dissenting.
I respectfully dissent.
The Supreme Court has made it clear that changes affecting parole decisions are within the purview of the ex post facto clause. In finding that the “no parole” provision of a statute was part of the offender’s punishment, the Court noted in Warden v. Marrero, 417 U.S. 653, 94 S.Ct. 2532, 41 L.Ed.2d 383 (1974), that “only an unusual prisoner could be expected to think that he was not suffering a penalty when he was denied eligibility for parole” and that “repealer of parole eligibility previously available to imprisoned offenders would clearly present the serious question under the ex post facto clause ... of whether it imposed a ‘greater or more severe punishment than was prescribed by law at the time of the ... offense.’ ” Id. at 662-63, 94 S.Ct. 2538 (quoting Rooney v. North Dakota, 196 U.S. 319, 325, 25 S.Ct. 264, 265-66, 49 L.Ed. 494 (1905) (emphasis and omission in Warden)). That changes in parole eligibility can violate the ex post facto clause is also clear in Weaver v. Graham, 450 U.S. 24,101 S.Ct. 960, 67 L.Ed.2d 17 (1981). The issue in Weaver was whether the application of a statute altering the method for computing “gain time” for good conduct to an inmate who committed a crime before enactment of the change violated the ex post facto clause. The Court first determined that regardless of whether the prospect of gain time was in some technical sense part of the offender’s sentence, the statute substantially altered the consequences attached to a crime already completed, changing the quantum of punishment, and thus could be constitutionally applied to offenders who committed crimes prior to enactment of the new statute only if it did not have a more onerous effect. Id. at 32-33, 101 S.Ct. at 966-67. The Court then determined that the new statute had a more onerous effect because it reduced the amount of gain time that could be earned for good conduct, thus lengthening the period that the offender had to spend in prison. Id. at 35-36, 101 S.Ct. at 967-68. Weaver makes it clear that retroactive changes to the parole process such as the change involved in this case violate the ex post facto clause.
Several courts have attempted to distinguish Weaver from cases involving retroactive application of the federal parole guidelines. These cases observe that in Weaver, the gain time credits for good conduct were automatic provided the offender avoid disciplinary infractions and perform assigned tasks, whereas in the case of the federal parole guidelines, the ultimate parole decision rests in the discretion of the Parole Commission. See, e.g., Inglese v. United States Parole Commission, 768 F.2d 932, 938 (7th Cir.1985). I find this a fragile distinction. The Court in Weaver expressly rejected the notion that a change in law can violate the ex post facto clause only if the change impairs a “vested right.” The Court stated, “When a court engages in ex post facto analysis, which is concerned solely with whether a statute assigns more disadvantageous criminal or penal consequences to an act than did the law in place when the act occurred, it is irrelevant whether the statutory change touches any vested rights.” Weaver, 450 U.S. at 29-30 n. 13, 101 S.Ct. at 964-65 n. 13.
Moreover, although Congress intended that the Parole Commission would retain discretion in making parole decisions, the Parole Commission and Reorganization Act places substantial constraints on the Commission’s exercise of it’s discretion.1 Prior *1302to the adoption of the Act, the Parole Board was vested with almost unlimited discretion in making parole determinations.2 Under the Act, however, the Parole Commission is required to promulgate written parole guidelines, 18 U.S.C. § 4203(a)(1), and to make parole decisions “pursuant to” those guidelines, 18 U.S.C. § 4206(a)(2).3 Although the Commission “may grant or deny release on parole notwithstanding the guidelines,” it may deviate from the guidelines only if it “determines there is good cause for so doing”4 and furnishes the offender “written notice stating with particularity the reasons for its determination, including a summary of the information relied upon.” 18 U.S.C. § 4206(c). Although individual parole determinations are “committed to agency discretion,” 18 U.S.C. § 4218(d), and therefore reviewable only for abuse of discretion, courts have held that the Commission must exercise its discretion in a manner consistent with the federal Constitution, applicable statutes, and its own published rules. Edwards v. United, 574 F.2d 937, 942 (8th Cir.1978); see also Briggs v. United States Parole Commission, 736 F.2d 446, 450 (8th Cir.1984); Joost v. United States Parole Commission, 698 F.2d 418, 419 (10th Cir.1983) (per curiam). This court has also made it clear that when the Parole Commission exceeds the guidelines, it must clearly state its reasons for determining that “good cause” exists for deviating from the guidelines and those reasons must be proper considerations under the applicable statutes and regulations. Briggs, 736 F.2d at 450; see also Joost, 698 F.2d at 419. Thus, although the Act does not eliminate the Commission’s discretion over parole decisions, the requirement that the Commission promulgate and use written guidelines in making parole determinations significantly curtails that discretion.
A review of the facts of this case clearly illustrates that retroactive application of the 1983 parole guidelines to Yamamoto had a more onerous result. Based on the 1983 guidelines, which recommended that an offender with Yamamoto’s salient factor score and offense severity rating serve one hundred or more months before release, the parole panel determined that Yama-moto should serve his full eighty-four month sentence less good time reductions. The 1979 guidelines in effect when Yama-moto committed his crime, however, recommended that an offender such as Yama-moto serve only forty to fifty-two months before parole. The Parole Commission could have required Yamamoto to serve his full eighty-four month sentence only if the panel had determined that there was “good cause” for an upward deviation from the guidelines. 18 U.S.C. § 4206(c). It is, in my judgment, difficult to rationalize that the 1983 parole guidelines are not therefore more onerous in this case.
The analogy to Weaver is readily apparent. Because the statute in effect at the *1303time Yamamoto committed his crime specified that Yamamoto would receive a parole decision within the parole guidelines unless there was “good cause” for a different result, as in Weaver, the retroactive change increased the likelihood that Yama-moto would have to serve more time in prison before release, thus resulting in a more onerous punishment. To say that the 1983 guidelines did not have a more onerous result in this case because the Parole Commission could have required Yama-moto to serve his full eighty-four month sentence under the 1979 guidelines had it determined that “good cause” existed for exceeding those guidelines ignores the admonition in Weaver that it is irrelevant that the same result might have been possible under another provision. Weaver, 450 U.S. at 34-36, 101 S.Ct. at 967-68. Thus, Weaver refutes the Commission’s argument, and the majority’s reference to this argument,5 that Yamamoto might have received a parole decision outside the 1979 guidelines. In any event, such an argument is mere speculation.
Prior decisions of this circuit holding that retroactive application of the parole guidelines did not result in a more onerous punishment in those particular cases are clearly distinguishable. In fact, these cases illustrate that use of the 1983 parole guidelines to determine Yamamoto’s release date had a more onerous effect. As the majority notes, the issue in Hayward was whether the use of the newly promulgated guidelines to determine the parole eligibility of an offender who committed a crime prior to the adoption of the guidelines violated the ex post facto clause. Because Hayward had no expectation as to the length of time he would serve before parole at the time he committed his crime, it was clear that adoption and use of the written guidelines did not result in a more onerous punishment. In contrast, at the time Yama-moto committed his crime, written parole guidelines were in effect and specified the length of time an offender such as Yama-moto would be required to serve before parole absent good cause for a deviation.6
Richardson v. United States Parole Commission, 729 F.2d 1154 (8th Cir.1984) (per curiam), is also distinguishable. In Richardson, we considered whether the ex post facto clause barred application of the 1980 federal parole guidelines to an offender who committed a crime at the time the 1976 guidelines were in effect. The 1976 guidelines recommended that an offender such as Richardson serve a minimum of seventy-two months before parole, whereas the 1980 guidelines recommended that such a prisoner serve one hundred or more months before parole. Applying the 1980 guidelines, the Parole Commission determined that Richardson should serve eighty-seven months before parole. Because a term of imprisonment of eighty-seven months prior to parole was fully contemplated by the 1976 guidelines, we held that use of the 1980 guidelines did not result in a more onerous punishment and thus did not violate the ex post facto clause. Richardson, 729 F.2d at 1155; see also Rush v. Petrovsky, 756 F.2d 675 (8th Cir.1985) (per curiam) (use of 1983 guidelines, recommending parole after fifty-two to sixty-four months, to determine parole eligibility of offender who committed crime in 1982, when 1982 guidelines recommending parole after thirty-six to forty-eight months were in effect, was not more onerous where Parole Commission determined offender should serve forty to forty-seven months *1304before parole).7 By contrast, under the 1979 guidelines, Yamamoto could not have received the parole determination he did unless there was “good cause” for exceeding the guidelines. The 1983 guidelines thus clearly had a more onerous result than the guidelines in effect when Yamamoto committed his crime.
Although the majority seemingly finds it unnecessary to consider whether or not the federal parole guidelines are laws,8 in view of my conclusion that the guidelines have a more onerous result in this case, I address this issue. In concluding that the federal parole guidelines are not laws, some courts have relied on the fact that the guidelines were promulgated by an agency rather than by Congress itself.9 I strongly disagree with this implication. As the majority correctly notes, Congress may clearly not, by delegation, escape constitutional limitations on its power.10
Other courts have relied on the fact that the Parole Commission retains substantial discretion in making parole determinations notwithstanding the guidelines.11 These cases, however, do not cite any authority for the novel proposition that guidelines promulgated pursuant to congressional mandate are not laws merely because the *1305agency may deviate from the guidelines if there is “good cause” for doing so. This reasoning would have some merit if the guidelines merely provided a presumptive release date from which the Parole Commission could deviate at its complete discretion and if courts had absolutely no authority to review the Commission’s parole determinations. The system of “checks” on the Commission’s discretion in making parole decisions discussed above, however, is inconsistent with the notion that the guidelines are not laws.
I conclude therefore that (1) the 1983 guidelines are laws, which when applied retroactively, as in this case, are (2) more onerous than the guidelines in effect when Yamamoto committed his crime. Under the circumstances, Yamamoto deserves immediate release from prison, because to hold him incarcerated for any additional time is to deny him his constitutional right to be free from an ex post facto application of the law.
. As this court stated in Edwards v. United States, 574 F.2d 937 (8th Cir.), cert. dismissed, 439 U.S. 1040, 99 S.Ct. 643, 58 L.Ed.2d 700 (1978):
The goals of the guideline system are (1) to reduce the disparity in sentences meted out to similar offenders committing identical offenses, and (2) to contribute more predictability and fairness to Parole Board determina*1302tions. It is clear that the guidelines do carry great, if not overwhelming, weight in Parole Board determinations.
Edwards, 574 F.2d 942-43 (footnote omitted).
. The prior statute authorizing the Parole Board to make parole decisions is set out in footnote 11 of the majority opinion.
. Section 4206(a) provides:
If an eligible prisoner has substantially observed the rules of the institution or institutions to which he has been confined, and if the Commission, upon consideration of the nature and circumstances of the offense and the history and characteristics of the prisoner, determines:
(1) that release would not depreciate the seriousness of his offense or promote disrespect for the law; and
(2) that release would not jeopardize the public welfare; subject to the provisions of subsections (b) and (c) of this section, and pursuant to guidelines promulgated by the Commission pursuant to section 4203(a)(1), such prisoner shall be released.
18 U.S.C. § 4206(a).
. “For the purposes of this section ‘good cause’ means substantial reason and includes only those grounds put forward by the Commission in good faith and which are not arbitrary, irrational, unreasonable, irrelevant or capricious." H.R.Rep. No. 838, 94th Cong., 2d Sess. 19, 27, reprinted in 1976 U.S. Code Cong. & Ad. News 351, 359.
. See footnote 12 of the majority opinion.
. The majority asserts that because the legislative history of the Parole Commission and Reorganization Act and the Parole Commission regulations state that the guidelines should be revised or modified if deviations from the guidelines are frequent, an offender has no expectation that the parole guidelines in effect at the time of the offense will be used to determine the offender's parole eligibility. This is a makeshift argument. It suggests that Congress or an agency may avoid the constitutional prohibition against ex post facto laws merely by adding to the statute or rule that it is subject to revision. All laws are subject to revision, and merely because one is put on notice of this obvious fact does not destroy a person’s expectation of having his or her conduct viewed under the law existing at the time of the crime.
. The court required Richardson to show "that he would have received a more favorable parole determination under the 1976 guidelines than he received under the 1980 guidelines.” Richardson, 729 F.2d at 1155. In determining that a retroactive change in the sentence for a particular crime from a discretionary term of six months to fifteen years to a mandatory sentence of fifteen years violated the ex post facto clause, however, the Supreme Court held in Lindsey v. Washington, 301 U.S. 397, 57 S.Ct. 797, 81 L.Ed. 1182 (1937):
It is true that petitioners might have been sentenced to fifteen years under the old statute. But the ex post facto clause looks to the standard of punishment prescribed by a statute, rather than to the sentence actually imposed ____ It is for this reason that an increase in the possible penalty is ex post facto, ... regardless of the length of the sentence actually imposed, since the measure of punishment prescribed by the later statute is more severe than that of the earlier, ____
Id. at 401, 57 S.Ct. at 799 (emphasis added) (citations omitted). In interpreting Lindsey, the Supreme Court recently stated,
We think the excerpted language from Lindsey must be read in the light of these facts to mean that one is not barred from challenging a change in the penal code on ex post facto grounds simply because the sentence he received under the new law was not more onerous than that which he might have received under the old.
Dobbert v. Florida, 432 U.S. 282, 300, 97 S.Ct. 2290, 2302, 53 L.Ed.2d 344 (1977). These decisions strongly suggest that an ex post facto inquiry must focus on the overall effect of a change in law, rather than on the actual result of the change in the particular case. The validity of Richardson and Rush in light of these Supreme Court cases is thus highly questionable.
. The majority approves decisions of other circuits that hold that the federal parole guidelines are not laws. See footnote 7 of the majority opinion. At the same time, the majority states that it does not decide this issue. The majority also states that "Congress may not by delegation escape constitutional limitations on its power.” The majority’s position is confusing because many of the decisions it approves rely on the specious reasoning that the federal parole guidelines are not passed by Congress. See infra note 9.
. See, e.g., Dufresne v. Baer, 744 F.2d 1543, 1549-50 (11th Cir.1984), cert. denied, — U.S. —, 106 S.Ct. 61, 88 L.Ed.2d 49 (1985) ("[The guidelines] are not promulgated by any legislature. Rather, they are stated policy rules that show how agency discretion is likely to be exercised.”); Ruip v. United States, 555 F.2d 1331, 1336 (6th Cir.1977) (in distinguishing cases suggesting that retroactive changes in statutes affecting parole determinations might violate ex post facto clause, the court stated, "Since the relevant provisions were statutes, there was no question that they were encompassed in the prohibition on ex post facto laws. The situation here is different.”); see also Inglese, 768 F.2d at 941 (Cudahy, J., concurring).
. Promulgation of and changes to the guidelines are expressly subject to the rulemaking procedures of the Administrative Procedure Act notwithstanding the Parole Commission's designation of the guidelines as general statements of policy. See 18 U.S.C. § 4218(a)-(c) (1982).
. See, e.g., United States ex rel. Forman v. McCall, 776 F.2d 1156, 1163 (3d Cir.1985); In-glese, 768 F.2d at 936; DiNapoli, 764 F.2d at 146-47; Dufresne, 744 F.2d at 1550. As one court stated, “The power to exercise discretion indicates that the guidelines are merely guides, and not laws: guides may be discarded where circumstances require; laws may not.” Inglese, 768 F.2d at 936.