Dexsil Corporation v. Commissioner of Internal Revenue

JOHN M. WALKER, Jr., Circuit Judge: Section 162(a) of the Internal Revenue Code, 26 U.S.C. § 162(a), provides that: 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;.... At issue in this case is whether the amount of salary and bonuses paid by Dexsil during the 1989 and 1990 tax years to Ted Lynn— Dexsil’s president, chief executive officer, treasurer, and chief financial officer — was reasonable compensation for his services and thus deductible by Dexsil as a business expense, or was instead to some degree unreasonable, with the unreasonable amount representing, in effect, a hidden dividend payment. The Tax Court found that Dex-sil’s deduction for compensation to Lynn for the 1989 and 1990 tax years was unreasonable in amounts of $76,540 and $168,000, respectively, and ordered Dexsil to pay the resulting deficiencies of $33,504 and $95,778. For the following reasons, we vacate the Tax Court’s judgment and remand for further proceedings. BACKGROUND Dexsil Corporation was founded in 1977 by Ted Lynn (“Lynn”) and John Churchill, who each made initial capital contributions of $7,000. The company was originally formed to manufacture and market Dexsil, a high-temperature polymer used for gas chromatography. In 1981, following unsuccessful attempts to lower the production costs of the polymer, Lynn purchased Churchill’s interest in the company for $20,000 and became the president of Dexsil. With Lynn at the helm, the company set out to develop, manufacture, and market field-test kits to be used for the on-site detection of hazardous contaminants regulated by the Environmental Protection Agency. By 1990, sales of these kits — named Clor-N-Oil, Clor-D-Tect, Clor-N-Soil, and Q4000 — comprised the bulk of Dexsil’s business. From 1981 until 1983, Dexsil was wholly owned by members of the Lynn family: In 1983 the company raised $105,000 of needed capital by selling approximately ten percent of the company’s shares to about a dozen outside investors at $200 per share. During the tax years in question, 1989 and 1990, Lynn owned 61.62% of Dexsil’s stock, other family members owned a combined total of 29.11%, and the balance, a little over 9%, was owned by non-family employees and outside investors. The company enjoyed tremendous growth and its workforce expanded from 2.5 employees in 1983 to at least 30 employees in 1990. Lynn functioned as the company’s president, chief executive officer, treasurer, and chief financial officer during this period, working *99 approximately 60 to 65 hours per week, Lynn’s annual salary and bonus grew in rough proportion to Dexsil’s annual gross sales. Both Lynn and Dexsil s accountant, Richard Kaczynsld, testif^^ that at least smce 1982, they had employed an informal „ , , ¿ T ,, , .j formula whereby Lynn would be paid approximately 11% of the year’s gross sales. With respect to its other employees, Dexsil instituted an incentive stock-option plan and a restricted stock awards plan for key non-officer employees and it disbursed bonuses to each of its employees. It appears, however, that shareholder-employees received bonuses ^hat were considerably larger than non-shareholder employees. According to Dex- ... . _ . , , , , , .. sirs audited financial statements and payroll r J reeords> the company s gross sales, return on equity, retained earnings, and dividends, as well as the salary and bonus it paid to Lynn were as follows: Fiscal Gross Sales Return on Retained Dividends Earnings Year Equity * Paid Lynn’s Salary and Bonus % OF Sales 9/77 5 ' 52,974 $ 2,684 $ 0 0% 9/78 68,391 4.7% 3,483 15.000 22 9/79 83,736 7.3 4,802 6,000 7 9/80 96,505 4.5 5,676 20.000 21 9/81 83,488 40.8 ■ 8,628 $5,676 13.000 16 9/82 96,868 (209.2) (15,683) 2,400 10.000 10 9/83 113,179 2.5 (14,439) 12,000 11 9/84 698,719 56.6 75,617 73,000 10 9/85 1,875,038 76.2 344,512 232,400 12 9/86 2,492,631 43.7 617,150 288,600 12 9/87 2,406,396 31.8 904,536 252.000 10 2,702,491 31.0 1,273,074 6,890 277,880 10 9/89 3,419,984 26.3 1,670,919 16,090 376,540 11 9/90 4,899,359 25.5 2,168,087 25,992 488.000 10 On November 17, 1992, the Commissioner served Dexsil with a notice of deficiency pursuant to 26 U.S.C. § 6216, alleging, inter alia, deficiencies in its federal income taxes for fiscal years 1989 and 1990 of $82,739 and $174,207, respectively. Dexsil was assessed an additional $8,710 for failure to file a timely return for tax year 1990. On January 21, 1993, Dexsil filed a petition for redetermination with the Tax Court, pursuant to 26 U.S.C. § 6213(a). Following a two-day trial, the Tax Court (Mary Ann Cohen, Jtidge) issued a memorandum opinion agreeing with respondent that the compensation paid to Lynn in 1989 and 1990 was unreasonable under § 162 of the Internal Revenue Code, 26 U.S.C. § 162, but concluding that a greater amount than that calculated by the Commissioner was reasonable compensation in light of Lynn’s significant contribution to Dexsil’s superior financial performance. This appeal followed. *100 DISCUSSION The Tax Court’s definition and application of the appropriate factors in the determination of reasonable compensation is reviewable de novo as a question of law. Rapco, Inc. v. Commissioner, 85 F.3d 950, 954 (2d Cir.1996); Rutter v. Commissioner, 853 F.2d 1267, 1271-72 (5th Cir.1988). If we find the Tax Court adequately considered the various factors as a matter of law, its finding as to the reasonableness of compensation paid is reviewable as an issue of fact under the clearly erroneous standard. Rutter, 853 F.2d at 1272. “A finding is ‘clearly erroneous’ when ... the reviewing court ... is left with the definite and firm conviction that a mistake has been committed.” Id. (internal quotations omitted); see also Owensby & Kritikos, Inc. v. Commissioner, 819 F.2d 1315, 1323-24 (5th Cir.1987). The IRS’s determination of reasonableness is presumed to be correct. Thus the taxpayer has the burden of showing that it is entitled to a larger compensation deduction than that allowed by the IRS. See Owensby & Kritikos, 819 F.2d at 1324. If petitioner proves that the IRS’s determination is wrong, the Tax Court then must decide for itself the amount of reasonable compensation. See Acme Constr. Co. v. Commissioner, 69 T.C.M. (CCH) 1596, 1598, 1995 WL 5855 (1995). Where a case involves a closely held corporation whose controlling shareholders set their own level of compensation, the reasonableness of the compensation paid to the shareholder-employee is subject to close scrutiny. See Rapco, 85 F.3d at 954. Such close examination is warranted because: it obviously is in the tax interest of all parties to characterize the amounts distributed to shareholder/officers as compensation rather than as dividends. Corporations can deduct salaries and bonuses under § 162(a)(1) but obviously not dividend payments because they are profits, not business expenses. The concern is to prevent the distribution of corporate profits through the payment of unreasonably large salaries and bonuses to controlling shareholder/officers. Rutter, 853 F.2d at 1270-71; see also Owensby & Kritikos, 819 F.2d at 1322-23 (same). a. Relevant Factors In Rapco, we outlined several factors to be considered in assessing the reasonableness of an employee’s compensation. 85 F.3d at 954. Those factors are, broadly speaking: (1) the employee’s role in the taxpaying company, “including the employee’s position, hours worked, and duties performed;” (2) comparison of the employee’s salary with those paid by similar companies for similar services; (3) the character and financial condition of the company; (4) potential conflicts of interest, such as the “ability to ‘disguise’ dividends as salary;” and (5) “[[Internal consistency in compensation ... throughout the ranks of the company.” 1 Id. (quoting Elliotts v. Commissioner, 716 F.2d 1241, 1245-48 (9th Cir.1983)). No single factor is dispositive and “the court should assess the entire tableau from the perspective of an independent investoi’ — that is, given the dividends and return on equity enjoyed by a disinterested stockholdex-, would that stockholder approve the compensation paid to the employee?” Id. at 954-55. b. Hypothetical or Independent Investor Test Petitionei*’s first argument on appeal is that the Tax Court erred in failing to consider Lynn’s compensation from the perspective of a hypothetical investor. Petitioner contends that had the Tax Coui't done so, Dex-sil’s extraordinary return on equity, the dividends it paid dui'ing the tax years at issue, and the testimony of Dexsil’s actual independent iixvestors would have strongly favored a finding that Lynn’s compensation was reasonable. Under the hypothetical or independent investor test, coux'ts assess the reasonableness of compensation in tei'ms of *101 “[w]hether an inactive, independent investor would be willing to compensate the employee as he was compensated. The nature and quality of the services should be considered, as well as the effect of those services on the return the investor is seeing on his investment.” Elliotts, 716 F.2d at 1245; see also Acme Constr., 69 T.C.M. at 1602-03. If the bulk of the corporation’s earnings are being paid out in the form of compensation, such that the corporate profits do not represent a reasonable return on the shareholder’s investment, then an independent investor would probably disapprove of the compensation arrangement. See Elliotts, 716 F.2d at 1247. Relevant factors to consider include the company’s return on equity, whether and to what extent the company has paid dividends, increases in the company’s net worth, and increased market value of the company’s stock. See, e.g., id. at 1247 (20% return on equity “would satisfy independent investor”); Kennedy v. Commissioner, 671 F.2d 167, 176 (6th Cir.1982) (six-fold increase in net worth over ten-year period indicated compensation was reasonable); L & B Pipe & Supply Co. v. Commissioner, 67 T.C.M. (CCH) 2798, 2805, 1994 WL 151310 (1994) (investor would have been happy with either 6% dividend return plus 10% growth in retained earnings; a 20% growth in shareholders’ equity; or a capital appreciation factor in excess of 30%). As noted, in this circuit the independent investor test is not a separate autonomous factor; rather, it provides a lens through which the entire analysis should be viewed. See Rapco, Inc., 85 F.3d at 954-55. Thus, if the company’s earnings on equity, when viewed in relation to such factors as the company’s overall performance and levels of compensation, “remain at a level that would satisfy an independent investor, there is a strong indication that management is providing compensable services and that profits are not being siphoned out of the company disguised as salary.” Elliotts, 716 F.2d at 1247; see also Owensby & Kritikos, 819 F.2d at 1327; L & B Pipe & Supply Co., 67 T.C.M. at 2805. Both documentary and expert evidence were presented at trial concerning Dexsil’s profitability in comparison to other companies, including return on equity and dividends paid. Nevertheless; the Tax Court made no mention of either the independent investor test or the specific considerations that courts have held to comprise the independent investor inquiry. “Although we accord deference to the Tax. Court’s special expertise, [the] definition of the'appropriate factors is reviewable by this court as a question of law.” Elliotts, Inc., 716 F.2d at 1245. The Tax Court’s apparent failure to consider Lynn’s compensation from the perspective of an independent investor was legal error. 'Accordingly, we must vacate and remand this case to allow it to afford such consideration. c. Contingent Compensation Formula Petitioner next contends that the Tax Court committed error by failing to consider petitioner’s assertion that Lynn was consistently paid pursuant to a compensation formula whereby his total compensation (salary plus bonus) roughly equaled' 11% of Dexsil’s gross sales. Dexsil argues, that the existence of such a long-standing formula is evidence of reasonableness. Contingent compensation formulas, which are typically connected to gross revenues or profits, are often adopted to provide incentive to the, employee. Courts have recognized that such arrangements “may tend to stimulate activity.” Kennedy, 671 F.2d at 174. In general, a formula will be upheld “if it was set up when the business started, or when the amount of future earnings was -questionable, and has been consistently followed through the ups and downs of the business.” • Id. at 175; see Mayson Mfg. Co. v. Commissioner, 178 F.2d 115, 121 (6th Cir.1949). Moreover, at least with respect to closely held companies, such agreements need not be in writing and may be informal. See, e.g., Mad Auto Wrecking, Inc. v. Commissioner, 69 T.C.M. (CCH) 2330, 2338, 1995 WL 149132 (1995); Acme Constr., 69 T.C.M. at 1602; L & B Pipe, 67 T.C.M. at 2806. In recognition of -, the fact that “[s]uch a formula may over compensate in *102 good years and undercompensate in bad years,” Elliotts, 716 F.2d at 1248, “[a]n incentive compensation agreement is to be judged by the circumstances existing at the time the agreement was made, not the time it is questioned.” Kennedy, 671 F.2d at 174 (citing 26 C.F.R. § 1.162-7(b)(2)). Thus, courts have upheld seemingly high salaries for executives where they have been calculated pursuant to a long-standing and consistently applied formula, under the rationale that “[wjhere a formula has proved reasonable over a long-period of time, ... it should normally not be deemed unreasonable solely because it has overcompensated in one or two years.” Elliotts, 716 F.2d at 1248 n. 8; see also Owensby & Kritikos, 819 F.2d at 1327-28. As the Sixth Circuit has explained, such an arrangement “does not lose its original good faith character because its automatic operation in later years under favorable business conditions returned materially higher compensation.” May son Mfg., 178 F.2d at 121. The Tax Court’s opinion is virtually silent with respect to the evidence proffered by Dexsil that, starting in 1982, it had consist tently compensated Lynn according to a formula of approximately 11% of sales. See Table, supra, at [p. 99]. Thus we are left to wonder whether the judge rejected Dexsil’s argument that a formula existed, found the formula to be unreasonable, or simply failed to consider it. There is some indication in the opinion that the Tax Court implicitly rejected the existence of the formula. In the opening paragraph discussing Lynn’s compensation, the Tax Court’s opinion refers to petitioner’s “alleged ‘consistently applied compensation plan.’” Later in the discussion of Lynn’s compensation, the Tax Court states, “[a]l-though the record indicates that petitioner paid bonuses to many other employees, there is no evidence of a set bonus formula, and, thus, we cannot determine whether the bonus awards to the employees of petitioner were determined pursuant to any longstanding, consistently applied compensation plan.” If, by this latter statement, the Tax Court was suggesting that in order for there to be a valid contingent compensation formula for Lynn, it must have been applied to Dexsil’s other employees, we find it to be erroneous as a matter of law. Certainly a comparison of the compensation paid to non-shareholder-employees and shareholder-employees is a relevant consideration in determining the reasonableness of compensation. See, e.g., Elliotts, 716 F.2d at 1247 (bonuses and salaries “paid to controlling shareholders are open to question if, when compared to [bonuses and] salaries paid non-owner management, they indicate that the level of compensation is a function of ownership, not corporate management responsibility”); Owensby & Kritikos, 819 F.2d at 1330-33 (finding patent disparity between compensation paid to nonshareholder-employees and shareholder-employees). But this is an inquiry separate from whether a contingent compensation formula used to determine a shareholder-officer’s compensation was reasonable. See id. at 1327-29; Elliotts, 716 F.2d at 1247. Moreover, the fact that only one or two officers received contingent compensation would not in itself render the formula unreasonable. See id.; Kennedy, 671 F.2d at 170, 174-75. Accordingly, and in light of our disposition of this case, on remand we direct the Tax Court to consider whether petitioner paid Lynn pursuant to an existing contingent compensation formula, whether the formula was reasonable at its inception, and whether it was applied consistently. d. Petitioner’s Many Roles Finally, petitioner argues that, in comparing his salary to the salaries paid at comparable companies, the Tax Court failed to consider properly the multiple roles that Lynn performed within Dexsil. The Tax Court appears to have rejected this argument, finding that: [w]e are satisfied that Lynn provided important management services that significantly contributed to the profitability of petitioner during the years at issue; however, we are not persuaded that the multiple “roles” that petitioner argues were occupied by Lynn are extraordinary for the president of a small corporation. On the other hand, the Tax Court ultimately found reasonable an amount considerably in excess of that paid to CEOs in comparable companies in recognition of Lynn’s significant contribution to Dexsil’s success. On *103 this record, we cannot determine if the court factored in Lynn’s roles or completely disregarded them. Given our disposition of this case, however, we think it worthwhile to note that while we agree with the Tax Court that the multiple roles performed by Lynn may not be extraordinary for the president of a small company, we nevertheless think that multiple roles are relevant in conducting a meaningful comparison to similar companies. After all, the inquiry at hand is whether the compensation paid was reasonable in light of the services performed. See Elliotts, 716 F.2d at 1246; Kennedy, 671 F.2d at 174-75; L & B Pipe, 67 T.C.M. at 2803-04. The Commissioner’s expert, David J. Bowering, prepared a report on the financial performance and compensation levels of, inter alia, seven “comparable companies.” The Tax Court found that this report constituted “the only objective evidence in the record of levels of compensation in companies that appear to be comparable.” However, Bowering’s report looked only at the compensation levels of the CEOs of each of these companies. It did not examine whether the CEOs at those companies worked as many hours and performed as many roles as Lynn did, or whether the companies paid separate compensation for the jobs of president, chief financial officer, or treasurer. We, think a meaningful comparison between the compensation paid by Dexsil and the compensation paid by similar companies requires such an analysis. See, e.g., Elliotts, 716 F.2d at 1245-46 (if employee, who worked 80 hours per week, “was performing the work of three people, the relevant comparison would be the combined salaries of those three people at another [company]”); L & B Pipe, 67 T.C.M. at 2803-04 (same for employees who worked 65-70 hours per week). CONCLUSION In conclusion, we find the Tax Court’s failure to assess the reasonableness of Lynn’s compensation from the perspective of a hypothetical or independent investor erroneous as a matter of law. Accordingly, we vacate and remand for reconsideration consistent with this opinion. The Tax Court is directed to make specific findings regarding the following questions: (1) whether a hypothetical investor would accept the compensation paid to Lynn; (2) whether Lynn was paid according to a long-standing and consistently applied contingent compensation formula, and if so, whether his salary was reasonable in light of this formula;, (3) whether Lynn’s compensation compared favorably with the compensation paid by similar companies for comparable services, given the many roles Lynn played at Dexsil; and (4) whether, after reconsideration of these factors, the balance of factors has shifted in favor of Dexsil such that it has met its burden of proving that Lynn’s compensation was reasonable. Footnotes [*] Calculated by dividing after-tax net income by year’s average shareholders’ equity [1] Other circuits have subdivided some of these factors and thus rely on nine, rather than five. The analysis, however, is similar. See, e.g., Rutter, 853 F.2d at 1271; Mayson Mfg. Co. v. Commissioner, 178 F.2d 115, 119 (6th Cir.1949).