In August 1999, Edwin David Wood, II was indicted on 22 counts of wire fraud, mail fraud, and money laundering. These charges arose out of two loan transactions that Wood arranged in 1994. Wood pleaded not guilty. After a jury found Wood guilty on 21 of the 22 counts, the district court sentenced him to 168 months of imprisonment and ordered him to pay approximately $570,000 in restitution. Wood appeals, arguing that (1) certain jury instructions were erroneous, (2) venue in the Western District of Michigan was improper with regard to five of the seven mail fraud charges, (3) the district court's restitution order was excessive, and (4) the government failed to produce sufficient evidence concerning one of the loan transactions to support his conviction. For the reasons set forth below, we AFFIRM Wood's conviction and sentence on all counts other than the mail fraud charges in Counts 17-21, REVERSE Wood's conviction on these latter counts for lack of proper venue, and, accordingly, REMAND for resentencing.

The charges against Wood arise out of two loan agreements executed by him in 1994. Wood orchestrated these transactions from the Metropolitan Correctional Center (MCC) in Chicago, a federal prison, between June 15 and October 6 of 1994. While at the MCC, Wood worked through First Financial Acceptance Company, Inc., which had offices in Battle Creek, Michigan. First Financial personnel did not disclose to the potential borrowers that Wood was in prison as a convicted felon.

To attract customers, First Financial advertised in publications such as the New York Times, the Wall Street Journal, and Investors Business Daily, offering to lend up to 90% of the value of any eligible stock that the prospective borrower owned. These ads generated a large number of telephone inquiries. Wood kept track of these inquiries by calling First Financial from prison and then having First Financial personnel connect him to the prospective borrower through call forwarding. Interested callers were sent a package of documents compiled by Wood. These documents included a Security and Pledge Agreement, which provided in pertinent part as follows:

Lender [First Financial] shall hold the pledged shares as security for the repayment of the loan and shall not encumber the shares except in accordance with the provisions of this Agreement.... In the event that Pledgor defaults in the performance of any of the terms of this Agreement, ... Lender may, upon five (5) days advance written notice to Pledgor, sent by certified mail, and without liability for any diminution in price which may have occurred, sell all of the pledged shares in such manner and for such price as Lender may determine, at any bona fide public sale, and Lender shall be free to purchase all or any part of the pledged shares at such sale.

First Financial utilized a technique called "selling short against the box" to fund the loans. (For a detailed description of short-against-the-box transactions, see Edward D. Kleinbard, Risky and Riskless Positions in Securities, TAXES, December 1993, at 788). Neither in the Agreement nor in any other communication with prospective borrowers did First Financial disclose the use of short-against-the-box transactions to fund the requested loans.

The first loan transaction at issue was with Robert Graham. Graham wanted to retire a debt with the $230,000 that he intended to borrow from First Financial. He pledged 70,000 shares of Sonic Environmental Systems stock as collateral. Graham received a loan commitment from First Financial for the $230,000 in June of 1994, but he did not receive any of the loan proceeds until November of that year. At that time, Graham received approximately $31,000. First Financial made no other disbursements on its $230,000 commitment.

Meanwhile, on September 13 and 14 of 1994, a broker acting on Wood's instructions sold short 35,000 shares of Graham's Sonic Environmental Systems stock for $173,191. The sale of Graham's stock was supposed to be a short-against-the-box sale, but because the broker was unable to borrow 35,000 shares of Sonic Environmental Systems stock in the open market, a short-against-the-box transaction could not be executed. Instead, the short position was closed by selling the 35,000 shares on the open market in October of 1994.

Graham was not told that half of his collateral had been sold when he attempted to get his stock back in late 1994. First Financial returned the remaining 35,000 shares of Graham's collateral to him in January of 1995. Graham later sold the stock and paid the approximately $31,000 balance due on his loan. He never received the rest of his collateral back.

The second transaction at issue in this case arises from a loan First Financial made to Gordon Miller and his mother, Ruth Miller. Ruth Miller pledged 8,167 shares of her Comerica Bank stock as collateral for the loan, and Gordon Miller received approximately $200,000 in loan proceeds in September of 1994. First Financial sold Ruth Miller's 8,167 shares of stock short against the box two days after closing on the Millers' loan. It received approximately $227,000 from the sale. The government presented evidence at trial that, at Wood's direction, First Financial closed its short position in Comerica by delivering Ruth Miller's stock in December of 1994 to the broker who provided the shares sold short. Although the stock was thus disposed of by First Financial within three months of closing the loan transaction, Gordon Miller continued to make loan payments through June of 1996. Ruth Miller never recovered her collateral from First Financial.

In August of 1999, a federal grand jury returned a 22 count indictment that charged Wood with 7 counts of wire fraud in violation of 18 U.S.C. § 1343, 7 counts of mail fraud in violation of 18 U.S.C. § 1341, 3 counts of money laundering in violation of 18 U.S.C. § 1956(a)(1)(A)(i), 2 counts of money laundering in violation of 18 U.S.C. § 1956(a)(1)(B)(i), and 3 counts of engaging in monetary transactions with proceeds of a specified unlawful activity in violation of 18 U.S.C. § 1957. Arguing that the district court lacked venue to try certain mail fraud charges, Wood sought dismissal of most of the mail fraud counts before trial, at the close of the government's case, and at the close of all of the evidence. The district court denied all of these motions. Wood also moved for acquittal under Rule 29 of the Federal Rules of Criminal Procedure at the end of the government's case and at the close of all the evidence. The district court denied these motions as well. After an eight-day trial in April of 2001, a jury convicted Wood on 21 of the 22 counts. He was later sentenced to 168 months of imprisonment and ordered to pay $570,025.83 in restitution. The district court entered judgment on October 26, 2001. This timely appeal followed.

Wood claims that the district court erred in instructing the jury that short sales against the box encumber the stock pledged as collateral. He maintains that, as a matter of law, such sales do not encumber the stock, thus making the court's instructions to the jury erroneous. Wood concludes that the jury "was charged in a manner that allowed them to convict him on insufficient grounds — for legal conduct."

Because Wood did not object to the jury instructions at trial, we review the instructions under the "plain error" standard, considering whether "the instructions, when taken as a whole, were so clearly wrong as to produce a grave miscarriage of justice." United States v. Sanderson, 966 F.2d 184, 187 (6th Cir.1992). Plain error is a stringent standard. See United States v. Cox, 957 F.2d 264, 267 (6th Cir.1992). This court in Cox noted that the Supreme Court and numerous federal courts caution that the plain error doctrine should be applied only "in exceptional circumstances, and solely to avoid a miscarriage of justice. Recourse may be had to the doctrine only on appeal from a trial infected with error so `plain' the trial judge and prosecutor were derelict in countenancing it." Id. (quotation marks omitted).

Wood claims that the following instruction was erroneous:

The government has alleged, in part, that Defendant Wood engaged in a scheme or plan to obtain money or property by means of false or fraudulent pretenses, representations, or promises and, more particularly, that a short sale against the box encumbers the securities in the box.

The government concedes that this portion of the district court's instruction suggests that a short sale against the box is a fraudulent transaction in and of itself. What followed this portion of the instruction, however, is significant in determining whether the instruction as a whole was erroneous. The court continued:

A short sale is made by selling a security that the seller does not own. This is different from a short sale against the box. A short sale is "against the box" to the extent that the separate account contemporaneously owns or has the right to obtain at no added cost securities identical to those sold. The owner of the account is considered the owner of the shares deposited "in the box."

The shares can be returned in one of two ways:

1. By delivering the shares held "in the box" — or in the account; or

2. By purchasing in the market identical shares. If the market value is higher than the amount received in the short sale, then the account owner must come up with more money than he received from the short sale.... If the market value of the short sale goes down, then the account owner must come up with less money than he received from the short sale.... [I]f the account owner is required to deliver the shares "in the box" to a third person, like Mrs. Miller, for example, after she has repaid her loan, he has to do two things:

a. Deliver the equivalent shares to the third person-that's Mrs. Miller and

b. Cover with cash or shares the shares that were in the box so that the broker is not at risk on the money paid to the account holder at the time of the original short sale.

This instruction, when read in its entirety, explained how the short sale against the box had to be covered so that the collateral could be returned once the loan was repaid. Moreover, the jury instructions as a whole detailed the elements of the alleged crimes and did not imply that a short sale against the box was inherently a fraudulent transaction. We therefore conclude that the jury instructions were not "so clearly wrong as to produce a grave miscarriage of justice." United States v. Sanderson, 966 F.2d 184, 187 (6th Cir.1992).

Wood claims that because the government did not prove by a preponderance of the evidence that the checks forming the basis of the mail fraud charges in Counts 17 through 21 were sent from, received in, or moved through the Western District of Michigan, venue in that district was improper. Accordingly, Wood urges this court to reverse his convictions on these counts.

Counts 17 through 21 allege that Wood, on five separate occasions, "did cause to be deposited, sent and delivered by mail ... from Battle Creek, Michigan, to Ruth Miller in Birmingham, Michigan, [checks] purporting to be Mrs. Miller's dividend check from her Comerica stock being used as collateral for the First Financial loan to her son." Before trial, Wood moved unsuccessfully for the dismissal of Counts 17 through 21. At the end of the government's case and again at the close of all evidence, Wood moved, pursuant to Rule 29 of the Federal Rules of Criminal Procedure, to dismiss Counts 17 through 21 because of improper venue. The district court denied these motions. Count 16 was also part of the Rule 29 motions, but Wood does not include that count in his argument on appeal.

Both the United States Constitution and Rule 18 of the Federal Rules of Criminal Procedure provide that a person can be tried for a crime only where that crime was committed. United States v. Cabrales, 524 U.S. 1, 6-7, 118 S. Ct. 1772, 141 L. Ed. 2d 1 (1998). The Constitution requires that all criminal trials "shall be held in the State where the said Crimes shall have been committed." U.S. CONST. art. III, § 2, cl. 3. A similar guarantee is found in the Sixth Amendment: "In all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury of the State and district wherein the crime shall have been committed." U.S. CONST. amend. VI. Rule 18 of the Federal Rules of Criminal Procedure likewise provides that "[u]nless a statute or these rules permit otherwise, the government must prosecute an offense in a district where the offense was committed."

"[T]he locus delecti [of the crime charged] must be determined from the nature of the crime alleged and the location of the act or acts constituting it." Cabrales, 524 U.S. at 6-7, 118 S. Ct. 1772. In determining the "locus delecti" of a crime, the Supreme Court directs us to "initially identify the conduct constituting the offense (the nature of the crime) and then discern the location of the commission of the criminal acts." United States v. Rodriguez-Moreno, 526 U.S. 275, 279, 119 S. Ct. 1239, 143 L. Ed. 2d 388 (1999). Venue is therefore appropriate only in the district where the conduct comprising the essential elements of the offense occurred. The government must prove that venue was proper as to each count charged. United States v. Crozier, 259 F.3d 503, 519 (6th Cir.2001). Because venue is an essential aspect of the government's case, "[i]f the government does not establish venue and the defendant objects at trial, then an appellate court must reverse the conviction." United States v. Scaife, 749 F.2d 338, 346 (6th Cir.1984).

For venue purposes, this court has held that mail fraud is a continuing offense governed by 18 U.S.C. § 3237(a). United States v. Holt, 899 F.2d 15, 1990 WL 37613, at *1-2 (6th Cir. April 3, 1990) (unpublished opinion). Section 3237 reads in pertinent part as follows:

(a) Except as otherwise expressly provided by enactment of Congress, any offense against the United States begun in one district and completed in another, or committed in more than one district, may be inquired of and prosecuted in any district in which such offense was begun, continued, or completed.

Any offense involving the use of the mails, transportation in interstate or foreign commerce, or the importation of an object or person into the United States is a continuing offense and, except as otherwise expressly provided by enactment of Congress, may be inquired of and prosecuted in any district from, through, or into which such commerce, mail matter, or imported object or person moves.

The substantive offense of mail fraud, on the other hand, is defined by the following portion of the federal mail fraud statute:

Whoever having devised or intending to devise any scheme or artifice to defraud ... for the purpose of executing such scheme or artifice or attempting so to do, places in any post office or authorized depository for mail matter, any matter or thing whatever to be sent or delivered by the Postal Service, or deposits or causes to be deposited any matter or thing whatever to be sent or delivered by any private or commercial interstate carrier, or takes or receives therefrom, any such matter or thing, or knowingly causes to be delivered by mail or such carrier according to the direction thereon, or at the place at which it is directed to be delivered by the person to whom it is addressed, any such matter or thing, shall be fined under this title or imprisoned not more than 20 years, or both.

18 U.S.C. § 1341.

Appropriate venue under the mail fraud statute is far from clear in the present case because the government did not allege that the checks at issue in mail fraud Counts 19-21 were either mailed from or received in the Western District of Michigan. Rather, the government argues that venue in the Western District of Michigan was appropriate because the offices of First Financial, located in the Western District of Michigan, were used at every step of the scheme to defraud.

As a starting point, we note that in discussing the mail fraud statute, court decisions have focused on the use of the mails. The Ninth Circuit, for example, has stated that the mail fraud statute protects the instrumentality of communication "making the use of the mails ... as part of a fraudulent scheme an independent offense quite separate from any other potentially illegal conduct." United States v. Garlick, 240 F.3d 789 (9th Cir.2001) (emphasis in original). See also United States v. Monostra, 125 F.3d 183, 187 (3d Cir.1997) (comparing the mail and wire fraud statute to the bank fraud statute and commenting that "the mail and wire fraud statutes do not penalize the victimization of specific persons; rather, they are directed at the instrumentalities of fraud.") (emphasis added) (citations and quotations marks omitted).

This concentration upon the use of the mails makes sense, given that the use of the mails is what establishes federal jurisdiction in these cases. Although "the use of the mails need not be an essential element of the [fraudulent] scheme," Schmuck v. United States, 489 U.S. 705, 710, 109 S. Ct. 1443, 103 L. Ed. 2d 734 (1989), such use is essential to federal jurisdiction, United States v. Mikell, 163 F. Supp. 2d 720, 729 (E.D.Mich.2001). Without the use of the mails, after all, the fraudulent scheme would be only a state-law offense.

We have found only three cases that have directly addressed a situation comparable to the present case where the "base" location for the fraudulent scheme was in a different district than the place of the mailings. Two of these cases rejected the government's expansive view of mail fraud venue. See Kreuter v. United States, 218 F.2d 532, 534 (5th Cir.1955) ("[T]he gravamen of the offense is the use of the mails. Therefore, [defendant] could be tried where he caused a letter to be either mailed or delivered in furtherance of the scheme. The place where the scheme is conceived or put in motion is immaterial, it is the place of mailing or delivery by mail."); United States v. Mikell, 163 F.Supp.2d at 728 (holding that venue was inappropriate "for a prosecution under § 1314 in a district where the mail matter did not move from, through, or into, but where an aspect of the scheme to defraud was devised and executed").

The third case is United States v. Olen, 183 F. Supp. 212 (S.D.N.Y.1960), which, although not a mail fraud case, discusses the general venue statute for continuing crimes, 18 U.S.C. § 3237(a). In Olen, the district court granted the defendant's request to transfer venue in a securities fraud case to the Southern District of Alabama where the scheme to defraud was located, even though the mailings in furtherance of that scheme occurred in the Southern District of New York. Id. at 218. The district court focused on the second paragraph of § 3237(a) that prescribes venue for mail fraud prosecutions, even though mail fraud was not charged, yet ignored the first paragraph of § 3237(a) that appears to justify the court's decision. We frankly find Olen's reasoning unpersuasive as it relates to the present case.

Also relevant to our inquiry are two Supreme Court decisions, neither of which involved the mail fraud statute, but both of which resolved venue challenges in a criminal case. One is United States v. Cabrales, 524 U.S. 1, 118 S. Ct. 1772, 141 L. Ed. 2d 1 (1998), where the government prosecuted Cabrales in Missouri for illegal financial transactions in Florida. The Supreme Court held that venue in Missouri was inappropriate, despite the fact that the source of the funds was from drug trafficking in that state, reasoning in part that the money laundering statutes proscribed "only the financial transactions (acts located entirely in Florida), not the anterior criminal conduct that yielded the funds allegedly laundered." Id. at 7, 118 S. Ct. 1772.

The other relevant decision is United States v. Rodriguez-Moreno, 526 U.S. 275, 119 S. Ct. 1239, 143 L. Ed. 2d 388 (1999), where a drug distributor and his henchmen kidnaped another drug dealer in Texas and held him captive as they traveled through several states, including New Jersey. Rodriguez was tried in New Jersey "for using or carrying a firearm during and in relation to any crime of violence." Id. at 276, 119 S. Ct. 1239 (citation and quotation marks omitted). The Supreme Court rejected Rodriguez's claim that venue was appropriate only in Maryland, where the gun was actually possessed. It reasoned that kidnaping is a continuing offense, causing venue to lie in any district where the offense "was begun, continued, or completed" as provided in the first paragraph of 18 U.S.C. § 3237(a). Id. at 282, 119 S. Ct. 1239. Aside from the fact that neither Cabrales nor Rodriguez-Moreno involved the mail fraud statute, they point in different directions as to the appropriate resolution of the case before us.

In the present case, the fraudulent scheme was not separable from the mailings, which is the very reason why Cabrales is not directly on point. If it were, then the venue issue would not be an open question for us to resolve. But Rodriguez-Moreno is not directly on point either, because the decision relied solely upon the first paragraph of 18 U.S.C. § 3237(a), which provides, in pertinent part, that "any offense against the United States begun in one district and completed in another, or committed in more than one district, may be inquired of and prosecuted in any district in which such offense was begun, continued, or completed." We are concerned here, however, with the second paragraph of § 3237(a), which specifically deals with any offense involving the use of the mails: "Any offense involving the use of the mails ... may be inquired of and prosecuted in any district from, through, or into which such ... mail matter ... moves."

The dissent focuses in part on a "plain reading of the text" to support its conclusion that venue for mail fraud may be based on the situs of the scheme to defraud. (Dissenting Op. 727) But the "text" that the dissent focuses on is § 1341, which creates the substantive offense of mail fraud, not § 3237(a), which deals specifically with the proper venue for "any offense involving the use of the mails." See United States v. Holt, 899 F.2d 15, 1990 WL 37613, at *1-2 (6th Cir. April 3, 1990) (unpublished opinion) (holding that 18 U.S.C. § 3237 governs venue in mail fraud cases); United States v. Reitmeyer, 356 F.3d 1313 (10th Cir.2004) (discussing venue in mail fraud cases and noting that "Congress specifically stated `any offense involving the use of the mails ... is a continuing offense' for venue purposes."); United States v. Loe, 248 F.3d 449, 465 (5th Cir.2001) ("As a `continuing offense,' mail fraud may be prosecuted in `any district in which such offense was begun, continued, or completed.'").

The dissent questions whether the offense of mail fraud is an "offense involving use of the mails...." 18 U.S.C. § 3237(a). (Dissenting Op. 721) To the contrary, we believe that the question virtually answers itself. If the plain meaning of language is to be given any efficacy at all, how can the offense of mail fraud not be an "offense involving the use of the mails...."? We find this court's application of § 3237(a) to the offense of mail fraud in Holt, as well as the Fifth and Tenth Circuits' discussion of the issue, more persuasive and in line with common sense that the contrary Second Circuit decision in United States v. Brennan, 183 F.3d 139 (2d Cir.1999), that is relied upon so heavily by the dissent. Consistent with this court's previous decision in Holt, as well as Reitmeyer and Loe, we believe that § 3237(a) is the controlling statute on venue for mail fraud cases.

A plain reading of the text thus leads us to the conclusion that venue in a mail fraud case is limited to districts where the mail is deposited, received, or moves through, even if the fraud's core was elsewhere. Whether or not Congress could constitutionally provide otherwise is not before us, because the present language of § 3237(a) is determinative.

The dissent also focuses in part on fairness to the defendant. (Dissenting Op. 726-27) We find this argument puzzling in light of the fact that the government is the party pressing for venue where the core of the fraudulent scheme took place, not Wood. If public policy supports accommodating the accused by promoting "fairness," then it should support Wood's desire to limit venue to where the mailings took place. The dissent's expansive interpretation that would impose a different venue on Wood thus appears contrary to public policy as expressed in § 3237(a).

Finally, we note the dissent's concern that "[b]y applying § 3237(a) to mail fraud, the majority could permit the government to hale a defendant into court in distant jurisdictions having virtually no relation to the underlying crime." (Dissenting Op. 723) We would first observe that this is not what happened in the case before us, nor is the dissent's concern consistent with Department of Justice Policy. United States Department of Justice, United States Attorneys' Manual 9-43.300 (1997) ("Department of Justice policy opposes mail fraud venue based solely on the mail matter passing through a jurisdiction."). Second, the dissent's "parade of horribles" is really a challenge to § 3237(a) itself, because the statute allows all "continuing offense[s]" to be prosecuted "in any district from, through, or into which such commerce, mail matter, or imported object or person moves." (Emphasis added.) We would note in this regard that the dissent's concern was evidently not shared by the Supreme Court in Rodriguez-Moreno, because the Court expressly held that the defendant could be prosecuted under § 3237(a) for the illegal possession of a firearm in New Jersey, through which he traveled on his kidnaping odyssey, even though the gun was actually possessed only in Maryland. United States v. Rodriguez-Moreno, 526 U.S. 275, 119 S. Ct. 1239, 143 L. Ed. 2d 388, (1999).

We now turn our attention to the mail fraud charged in Counts 17 and 18. Unlike Counts 19-21 discussed above, where the government based venue solely on the existence of a fraudulent scheme with a significant nexus to the Western District of Michigan, the government argues that it proved by a preponderance of the evidence that the checks involved in Counts 17 and 18 were actually mailed from Battle Creek, which is located within the district. Counts 17 and 18 refer to checks signed by Jacqueline Johnson, an attorney based in Columbus, Ohio, who occasionally worked for Wood. The checks are dated May 31, 1995, and August 28, 1995. Johnson testified that she traveled to Michigan "quite a bit" during "March, April, May, through probably late 1995...." She noted that the bank account from which the two checks were drawn was a Battle Creek account and that any checks written when she was frequently traveling to Michigan "were probably all written while we were visiting the offices in Battle Creek." Based upon this testimony, the government argues that it proved that venue is proper as to Counts 17 and 18 by a preponderance of the evidence.

Wood counters that the undisputed evidence at trial showed that the checks were sent to Ruth Miller in Birmingham, Michigan, which is located in the Eastern District of Michigan. He also points out that the evidence at trial showed that the checkbook for the Battle Creek account on which the checks were drawn was kept in Ohio and that the dividend checks were generally sent from Columbus. Moreover, Wood was in federal custody in Chicago during the relevant time period and was therefore not personally sending or receiving mail in the Western District of Michigan.

Johnson's testimony that any checks written while she was traveling in Michigan were "probably all written while we were visiting the offices in Battle Creek" does not answer the question at hand. In the first place, she had no specific recollection that the particular two checks in question were in fact written while she was in Michigan. Nor was she asked if, after writing any checks in Battle Creek, she then mailed them from Battle Creek. Under these ambiguous circumstances, we conclude that the government failed to prove by a preponderance of the evidence that Johnson actually mailed the two checks at issue from the Western District of Michigan. This precludes a finding that venue was proper with respect to Counts 17 and 18. We therefore reverse Wood's conviction on these counts, as well as on Counts 19-21, and remand for resentencing.

Wood also claims that the district court erred in imposing restitution as a penalty for the Graham and Miller loan transactions. The district court ordered Wood to pay $125,000 in restitution to Graham and approximately $445,000 in restitution to the Millers. We review de novo whether a restitution order is permitted under the law. United States v. Dunigan, 163 F.3d 979, 981 (6th Cir.1999). If it is, then the amount of restitution ordered is reviewed under the "abuse of discretion" standard. Id. Because Wood did not object to the restitution awards at sentencing, however, we will not set aside the district court's determination unless it constitutes plain error. United States v. Bondurant, 39 F.3d 665, 668 (6th Cir.1994).

The district court had authority under 18 U.S.C. § 3663 to order Wood to pay restitution. This statute provides in pertinent part as follows:

(B)(i) The court, in determining whether to order restitution under this section, shall consider — (I) the amount of the loss sustained by each victim as a result of the offense; and (II) the financial resources of the defendant, the financial needs and earnings ability of the defendant and the defendant's dependents, and such other factors as the court deems appropriate.

Wood argues that the district court failed to consider his financial situation as the statute mandates. The government counters that because Wood refused to complete his personal financial statement, he did not meet his burden under 18 U.S.C. § 3664(e), which provides in pertinent part as follows:

Any dispute as to the proper amount or type of restitution shall be resolved by the court by the preponderance of the evidence.... The burden of demonstrating the financial resources of the defendant and the financial needs of the defendant's dependents, shall be on the defendant.

Although he failed to complete his personal financial statement, Wood's presentence report reflects that "[f]rom April 1996 to the present [the date of the presentence report], ... he received $20,000 a year." Wood's failure to cooperate and to complete his personal financial statement undercuts his argument that the district court should have taken his financial situation into account at sentencing.

The district court appears, however, to have had some awareness of Wood's financial situation, even without Wood's personal financial statement. After ordering Wood to pay approximately $570,000 in restitution and a $5,000 fine, the court waived the fine because it found that Wood "[did] not have the ability to pay a fine, because he can't even pay restitution probably. But he doesn't have the financial wherewithal to pay both certainly. So I'll reverse myself and waive the fine in this case." Thus the district court apparently had some understanding of Wood's financial situation.

This court has taken a clear position against restitution orders in amounts "that a defendant cannot possibly pay," reasoning that such restitution orders threaten "respect for judicial orders generally and provide[ ] the defendant with less incentive to seek remunerative, rehabilitative, and non-criminal employment." United States v. Dunigan, 163 F.3d 979, 982 (6th Cir.1999) (quotation marks omitted). In the present case, however, the record provides no basis to conclude that the district court definitively knew that Wood could not pay the award. Wood bore the responsibility of providing the court with details of his financial situation. See United States v. Hall, 71 F.3d 569, 573-74 (6th Cir.1995) ("The Sixth Circuit has held repeatedly that the district court is not required to make findings on the defendant's financial condition."). We conclude that the district court's restitution order does not rise to the level of plain error because Wood's lack of cooperation belies his argument that the order "affected the fairness or integrity of the judicial proceeding." United States v. Koeberlein, 161 F.3d 946, 952 (6th Cir.1998).

Wood makes two additional arguments as to why the restitution awards were erroneous. He contends that the district court miscalculated the loss sustained by the Millers because it failed to consider the Millers' receipt and use of loan proceeds that were never repaid. Wood also argues that the court failed to consider that the Millers and Wood had settled their claims against each other in a prior civil action. These arguments fail, however, because evidence in the record shows that the court did take these considerations into account when ordering the restitution awards. And even if the court had not taken these factors into account, we cannot say that this would rise to the level of plain error in light of Wood's failure to object to the awards at sentencing. Accordingly, we affirm the district court's restitution awards.

Wood's final argument is that, with respect to the Miller transaction, the government failed to present sufficient evidence for a rational trier of fact to convict him on Counts 7-11 and 16-22 of the indictment. He contends that (1) the Millers breached their loan agreement, thus giving First Financial the right under the Agreement to sell the Millers' stock; (2) there was no sale as alleged in the indictment, only a "short sale against the box"; and (3) the government failed to show that Wood was being deceitful when he claimed in a telephone conversation in August of 1996 to have 2,950 shares of Ruth Miller's Comerica stock. When reviewing the sufficiency of the evidence to support a criminal conviction, the "relevant question is whether, after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt." Jackson v. Virginia, 443 U.S. 307, 319, 99 S. Ct. 2781, 61 L. Ed. 2d 560 (1979) (emphasis in original); see also United States v. Kincaide, 145 F.3d 771, 781 (6th Cir.1998) (employing the Jackson standard of review).

Wood first contends that, because the evidence at trial demonstrated that the Millers defaulted on their loan, First Financial had the right to sell the Millers' Comerica stock under the terms of the Agreement. The government counters that the parties treated the loan as an outstanding debt through 1996, yet First Financial had in fact sold Miller's collateral long before it ever claimed that the Millers were in default.

First Financial closed on the Millers' loan for $200,000 on September 19, 1994. The government presented evidence at trial that First Financial sold Ruth Miller's 8,167 shares of Comerica stock short against the box two days later. According to the government, Wood directed First Financial to cover its short position in Comerica stock by delivering Ruth Miller's stock to a broker, on December 8, 1994. This means that less than three months after the Millers entered into the loan agreement, First Financial had disposed of their collateral.

The earliest default notice First Financial sent to Gordon Miller was in the spring of 1995 and the latest was in early 1996. Gordon Miller made 12 payments to First Financial between October of 1994 and June of 1996, 10 of which were made after Ruth Miller's stock had been sold in December of 1994. Viewed in the light most favorable to the government, this proof was sufficient for a jury to find that the Millers were not in default at the time that First Financial sold their collateral.

Wood next argues that the district court erred in denying his Rule 29 motion to dismiss Count 8 of the indictment. Count 8 alleged wire fraud due to the sale of the Ruth Miller's 8,167 shares of Comerica stock on September 21, 1994. Wood claims that there was no evidence of a "sale" of the Comerica stock. According to Wood, the sale alleged in the indictment never occurred because Ruth Miller's Comerica stock was used as collateral for a short sale of someone else's Comerica stock. "Therefore, the Comerica stock posted by the Millers as collateral for the loan from First Financial was held by First Financial in a brokerage account in a short against the box transaction." Wood's argument may be semantically correct, but it is ultimately unpersuasive.

In response to Wood's Rule 29 motion, the government argued at trial that "the term `sale' is sufficiently broad to cover a sale short against the box at that time where Ms. Miller's stock was the collateral standing behind the loan and became, in a sense, encumbered to fulfill the obligation to restore the stock where it had been borrowed." (Emphasis added.) The government presented evidence at trial that Ruth Miller's Comerica stock was in fact disposed of on December 8, 1994, to cover the short position created by the September 21, 1994, transaction. This evidence was sufficient to support Count 8 of the indictment. We therefore find no error in the district court's denial of Wood's Rule 29 motion on Count 8.

Wood's final argument concerns the sufficiency of the evidence supporting the wire fraud charged in Count 22 of the indictment. Count 22 alleged that Wood committed wire fraud during a telephone call when he represented that 2,950 shares of Comerica stock in the account of Keystone Financial, a successor to First Financial, was a portion of Ruth Miller's stock, when in fact Wood knew that the stock was not hers. Wood argues that the government presented no evidence at trial that the stock did not belong to Ruth Miller.

Ruth Miller testified at trial, however, that she became suspicious about her stock's whereabouts when her dividend checks "were coming very late and they were not directly from Comerica, which had been the practice up until then." She explained that no one had told her where her stock was being kept and that she attempted to track it down on her own. Miller finally engaged the help of an attorney, Michael Herzoff, to assist her. Herzoff testified at trial that after attempting to find Miller's stock, his "beliefs were ... confirmed ... that the stock was not in existence any more." In addition to Ruth Miller's and Michael Herzoff's testimony, Gordon Miller testified that his mother's stock was never returned to her.

This evidence, when viewed in the light most favorable to the government, supports the allegation that Wood's telephone conversation, which is the subject of Count 22, furthered a scheme to defraud because "[i]t lulled [Miller] into believing her collateral was safe ... when in fact it was completely gone." Accordingly, we find no error in the district court's denial of Wood's Rule 29 motion with regard to Count 22.

For all of the reasons set forth above, we AFFIRM Wood's conviction and sentence on all counts other than the mail fraud charges in Counts 17-21, REVERSE Wood's conviction on these latter counts, and, accordingly, REMAND for resentencing.

GWIN, District Judge, concurring, in part and dissenting, in part.

While agreeing with the well-stated opinion of my colleagues on other issues, I respectfully dissent regarding their determination that venue did not lie in the Western District of Michigan under the mail fraud statute, 18 U.S.C. § 1341. In that district, the defendant, through collaborators, engaged in a large portion of the fraudulent activities involved with this case. Further, while I concur with the majority opinion's denial of a motion for acquittal, I offer a somewhat different explanation why Defendant Wood's transactions were sales or encumbrances and not "short sales against the box." I begin by offering supplemental reasoning regarding Wood's claim that he engaged in short sales against the box and that such transactions were neither a sale nor an encumbrance of any equities. I then turn to the majority's finding that, in mail fraud prosecutions, venue lies in the district where the participants have mailed or delivered or through which materials passed but not within the district that was the focus of the fraudulent scheme.

On appeal, Defendant Wood argues that, as a matter of law, a short sale against the box is neither a sale nor an encumbrance. First, the defendant argues that the jury instructions that describe a short sale against a box as a "sale" or an "encumbrance" were invalid. Second, the Defendant argues that the evidence was insufficient to convict him because his activities could not be a sale or an encumbrance.

The majority finds the trial court's instructions were balanced, neither suggesting nor rejecting that a short sale against the box could be an encumbrance or a sale. I agree with this finding, although I believe we should reject Wood's argument for a more fundamental reason — he offered insufficient evidence to support a finding that the Graham and Miller transactions were short sales against the box transactions.

Long sales of securities occur when the owner of certain stocks sells them, delivers them, and consequently pays taxes on the profits. In short sales, an investor borrows stock, typically from his broker, and sells that stock in the hopes that he can later return the stock the investor had borrowed from the broker with stock purchased at a lower price. If the price of the security declines, the investor profits by the difference between the price at which he sold the stock and the price at which he later purchases the stock (minus any sales commissions and margin fees that his broker assesses). Short sales can also help customers avoid taxes. Short sale transactions close when the short seller returns an equal number of shares to the lender.

Short sales are risky. The short seller loses when the price of the security rises and the cost of covering the borrowed securities increases. Because no metaphysical cap exists on how high a security price can rise, losses continue to increase as the stock borrowed for the short sale gains value. The loss is theoretically unlimited. As a result, "[a] material misrepresentation concerning the risks of a short sale transaction... can form the basis of a private suit under SEC Rule 10b-5."1

Short sales "against the box" differ from short sales by attempting to reduce the risk associated with borrowing stocks and agreeing to replace the borrowed stocks. In these transactions, an investor already owns equities, but borrows (generally through a broker) an equal number of shares of the same class of stock to sell short in the open market. The investor posts the "long" (owned) shares as collateral. The broker then generally holds the shares in a safety deposit box — thus the name, "against the box." The broker uses the shares from the box to complete the short sale only if the investor is unable or unwilling to return the borrowed stock.

Theoretically, those who invest "against the box" perfectly hedge against market risk. If the market price of the stock increases, this increases the value of their long (owned) shares. Yet, if the market price decreases, this increases the value of the short (borrowed) shares. While an investor will not profit from a short against the box transaction — since losses in the short or long transaction offset any gains in the other — such transactions can allow investors to liquidate stock without incurring taxes on the original sale.

Defendant Wood claims he did not defraud First Financial customers because securities taken from customers' accounts were covered by an "in the box" transaction. With his argument, the Defendant turns the idea of a short sales against the box on its head. Most importantly, Wood's argument fails because he never had any long shares "in the box" to cover the shares he borrowed in the short sale. Without permission, Defendant Wood sold securities that customers had pledged to secure loans the customers had received. At the time he sold the pledged securities, no evidence showed that he placed equivalent securities "in the box," as required for a short sale against the box.

Although he never covered the securities used, Wood tries to overcome this problem by arguing that he could have purchased shares to place "in the box" if his customers' loans were repaid: "Once borrowers from First Financial had repaid their loans, First Financial had a pool of capital with which to purchase securities in the open market to close the short position, which would in turn free the borrower's pledged securities from `the box' and allow their return to the borrower." (Def. br. at 12).

This argument fails. Most important, the presence of collateral to satisfy the short seller's obligation to return the borrowed shares distinguishes an "in the box" transaction. Wood showed no evidence that he placed equivalent otherwise-unencumbered securities "into the box." For example, the shorted security, Sonic Environmental Systems, was very thinly traded and not readily obtainable. In essence, Wood makes the argument that a transaction is a short sale against the box if funds are available to cover the borrowed security. If we accepted this argument, any trustee who absconded with securities could characterize any short sale as no sale. Even if Wood's unsupportable argument were allowed, he loses for another simple reason. Wood showed no evidence that he dedicated funds to obtain replacement shares for the Sonic Environmental Systems shares he had shorted. The evidence showed that Wood used the Graham proceeds to cover the loan to the Millers. In addition, Wood offered no evidence that he maintained unencumbered assets of any kind that would allow the purchase of cover for the shorted stock.

Moreover, I am not convinced that it would make a difference even if Defendant Wood could show the transactions were short sales against the box. Even if his assertions are true, the Government could still convict him for fraudulent behavior.

The defendant argues that a short sale "against the box" does not encumber the stock posted or make up a sale as a matter of law. The defendant cites no authority for his assertion that a short sale against the box can never be a "sale" violating the Security and Pledge Agreement. Cf. Rubin v. United States, 449 U.S. 424, 429, 431, 101 S. Ct. 698, 66 L. Ed. 2d 633 (1981) (expansively interpreting the words "sale" to hold that a pledge of stock as collateral for a loan is a "sale" under the Securities Act, even without a default on the loans).

The defendant goes on to assert that a short sale against the box never amounts to an encumbrance, citing Schreiber v. First Fin. Acceptance, 965 F. Supp. 397 (E.D.N.Y.1997). The Schreiber case also involved Defendant Wood. The Schreiber court found that "the question of encumbrance amounts to whether the defendants could have returned the securities to [plaintiff], at any given time, had [plaintiff] repaid the loan in full." Id. at 400. The Schreiber court opined that short sales against the box do not encumber the stock if the defendant maintained "sufficient equity in the account" to satisfy margin requirements. Id. This holding confuses short sales with short sales against the box. While the former transactions depend upon margin requirements, the later transactions revolve around posted shares, not the availability of other assets that could be used to satisfy the obligation to cover the borrowed shares. Furthermore, the Eastern District of New York engaged in a unique definition of equity when it considered outstanding payments on Defendant Wood's loan as collateral for the obligation to post securities. Id. This definition, which Defendant Wood urges the Court to adopt in the present case, is not persuasive.

Typically, to engage in a short sale, one must maintain a margin account related to the market value of the borrowed stock. Until Schreiber, speculative future earnings have never been considered part of posted equity for purposes of fulfilling margin requirements. Essentially, the Schreiber court redefines securities law with its anomalous view of an "encumbrance." As a result, commentators have criticized Schreiber as "flagrantly wrong." See, e.g., Kenneth C. Kettering, Repledge and Pre-Default Sale of Securities Collateral under Revised Article 9, 74 CHICAGO-KENT L.REV. 1109, n. 13 (1999) ("Schreiber seems flagrantly wrong and is explicable only on the supposition that the court had little patience with the suit because the secured party was in fact solvent.").

Moreover, Schreiber breaks from the momentum of Supreme Court precedent. Recall that, in Rubin v. United States, 449 U.S. 424, 429, 431, 101 S. Ct. 698, 66 L. Ed. 2d 633 (1981), the Supreme Court held that a pledge of stock as collateral for a loan is a "sale" under the Securities Act, even without a default on the loans. Extending this holding to conclude that a short sale against the box is an encumbrance of the stock would be logical.

Even if this court were to apply the questionable reasoning of Schreiber, the facts of the Graham loan would still amount to an encumbrance. The Schreiber court says that a defendant must be prepared to return the securities upon payment of the loan in full. Yet, Graham did repay his loan and Defendant Wood did not produce his 35,000 shares of Sonic stock that Graham had posted to secure the loan.

An analysis of the Miller case, under the common definition of margin requirements, also reveals an encumbrance. Wood sold the Millers' Comerica stock on December 8, 1994, less than three months after the Millers pledged the stock to secure the loan to them. Defendant Wood did not maintain long shares of Comerica stock "in the box" nor did he post margin requirements. By the end of December 1994, all that remained in that brokerage account was $1,117. Because the price of stock can increase dramatically, the only way to insure an economically feasible purchase of stock to close a short sale is by posting margin requirements before-the-fact. This court would need to engage in pure fantasy to pretend that Wood did not encumber the Miller stock that Defendant Wood could magically come up with identical shares when the Millers repaid the loan. That the Comerica stock tripled in value by May 1996 and Wood failed to produce replacement Comerica stock drives this point home.

Ultimately, Defendant Wood argues that by considering a short sale against the box as an unlawful sale or encumbrance, the lower court allowed the jury to find that these types of transactions are illegal. Nevertheless, the jury instructions never say that a short sale against the box is per se an illegal transaction. The jury instructions explain that the jury cannot convict the Defendant apart from fraudulent behavior. Nothing in the jury instructions says that engaging in a short sale against the box, by itself, is fraudulent behavior. Rather, the Grand Jury charged Defendant Wood with knowingly deceiving investors about the status of their stock. The jury convicted the Defendant of fraud, not for engaging in short sales against the box, assuming we could so characterize his transactions. Summarizing, Wood fails to show that he ever posted covering shares for the stock that he sold short. Additionally, he fails to offer evidence that he had the resources to provide covering shares for the stock that he sold short. Having failed to give any evidence that his short sale of the pledged securities was covered, he cannot argue that there was no sale or encumbrance of the securities that Miller and Graham had pledged to secure their loans.

In a criminal case, the question of venue has not only pragmatic, but also constitutional implications. Article III requires that "[t]he Trial of all Crimes ... shall be held in the State where the said Crimes shall have been committed." U.S. CONST. art. III § 2 cl. 3. The Sixth Amendment takes this requirement a step further by requiring that the trial take place in the same district as that in which the crime was allegedly committed. U.S. CONST. amend. VI. ("In all criminal prosecutions, the accused shall enjoy the right to ... trial, by an impartial jury of the State and district wherein the crime shall have been committed[.]").2 Various federal statutes further define these provisions. See, e.g., Fed.R.Crim.P. 18 ("Except as otherwise permitted by statute or by these rules, the prosecution shall be had in a district in which the offense was committed."); 18 U.S.C. § 3237, infra. Read as a whole, these provisions manifest a strong constitutional policy disfavoring trials removed from the situs of the alleged criminal activity.

The general venue provisions for continuing offenses are found at 18 U.S.C. § 3237. Section 3237(a) reads, in relevant part, "Any offense involving use of the mails ... is a continuing offense and ... may be ... prosecuted in any district from, through, or into which such ... mail matter ... moves." The majority states that mail fraud is such a continuing offense. After referencing that provision, the majority finds that venue exists where mail related to the fraud "is deposited, received, or moves through, even if the fraud's core was elsewhere." Majority opinion at 713. I cannot agree.

Section 3237 has a limited application. The Supreme Court has long noted that § 3237 is inapplicable to statutes which contain their own specific venue provisions. See Travis v. United States, 364 U.S. 631, 636-37, 81 S. Ct. 358, 5 L. Ed. 2d 340 (1961) ("[V]enue should not be made to depend upon the chance use of the mails, when Congress has so carefully indicated the locus of the crimes."); see also United States v. Brennan, 183 F.3d 139, 147 (2d Cir.1999). Furthermore, the second paragraph of § 3237(a) broadly refers to offenses involving the "use of the mails." Thus, any statutes which do not so broadly criminalize acts involving the "use of the mails" are not likely covered under § 3237. See infra.

In United States v. Brennan, 183 F.3d 139, the Second Circuit spoke to the applicability of § 3237 to the mail fraud statute. In Brennan, the government argued that venue was appropriate under § 3237(a) because mail had moved through the Eastern District of New York. The Second Circuit rejected this argument and found that § 3237(a) does not establish venue in mail fraud cases:

Defendants maintain ... that § 3237(a) does not apply to mail fraud prosecutions. Their argument rests on the contention that the mail fraud statute does not proscribe conduct involving "the use of the mails" within the meaning of § 3237(a). We agree. Though perhaps surprising, this conclusion is strongly supported by consideration of the history and purpose of § 3237(a) and the constitutional protection of defendants' venue rights.

Id. at 146.

Examining the legislative history of § 3237(a), the Second Circuit found that this venue provision was enacted in response to the Supreme Court's opinion in United States v. Johnson, 323 U.S. 273, 65 S. Ct. 249, 89 L. Ed. 236 (1944), where the Court articulated a rule favoring restrictive construction of venue provisions. In Johnson, the defendant had been charged with an offense involving "use [of] the mails or any instrumentality of interstate commerce." Finding there was insufficient venue under the continuing violation provision of the first paragraph of § 3237(a), the Court overturned the conviction. In response to the Court's ruling in Johnson, Congress expanded § 3237(a) by adding a second paragraph expanding continuing violations to include "the use of the mails." Brennan, 183 F.3d at 146.

Against this backdrop, the Second Circuit in Brennan held that mail fraud is not a continuing violation and, consequently, that § 3237 is inapplicable to the mail fraud statute:

[W]e agree with defendants that § 3237(a) is best read as not applying to statutes, like the mail fraud statute, that specify that a crime is committed by the particular acts of depositing or receiving mail, or causing it to be delivered, rather than by the more general and ongoing act of "us[ing] the mails." Rather than make a defendant like Brennan subject to prosecution in any district through which a mail truck carrying his mail happened to drive (or perhaps even in any district over which an airplane carrying the mail happened to fly, or in which it happened to make an interim stop), we think Congress's more particularized and careful phrasing in the mail fraud statute takes it outside the scope of § 3237(a) and is best read less expansively.

Id. at 147 (brackets in original). This well-reasoned decision also is consistent with dicta from the Third Circuit. See United States v. Turley, 891 F.2d 57, 60 (3d Cir.1989) (acknowledging and approving government concession that "18 U.S.C. § 3237 ... is not applicable to mail fraud.").

A close comparison of the text of § 3237(a) with that of the mail fraud statute, 18 U.S.C. § 1341, compels this conclusion. Recall that § 3237(a) applies to crimes involving "use of the mails." As the Brennan court noted, "[t]here are today many ... statutes that expressly prohibit `use of the mails' in connection with various activities for various purposes." 183 F.3d at 147.3 Mail fraud, however, "specif[ies] that a crime is committed by the particular acts of depositing or receiving mail, or causing it to be delivered, rather than by the more general and ongoing act of `us[ing] the mails.'" Id. (second brackets in original). Had Congress intended to include mail fraud as a "continuing offense" under § 3237(a), it would have defined mail fraud as involving "use of the mails." Its refusal to do so, especially considered in light of the numerous statutes invoking "use of the mails," is a compelling reason not to apply § 3237(a) to mail fraud.

Furthermore, § 3237 is inapplicable to statutes which contain their own venue provisions. See Travis, 364 U.S. at 636-37, 81 S. Ct. 358 (1961). Section 1341 specifically defines the locus of mail fraud prosecutions as the situs of "scheme or artifice to defraud," or the location where the defendant "places", "causes to be deposited", or "takes or receives" an item for delivery through the Postal Service. 18 U.S.C. § 1341. These provisions expressly contemplate the acts which trigger venue and preempt the more general venue provisions contained in § 3237(a). See id. at 147.

Based on these considerations, the United States Department of Justice policy opines that § 3237 is inapplicable to the mail fraud statute: "The locus for mail fraud prosecutions is specifically set forth in section 1341; since Congress has `otherwise expressly provided [the situs for venue],' section 3237 is inapplicable to mail fraud." United States Department of Justice, Criminal Resource Manual 966 (1997).

Moreover, the majority's holding that a prosecutor can establish venue in any district that the mail moves through is extremely expansive, and invites conflict with the Constitution's guarantee that trials be held at the location of the crime. The Sixth Amendment's requirement that defendants be tried in the district where their alleged crimes took place reveals a policy against enhancing the already-onerous burdens of trial by establishing venue in a far-off locale. The majority's reasoning impairs this important constitutional policy.4 By applying § 3237(a) to mail fraud, the majority could permit the government to hale a defendant into court in distant jurisdictions having virtually no relation to the underlying crime. To illustrate, consider a defendant who initiated a Ponzi scheme by mailing letters from Cleveland soliciting "investments" to residents of California. Suppose further that the U.S. Postal Service routs mail from Cleveland to California through a postal center in Kansas City. The defendant could potentially be forced to stand trial in Kansas City despite the fact that his mail fraud concerned nary a Missourian.5 This example shows how Government could use this expansive interpretation of venue to increase the burden of defending a case in federal court.

The majority argues that using § 3237 does not expand Wood's vulnerability to suit in far-away jurisdictions. Although true in this case, the majority's holding greatly expands the number of locations other defendants may be summoned to. Indeed, future defendants will likely be summoned to trial in a distant jurisdiction merely because the postal system happened to route a piece of mail though that location. As I see it, this outcome plainly violates the constitutional policy, expressed in Article III and the Sixth Amendment, against exacerbating the burdens of trial by establishing venue in far-away locations.6

B. Appropriateness of Venue Where the Scheme Was Hatched

At the outset of this discussion, I wish to highlight that this case presents an unusual factual setting for discussing venue under the mail fraud statute. Typically, the district where a defendant set up and carried the scheme to defraud also is the place from which mail was sent or received. Thus, most federal courts have never addressed whether venue under § 1341 is also appropriate where the defendant carried out the scheme to defraud, in the rare circumstances that a mailing is not easily traceable to that area. Only a handful of district courts have dealt directly with the matter and they have not come to any consensus. Compare United States v. Mikell, 163 F. Supp. 2d 720, 728 (E.D.Mich.2001) (holding that venue was proper in any district through which the mail moved), with United States v. Olen, 183 F. Supp. 212, 218 (S.D.N.Y.1960) (holding that venue was proper in the district where the defendants initially devised their fraudulent scheme). As district courts take opposing sides on the issue, the only thing that is clear is that the law is unclear.

Thus, no binding precedent exists from the Sixth Circuit Court of Appeals (or any other circuit) addressing the appropriateness of venue in the present matter.7 The Eastern District of Michigan, in Mikell, notes that the Sixth Circuit never before has decided the matter: "This issue is a novel one, especially in light of the Supreme Court decision in Rodriguez-Moreno." 163 F. Supp. 2d 720, 728 (reversed on other grounds). Further, in Brennan the Second Circuit — the only other Circuit to decide (albeit in passing) whether venue under § 1341 can be based on the scheme to defraud described the lack of clarity in this area of the law: "We cannot tell whether the reasoning of [the Supreme Court in] Rodriguez-Moreno would make venue appropriate for a prosecution under § 1341 not only in districts where mail matter was sent or received in furtherance of a fraud [sic] scheme, but also in any district where any aspect of the `scheme or artifice to defraud,' 18 U.S.C. § 1341, was practiced." 183 F.3d 139, 145 (2d Cir.1999) (emphasis added).

Given that this ruling sets precedent regarding the scope of venue under § 1341, we should look to the text of the statute, the legislative history, the Supreme Court's mandates for how to interpret the statutory text in such situations, and public policy concerns. I fear that the majority adopts a frame of analysis that ignores great inconsistencies and practical difficulties created by its approach.

In United States v. Rodriguez-Moreno, 526 U.S. 275, 119 S. Ct. 1239, 143 L. Ed. 2d 388 (1999), the Supreme Court addressed how to decide where venue lies. To decide whether venue is appropriate "a court must initially identify the conduct making up the offense (the nature of the crime) and then discern the location of the commission of the criminal acts." Id. at 279, 119 S. Ct. 1239. The Supreme Court has said that the "locus delicti [of the charged offense] must be determined from the nature of the crime alleged and the location of the act or acts constituting it." United States v. Cabrales, 524 U.S. 1, 6-7, 118 S. Ct. 1772, 141 L. Ed. 2d 1 (1998) (quoting United States v. Anderson, 328 U.S. 699, 703, 66 S. Ct. 1213, 90 L. Ed. 1529 (1946)). The Rodriguez-Moreno Court recognized that essential conduct elements may be "embedded in a prepositional phrase and not expressed in verbs," and asserted that verbs are not "the sole consideration in identifying conduct that constitutes an offense." Id. at 280, 119 S. Ct. 1239. A study of the mail fraud statute will clarify the essential conduct elements of the statute.

The mail fraud statute describes the central acts forbidden: "Whoever, having devised or intending to devise any scheme or artifice to defraud ... places, ... deposits, or causes to be deposited ... or knowingly causes to be delivered [mail matter] ... shall be fined under this title or imprisoned not more than 20 years...." 18 U.S.C. § 1341. The mail fraud statute therefore involves (1) a scheme to defraud and (2) a deposit or a delivery of mail. The majority asserts that venue is proper where mail is placed, deposited, travels through or is delivered through the postal system. The majority does not pay heed to the critical phrase of § 1341 — "[w]hoever, having devised or intending to devise any scheme or artifice to defraud." The ultimate dispute is one of textual interpretation.

First of all, the text of the statute establishes that venue is appropriate in the district encompassing the situs of the scheme to defraud. Applying the standard established in Rodriguez-Moreno, the scheme to defraud is the more essential conduct element in mail fraud. As the statute reads, the mailings themselves are illegal only to the extent that they further an ongoing fraudulent scheme. See, e.g., United States v. Griffith, 17 F.3d 865, 874 (6th Cir.1994). In this sense, the mailing is merely a jurisdictional "hook," secondary in importance to the substantive scheme.

Additionally, the mail fraud statute's legislative history supports the conclusion that the fraud, and not the mailing, is the central conduct of the mail fraud offense. The legislative history that exists establishes that, when enacting the mail fraud statute, Congress primarily concerned itself with the alleviation of fraud, not the protection of the mails. In 1948, Congress passed the current mail fraud statute, codified in part at 18 U.S.C. § 1341. The modern statute stems from a predecessor statute, originally enacted in 1872 as part of the recodification of the postal laws. The legislative history for the mail fraud statute is sparse. However, in 1870 the sponsor of the original statute noted that the mail fraud component was intended "to prevent the frauds which are mostly gotten up in the large cities... by thieves, forgers, and rapscallions generally, for the purpose of deceiving and fleecing the innocent people in the country." McNally v. United States, 483 U.S. 350, 356, 107 S. Ct. 2875, 97 L. Ed. 2d 292 (1987) (quoting Remarks of Rep. John Franklin Farnsworth, Cong. Globe, 41st Cong., 3d Sess. 35 (1870) (emphasis added)).8 Commentators who have studied the legislative history of the mail fraud statute also have described the purpose of the statute as seeking to eliminate fraud: "There existed a perceived need for federal intervention to dispel widespread fraud." Jed S. Rakoff, The Federal Mail Fraud Statute (Part I), 18 DUQ. L.REV. 771 (1980) (emphasis added).

Third, contemporary case law supports the view that the scheme to defraud comprises the integral part of mail fraud. For example, the Supreme Court has minimized the importance of mailings in establishing a mail fraud offense:

"To be part of the execution of the fraud, however, the use of the mails need not be an essential element of the scheme. It is sufficient for the mailing to be `incident to an essential part of the scheme,' or `a step in [the] plot.'" Schmuck v. United States, 489 U.S. 705, 710-11, 109 S. Ct. 1443, 103 L. Ed. 2d 734 (1989) (internal citations omitted). The Sixth Circuit also has downplayed the importance of mailings in establishing mail fraud, noting that a legal mailing will violate the federal mail fraud statute if it furthers a fraudulent scheme. United States v. Hopkins, 357 F.2d 14, 17 (6th Cir.1966).

Indeed, the Southern District of New York also has explained that the mailing is merely the hook to federal jurisdiction, while the fraud is the essence of the offense:

It is [] true ... that the "gist" of a mail fraud case is the mailing. This statement, however, is true only in the sense that until a use of the mail occurs, no federal jurisdiction exists. The evil to be combated, of course, is the fraud and I have no doubt that Congress could constitutionally provide for trial of a mail fraud case in the district where the scheme was hatched even though all mailings took place in another district.

Olen, 183 F.Supp. at 218. For these reasons, the scheme to defraud, not the protection of the mails, is the central prohibited activity of fraudulent mailing under § 1341.

After considering the conduct making up the offense of mail fraud, it is difficult to find justification for holding Wood's trial in the Southern District of Ohio or the Eastern District of Michigan. At best, the only connections with those places are the incidental mailing of less important documents. The fraud's core centered in the Western District of Michigan and most of the evidence and connections lay there. It is notable that the "independent value in trial at the place of offense" is that "this may be where the witnesses will be located." Charles Alan Wright, Federal Practice and Procedure: Criminal § 301(3d ed.2000). Most of the witnesses and evidence are found near the situs of the fraudulent scheme — the obvious location for a fair trial. Further, the majority's holding implicates judicial economy and fairness to the defendant. It is unnecessarily duplicitous if federal prosecutors pursue — and judges must try — separate cases for each location where defendants sent or received mail. Four decades ago, the Second Circuit noted that modern policy concerns may merit an expansion of venue under the mail fraud statute. Such concerns are even more pressing today:

[W]e cannot ignore the fact that the venue rule for mail fraud cases was developed many decades ago, when more restricted notions of the limits of federal power prevailed than exist today. Were the issue of the appropriate place of trial in such cases to arise [de] novo in 1960, it is by no means clear that trial would not be permitted at the place where the fraudulent scheme was developed as well as at the place where the mailing has its impact.

United States v. Cashin, 281 F.2d 669, 674 (2d Cir.1960); see also Wright, supra, § 303, at 323 ("A similar principle should apply as to use of the mails to defraud.... [I]t can be fairly argued that the place where the scheme to use the mails to defraud was devised is also a proper venue.").

I would therefore find that venue exists where the fraudulent conduct is centered, where items of mail are deposited, where they are received, and where they are caused to be delivered. In this case, the Western District of Michigan is where the scheme was centered, where Wood sent out advertising to national publications soliciting customers, sent out packets of information in response to inquiries from potential customers, took telephone messages and offer excuses why Defendant could not answer a customer's call, prepared loan documents; disbursed loan proceeds, accepted the stock offered as collateral, directed the distribution of the proceeds from the sale of the stock, and sent letters to traders concerned about the fate of their collateral.

In summation, when considering the mail fraud statute in light of legislative history, a plain reading of the text, the Supreme Court's mandates for textual interpretation, and public policy, venue for mail fraud may be based on the situs of the scheme to defraud.

To summarize, I respectfully dissent from the majority's conclusion that venue was improper as to the mail fraud counts. While I agree with all of the other legal conclusions of this Court, I have provided supplemental reasoning regarding why Defendant Wood's transactions are not protected under the rubric of "short sales against the box."