This appeal mainly involves the issue of whether the losses and fees from certain “forward straddles” are deductible. The Tax Court disallowed these deductions and assessed a negligence penalty because it held that the forward contracts were sham transactions. The particular facts are exhaustively detailed in the published opinion below. Forseth v. Commissioner, 85 T.C. 127 (1985). The Enriéis and the Easter-lings, taxpayer-petitioners before the Tax Court, appeal.

In essence, the Tax Court held that Interact, LMEI and LMEC were taking advisory fees and margin “deposits” from the taxpayers; entering into sham forward contracts with the client taxpayers at terms unconnected to any real market, and without laying off the contracts with any real third-party brokers; creating artificial losses for the taxpayers by closing out “losing” positions at prices and terms set by LMEI/LMEC to generate a tax loss that largely offset the type and amount of in come the taxpayers needed to shelter; and closing out the “winning” position in a subsequent year at a price set by LMEI/LMEC to generate a net loss that conveniently equaled the margin deposit, which was really a disguised fee for creating artificial tax losses.

The Tax Court’s finding that the forward contracts were shams, although perhaps a mixed issue of fact and law, is reviewable under the clearly erroneous standard because it “is essentially a factual determination.” Thompson v. Commissioner, 631 F.2d 642, 646 (9th Cir.1980), cert. denied, 452 U.S. 961 , 101 S.Ct. 3110 , 69 L.Ed.2d 972 (1981); see also United States v. McConney, 728 F.2d 1195, 1202 (9th Cir.) (en banc), cert. denied, 469 U.S. 824 , 105 S.Ct. 101 , 83 L.Ed.2d 46 (1984). The clearly erroneous standard of review would govern even if the finding that the transactions were shams was inferred from undisputed basic facts concerning the transactions because the inference is still essentially factual. See Commissioner v. Duberstein, 363 U.S. 278, 291 , 80 S.Ct. 1190, 1199 , 4 L.Ed.2d 1218 (1960). The clearly erroneous standard is especially appropriate here where the finding was based in part on the Tax Court’s evaluation of conflicting evidence and live testimony. Likewise, we defer to the Tax Court’s finding that the Easterlings negligently claimed their deduction unless such finding is clearly erroneous. Lysek v. Commissioner, 583 F.2d 1088, 1094 (9th Cir.1978); see also McConney, 728 F.2d at 1204 .

The Tax Court based its finding that the transactions were shams on six factors.

(1) The close correlation of tax needs to losses. Tax losses were correlated to tax needs in two ways. One, the amount of losses “suffered” by each taxpayer largely equaled the amount of income each taxpayer needed to shelter. Two, the type of loss created (capital versus ordinary) matched the type of income needed to be sheltered. Thus, the Easterlings, who had ordinary income, received losses that were created in a manner that purportedly created ordinary losses, whereas the Enricis, who had capital gains income, received losses in a manner that purportedly created capital losses.

(2) Interacts ability to predict tax losses. Interact distributed informational packages that informed the taxpayers of the “possible” tax deductions that could be expected at various levels of margin deposit. Their accuracy in “predicting” these losses was suspicious.

(3) LMEI/LMEC “traded” before they had received margin deposits, which were supposedly their “insurance” against risk.

(4) The margin accounts were “zeroed-out” by closing out the taxpayers’ gains at a price that conveniently made the taxpayers’ overall losses equal or nearly equal the margin deposits the taxpayers initially placed with LMEI and LMEC.

(5) The record contained no documentary evidence to show that opposite positions were actually entered into with market-making brokers to lay-off trades and in fact did not even identify the supposed market-makers with whom LMEI/LMEC allegedly dealt.

(6) The trading records seemed to have been manipulated in certain circumstances.

Other evidence also suggested that no real transactions were being made, but the foregoing are the main grounds the tax court relied on. These grounds provide ample reason to infer that no real transactions were taking place. The taxpayers' main argument is that many of these grounds would also have existed even if the taxpayers had entered into straddles on the open commodities market. Further, they argue that the losses suffered in their forward straddles were in fact correlated to the losses and gains they would have suffered in the American commodities market. However, even if these factual assertions are accepted as true, the Tax Court was entitled to infer from both the precision of the tax advantages obtained and the undocumented nature of the transactions that the parties were merely rigging paper prices, losses, and gains to effectuate a sale of generated tax losses. The fact that Congress may have permitted similar tax advantages to be obtained from real commodities straddles does not make the losses from these sham transactions deductible. The absolute power to manipulate and dictate the price and timing of the artificial transactions, even if kept within the general range of the marketplace, allows a taxpayer (especially if he sets the price and timing after the stated date of the transaction) to reap larger and surer tax advantages with much less economic risk than he would have had he entered into real transactions. Thus, the claiming of deductions from artificial straddles that are designed to look similar to marketplace straddles, if left unfettered, has a much greater potential for abuse than deductions from marketplace straddles. The fact that the deductions allowed (at the time) from verifiable futures straddles were themselves abusive is hardly good reason for extending and intensifying that abusive potential by allowing deductions from sham straddles.

The Tax Court’s determination that the transactions were shams is thus upheld as not clearly erroneous. Accordingly, losses claimed on the forward contracts cannot be deducted. The taxpayers argue that even if the forward contracts were shams and the losses nondeductible, the fees paid to Interact should be deductible under I.R.C. § 212. But if the transactions with LMEI/LMEC were shams, then the fees were not spent for the production of income, nor for informational services concerning bona fide investments, nor for tax advice. They were spent for the generation of artificial tax losses and are nondeductible.

Finally, we affirm the Tax Court’s determination that Easterling is liable for a negligence penalty under I.R.C. § 6653(a). When, as in this case, the Commissioner assesses a negligence penalty under § 6653(a), the taxpayer has the burden of showing that he was not negligent. See Rapp v. Commissioner, 774 F.2d 932, 935 (9th Cir.1985). The determination of the Tax Court that Easterling did not meet this burden is not clearly erroneous in light of the fact that Easterling, if he was not a knowing participant, obviously was paying no attention to the rather suspect workings of LMEI/LMEC.

AFFIRMED.

. Because the Tax Court held the transactions were shams, it never reached the issues of whether they were entered into for profit (even losses on real straddles cannot be deducted unless the investor has a reasonable expectation of profit, see Wehrly v. United States, 808 F.2d 1311, 1312 (9th Cir.1986)) or whether the losses were capital or ordinary losses. See 85 T.C. at 166 n. 43. . Interact and LMEC/LMEI asserted that losses created by cancellation could be claimed as ordinary losses whereas losses created by offset could only be claimed as capital losses. Whether or not this assertion is legally correct has not been decided.