Greenholtz v. Inmates of the Nebraska Penal & Correctional Complex

Mr. Justice Marshall,

with whom Mr. Justice Brennan and Mr. Justice Stevens join,

dissenting in part.

My disagreement with the Court’s opinion extends to both its analysis of respondents’ liberty interest and its delineation of the procedures constitutionally required in parole release proceedings. Although it ultimately holds that the Nebraska statutes create a constitutionally protected “expectation of parole,” the Court nonetheless rejects the argument that criminal offenders have such an interest whenever a State establishes the possibility of parole. This gratuitous commentary reflects a misapplication of our prior decisions and an unduly narrow view of the liberty protected by the Fourteenth Amendment. Since the Court chooses to address the issue, I must register my opinion that all prisoners potentially eligible for parole have a liberty interest of which they may not be deprived without due process, regardless of the particular statutory language that implements the parole system.

The Court further determines that the Nebraska Board of Parole already provides all the process that is constitutionally due. In my view, the Court departs from the analysis adopted in Morrissey v. Brewer, 408 U. S. 471 (1972), and Mathews v. Eldridge, 424 U. S. 319, 335 (1976), and disregards considerations that militate for greater procedural protection. To supplement existing procedures, I would require that the Parole Board give each inmate reasonable notice of hearing dates and the factors to be considered, as well as a written statement of reasons and the essential facts underlying adverse decisions.

*23 I

A

It is self-evident that all individuals possess a liberty interest in being free from physical restraint. Upon conviction for a crime, of course, an individual may be deprived of this liberty to the extent authorized by penal statutes. 1 But when a State enacts a parole system, and creates the possibility of release from incarceration upon satisfaction of certain conditions, it necessarily qualifies that initial deprivation. In my judgment, it is the existence of this system which allows prison inmates to retain their protected interest in securing freedoms available outside prison. 2 Because parole release proceedings clearly implicate this retained liberty interest, the Fourteenth Amendment requires that due process be observed, irrespective of the specific provisions in the applicable parole statute.

This Court’s prior decisions fully support the conclusion that criminal offenders have a liberty interest in securing parole release. In Morrissey v. Brewer, supra, the Court held that all persons released on parole possess such an interest in remaining free from incarceration. Writing for the Court, Mr. Chief Justice Burger stated that the appli *24 cability of due process protections turns "on the extent to which an individual will be 'condemned to suffer grievous loss/ ” citing Joint Anti-Fascist Refugee Committee v. McGrath, 341 U. S. 123, 168 (1951) (Frankfurter, J., concurring), and on the “nature of the interest.” 408 U. S., at 481 . In assessing the gravity and nature of the loss caused by parole revocation, Morrissey relied on the general proposition that parole release enables an individual “to do a wide range of things open to persons who have never been convicted of any crime.” Id., at 482 . 3 Following Morrissey, Gagnon v. Scarpelli, 411 U. S. 778 (1973), held that individuals on probation also retain a liberty interest which cannot be terminated without due process of law. Nowhere in either opinion did the Court even intimate that the weight or nature of the criminal offender’s interest in maintaining his parole release or probation depends upon the specific terms of any statute, for in both cases the Court disregarded the applicable statutory language. 4 Rather, this liberty interest derived solely from the *25 existence of a system that permitted criminal offenders to serve their sentences on probation or parole.

Wolff v. McDonnell, 418 U. S. 539 (1974), adopted a similar approach. There, the Court concluded that abrogation of a prisoner’s good-time credits implicates his interest in subsequently obtaining release from incarceration. Although the Court recognized that Nebraska was not constitutionally obligated to establish a credit system, by creating “a right to a shortened prison sentence through the accumulation of credits for good behavior,” id., at 557 , the State had allowed inmates to retain a liberty interest that could be terminated only for “serious misbehavior.” This liberty interest derived from the existence of a credit system, not from the specific language of the implementing statute, see id., at 555-558 , as decisions applying Wolff have consistently recognized. 5

B

A criminal offender’s interest in securing release on parole is therefore directly comparable to the liberty interests we *26 recognized in Morrissey, Scarpelli, and Wolff . However, because the Court discerns two distinctions between “parole release and parole revocation,” ante, at 9, it refuses to follow these cases here. In my view, the proffered distinctions do not support this departure from precedent.

First, the Court finds a difference of constitutional dimension between a deprivation of liberty one has and a denial of liberty one desires. Ibid. While there is obviously some difference, it is not one relevant to the established constitutional inquiry. Whether an individual currently enjoys a particular freedom has no bearing on whether he possesses a protected interest in securing and maintaining that liberty. The Court acknowledged as much in Wolff v. McDonnell, supra, when it held that the loss of good-time credits implicates a liberty interest even though the forfeiture only deprived the prisoner of freedom he expected to obtain sometime hence. See Drayton v. McCall, 584 F. 2d 1208, 1219 (CA2 1978). And in other contexts as well, this Court has repeatedly concluded that the Due Process Clause protects liberty interests that individuals do not currently enjoy. 6

The Court’s distinction is equally unrelated to the nature *27 or gravity of the interest affected in parole release proceedings. The nature of a criminal offender’s interest depends on the range of freedoms available by virtue of the parole system’s existence. On that -basis, Morrissey afforded constitutional recognition to a parolee’s interest because his freedom on parole includes “many of the core values of unqualified liberty.” 408 U. S., at 482 . This proposition is true regardless of whether the inmate is presently on parole or seeking parole release. As the Court of Appeals for the Second Circuit has recognized, “[w]hether the immediate issue be release or revocation, the stakes are the same: conditional freedom versus incarceration.” United States ex rel. Johnson v. Chairman of New York State Board of Parole, 500 F. 2d 925, 928 , vacated as moot sub nom. Regan v. Johnson, 419 U. S. 1015 (1974).

The Court’s second justification for distinguishing between parole release and parole revocation is based on the “nature of the decision that must be made in each case.” Ante, at 9. The majority apparently believes that the interest affected by parole release proceedings is somehow diminished if the administrative decision may turn on “subjective evaluations.” Yet the Court nowhere explains why the nature of the de-cisional process has even the slightest bearing in assessing the nature of the interest that this process may terminate. 7 Indeed, the Court’s reasoning here is flatly inconsistent with its subsequent holding that respondents do have a protected liberty interest under Nebraska’s parole statutes, which require a decision that is “subjective in part and predictive in part.” Ante, at 13. For despite the Parole Board’s argument that such an interest exists “only if the statutory con *28 ditions for [denying parole are] essentially factual, as in Wolff and Morrissey , rather than predictive,” ante, at 12, the Court nonetheless concludes that respondents’ interest is sufficient to merit constitutional protection.

But even assuming the subjective nature of the decision-making process were relevant to due process analysis in general, this consideration does not adequately distinguish the processes of granting and revoking parole. See Morrissey v. Brewer, 408 U. S., at 477-480 ; Gagnon v. Scarpelli, 411 U. S., at 781-782 . Contrary to the Court’s assertion that the decision to revoke parole is predominantly a “ 'retrospective factual question,’ ” ante, at 9, Morrissey recognized that only the first step in the revocation decision can be so characterized. And once it is

“determined that the parolee did violate the conditions [of parole, a] second question arise [s]: should the parolee be recommitted to prison or should other steps be taken to protect society and improve chances of rehabilitation? The first step is relatively simple; the second is more complex. The second question involves the application of expertise by the parole authority in making a prediction as to the ability of the individual to live in society without committing antisocial acts. . . . [T]his second step, deciding what to do about the violation once it is identified, is not purely factual but also predictive and discretionary.” 408 U. S., at 479-480 (emphasis added).

Morrissey thus makes clear that the parole revocation decision includes a decisive subjective component. Moreover, to the extent parole release proceedings hinge on predictive determinations, those assessments are necessarily predicated on findings of fact. 8 Accordingly, the presence of subjective *29 considerations is a completely untenable basis for distinguishing the interests at stake here from the liberty interest recognized in Morrissey .

C

The Court also concludes that the existence of a parole system by itself creates “no more than a mere hope that the benefit will be obtained,” ante, at 11, and thus does not give rise to a liberty interest. This conclusion appears somewhat gratuitous, given the Court’s ultimate holding that the Nebraska statutes do generate a “legitimate expectation of [parole] release” which is protected by the Due Process Clause. Ante, at 12. Moreover, it is unclear what purpose can be served by the Court’s endeavor to depreciate the expectations arising solely from the existence of a parole system. The parole statutes in many jurisdictions embody the same, standards used in the Model Penal Code, upon which both the Nebraska and federal provisions are patterned, and the Court’s analysis of the Nebraska statutes would therefore suggest that the other statutes must also create protectible expectations of release. 9

*30 Furthermore, in light of the role that parole has assumed in the sentencing process, I believe the Court misapplies its own test, see ante, at 11-12, by refusing to acknowledge that inmates have a legitimate expectation of release whenever the government establishes a parole system. As the Court observed in Morrissey :

“During the past 60 years, the practice of releasing prisoners on parole before the end of their sentences has become an integral part of the penological system. . . . Rather than being an ad hoc exercise of clemency, parole is an established variation on imprisonment of convicted criminals.” 408 U. S., at 477 .

Indeed, the available evidence belies the majority’s broad assumptions concerning inmate expectations, at least with respect to the federal system, and there is no suggestion that experience in other jurisdictions is significantly different. 10

Government statistics reveal that substantially less than one-third of all first-time federal offenders are held in prison until mandatory release. 11 In addition, 88% of the judges responding to a recent survey stated that they considered the availability of parole when imposing sentence, and 47% acknowledged their expectation that defendants would be re *31 leased on parole after serving one-third of their sentences. 12 In accord with these views, the Administrative Conference of the United States has advised Congress that courts set maximum sentences anticipating “that a prisoner who demonstrates his desire for rehabilitation will not serve the maximum term or anything approaching the maximum.” 13 And in discussing the sentencing provisions of the proposed revision of the Federal Criminal Code, S. 1437, the Senate Judiciary Committee observed:

“A federal judge who today believes that an offender should serve four years in prison may impose a sentence in the vicinity of ten years, knowing that the offender is eligible for parole release after one third of the sentence.” S. Rep. No. 95-605, p. 1169 (1977).

Thus, experience in the federal system has led both judges and legislators to expect that inmates will be paroled substantially before their sentences expire. Insofar as it is critical under the Court’s due process analysis, this understanding would certainly justify a similar expectation on the part of the federal inmates. Hence, I believe it is unrealistic for this Court to speculate that the existence of a parole system provides prisoners “no more than a mere hope” of release. Ante, at 11.

II

A

I also cannot subscribe to the Court’s assessment of the procedures necessary to safeguard respondents’ liberty interest. Although the majority purports to rely on Morrissey v. *32 Brewer and the test enunciated in Mathews v. Eldridge, 424 U. S. 319 (1976), its application of these standards is fundamentally deficient in several respects.

To begin with, the Court focuses almost exclusively on the likelihood that a particular procedure will significantly reduce the risk of error in parole release proceedings. Ante, at 14-16. Yet Mathews advances three factors to be considered in determining the specific dictates of due process:

“First, the private interest that will be affected by the official action; second, the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards; and finally, the Government’s interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail.” 424 U. S., at 335 .

By ignoring the other two factors set forth in Mathews , the Court skews the inquiry in favor of the Board. For example, the Court does not identify any justification for the Parole Board’s refusal to provide inmates with specific advance notice of the hearing date or with a list of factors that may be considered. Nor does the Board demonstrate that it would be unduly burdensome to provide a brief summary of the evidence justifying the denial of parole. To be sure, these measures may cause some inconvenience, but “the Constitution recognizes higher values than speed and efficiency.” Stanley v. Illinois, 405 U. S. 645, 656 (1972); accord, Frontiero v. Richardson, 411 U. S. 677, 690 (1973); Bell v. Burson, 402 U. S. 535, 540-541 (1971). Similarly lacking in the Court’s analysis is any recognition of the private interest affected by the Board’s action. Certainly the interest in being released from incarceration is of sufficient magnitude to have some bearing on the process due. 14

*33 The second fundamental flaw in the Court’s analysis is that it incorrectly evaluates the only factor actually discussed. The contribution that additional safeguards will make to reaching an accurate decision necessarily depends on the risk of error inherent in existing procedures. See Mathews v. Eldridge, supra, at 334-335, 343-347 . Here, the Court finds supplemental procedures to be inappropriate because it assumes existing procedures adequately reduce the likelihood that an inmate’s files will contain incorrect information which could lead to an erroneous decision. No support is cited for this assumption, and the record affords none. In fact, researchers and courts have discovered many substantial inaccuracies in inmate files, and evidence in the instant case revealed similar errors. 15 Both the District Court and the Court of Appeals *34 found additional procedures necessary to decrease the margin of error in Nebraska’s parole release proceedings. Particularly since the Nebraska statutes tie the parole decision to a number of highly specific factual inquiries, see ante, at 16-18, I see no basis in the record for rejecting the lower courts’ conclusion.

Finally, apart from avoiding the risk of actual error, this Court has stressed the importance of adopting procedures that preserve the appearance of fairness and the confidence of inmates in the decisionmaking process. The Chief Justice recognized in Morrissey that “fair treatment in parole revocations will enhance the chance of rehabilitation by avoiding reactions to arbitrariness,” 408 U. S., at 484 (citation omitted), a view shared by legislators, courts, the American Bar Association, and other commentators. 16 This consideration is equally significant whether liberty interests are extinguished in parole release or parole revocation proceedings. As Mr. Justice Frankfurter argued in Joint Anti-Fascist Refugee Committee v. McGrath, 341 U. S., at 171-172 (concurring opinion):

“The validity and moral authority of a conclusion largely depend on the mode by which it was reached. Secrecy is not congenial to truth-seeking and self-righteousness gives too slender an assurance of rightness. No better instrument has been devised for arriving at truth than to give a person in jeopardy of serious loss notice of the *35 case against him and opportunity to meet it. Nor has a better way been found for generating the feeling, so important to a popular government, that justice has been done.”

In my judgment, the need to assure the appearance, as well as the existence, of fairness supports a requirement that the Parole Board advise inmates of the specific dates for their hearings, tho criteria to be applied, and the reasons and essential facts underlying adverse decisions. For “‘[o]ne can imagine nothing more cruel, inhuman, and frustrating than serving a prison term without knowledge of what will be measured and the rules determining whether one is ready for release.’ ” K. Davis, Discretionary Justice: A Preliminary Inquiry 132 (1969).

B

Applying the analysis of Morrissey and Mathews , I believe substantially more procedural protection is necessary in parole release proceedings than the Court requires. The types of safeguards that should be addressed here, however, are limited by the posture of this case. 17 Thus, only three specific issues need be considered.

*36 While the question is close, I agree with the majority that a formal hearing is not always required when an inmate first becomes eligible for discretionary parole. Ante, at 14^15. The Parole Board conducts an initial parole review hearing once a year for every inmate, even before the inmate is eligible for release. Although the scope of this hearing is limited, inmates are allowed to appear and present letters or statements supporting their case. If the Board concludes that an eligible inmate is a good candidate for release, it schedules a final and substantially more formal hearing.

The Court of Appeals directed the Parole Board to conduct such a formal hearing as soon as an inmate becomes eligible' for parole, even where the likelihood of a favorable decision is negligible, but the court required no hearing thereafter. 576 P. 2d 1274, 1285 (CA8 1978). From a practical standpoint, this relief offers no appreciable advantage to the inmates. If the Board would not have conducted a final hearing under current procedures, inmates gain little from a requirement that such a hearing be held, since the evidence almost certainly would be insufficient to justify granting release. And because the Court of Appeals required the Board to conduct only one hearing, inmates risk losing the right to a formal proceeding at the very point additional safeguards may have a beneficial impact. The inmates’ interest in this modification of the Board’s procedures is thus relatively slight. 18 Yet the burden *37 imposed on the Parole Board by the additional formal hearings would be substantial. Accordingly, I believe the Board’s current practice of combining both formal and informal hearings is constitutionally sufficient.

However, a different conclusion is warranted with respect to the hearing notices given inmates. The Board currently informs inmates only that it will conduct an initial review or final parole hearing during a particular month within the next year. The notice does not specify the day or hour of the hearing. Instead, inmates must check a designated bulletin board each morning to see if their hearing is scheduled for that day. In addition, the Board refuses to advise inmates of the criteria relevant in parole release proceedings, despite a state statute expressly listing 14 factors the Board must consider and 4 permissible reasons for denying parole. See Neb. Rev. Stat. § 83-1 ,114 (1976), quoted ante, at 11, 16-18.

Finding these procedures insufficient, the District Court and the Court of Appeals ordered that each inmate receive written advance notice of the time set for his hearing, along with a list of factors the Board may consider. 576 F. 2d, at 1285. 19 Although the Board has proffered no justification for refusing to institute these procedures, id., at 1283, the Court sets aside the relief ordered below on the ground that “[t]here is no claim that either the timing of the notice or its substance seriously prejudices the inmate’s ability to prepare adequately for the hearing.” Ante, at 14 n. 6. But respondents plainly have contended throughout this litigation that reasonable advance notice is necessary to enable them to organize their evidence, call the witnesses permitted by the Board, and notify private counsel allowed to participate in the *38 hearing, see Brief for Respondents 65-66; Answer Brief for Appellee Inmates in No. 77-1889 (CA8), pp. 6, 8-9, 25, 28; Trial Brief for Inmates in Civ. 72-L-335 (Neb.), pp. 17-18; and the courts below obviously agreed. See 576 F. 2d, at 1283 ; Mem. Op. in Civ. 72-L-335 (Neb., Oct. 21, 1977), App. to Pet. for Cert. 25, 39, 45-47. Given the significant private interests at stake, and the importance of reasonable notice in preserving the appearance of fairness, I see no reason to depart here from this Court’s longstanding recognition that adequate notice is a fundamental requirement of due process, e. g., Memphis Light, Gas & Water Division v. Craft, 436 U. S. 1, 13 (1978); Mullane v. Central Hanover Trust Co., 339 U. S. 306, 314 (1950), a principle heretofore found equally applicable in the present context. Wolff v. McDonnell, 418 U. S., at 563-564 ; Gagnon v. Scarpelli, 411 U. S., at 786 ; Morrissey v. Brewer, 408 U. S., at 486-487, 489 .

Finally, I would require the Board to provide a statement of the crucial evidence on which it relies in denying parole. 20 At present, the Parole Board merely uses a form letter noting the general reasons for its decision. In ordering the Board to *39 furnish as well a summary of the essential facts underlying the denial, the Court of Appeals made clear that “ 'detailed findings of fact are not required.’ ” 576 F. 2d, at 1284. The majority here, however, believes even this relief to be unwarranted, because it might render parole proceedings more adversary and equate unfavorable decisions with a determination of guilt. Ante, at 15-16.

The Court nowhere explains how these particular considerations are relevant to the inquiry required by Morrissey and Mathews . Moreover, it is difficult to believe that subsequently disclosing the factual justification for a decision will render the proceeding more adversary, especially when the Board already provides a general statement of reasons. 21 And to the extent unfavorable parole decisions resemble a determination of guilt, the Board has no legitimate interest in concealing from an inmate the conduct or failings of which he purportedly is guilty.

While requiring a summation of the essential evidence might entail some administrative inconvenience, in neither Morrissey v. Brewer, supra, at 489 ; Gagnon v. Scarpelli, supra, at 786 ; nor Wolff v. McDonnell, supra, at 563, 564-565 , did the Court find that this factor justified denying a written statement of the essential evidence and the reasons underlying a decision. It simply is not unduly

“burdensome to give reasons when reasons exist. Whenever an application ... is denied . . . there should be some reason for the decision. It can scarcely be argued that government would be crippled by a requirement that the reason be communicated to the person most directly affected by the government’s action.” Board of Regents v. Roth, 408 U. S. 564, 591 (1972) (Marshall, J., dissenting).

*40 See Mathews v. Eldridge, 424 U. S., at 345-346 ; SEC v. Chenery Corp., 318 U. S. 80 (1943). And an inability to provide any reasons suggests that the decision is, in fact, arbitrary. 22

Moreover, considerations identified in Morrissey and Mathews militate in favor of requiring a statement of the essential evidence. Such a requirement would direct the Board’s focus to the relevant statutory criteria and promote more careful consideration of the evidence. It would also enable inmates to detect and correct inaccuracies that could have a decisive impact. 23 And the obligation to justify a decision publicly would provide the assurance, critical to the appearance of fairness, that the Board’s decision is not capricious. Finally, imposition of this obligation would afford inmates instruction on the measures needed to improve their prison behavior and prospects for parole, a consequence surely consistent with rehabilitative goals. 24 Balancing these con *41 siderations against the Board’s minimal interest in avoiding this procedure, I am convinced that the Fourteenth Amendment requires the Parole Board to provide inmates a statement of the essential evidence as well as a meaningful explanation of the reasons for denying parole release. 25

Because the Court’s opinion both depreciates inmates’ fundamental liberty interest in securing parole release and sanctions denial of the most rudimentary due process protection, I respectfully dissent.

A criminal conviction cannot, however, terminate all liberty interests. Wolff v. McDonnell, 418 U. S. 539, 555-556 (1974); see, e. g., Procunier v. Navarette, 434 U. S. 555 (1978); Bounds v. Smith, 430 U. S. 817 (1977); Pell v. Procunier, 417 U. S. 817, 822 (1974); Cruz v. Beto, 405 U. S. 319 (1972); Wilwording v. Swenson, 404 U. S. 249 (1971); Cooper v. Pate, 378 U. S. 546 (1964); Ex parte Hull, 312 U. S. 546 (1941); Weems v. United States, 217 U. S. 349 (1910). See also Carmona v. Ward, 439 U. S. 1091 (1979) (Marshall, J., dissenting).

See Bell v. Wolfish, 441 U. S. 520, 568-571 (1979) (Marshall, J., dissenting); id., at 580-584 (Stevens, J., dissenting); Leis v. Flynt, 439 U. S. 438, 448-453 (1979) (Stevens, J., dissenting); Meachum v. Fano, 427 U. S. 215, 230 (1976) (Stevens, J., dissenting); cf. Bell v. Wolfish, supra, at 535-536, 545 . See generally Smith v. Organization of Foster Families, 431 U. S. 816, 842-847 (1977).

Because parolees’ enjoyment of these freedoms was subject to a number of restrictions, the Court characterized their liberty interest as “conditional.” See 408 U. S., at 480. The risk that violation of those conditions could lead to termination of parole status, however, did not diminish the significance of the parolees’ interest, since the Due Process Clause anticipates that most liberty interests may be abrogated under proper circumstances. So, too, here, respondents’ interest does not forfeit constitutional protection simply because their freedom would also be subject to conditions or because of the possibility that the Nebraska Parole Board will deny release after providing due process of law.

The state law in Morrissey , quoted only in the dissenting opinion, provided that “ '[a]ll paroled prisoners . . . shall be subject, at any time, to be taken into custody and returned to the institution ....’” 408 U. S., at 493 n. 2 (Douglas, J., dissenting in part). The statute specified no other criteria for parole revocation. Thus, had the Court relied solely on particular statutory language, it could not have held that parolees possess a constitutionally protected interest in continuing their status. In Scarpdli, the Court completely ignored the pertinent statutory language. See 411 U. S., at 781-782.

Cf. Baxter v. Palmigiano, 425 U. S. 308, 323-324 (1976). Lower courts have understood Wolff to require due process safeguards whenever good-time credits are revoked, and have not focused on the language of various statutory provisions. See, e. g., Franklin v. Shields, 569 F. 2d 784 , 788-790, 800-801 (CA4) (en banc), cert. denied, 435 U. S. 1003 (1978); United States ex rel. Larkins v. Oswald, 510 F. 2d 583 (CA2 1975); Gomes v. Travisono, 510 F. 2d 537 (CA1 1974); Willis v. Ciccone, 506 F. 2d 1011, 1017 (CA8 1974); Workman v. Mitchell, 502 F. 2d 1201 (CA9 1974). See also United States ex rel. Miller v. Twomey, 479 F. 2d 701, 712-713 (CA7 1973) (Stevens, J.), cert. denied sub nom. Gutierrez v. Department of Public Safety of Ill., 414 U. S. 1146 (1974).

Meachum v. Fano, 427 U. S. 215 (1976), signals no' departure from the basic principles recognized in Morrissey, Gagnon , and Wolff . While the majority in Meachum concluded that the prisoners did not have a protected liberty interest in avoiding transfers between penal institutions, the Court’s opinion rested on the absence of any limitation on such transfers rather than on particular statutory language. 427 U. S., at 225-228 . See Tracy v. Salamack, 572 F. 2d 393 , 395 n. 9 (CA2 1978); Four Certain Unnamed Inmates v. Hall, 550 F. 2d 1291, 1292 (CA1 1977).

See, e. g., Willner v. Committee on Character and Fitness, 373 U. S. 96 (1963); Speiser v. Randall, 357 U. S. 513 (1958); Konigsberg v. State Bar, 353 U. S. 252 (1957); Schware v. Board of Bar Examiners, 353 U. S. 232 (1957); Simmons v. United States, 348 U. S. 397 (1955); Goldsmith v. Board of Tax Appeals, 270 U. S. 117 (1926).

The Second Circuit has characterized the attempt to differentiate between a liberty interest currently enjoyed but subject to termination, and an interest that can be enjoyed in the future following an administrative proceeding, as actually “nothing' more than a reincarnation of the right-privilege dichotomy in a not-too-deceptive disguise.” United States ex rel. Johnson v. Chairman of New York State Board of Parole, 500 F. 2d 925 , 927-928, n. 2, vacated as moot sub nom. Regan v. Johnson, 419 U. S. 1015 (1974), construing United States ex rel. Bey v. Connecticut Board of Parole, 443 F. 2d 1079, 1086 (CA2 1971), which the Court quotes ante, at 10; see Comment, The Parole System, 120 U. Pa. L. Rev. 282 , 363 (1971).

Government decisionmakers do not gain a “license for arbitrary procedure” when legislators confer a “substantial degree of discretion” regarding the assessment of subjective considerations. Kent v. United States, 383 U. S. 541, 553 (1966); see Thorpe v. Housing Authority of City of Durham, 386 U. S. 670, 678 (1967) (Douglas, J., concurring).

See Franklin v. Shields, 569 F. 2d, at 791; Dawson, The Decision to Grant or Deny Parole: A Study of Parole Criteria in Law and Practice, 1966 Wash. U. L. Q. 243, 248-285; cf. Morrissey v. Brewer, 408 U. S., at *29 479-480 . The Nebraska statutes, in particular, demonstrate the factual nature of the parole release inquiry. One provision, quoted ante, at 16-18, enumerates factual considerations such as the inmate’s intelligence, family status, and employment history, which bear upon the four predictive determinations underlying the ultimate parole decision. See ante, at 11.

The parole statutes of 47 States establish particular standards, criteria, or factors to be applied in parole release determinations. A list of these statutes is set out in the Brief for Jerome N. Frank Legal Services Organization et al. as Amici Curiae 30-31, 23a-26a. These criteria presumably will be a significant source of inmates’ “legitimate expectations” regarding the availability of parole. Expectations would also be shaped by the role that parole actually assumes in a jurisdiction’s penological system, see infra, at 30-31. It is in these respects that most parole statutes are similar. While there are some differences in statutory language among jurisdictions, it is unrealistic to believe that variations such as the use of “may” rather than “shall,” see ante, at 11-12, could negate the expectations derived from *30 experience with a parole system and the enumerated criteria for granting release.

The New York State Parole Board, for-example, granted parole in 75.4% of the cases it considered during 1972. See United States ex rel. Johnson v. Chairman of New York State Board of Parole, 500 F. 2d, at 928 . In addition, recent studies show that parole is the method of release for approximately 70% of all criminal offenders returned each year to the community. Uniform Parole Reports, Parole in the United States: 1976 and 1977, p. 55 (1978). In some States, the figure is as high as 97%. See Kastenmeier & Eglit, Parole Release Decision-Making: Rehabilitation, Expertise, and the Demise of Mythology, 22 Am. U. L. Rev. 477 , 481-482 (1973).

See Brief for United States in United States v. Addonizio, O. T. 1978, No. 78-156, p. 55 n. 47.

Project, Parole Release Decisionmaking and the Sentencing Process, 84 Yale L. J. 810, 882 n. 361 (1975).

Hearings on H. R. 1598 and Identical Bills before the Subcommittee on Courts, Civil Liberties, and the Administration of Justice of the House Committee on the Judiciary, 93d Cong., 1st Sess., 163-164, 193 (1973) (testimony and statement of Antonin Scalia, Chairman of the Administrative Conference of the United States).

While the severity of a loss does not of itself establish that an interest deserves constitutional protection, this factor does weigh heavily in deter *33 mining the procedural safeguards mandated by the Fourteenth Amendment. See Goss v. Lopez, 419 U. S. 565, 575-576 (1975); Board of Regents v. Roth, 408 U. S. 564 (1972).

In this case, for example, the form notifying one inmate that parole had been denied indicated that the Board believed he should enlist in a self-improvement program at the prison. But in fact, the inmate was already participating in all such programs available. Tr. 38-39. Such errors in parole files are not unusual. E. g., Kohlman v. Norton, 380 F. Supp. 1073 (Conn. 1974) (parole denied because file erroneously indicated that applicant had used gun in committing robbery); Leonard v. Mississippi State Probation and Parole Board, 373 F. Supp. 699 (ND Miss. 1974), rev’d, 509 F. 2d 820 (CA5), cert. denied, 423 U. S. 998 (1975) (prisoner denied parole on basis of illegal disciplinary action); In re Rodriguez, 14 Cal. 3d 639 , 537 P. 2d 384 (1975) (factually incorrect material in file led parole officers to believe that prisoner had violent tendencies and that his “family rejectfed] him”); State v. Pohlabel, 61 N. J. Super. 242, 160 A. 2d 647 (1960) (files erroneously showed that prisoner was under a life sentence in another jurisdiction); Hearings on H. R. 13118 et al. before Subcommittee No. 3 of the House Judiciary Committee, 92d Cong., 2d Sess., pt. VII-A, p. 451 (1972) (testimony of Dr. Willard Gaylin: “I have seen black men listed as white and Harvard graduates fisted with borderline IQ’s”); S. Singer & D. Gottfredson, Development of a Data Base for Parole Decision-Making 2-5 (NCCD Research Center, Supp. Report 1, 1973) (information provided by FBI often fists same charge six or seven times without showing a final disposition).

See, e. g., S. Rep. No. 94-369, p. 19 (1975) (“It is essential, then, that parole has both the fact and appearance of fairness to all. Nothing less is necessary for the maintenance of the integrity of our criminal justice institutions”); United States ex rel. Johnson v. Chairman of New York State Board of Parole, 500 F. 2d, at 928 ; Phillips v. Williams, 583 P. 2d 488, 490 (Okla. 1978), cert. pending, No. 78-1282; ABA, Standards Relating to the Legal Status of Prisoners (Tent. Draft 1977), in 14 Am. Crim. L. Rev. 377 , 598 (1977); K. Davis, Discretionary Justice: A Preliminary Inquiry 126-133 (1969); Official Report of the New York State Special Commission on Attica 97, 98 (Bantam ed. 1972).

In accordance with the majority opinion, ante, at 16 n. 8, I do not address whether the Court of Appeals was correct in holding that the Nebraska Parole Board may not abandon the procedures it already provides. These safeguards include permitting inmates to appear and present documentary support at hearings, and providing a statement of reasons when parole is denied or deferred. Because the inmates failed to seek review of the Court of Appeals’ decision, I also express no view on whether it correctly held that the Board’s practice of allowing inmates to present witnesses and retain counsel for final parole hearings was not constitutionally compelled. Finally, it would be inappropriate to consider the suggestion advanced here for the first time that inmates should be allowed access to their files in order to correct factual inaccuracies. Cf. ante, at 15 n. 7.

Nevertheless, the range of protections currently afforded does affect whether additional procedures are constitutionally compelled. The specific dictates of due process, of course, depend on what a particular situation *36 demands. See Cafeteria & Restaurant Workers v. McElroy, 367 U. S. 886, 895 (1961). Nebraska’s use of formal hearings when the possibility of granting parole is substantial and informal hearings in other cases, for example, combined with provision of a statement of reasons for adverse decisions, obviously reduces the need for supplemental procedures.

Although a formal hearing at the point of initial eligibility would reduce the risk of error and enhance the appearance of fairness, providing. a summary of essential evidence and reasons, see n. 25, infra, together with allowing inmates to appear at informal hearings, decreases the justification for requiring the Board to conduct formal hearings in every case. See n. 17, supra.

The courts below found that 72 hours’ advance notice ordinarily would enable prisoners to prepare for their appearances. 576 F. 2d, at 1283. The Court of Appeals further determined that the statutory criteria were sufficiently specific that the Board need only include a list of those criteria with the hearing notices or post such a list in public areas throughout the institution. Ibid.

Every other Court of Appeals holding the Due Process Clause applicable to parole release proceedings has also concluded that the parole board must advise the inmates in writing of the reasons for denying parole. See Franklin v. Shields, 569 F. 2d, at 800-801 (en banc); United States ex rel. Richerson v. Wolff, 525 F. 2d 797 (CA7 1975), cert. denied, 425 U. S. 914 (1976); Childs v. United States Board of Parole, 167 U. S. App. D. C. 268, 511 F. 2d 1270 (1974); United States ex rel. Johnson v. Chairman of New York State Board of Parole, 500 F. 2d 925 (CA2), vacated as moot, 419 U. S. 1015 (1974). The parties to Franklin v. Shields did not request that the Parole Board also be required to provide a summary of the essential facts, see 569 F. 2d, at 787, 797, and the Fourth Circuit did not address the issue. The Second Circuit in Johnson expressly held that the statement of reasons must be supplemented by a summary of the “essential facts upon which the Board’s inferences are based.” 500 F. 2d, at 934 . Richerson and Childs also indicated that the notice of reasons should include a description of the crucial facts. See 525 F. 2d, at 804 ; 511 F. 2d, at 1281-1284 , aff’g 371 F. Supp. 1246, 1247 (1973).

Contrary to its supposition here, in Wolff v. McDonnell, 418 U. S., at 565 , the Court could perceive no “prospect of prison disruption that can flow from the requirement of these statements.”

See Hirschkop & Millemann, The Unconstitutionality of Prison Life, 55 Va. L. Rev. 795 , 811-812, 839 (1969).

The preprinted list of reasons for denying parole is unlikely to disclose these types of factual errors. Out of 375 inmates denied parole during a 6-month period, the only reason given 285 of them was: “Your continued correctional treatment, vocational, educational, or job assignment in the facility will substantially enhance your capacity to lead a law-abiding life when released at a later date.” App. 40-42. Although the denial forms also include a list of six “[r]ecommendations for correcting deficiencies,” such as “[e]xhibit some responsibility and maturity,” the evidence at trial showed that all six items were checked on 370 of the 375 forms, regardless of the facts of the particular case. App. 42; Tr. 38-39, 45-46.

See, e. g., cases cited in n. 20, supra; Candarini v. Attorney General of United States, 369 F. Supp. 1132, 1137 (EDNY 1974); Monks v. New Jersey State Parole Board, 58 N. J. 238, 249, 277 A. 2d 193, 199 (1971); K. Davis, Discretionary Justice: A Preliminary Inquiry 126-133 (1969); M. Frankel, Criminal Sentences 40-41 (1972); Dawson, The Decision to Grant or Deny Parole: A Study of Parole Criteria in Law and Practice, 1966 Wash. U. L. Q. 243, 302; Comment, 6 St. Mary’s L. J. 478, 487 (1974).

This statement of reasons and the summary of essential evidence should be provided to all inmates actually eligible for parole, whether the adverse decision is rendered following an initial review or a final parole hearing.