Lesson From The Tax Court: No Deduction For Disguised Dividends

A lesson that comes up often in my tax class is how economic substance trumps transactional form.  One common example is when a corporate taxpayer seeks to deduct payments that seem to be compensation payments.  Sometimes, however, not all is what it seems and the corporation is really distributing corporate profits rather than incurring a corporate expense.

In last week’s case of Aspro, Inc. v. Commissioner, T.C. Memo. 2021-8 (Jan. 21, 2021), Judge Pugh’s clear and crisp opinion teaches us how the Court decides whether compensation payments are really disguised dividends, a lesson we can use to help clients avoid the mess that this taxpayer got into.

Closely held corporations are especially tempted to disguise dividend distributions as compensation.  That is because it is cheaper for the corporation to pay out compensation (which is an expense and, hence, deductible) than it is to pay out dividends (which represent the distribution of after-tax corporate profits and, hence, are not deductible).  Even though the corporation might also pay employment tax on compensation, that is still cheaper than paying income tax on the money used to pay dividends.  The shareholders/employees are also tempted because disguising corporate dividends as compensation similarly works out better for their bottom line, again even after accounting for employment taxes.

It becomes important, then, for courts to detect when compensation is really a disguised dividend distribution.  The law here is messy, but I’ll try to give you the gist of his here.

Law: Detecting When Compensation is Disguised Dividends
Section 162(a) allows taxpayers to deduct the ordinary costs of carrying on a trade or business.  The statute explicitly lists one of those costs as “reasonable salaries or other compensation” and it has long been settled that similar payments to non-employees for services are also among the deductions authorized by §162.

The word “reasonable” is important.  Treas. Reg. 1.162-7(a) provides that “The test of deductibility in the case of compensation payments is whether they are reasonable and are in fact payments purely for services.”  Treas. Reg. 1.162-7(b) then explicates that general rule by explaining what compensation payments might be disguising, such as dividends or sales proceeds.

Thus, to be deductible, courts engage in a two-step inquiry: (1) were the payments actually compensation for services; and (2) were the amounts reasonable?  The regulations allow for a partial exception to the reasonableness requirement if payments are made under a contingent salary contract that was a result of a “free bargain” at the time of its formation, even if the actual payments under the contract “prove to be greater than the amount which would ordinarily be paid.”  Treas. Reg. 1.162-7(b)(2).  See Harold’s Club v. Commissioner, 340 F.2d 861 (9th Cir. 1965).

When considering employee salaries, a majority of courts have re-worked the ideas contained in the regulation into various indeterminate multi-factor tests.  Today’s case would go to the 8th Circuit which has declared that such factors include:

“the employee's qualifications; the nature, extent and scope of the employee's work; the size and complexities of the business; a comparison of salaries paid with the gross income and the net income; the prevailing general economic conditions; comparison of salaries with distributions to stockholders; the prevailing rates of compensation for comparable positions in comparable concerns; the salary policy of the taxpayer as to all employees; and in the case of small corporations with a limited number of officers the amount of compensation paid to the particular employee in previous years.” Schneider v. Commissioner, 500 F.2d 148, 152 (8th Cir. 1974).

A minority of courts object to the multi-factor indeterminate approach.  The brilliant Judge Richard Posner of the 7th Circuit Court of Appeals gave a powerful take-down of that approach in Exacto Spring Corporation v. Commissioner, 196 F.3d 833 (7th Cir. 1999).  He proposed that courts evaluate reasonableness of compensation by using what he called the “independent investor test.”  The idea is to ask whether a hypothetical investor in the taxpayer would have received a sufficient return on investment (ROI) given the purported salaries.  After all, excessive salaries are by definition not available for either reinvestment in the business or dividends, the two things investors care about, says Posner.  Posner's test requires “comparing the corporation's reported income with that of similar corporations, the comparison being stated in terms of percentage return on equity, the standard measure of corporate profitability.”  Mulcahy, Pauritsch, Salvador & Co., v. Commissioner, 680 F.3d 867 at 870 (7th Cir. 2012).  The problem, as even Posner admits, is that if “the company cannot be valued, neither can the returns to the shareholders be calculated as a percentage of that value, and so the independent investor test is difficult to run.”  Id.  Yup.

An additional problem with the independent investor test is that few of use are as smart as Judge Posner.  Sure, it’s no trouble for him to figure it out, but other judges will need to rely on expert testimony.  This risks throwing the decisional authority from a duly appointed court to an unduly partisan expert.  A good example of this problem is in Wagner Construction v. Commissioner, T.C. Memo. 2001-160, where an exasperated Judge Carolyn Miller Parr threw out both parties’ expert reports for “patently favor[ing] their respective clients,” noting that “their reports were designed to support their conclusions.”  Judge Parr then proceeded to use a multi-factor test, one prong of which did, however, include considering a hypothetical investor’s ROI.

In short, the actual cases where this issue comes up typically involve closely held corporations and courts have decided that the difficulties in creating a counter-factual analysis regarding what a hypothetical investor would think is more difficult than the traditional approach of looking at the facts and circumstances of what actually happened.

This case is a lesson in how that reasoning works and how to help taxpayers do it right.

Facts
Aspro, Inc. is a C Corp. in the asphalt paving business in Waterloo, Iowa.  For the years in question (2012-2014) it employed between 66 and 75 people.  Its stock was held as follows: (1) Jackson Enterprises Corp. owned 40%; (2) Manatt’s Enterprises, Ltd. owned another 40%; and (3) Mr. Milton Dakovich owned 20%.   Thus, stock was held by two entities and one individual.  During the company’s existence it had never declared a dividend for its shareholders.

During the years in question, the taxpayer took deductions for “management fees” that it paid each of its three shareholders as follows:

Tax Year	Jackson Enterprises	Manatt’s Enterprises	Mr. Dakovich	Total
2012	$500,000	$500,000	$166,000	$1,166,000
2013	$800,000	$800,000	$150,000	$1,750,000
2014	$800,000	$800,000	$200,000	$1,800,000

The fees were paid in a lump sum at the conclusion of each year.  By taking these deductions, the company reduced its taxable income to nubbins.  Judge Pugh found that “The management fees...thus eliminate 89%, 86%, and 77% of what would have been petitioner’s taxable income for tax years 2012-2014, respectively.” Op. at 26.

Mr. Dakovich was also an employee of the company.  He had been its President for 20 years and was responsible for its operations.  He received compensation for his services through a salary, a yearly bonus from an employee bonus pool based on company profits, and a Director’s fee for his service on the Aspro board of directors.  Those amounts added up to roughly $580,000 in 2012, $404,000 in 2013 and $538,000 in 2014.  The variation is chiefly due to the amount of the yearly bonus.  The management fees to him were additional payments.

On audit the IRS disallowed the deduction for management fees.  The Tax Court agreed.

Judge Pugh held the management fee payments were not deductible compensation.  She gave two basic reasons: (1) they were not payments for any sufficiently identified services and (2) they were not reasonable.  Her first reason was more aimed at the payments to the two corporate entities and not so much Mr. Dakovich.  Her second reason was more aimed at the payments to Mr. Dakovich and not so much to the two entities.

Each part of the opinion contains a good lesson for what factors the Court will look for.  Each lesson thus teaches us what we can do to help clients properly structure compensation packages.

Lesson 1: Link Payments to Documented Services
The biggest problem here was that the taxpayer flunked the regulation’s requirement that the payments were in fact payments purely for services.  Treas. Reg. 1.162-7(a). As Judge Pugh put it: “Petitioner has to connect the dots between the services performed and the management fees it paid.  Petitioner failed to do so.”

There is no doubt that Jackson Enterprises and Manatt’s Enterprises provided services to Aspro in each year.  The three companies were inter-related in the sense that their principal owners had all done business together over the years in Waterloo and engaged in cooperative behaviors to help each other’s businesses.  But their business were all different.  Aspro did asphalt paving.  Jackson Enterprises was into concrete paving.  And Manatt’s Enterprises was a farming operation.  Still, each entity had overlapping business interests and informally shared each others’ employees for those overlapping interests.

It was the informal nature of the cooperation that doomed their attempted deduction.

First, Aspro had no management contracts with either entity, and there was zero documentation of what services were either expected to be provided or were actually provided and paid for.  As Judge Pugh noted:  “No management fee rate or billing structure was negotiated or agreed to between the shareholders and petitioner at the beginning of any of the years in issue. And none of the shareholders invoiced or billed petitioner for any services provided.” Op. at 15.

Instead, the Aspro board of directors simply met late in the tax year and decided on an amount to pay ex post.  The board minutes reflected no consideration of actual services but instead reflected discussion on “how the company was going to perform and how much earnings the company should retain.”  Op. at 15.  Hmmm, that sure sounds like a dividend discussion!  And it sure did not help matters that after that discussion, the Aspro board regularly decided on an amount of “management fees” roughly in proportion to each recipient's stock holdings and sufficient to mostly wipe out Aspro’s taxable income.

Second, the taxpayer failed to show at trial what services were in fact provided.  It introduced substantial evidence that various employees of Jackson Enterprises, Manatt’s Enterprises and their wholly owned subsidiaries provided various services to Aspro each year.  But the services themselves did not appear very substantial.  And the services could not be linked to payments.  For example, one employee of a Manatt’s Enterprises subsidiary sometimes visited Aspro project sites to give a safety review.  And two employees of a Jackson Enterprises subsidiary supervised a self-insured health plan that Aspro participated in with both other companies.  But, again, the big problem was that all of this information came up ex post.  There was nothing ex ante.

To fix this problem would not seem onerous.  The company could set up a contract that provided for reasonable payment of a base amount of potential services.  The contract would list the panoply of services actually provided each year, set a base amount for that, and then add an amount for “other services as needed...”  The contract would be a flat fee, paying for potential.

Paying for potential is common.  Think lifeguards.  We lawyers do this all the time when we bill a set fee for a project.  For example, when I was in private practice, my firm had a standard $10,000 fee agreement for start-up corporations where we agreed to provide all legal services (except litigation services) for a one year period.  The client was paying for the potential needs and not actual services (and it was nice not to have to bill time for those clients).

Of course, doing this ex ante means the company may or may not have a profit to share with shareholders at the end of the year.  After all, the yearly contract amount should not fluctuate and should be linked not to each entities’ ownership share in the company (!!) but to the expected value of the potential services it would provide.  That was another problem that bothered Judge Pugh: there was simply no rational explanation for the fluctuations in “management fees” in this case.  If well done, an upfront consulting contract would reduce but not eliminate dividend distributions.  Remember, the idea here is not to disguise dividends and not to get greedy, but to create a reasonable basis for accounting for reasonable compensation.

Lesson 2:  Of Pigs and Hogs
The taxpayer also flunked the second requirement of the regulation, that even if payments were linked to services, the taxpayer was unable to show that the payments were “reasonable compensation.”  This part of Judge Pugh’s opinion went mostly to the supposed management fees paid to Mr. Dakovich.  While she does also walk through the services provided by employees of the other two entity shareholders, but again that is a problem that can be fixed by a proper ex ante management services contract by paying for potential.  Mr. Dakovich's problem cannot be so easily solved, because he was already providing services.

As Mr. Dakovich was President of Aspro, there was “no reason to question whether it was ordinary or necessary for petition to compensate its president” (Op. at 39) regardless of how the compensation was labeled.  Instead, the question was whether his total compensation was reasonable.  Judge Pugh followed Judge Parr’s approach of blending a multi-factor test with the independent investor test and, unlike Judge Parr, Judge Pugh was able to rely upon the IRS expert’s report.

The biggest factor against the claim that the management fees were compensation to Mr. Dakovich for services was that all the other payments made to Mr. Dakovich for those same services already greatly exceeded the prevailing rates of compensation for comparable positions in comparable companies.  Ignoring the management fees, the IRS expert testified that that Mr. Dakovich’s other compensation exceeded both the industry average and median by over $200,000.  Based on that, Judge Pugh concluded, “because of the management fees paid to Mr. Dakovich were ostensibly additional compensation for services that he performed as petitioner’s president---services for which he was already highly compensated comparison to appear companies---he entire amount of management fees paid to Mr. Dakovich appears unreasonable.” Op. at 44.

The other factor that worked against reasonableness was a comparison of the management fees and the company’s choices regarding dividends and retaining earnings.  The problem was that Aspro had never paid dividends.  That alone was not a killer.  But the expert testified that Aspro’s operating margins were strong before it paid the management fees, but were significantly below peers once it paid out the management fees.  That persuaded Judge Pugh that a hypothetical investor would not find investment returns reasonable in light of the management fees.  Now if they had been dividends.... 

The fix to this problem requires moderation.  It seems to be that there was a degree of inter-connectedness between Aspro and the two entity shareholders that would justify some payment of compensation for services, whether labeled “management services” or not.  So in addition to creating an ex ante contract paying for potential, some like sum could be added to Mr. Dakovich’s base salary.  As in many other areas of tax, it is not so much a question of all or nothing as a question of more or less.  The lesson here is moderation: Aspro should in fact declare dividends in years when appropriate and not try to disguise company profits as company expenses.  At the same time, it appears Aspro could more accurately identify its operational costs, thus more accurately identifying the company's true overall profit after accounting for the value of the informal cross-company services.

Coda 1: The new top corporate tax rate of 21% would appear to lessen the incentive to re-characterize dividends as compensation.  However, one also needs to factor in the cost of employment taxes to the company. 

Bryan Camp is the George H. Mahon Professor of Law at Texas Tech University School of Law.  He invites readers to return to TaxProfBlog each Monday for a new Lesson From The Tax Court.