This case presents the question whether the exclusionary rule applies in federal supervised release revocation hearings. For the reasons that follow, we hold that it does not.

In 1992, appellant David Mark Armstrong pled guilty in the United States District Court for the Eastern District of Virginia to conspiracy to possess and distribute cocaine and cocaine base. He was sentenced to 210 months in prison and five years of supervised release. Pursuant to a government motion under Fed. R. Crim. P. 35, his prison sentence was subsequently reduced to 70 months.

On January 17, 1997, Armstrong was released from custody and began his term of supervised release. On April 22, 1998, however, he was arrested in Washington, D.C., after officers searching his vehicle following a routine traffic stop discovered a loaded.38-caliber revolver and 36 plastic bags containing 3.1 grams of cocaine base. Armstrong was subsequently charged in the United States District Court for the District of Columbia with possession with intent to distribute cocaine base, possession of a firearm by a convicted felon, and using or carrying a firearm during a drug trafficking offense. Before that court, Armstrong moved to suppress the evidence obtained during the search of his car on the ground that the search was unreasonable under the Fourth Amendment. On October 1, 1998, the district court granted the motion, and soon thereafter the government dismissed the charges against Armstrong.

Meanwhile, the government had filed a petition before the United States District Court for the Eastern District of Virginia, seeking revocation of Armstrong's supervised release. In response, Armstrong contended that the exclusionary rule should operate in the revocation proceeding and that, because the district court in the District of Columbia had held that the search of Armstrong's car was unreasonable, the government was collaterally estopped from challenging the constitutionality of the search in the revocation proceeding. The district court held that the exclusionary rule does not operate in federal supervised release revocation proceedings, and therefore granted the government's petition. See United States v. Armstrong, 30 F. Supp. 2d 901, 906 (E.D. Va. 1998). Armstrong appeals.

The district court based its conclusion that the exclusionary rule does not apply in federal supervised release revocation proceedings on the Supreme Court's recent decision in Pennsylvania Bd. of Probation and Parole v. Scott, 118 S. Ct. 2014 (1998). We agree with the district court that the Supreme Court's decision in Scott compels such a conclusion.

In Scott, the Supreme Court held that the exclusionary rule does not apply in a state parole revocation hearing. See id. at 118 S.Ct. at 2022. In so deciding, the Court undertook the familiar task of weighing the costs of applying the exclusionary rule in a particular context against its benefits. See, e.g., United States v. Calandra, 414 U.S. 338, 349-52 (1974).

The Court began by considering the costs of applying the exclusionary rule. The Court first noted that, because of the high costs of applying the exclusionary rule generally, it had"repeatedly declined to extend the exclusionary rule to proceedings other than criminal trials." Scott, 118 S. Ct. at 2019.1 The Court then proceeded to observe that "[t]he costs of excluding reliable, probative evidence are particularly high in the context of parole revocation proceedings." Id. 118 S.Ct. at 2020. In support of this conclusion, the Court cited two rationales. First, it observed that parole was simply a more lenient variation on imprisonment, in which the government extended a limited degree of freedom to a parolee in return for the parolee's assurance that he would comply with the terms of his parole. See id. The Court reasoned that the application of the exclusionary rule would hamper the government's ability to ensure such compliance and even deter the government from offering parole in the first place, and it further noted that the application of the exclusionary rule in the parole context would be especially troublesome because parolees are more likely to commit criminal offenses than are average citizens. See id. Second, the Court concluded that the application of the exclusionary rule would be "incompatible with the traditionally flexible, administrative procedures of parole revocation." Id. Specifically, the Court reasoned that the application of the exclusionary rule would often necessitate extensive litigation and would thereby alter the relatively less adversarial nature of parole revocation proceedings. See id. 118 S.Ct. at 2021.

The Court then considered the benefits of applying the exclusionary rule. The Court reasoned that applying the exclusionary rule would have minimal, if any, deterrence benefits, since an officer would already be deterred from violating a parolee's Fourth Amendment rights because he would know that any violation would lead to the application of the exclusionary rule in any subsequent criminal proceedings for the offense being investigated, even if it did not in any collateral parole revocation proceedings. See id. at 2021-22. The Court concluded that any deterrence benefits did not outweigh the substantial costs, and therefore refused to extend the exclusionary rule to parole revocation proceedings. See id. at 2022.

Appellant contends that the reasoning of Scott is inapplicable to the instant case on two grounds. First, appellant asserts that Scott involved a parole revocation proceeding, whereas the instant case involves a supervised release revocation proceeding. However, as we have previously noted, parole and supervised release are "analogous contexts." United States v. Woodrup , 86 F.3d 359, 361 (4th Cir. 1996). Indeed, Congress designed supervised release as the successor to parole in the federal criminal system, because it believed that the parole system provided inadequate supervision. See generally United States v. Montenegro-Rojo, 908 F.2d 425, 432-33 (9th Cir. 1990) (discussing history of supervised release statute). For purposes of the rule established in Scott, moreover, parole and supervised release are not just analogous, but virtually indistinguishable. As the district court correctly noted, the costs and benefits of applying the exclusionary rule to revocation proceedings are almost identical in the parole and supervised release contexts. Although supervised release revocation proceedings, unlike parole revocation proceedings, do take place before a judge, they are characterized by the same"flexibility" that the Supreme Court found significant in Scott. As in parole revocation proceedings, findings of fact are made under a preponderance-of-theevidence, rather than reasonable-doubt, standard, see 18 U.S.C. § 3583(e)(3); the traditional rules of evidence are inapplicable, see Fed. R. Evid. 1101(d)(3); United States v. Frazier, 26 F.3d 110, 11214 (11th Cir. 1994) (concluding that Fed. R. Evid. 1101(d)(3) extends to supervised release revocation proceedings); and the "full panoply of constitutional protections afforded a criminal defendant" is not available, Woodrup, 86 F.3d at 361.

Second, appellant contends that Scott is distinguishable because it involved a state, rather than a federal, revocation proceeding. In Scott, however, the Court did not find the fact that the proceeding was a state proceeding determinative. Although the Court did suggest that the application of the exclusionary rule would work an intrusion into "States' correctional schemes," id. at 2021, the Court relied more broadly on the "traditionally flexible, administrative" nature of parole revocation schemes, id. at 2020 -a description that, as noted above, would fit federal revocation proceedings just as well as state ones.

In sum, because the reasoning of Scott applies equally to supervised release revocation proceedings as to parole revocation proceedings, and to federal proceedings as to state proceedings, we agree with the district court that Scott requires that the exclusionary rule not be extended to federal supervised release revocation proceedings.

The district court further concluded that the Supreme Court's decision in Scott effectively superseded our decision in United States v. Workman, 585 F.2d 1205 (4th Cir. 1978). We agree.

In Workman, we held that the exclusionary rule applies in federal probation revocation hearings. See id. at 1211. Like the Supreme Court in Scott, we applied the cost-benefit balancing test outlined by the Court in Calandra. See id. at 1209. We reasoned that "an inferior court," rather than engaging in such a balancing inquiry from scratch, "should . . . compar[e] revocation hearings with other instances where the Supreme Court has considered the rule's application." Id. at 1210. Because "the Supreme Court ha[d] never exempted from the operation of the exclusionary rule any adjudicative proceeding in which the government offers unconstitutionally seized evidence in direct support of a charge that may subject the victim of a search to imprisonment," id. at 1211, we concluded that the exclusionary rule was applicable in federal probation revocation hearings, see id.

At the outset, we acknowledge that probation, like parole, is "analogous" to supervised release, see Woodrup, 86 F.3d at 361, and therefore that the rule in Workman would appear to be applicable in this case. As the district court correctly noted, however, "[i]t follows inescapably from Workman's teachings that the Fourth Circuit panel in Workman, under its own analysis, would reach a different result today." Armstrong, 30 F. Supp. 2d at 904. Indeed, by suggesting that "an inferior court should . . . compar[e] revocation hearings with other instances where the Supreme Court has considered the rule's application," Workman, 585 F.2d at 1210, we expressly contemplated the possibility that the rule we set out in Workman would evolve in tandem with the Supreme Court's exclusionary-rule jurisprudence. And today, our assertion in Workman that "the Supreme Court has never exempted from the operation of the exclusionary rule any adjudicative proceeding in which the government offers unconstitutionally seized evidence in direct support of a charge that may subject the victim of a search to imprisonment," id. at 1211, is unquestionably no longer correct. Even before the Supreme Court's decision in Scott, the Court had substantially limited the applicability of the exclusionary rule, even in the context of criminal trials themselves. See Scott, 118 S. Ct. at 2020 n.4 (listing cases, most notably United States v. Leon).2 We therefore conclude that our decision in Workman has effectively been superseded by the Supreme Court's recent decision in Scott, if it had not been already by the Supreme Court's prior decisions. See generally Smith v. Moore, 137 F.3d 808, 821 (4th Cir.) (noting that "[i]t is well established that a decision of this Court is binding on other panels unless it is overruled by . . . an intervening decision of the United States Supreme Court"), cert. denied, 119 S. Ct. 199 (1998). Accordingly, we reaffirm our conclusion, on the basis of Scott, that the exclusionary rule does not apply in supervised release revocation proceedings.3

The district court further concluded that appellant was in constructive possession of the gun and drugs when they were discovered in the course of the search of his car. We agree.

A person has constructive possession over contraband when he has ownership, dominion, or control over the contraband itself or over the premises or vehicle in which it was concealed. See, e.g., United States v. Blue, 957 F.2d 106, 107 (4th Cir. 1992) (en banc). In this case, as the district court noted, not only did appellant own the car in which the contraband was concealed, but appellant was the only person in the car when it was stopped, the gun was under a floormat on the driver's side, and the drugs were located in a compartment to the left of the steering wheel. When taken together, these findings amply support the district court's conclusion that appellant constructively possessed the gun and drugs found during the search. The judgment of the district court is affirmed.