996 F.2d 919
93-2 USTC P 50,368
UNITED STATES of America, Plaintiff-Appellee,
James HARVEY, Defendant-Appellant.
United States Court of Appeals,
Argued June 1, 1993.
Decided June 22, 1993.
William T. Grimmer, Asst. U.S. Atty. (argued), Office of the U.S. Atty., South Bend, IN, for plaintiff-appellee.
Mark S. Lenyo (argued), Allen, Fedder, Herendeen & Kowals, South Bend, IN, for defendant-appellant.
Before POSNER, COFFEY, and EASTERBROOK, Circuit Judges.
EASTERBROOK, Circuit Judge.
The Sentencing Guidelines use "tax loss" as the foundation for determining the appropriate sentence for tax evasion and other violations of the revenue laws. The Guidelines define "tax loss" as "28 percent of the amount by which the greater of gross income and taxable income was understated, plus 100 percent of the total amount of any false credits claimed against tax. If the taxpayer is a corporation, use 34 percent in lieu of 28 percent." U.S.S.G. § 2T1.3(a). This rough-and-ready calculation applies the highest marginal rate to the amount of concealed income, disregarding deductions that would have been available had the taxpayer filed an honest return.
Our case presents the question: What happens when a single crime causes both corporate and personal income to be understated? One common tax crime occurs when the manager and dominant investor in a closely held corporation distributes the firm's income to himself, not reporting it at either the corporate or personal level. Here is a simple example: the manager and principal shareholder of a corporation sells some of the corporation's inventory, producing a $100,000 profit that he diverts to his own use. This is a disguised dividend from the corporation, with income taxable at two levels: the corporation is supposed to pay an income tax on its profits, and the manager owes tax on the dividend income. United States v. Mews, 923 F.2d 67 (7th Cir.1991). Suppose the manager, responsible for both the corporation's and his own failure to pay, is convicted of tax evasion. There are at least three ways to compute the "tax loss": (a) apply the 28% rate to the manager's income of $100,000, for a tax loss of $28,000; (b) apply the corporate rate of 34% to the unreported profit of $100,000 and the personal rate of 28% to the unreported income of the same $100,000, for a tax loss of $62,000; (c) apply the corporate rate of 34% to the unreported profit of $100,000 and reduce the imputed dividend by the amount of imputed taxes, then apply the personal rate of 28% to the reduced dividend of $66,000, for a tax loss of $52,480. Method (c) assumes that the corporation, having paid $34,000 to the United States, distributes as a "dividend" only $66,000, the after-tax profit.
James Harvey pleaded guilty to tax evasion patterned after our example, except that the corporation distributed scrap aluminum rather than cash. Harvey sold the aluminum and kept the proceeds (some $81,000), reporting neither corporate nor personal income. The district court applied Method (b) to this amount. Insisting that Method (b) is "double taxation", Harvey argues for Method (a). We conclude that Method (c) is more accurate than either.
One may debate the wisdom of the corporate income tax, but our current system calls for sequential taxation of the same income. Corporate profits are taxed to the corporation, then taxed as personal income after distribution to the investors. Dividends, unlike repayments of debt, are not deductible at the corporate level even though both reflect the firm's cost of capital. When a single transaction bypasses both corporate and personal taxes, any effort to determine the "tax loss" must include both. (Harvey's corporations did not qualify under Subchapter S of the Internal Revenue Code, which ascribes corporate income directly to the investors and avoids sequential taxation.) Harvey refers to the last sentence of § 2T1.3(a): "If the taxpayer is a corporation, use 34 percent rather than 28 percent." This means, he insists, that the judge cannot apply both 28% and 34% to the same income. True enough: the personal rate is 28%, and the corporate rate 34%; no one has a rate of 62%. Still, the sentence speaks of "the taxpayer", not "the defendant" and does not speak directly to the person whose crime causes two taxpayers to understate their income and pay less to the Internal Revenue Service.
Application Note 3 to § 2T1.3 implies an answer to the multiple-taxpayer case in remarking: "In determining the total tax loss attributable to the offense ..., all conduct violating the tax laws should be considered as part of the same course of conduct or common scheme or plan unless the evidence demonstrates that the conduct is clearly unrelated." In other words, the Sentencing Commission wants the judge to consider the entire tax loss produced by the defendant's criminal conduct. If one person causes two taxpayers to understate their incomes, both underpayments count.
Reaching this point, the district court applied Method (b), which the prosecutor proposed and the presentence report endorsed. The difficulty is that Method (b) overstates the revenues lost to the Treasury. If the corporation has a profit of $100,000, it must pay $34,000 to the United States. (Here we disregard, as the tax loss table does, all deductions and credits.) Only $66,000 remains for distribution, and only the amount distributed as a dividend is subject to personal income taxation. If the investor-manager obeys the tax laws, the tax revenues will be $52,480: a 34% tax on $100,000 at the corporate level, and a 28% tax on the $66,000 the stockholder receives as a dividend. If this corporation nonetheless distributes $100,000 to an investor-manager, the $34,000 difference must be a return of capital, which is not taxable income (although it reduces the investor's basis in his stock). If we understand the extra $34,000 as an implicit wage or bonus to the recipient in his role as manager rather than an implicit dividend to the recipient in his role as investor, then this sum is deductible to the corporation, which may deduct compensation to its employees, and again sequential taxation does not occur. Method (c) will not be precise; characterizing the $34,000 increment in this example as a bonus rather than a return of capital would reduce the marginal rate from 34% to 28%; but the tax loss is not supposed to be a perfect reflection of actual tax revenues foregone. All in all, Method (c) approximates the loss better than Method (b). The district judge therefore should have deducted the (imputed) corporate income tax before determining the personal taxable income that Harvey was required to report.
As it happens, the difference between Method (b) and Method (c) does not affect Harvey's sentence. Explaining why requires a short tour through the numbers. The scrap aluminum that should have been reported as corporate income fetched Harvey $80,613, which also represented unreported personal income to Harvey. Applying Method (b) to the $80,613 yields a tax loss of almost $50,000. Under Method (a) the tax loss is $22,571. Method (c) implies a tax loss of $42,306. Harvey also concealed approximately $344,000 in additional personal income, representing a tax loss of roughly $96,000. Adding that $96,000 to the tax losses under each method comes to roughly $118,600 for Method (a), $146,200 for Method (b), and $138,500 for Method (c). The table in U.S.S.G. § 2T4.1 specifies that a tax loss between $70,000 and $120,000 translates to 12 offense levels, while a tax loss between $120,000 and $200,000 calls for 13 offense levels. Harvey thus gets the extra level under either Method (b) or Method (c). Nothing in the district judge's thorough sentencing memorandum suggests that the choice of a sentence within the guideline range turned on the precise calculation of the tax loss. Indeed, Harvey received the lowest penalty within the sentencing range. A remand is accordingly unnecessary--unless there is some other problem in the calculation.
Harvey insists that there is. The district court included about $58,000 of unreported personal income from 1985, although Harvey was not charged with any crime in connection with that tax year. Deleting all of the 1985 income would not depress the tax loss below $120,000, so this contention, too, turns out not to matter. The $58,000 accounts for less than $16,300 of the tax loss; subtract that from the $138,500 we have calculated, and $122,200 remains. We add for completeness that there was no error. Tax offenses, like embezzlements and drug crimes, fall within the rule that relevant conduct includes the whole scheme. Application Note 3 to § 2T1.3 refers to § 1B1.3(a)(2), the cornerstone of the relevant-conduct rules that have led to the inclusion of quantities of drugs in addition to those charged in the offenses of conviction. United States v. White, 888 F.2d 490, 497 (7th Cir.1989), explains how the Guidelines produce this result for drug offenses. Tax offenses work the same way. United States v. Daniel, 956 F.2d 540 (6th Cir.1992), does not disagree. That case holds only that the tax loss attributable to criminal activity for a given year must be reliably computed, and that civil tax liability is not an adequate substitute for "tax loss" under the Guidelines, which is limited to criminal understatements. Harvey does not contest the computation for 1985 or the presentence report's conclusion that failure to report that income was a crime.