Tax Notes logo

Would Management Fees by Any Other Name Still Be Deductible?

Posted on Oct. 25, 2021
Christopher Yan
Christopher Yan
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. Christopher Yan is a senior legal research associate at Blue J Legal.

In this article, Alarie and Yan examine Aspro and use machine-learning models to evaluate the strength of the appellant’s arguments in its appeal to the Eighth Circuit concerning the deductibility of management fees the business paid to its shareholders.

Copyright 2021 Benjamin Alarie and Christopher Yan.
All rights reserved.

I. Introduction

Although most deductions taken by taxpayers are uncontroversial and routinely permitted, some kinds of payments are categorically not deductible. Dividends, as a distribution of after-tax profits to corporate shareholders, are the traditional example. Depending on the differential in tax rates in play, it can be tempting to try to recharacterize payments that would otherwise be regarded as dividends as something else that would be deductible, such as interest payments (as compensation for the use of borrowed capital) or fees for services rendered (as reasonable compensation for work that was done independently of one’s status as a shareholder).

Indeed, in many cases it is plausible that a shareholder has performed bona fide valuable services for a corporation. In these circumstances, the corporation may seek to deduct for tax purposes “management fees” that might in different circumstances be treated as dividends. If not backed up by economic substance, these recharacterizations of dividends as payment for services can lead to trouble. Tax authorities are rightly skeptical of claims that a payment is for services rather than dividends, particularly when there is a clear tax motivation on the part of the parties for doing so. Consequently, tax authorities will scrutinize the substance of these arrangements to determine the true legal nature of the payments.

In last month’s installment of Blue J Predicts, we considered how Blue J’s machine learning technology could have been used by counsel to assess the likelihood of success in an appeal to the Seventh Circuit on the issue of the taxpayer’s entitlement to innocent spouse relief.1 In this month’s column, we examine an appeal to the Eighth Circuit from the Tax Court’s January 21 decision in Aspro.2

Aspro involves a dispute over the deductibility of management fees the business paid to its three shareholders: two corporations, each with 40 percent ownership, and an individual with 20 percent ownership. The Tax Court sided with the IRS in disallowing Aspro’s deductions for management fees paid to the three shareholders. The case is now on appeal.3 Aspro seeks either to have the case remanded to the Tax Court for a redetermination or to have the opinion of the Tax Court modified to reflect the deductibility of the portion of fees Aspro paid, asserting that the Tax Court committed errors of law in denying the deductions.

The IRS’s position is that Aspro’s payments to its three shareholders are not in substance management fees payable for services but represent distributions of after-tax profits (that is, dividends). To deduct the management fees paid to its shareholders, Aspro must show that: (1) the fees paid to the shareholders were for ordinary and necessary services performed for Aspro by or on behalf of the shareholders, and (2) the fees paid to the shareholders were reasonable in their amounts. In the alternative, Aspro may contend that even if the services were valid but some portions of the fees were excessive, the reasonable portion of the fees ought to be deductible.

Several issues in the case are interconnected. For example, if Aspro cannot show that the Tax Court erred in its determination that the payments are not sufficiently connected with the services, then the remainder of the analysis may be effectively moot. Payments in exchange for services that are not connected with the services performed would not be deductible. Similarly, if the Eighth Circuit upholds the Tax Court’s finding that these payments were disguised distributions, then whether the payments were ordinary, necessary, or reasonable also becomes moot because disguised distributions are not deductible. To complicate matters further for Aspro, the IRS took issue with the sufficiency of the evidence for the services provided and the payments for them; the Tax Court reiterated those concerns in its opinion.

However, for this analysis, we focus on the question of whether the payments to the corporate shareholders are likely to be considered ordinary and necessary expenses. While answering this question in the affirmative would not necessarily entitle Aspro to the full deductions claimed, it is an important threshold issue and controlling concept that warrants analysis.

The precise nature of the services provided to Aspro by its two corporate shareholders differed, with important implications for predicting the validity of the deduction of the payments. Although the Tax Court issued a single ruling that Aspro had failed to establish that any of the fees paid to the two corporate shareholders were ordinary, necessary, and reasonable, the services in question arguably differ meaningfully. Specifically, the Tax Court found that some of the services alleged to have been provided by the shareholders were “not customary or usual.” But the Tax Court declined to make this finding about some of the other services, suggesting that those other services were of a type that was more customary or usual. In analyzing hundreds of previously decided cases and making a prediction about the likely outcome on appeal in Aspro, Blue J’s algorithm rendered multiple predictions and distinguished between the services and the likelihood of success for the taxpayer based on the differing “customary or usual” findings.

For services that were not customary or usual, Blue J predicts with 56 percent confidence that expenses connected with those services will not be considered ordinary and necessary. For services that the Tax Court declined to find were not customary or usual, Blue J predicts with 74 percent confidence that the expenses will be treated as ordinary and necessary. Of course, establishing that expenses are for services that are of an ordinary and necessary nature is not, by itself, sufficient to result in entitlement to a deduction; Aspro must still establish that the amounts it seeks to deduct in connection with these services are reasonable (an issue that might need to be redetermined by the Tax Court).

II. Background

Section 162(a) permits a taxpayer to deduct ordinary and necessary expenses paid or incurred in carrying on a trade or business, which includes a reasonable allowance for salaries or other compensation for personal services rendered.

Reg. section 1.162-7 elaborates on this by requiring that the reasonable allowance for salaries or compensation must be for personal services rendered, and that the test for deductibility of compensation payments is whether they are reasonable and in fact payments purely for services. Moreover, reg. section 1.162-7 also contemplates situations in which salaries paid by closely held corporations may be a disguised distribution of dividends on stock when the payments are more than those ordinarily paid and the excessive payments correspond or bear a close proportional relationship to the stockholdings of the officers or employees.

Aspro is a C corporation engaged in the asphalt paving business and has three shareholders: an individual owning 20 percent of its stock and two corporate shareholders, each owning 40 percent of its stock. From 2012 through 2014 Aspro did not make any dividend distributions to its shareholders but paid management fees to each at the end of each tax year.

The individual shareholder was an employee and the president of Aspro and was paid management fees on that basis in addition to other compensation in connection with his employment. The corporate shareholders were paid fees for various services ostensibly performed by affiliates that included assistance with placing alternate bids, self-insured health plan expenses, human resources assistance, equipment advice, lobbying activity, investment management, environmental advice, safety advice, bonding, meeting best practices, dredging, and enjoyment of discounts. Aspro deducted the management fees paid to each shareholder for 2012, 2013, and 2014, thereby reducing its taxable income in each of those tax years. The IRS denied the deductions.

A. Tax Court Decision

The Tax Court sustained the IRS’s denial of the deductions for management fees on the basis that the payments to shareholders were disguised distributions. The court found that the payments to the corporate shareholders were not ordinary, necessary, and reasonable, and that the payments to the individual shareholder, who also acted as president of the business, were ordinary and necessary but not reasonable in quantum. Given that the question of whether management fees paid to the individual shareholder are ordinary and necessary is not under appeal (although the reasonableness of the management fees is subject to appeal), the remainder of this analysis focuses on whether the management fees paid to the corporate shareholders were ordinary and necessary.

Generally speaking, taxpayers may sometimes prefer to characterize payments to shareholders as compensation payments (which are typically tax-deductible expenses under section 162) rather than dividend distributions (which represent after-tax corporate profits and are therefore not a tax-deductible expense).

The Tax Court scrutinized many aspects of the management fee payment arrangement between Aspro and its shareholders in finding that the expenses were not deductible, highlighting the following:

  • Aspro did not have a history of making distributions to its shareholders, and instead, had always paid them in the form of management fees.

  • The amount of the management fees, if deductible, would have eliminated most of Aspro’s taxable income (from 77 percent to 89 percent depending on the tax year).

  • The management fees were paid in amounts that roughly corresponded to each shareholder’s ownership interest, suggesting that payments arose out of each shareholder’s equity in the business. Moreover, both corporate shareholders that had identical share ownership in Aspro were paid identical amounts in management fees for all the tax years in dispute.

  • There was little to no evidence of any method being used to determine the amount of the management fees that were payable. The fees were not set in advance of the services that were performed.

  • The management fees were not paid as the services were provided, but as a single lump sum at the end of the tax year.

  • The purported services were performed by individuals and entities affiliated with the corporate shareholders while the payments were made directly to the corporate shareholders.

  • There were no documents or written agreements regarding the services to be performed and no invoices were sent to Aspro for services rendered.

  • There was no evidence of what similarly situated businesses would pay for the services, nor was there a breakdown for how each particular service corresponded with fees that were paid.

  • There was little evidence of whether the expenses were customary or usual in relation to its business.

As a result, the Tax Court found that the fees paid to the corporate shareholders were disguised distributions and concluded that they were not ordinary, necessary, and reasonable. On that basis, the Tax Court denied the entire deduction for management fees paid to Aspro’s corporate shareholders.

B. Taxpayer’s Position on Appeal

Aspro claims in its appellate brief4 that the Tax Court erred in holding that it lacked a compensatory intent in paying management fees to its shareholders and that all the management fee payments were thus nondeductible, even if the payments were reasonable in amount. In other words, Aspro asserts that it should only be disallowed from deducting fees found to be in excess of what is reasonable rather than be wholly denied the deduction of all management fees, including those fees paid in connection with services that provided legitimate value. Aspro contends that the Tax Court should have determined what portion of the compensation should be appropriately treated as a disguised dividend and confined its denial of the deduction to only that portion.

Aspro maintains that the Tax Court erred in finding it lacked compensatory intent given that its directors were aware of the services provided by the shareholders over many years in exchange for management fee payments. Aspro further claims that the court erred in holding that none of the management fees it paid were reasonable in amount and erroneously disregarded the IRS expert’s concessions that at least some portions of the fees were reasonable. The IRS’s expert conceded that at least $77,340 of value was generated by some of the services performed.5

Aspro also claims that the Tax Court erred by applying a “customary or usual” standard to determine if the fees paid were deductible. There appear to be two arguments embedded in this claim. First, Aspro is effectively asking the court to bypass the ordinary and necessary analysis and jump right into reasonableness when management fees are concerned, because compensation for personal services rendered is an ordinary and necessary business expense. Second, Aspro argues that the “customary or usual” standard applied by the Tax Court set too high a bar when it excluded otherwise obviously deductible expenses such as consultations on the alternate bid, management of employee participation in a self-insured insurance plan, consulting services by a senior-level executive, or the management of a multimillion-dollar investment fund. The IRS’s expert conceded that these services had at least some value and it is common sense that they were beneficial to Aspro’s operations.

Aspro defends its lack of documentation by citing precedent for relaxed requirements for closely held corporations and claims that the Tax Court erred by ignoring precedent that permits this. Aspro also claims that the court erred in holding that fees paid to the corporate shareholders were not deductible because some of the services were provided by the shareholders’ affiliates. There is no requirement in the law that management fees cannot be deducted when paid based on services provided by affiliated companies.

C. IRS’s Position on Appeal

The IRS’s brief largely defends the position and analysis set out in the Tax Court opinion. It reiterates the problematic nature of the deductions sought for services that were provided by persons and businesses loosely affiliated with the entities to which management fees were paid.

The IRS also justifies why the Tax Court remained unpersuaded that a connection existed between the fees and the services, including the fact that (1) no management fee agreements existed between Aspro and its shareholders, (2) there was no documentation that detailed the services provided, (3) Aspro paid its shareholders instead of the service providers directly, (4) Aspro had no structured method for determining the amounts of management fees and paid amounts that were not shown to be related to the value of the services, and (5) Aspro paid or authorized the fees in lump sums rather than periodic payments as services were provided.

The IRS cites precedent to support its position that payments must be intended to be compensation for services rendered to be deductible and that showing that the fees are reasonable alone is not sufficient. The agency also defends the Tax Court’s decision to deny the deductions in their entirety because Aspro failed to establish that it paid the management fees purely for services.

The IRS defends the Tax Court’s application of the customary or usual standard in determining whether an expense was ordinary. Moreover, the agency contends that the ordinariness requirement extends beyond whether the services were ordinary expenses and should include an examination of whether the form and method of payment were ordinary, and whether any exaggerated quantum of payment was reasonable.

The IRS rejects the notion that closely held companies are excused from their statutory recordkeeping requirements when claiming deductions and emphasizes that the taxpayer ultimately bears the burden of proof that the agency’s determination is incorrect.

The IRS claims that Aspro misapprehends the Tax Court’s decision in asserting that the court overlooked authority permitting a corporation to deduct a payment to one entity in return for benefits received from a related entity. Ultimately, the IRS points to the fact that Aspro had failed to “connect the dots between the services performed and the management fees it paid” and that the undocumented nature of the arrangement between affiliated parties was just one aspect that contributed to the company’s overall inability to connect the dots.

III. Legal Technology Insights

While a deceptively simple test on its face, the question of whether an expense is ordinary and necessary has been the subject of extensive litigation, especially considering the vast array of possible circumstances under which an expense can be assessed in the context of different trades or businesses. Advances in computing power and machine learning are particularly well suited to handle the sizable body of case law and the fact-intensive inquiry required to answer this legal question.

Blue J uses machine learning to assess and model a data set of more than 300 cases that consider whether an expense is an ordinary and necessary business expense under section 162(a). The facts and circumstances used in Blue J’s predictions are the kinds that judges refer to and rely on in their decisions. After a user inputs all the relevant facts of the case, Blue J’s model produces a prediction of how likely it is that a court would rule that an expense is ordinary and necessary based on those stipulated facts and circumstances. Blue J also discloses the degree of confidence it has in that prediction.

A. Ordinary and Necessary

Although the issue of whether the management fees paid were ordinary and necessary expenses is just one of many live issues, the parties disagree at a basic level on whether and to what extent the test applies to management fees. As noted, Aspro claims that the Tax Court erred in law and should not have conducted an inquiry into whether specific services provided to it were ordinary and necessary, and should have instead applied the Charles Schneider factors6 to determine the reasonableness of the management fees paid.

Section 162(a)(1) provides:

(a) In general

There shall be allowed as a deduction all the ordinary and necessary expenses paid or incurred during the taxable year in carrying on any trade or business, including —

(1) a reasonable allowance for salaries or other compensation for personal services actually rendered.

On appeal, Aspro reasons, based on a plain reading of the statute, that because compensation for personal services rendered (that is, management fees such as those paid) is on its face an ordinary and necessary business expense under section 162, the Tax Court did not need to conduct any further analysis on “ordinary and necessary” and should have proceeded straight into an examination of reasonableness. While this is a tempting interpretation on its face, Aspro’s position seems misplaced for a few reasons.

First, Aspro is essentially asking the court to sidestep one of the most crucial components of the legal test in determining the deductibility of expenses. While compensation for personal services rendered is one example of an ordinary and necessary business expense, a contextual reading of the statute suggests that personal services compensation is still subject to the controlling concept that the personal services rendered must still be ordinary and necessary. Aspro’s interpretation that compensation for personal services rendered is by default ordinary and necessary would suggest that all payments for personal services rendered are ordinary and necessary, which cannot be true.

Second, the question of whether payment to the shareholders was made in connection with personal services remains a live issue. We know that payment was made and services were rendered, but the Tax Court as well as the IRS in its submissions on appeal reject the notion that the evidentiary record supports that these payments could be connected to the services rendered in any meaningful way. On the contrary, the Tax Court found that the payments were in fact disguised distributions and were not payments made purely for services.

Third, the personal services in question that collectively fall under Aspro’s categorization of management fees actually consist of a variety of disparate and unrelated services performed by different individuals and entities. If the services were relatively homogenous and provided similar types of value, it might have an easier time requesting homogenous treatment that all services are personal services that are ordinary and necessary. But this is not the case here.

Therefore, Aspro may have been better served by arguing that each of the services for which it seeks to claim a deduction was an ordinary and necessary expense by conducting the multifactor analysis set out in the case law on ordinary and necessary, rather than seeking to bypass the analysis altogether.

However, Aspro’s accompanying argument that the Tax Court’s application of the ordinary and necessary standard sets the bar too high appears to be more forceful. The IRS interprets the meaning of ordinary to include a requirement that the form and arrangement of payment must also be ordinary, and not just whether the type of expense is ordinary. In considering the services individually, the Tax Court found that Aspro failed to establish that some services provided were customary or usual.

Table 1. Tax Court Findings on Aspro’s Services and Expenses


Tax Court’s Findings on Whether Aspro Established That the Service or Expense Was ‘Customary or Usual’

Alternate bid assistance


Self-insured health plan


Human resources assistance

Equipment advice

Tim Manatt’s advice


Lobbying activity


Investment management


Environmental advice

Safety advice



Best practices




Various discounts


aThe Tax Court’s finding in some cases applied to a failure to establish that the method, choice, or recipient of payment was customary or usual rather than a failure to establish that the type of expense itself was customary or usual.

B. Applying Machine Learning

In part, Blue J’s algorithm considers the impact of whether an expense is “customary or usual” by asking a user whether it is likely that the taxpayer’s competitors have incurred this type of expense previously. When inputting the various other remaining facts of the case into the questionnaire, Blue J’s algorithm reveals the effect of this factor alone (while holding the other 22 factors constant):

Table 2. Effect on Outcome of Whether Competitors Likely Incurred This Type of Expense


Competitors Likely Incurred This Type of Expense

Ordinary and Necessary?

Scenario 1


Yes, 74%

Scenario 2


No, 56%

aServices for which the Tax Court did not find that the taxpayer failed to establish that the service or expense was customary or usual.

bServices for which the Tax Court found that the taxpayer failed to establish that the service or expense was customary or usual.

It should be noted that the algorithm does not assign static weights to any of the facts and circumstances, but those factors are weighted dynamically based on how other factors tend to be interpreted and addressed by the courts. Based on the particular facts and circumstances of this case, the answer to whether competitors likely incurred this type of expense changed the algorithm’s confidence by 30 percent, enough to change the predicted outcome in the appeal. In a different set of facts, this question may change the algorithm’s confidence by a different percentage and may or may not be pivotal.

This one-factor experiment demonstrates that whether an expense is ordinary and necessary can turn on whether a taxpayer succeeds or fails to establish that the expense incurred is the type of expense that competitors have previously incurred. Admittedly, neither prediction is a confident one, because of conflicting factors in the fact pattern.

Therefore, even based on the Tax Court’s findings alone (as summarized in Table 1), there is a meaningful difference between services that the taxpayer has failed to establish are customary or usual versus otherwise. In this case, one set of services is more likely to be found to be ordinary and necessary than the other set.

The other takeaway is that this simple experiment demonstrates a way to quantify the risk of failing on a legal determination based on whether the taxpayer can adduce evidence to demonstrate that a particular type of expense is customary or usual in the industry. Moreover, there are other problems caused by insufficient evidence that can be illustrated in our next example.

C. Sufficiency of Evidence Problem

While Blue J’s machine learning algorithm does not evaluate the sufficiency of evidence and relies on the tax practitioner to exercise her skill and judgment in making a determination, the algorithm is able to identify issues with evidence sufficiency and quantify the risk of not having the evidence to substantiate a tax position.

Another question that Blue J’s algorithm asks is whether there is a substantially more cost-effective way of achieving the intended outcome of the transaction or activity. It is clear in Aspro that this information is missing, as the taxpayer failed to substantiate the cost of comparable services. This puts the taxpayer at a distinct disadvantage, because the taxpayer cannot confidently speak to whether the expense was truly ordinary and necessary without presenting comparable options. Even though evidentiary deficiency is obvious and discussed at length in the Tax Court’s opinion, the algorithm is able to quantify the effect of this particular deficiency, using an examination of the previous two scenarios (the previously reported results are the “unknown” base cases of scenarios 1 and 2 in Table 3):

Table 3. Cumulative Effect on Outcome of Changes to Multiple Factors


Competitors Likely Incurred This Type of Expense

Substantially More Cost-Effective Way of Achieving Outcome

Ordinary And Necessary

Scenario 1



Yes, 79%


Yes, 74%


Yes, 69%

Scenario 2



Yes, 60%


No, 56%


No, 72%

Three observations can be made from these results. First, in a situation in which the evidence supports that there is a substantially more cost-effective way of achieving the intended outcome, the algorithm is less confident about the taxpayer’s chances of success in establishing that the expense was ordinary and necessary, even when competitors likely incurred this type of expense. However, the algorithm is more confident that the taxpayer will likely fail to establish that the expense was ordinary and necessary when competitors were not likely to have incurred this type of expense.

Second, being able to confidently present evidence that there is not a substantially more cost-effective way improves a taxpayer’s odds of claiming that the expense is ordinary and necessary. Even in a scenario in which the taxpayer cannot establish that competitors likely incurred this type of expense, the taxpayer’s odds have shifted from being more likely to fail on this legal issue to more likely than not to succeed on this legal issue (albeit with narrow margins).

Third, the results reveal a point that was made earlier: that the weight and effect of each factor are dynamic and interact with other factors in the scenario. Here, the 30 percent difference between whether the competitors likely incurred this type of expense between scenarios 1 and 2 shrunk to a 19 percent difference when the taxpayer could confidently claim that there are no other more cost-effective ways of achieving the same outcome but expanded to a 41 percent difference when the evidence shows that there was a substantially more cost-effective way of achieving the intended outcome.

D. Examining Case Data

The process by which the team at Blue J builds predictive algorithms involves a rigorous data collection process to translate unstructured data from the jurisprudence into structured data that can be used to train machine learning models. The structured data has also been repurposed into a separate queryable database that allows tax practitioners to examine the case data from a bird’s-eye view. Let’s examine how we can apply this to the case at hand in formulating legal arguments on appeal.

The facts in Aspro suggest that at least some portion of the payment to shareholders was made outside of payment for services. The taxpayer argues on appeal that the existence of these types of payments made to shareholders (that is, disguised distributions) should not be fatal to the entire claim for deductible expenses. This specific circumstance is a factor in Blue J’s algorithm; Blue J considers cases identifying this circumstance as a relevant factor in its predictive model. A query of the Blue J database supports the taxpayer’s contention on this front. In the 43 rulings in which payment was made to a shareholder or partner excluding payment for services provided, courts found that some part of the expense was ordinary and necessary in just more than half the cases (22 of 43).

The Blue J platform also allows users to immediately access the full text of these decisions to narrow the list of relevant cases that merit further research and scrutiny. This process cuts down the amount of time spent trying to find relevant cases and allows tax practitioners to focus their attention on the most germane and relevant cases.

IV. Conclusion

Blue J predicts with 74 percent confidence that the expenses in connection with the set of services provided to Aspro that are customary or usual will be found to be ordinary and necessary expenses. Blue J also predicts with 56 percent confidence that expenses in connection with the set of services that Aspro has failed to establish are customary or usual will be found not to be ordinary and necessary expenses.

Ultimately, regardless of which way the Aspro case goes, it should serve as a cautionary tale that taxpayers who seek to claim deductions must take special care in substantiating the form and labels they choose to characterize their transactions. It is a critical practice point to note that taxpayers are best served by developing contemporaneous documentation, even if (perhaps especially if) they operate as closely held corporations.

Even if Aspro successfully convinces the Eighth Circuit that only deductions in excess of reasonable payments should be denied, it may still be difficult for a court to retroactively determine which portion of the fees are reasonable and which portions are disguised distributions, especially in the absence of contemporaneous documentation. Ex post valuations generated by the taxpayer’s experts may be seen as self-serving, inaccurate, or unreliable.

In the context of tax and business planning, practitioners ought to use legal technology to identify relevant practices and develop robust documentation to withstand scrutiny from tax authorities. In the context of litigation, practitioners can leverage machine learning to quantify the risk of going to trial while homing in on the most salient factors to focus on.


1 Benjamin Alarie and Stefanie Di Giandomenico, “Seventh Circuit Affirms Spouse Is Not So Innocent on Appeal,” Tax Notes Federal, Sept. 27, 2021, p. 2149.

2 Aspro Inc. v. Commissioner, T.C. Memo. 2021-8.

3 The parties have filed their respective briefs as of the date of this writing.

4 Aspro Inc. v. Commissioner, No. 21-1996 (8th Cir. 2021), Appellant’s Brief (July 16, 2021).

5 Id. at 32.

6 Charles Schneider & Co. v. Commissioner, 500 F.2d 148 (8th Cir. 1974).


Copy RID