ORDER

Dennis Cesario pleaded guilty to bank fraud, 18 U.S.C. § 1344, and conspiracy to defraud the United States by impeding the IRS from collecting tax revenue, id. § 371. The district court imposed concurrent 40-month terms of imprisonment, and Cesario filed a notice of appeal. His lawyer, however, now moves to withdraw on the ground that he cannot discern a nonfrivolous basis for appeal. See Anders v. California, 386 U.S. 738, 87 S.Ct. 1396, 18 L.Ed.2d 493 (1967). Counsel’s brief is facially adequate, and Cesario has not responded under Circuit Rule 51(b). We therefore limit our review to the potential issues flagged by counsel. See United States v. Maeder, 326 F.3d 892, 893 (7th Cir. 2003) (per curiam); United States v. Tabb, 125 F.3d 583, 584 (7th Cir. 1997) (per curiam).

The two possibilities identified by counsel are related to Cesario’s sentence for conspiring to impede revenue collection. The two-year conspiracy started in 1995 when Cesario and his codefendant, Alex Leito, began targeting delinquent taxpayers with offers to negotiate their tax liabilities with the IRS. Styling themselves as tax professionals, Cesario and Leito in each case proposed to submit an “offer in compromise” on behalf of the taxpayer and, a few months later, produced phony documents suggesting that the IRS had accepted the deal.

The “offers” actually were rejected by the IRS or simply never made, but the taxpayers, unaware of the scam, still transferred funds to Cesario and Leito that were intended for the benefit of the IRS. The pair collected a total of more than $600,000-money that they promised to turn over to the IRS but pocketed instead. Cesario and Leito meanwhile had a corrupt IRS agent place a “stay-up” on each taxpayer’s file, which temporarily halted collection efforts and made it appear that the taxpayer’s dispute indeed had been settled. The defendants also furnished forged papers to four of the taxpayers purporting to release tax liens on their property. The taxpayers then used these documents (presumably by recording them) to sell the encumbered property, which extinguished the liens.

At sentencing the district court increased Cesario’s offense level by four on account of two offense characteristics listed in the 1997 version of U.S.S.G. § 2T1.4. (The parties agreed in the district court that the 1997 guidelines govern, see id. § lBl.ll(b)(l), and Cesario conceded in his plea agreement that § 2T1.4 applies to his conspiracy conviction, see id. § 2T1.9(a)(l).) First, the district court concluded that Cesario’s crimes were “part *358of a pattern or scheme from which he derived a substantial portion of his income.” Id. § 2T1.4(b)(l)(A). Second, the judge determined that Cesario used “sophisticated means” in an effort “to impede discovery of the existence or extent of the offense.” Id. § 2T1.4(b)(2).

In his brief counsel considers whether to argue that the district court erroneously reached both conclusions. On the question whether a substantial portion of Cesario’s income came from the tax fraud, the presentence investigation report reflects that from 1995 to 1997 his legitimate businesses generated negligible income. And at sentencing Cesario introduced no evidence to contradict this assessment. Indeed, he testified that during the course of the fraudulent scheme he received at least $300,000 from the taxpayers and had no legitimate income.

Cesario did object at sentencing that he had no obligation to estimate what portion of his income came from legitimate sources. To this end, he observed that the government, not the defendant, ordinarily must prove factors justifying a heavier sentence. See United States v. Foutris, 966 F.2d 1158, 1160 (7th Cir. 1992). But Cesario effectively withdrew that objection when he finally agreed to speak about his sources of income. See United States v. Vivit, 214 F.3d 908, 916 (7th Cir. 2000) (regardless of the government’s burden of proof, the district court may consider any reliable evidence at sentencing). And in any event, the district court could infer from the PSR alone that Cesario had little legitimate income. We therefore agree with counsel that it would be frivolous to argue that the district court’s assessment of Cesario’s income was clearly erroneous. See, e.g., United States v. Searan, 259 F.3d 434, 448-49 (6th Cir. 2001) (having negligible legitimate earnings is “conclusive evidence” that illegitimate earnings composed a substantial portion of the defendant’s income).

Counsel also considers whether to contest the district court’s finding that Cesario used “sophisticated means” to conceal the tax fraud. This finding also would be reviewed for clear error, United States v. Kontny, 238 F.3d 815, 821 (7th Cir. 2001); United States v. Friend, 104 F.3d 127, 129 (7th Cir. 1997), and requires evidence of concealment beyond what is inherent in ordinary tax frauds, Kontny, 238 F.3d at 820-21; see also United States v. Furkin, 119 F.3d 1276, 1284-85 (7th Cir. 1997); Friend, 104 F.3d at 129-30. Standard examples include hiding assets in offshore bank accounts and conducting business through corporate shells or fictitious entities. See U.S.S.G. § 2T1.4, comment. (n.3) (1997). The cases also show that fabricating receipts, tax returns, business records, and other documents can be sophisticated methods of hiding tax fraud. United States v. Utecht, 238 F.3d 882, 889 (7th Cir. 2001); United States v. Madoch, 108 F.3d 761, 766 (7th Cir. 1997); United States v. Wu, 81 F.3d 72, 73-74 (7th Cir. 1996).

Here Cesario and Leito provided some of the taxpayers with documents purporting to release tax liens on their property. The phony releases were made with seals and signature stamps stolen from IRS agents who actually had authorization to release tax liens. According to the government, these documents hampered discovery of the fraud by convincing the taxpayers that Cesario and Leito had resolved their obligations to the IRS. The district court accepted this argument, noting that as long as the taxpayers remained unaware of the defendants’ scheme, they would have had no reason to contact the IRS and report the fraud.

On a different record, the district court’s analysis might have raised nonfrivolous *359questions. The PSR, for example, reflects that government agents viewed all but one of the taxpayers as “willing victims”-which could be understood to mean that the taxpayers did not care how the defendants went about reducing their tax liabilities. This characterization could have undercut the district court’s conclusion that the forged releases prevented the taxpayers from promptly discovering and reporting the fraud. For if the taxpayers expected Cesario and Leito to use their money illegally (by bribing IRS officials, for instance), they probably would have thought twice before rushing to authorities with evidence that the defendants’ scam frustrated their own plans to commit tax fraud. (The one taxpayer who was not a “willing victim” did not receive a phony release of his tax liens.)

Yet there was no suggestion in the district court that the taxpayers would have had any reason to ignore the defendants’ misconduct if they learned of it. The sums collected during the fraud (at a minimum nearly 20% of each taxpayer’s liability excluding interest and penalties) do not suggest that Cesario and Leito made extravagant promises, which might have raised a question whether the taxpayers believed that legitimate negotiations with the IRS were in store. And Cesario’s plea agreement makes clear that he and Leito promised to tender the taxpayers’ money to the IRS to pay their deficiencies. So we agree with counsel that the district court’s finding rests on a permissible view of the evidence, which would render frivolous any contention that the court committed clear error.

The motion to withdraw is GRANTED, and the appeal is DISMISSED.