Marino Amador-Leal appeals his conviction and sentence pursuant to a guilty plea on one count of possession with intent to distribute cocaine base in violation of 21 U.S.C. § 841(a)(1). The magistrate judge did not explain the potential immigration consequences of Amador-Leal’s conviction when the plea was taken. The question presented here is whether immigration consequences are collateral, as we held in
Fruchtman v. Kenton,
I
Amador-Leal is an illegal alien who was caught selling crack in February, 2000. He was charged with conspiracy to possess cocaine base in violation of 21 U.S.C. § 846, and with possession with intent to distribute in violation of 21 U.S.C. § 841(a)(1). The indictment also included criminal forfeiture charges under 21 U.S.C. § 853.
Pursuant to a written plea agreement which, among other things, waived appeal, Amador-Leal entered a plea of guilty to the possession count on May 18, 2000. The magistrate judge reviewed the charges with Amador-Leal, explained his rights and the implications of his plea, and found that he knowingly and voluntarily agreed to the plea. Based on the plea agreement and Amador-Leal’s cooperation, the district court accepted the government’s recommendation for a two level downward departure and sentenced Ama-dor-Leal to 87 months in custody. Remaining charges were dismissed, and this timely appeal followed.
II
The government urges us to enforce Amador-Leal’s waiver of the right to appeal in the plea agreement, arguing that prosecutors — unlike the district court— have no procedurally — imposed duty to engage in a direct/collateral effects analysis under Rule 11(c) of the Federal Rules of Criminal Procedure. It reasons that when a plea agreement is entered into voluntarily (as here), there is nothing to suggest that the acceptance is invalid. But this does not follow, for whether or not the
agreement
is knowing and voluntary, acceptance
and entry of the plea
may not be if the plea is not taken in compliance with Rule 11. Accordingly, we may hear Amador-Leal’s appeal in order to determine whether his plea conforms to the Rule.
United States v. Portillo-Cano,
*514 III
Amador-Leal submits that his plea neither conforms to Rule 11 nor comports with due process because, for a plea to be considered voluntary, district courts must inform defendants pleading guilty of the direct consequences of their plea and resulting conviction; immigration consequences are direct as they flow automatically from a statute; and
United States v. Littlejohn,
To start with the obvious, guilty pleas must be knowing and voluntary. As we explained in Torrey v. Estelle:
A plea is voluntary only if it is entered by one fully aware of the direct consequences of his plea of guilty.... This court, in harmony with other circuits, has held that although a defendant is entitled to be informed of the direct consequences of the plea, the court need not advise him of all the possible collateral consequences.
As Amador-Leal recognizes, we decided in
Fruchtman
that immigration consequences are collateral. In
Fruchtman,
the defendant was an alien who also was not advised by the district court that his conviction would subject him to deportation proceedings. Fruchtman argued on appeal that his plea was not voluntary because deportation is a drastic measure of which he should have been advised before his plea was accepted. We recognized that administration of Rule 11 “requires the development of some limiting guide to define the nature of the consequences of which a defendant must be advised so that the requirements of the rule shall have been met,” and that the “common distinction” drawn is between consequences that are direct and those that are collateral.
Fruchtman,
We must, of course, follow Fruchtman unless it is distinguishable or undermined. First, relying on Littlejohn, Amador-Leal suggests that the proper inquiry for determining whether a statutorily created ineligibility to apply for some form of gov *515 ernment benefit is a direct or collateral consequence is not whether the benefit is denied later by some agency other than the court, but whether the ineligibility is an automatic consequence of a statute. However, Littlejohn did not purport to change the criterion for determining what is direct or collateral, and we decline to embrace the test that Amador-Leal proposes. No court has, and it would be totally unworkable in practice.
In Littlejohn, the defendant pled guilty to a drug offense. His conviction automatically rendered him ineligible for certain food stamp and social security benefits pursuant to 21 U.S.C. §§ 862(a) and 862a, but the court did not say so during the Rule 11 colloquy. 2 On appeal, Littlejohn argued that his plea was involuntary because he had not been warned that he would suffer ineligibility, and we agreed. We contrasted direct consequences, described, as in Torrey, as consequences that have G
“a definite, immediate and largely automatic effect on the range of the defendant’s punishment,” [with collateral consequences that] have included the possibility of a felony prosecution for reentry following deportation; imposition of a consecutive rather than concurrent sentence where the district court has discretion to choose between the two; the possibility of being resen-tenced to a maximum term if a state agency determines that the defendant is not amenable to treatment; exposure to potential civil tax litigation; revocation of parole from a separate convietion where such revocation is within the power of a parole board; and the potential of deportation, where a separate agency has authority over such deportation, Fruchtman v. Kenton.
Littlejohn,
Littlejohn itself puts the potential for deportation firmly within the category of collateral consequences. Indeed, our decision in Littlejohn turned on the rationale articulated by Fruchtman for determining whether a consequence is direct or collateral. As we summed up in Littlejohn, “where the consequence is contingent upon action taken by an individual or individuals other than the sentencing court — such as another governmental agency or the defendant himself — the consequence is generally *516 ‘collateral.’ ” Id. at 965. Thus, the Fruchtman test remains intact.
Nevertheless, Amador-Leal maintains,
Fruchtman
can no longer control in the context of deportation because AEDPA and IIRIRA have changed the landscape for illegal aliens convicted of aggravated felonies. No doubt the landscape has changed, because it is now virtually certain that an aggravated felon will be removed.
See
8 U.S.C. § 1228(a)-(b) (2001) (providing expedited removal proceedings for convicted aliens and eliminating agency discretion to forego removal); 8 U.S.C. § 1228(c) (2001) (adopting conclusive presumption of deportability of convicted aliens and denying judicial review of removal orders); 8 U.S.C. § 1229b (2001) (prohibiting INS from providing convicted aliens with cancellation of removal);
see also INS v. St. Cyr,
Amador-Leal points out that one court has expressed the view that deportation is no longer a consequence that will be determined by an agency outside of the trial court’s control.
United States v. El-Nobani,
What renders the plea’s immigration effects “collateral” is not that they arise “virtually by operation of law,” but the fact that deportation is “not the sentence of the court which aceept[s] the plea but of another agency over which the trial judge has no control and for which he has no responsibility.” However “automatically” [the defendant’s ] deportation — or administrative detention— might follow from his conviction, it remains beyond the control and responsibility of the district court in which that conviction was entered and it thus remains a collateral consequence thereof.
Gonzalez,
IIRIRA did not so substantially alter the treatment of individuals in [the defendant’s] situation as to warrant recon *517 sideration of whether deportation is still a collateral consequence of conviction. Even prior to IIRIRA’s passage, any alien who committed an aggravated felony was considered deportable, and the Attorney General was directed to apprehend any such alien upon his or her release from prison.
Id. at 26 (citation omitted). We agree with this analysis, and reaffirm that our decision in Fruchtman remains good law in this circuit as well.
Therefore, immigration consequences continue to be a collateral consequence of a plea and the resulting conviction. This means that district courts are not constitutionally
required
to warn defendants about potential removal in order to assure voluntariness of a plea; but it does not mean that they
should
not do so. Many district judges comment in their Rule 11 colloquy that a plea of guilty and resulting conviction may affect an alien’s status in this country, and inquire whether the defendant understands the possible immigration consequences of his plea. Although not required by Rule 11 or due process, we commend this sort of dialogue for there is no question that immigration consequences of a conviction are important to aliens contemplating a plea.
See St. Cyr,
AFFIRMED.
Notes
.
United States v. Gonzalez,
. The issue came to light in the Pre-Sentence Report, which included a section on "Denial of Federal Benefits” referring to § 862(a) and its implementing guideline, U.S.S.G. § 5F1.6. As we explained in the opinion, there are actually two statutes with similar numbers that have essentially the same effect, § 862(a) and § 862a.
. The same feasibility concerns that animated this different treatment in Littlejohn are also present here: district courts generally have no information about a defendant’s immigration status at the Rule 11 proceeding.
. Many states have enacted
statutes
that require courts to advise a defendant of the immigration consequences of a plea,
see St. Cyr,