BROWN

Appeal from the United States District Court for the District of Columbia (No. 1:12–cv–00498).
Matthew A. Seligman, appointed by the court, argued the cause and filed the briefs for appellant. Jane M. Lyons, Assistant U.S. Attorney, argued the cause for appellees. With her on the brief were Ronald C. Machen, Jr., U.S. Attorney, and R. Craig Lawrence, Assistant U.S. Attorney.
Before: HENDERSON, ROGERS and BROWN, Circuit Judges.

Opinion for the court filed by Circuit Judge BROWN.
Opinion concurring in part and dissenting in part filed by Circuit Judge ROGERS.

BROWN, Circuit Judge:

Appellant challenges the United States Parole Commission's (USPC) denial of his 2010 and 2012 applications for parole. In particular, he asserts the USPC violated the Constitution's prohibition on ex post facto laws, U.S. CONST. art. I, § 9, cl. 3, by incorrectly applying the regulations in place at the time of appellant's underlying offense. The district court dismissed appellant's complaint for failure to state a claim. On review, we find that the USPC's denial of appellant's requests for parole was a valid exercise of parole authority as it existed at the time of his offense. In addition, the USPC did not rely on the retroactive application of any law, regulation, or guideline to justify its decisions, and therefore could not have violated the Ex Post Facto Clause. See Fletcher v. District of Columbia (Fletcher II), 391 F.3d 250, 251 (D.C.Cir.2004). Accordingly, we affirm the judgment of the court below.

I.

Ari Bailey is currently serving a fifteen- to forty-five-year sentence for a rape he committed in December 1993. In 2004, after Bailey had served ten years of his sentence, he became eligible for parole. After an initial parole hearing before the USPC in September 2004, Bailey was denied parole. In 2007, 2010, and 2012, Bailey again applied for parole.

After rehearings, the USPC denied each of Bailey's applications.

Between the time Bailey committed his crime and the time he became eligible for parole, the law governing parole for individuals convicted of criminal violations of the D.C.Code underwent several changes. In 1993, at the time of Bailey's offense, the D.C. Parole Board (“Board”) made parole determinations for D.C. offenders. D.C. CODE §§ 24–201.1—201.3 (1989), superseded by § 24–131 (2001). The Board exercised its authority pursuant to section 24–204 of the D.C. Code, which provided:

Whenever it shall appear to the Board of Parole that there is a reasonable probability that a prisoner will live and remain at liberty without violating the law, that his release is not incompatible with the welfare of society, and that he has served the minimum sentence imposed or the prescribed portion of his sentence, as the case may be, the Board may authorize his release on parole upon such terms and conditions as the Board shall from time to time prescribe.
Id. § 24–204(a) (1989), superseded by § 24–404(a) (2009).

In 1987, the Board promulgated guidelines to govern its evaluation of a prisoner's suitability for parole. SeeD.C. MUN. REGS. tit. 28, §§ 100, et seq. (1987) (“1987 Guidelines”), superseded by28 C.F.R. §§ 2.70, et seq. (“2000 Guidelines”). The 1987 Guidelines created a point system focused on offender history, offense characteristics, and behavior while in prison. The resulting point total determined whether parole would be granted. Id. § 204.19. However, the Guidelines also allowed the Board to override the point-based determination in “unusual circumstances.” Id. § 204.22. See Daniel v. Fulwood, 766 F.3d 57, 59 (D.C.Cir.2014). In 1991, in an effort to “facilitate consistency in Guideline application,” the Board also issued an unpublished policy guideline that provided definitions of criteria, parameters, and terms used in the 1987 Guidelines. Policy Guideline, D.C. Board of Parole (Dec. 16, 1991) (“1991 Policy Guideline”).

In 1997, Congress abolished the Board and directed the USPC to conduct parole hearings for D.C. offenders. National Capital Revitalization and Self–Government Improvement Act, Pub.L. No. 105–33, § 11231(a)-(c), 111 Stat. 712, 745 (1997), codified atD.C. CODE § 24–131 (2001). Like the Board it replaced, the USPC was given authority to grant parole “where there is a reasonable probability that a prisoner will live and remain at liberty without violating the law, [and where] ... his or her release is not incompatible with the welfare of society.” D.C. CODE § 24–404 (2009). In 2000, the USPC promulgated its own parole guidelines, the 2000 Guidelines, which initially applied to all D.C. offenders who became eligible for parole on or after August 5, 1998. 28 C.F.R. §§ 2.70, et seq.

In Fletcher v. Reilly (Fletcher III), 433 F.3d 867 (D.C.Cir.2006), this Court recognized that the 1987 Guidelines and the 2000 Guidelines were “substantially different.” Id. at 877–78. As a result, the Court warned, retroactive application of the 2000 guidelines could give rise to a violation of the Ex Post Facto Clause. Id. at 878–79. Subsequently, in Sellmon v. Reilly, 551 F.Supp.2d 66 (D.D.C.2008), the district court ruled in favor of four prisoner-plaintiffs who argued they “faced a significantly increased risk of lengthier incarceration due to the [retroactive application of the] 2000 Guidelines.” Id. at 91. The district court therefore ordered the USPC to reevaluate the prisoner-plaintiffs' parole applications under the 1987 Guidelines. Id. at 99.

In light of these rulings and others, the USPC promulgated a new rule—sometimes referred to as the Sellmon Rule—to address retroactive applications of the 2000 Guidelines. 28 C.F.R. § 2.80( o ). Under the Sellmon Rule, the USPC applies the 1987 Guidelines when reviewing parole applications filed by a D.C. offender who committed his offense between March 4, 1985 and August 4, 1998. Id.; see also, e.g., Taylor v. Reilly, 685 F.3d 1110, 1112 (D.C.Cir.2012). Accordingly, the 1987 Guidelines governed appellant's 2010 and 2012 parole rehearings—the two rehearings at issue in this case. See March 1, 2010 Notice of Action, J.A. 74; March 19, 2012 Notice of Action, J.A. 79.

On March 1, 2010, the USPC informed Bailey that his request for parole was denied. As the Commission explained:

The Commission has applied the D.C. Board of Parole's 1987 guidelines to ... your case. You have a total point score of 2 under the guidelines for D.C. offenders. The guidelines indicate that parole should be granted at this time. However, a departure from the guidelines at this consideration [sic] is found warranted because the Commission finds there is a reasonable probability that you would not obey the law if released and your release would endanger the public safety.
March 1, 2010 Notice of Action, J.A. 74. Specifically, the Commission provided that its decision was based on the fact that (1) Bailey had “not completed any programs that address the underlying cause of [his] criminal conduct of rape;” (2) he “continued to deny the offense conduct;” (3) he “never expressed an interest in participating in relevant programming to address [his] criminal conduct;” (4) in the two prior years he had “completed no other rehabilitative programs that would indicate [his] risk to the community has been lessened;” and (5) he “continued to incur incident reports for threatening and assaultive conduct.” Id.

In 2012, after another rehearing, the USPC again denied appellant parole. March 19, 2012 Notice of Action, J.A. 79. As in 2010, the Commission concluded there was “a reasonable probability [he] would not obey the law if released and [his] release would endanger public safety.” Id. The Commission explained its denial was based on the fact that (1) Bailey had “not completed any programs that address the underlying cause of [his] criminal conduct of rape;” (2) at the time he committed rape in DC “there was an outstanding warrant for [his] arrest on another rape [charge] in Baltimore, Maryland;” and (3) he had “been confined to a closed prison setting in the past two years based on [his] prior institution misconduct” and had not “continued significant programming since that time.” Id.

On March 30, 2012, appellant filed a complaint arguing the 2010 and 2012 parole decisions violated his rights under the Ex Post Facto Clause. On May 20, 2013, the district court granted the government's motion to dismiss after concluding that “[t]here is no ex post facto violation where, as here, the USPC applied the regulations that were in effect at the time the plaintiff committed the underlying criminal offense.” Bailey v. Fulwood (Bailey I), 945 F.Supp.2d 62, 63 (D.D.C.2013). Thereafter, Bailey filed a timely notice of appeal.

II.

Appellant contends the USPC “violated the Ex Post Facto Clause of the Constitution ... by denying Mr. Bailey parole on the basis of factors that were impermissible under the Board's 1987 Guidelines and 1991 Policy Guideline but are permissible under the Commission's [2000] Guidelines.” Opening Brief of Court-appointed Amicus Curiae in Support of Appellant at 15. This argument fails in two respects. First, the USPC's decisions were a permissible exercise of its statutory discretion, which was cabined neither by the 1987 Guidelines nor by the 1991 Policy Guideline. Second, a violation of the Ex Post Facto Clause requires retroactive application of a law, regulation, or guideline. Here, the USPC did not base its denial on an application of the 2000 Guidelines. Rather, it explicitly relied on the 1987 Guidelines as the basis for its actions. Accordingly, the USPC did not violate the Ex Post Facto Clause.

A.

It is clearly established under D.C. law that the factors set forth in the 1987 Regulations and the definitions articulated in the 1991 Policy Guideline never constrained the discretion of the Board or the USPC. As the D.C. Court of Appeals explained in McRae v. Hyman, 667 A.2d 1356 (D.C.1995), our analysis of this issue must begin with the governing statute, id. at 1359, which provides that “the Board may authorize [a prisoner's] release on parole” if it determines “there is a reasonable probability that a prisoner will live and remain at liberty without violating the law ... [or endangering] the welfare of society.” D.C.Code § 24–204(a) (1989) (emphasis added). As the Court observed in McRae, this statute is phrased “in discretionary terms.” 667 A.2d at 1360. In turn, the 1987 Guidelines “incorporate this discretionary approach,” id. at 1359, and set forth a system “to guide the Board in making the decision whether to grant or deny the parole”—not to limit it, id. at 1360 (emphasis added). Where the Board exercises its discretion to depart from this numerical system, it may do so as long as it “specif[ies] in writing those factors which it used. Departures must be explained, but they are not proscribed.” Id. (emphasis added). As the McRae Court concluded, the Board need not render a decision based on a strict application of the system set forth in the 1987 Regulations. Rather, it must simply adhere to “the words of the governing statute, § 24–204(a), [and determine whether a prisoner is able to] live and remain at liberty without violating the law such that release would be compatible with the welfare of society.” Id. at 1361; see also id. at 1360–61 (holding the Board “is not required to either grant or deny parole based upon the score attained” and it may “ignore the results of the scoring system and either grant or deny parole in the individual case” so long as it specifies the reasons in writing).

The holding in McRae does not stand alone. In two prior cases, the D.C. Court of Appeals had already explained the broad discretion retained by the Board under the 1987 Regulations. Davis v. Henderson, 652 A.2d 634 (D.C.1995); White v. Hyman, 647 A.2d 1175, 1180 (D.C.1994) (“[T]he statute and Regulation vest in the Board substantial discretion in granting or denying parole.”). In Davis, the Court compared the Board's discretion under the 1987 Regulations to its extensive discretion under the 1980 Regulations prior to the introduction of the formalized scoring system and concluded “[t]he discretion conferred by the 1980 guidelines ... survived in the 1987 revisions.” 652 A.2d at 635. Read together, these cases compel the conclusion that the 1987 Guidelines did not diminish the broad discretion to deny parole afforded to the Board under section 24–204(a) of the D.C.Code.

This conclusion is further bolstered by our own “independent review of the regulations” in Ellis v. District of Columbia, which concluded that even section 204.1 of the 1987 Guidelines does not constrain the Board's discretion. 84 F.3d 1413, 1419 (D.C.Cir.1996). Section 204.1 provides: “Any parole release decision falling outside the numerically determined guideline shall be explained by reference to the specific aggravating or mitigating factors as stated in Appendices 2–1 and 2–2.” 1987 Guidelines, § 204.1. Among the factors listed in “Appendices 2–1 and 2–2” are “Other,” id. app. 2–1, and “Other change in circumstances,” id., app. 2–2. As this court explained, the inclusion of such “scarcely constraining language” suggests that the “Board, in the exercise of its discretion,” may deny a prisoner parole if it “believes there is some ... reason for not granting him parole.” Ellis, 84 F.3d at 1419; see also id. (“Under § 204.22 of the [1987 Guidelines], if the Board wished to disregard the results of the scoring system it merely had to say so in writing.”) (emphasis added); Phillips v. Fulwood, 616 F.3d 577, 582 (D.C.Cir.2010) (“The 2000 regulations permit the Commission, in ‘unusual circumstances,’ to depart upward based on a prisoner's risk to society. But so too did the 1987 regulations.”). To emphasize the extent of this discretion, we held that under D.C. law, “parole is never required ”—even where “the Board determines that [a prisoner meets] the necessary prerequisites” for release. Ellis, 84 F.3d at 1420. Thus, “a prisoner's low total point score [under the 1987 Regulations] does not compel the Board to grant a prisoner release.” Id.

To be sure, as Judge Tatel noted in his concurrence in Ellis, one could read the 1987 Regulations differently. Ellis, 84 F.3d at 1427–28 (Tatel, J., concurring in part). Specifically, we could interpret the 1987 Guidelines as constraining the Board's discretion by requiring it to “apply[ ] a set of specified standards” in denying parole to prisoners with qualifying scores. Id. at 1428. However, as Judge Tatel ultimately concluded, in light of the rulings of the D.C. Court of Appeals, we may not adopt such an interpretation. Id. at 1429; see also id. at 1420 (“Although we are not bound by the D.C. Court of Appeals's interpretation of the Constitution, we must respect its construction of D.C. law.”). Accordingly, the 1987 Guidelines did not constrain the Board's discretion, nor do they now constrain the USPC's.

Similarly, the 1991 Policy Guideline does not limit the USPC's discretion. First, though none of the opinions mentioned expressly discuss the 1991 Policy Guideline, such an omission is telling.

For judges, a significant change to a legal regime is like an alligator in one's bathtub; it cannot prudently be ignored. If the 1991 Policy Guideline had altered the scope of the Board's discretion, the judicial response would not have been silence. Second, in the context of parole set-offs, the D.C. Court of Appeals has explained “[t]he [1991 Policy] Guideline[ ] undoubtedly reflect[s], rather than limit[s], the discretion that the District's municipal regulations and parole statute already vest in the Board's set-off determinations.” Hall v. Henderson, 672 A.2d 1047, 1053 (D.C.1996). There is no reason to believe this conclusion, for which the Court cites both White and McRae as authority, does not apply with the same force to parole determinations themselves. Id. at 1054. Finally, it would be somewhat unusual to deem that the unpublished 1991 Policy Guideline trumps a statutory enactment or published regulation. Cf. San Luis Obispo Mothers for Peace v. U.S. Nuclear Regulatory Comm'n, 789 F.2d 26, 33 (D.C.Cir.1986) (“To accept petitioners' argument, therefore, we would have to hold that NUREG–0654, a staff document intended as guidance, supersedes the regulation itself. The only virtue of that approach is novelty.”); Edward v. D.C. Taxicab Comm'n, 645 A.2d 600, 603 n. 10 (D.C.1994) (“This court has often applied federal administrative law principles to local administrative agency proceedings.”).

Accordingly, we are bound to conclude that the 1987 Guidelines and the 1991 Policy Guideline do not constrain the discretion of the USPC. Therefore, the USPC did not violate either of them when it denied Bailey parole after finding “a reasonable probability that [he] would not obey the law if released and [that his] release would endanger the public safety.” March 1, 2010 Notice of Action, J.A. 74; March 19, 2012 Notice of Action, J.A. 79.

B.

1.

Even if the USPC had failed in its effort properly to apply the 1987 Regulations and 1991 Policy Guideline, such a mistake could not be the basis for a claim under the Ex Post Facto Clause. Indeed, the government did not rely on the retroactive application of any statute, regulation, or guideline to justify its denial of Bailey's requests for parole.

As the Supreme Court explained in Garner v. Jones, 529 U.S. 244, 120 S.Ct. 1362, 146 L.Ed.2d 236 (2000), “the Ex Post Facto Clause ... bar [s] enactments which, by retroactive operation, increase the punishment for a crime after its commission ... [and] [r]etroactive changes in laws governing parole or prisoners, in some instances, may be violative of this precept.” Id. at 249–50, 120 S.Ct. 1362 (emphasis added). Admittedly, such “[r]etroactive changes in laws” can include changes to regulations and even guidelines governing the parole process. Id. at 250, 252, 120 S.Ct. 1362; see also Fletcher II, 391 F.3d at 251 (“The Supreme Court [in Garner ] thus foreclosed our categorical distinction between a measure with the force of law and guidelines [that] are merely policy statements from which the Commission may depart in its discretion.”). However, a necessary feature of any ex post facto claim is a rule to which the government seeks to give retroactive effect. See, e.g., Garner, 529 U.S. at 252, 120 S.Ct. 1362 (“The case turns on the [retroactive] operation of the amendment to Rule 475–3–.05–2.”); Daniel, 766 F.3d at 61 (“In order to prevail on the merits of an ex post facto claim with regard to parole guidelines, a plaintiff must show that retroactive application of new guidelines creates a significant risk of prolonging his incarceration as compared to application of the prior guidelines.”); Phillips, 616 F.3d at 578 (“Although he has been eligible for parole since 2003, the United States Parole Commission—applying regulations it issued in 2000—has repeatedly found him unsuitable for release. Phillips contends that the Commission should have applied the parole rules that were in effect when he committed his crimes, and that its failure to do so violated the Ex Post Facto Clause of the Constitution.”); Fletcher III, 433 F.3d at 876 (“[Fletcher] contends that the Commission's retroactive application of the new federal reparole regulations, rather than the Board's regulation, during his 2000 reparole hearing, creates a significant risk of increased punishment, and is thus an unconstitutional ex post facto law.”); see also Taylor, 685 F.3d at 1112–13 (noting plaintiff did not contest his claim for injunctive and declaratory relief under the Ex Post Facto Clause became moot when the USPC agreed to grant him a new parole hearing and apply the 1987 Guidelines). Here, the USPC justified its decisions without invoking any retroactive laws.

2.

Appellant's court-appointed amicus calls into question this bedrock principle of Ex Post Facto Clause jurisprudence—that the clause only applies where the government seeks to give retroactive effect to a legal rule. Citing Sellmon, amicus argues that “[d]efendants may not avoid a constitutional challenge simply by citing the correct rules, while in fact following a federal practice that is inapplicable to [Mr. Bailey].” Opening Brief of Court-appointed Amicus Curiae in Support of Appellant at 34–35 (quoting Sellmon, 551 F.Supp.2d at 96) (second alteration in original). According to amicus, “[a]llowing the Commission to evade the requirements of the Ex Post Facto Clause merely by asserting that it applied the 1987 Guidelines while its analysis in practice applied the 2000 Guidelines would eviscerate the protections afforded by the Clause.” Id. at 35. However, despite amicus' arguments to the contrary, as the district court observed, the Ex Post Facto Clause asks whether the USPC “applied the correct parole guidelines” and not whether the USPC correctly applied the parole guidelines. Bailey I, 945 F.Supp.2d at 64.

The Ex Post Facto Clause constrains the government's ability to use retroactive legal rules to justify criminal punishment. Where, as here, a prisoner believes the USPC has mis-applied a prospective legal rule, the Clause simply does not apply. Of course, this is not to say the prisoner has no legal recourse to challenge an alleged mis-application of law. Prisoners have avenues to challenge the unlawful denial of a request for parole—even where this denial does not violate the Ex Post Facto Clause.

See, e.g., Furnari v. U.S. Parole Comm'n, 531 F.3d 241, 247–48 (3d Cir.2008) (holding that in reviewing a USPC decision to deny parole on a petition for a writ of habeas corpus, a federal court inquires into “whether there is a rational basis in the record for the [Parole Commission's] conclusions embodied in its statement of reasons”); see also Doe v. U.S. Parole Comm'n, 602 Fed.Appx. 530 (D.C.Cir.2015) (unpublished per curiam) (discussing various causes of action an offender may assert to challenge USPC decisions). In fact, Bailey himself has already made use of one such avenue by bringing a habeas action in the Middle District of Pennsylvania.

See Bailey v. Fulwood, Civil No. 3:CV–11–435, 2012 WL 5928302 (M.D.Pa. Nov. 26, 2012).

To extend the Ex Post Facto Clause to cases like the one at bar would be deleterious to the proper functioning of the criminal justice system. The Clause strikes a careful balance; it prohibits retroactive application of parole regulations that “create[ ] a significant risk of prolonging [an inmate's] incarceration,” Garner, 529 U.S. at 251, 120 S.Ct. 1362, while preserving for parole boards some flexibility in the way they exercise their discretion prospectively. As the Garner Court explained:

The danger that legislatures might disfavor certain persons after the fact is present even in the parole context, and the Court has stated that the Ex Post Facto Clause guards against such abuse. On the other hand, to the extent there inheres in ex post facto doctrine some idea of actual or constructive notice to the criminal before commission of the offense of the penalty for the transgression we can say with some assurance that where parole is concerned discretion, by its very definition, is subject to changes in the manner in which it is informed and then exercised. The idea of discretion is that it has the capacity, and the obligation, to change and adapt based on experience. New insights into the accuracy of predictions about the offense and the risk of recidivism consequent upon the offender's release, along with a complex of other factors, will inform parole decisions.
Id. at 253, 120 S.Ct. 1362.

The rule suggested by the district court in Sellmon and advocated by appellant would subject any change to a parole board's exercise of discretion to constitutional inquiry—even where the change is explicitly made prospective. Any inmate denied parole could point to a prospective policy that the board did not rely on in justifying its decision and argue that this policy violated his right to be free from “[r]etroactive changes in laws.” Id. at 250, 120 S.Ct. 1362. Parole boards would have no choice but to “freeze in time” their discretion to insulate themselves from such absurd challenges. Id. at 259, 120 S.Ct. 1362 (Scalia, J., concurring). In essence, courts would deprive parole boards of “the capacity, and the obligation, to change and adapt based on experience.” Id. at 253, 120 S.Ct. 1362 (majority opinion). We decline to adopt such a rule.

3.

Here, the USPC did not rely on the 2000 Guidelines to justify its 2010 and 2012 parole decisions. Rather, “[i]t is apparent from the Court's review of the record that the USPC applied the Parole Board's 1987 Regulations—not the 2000 Guidelines—in 2010 and again in 2012.” Bailey I, 945 F.Supp.2d at 63. As the district court aptly noted, “[t]here is no [ Ex Post Facto Clause] violation where, as here, the USPC applied the regulations which were in effect at the time the plaintiff committed the underlying criminal offense.” Id.

Accordingly, the judgment of the court below is

Affirmed.

ROGERS, Circuit Judge, concurring in part and dissenting in part.

This appeal and a related appeal

arise as a result of Congress's transfer of parole responsibilities from the D.C. Board of Parole to the U.S. Parole Commission in 1998. See National Capital Revitalization and Self–Government Improvement Act, Pub.L. No. 105–33, § 11231(a)-(c), 111 Stat. 712, 745 (1997), codified atD.C.Code § 24–131. These plaintiffs, who sued under 42 U.S.C. § 1983, were convicted when the D.C. Board was still in charge of parole decisions and they contend on appeal that the U.S. Commission is carrying out stricter parole practices than did the D.C. Board. Bailey contends, adopting the brief of amicus curiae, that the U.S. Commission violated the Ex Post Facto Clause of the Fifth Amendment to the Constitution by denying him parole based on factors that were impermissible under the D.C. Board's policy in place at the time of his offense, but permissible under the U.S. Commission's subsequent policy. Because the U.S. Commission properly applied the D.C. Board's earlier policy, I concur in holding that Bailey's ex post facto challenge fails. Because that holding disposes of the issue Bailey has presented, I dissent from the court's alternative analysis prematurely announcing a broader ex post facto principle for future cases.

I.

In interpreting the authority of the D.C. Board of Parole, the District of Columbia Court of Appeals has held that when acting pursuant to the 1987 Regulations, the Board retains discretion under D.C.Code § 24–204(a) to depart from numerical recommendations set forth in the regulations. See McRae v. Hyman, 667 A.2d 1356, 1359–61 (D.C.1995); Davis v. Henderson, 652 A.2d 634, 635–38 (D.C.1995); White v. Hyman, 647 A.2d 1175, 1180 (D.C.1994). This court has adopted the D.C. Court of Appeals's interpretation of D.C.Code § 24–204(a) and the parole system established in the D.C. regulations. See Phillips v. Fulwood, 616 F.3d 577, 582 (D.C.Cir.2010); Ellis v. Dist. of Columbia, 84 F.3d 1413, 1420 (D.C.Cir.1996). Although the cited D.C. cases either do not discuss the Board's 1991 Guideline or do so as to other issues, the Guideline merely provided definitions for certain terms in the 1987 Regulations and did not purport to limit the D.C. Board's discretion under the 1987 Regulations.

Therefore, essentially for the reasons stated by the court, I join the court in holding that because “the [U.S. Commission]'s decisions were a permissible exercise of its statutory discretion, which was cabined neither by the 1987 Guidelines nor by the 1991 Policy Guideline,” Op. 131–32, Bailey's ex post facto challenge fails.

II.

The court insists on going further, concluding alternatively that a prisoner can never present a claim under the Ex Post Facto Clause where a parole agency cites the correct regulation or guidelines. See Op. 132, 134–37. The court normally “do[es] not reach out to decide” constitutional issues when the appeal does not require it. Pub. Citizen Health Research Grp. v. Tyson, 796 F.2d 1479, 1507 (D.C.Cir.1986). The court's reason for doing so today is to express its view that Sellmon v. Reilly, 551 F.Supp.2d 66, 96 (D.D.C.2008), misstates the law. Op. 135, 136–37. In Sellmon, the district court was “unpersuaded that the [Commission's] reference to the 1987 Regulations, standing alone, is sufficient to bar an ex post facto challenge.” 551 F.Supp.2d at 96 (emphasis added). This court's categorical rejection of Sellmon goes far beyond the facts of Bailey's case. In denying parole, the Commission did not simply reference the 1987 Regulations, it correctly applied them. That conclusion fully resolves the issue Bailey has presented in this appeal.

The court's alternative conclusion is troubling for two reasons. First, it may not be correct, and the factual record and briefing in this case did not focus on the issue so as to allow for careful consideration. Second, the policy considerations on which the court relies are dubious at best. The more prudent course would be to leave consideration of this issue for a case that actually presented it.

1. The Ex Post Facto Clause prohibits the passing of an “ex post facto Law.” U.S. CONST. art. I, § 9, cl. 3. In the parole context, this prohibition applies to new regulations and policy guidelines that “create[ ] a significant risk of prolonging [an inmate's] incarceration.” Garner v. Jones, 529 U.S. 244, 251, 120 S.Ct. 1362, 146 L.Ed.2d 236 (2000); see Phillips, 616 F.3d at 580. The analysis involves two questions: did the parole agency apply a retroactive policy (the retroactivity question), and, if so, did the retroactive application carry a significant risk of increased punishment (the risk question). Answering the retroactivity question will generally be easy when the parole agency states that it is applying a later policy. In such cases, the court will only have to answer the risk question, which requires the court to conduct “a searching comparison of the old and new” policies to determine whether application of the later policy “create[s] a significant risk” of lengthier incarceration. Fletcher v. Reilly, 433 F.3d 867, 879 (D.C.Cir.2006) (“ Fletcher III ”). The court's alternative analysis speaks to the opposite situation, in which the parole agency denies parole using an analysis available under the new policy but not the old (thereby demonstrating a “significant risk” that applying the new policy will prolong incarceration) but claims to be applying the prospective policy only. That situation requires the court to answer the first question: which policy did the parole agency, in fact, apply?

In addressing the risk question, the Supreme Court's ex post facto precedent has eschewed formalism. It has instructed courts to look at an agency's “policy statements, along with the [agency's] actual practices,” to determine “the manner in which it is exercising its discretion” under both policies, looking to “evidence drawn from the rule's practical implementation.” Garner, 529 U.S. at 256, 120 S.Ct. 1362; see id. at 255, 120 S.Ct. 1362. “The controlling inquiry under Garner is how the Board or the Commission exercises discretion in practice, and whether differences between the exercise of discretion in [the] two systems actually” create a significant risk of increased punishment. Fletcher III, 433 F.3d at 876–77. “[T]he question is one of practical effect,” not labels. Fletcher v. Dist. of Columbia, 391 F.3d 250, 251 (D.C.Cir.2004) (“ Fletcher II ”). Thus, even if a later policy is facially similar to the one in place at the time of offense, courts are to scrutinize whether, in practice, the later policy reflects a stricter implementation of statutory discretion. In Bailey's case, the court has effectively answered that question in the negative, holding that, as exercised, the discretion is the same under both, and consequently there is no difference in the risk of punishment under the two regimes. See also Phillips, 616 F.3d at 582–83.

The court proceeds to answer the retroactivity question as well. With virtually no analysis of the Ex Post Facto Clause itself, the court opines that the parole agency's characterization of its action is conclusive on the question of which policy was, in fact, applied. Thus, in the language of Sellmon, a simple “reference” to the correct policy, “standing alone,” is sufficient to bar an[y] ex post facto challenge. 551 F.Supp.2d at 96 (emphasis added). But if, as the Supreme Court has instructed, the risk question must take account of actual practice, then it is not clear why the retroactivity question categorically may not. The Ex Post Facto Clause prohibits “[e]very law that changes the punishment, and inflicts a greater punishment, than the law annexed to the crime, when committed.” Calder v. Bull, 3 U.S. (3 Dall.) 386, 390, 1 L.Ed. 648 (1798) (Chase, C.J.); see Garner, 529 U.S. at 249–50, 120 S.Ct. 1362. It “forbids the imposition of punishment more severe than the punishment assigned by law when the act to be punished occurred.” Weaver v. Graham, 450 U.S. 24, 30, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981) (emphasis added). When a parole agency prolongs incarceration using policy considerations that were not adopted until after the offense, it contravenes that prohibition regardless of the label it affixes to its decision. As the Supreme Court has explained, the Clause “safeguards a fundamental fairness interest in having the government abide by the rules of law it establishes to govern the circumstances under which it can deprive a person of his or her liberty or life.” Peugh v. United States, –––U.S. ––––, 133 S.Ct. 2072, 2085, 186 L.Ed.2d 84 (2013) (quotation marks and alterations omitted). That purpose is not satisfied by a bare disavowal of any ex post facto violation.

Imagine a sentencing judge who considers aggravating factors enacted after the defendant's offense, but concludes by stating “I applied the earlier sentencing law.” Or imagine a parole agency that changes the factors it will consider from A, B, C to X, Y, Z, then denies parole based only on the latter set of factors, but states that it applied the earlier policy. In those situations, it is unclear why mere averral of compliance with the Ex Post Facto Clause should override clear facts of the case to the contrary. The Clause “deals with substance, not shadows. Its inhibition was levelled at the thing, not the name. It intended that the rights of the citizen should be secure against deprivation for past conduct by legislative enactment, under any form, however disguised.” Weaver, 450 U.S. at 31 n. 15, 101 S.Ct. 960 (quoting Cummings v. Missouri, 71 U.S. (4 Wall.) 277, 325, 18 L.Ed. 356 (1866)). If the risk question requires an examination of “actual practices” and “practical effect,” why should the retroactivity question be blind to practical realities altogether? The court provides no answer. With no need to reach the question, the court relegates a core constitutional protection to an easily evaded formalism.

Granted, not every misapplication of a parole policy constitutes an ex post facto violation. Plaintiffs in the companion case, supra note 1, maintained that the improper application of a permissible factor amounted to an “unwritten” policy change, and thereby violated the Ex Post Facto Clause. An incorrect application of a prospective policy is not necessarily an ex post facto violation. But neither is the prospective policy necessarily the one that was applied, for purposes of the Clause, merely because of a statement to that effect by the parole agency. In some cases (and again, neither Bailey's case nor the companion case require this court to decide which ones), the facts may be sufficiently clear to establish that a later rule was applied, despite an agency's contrary statement.

2. How, then, does the court reach its conclusion that a bare assertion cures any possible ex post facto violation? Two policy reasons. First, the court suggests that the Clause is an unnecessary prophylactic because prisoners have other “avenues to challenge the unlawful denial of a request for parole.” Op. 135. The availability of other legal theories or causes of action is irrelevant to the meaning of the Ex Post Facto Clause. The Clause exists to prevent retroactive punishment, period. See Peugh, 133 S.Ct. at 2081 (quoting Calder, 3 U.S. (3 Dall.) at 390). It does not exist merely to prevent retroactive punishment that is not prevented by other means. Would the majority hold that the Fourth Amendment does not prohibit unreasonable searches and seizures that can be challenged under state tort law? Moreover, the court's premise may be wrong, as the court's citations cast doubt on the availability of other remedies. See Op. 135–36 (citing Doe v. U.S. Parole Comm'n, 602 Fed.Appx. 530 (D.C.Cir.2015), where the court highlights the barriers to bringing a direct appeal, federal habeas, state habeas, and claims under 42 U.S.C. § 1983 and the Administrative Procedure Act). Even if other legal theories are available, the scope of relief may depend on whether or not a claim is constitutional. See, e.g., 42 U.S.C. §§ 1983, 1988(b).

Second, the court is concerned that enforcing the Ex Post Facto Clause despite absolving boilerplate might “freeze in time” the Board's discretion. Op. 136–37 (quoting Garner, 529 U.S. at 259, 120 S.Ct. 1362 (Scalia, J., concurring in part in the judgment)). It is not clear what this means here. Surely the court does not mean to suggest that a parole agency is free to apply a later policy when doing so significantly increases the risk of incarceration. The Supreme Court and this court have held that parole agencies cannot retroactively apply stricter regulations or guidelines, even when their statutory discretion is unchanged. See Peugh, 133 S.Ct. at 2081, 2086; Garner, 529 U.S. at 253–55, 120 S.Ct. 1362; Phillips, 616 F.3d at 580; Fletcher III, 433 F.3d at 876–77; Fletcher II, 391 F.3d at 251. In fact, the “freezing” concern comes from a separate opinion criticizing that conclusion in Garner, see Garner, 529 U.S. at 257–59, 120 S.Ct. 1362 (Scalia, J., concurring in part in the judgment); but as the court acknowledges, see Op. 134–35, that bridge has been crossed. Perhaps the court's “freezing” concern is targeted at the “unwritten” policy theory advanced in the companion case. But rejecting that theory does not require the court's broad alternative analysis. Where a parole agency applies a later policy retrospectively, no concern for “freezing” the agency's discretion can overcome the Supreme Court's holding in Garner and our cases applying it.

The court has no reason to make new ex post facto law on such a weak foundation. As a result of our holding that the D.C. 1987 Regulations and 1991 Guideline do not constrain the U.S. Commission's discretion, no prisoner could reasonably anticipate success in filing an action like Bailey's because it will be clear that the U.S. Commission's exercise of reasoned discretion is not constrained under the earlier D.C. policy.

The court's alternative analysis thus bears all the worst hallmarks of an advisory141opinion. See United States v. Fruehauf, 365 U.S. 146, 157, 81 S.Ct. 547, 5 L.Ed.2d 476 (1961); Alabama State Fed'n of Labor v. McAdory, 325 U.S. 450, 461–62, 65 S.Ct. 1384, 89 L.Ed. 1725 (1945) (collecting cases). Lacking the focus that an actual controversy presents, the court adopts an ill-considered position with implications well beyond the facts of Bailey's case and possibly beyond the parole context altogether. The parties have barely briefed this issue, instead treating the ex post facto question as turning on whether the factors on which the U.S. Commission relied were permissible under the D.C. Parole Board's 1987 Regulations and 1991 Guideline. See McBride v. Merrell Dow & Pharm., Inc., 800 F.2d 1208, 1211 (D.C.Cir.1986). The court answered that question in the affirmative, and I concur. Because the court's alternative analysis is unnecessary (and unpersuasive), I respectfully dissent from Part II.B of the court's opinion (and the sentences introducing it on page 7) and would leave the operation of the retroactivity question for another day.