at arm’s length. To decide on any of these three factors, the Tax Court looked to the evidentiary record that either supported or was at odds with each.

For the three arguments advanced by Reserve on appeal, we examined one key factor underpinning each to demonstrate the effect these factors have on the outcome. For the first argument, that the Tax Court erred in characterizing the policies as only “excess” insurance, we examined the factor of whether the insured maintained complete coverage for the risks from other unrelated insurers. For the second argument, that the Tax Court erred in finding that the transactions with PoolRe were not bona fide insurance arrangements, we examined the factor of whether there was a circular flow of funds between the coinsureds in the PoolRe transactions. Finally, for the third argument, that the Tax Court erred in finding that the premiums charged were unreasonable, we examined the factor of whether the premiums were based on actuarial data or customary industry rating formulas.

VII. Blue J Insights on Key Factors

By analyzing the effect of these three factors using the Blue J machine-learning model, we can determine the likely contribution of each factor to the ultimate outcome of Reserve’s appeal. To this end, we ran a series of hypothetical scenarios through Blue J’s insurance arrangement module to elicit predictions of their validity. Because we are analyzing the presence or absence of three independent characterization questions, each of which could be toggled on or off, there are a total of eight different scenarios. Each of these is identical to Reserve’s scenario, except for the three questions of interest. For each of the eight scenarios, we have recorded the predicted outcome of whether a court is likely to find there was a valid insurance arrangement or not, on a scale of 1 percent likelihood to 99 percent likelihood. The results are summarized in Table 1, ordered from the least favorable to the most favorable scenario for Reserve.

Table 1. Predicted Outcome of Court Finding Valid Insurance Arrangement


Factor 1: Whether complete coverage was maintained by other insurers

Factor 2: Whether there was a circular flow of funds

Factor 3: Whether premiums were objectively determined

Likelihood of a valid insurance arrangement

1 (Original prediction)

Yes. Substantially complete coverage.

Circular flow

Not objectively determined




Circular flow

Not objectively determined




Not circular flow

Not objectively determined




Circular flow

Objectively determined




Not circular flow

Not objectively determined




Circular flow

Objectively determined




Not circular flow

Objectively determined




Not circular flow

Objectively determined


One caveat to the results presented here is that to run this type of analysis to test the effect of each factor, we had to control for whether Reserve would lose simply because it does not meet the threshold of distributing risk to 30 percent unrelated insureds. The courts have not stated that 30 percent is a bright-line rule; however, this is the lowest judicially approved threshold. For simplicity and the sake of analysis, we will treat it as a bright-line rule. Whether 30 percent is the correct threshold was not at issue in this appeal. Both parties agreed that if Reserve failed to meet the 30 percent threshold, then the risk distribution requirement would not have been satisfied. Thus, to run the analysis on the other key factors, we assumed that Reserve could satisfy the 30 percent threshold so that its appeal would not be dismissed solely on that basis.

Hypothetical 1 is Blue J’s originally predicted outcome. In August 2021 we predicted that the Tenth Circuit would likely agree with the findings of the Tax Court on these three key factors, so the likelihood of Reserve succeeding on appeal was only 23 percent. In other words, Blue J’s algorithm predicted with 77 percent confidence that the Tenth Circuit would dismiss the appeal. Note that we are only focusing on three key factors. Insurance arrangement cases contain many more relevant features. Blue J’s algorithm examines a total of 19 factors in its prediction. In focusing on just the above three factors, we continue to hold the other 16 factors constant.

This analysis reveals that while there are five scenarios in which Reserve stood to prevail in its appeal on a more-likely-than-not basis, only hypothetical scenarios 6 through 8 clearly favor the taxpayer. Note that in all the taxpayer-favorable scenarios, the data reveal that factor 3 (whether the premiums were set based on objective methods) was present. The corollary observation is that if Reserve failed to convince the Tenth Circuit that the premiums had been set using objective methods, the greatest chance of success Reserve could have expected was just 53 percent. Therefore, succeeding only in convincing the Tenth Circuit that the premiums were objectively determined would have been just enough to give Reserve a more-likely-than-not chance of prevailing on appeal. This factor was, therefore, one that the taxpayer should have been loath to lose.

In contrast, whether Peak maintained complete coverage from other insurers so that Reserve’s insurance policies were characterized as excess coverage is a significantly less determinative factor. In hypotheticals 7 and 8, controlling for all other factors, whether the coverage is found to be “excess” had just a 5 percent effect on the overall outcome. In other words, on the optimistic end of the spectrum, if Reserve were to succeed on the other two factors, successfully convincing the appellate court that the coverage is primary coverage rather than excess coverage only brings up Reserve’s ultimate chance of success by 5 percent. The same pattern emerges on the other end of the spectrum. Hypotheticals 1 and 2 demonstrate that whether Reserve convinces the court that Peak did not maintain complete coverage for the same risks from other insurers has only a 5 percent effect on the outcome when assuming that Reserve fails on the other two factors.

VIII. Practical Takeaways

Practitioners should be sensitive to potential blind spots to improve their chances of successfully obtaining a desired tax outcome, including in the context of captive insurance arrangements. Sometimes, the blind spot may be underestimating the significance of a particularly weighty factor. Other times, it may be missing important opportunities to shore up the evidence supporting a particularly weighty factor. In any case, it’s crucial to uncover these patterns and insights from the corpus of past legal decisions on the issue.

To do so, particular attention needs to be paid to identifying cases with sufficiently comparable transactions and engaging sufficiently comparable policies. A simple analysis of the legislation and regulations — and even identifying earlier IRS revenue rulings and private letter rulings — is not necessarily enough to provide completely informed advice. For example, in Reserve Mechanical the taxpayer pointed to 39 determination letters that the IRS issued approving tax-exempt status to unrelated insurers under section 501(c)(15). Reserve contended that these previous determination letters were sufficiently like its own captive arrangement to lead to the same exemption.13 However, the Tax Court dismissed these determinations because they are not precedential. Our view is that the most reliable guidance for tax planning and tax litigation is to systematically consider all the existing case law, leveraging machine learning, in addition to performing conventional legal analysis.

A systematic analysis of the existing authorities is wise because not all factors will have the same weight before the courts in different contexts. The central interpretive question that arises in a planning context is which factors are likely to be weighted more heavily than others — and to then take these weights into account upfront in a context-sensitive manner. In a litigation context, it would be a mistake to spend identical resources on each factor or ground of appeal as a matter of sound advocacy. With the scarcity of time, identifying the most significant factors can make the difference between a successful tax planning or tax advocacy strategy and one that misses the mark. Of course, long before litigation arises, spending time obtaining the proper evidence and ensuring that the proper events occur can substantially reduce the risk that the taxpayer will later encounter government resistance or challenges.

IX. Conclusion

In dismissing the taxpayer’s appeal, the Tenth Circuit explicitly reiterated that its decision is not an indictment of microcaptive insurance arrangements generally. The Tenth Circuit stated that insurance pools like the ones with PoolRe may be perfectly legitimate in other circumstances. But regarding the specific evidence presented in this case, it held that the Tax Court did not err in its holding.

Taxpayers should be aware that the IRS appears to have adopted a bolder and more combative stance on the validity of microcaptive insurance arrangements in general. The IRS has stated that it “will use all available legal options to challenge improper attempts to avoid or evade U.S. income tax, regardless of how long it takes for these cases to wind their way through the courts. The IRS will also aggressively pursue penalties for all participants in these abusive transactions.”14

At the tax planning stage, taxpayers should be aware of the factors that will likely have a significant effect on the overall outcome in addition to the types of evidence that will most support those factors. In this appeal, the evidence put forward by Reserve was not enough to convince the Tax Court or the Tenth Circuit on the three key factors discussed in this article. Practitioners should be aware of the deficiencies the courts identified with the evidence so that careful consideration may be given to the types of documentation that need to exist and events that need to occur to successfully deploy and defend a microcaptive insurance arrangement.


1 Reserve Mechanical Corp. v. Commissioner, No. 18-9011 (10th Cir. 2022).

2 Benjamin Alarie and Bettina Xue Griffin, “Captive Insurance Appeal in Reserve Mechanical Will Likely Fail,” Tax Notes Federal, Aug. 30, 2021, p. 1431.

3 Chandra Wallace, “IRS Doubles Down on Microcaptives Enforcement,” Tax Notes Federal, June 13, 2022, p. 1765.

4 IR-2022-118.

5 IR-2015-19 (IRS “Dirty Dozen” list).

6 Visit for more information.

7 Although for some life insurance purposes an “insurance company” is defined under section 816(a) as any company “more than half of the business of which during the taxable year is the issuing of insurance . . . or the reinsuring of risks underwritten by insurance companies.”

8 Harper Group v. Commissioner, 96 T.C. 45, 58 (1991), aff’d, 979 F.2d 1341 (9th Cir. 1992); Amerco v. Commissioner, 96 T.C. 18, 38 (1991), aff’d, 979 F.2d 162 (9th Cir. 1992).

9 In Harper Group, the court found the existence of risk distribution had been satisfied when more than 30 percent of the captive’s gross premiums were derived from insuring unrelated parties.

10 Reserve Mechanical, No. 18-9011, petitioner-appellant’s opening brief at 47.

11 Reserve Mechanical Corp. v. Commissioner, T.C. Memo. 2018-86 at 59.

12 Alarie and Griffin, supra note 2.

13 Reserve Mechanical, T.C. Memo. 2018-86 at 49.

14 IR-2022-118.


Copy RID