United States v. Ledesma, 313 F. Supp. 2d 662 (S.D. Tex. 2004)

*663 Michael Rodriguez, Attorney at Law, Brownsville, TX, for Gustavo Ledesma.

Reynaldo S. Cantu, Assistant Federal Public Defender, Brownsville, TX, for Francisco Ledesma.

David H. Peck, Assistant United States Attorney, Houston, TX, for United States of America.

HANEN, District Judge.

On the 8th day of March, 2004, came to be heard the final sentencing hearing on the above referenced case. Gustavo and Francisco Ledesma ("the defendants") pleaded guilty to a violation of Title 18 U.S.C. § 371 conspiracy to defraud the United States by preparation of false tax returns. Both the Government and the defendants objected to the section of the Presentence Investigation Report ("PSI") that assigns an eighteen base offense level, predicated upon the finding that the tax fraud scheme implemented by the defendants resulted in a total tax loss of $626,905.60. Both the Government and the defendants proposed to the court a mode of calculating the tax loss. For the reasons stated below, the court holds that the proper method of computing the loss in this particular case is to add the figure of $198,832 which represents the tax loss actually calculated by the Internal Revenue Service ("IRS") using eighty-five actual audited returns of the 268 returns at issue to the figure of $382,853.98 which is the net amount of the Rapid Anticipation Loan ("RAL") checks illegitimately paid to the defendants for a total tax loss of $581,685.98. This figure compels a base offense level of eighteen, which the court utilized in assessing the guideline application of each defendant.

In a factual nutshell, the defendants acted as tax return preparers in Raymondville, Texas. They operated under the business name of Ledesma Fast Filing. In that capacity, they prepared a number of tax returns over a period of years in the 1990s and, in the late 1990s, adopted a practice of electronically filing such returns. As part of the scheme, at the defendants' behest, the taxpayers signed a blank United States Individual Income Tax Declaration for Electronic Filing form (IRS Form 8453). The Ledesma clients routinely took advantage of a local bank's Refund Anticipation Loan ("RAL") program. Under the RAL program, a taxpayer, on the basis of his income tax return, could borrow money from a financial institution based upon the refund he was eventually to receive.

*664 While there were exceptions to the fact pattern utilized, the defendants had a basic scheme they employed to perpetrate this tax fraud. The defendants would give their address (Ledesma Fast Filing) as the taxpayer's home address. Next, the defendants would prepare false tax returns by illegally maximizing the client's earned income credit ("EIC"). They would do this by including children (real children with existing social security numbers), albeit not the taxpayer's children, on these returns to maximize these credits. Most of the evidence indicates that the taxpayers would receive a made-up or false copy of what would be an accurate return, but that the actual e-filed return would contain these additional dependents. (Certain evidence implies that some of the taxpayers may have known about this scheme.) The inclusion of these fictional dependents effectively maximized the EIC, lessened the tax liability, and ultimately resulted in a larger return.

The defendants could then obtain the RAL checks from Bank One in Raymondville. When the RAL checks would arrive, the defendants would forge the endorsement and deposit the checks into their own account. Sometimes the taxpayers would receive the first RAL check and the defendants would receive a later check. Other times the defendants would merely pay the taxpayers the amount the taxpayers anticipated from the return and then pocketed the amount received from the fraudulent return. The scheme varied in certain details from taxpayer to taxpayer, but ultimately the end result was a fraudulent tax return resulting in the under-estimation of tax liability and overpayment of refunds to the benefit of the defendants.

The IRS identified 268 falsely prepared tax returns spanning the period of 1997 through 1999. The IRS was able to audit eighty-five of those returns and from those audited returns determine a total tax loss of $198,832. Due to time constraints and the number and mobility of the taxpayers in question, it became impractical for all taxpayers to be audited.

The court held an initial hearing on sentencing, where testimony on the issues regarding the sentencing of both defendants was presented and also held a second hearing on March 8, 2004. Both defendants had originally lodged various objections to the PSI; however, both withdrew and waived those objections (with one exception) at the start of the hearing. The sole remaining objection was aimed at the portion of both reports which set the base offense level at eighteen, based upon the finding that the tax fraud scheme perpetrated by the defendants resulted in a tax loss of $626,905.60. The guideline in question for the defendants' violation of 18 U.S.C. § 371 is found in the 1998 United States Sentencing Guidelines at Section 2T1.9.

The Government proposed two different methods to measure the impact of the defendants' conduct. One method results in a calculated tax loss of $626,905.60, the number adopted by the PSI. The Government arrived at this number by suggesting that the court adopt the figure of $198,832 (the figure calculated by the IRS as the loss for the eighty-five audited returns) and also calculate the loss for the remaining 183 returns and add that figure to the $198,832 loss. Specifically, the government asked the court to calculate the tax loss as follows:

  $   198,832  the amount of tax loss determined
                in the audit of eighty-five returns
                plus
  $426,073.60  representing the 183 unaudited
                tax returns multiplied by
                $2,239.50, which is the average
                loss per return in the audited
                batch determined by dividing the
                known loss of $198,832 by the
                eighty-five audited returns
  -----------
  $626,905.60   Total loss

*665 The second method suggested by the Government results in a lower number, but one which also falls within the same guideline level. The defendants objected to the use of any figure above $198,832, which they claim is the actual known loss determined by audit. They requested that no amount be added for the 183 unaudited tax returns because they consider any figure to be too speculative to be meaningful.

The court, after holding two hearings on this matter, overrules the defendants' objection based upon Section 2T1.9 et sq. of the United States Sentencing Guidelines ("U.S.S.G." or "Guidelines"). Specifically, that section of the Guidelines provides for penalties for offenses involving taxation. Tax offenses are governed by Section 2T1.9 of the Guidelines, which states that offense levels are determined by Section 2T1.1 or Section 2T1.4. See U.S.S.G. § 2T1.9. Section 2T1.4, which details the offense levels for "Aiding, Assisting, Procuring, Counseling, or Advising Tax Fraud," refers the court to Section 2T1.1 for defining tax loss. See U.S.S.G. § 2T1.4(a).

Section 2T1.1(c) (1) states that "[i]f the offense involved tax evasion or a fraudulent or false return ... the tax loss is the total amount of loss that was the object of the offense." U.S.S.G. § 2T1.1(c) (1). No fixed formula is set out for determining this amount, nor is the word "loss" defined. Nevertheless, guidance is found in the commentary wherein it states that "[i]n some instances, such as when indirect methods of proof are used, the amount of tax loss may be uncertain; the guidelines contemplate that the court will simply make a reasonable estimate based upon the available facts." U.S.S.G. § 2T1.1, cmt.n.1

The case law in other Circuits supports the approach proposed by the Government in this case. In fact, in United States v. Bryant, 128 F.3d 74 (2nd Cir. 1997), the Second Circuit instructed that "... it is permissible for the sentencing court, in calculating a defendant's offense level, to estimate the loss resulting from his offenses by extrapolating the average amount of loss from known data and applying that average to transactions where the exact amount in unknown."^[1]Id. at 76 (emphasis added). More directly on point, albeit in unpublished opinions, both the Second and Fourth Circuits upheld the approach of using known figures obtained from audited figures as a means to estimate unknown figures from unaudited returns. United States v. Maye, 205 F.3d 1335, 2000 WL 223344 (4th Cir.2000) (unpublished); United States v. Bruno, 234 F.3d 1263, 2000 WL 1715254 (2nd Cir.2000) (unpublished).

Thus, the approach and figures suggested by the Government are perfectly reasonable and would be acceptable to the court absent the fact that in this particular instance there may be a more exact way of calculating the losses incurred. Accordingly, while it has overruled the defendants' objection, the court does not adopt the Government's proposed methodology in toto.

The court accepts, as do the defendants, the figure of $198,832, which represents the exact losses determined by the eighty-five audits performed by the IRS. The IRS, however, also determined the identity of unaudited returns by tracking the payment of the RAL checks, which was the means utilized by the defendants to perpetuate the fraud. The amount of the RAL checks for the unaudited accounts is $433,393.48. Further, the IRS identified $50,539.50 as the amount that the taxpayers *666 were legitimately owed and paid, leaving $382,853.98 as the amount that was paid illegitimately to the defendants. The court also adopts this figure of $382,853.98 as the losses resulting from the scheme utilized in the 183 unaudited returns.

Both sides have suggested to the court possible problems with this approach. The defendants, while maintaining that the court should not assign any tax loss to the 183 unaudited returns, suggested that they made more payments (in cash) to clients that were not included in the figure of $50,539.50. Their evidence in this regard is extremely sketchy, and the court finds it too speculative and vague to consider. The defendants further contended that the court should not include all of the RAL checks, yet put forth no credible reason in support of this position other than arguing an audit of all returns would be the only way to produce a precise number. While this might, in fact, be true, it is not required by the Guidelines, nor is it reasonable in light of the kind of offense in question. Indeed, if absolute precision were required, one could also attack the audited numbers, by arguing that different auditors might come up with different results.

The Government also questions the court's approach, arguing in favor of its own averaging method. The basis for its objection is that the court is combining two different calculations resulting in two different kinds of loss actual calculated losses ($198,832) and actual ill-gotten dollars ($382,853.98) for a total loss of $581,685.98. This court readily admits it is doing just that, but it is doing so because it will result in a more precise less estimated number. (The court also notes that the Government alternatively argued in favor of a similar procedure in at least a portion of it pleadings.)

In adopting this methodology, the court holds that the approach put forth by the Government one which estimates an unknown amount by using an average derived from known calculated losses is reasonable and is sanctioned both by case law and by the guidelines. Nevertheless, the court chooses not to utilize the averaging method in this specific case solely because the loss derived using its methodology is much more accurate (and thus fairer to the defendants) and reduces the imprecision that might result from utilizing an average. The Guidelines prescribe that "[t]he court should use any method of determining the tax loss that appears appropriate to reasonably calculate the loss ...." U.S.S.G. § 2T1.1 App. 1 Note 1. Given the facts of this case, the method this court has employed in this case is the superior of the two, both of which comport with the guidelines.

Accordingly, the court holds that $581.685.98 represents the total tax loss. Under the applicable guidelines, this loss figure results in a base offense level of eighteen, which the court hereby uses in calculating the guideline application of each defendant.

[1] The Second Circuit had earlier adopted such a procedure in United States v. Sutton, 13 F.3d 595 (2nd Cir.1994), in a case governed by Chapter 2F of the Guidelines.