We decide whether Congress can, under its Commerce Clause power, prohibit the mere possession of homemade machineguns.

Robert W. Stewart sold parts kits for the manufacture and assembly of Maadi-Griffin.50 caliber rifles; he advertised the kits on the Internet and in Shotgun News, a national firearms magazine. Stewart believed the kits were legal to sell because the receivers on the rifles had not yet been completely machined and the rifles were thus not usable as firearms. An agent of the Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) discovered that Stewart had a prior conviction for possession and transfer of a machinegun and decided to investigate Stewart's business. Another agent, acting undercover, purchased one of Stewart's kits and determined that it could be "readily ... converted" into an unlawful firearm, in violation of 18 U.S.C. § 922(a)(1)(A) and § 921(a)(3)(A). Based on this information, the ATF agent secured a federal search warrant for Stewart's residence.

In addition to numerous rifle kits, the ATF search also turned up thirty-one firearms, including five machineguns. The machineguns had been machined and assembled by Stewart. Stewart was charged and convicted of one count of felony possession of firearms in violation of 18 U.S.C. § 922(g)(1) and § 924(a)(2), and five counts of unlawful possession of a machinegun in violation of 18 U.S.C. § 922(o). No charges were brought against Stewart regarding the advertised parts kits that were initially the subject of the investigation. Stewart appeals his conviction for unlawful possession of machineguns, claiming that 18 U.S.C. § 922(o) is an invalid exercise of Congress's commerce power and violates the Second Amendment; he appeals his conviction for possession of a firearm by a felon on Second Amendment grounds.1

Section 922(o) makes it unlawful to "transfer or possess a machinegun." Notably absent from this provision is any jurisdictional requirement that the machinegun has traveled in or substantially affected interstate commerce. We decide whether this statute, as applied to Stewart, offends the Commerce Clause.

1. There are three categories of activity that Congress can regulate under its commerce power: (1) "the use of the channels of interstate commerce"; (2) "the instrumentalities of interstate commerce, or persons or things in interstate commerce, even though the threat may come only from intrastate activities"; and (3) "those activities having a substantial relation to interstate commerce." United States v. Lopez, 514 U.S. 549, 558-59, 115 S.Ct. 1624, 131 L.Ed.2d 626 (1995). In United States v. Rambo, 74 F.3d 948 (9th Cir.1996), we held that section 922(o) was "a regulation of the use of the channels of interstate commerce" because "there can be `no unlawful possession under section 922(o) without an unlawful transfer.'" Id. at 952 (quoting United States v. Kirk, 70 F.3d 791, 796 (5th Cir.1995)). We elaborated that, "`[i]n effect, the ban on such possession is an attempt to control the interstate market for machineguns by creating criminal liability for those who would constitute the demand-side of the market, i.e., those who would facilitate illegal transfer out of the desire to acquire mere possession.'" Id. (quoting Kirk, 70 F.3d at 796). Rambo thus held section 922(o) was a valid exercise of the commerce power because a transfer or sale must have preceded the criminalized possession.

Stewart's case reveals the limits of Rambo's logic. Contrary to Rambo's assumption that an unlawful transfer must precede unlawful possession, Stewart did not acquire his machineguns from someone else: He fabricated them himself. The government has never contested Stewart's claim that the machineguns were entirely homemade, and the evidence supports his claim. The chief of the ATF Firearms Technology Branch, referring to one of the machineguns, testified that it was "a unique type of firearm." Tr. of Trial at 562 (emphasis added). He explained that the machineguns were "based on a ... Sten gun design," which is a type of British machinegun, and had "certain [Sten gun] parts," id. at 558, but "the rest of the parts ... [were] not ... conventional Sten gun parts," id. at 562. He also testified that one of the machineguns had "some Sten gun parts on it, but then it also ha[d] parts which [were] not original Sten gun parts." Id. at 550. He continued: "And I've seen many Sten guns assembled from Sten gun parts kits, but I had never previously seen one that was assembled with these other parts on it." Id. at 550-51. None of the machineguns had original Sten receiver tubes (the part of the gun that houses the cartridge when the weapon is fired), and at least one was identified as having a "homemade receiver tube." Id. at 567. On some of the machineguns, the trigger was "quite different" from "an ordinary Sten gun trigger." Id. at 561. The ATF chief testified that "[t]he only time [he'd] ever seen ... this [type of mechanism was] in conjunction with [a] .... single-shot rifle." Id. at 561-62.

The district court ruled against Stewart's Commerce Clause argument, reasoning that "the parts, at least, moved in interstate commerce." Id. at 626. Indeed, some of the machinegun parts did move in interstate commerce. At some level, of course, everything we own is composed of something that once traveled in commerce.2 This cannot mean that everything is subject to federal regulation under the Commerce Clause, else that constitutional limitation would be entirely meaningless. As Lopez reminds us, Congress's power has limits, and we must be mindful of those limits so as not to "`obliterate the distinction between what is national and what is local and create a completely centralized government.'" Lopez, 514 U.S. at 557, 115 S.Ct. 1624 (quoting NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1, 37, 57 S.Ct. 615, 81 L.Ed. 893 (1937)). Our sister circuits have also recognized that section 922(o) must have certain implicit limits, noting that, "because § 922(o) has no jurisdictional element, it has the potential to criminalize the possession of such guns that have never traveled in interstate commerce." United States v. Wright, 117 F.3d 1265, 1270 (11th Cir.1997), vacated in irrelevant part by 133 F.3d 1412 (11th Cir.1998). The difficult question is where to draw the line between a regulated object and the matter from which that object was created.

In United States v. McCoy, 323 F.3d 1114 (9th Cir.2003), we confronted a similar line-drawing problem. McCoy held that a statute criminalizing possession of child pornography was unconstitutional as applied to a woman who posed nude with her child for her husband's camera. The photographs were intended only for home use. The statute contained a jurisdictional element allowing prosecutions even where the pornographic material "was produced using materials which have been mailed or... shipped or transported" in interstate commerce, 18 U.S.C. § 2252(a)(4)(B); this would seem to include any of the film, paper, cameras, computers or other technology needed to produce pornographic images. However, because "`all but the most self-sufficient child pornographers will rely on film, cameras, or chemicals that traveled in interstate commerce,'" McCoy, 323 F.3d at 1125 (quoting United States v. Rodia, 194 F.3d 465, 473 (3d Cir.1999)), McCoy held that the jurisdictional element "provide[d] no support for the government's assertion of federal jurisdiction," id. at 1126. McCoy thus recognized that, just because certain of the elements that make up an object have traveled interstate at one time or another, this does not necessarily mean Congress can regulate that object under the Commerce Clause.

Some components of Stewart's machineguns had crossed state lines, but these components did not add up to a gun. Not even close. Even more than in McCoy, many additional parts and tools, as well as expertise and industry, were needed to create functioning machineguns. This is quite different than if Stewart had ordered a disassembled gun and simply put the parts together, the way one might assemble a chair from IKEA. These machineguns were a "unique type of firearm," with legal parts mixed and matched from various origins; they required more than a simple turn of a screw-driver or a hit of a hammer to become machineguns. We therefore cannot say that the machineguns themselves—in any recognizable form— traveled in interstate commerce.

Because these firearms were genuinely homemade, we find that Stewart did not obtain his machineguns by "us[ing] the channels of interstate commerce." Thus, although Rambo found section 922(o) to be generally valid under the Commerce Clause, Rambo's reasoning does not cover Stewart's case.

2. Even if Stewart did not use the channels of interstate commerce, his possession of machineguns may still have substantially affected interstate commerce. Several courts of appeals have held section 922(o) constitutional on this ground. Wright, 117 F.3d at 1268-71; United States v. Rybar, 103 F.3d 273, 276-85 (3d Cir.1996); United States v. Kenney, 91 F.3d 884, 890-91 (7th Cir.1996). We cannot agree that simple possession of machineguns— particularly possession of homemade machineguns—has a substantial effect on interstate commerce.

In United States v. Morrison, 529 U.S. 598, 120 S.Ct. 1740, 146 L.Ed.2d 658 (2000), the Supreme Court set out the controlling test for determining whether a regulated activity "substantially affects" interstate commerce: We must consider (1) whether the regulated activity is commercial or economic in nature; (2) whether an express jurisdictional element is provided in the statute to limit its reach; (3) whether Congress made express findings about the effects of the proscribed activity on interstate commerce; and (4) whether the link between the prohibited activity and the effect on interstate commerce is attenuated. Id. at 610-12, 120 S.Ct. 1740.

We start by considering the first and fourth prongs of the Morrison test, as we have deemed them the most important. See McCoy, 323 F.3d at 1119. The first prong is not satisfied here. Possession of a machinegun is not, without more, economic in nature. Just like the statute struck down in Lopez, section 922(o) "is a criminal statute that by its terms has nothing to do with `commerce' or any sort of economic enterprise, however broadly one might define those terms." Lopez, 514 U.S at 561, 115 S.Ct. 1624. Unlike in Wickard v. Filburn, 317 U.S. 111, 63 S.Ct. 82, 87 L.Ed. 122 (1942), where growing wheat in one's backyard could be seen as a means of saving money that would otherwise have been spent in the open market, a homemade machinegun may be part of a gun collection or may be crafted as a hobby. Or it may be used for illegal purposes. Whatever its intended use, without some evidence that it will be sold or transferred—and there is none here—its relationship to interstate commerce is highly attenuated.

Moreover, the regulation itself does not have an economic purpose: whereas the statute in Wickard was enacted primarily to control the market price of wheat, id. at 115, 63 S.Ct. 82, there is no evidence that section 922(o) was enacted to regulate commercial aspects of the machinegun business. More likely, section 922(o) was intended to keep machineguns out of the hands of criminals—an admirable goal, but not a commercial one.

We can also say with some confidence that the effect of Stewart's possession of homemade machineguns on interstate commerce was attenuated under the fourth prong of the Morrison test. Lopez already rejected the reasoning that, because the cost of violent crimes is spread through insurance, regulations intended to prevent violent crimes significantly affect the national economy. Lopez, 514 U.S. at 563-64, 115 S.Ct. 1624. Nor did Lopez buy the argument that violent crime substantially affects commerce by reducing people's willingness to travel to unsafe areas of the country. Id. at 564, 115 S.Ct. 1624. Though prohibition of all machinegun possession may have a greater chance of reducing violent crime than a prohibition that extends only to school zones, this does not change what the Court said in Lopez: that under these expansive theories, "it is difficult to perceive any limitation on federal power, even in areas such as criminal law enforcement ... where States historically have been sovereign." Id.; see also Morrison, 529 U.S. at 615-16, 120 S.Ct. 1740 (rejecting the argument, supported by legislative history, that the effect of gender-motivated violence on the national economy was not attenuated); McCoy, 323 F.3d at 1124 ("It is particularly important that in the field of criminal law enforcement, where state power is preeminent, national authority be limited to those areas in which interstate commerce is truly affected."); United States v. Ballinger, 312 F.3d 1264, 1271 (11th Cir.2002) ("To allow Congress to regulate local crime on a theory of its aggregate effect on the national economy would give Congress a free hand to regulate any activity, since, in the modern world, virtually all crimes have at least some attenuated impact on the national economy."). This "cost of crime" rationale thus cannot save the government's case.

Our most recent child pornography case, United States v. Adams, 343 F.3d 1024 (9th Cir.2003), used a different approach to link simple possession of child pornography to interstate commerce. Adams reasoned that prohibiting possession of child pornography "could strike a blow to the [child pornography] industry ... `because those who possess and view child pornography encourage its continual production and distribution.'" Id. at 1032. Thus, a law limiting only possession was "part of a larger regulation of economic activity." Id. (quoting Lopez, 514 U.S. at 561, 115 S.Ct. 1624) (internal quotation marks omitted).

Adams, however, is distinguishable from Stewart's case because Adams involved commercial child pornography that had been bought in the open market. Id. at 1030. Purchase of these illegal materials thus stimulates the demand for others to produce and sell them. By contrast, Stewart's homemade machineguns did not stimulate a demand for anything illegal — all the components he bought were legally available from commercial sources. This case is much closer to McCoy, where McCoy's photographs, which were intended "for her own personal use," did not "`compete' with other depictions exchanged, bought or sold in the illicit market for child pornography and did not affect their availability or price." McCoy, 323 F.3d at 1122. Similarly, by crafting his own guns and working out of his own home, Stewart functioned outside the commercial gun market. His activities obviously did not increase machinegun demand. Nor can we say that Stewart's homemade machineguns reduced overall demand. Unlike wheat, for example, which is a staple commodity that Filburn would probably have had to buy, had he not grown it himself, there is no reason to think Stewart would ever have bought a machinegun from a commercial source, had he been precluded by law from building one himself.3 In fact, the evidence suggests that Stewart was cognizant of the law and made careful efforts not to come into conflict with it.4 Thus, the link between Stewart's activity and its effect on interstate commerce is simply too tenuous to justify federal regulation.

This case fails Morrison's other requirements as well. As we stated earlier, section 922(o) contains no jurisdictional element anchoring the prohibited activity to interstate commerce. Congress also failed to make any legislative findings when it enacted the statute. While neither Lopez nor Morrison requires Congress to make findings every time it passes a law under its Commerce Clause power, the Supreme Court did note the importance of findings where—as here—such findings would "enable [a court] to evaluate the legislative judgment that the activity in question substantially affected interstate commerce, even though no such substantial effect was visible to the naked eye." Lopez, 514 U.S. at 563, 115 S.Ct. 1624.

The Third Circuit looked at the legislative findings of other federal firearms statutes as evidence of a nexus between machinegun possession and interstate commerce, because "the subject matter of § 922(o) is sufficiently similar to that of the other legislation accompanied by these findings so as to be a reliable statement of the rationale for Congress' authority to pass § 922(o)." Rybar, 103 F.3d at 279. Putting aside whether it is ever appropriate to shuttle legislative findings from one statute to another in order to establish a Commerce Clause nexus, we cannot see how the findings imported by the Third Circuit have any bearing on the constitutionality of section 922(o).

Section 922(o) is quite different from previous firearms regulations. Whereas section 922(o) addresses possession of machineguns, all of the earlier legislation cited by the Third Circuit deals with transactions, sales or deliveries of firearms, and nearly all of the provisions specifically require that the transaction, sale or delivery be conducted interstate.5 All of these provisions are cut from the ordinary cloth of Commerce Clause regulation of interstate commerce, while section 922(o) is much closer to the statute struck down in Lopez. That statute criminalized gun possession in a particular location—a school zone. Section 922(o) criminalizes possession of a particular type of firearm—a machinegun. The latter no more has an inherent link to interstate commerce than the former. The Supreme Court found that the school zones statute "`plow[ed] thoroughly new ground and represent[ed] a sharp break with the long-standing pattern of federal firearms legislation,'" Lopez, 514 U.S. at 563, 115 S.Ct. 1624 (quoting United States v. Lopez, 2 F.3d 1342, 1366 (5th Cir.1993)). As a result, the Court found it was "especially inappropriate" to import previous legislative findings to justify the statute there. Id. Section 922(o) is no less of a "sharp break" from previous regulations.

Moreover, nothing in the legislative history of any of the earlier firearms statutes speaks to the relationship between mere possession of firearms and interstate commerce. Instead, the legislative findings focus primarily on the need for federal enforcement where firearms cross state and international borders, and are thus difficult for individual states to regulate on their own. The legislative findings supporting the Omnibus Act, for example, address the need for federal regulation to "adequately enable the States to control the firearms traffic within their own borders through the exercise of their police power." S.Rep. No. 90-1097 (1968), reprinted in 1968 U.S.C.C.A.N. 2112, 2197 (emphasis added). More specifically, Congress found that "the United States has become the dumping ground of the castoff surplus military weapons of other nations," which has "contributed greatly to lawlessness and to the Nation's law enforcement problems." 1968 U.S.C.C.A.N. at 2199 (emphasis added). Congress also found that "the lack of adequate Federal control over interstate and foreign commerce in highly destructive weapons ... has allowed such weapons and devices to fall into the hands of lawless persons, ... thus creating a problem of national concern." Id. The Gun Control Act's findings similarly discuss only the need "to strengthen Federal controls over interstate and foreign commerce in firearms and to assist the States effectively to regulate firearms traffic within their borders." H.R.Rep. No. 90-1577 (1968), reprinted in 1968 U.S.C.C.A.N. 4410, 4411 (emphases added). Nothing in the legislative history suggests that Congress ever considered the impact of purely intrastate possession of homemade machineguns on interstate commerce, and there is no reason to assume that prohibiting local possession of machineguns would have the same national and commercial consequences as prohibiting the interstate and foreign traffic in firearms. We therefore cannot import these earlier legislative findings to give section 922(o) constitutional grounding.

Based on the four-factor Morrison test, section 922(o) cannot be viewed as having a substantial effect on interstate commerce. We therefore conclude that section 922(o) is unconstitutional as applied to Stewart.

3. This raises the question posed by the dissent in McCoy—whether claims under the Commerce Clause are susceptible to as-applied challenges at all. McCoy, 323 F.3d at 1133 (Trott, J., dissenting). According to the McCoy dissent, once it is determined that a particular statute is a legitimate exercise of congressional power under the Commerce Clause, an individual may not escape the statute's sweep by showing that his particular activities lack an interstate nexus. McCoy, of course, found an as-applied violation and thus controls this case. However, because the McCoy majority did not address the dissent's superficially plausible arguments, we do so here.

The dissent in McCoy asserted that as-applied challenges cannot be brought under the Commerce Clause, relying on a single sentence from Lopez for support: "[W]here a general regulatory statute bears a substantial relation to commerce, the de minimis character of individual instances arising under that statute is of no consequence." Id. at 1134 (citing Lopez, 514 U.S. at 558, 115 S.Ct. 1624 (emphasis omitted)). The McCoy dissent took this sentence entirely out of context.

Lopez itself borrowed this sentence from a footnote in Maryland v. Wirtz, 392 U.S. 183, 88 S.Ct. 2017, 20 L.Ed.2d 1020 (1968)—a case that had nothing to do with as-applied challenges, but instead announced the so called "enterprise concept," which allows Congress to exercise authority over a large enterprise or industry by regulating its smaller components, even those components that bear no relation to interstate commerce on their own. See id. at 188, 196-97 n. 27, 88 S.Ct. 2017 (discussing the definition of the term "enterprise").6 Wirtz held that Congress could regulate a group of employees who had no direct connection to interstate commerce, reasoning that labor-related "strife disrupting an enterprise involved in commerce may disrupt commerce," and that "substandard labor conditions among any group of employees, whether or not they are personally engaged in commerce or production, may lead to strife disrupting an entire enterprise." Id. at 192, 88 S.Ct. 2017. The Court in Wirtz was careful to explain that, although the employees' activities were not themselves in interstate commerce, Congress had reasonably determined they had a material effect on interstate commerce because of their participation in the larger enterprise. The Court employed a similar mode of analysis in Wickard. It held that, though Wickard's homegrown wheat may not have traveled interstate, it had a material effect on the interstate price of wheat: "[T]aken together with [the homegrown wheat] of many others similarly situated," it had an aggregate effect on commerce that was "far from trivial." Wickard, 317 U.S. at 128, 63 S.Ct. 82.7

Read in context, the sentence quoted by the McCoy dissent can only mean that, where a general regulatory statute governs a large enterprise, it does not matter that its components have a de minimis relation to interstate commerce on their own. What does matter is that the components could disrupt the enterprise, and could thus interfere with interstate commerce. In the Wirtz situation, then, the enterprise is the mechanism through which a multitude of the intrastate effects are consolidated and amplified so that they have an effect on interstate commerce. This obviously has no bearing at all on a case such as ours where the activity in question is not part of a large enterprise that itself has an effect on interstate commerce.

Our Commerce Clause jurisprudence supports this reading. Before cases like Wirtz, the Court drew a much sharper line between local and interstate commerce, holding that certain activities such as production, manufacturing and mining were exclusively the province of state governments. See, e.g., United States v. E.C. Knight Co., 156 U.S. 1, 12, 15 S.Ct. 249, 39 L.Ed. 325 (1895) (holding that manufacturing is not commerce and thus is not subject to Congress's commerce power). Cases like Wirtz and Wickard were thus quite radical in their expansive conception of the Commerce Clause, because they first articulated Congress's power to regulate persons and things twice and thrice removed from interstate commerce. See Lopez, 514 U.S. at 556, 115 S.Ct. 1624 (describing Wickard as "usher[ing] in an era of Commerce Clause jurisprudence that greatly expanded the previously defined authority of Congress under that Clause"). But this is entirely different than saying Congress can regulate someone with no relation to interstate commerce at all — such as a person who builds a machinegun from scratch in his garage — so long as there is an otherwise valid statute that covers his activities. There is nothing in Wirtz, Wickard, Lopez, or in any of our cases — not even buried in a footnote — suggesting this understanding of the Commerce Clause is plausible.

Quite the contrary, the Supreme Court has always entertained as-applied challenges under the Commerce Clause. In Heart of Atlanta Motel, Inc. v. United States, 379 U.S. 241, 85 S.Ct. 348, 13 L.Ed.2d 258 (1964), for example, the Court found Title II of the Civil Rights Act of 1964 was valid "as applied ... to a motel which concededly serves interstate travelers." Id. at 261, 85 S.Ct. 348. In Katzenbach v. McClung, 379 U.S. 294, 85 S.Ct. 377, 13 L.Ed.2d 290 (1964), the Court found the same statutory provision valid "as applied to a restaurant annually receiving about $70,000 worth of food which has moved in commerce." Id. at 298, 85 S.Ct. 377. If the dissent in McCoy were right, we would have only needed one case to say Title II is valid, period. There would have been no need to consider — as the Court did — whether a single hotel or restaurant had a sufficient nexus to interstate commerce, and could thus be federally regulated. Wickard was also an as-applied challenge: Had the Court deemed regulation of the business of agriculture a sufficient basis for upholding the application of the Agricultural Adjustment Act to Filburn, there would have been no need for it to analyze how his particular activities affected interstate commerce.

Indeed, it is hard to believe the Court would ever eliminate as-applied challenges for one particular area of constitutional law. As Professor Fallon explains, "[a]s-applied challenges are the basic building blocks of constitutional adjudication." Richard H. Fallon, Jr., As-Applied and Facial Challenges and Third-Party Standing, 113 Harv. L.Rev. 1321, 1328 (2000). An as-applied challenge asks a court to consider whether a statute's application to a particular litigant is a valid one. Whereas the "enterprise concept" is only relevant when a party is regulated in relation to a large industry or enterprise, whether a given statute can constitutionally be applied to a claimant is an inquiry that occurs in every constitutional case:

In order to raise a constitutional objection to a statute, a litigant must always assert that the statute's application to her case violates the Constitution. But when holding that a statute cannot be enforced against a particular litigant, a court will typically apply a general norm or test and, in doing so, may engage in reasoning that marks the statute as unenforceable in its totality. In a practical sense, doctrinal tests of constitutional validity can thus produce what are effectively facial challenges. Nonetheless, determinations that statutes are facially invalid properly occur only as logical outgrowths of rulings on whether statutes may be applied to particular litigants on particular facts.

Id. at 1327-28. Professor Fallon also notes that "[t]raditional thinking has long held that the normal if not exclusive mode of constitutional adjudication involves an as-applied challenge." Id. at 1321 (citing United States v. Raines, 362 U.S. 17, 20-21, 80 S.Ct. 519, 4 L.Ed.2d 524 (1960), and Yazoo & Miss. Valley R.R. v. Jackson Vinegar Co., 226 U.S. 217, 219-20, 33 S.Ct. 40, 57 L.Ed. 193 (1912)); see also United States v. Salerno, 481 U.S. 739, 745, 107 S.Ct. 2095, 95 L.Ed.2d 697 (1987) ("A facial challenge to a legislative Act is, of course, the most difficult challenge to mount successfully...."). We therefore cannot agree with the bold assertion in the McCoy dissent that an as-applied challenge is inapposite in cases such as this.

Finally, Stewart argues that the Second Amendment guarantees him the right to possess machineguns, as well as the right to possess firearms generally despite his former felony conviction—as charged in count one of Stewart's indictment. We have held that the Second Amendment "was not adopted in order to afford rights to individuals with respect to private gun ownership or possession." Silveira v. Lockyer, 312 F.3d 1052, 1087 (9th Cir.2002). Thus, there is no Second Amendment limitation on "legislation regulating or prohibiting the possession or use of firearms." Id. Stewart's Second Amendment argument must therefore fail. We reverse Stewart's conviction for machinegun possession under section 922(o) as an unlawful extension of Congress's commerce power and affirm his conviction for possession of firearms by a felon.

AFFIRMED in part and REVERSED in part.

RESTANI, Judge, concurring in part, dissenting in part:

I dissent from that part of the majority's opinion which finds 18 U.S.C. § 922(o) unconstitutional as applied to a machine gun partially home manufactured from legal parts. I agree that this case is not controlled by prior circuit precedent, which relies on earlier illegal transfers. See, e.g., United States v. Rambo, 74 F.3d 948 (9th Cir.1996). Rather, I adopt the reasoning of the Seventh Circuit in United States v. Kenney, 91 F.3d 884 (7th Cir.1996), which finds that the regulation of possession, as well as transfer, of machine guns is part of Congress's long standing efforts to regulate the trade in machine guns, that is, to regulate the whole of the economic activity of trade in machine guns. Id. at 890 (upholding the constitutionality of § 922(o) as a regulation of activity substantially affecting interstate commerce).

Unlike the majority, and like the court in Kenney, I find Wickard v. Filburn, 317 U.S. 111, 63 S.Ct. 82, 87 L.Ed. 122 (1942) controlling. Possession of machine guns, home manufactured or not, substantially interferes with Congress's long standing attempts to control the interstate movement of machine guns by proscribing transfer and possession. Congress's chosen method in § 922(o) was to totally eliminate the demand side of the economic activity by freezing legal possession at 1986 levels, "an effect that is closely entwined with regulating interstate commerce" even as applied to purely intrastate possession of machine guns resulting from home manufacture. Kenney, 91 F.3d at 890. Allowing home manufacture is clearly not within the intent of § 922(o) and would upset Congress's entirely lawful plan to regulate trade in machine guns. Accordingly, I dissent in part.