Moreland v. Federal Bureau of Prisons

363 F.Supp.2d 882 (2005)

Ellen Jeanette MORELAND, Petitioner,
v.
THE FEDERAL BUREAU OF PRISONS, Respondent.

No. CIV.A. H-04-3658.

United States District Court, S.D. Texas, Houston Division.

April 1, 2005.

*883 *884 Brent Evan Newton, Asst. Federal Public Defender, Houston, TX, for Petitioner.

James L. Turner, U.S. Attorney's Office, Houston, TX, for Respondents.

Order

HUGHES, District Judge.

The memorandum and recommendation entered March 30, 2005, by Magistrate Judge Stephen Wm. Smith is adopted as this court's opinion.

MEMORANDUM AND RECOMMENDATION

SMITH, United States Magistrate Judge.

This is a petition for habeas corpus under 28 U.S.C. § 2241, which has been referred to this magistrate judge for report and recommendation. (Dkt.3). The case involves the proper application of good conduct time (GCT) to the sentence of a federal prisoner under 18 U.S.C. § 3624(b).

Background

On February 14, 1990, petitioner Ellen Jeanette Moreland committed two drug trafficking offenses in Milwaukee, Wisconsin, for which she was arrested and then convicted on January 21, 1991. The United States District Court for the Eastern District of Wisconsin sentenced her to 210 months in prison, and she has been confined to the Federal Prison Camp in Bryan, Texas, for most of that term. Moreland has actually been in federal custody since August 29, 1990, and has received 157 days jail credit against her sentence, which is not at issue here.

Moreland has apparently been a model prisoner, earning the maximum number of good conduct time credits each year under 18 U.S.C. § 3624(b). Moreland contends that she is entitled to a GCT of 54 days for each year of the sentence imposed. She was sentenced to 17.5 years. The Bureau *885 of Prisons disagrees, declaring that the credit is 54 days for each year of the sentence actually served. The Bureau's method of calculation, which requires numerous pages of mathematical explanation, yields 47 days of GCT each year of the sentence, seven days less than Moreland's method. The difference in calculation has a significant impact upon Moreland's expected release date. According to the Bureau, Moreland's maximum GCT is 810 days,[1] resulting in a projected release date of November 17, 2005. Moreland calculates her maximum GCT credit to be 945 days, resulting in her release on July 18, 2005, some four months sooner.

Resolution of this dispute hinges entirely upon a matter of statutory interpretation: does section 3624(b) award good conduct time credit based on the sentence imposed, or on time actually served?

Analysis

As a preliminary matter, the Bureau contends that Moreland has failed to exhaust her administrative remedies.[2] This contention has no merit, because "the exhaustion requirements of 42 U.S.C. § 1997e(a) do not apply to a properly filed section 2241 petition." Mayberry v. Pettiford, 74 Fed.Appx. 299, 299 (5th Cir.2003) (unpublished); see also Davis v. Fechtel, 150 F.3d 486, 487 (5th Cir.1998) (finding the Prison Litigation Reform Act does not apply to section 2241 petitions). While there is a judicially created exhaustion requirement with respect to section 2241, exceptions to the exhaustion requirement are appropriate where the attempt to exhaust such remedies would be patently futile. See Fuller v. Rich, 11 F.3d 61, 62 (5th Cir.1994). The Bureau has adopted a uniform policy to calculate the amount of good time awarded per year under 18 U.S.C. § 3624(b)(1), expressed in 28 C.F.R. § 523.20. In light of the Bureau's established policy, it would be patently futile to require Moreland to continue seeking an administrative remedy. Thus, the court will review the merits of Moreland's petition.

1. Text of the Statute

Title 18 U.S.C. § 3624(b) reads:

(b) Credit toward service of sentence for satisfactory behavior. —

(1) Subject to paragraph (2), a prisoner who is serving a term of imprisonment of more than 1 year other than a term of imprisonment for the duration of the prisoner's life, may receive credit toward the service of the prisoner's sentence, beyond the time served, of up to 54 days at the end of each year of the prisoner's term of imprisonment, beginning at the end of the first year of the term, subject to determination by the Bureau of Prisons that, during that year, the prisoner has displayed exemplary compliance with institutional disciplinary regulations.

18 U.S.C. § 3624(b)(1) (emphasis supplied).

The key phrase in the italicized passage is "term of imprisonment," which is not defined by the statute. Moreland argues this phrase means "sentenced imposed," so that the maximum credit she could earn for good conduct is 54 days for each of the 17.5 years of her sentence. On the other hand, the government contends this phrase *886 means "time served," which, after a complex mathematical computation requiring dozens of pages of explanation in the Bureau's Sentence Computation Manual yields a maximum of 47 days for each year of the sentence.

When construing a statute, a court must consider the statute as a whole. Crandon v. United States, 494 U.S. 152, 158, 110 S.Ct. 997, 108 L.Ed.2d 132 (1990). A fundamental canon of statutory construction is that identical terms within an act should be given the same meaning. Sorenson v. Sec'y of Treasury, 475 U.S. 851, 860, 106 S.Ct. 1600, 89 L.Ed.2d 855 (1986). This rule of statutory consistency is at its "most vigorous" when a term is repeated within a given sentence. Brown v. Gardner, 513 U.S. 115, 118, 115 S.Ct. 552, 130 L.Ed.2d 462 (1994).

The phrase "term of imprisonment" appears three times in the first sentence of section 3624(b)(1). On the first two occasions, the phrase undoubtedly means sentence imposed, as several courts have observed. See, e.g., White v. Scibana, 390 F.3d 997, 1001 (7th Cir.2004); Esposito v. Ashcroft, 2005 WL 119872, at *4 (N.D.W.Va.2005); Williams v. Dewalt, 351 F.Supp.2d 412, 416 (D.Md.2004). There is no evidence that Congress intended a different meaning the third time around. On the contrary, Congress used the phrase "time served" elsewhere within this same contested sentence, fully demonstrating its drafting ability to distinguish between the two terms when it chose to do so. See, e.g., Williams, 351 F.Supp.2d at 417 ("In drafting § 3624(b), Congress used the phrase `time served' when it meant time served"). The Bureau's interpretation not only renders the third instance of the phrase "term of imprisonment" inconsistent with its plain meaning in the two previous appearances, but also transforms its meaning to that of a contrary phrase within the same sentence. Sloppy draftsmanship is not difficult to find in the U.S. Code,[3] but the Bureau's interpretation plumbs new depths of linguistic confusion.

To be sure, broadening the focus to the rest of section 3624 does reveal a single instance in which "term of imprisonment" is apparently intended to denote "time served."[4] Against this lone example, three other instances may be cited in which the phrase in question can only mean "sentence imposed."[5] While the Bureau may be correct that Congress was not perfectly consistent in other parts of section 3624, the canon of statutory consistency favors Moreland's interpretation, at least insofar as subsection (b) is concerned.

The Bureau points to other language in the contested sentence to support its position: "up to 54 days at the end of each year of the prisoner's term of imprisonment." 18 U.S.C. § 3624(b)(1) (emphasis supplied). According to the Bureau, this language can only mean that the GCT *887 must be applied after the end of each year. But this argument is linguistically unsound, because it incorrectly interprets "at" to mean "after." No support for this substitution has been found in any standard dictionary. On the contrary, Webster's first definition of "at" is: "1. on; in; near; by; as, at the office, at heart." Webster's New Universal Unabridged Dictionary 17 (2d ed.1983). In fact, the same dictionary explicitly equates the two phrases "at the end" and "in the end." Id. at 599. Substituting the listed prepositions in, near, or by in place of at in the contested passage clearly conveys the meaning that GCT is to be applied during the last part of the year, not after the year is over.

Normal usage readily demonstrates that the prepositions "at" and "after" are not fungible, and generate quite different meanings when used in a temporal setting. King Lear dies at the end of the play, not after the play. The fat lady sings at the end of the opera, not after it. The two-minute warning occurs at the end of the game, not after it. Halloween comes at the end of October, not after. This common usage of "at" extends to non-temporal settings as well: the barb is at the end of the hook; the tail is at the end of the dog; the filter is at the end of the cigarette; the caboose is at the end of the train. In each of these examples, the first item mentioned is included within the object referenced, not beyond or outside it.

An analogy based on a familiar childhood experience illustrates the point. A child is told that if she is "nice" she will be rewarded with Christmas presents at the end of the year. Christmas morning comes, the child has been nice, but no presents are under the tree. The child's parents (doubtless BOP employees) tell the distraught child that Christmas really comes after the end of the year, and so the presents are not due until January. In like fashion, the Bureau's position can fairly be dubbed the "Christmas-in-January" approach to GCT.

Moreland interprets the phrase "at the end of each year" to mean that good conduct time credit is included within the last part of the year, like so:

311 days actually served + 54 days GCT = one year of the sentence imposed.

Like Christmas in December, this "inclusive year" approach best conforms to our ordinary understanding of these words. The Bureau's interpretation conflicts with plain meaning, normal usage, and, as we have seen, accepted canons of statutory construction.

2. Legislative History

A review of the legislative history also shows that Moreland's interpretation of section 3624(b) better accords with congressional design than the Bureau's. Section 3624(b) was enacted in 1984 as part of the Sentencing Reform Act. Pub.L. No. 98-473, Title II, § 212(a)(2). The predecessor statute, codified at 18 U.S.C. § 4161, credited "good time" based upon the sentence imposed. Indeed, for most of the twentieth century, "good conduct time was computed by multiplying the number of months of a sentence as imposed by the court by the appropriate number of days as prescribed in the statute. The resultant total was then credited to the account of the prisoner and a tentative release date established." H.R. Rep. 86-935, reprinted in 1959 U.S.C.C.A.N. 2518, at 2519-20.

Congress changed the "good time" statute in 1984 to achieve two main objectives: (1) to award a uniform potential of 15 percent per year as an incentive for good conduct to replace the confusing multi-tiered system; and (2) to create a simpler, easier to understand system. As the Senate committee report explained: "Computation of credit toward early release pursuant *888 to section 3624(b) will be considerably less complicated than under current law in many respects. Current law provides a different rate of credit for good behavior for different lengths of prison terms, while section 3624(b) provides a uniform maximum rate of 36 days a year [later amended to 54] for all time in prison beyond the first year." S. Rep. 98-224, reprinted in 1984 U.S.C.C.A.N. 3182, at 3329-30. Congress decided to replace this complicated, multi-tiered "good time" system with one calculated at a uniform rate more readily comprehended by prisoners. "The result of the complexity of current law provisions concerning good time allowances is to increase the uncertainty of the prisoner as to his release date, with a resulting adverse effect on prisoner morale." Id. at 3330. The report continued:

It is the belief of the committee that the simplified provisions of section 3624(b) will have a positive effect on prisoner behavior. The credit toward early release is earned at a steady and easily determined rate that will have an obvious impact on the prisoner's release date. The rate is sufficiently high (approximately 10 percent of the part of a term of imprisonment that exceeds one year [later amended to 15 percent]) to provide an incentive for good institutional behavior, yet not so high that it will carry forward the uncertainties as to release dates that occur under current law.

Id. Congress assumed that section 3624(b) would rectify the complications of the old "good time" law: "The new provisions will also be easier (and probably cheaper) to administer than those under current law." Id.

While not conclusive in themselves, specific references in the legislative history reinforce Moreland's interpretation that Congress intended to award a potential credit of 15 percent per year of the sentence imposed. For instance, in amendment 130 of House Conference Report 98-1159, Congress "increas[ed] `good time' that accrues from 10 percent to 15 percent." H.R. Conf. Rep. 98-1159, reprinted in 1984 U.S.C.C.A.N. 3710, at 3711 (Oct. 10, 1984). And a 1986 committee report on related federal sentencing legislation summarized the good time provisions of the 1984 Act this way: "A defendant can reduce a prison term by no more than 15 percent." H.R. Rep. 99-614, reprinted in 1986 U.S.C.C.A.N. 1762, at 1763 (May 28, 1986). In the footnote to that sentence, the committee report elaborated that "18 U.S.C. 3624(b) as enacted by the Sentencing Reform Act ... provides that a prisoner can earn up to 54 days per year-15 percent-for good behavior.... [U]nder present law, a prisoner can reduce a prison term by about one-third for good behavior." 1986 U.S.C.C.A.N. 1762, at 1763 n. 4 (emphasis and underscoring supplied). The views of a subsequent congressional committee on the meaning of a statute are entitled to some weight. See Seatrain Shipbuilding Corp. v. Shell Oil Co., 444 U.S. 572, 596, 100 S.Ct. 800, 63 L.Ed.2d 36 (1980).

Although entitled to less weight, a similar understanding of the statute was later expressed by Senator Joseph Biden, co-sponsor of the legislation: "I was coauthor of that bill. In the Federal courts, if a judge says you are going to go to prison for 10 years, you know you are going to go to prison for at least 85 percent of that time-8.5 years, which is what the law mandates. You can get up to 1.5 years in good time credits, but that is all." 141 Cong. Rec. S.2348-01, at S.2349, 1995 WL 50344 (Feb. 9, 1995).

The Bureau's "time served" interpretation of section 3624(b) is inconsistent with congressional objectives in two ways. First, the Bureau's Program Statement 5880.028, which implements the "time *889 served" policy, is well-over 200 pages long, contains cumbersome and confusing formulas that even the Bureau describes as "arithmetically complicated,"[6] and which few, if any, prisoners could ever be expected to decipher. Second, nothing in the legislative history suggests that Congress intended to abandon the long-standing practice of awarding "good conduct time" credit based on the sentence imposed. Congress had specifically rejected using "time served" as the basis for determining "good time" in 1959. As related by the court in Perez-Olivo v. Chavez, 394 F.3d 45, 50 (1st Cir.2005):

The legislative history of the repealed GCT statute reveals a clear congressional intent to calculate GCT based on the "sentence imposed" rather than the "time served." See, e.g., H. Rep. 86-935 (1959), reprinted in 1959 U.S.C.C.A.N. 2518, 2518-19 (discussing 1959 amendment to § 4161 intended in part to reverse a 1952 court decision [Hunter v. Facchine, 195 F.2d 1007 (10th Cir.1952)] interpreting the statute as requiring GCT to be calculated based on time served rather than sentence imposed).

At that time, Congress determined that basing credit on "time served" would require well-behaved prisoners to serve longer periods of confinement than they otherwise should. See H.R. Rep. 86-935, reprinted in 1959 U.S.C.C.A.N. 2518, at 2519 (Aug. 18, 1959). No evidence supports the inference that Congress had changed its mind about this aspect of good time credit in 1984.

3. Rule of Lenity

Moreland's "sentence imposed" interpretation is further bolstered by the rule of lenity. This venerable rule, which traces it origin to the days of the Marshall court, compels a court to resolve ambiguity in a criminal statute in favor of the defendant. See United States v. R.L.C., 503 U.S. 291, 305, 112 S.Ct. 1329, 117 L.Ed.2d 559 (1992); United States v. Wiltberger, 5 Wheat. 76, 18 U.S. 76, 95, 5 L.Ed. 37 (1820). This rule serves two important policies: (1) to give fair warning to the world, "in language that the common world will understand, of what the law intends to do if a certain line is passed;" and (2) to assure that, because of the seriousness of criminal penalties, legislatures and not courts define criminal activity. United States v. Bass, 404 U.S. 336, 348, 92 S.Ct. 515, 30 L.Ed.2d 488 (1971). The rule embodies "`the instinctive distaste against men languishing in prison unless the lawmaker has clearly said they should.'" Id. (quoting H. Friendly. Mr. Justice Frankfurter and the Reading of Statutes, in Benchmarks 196, 209 (1967)). The rule has been associated with the criminal law for so long that the current edition of Black's Law Dictionary places a 19th century summary of the rule immediately after its definition of "penal statute."[7]

It is settled that the rule of lenity applies not only to the substantive scope of criminal prohibitions, but also to *890 questions about the severity of sentencing. Bifulco v. United States, 447 U.S. 381, 387, 100 S.Ct. 2247, 65 L.Ed.2d 205 (1980); see generally Phillip M. Spector, The Sentencing Rule of Lenity, 33 U. TOL. L. REV. 511, 513 (Spring 2002) (nearly half of all recent cases in which the Supreme Court has invoked the rule of lenity have been sentencing cases); accord Leocal v. Ashcroft, ___ U.S. ___, ___ n. 8, 125 S.Ct. 377, 384 n. 8, 160 L.Ed.2d 271 (2004) (the rule of lenity applies where a statute has both criminal and noncriminal applications); United States v. Thompson/Center Arms Co., 504 U.S. 505, 518 n. 10, 112 S.Ct. 2102, 119 L.Ed.2d 308 (1992) (same). The Bureau rightly does not contest the proposition that section 3624(b) is a penal statute. See Dkt. 14, at 6. Unquestionably, then, any ambiguity in the good time statute should generally be resolved in favor of the federal prisoner.

Although the rule of lenity is firmly entrenched as a substantive rule of statutory interpretation, there are limits upon its use. For example, the rule cannot be used to create ambiguity in the face of otherwise clear statutory language. See Callanan v. United States, 364 U.S. 587, 596, 81 S.Ct. 321, 5 L.Ed.2d 312 (1961) (rule "only serves as an aid for resolving ambiguity; it is not to be used to beget one"). Nor is the rule to be invoked by a mere grammatical possibility. See Lopez v. Davis, 531 U.S. 230, 244 n. 7, 121 S.Ct. 714, 148 L.Ed.2d 635 (2001) (citing Caron v. United States, 524 U.S. 308, 316, 118 S.Ct. 2007, 141 L.Ed.2d 303 (1998) (rule does not apply "if the ambiguous reading relied on is an implausible reading of congressional purpose")). It is sometimes said that lenity applies only in situations where the court can make "no more than a guess as to what Congress intended." Reno v. Koray, 515 U.S. 50, 65, 115 S.Ct. 2021, 132 L.Ed.2d 46 (1995) (citation omitted). But the Supreme Court has often invoked the rule to reject a possible but implausible interpretation of a criminal statute unfavorable to criminal defendants. See, e.g., Scheidler v. NOW, Inc., 537 U.S. 393, 408-09, 123 S.Ct. 1057, 154 L.Ed.2d 991 (2003) (rule applied to support narrow construction against a contrary line of precedent calling for more expansive interpretation of Hobbs Act); see generally Federal Statutes and Regulations, 117 HARV. L. REV. 459, 459-69 (Nov.2003). The Fifth Circuit has applied the rule even when the government's interpretation of an ambiguous criminal statute is "not unreasonable." United States v. Phipps, 319 F.3d 177, 187 (5th Cir.2003).

As demonstrated above, the Bureau's "time served" interpretation is a possible but implausible construction of section 3624(b). To the extent there remains any ambiguity in the statute after considering its most natural linguistic meaning and the legislative history, the rule of lenity eliminates all doubt: good conduct time must be based on the sentence imposed, rather than time served.

4. Chevron Considerations

The Bureau argues that, as the agency responsible for administering the good time statute, its interpretation is entitled to deference under Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). There are several difficulties with this argument.

As a general matter, many courts have questioned whether deference is owed to the executive branch's interpretation of a criminal statute. For example, in Dolfi v. Pontesso, 156 F.3d 696, 699-700 (6th Cir.1998), the court refused deference to the U.S. Parole Commission's interpretation of the phrase "special parole term" in 21 U.S.C. § 841(c), relying heavily upon Justice Scalia's concurring opinion in Crandon v. United States, 494 U.S. 152, 177, 110 *891 S.Ct. 997, 108 L.Ed.2d 132 (1990) ("[W]e have never thought that the interpretation of those charged with prosecuting criminal statutes is entitled to deference. Besides being unentitled to what might be called ex officio deference under Chevron, [the government's] expansive administrative interpretation of § 209(a) is not even deserving of any persuasive effect"). Similarly, the Seventh Circuit denied deference to the Parole Commission's interpretation of the same criminal statute in Evans v. United States Parole Comm'n, 78 F.3d 262, 265 (7th Cir.1996), declaring that "we have substantial doubt that the Judicial Branch owes any deference to the Executive Branch when the question concerns the maximum term of imprisonment; certainly judges do not defer to the Attorney General's interpretation of Title 18." Id.

The Bureau's authority to issue regulations under the good time statute is derived not from Congress, but rather from authority delegated by the Attorney General. See 28 C.F.R. § 0.96(s). Deference to such regulations would appear to raise a substantial separation of powers issue in light of the Crandon concurrence.

A second difficulty with Chevron deference here is that, at the time Moreland committed her offenses in 1990, the Bureau's "time served" interpretation was embodied only in a series of memos issued by the Bureau's general counsel Claire Cripe. See Dkt. 14, at 4-5. Two years later the Bureau issued internal agency guidelines in the form of a "Program Statement," which contained the "time served" interpretation of section 3624(b). U.S. Dep't of Justice, Bureau of Prisons Program Statement 5880.028, Sentence Computation Manual (promulgated on February 21, 1992 (Dkt. 14, at 2)). And the Bureau did not formally adopt any regulation on the subject until 1997. See 28 C.F.R. § 523.20 (promulgated in 62 Fed.Reg. 50, 786 (Sept. 26, 1997)). Accordingly, Moreland's sentence, including good time credit, is governed by the general counsel memoranda in effect when she committed the crimes in 1990. Cf. Weaver v. Graham, 450 U.S. 24, 29-33, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981) (state law adversely changing good time system violates ex post facto clause by substantially altering the consequences attached to a crime already committed).

Opinions by agency counsel do not merit full Chevron deference. Christensen v. Harris County, 529 U.S. 576, 587, 120 S.Ct. 1655, 146 L.Ed.2d 621 (2000). Interpretations contained in formats such as opinion letters or policy statements may sometimes be given deference, but only to the extent that those interpretations have the "power to persuade." Id.; Bolen v. Dengel, 340 F.3d 300, 310 (5th Cir.2003). By that measure, the Cripe memoranda are unsatisfactory; they offer no justification for adopting the "time served" standard, nor do they even acknowledge that a competing and more natural interpretation of the statutory text was possible.

But even if these difficulties could be reconciled in the Bureau's favor, an insurmountable hurdle remains. Under Chevron's familiar two-step test, a court must first examine the statute for ambiguity using the "traditional tools of statutory construction." Chevron, 467 U.S. at 842, 104 S.Ct. 2778. If ambiguity remains after applying such rules of construction, then a court defers to an agency's reasonable interpretation of the statute. See General Dynamics Land Sys., Inc. v. Cline, 540 U.S. 581, 600, 124 S.Ct. 1236, 157 L.Ed.2d 1094 (2004) (Chevron deference applies only if other "devices of judicial construction have been tried and found to yield no clear sense of congressional intent"); Mississippi Poultry Ass'n v. Madigan, 31 F.3d 293, 299 (5th Cir.1994) ("If, but only *892 if, the language of the statute is determined to be either ambiguous or silent on the particular issue is the reviewing court to proceed to the second Chevron inquiry"). The rule of lenity is unquestionably a traditional rule of statutory construction. United States v. Thompson/Center Arms Co., 504 U.S. 505, 518 n. 10, 112 S.Ct. 2102, 119 L.Ed.2d 308 (1992) ("The rule of lenity ... is a rule of statutory construction whose purpose is to help give authoritative meaning to statutory language").

As we have seen, the normal tools of statutory construction, including the rule of lenity, eliminate any possible ambiguity in the text of section 3624(b). For Chevron purposes, "that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress." 467 U.S. at 842-43, 104 S.Ct. 2778. Deference to the Bureau's construction of section 3624(b) is, therefore, unwarranted.

5. Other Cases Distinguished

The court recognizes that the "sentence imposed" interpretation of the good time statute runs counter to the great majority of reported decisions.[8] Indeed, only two district courts have so far rejected the Bureau's interpretation of section 3624(b), and one of those was reversed on appeal. White v. Scibana, 314 F.Supp.2d 834 (W.D.Wis.2004), rev'd, White v. Scibana, 390 F.3d 997 (7th Cir.2004). Appellate review of the other-Williams v. Dewalt, 351 F.Supp.2d 412 (D.Md.2004)-is pending. While the Fifth Circuit has not yet issued a published opinion on the matter,[9] a decent respect to the opinions of other jurisdictions requires further explanation for separating from their distinguished company.

First, some of those decisions hold the rule of lenity inapplicable on the grounds that 18 U.S.C. § 3624(b) is not a criminal statute. See, e.g., Perez-Olivo v. Chavez, 394 F.3d 45, 53-54 (1st Cir.2005); Harvey v. Joslin, 2004 WL 2645550, at *3 (N.D.Tex.2004); Martinez v. Wendt, 2003 WL 22456808, at *2 (N.D.Tex.2003). Here, the Bureau has candidly conceded that the statute is indeed a penal statute, and as noted above, this concession is fully justified. The "good time" statute was part of the Comprehensive Crime Control Act of 1984. It is found in Title 18 of the United States Code, which is entitled Crimes and Criminal Procedure. Its subject is good time credit toward service of a sentence, which is the final factor determining the duration of punishment ultimately meted out for criminal behavior. And, as the Bureau observes, section 3624(b) fits comfortably within the standard definition of penal statute as "A law that defines an offense and prescribes its corresponding fine, penalty, or punishment." *893 BLACK'S LAW DICTIONARY 1449 (8th ed.2004).

The second distinguishing feature here is that, unlike most reported cases, 28 C.F.R. § 523.20 does not properly govern the petitioner's GCT calculation. Moreland was sentenced in January 1991, for crimes committed in February 1990. The Bureau did not implement this regulation via the Federal Register notice-and-comment procedure until 1997, more than seven years after the subject offenses. In Pacheco-Camacho, the Ninth Circuit rejected the rule of lenity because the Bureau had resolved the statutory ambiguity through a "valid regulation" which "gives the public sufficient warning to ensure that nobody mistakes the ambit of the law or its penalties." 272 F.3d at 1272. When Moreland committed her offenses, the Bureau's policy on calculating GCT was not embodied in a valid regulation, but only in memoranda written by the Bureau's then general counsel Claire Cripe. Agency counsel opinions contained in internal memos hardly constitute the sort of "fair warning" contemplated by the rule of lenity.

The final point of departure is really a matter of bookkeeping. A surprisingly large number of courts have expressed concern that the inclusive year interpretation gives a "windfall" to the prisoner, because she could be awarded GCT for a year during which she served no time. See, e.g., Pacheco-Camacho v. Hood, 272 F.3d 1266, 1268 (9th Cir.2001); Germany v. Smith, 2005 WL 428585, at *5 (M.D.Pa.2005). A typical expression of this concern is found in Graves v. Bledsoe, 334 F.Supp.2d 906, 908 (W.D.Va.2004):

Under Grave's interpretation, an inmate serving a ten-year sentence would be credited with 540 days of GCT, even though, due to the GCT, the inmate would not serve the entire ten-year sentence. Therefore, the inmate would receive GCT for time in advance of the year for which the GCT was to be earned. Essentially, during his ninth year of incarceration, the BOP would have to grant the inmate 54 days of GCT for his tenth year of imprisonment, even though he would not serve any of this tenth year of imprisonment. Logically, this conflicts with the statute's plain language, which requires the BOP to credit the inmate only "at the end of each year."

In other words, these courts perceive a logical conflict between the inclusive year interpretation and the retrospective yearly evaluation of a prisoner's conduct contemplated by the statute.

But this conflict disappears upon a proper accounting for the GCT. Moreland's position is that she should complete her term of imprisonment by serving 85 percent of her sentence with good behavior, earning enough GCT to wipe out the remaining 15 percent of her term. The same is true with respect to each particular year of her sentence; upon serving 311 days of actual time, she earns 54 days credit, which immediately vests[10] and thus wipes out the remainder of that year. Effectively, then, Moreland's first year of imprisonment is complete at the close of the 311th day; her second year of imprisonment begins on the 312th day. Assuming she serves the next 311 days with good behavior, she would receive another 54 days credit, wiping out the remainder of that year. The second year of her sentence is thus complete at the end of the 622nd day, the third year begins on the 623rd, and so on until she has served 85 percent of her 17.5 year sentence.

*894 In other words, this is purely a matter of bookkeeping. The evaluation date (or "good time action date" in Bureau parlance) must be adjusted each year to take into account the GCT already earned and vested. When this is done, it is readily seen that the model prisoner has fully earned 54 days credit for each 311 days served, and no windfall is occurring in the last year of imprisonment. The real problem here is that the Bureau's current practice does not permit an annual adjustment of the "good time action date"; the Cripe memorandum decrees that the good time action date "remain the same for each full year in custody." Dkt. 14, at 3. Such inflexibility gives rise to an absurdity of its own-under the Bureau's scheme, a prisoner must actually serve the 54 GCT days in order to get credit for those very same days. This makes sense only if one believes that Christmas comes in January.

Conclusion

The Bureau's approach to the good time statute has almost nothing to recommend it. Besides generating additional expense due to longer prison terms, the Bureau's interpretation violates: (1) plain meaning and conventional usage, by translating "at" to mean "after;" (2) the canon of statutory consistency, by giving the same phrase different meanings within the same sentence; (3) the venerable rule of lenity, by construing a penal statute more harshly against the prisoner; and (4) congressional intent as reflected in the legislative history, by devising a more complicated good time system beyond the ready comprehension of inmates. True, the Bureau has consistently adhered to its view of good time credit over the years, but faithful misinterpretation of a statute over time does not alter its original meaning.

For these reasons, it is recommended that Moreland's petition for writ of habeas corpus be granted, and that the Bureau of Prisons be directed to release Moreland based on a calculation of good time credit consistent with this opinion.

March 30, 2005.

NOTES

[1] See Dkt. 9, at 3. This number should probably be closer to 823, even under the Bureau's "time served" interpretation of section 3624(b). See Dkt. 9, Ex. A.

[2] Moreland filed an administrative remedy request at the first, or institutional level, which was denied on July 27, 2004. See Dkt. 9, Ex. B. She then attempted to file at the second, or regional office level, on August 30, 2004, but this was rejected as untimely. See Dkt. 9, Ex. B.

[3] See, e.g., United States v. Granderson, 511 U.S. 39, 60, 114 S.Ct. 1259, 127 L.Ed.2d 611 (1994) (Scalia, J., concurring) (referring to 18 U.S.C. § 3565(a) as "this wretchedly drafted statute").

[4] See 18 U.S.C. § 3624(d) ("Upon the release of a prisoner on the expiration of the prisoner's term of imprisonment, the [Bureau shall furnish suitable clothing, an amount of money]").

[5] See 18 U.S.C. §§ 3624(a) ("A prisoner shall be released by the Bureau of Prisons on the date of the expiration of the prisoner's term of imprisonment, less any time credited toward the service of the prisoner's sentence as provided in subsection (b)"); 3624(b)(1) ("[A] prisoner who is serving a term of imprisonment of more than 1 year other than a term of imprisonment for the duration of the prisoner's life."); 3624(c) ("[The Bureau] shall ... assure that a prisoner serving a term of imprisonment spends a reasonable part [of the term in pre-release custody]").

[6] Program Statement 5880.028, at 1-44. This Program Statement may be viewed at http://www.bop.gov/policy/progstat/5880_028.pdf.

[7] It is a familiar and well-settled rule that penal statutes are to be construed strictly, and not extended by implications, intendments, analogies, or equitable considerations. Thus, an offense cannot be created or inferred by vague implications. And a court cannot create a penalty by construction, but must avoid it by construction unless it is brought within the letter and the necessary meaning of the act creating it.

Henry Campbell Black, Handbook on the Construction and Interpretation of the Laws 287 (1896), reprinted in BLACK'S LAW DICTIONARY 1449 (8th ed.2004).

[8] See Perez-Olivo v. Chavez, 394 F.3d 45 (1st Cir.2005); White v. Scibana, 390 F.3d 997 (7th Cir.2004); Pacheco-Camacho v. Hood, 272 F.3d 1266 (9th Cir.2001), cert. denied, 535 U.S. 1105, 122 S.Ct. 2313, 152 L.Ed.2d 1067 (2002); Carter v. Jeter, 2005 WL 65960 (N.D.Tex.2005); Sash v. Zenk, 344 F.Supp.2d 376 (E.D.N.Y.2004); Harvey v. Joslin, 2004 WL 2645550 (N.D.Tex.2004); Pollard v. Van Buren, 2004 WL 2645548 (N.D.Tex.2004); Graves v. Bledsoe, 334 F.Supp.2d 906 (W.D.Va.2004).

[9] The Fifth Circuit recently issued an unpublished opinion dismissing a challenge to the Bureau's GCT policy for lack of jurisdiction. Sample v. Morrison, 2005 WL 648291 (5th Cir.2005). Dicta in that opinion agreed with the reasoning of the Seventh and Ninth Circuits in White and Pacheco-Camacho. See id. at *3. However, the factual circumstances of Moreland's claim are materially different, and the Sample opinion does not address the legislative history, the rule of lenity, or the deference owed the Cripe memoranda. In any event, unpublished decisions are not binding precedent in this circuit. See Fifth Circuit Rule 47.5.4 [Sample has subsequently been published. — Editor's note.].

[10] See Dkt. 14, at 3.