We have taken this case en banc because of a conflict among our decisions in the application of the concurrent sentence doctrine. In doing so, we considered the propriety of using the doctrine as a discretionary way to avoid reviewing criminal convictions. However, the doctrine's disadvantages outweigh its advantages. Therefore, we reject its application in this and future cases.

* DeBright was convicted of conspiring to import heroin, illegally importing heroin, conspiring to possess heroin with intent to distribute, and illegally possessing heroin with intent to distribute in violation of 21 U.S.C. Secs. 963, 952(a), 960(a)(1), 841(a)(1), and 18 U.S.C. Sec. 2. She was sentenced to concurrent six-year sentences on all four counts. DeBright's arguments on appeal focused primarily on her conspiracy to import conviction in Count One. Because DeBright virtually conceded her guilt under the other three counts, the panel hearing this appeal concluded that it was unnecessary to reach the merits of her challenge to the conspiracy conviction under Count One. United States v. DeBright, 710 F.2d 1404, 1405 (9th Cir.1983) (DeBright ). The panel relied on the concurrent sentence doctrine, which allows:

the appellate court, as a matter of discretion, [to decline] review [of] a conviction under one count if a conviction under another count is affirmed and the sentences run concurrently and no adverse collateral legal consequences for the appellant result from the additional conviction.

United States v. Martin, 599 F.2d 880, 887 (9th Cir.), cert. denied, 441 U.S. 962, 99 S. Ct. 2407, 60 L. Ed. 2d 1067 (1979). The panel vacated DeBright's unreviewed conspiracy conviction.

In our discretionary use of the concurrent sentence doctrine over nearly four decades, see Haid v. United States, 157 F.2d 630, 631 n. 2 (9th Cir.1946); Maxfield v. United States, 152 F.2d 593, 595 (9th Cir.1945), cert. denied, 327 U.S. 794, 66 S. Ct. 821, 90 L. Ed. 1021 (1946), we have, with one exception, always affirmed the unreviewed conviction. In United States v. Fishbein, 446 F.2d 1201, 1205-06 (9th Cir.1971), we vacated the unreviewed conviction. In the subsequent twelve years until DeBright, we had never followed Fishbein but instead invariably adhered to our long-standing practice of affirming the unreviewed conviction. See, e.g., United States v. Ford, 632 F.2d 1354, 1370 & n. 16 (9th Cir.1980), cert. denied, 450 U.S. 934, 101 S. Ct. 1399, 67 L. Ed. 2d 369 (1981); United States v. Weislow, 485 F.2d 560, 562 (9th Cir.1973), cert. denied, 415 U.S. 933, 94 S. Ct. 1447, 39 L. Ed. 2d 491 (1974). The resurrection of Fishbein by DeBright caused us to face the initial question of whether we should vacate the unreviewed conviction when we apply the concurrent sentence doctrine.II

In determining to vacate rather than affirm DeBright's unreviewed conviction, the panel relied heavily on the decision of the Court of Appeals for the District of Columbia Circuit in United States v. Hooper, 432 F.2d 604 (D.C.Cir.1970) (Hooper ). There the court concluded that neither the government nor the public had any significant interest in maintaining the judgment under a count that was rendered "superfluous" by the imposition of concurrent sentences. Id. at 606. In recent decisions, the Fifth and Eleventh Circuits have followed Hooper. United States v. Butera, 677 F.2d 1376, 1386 (11th Cir.1982), cert. denied, --- U.S. ----, 103 S. Ct. 735, 74 L. Ed. 2d 958 (1983); United States v. Cardona, 650 F.2d 54, 57-58 (5th Cir.1981).

We conclude that vacating unreviewed sentences under the concurrent sentence doctrine is a fundamentally erroneous practice. Under the separation of governmental powers established by the Constitution, the executive branch has the primary responsibility for determining which violations of the law shall be prosecuted. See United States v. Miller, 722 F.2d 562, 565 (9th Cir.1983) ("[S]eparation of powers requires that the judiciary remain independent of executive affairs.... Charging decisions are generally within the prosecutor's exclusive domain."); see also Gray v. Bell, 712 F.2d 490, 513 (D.C.Cir.1983) ("[P]rosecutorial discretion is exercised pursuant to expansive constitutional and statutory authority delegated to the executive."). Courts may exercise supervisory power over the administration of the criminal justice system and encroach upon the prerogatives of the executive branch by dismissing individual indictments only where there is "a clear basis in fact and law for doing so." United States v. Chanen, 549 F.2d 1306, 1313 (9th Cir.), cert. denied, 434 U.S. 825, 98 S. Ct. 72, 54 L. Ed. 2d 83 (1977); see also United States v. Hasting, --- U.S. ----, 103 S. Ct. 1974, 1978-79, 76 L. Ed. 2d 96 (1983) (describing the limited situations that justify use of courts' supervisory powers) ("The purposes underlying use of the supervisory powers are threefold: to implement a remedy for violation of recognized rights, ... to preserve judicial integrity by ensuring that a conviction rests on appropriate considerations validly before the jury, ... and finally, as a remedy designed to deter illegal conduct, ....") (citations omitted). Adopting a rule which allows vacating convictions without reviewing their merits is not an appropriate use of our supervisory authority. It would impermissibly infringe on the prosecutorial function of the executive branch. Cf. United States v. Miller, 722 F.2d at 565 ("[C]ourts ... should avoid creating broad rules that limit traditional prosecutorial independence."); United States v. Real, 446 F.2d 40, 40 (9th Cir.1971) (judicial discretion may not be substituted for prosecutorial discretion in determining whether to dismiss an indictment).

We also reject Hooper's basic assumption that the government and public lack an interest in retaining unreviewed convictions under the concurrent sentence doctrine. The convictions represent the expenditure of society's resources in the investigation and prosecution of criminal conduct. Our society has established as one possible consequence of a criminal conviction the necessity of living with the record of having been convicted. We find it inappropriate for us to remove this socially imposed sanction merely because it serves the interest of judicial economy.

Finally, there are substantial practical difficulties with the implementation of Hooper. Under Hooper, the judgment of conviction is vacated but the jury verdict remains intact. The government is then invited to seek reinstatement of the judgment at any point in the future if new or changing circumstances make a conviction significant. Hooper, 432 F.2d at 606 n. 8. If the conviction were reinstated, it would then be subject to appellate review. Id. This unwieldy procedure is fraught with the potential for delay and the wasting of judicial resources. Inevitably, it would result in fragmented appeals of issues that could have been disposed of by the original reviewing court. This practice also places an onerous burden on the government to monitor ongoing circumstances that might be relevant to the previously unreviewed convictions. Consider, for example, a Parole Commission which considers a ten-year-old unreviewed conviction relevant to its determination of a parole date. The vacated conviction would then be significant to the government under Hooper. Presumably, the Parole Commission would then recess, while the government seeks reinstatement of the judgment of conviction, after which the defendant would appeal. If the previously unreviewed conviction were affirmed, a year or so later, the Parole Commission would resume consideration of a parole date. If the conviction were instead reversed, the government and defendant would attempt to find witnesses and evidence to retry a now stale case. We reject this procedure as impractical.

In some appeals from multiple count criminal convictions with concurrent sentences, the societal interests and those of the administration of justice may point towards the wisdom of dismissal of a count. But that can best be accommodated in our system by an appropriate policy instituted by the Department of Justice, not by the ad hoc actions of an appellate court.

Our rejection of the Hooper procedure does not end our en banc mission. We now consider application of the concurrent sentence doctrine in its original form: affirming the unreviewed conviction. It is clear that the concurrent sentence doctrine is only a rule of judicial convenience. See Benton v. Maryland, 395 U.S. 784, 791, 89 S. Ct. 2056, 2060, 23 L. Ed. 2d 707 (1969) (Benton ). Its discretionary use is justifiable only if the unreviewed conviction has no "adverse collateral legal consequences" for the convicted individual. See United States v. Martin, 599 F.2d at 887. We do not question the propriety of this limit on the doctrine's use. We have serious doubts, however, about our ability to ascertain all the adverse collateral legal consequences of unreviewed convictions and, assuming we have that ability, the practicality of attempting to ascertain those consequences.

Cases involving the concurrent sentence doctrine have frequently enumerated specific adverse collateral legal consequences of criminal convictions that should be considered in determining whether to apply the doctrine. E.g., Benton, 395 U.S. at 790-91, 89 S.Ct. at 2060; United States v. Vargas, 615 F.2d 952, 959-60 (2d Cir.1980). In Vargas, the Second Circuit listed five possible adverse collateral legal consequences of unreviewed convictions: effect on parole; possible application of recidivist statutes; potential impeachment in future trials; effect on pardon; and possible stigma. Id. Although not exhaustive, this list effectively demonstrates that making the necessary exploration of all possible adverse collateral legal consequences before affirming a conviction under the concurrent sentence doctrine would ordinarily be very time-consuming. Judicial economy, however, is the only justification for the doctrine. See id. at 960; see also Note, The Concurrent Sentence Doctrine After Benton v. Maryland, 7 U.C.L.A.-Alaska L.Rev. 282, 301-03 (1978). The inconsistency is obvious.

Application of the concurrent sentence doctrine inherently requires the assessment of all adverse collateral legal consequences of unreviewed convictions at the time the conviction is appealed. As a practical matter, it may be impossible at that time to foretell all future adverse collateral legal consequences for the appellant. Thus, our application of the concurrent sentence doctrine has been based only on a determination that no collateral consequences were apparent. This approach, unfortunately, placed the risk of our lack of omniscience on the party who will, without present or future review, suffer from the mistake.

Some circuits have attempted to avoid this potentially unjust result through a rough balancing technique. They simply presume that a conviction will have adverse collateral legal consequences substantial enough to preclude the concurrent sentence doctrine's application. See, e.g., United States v. Webster, 639 F.2d 174, 182-83 (4th Cir.) ("[W]e cannot predict with certainty whether the ... convictions will at some future point work some unforeseen harm to the defendants .... We refuse to place the risks of future collateral consequences upon the defendants."), cert. denied, 454 U.S. 857, 102 S. Ct. 307, 70 L. Ed. 2d 152 (1981); Gentry v. United States, 533 F.2d 998, 1000-01 (6th Cir.1976). This approach leaves the door open to application of the doctrine in cases where it is sufficiently certain that no adverse consequences will result. See, e.g., United States v. Truong Dinh Hung, 629 F.2d 908, 931-32 (4th Cir.1980) (Russell, J., concurring and dissenting), cert. denied, 454 U.S. 1144, 102 S. Ct. 1004, 71 L. Ed. 2d 296 (1982).

Rather than expend our time and efforts attempting to satisfy ourselves that a particular conviction does or does not harbor any future adverse collateral legal consequences, we conclude that it is preferable to address ourselves to the merits of all convictions before us on appeal. This will guarantee that no individual will suffer because of our inability to foretell the future effects of an unreviewed conviction. It will also protect society's interest in holding convicted criminals accountable for each of their convictions unless there is an adequate legal reason to set them aside. We overrule our prior cases which have authorized use of the concurrent sentence doctrine.

An additional reason counsels against maintenance of the doctrine in any form. Every federal criminal defendant has a statutory right to have his or her conviction reviewed by a court of appeals. 28 U.S.C. Sec. 1291; Coppedge v. United States, 369 U.S. 438, 82 S. Ct. 917, 8 L. Ed. 2d 21 (1962). The statutory right to appeal is deemed so important that a district court judge is required to inform a defendant specifically of that right after trial and sentencing. See Fed.R.Crim.P. 32(a)(2). The right to appeal is from the judgment of conviction and not from the sentence. See Coppedge, 369 U.S. at 441, 82 S.Ct. at 919 ("Present federal law has made an appeal from a District Court's judgment of conviction in a criminal case what is, in effect, a matter of right.") (emphasis added). That right encompasses all convictions. The concurrent sentence doctrine, by permitting an appellate court to decline to review a conviction for reasons of judicial economy, impinges upon the defendant's statutory right.

We recognize that in rejecting the concurrent sentence doctrine, we eliminate a tool that we adopted to direct our limited time for judicial review to other cases. The one potential disadvantage of the approach we now adopt is that it may increase our workload and slow the judicial process. As we concluded above, however, it is not at all clear that the doctrine, if carefully applied, actually does serve the interest of judicial economy. Moreover, we have applied the concurrent sentence doctrine in fewer than ten reported cases during the last four years. See United States v. Barker, 675 F.2d 1055, 1059 (9th Cir.1982); United States v. Kaiser, 660 F.2d 724, 732 (9th Cir.1981), cert. denied, 455 U.S. 956, 102 S. Ct. 1467, 71 L. Ed. 2d 674 (1982); United States v. Lipps, 659 F.2d 960, 962-63 (9th Cir.1981); United States v. Federico, 658 F.2d 1337, 1344 (9th Cir.1981); United States v. Ford, 632 F.2d 1354, 1370 n. 16 (9th Cir.1980), cert. denied, 450 U.S. 934, 101 S. Ct. 1399, 67 L. Ed. 2d 369 (1981); United States v. Hughes, 626 F.2d 619, 626-27 (9th Cir.), cert. denied, 449 U.S. 1065, 101 S. Ct. 793, 66 L. Ed. 2d 611 (1980); United States v. Price, 623 F.2d 587, 593 (9th Cir.), cert. denied, 449 U.S. 1016, 101 S. Ct. 577, 66 L. Ed. 2d 475 (1980); United States v. Ponticelli, 622 F.2d 985, 992 (9th Cir.), cert. denied, 449 U.S. 1016, 101 S. Ct. 578, 66 L. Ed. 2d 476 (1980); United States v. Jabara, 618 F.2d 1319, 1329 (9th Cir.), cert. denied, 449 U.S. 856, 101 S. Ct. 154, 66 L. Ed. 2d 70 (1980). It is questionable, therefore, whether our rejection of the concurrent sentence doctrine will appreciably affect our workload. But even if it does, we cannot conclude in good conscience that we possess the ability to predict with sufficient certainty all the adverse collateral legal consequences of unreviewed convictions. The doctrine is thus unfair to defendants and inappropriate in our criminal justice system. We therefore conclude that the advantages of reviewing each conviction on its merits substantially outweigh the threatened harm from any resulting decrease in judicial efficiency.

We reject the use of the concurrent sentence doctrine as a discretionary means of avoiding the review of criminal convictions. We remand this case to the panel to allow consideration of the merits of DeBright's challenges to her conviction under Count One of the indictment.