This interlocutory appeal presents the issue of whether, under the particular circumstances of this case, the fair use defense to copyright infringement applies to the photocopying of articles in a scientific journal. This issue arises on the appeal of defendant Texaco Inc. from the July 23, 1992, order of the United States District Court for the Southern District of New York (Pierre N. Leval, Judge) holding, after a limited-issue bench trial, that the photocopying of eight articles from the Journal of Catalysis for use by one of Texaco’s researchers was not fair use. See American Geophysical Union v. Texaco Inc., 802 F.Supp. 1 (S.D.N.Y.1992). Though not for precisely the same reasons, we agree with the District Court’s conclusion that this particular copying was not fair use and therefore affirm.
Background
The District Court Proceedings. Plaintiffs American Geophysical Union and 82 other publishers of scientific and technical journals (the “publishers”) brought a class action claiming that Texaco’s unauthorized photocopying of articles from their journals constituted copyright infringement. Among other defenses, Texaco claimed that its copying was fair use under section 107 of the Copy*915right Act, 17 U.S.C. § 107 (1988). Since it appeared likely that the litigation could be resolved once the fair use defense was adjudicated, the parties agreed that an initial trial should be limited to whether Texaco’s copying was fair use, and further agreed that this issue would be submitted for decision on a written record.
Although Texaco employs 400 to 500 research scientists, of whom all or most presumably photocopy scientific journal articles to support their Texaco research, the parties stipulated — in order to spare the enormous expense of exploring the photocopying practices of each of them — that one scientist would be chosen at random as the representative of the entire group. The scientist chosen was Dr. Donald H. Chickering, II, a scientist at Texaco’s research center in Beacon, New York. For consideration at trial, the publishers selected from Chickering’s files photocopies of eight particular articles from the Journal of Catalysis.
In a comprehensive opinion, reported at 802 F.Supp. 1, the District Court considered the statutory fair use factors identified in section 107, weighed other equitable considerations, and held that Texaco’s photocopying, as represented by Chickering’s copying of these eight articles did not constitute fair use. The District Court certified its ruling for interlocutory appeal under 28 U.S.C. § 1292(b) (1988).
Essential Facts. Employing between 400 and 500 researchers nationwide, Texaco conducts considerable scientific research seeking to develop new products and technology primarily to improve its commercial performance in the petroleum industry. As part of its substantial expenditures in support of research activities at its Beacon facility, Texaco subscribes to many scientific and technical journals and maintains a sizable library with these materials. Among the periodicals that Texaco receives at its Beacon research facility is the Journal of Catalysis (“Catalysis ”), a monthly publication produced by Academic Press, Inc., a major publisher of scholarly journals and one of the plaintiffs in this litigation. Texaco had initially purchased one subscription to Catalysis for its Beacon facility, and increased its total subscriptions to two in 1983. Since 1988, Texaco has maintained three subscriptions to Catalysis.
Each issue of Catalysis contains articles, notes, and letters (collectively “articles”), ranging in length from two to twenty pages. All of the articles are received by the journal’s editors through unsolicited submission by various authors. Authors are informed that they must transfer the copyright in their writings to Academic Press if one of their articles is accepted for publication, and no form of money payment is ever provided to authors whose works are published. Academic Press typically owns the copyright for each individual article published in Catalysis, and every issue of the journal includes a general statement that no part of the publication is to be reproduced without permission from the copyright owner. The average monthly issue of Catalysis runs approximately 200 pages and comprises 20 to 25 articles.
Chickering, a chemical engineer at the Beacon research facility, has worked for Texaco since 1981 conducting research in the field of catalysis, which concerns changes in the rates of chemical reactions. To keep abreast of developments in his field, Chicker-ing must review works published in various scientific and technical journals related to his area of research. Texaco assists in this endeavor by having its library circulate current issues of relevant journals to Chickering when he places his name on the appropriate routing list.
The copies of the eight articles from Catalysis found in Chickering’s files that the parties have made the exclusive focus of the fair use trial were photocopied in their entirety by Chickering or by other Texaco employees at Chickering’s request. Chickering apparently believed that the material and data found within these articles would facilitate his current or future professional research. The evidence developed at trial indicated that Chickering did not generally use the Catalysis articles in his research immediately upon copying, but placed the photocopied articles in his files to have them available for later reference as needed. Chickering became aware of six of the photocopied articles when the original issues of Catalysis containing the articles were circulated to him. He learned *916of the other two articles upon seeing a reference to them in another published article. As it turned out, Chickering did not have occasion to make use of five of the articles that were copied.
Discussion
I. The Nature of the Dispute
The parties and many of the amici curiae have approached this case as if it concerns the broad issue of whether photocopying of scientific articles is fair use, or at least the only slightly more limited issue of whether photocopying of such articles is fair use when undertaken by a research scientist engaged in his own research. Such broad issues are not before us. Rather, we consider whether Texaco’s photocopying by 400 or 500 scientists, as represented by Chickering’s example, is a fair use. This includes the question whether such institutional, systematic copying increases the number of copies available to scientists while avoiding the necessity of paying for license fees or for additional subscriptions. We do not deal with the question of copying by an individual, for personal use in research or otherwise (not for resale), recognizing that under the fair use doctrine or the de minimis doctrine, such a practice by an individual might well not constitute an infringement. In other words, our opinion does not decide the case that would arise if Chickering were a professor or an independent scientist engaged in copying and creating files for independent research, as opposed to being employed by an institution in the pursuit of his research on the institution’s behalf.
Fair use is a doctrine the application of which always depends on consideration of the precise facts at hand, see Campbell v. Acuff-Rose Music, Inc., — U.S. -, -, 114 S.Ct. 1164, 1170, 127 L.Ed.2d 500 (1994); Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 549, 105 S.Ct. 2218, 2225, 85 L.Ed.2d 588 (1985); Wright v. Warner Books, Inc., 953 F.2d 731, 740 (2d Cir.1991); H.R.Rep. No. 1476, 94th Cong., 2d Sess. 65-66 (1976) (“no generally applicable definition [of fair use] is possible, and each case raising the question must be decided on its own facts”), and in this case the parties have helpfully circumscribed the scope of the issue to be decided by tendering for the District Court’s decision the facts concerning the copying of eight particular articles. Our concern is whether the copying of these eight articles, as representative of the systematic copying that Texaco encouraged, was properly determined not to be fair use. Thus, the many background details stressed by each side are of only limited relevance in resolving this specific case.1
A. Fair Use and Photocopying
We consider initially the doctrine of fair use and its application to photocopying of documents. Seeking “to motivate the creative activity of authors ... by the provision of a special reward,” Sony Corporation of America v. Universal City Studios, Inc., 464 U.S. 417, 429, 104 S.Ct. 774, 782, 78 L.Ed.2d 574 (1984), copyright law grants certain exclusive rights in original works to authors, see 17 U.S.C. §§ 102(a), 106, 201(a). However, the fair use doctrine “tempers the protection of copyright by allowing ... [the] use [of] a limited amount of copyrighted material under some circumstances.” Twin Peaks Productions, Inc. v. Publications International, Ltd., 996 F.2d 1366, 1373 (2d Cir.1993). Traditionally conceived as based on authors’ implied consent to reasonable uses of their works, see Harper & Row, 471 U.S. at 549-50, 105 S.Ct. at 2224-25, or on an exception to authors’ monopoly privileges needed in order to fulfill copyright’s purpose to promote the arts and sciences, see Campbell, — U.S. at -, 114 S.Ct. at 1169, the fair use doctrine has a lengthy and rich common-law history, see William F. Patry, The Fair Use Privilege in Copyright Law 1-63 (1985) [hereinafter Patry, The Fair Use *917Privilege ], and is now codified in section 107 of the Copyright Act, 17 U.S.C. § 107.2
As with the development of other easy and accessible means of mechanical reproduction of documents, the invention and widespread availability of photocopying technology threatens to disrupt the delicate balances established by the Copyright Act. See 3 Melville B. Nimmer & David Nimmer, Nim-mer on Copyright § 13.05[E][1], at 13-225 to 13-226 (1994) [hereinafter Nimmer on Copyright ] (noting that “unrestricted photocopying practices could largely undercut the entire law of copyright”); see also Sony, 464 U.S. at 467-68 n. 16,104 S.Ct. at 801-02 n. 16 (Blackmun, J., dissenting) (recognizing that the “advent of inexpensive and readily available copying machines ... has changed the dimensions” of the legal issues concerning the practice of making personal copies of copyrighted materials). As a leading commentator astutely notes, the advent of modern photocopying technology creates a pressing need for the law “to strike an appropriate balance between the authors’ interest in preserving the integrity of copyright, and the public’s right to enjoy the benefits that photocopying technology offers.” 3 Nimmer on Copyright § 13.05[E][1], at 13-226.
Indeed, if the issue were open, we would seriously question whether the fair use analysis that has developed with respect to works of authorship alleged to use portions of copyrighted material is precisely applicable to copies produced by mechanical means. The traditional fair use analysis, now codified in section 107, developed in an effort to adjust the competing interests of authors — the author of the original copyrighted work and the author of the secondary work that “copies” a portion of the original work in the course of producing what is claimed to be a new work. Mechanical “copying” of an entire document, made readily feasible and economical by the advent of xerography, see SCM Corp. v. Xerox Corp., 463 F.Supp. 983, 991-94 (D.Conn.1978), aff'd, 645 F.2d 1195 (2d Cir.1981), cert. denied, 455 U.S. 1016, 102 S.Ct. 1708, 72 L.Ed.2d 132 (1982), is obviously an activity entirely different from creating a work of authorship. Whatever social utility copying of this sort achieves, it is not concerned with creative authorship.
Though we have been instructed to defer to Congress “when major technological innovations alter the market for copyrighted materials,” Sony, 464 U.S. at 431, 104 S.Ct. at 783, Congress has thus far provided scant guidance for resolving fair use issues involving photocopying, legislating specifically only as to library copying, see 17 U.S.C. § 108, and providing indirect advice concerning classroom copying.3 See generally 3 Nim-mer on Copyright § 13.05[E]. However, we learn from the Supreme Court’s consideration of copying achieved by use of a videotape recorder that mechanical copying is to be assessed for fair use purposes under the traditional mode of analysis, including the four statutory factors of section 107. See Sony, 464 U.S. at 447-56, 104 S.Ct. at 791-96. We therefore are obliged to apply that analysis to the photocopying that occurred in this case.
B. The Precise Copyrights at Issue
We must first identify precisely the copyrighted works alleged to be infringed, *918since certain arguments made on appeal seem to focus on different works. The publishers typically hold two separate sets of copyrights in their journal publications. As a consequence of the publishers’ requirement that authors transfer their copyrights when their articles are accepted for publication, the publishers usually possess the copyrights that subsist in each individual article appearing within their journals.4 Moreover, to the extent that the compilation of a journal issue involves an original work of authorship, the publishers possess a distinct copyright in each journal issue as a collective work, see 17 U.S.C. § 103; see also 17 U.S.C. § 101 (defining “compilation” and “collective work”). See generally Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340, 356-61, 111 S.Ct. 1282, 1293-96, 113 L.Ed.2d 358 (1991) (discussing extent of copyright protection in compilations and collective works).
From the outset, this lawsuit concerned alleged infringement of the copyrights in individual journal articles, copyrights assigned by the authors to the publishers. More specifically, by virtue of the parties’ stipulation, this case now concerns the copyrights in the eight articles from Catalysis found in Chick-ering’s files, copyrights now owned by Academic Press. There are no allegations that raise questions concerning Academic Press’s potential copyrights in whole issues or annual volumes of Catalysis as collective works.
C. Burdens of Proof and Standard of Review
Fair use serves as an affirmative defense to a claim of copyright infringement, and thus the party claiming that its secondary use of the original copyrighted work constitutes a fair use typically carries the burden of proof as to all issues in the dispute. See Campbell, — U.S. at -, 114 S.Ct. at 1177. Moreover, since fair use is a “mixed question of law and fact,” Harper & Row, 471 U.S. at 560, 105 S.Ct. at 2230, we review the District Court’s conclusions on this issue de novo, though we accept its subsidiary findings of fact unless clearly erroneous, see Twin Peaks, 996 F.2d at 1374.
II. The Enumerated Fair Use Factors of Section 107
Section 107 of the Copyright Act identifies four non-exclusive factors that a court is to consider when making its fair use assessment, see 17 U.S.C. § 107(1) — (4). The District Court concluded that three of the four statutory factors favor the publishers. As detailed below, our analysis of certain statutory factors differs somewhat from that of the District Court, though we are in agreement on the ultimate determination. Our differences stem primarily from the fact that, unlike the District Court, we have had the benefit of the Supreme Court’s important decision in Campbell, decided after Judge Leval issued his opinion.
A. First Factor: Purpose and Character of Use
The first factor listed in section 107 is “the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes.” 17 U.S.C. § 107(1). Especially pertinent to an assessment of the first fair use factor are the precise circumstances under which copies of the eight Catalysis articles were made. After noticing six of these articles when the original copy of the journal issue containing each of them was circulated to him, Chickering had them photocopied, at least initially, for the same basic purpose that one would normally seek to obtain the original — to have it available on his shelf for ready reference if and when he needed to look at it. The library circulated one copy and invited all the researchers to make their own photocopies. It is a reasonable inference that the library staff wanted each journal issue moved around the building quickly and returned to the library so that it would be available for others to look at. Making copies enabled all researchers who might one day be interested in examining the contents of an article in the issue to have the article *919readily available in their own offices. In Chickering’s own words, the copies of the articles were made for “my personal convenience,” since it is “far more convenient to have access in my office to a photocopy of an article than to have to go to the library each time I wanted to refer to it.” Affidavit of Donald Chickering at 11 (submitted as direct trial testimony) [hereinafter Chickering testimony]. Significantly, Chickering did not even have occasion to use five of the photocopied articles at all, further revealing that the photocopies of the eight Catalysis articles were primarily made just for “future retrieval and reference.” Id.
It is true that photocopying these articles also served other purposes. The most favorable for Texaco is the purpose of enabling Chickering, if the need should arise, to go into the lab with pieces of paper that (a) were not as bulky as the entire issue or a bound volume of a year’s issues, and (b) presented no risk of damaging the original by exposure to chemicals. And these purposes might suffice to tilt the first fair use factor in favor of Texaco if these purposes were dominant. For example, if Chickering had asked the library to buy him a copy of the pertinent issue of Catalysis and had placed it on his shelf, and one day while reading it had noticed a chart, formula, or other material that he wanted to take right into the lab, it might be a fair use for him to make a photocopy, and use that copy in the lab (especially if he did not retain it and build up a mini-library of photocopied articles). This is the sort of “spontaneous” copying that is part of the test for permissible nonprofit classroom copying. See Agreement on Guidelines for Classroom Copying in Noh-For-Profit Educational Institutions, quoted in Patry, The Fair Use Privilege, at 308.5 But that is not what happened here as to the six items copied from the circulated issues.
As to the other two articles, the circumstances are not quite as clear, but they too appear more to serve the purpose of being additions to Chickering’s office “library” than to be spontaneous copying of a critical page that he was reading on his way to the lab. One was copied apparently when he saw a reference to it in another article, which was in an issue circulated to him. The most likely inference is that he decided that he ought to have copies of both items — again for placement on his shelf for later use if the need arose. The last article was copied, according to his affidavit, when he saw a reference to it “elsewhere.” Chickering testimony at 22. What is clear is that this item too was simply placed “on the shelf.” As he testified, “I kept a copy to refer to in case I became more involved in support effects research.” Id.
The photocopying of these eight Catalysis articles may be characterized as “archival”— i.e., done for the primary purpose of providing numerous Texaco scientists (for whom Chickering served as an example) each with his or her own personal copy of each article without Texaco’s having to purchase another original journal.6 The photocopying “merely *920‘supersede^] the objects’ of the original creation,” Campbell, — U.S. at -, 114 S.Ct. at 1171 (quoting Folsom v. Marsh, 9 F.Cas. 342, 348 (No. 4,901) (C.C.D.Mass.1841)), and tilts the first fair use factor against Texaco. We do not mean to suggest that no instance of archival copying would be fair use, but the first factor tilts against Texaco in this case because the making of copies to be placed on the shelf in Chickering’s office is part of a systematic process of encouraging employee researchers to copy articles so as to multiply available copies while avoiding payment.
Texaco criticizes three aspects of the District Court’s analysis of the first factor. Relying largely on the Supreme Court’s discussion of fair use in Sony, the District Court suggested that a secondary user will “win” this first factor by showing a “transformative (or productive) nonsuperseding use of the original, or [a] noncommercial use, generally for a socially beneficial or widely accepted purpose.” 802 F.Supp. at 12. The District Court then concluded that Texaco’s copying is “neither transformative nor noncommercial,” id. at 13: not transformative because Texaco “simply makes mechanical photocopies of the entirety of relevant articles” and the “primary aspect” of Texaco’s photocopying is to multiply copies, see id. at 13-15; and not noncommercial because, though it facilitates research, this research is conducted solely for commercial gain, see id. at 15-16.
Texaco asserts that the District Court mis-charaeterized the inquiry under the first factor and overlooked several relevant considerations. First, Texaco contends that the District Court inappropriately focussed on the character of the user rather than the nature of the use in labeling Texaco’s copying as commercial. Texaco claims that its status as a for-profit corporation has no bearing on the fair use analysis, and that its use should be considered noncommercial since it photocopied articles in order to aid Chickering’s research. Texaco emphasizes that “research” is explicitly listed in the preamble of section 107, a circumstance that Texaco contends should make its copying favored under the first factor and throughout the entire fair use analysis.7
Second, Texaco contends that the District Court put undue emphasis on whether its use was “transformative,” especially since the Supreme Court appears to have rejected the view that a use must be transformative or productive to be a fair use. See Sony, 464 U.S. at 455 n. 40, 104 S.Ct. at 795 n. 40 (“The distinction between ‘productive’ and ‘unproductive’ uses may be helpful in calibrating the balance [of interests], but it cannot be wholly determinative.”). Texaco asserts that the “transformative use” concept is valuable only to the extent that it focusses attention upon whether a second work unfairly competes with the original. Texaco states that in this case, where the photocopies it made were not sold or distributed in competition with the original, the nontransformative nature of its copying should not prevent a finding of fair use. Texaco also suggests that its use should be considered transforma-tive: photocopying the article separated it from a bulky journal, made it more amenable to markings, and provided a document that could be readily replaced if damaged in a laboratory, all of which “transformed” the original article into a form that better served Chickering’s research needs.
Finally, Texaco claims that it should prevail on the first factor because, as the District Court acknowledged, the type of photocopying it conducted is widespread and has long been considered reasonable and customary. Texaco stresses that some courts and commentators regard custom and common usage as integral to the fair use analysis. See, e.g., Williams & Wilkins Co. v. United States, 487 F.2d 1345, 1353-56, 203 Ct.Cl. 74 (1973), aff'd by equally divided Court, 420 *921U.S. 376, 95 S.Ct. 1344, 43 L.Ed.2d 264 (1975); Lloyd L. Weinreb, Fair’s Fair: A Comment on the Fair Use Doctrine, 103 Harv.L.Rev. 1137, 1140 (1990) [hereinafter Weinreb, Fair’s Fair], We consider these three lines of attack separately.
1. Commercial use. We generally agree with Texaco’s contention that the District Court placed undue emphasis on the fact that Texaco is a for-profit corporation conducting research primarily for commercial gain. Since many, if not most, secondary users seek at least some measure of commercial gain from their use, unduly emphasizing the commercial motivation of a copier will lead to an overly restrictive view of fair use. See Campbell, — U.S. at -, 114 S.Ct. at 1174; see also Maxtone-Graham v. Burtchaell, 803 F.2d 1253, 1262 (2d Cir.1986) (noting that if “commercial” nature of a secondary use is overemphasized in the analysis, “fair use would be virtually obliterated”), cert. denied, 481 U.S. 1059, 107 S.Ct. 2201, 95 L.Ed.2d 856 (1987). See generally 3 Nim-mer on Copyright § 13.05[A][l][c], at 13-162 to 13-163 (categorical rule against commercial uses unwarranted since this “would cause the fair use analysis to collapse in all but the exceptional case of nonprofit exploitation”). Though the Supreme Court had stated in Sony that every commercial use was “presumptively” unfair, see 464 U.S. at 451, 104 S.Ct. at 793, that Court and lower courts have come to explain that the commercial nature of a secondary use simply “ ‘tends to weigh against a finding of fair use.’ ” Campbell, — U.S. at -, 114 S.Ct. at 1174 (quoting Harper & Row, 471 U.S. at 562, 105 S.Ct. at 2231); accord Rogers v. Koons, 960 F.2d 301, 309 (2d Cir.), cert. denied, — U.S. -, 113 S.Ct. 365, 121 L.Ed.2d 278 (1992); Sega Enterprises Limited v. Accolade, Inc., 977 F.2d 1510, 1522 (9th Cir.1992); Maxtone-Graham, 803 F.2d at 1262.
Indeed, Campbell warns against “elevating] ... to a per se rule” Sony’s language about a presumption against fair use arising from commercial use. — U.S. at -, 114 S.Ct. at 1174. Campbell discards that language in favor of a more subtle, sophisticated approach, which recognizes that “the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use.” Id. [at —, 114 S.Ct.] at 1171. The Court states that “the commercial or nonprofit educational purpose of a work is only one element of the first factor enquiry” id. [at -, 114 S.Ct.] at 1174, and points out that “[i]f, indeed, commerciahty carried presumptive force against a finding of fairness, the presumption would swallow nearly all of the illustrative uses listed in the preamble paragraph of § 107....” Id.
We do not mean to suggest that the District Court overlooked these principles; in fact, the Court discussed them insightfully, see 802 F.Supp. at 12-13. Rather, our concern here is that the Court let the for-profit nature of Texaco’s activity weigh against Texaco without differentiating between a direct commercial use and the more indirect relation to commercial activity that occurred here. Texaco was not gaining direct or immediate commercial advantage from the photocopying at issue in this case — i.e., Texaco’s profits, revenues, and overall commercial performance were not tied to its making copies of eight Catalysis articles for Chickering. Cf. Basic Books, Inc. v. Kinko’s Graphics Corp., 758 F.Supp. 1522 (S.D.N.Y.1991) (revenues of reprographic business stemmed directly from selling unauthorized photocopies of copyrighted books). Rather, Texaco’s photocopying served, at most, to facilitate Chickering’s research, which in turn might have led to the development of new products and technology that could have improved Texaco’s commercial performance. Texaco’s photocopying is more appropriately labeled an “intermediate use.” See Sega Enterprises, 977 F.2d at 1522-23 (labeling secondary use “intermediate” and finding first factor in favor of for-profit company, even though ultimate purpose of copying was to develop competing commercial product, because immediate purpose of copying computer code was to study idea contained within computer program).
We do not consider Texaco’s status as a for-profit company irrelevant to the fair use analysis. Though Texaco properly contends that a court’s focus should be on the use of the copyrighted material and not simply on *922the user, it is overly simplistic to suggest that the “purpose and character of the use” can be fully discerned without considering the nature and objectives of the user.8
Ultimately, the somewhat cryptic suggestion in section 107(1) to consider whether the secondary use “is of a commercial nature or is for nonprofit educational purposes” connotes that a court should examine, among other factors, the value obtained by the secondary user from the use of the copyrighted material. See Rogers, 960 F.2d at 309 (“The first factor ... asks whether the original was copied in good faith to benefit the public or primarily for the commercial interests of the infringer.”); MCA, Inc. v. Wilson, 677 F.2d 180, 182 (2d Cir.1981) (court is to consider “whether the alleged infringing use was primarily for public benefit or for private commercial gain”). The commercial/nonprofit dichotomy concerns the unfairness that arises when a secondary user makes unauthorized use of copyrighted material to capture significant revenues as a direct consequence of copying the original work. See Harper & Row, 471 U.S. at 562, 105 S.Ct. at 2231 (“The crux of the profit/nonprofit distinction is ... whether the user stands to profit from exploitation of the copyrighted material without paying the customary price.”).
Consistent with these principles, courts will not sustain a claimed defense of fair use when the secondary use can fairly be characterized as a form of “commercial exploitation,” ie., when the copier directly and exclusively acquires conspicuous financial rewards from its use of the copyrighted material. See Harper & Row, 471 U.S. at 562-63, 105 S.Ct. at 2231-32; Twin Peaks, 996 F.2d at 1375; Rogers, 960 F.2d at 309; Iowa State University Research Foundation, Inc. v. American Broadcasting Companies, Inc., 621 F.2d 57, 61 (2d Cir.1980); Meeropol v. Nizer, 560 F.2d 1061, 1069 (2d Cir.1977) (examining whether use was “predominantly for commercial exploitation”), cert. denied, 434 U.S. 1013, 98 S.Ct. 727, 54 L.Ed.2d 756 (1978). Conversely, courts are more willing to find a secondary use fair when it produces a value that benefits the broader public interest. See Twin Peaks, 996 F.2d at 1375; Sega Enterprises, 977 F.2d at 1523; Rosemont Enterprises, Inc. v. Random House, Inc., 366 F.2d 303, 307-09 (2d Cir.1966), cert. denied, 385 U.S. 1009, 87 S.Ct. 714, 17 L.Ed.2d 546 (1967). The greater the private economic rewards reaped by the secondary user (to the exclusion of broader public benefits), the more likely the first factor will favor the copyright holder and the less likely the use will be considered fair.
As noted before, in this particular ease the link between Texaco’s commercial gain and its copying is somewhat attenuated: the copying, at most, merely facilitated Chicker-ing’s research that might have led to the production of commercially valuable products. Thus, it would not be accurate to conclude that Texaco’s copying of eight particular Catalysis articles amounted to “commercial exploitation,” especially since the immediate goal of Texaco’s copying was to facilitate Chickering’s research in the sciences, an objective that might well serve a broader public purpose. See Twin Peaks, 996 F.2d at 1375; Sega Enterprises, 977 F.2d at 1522. Still, we need not ignore the for-profit nature of Texaco’s enterprise, especially since we can confidently conclude that Texaco reaps at least some indirect economic advantage from its photocopying. As the publishers emphasize, Texaco’s photocopying for Chickering could be regarded simply as another “factor of production” utilized in Texaco’s efforts to develop profitable products. Conceptualized in this way, it is not obvious why it is fair for Texaco to avoid having to pay at least some price to copyright holders for the right to photocopy the original articles.
2. Transformative Use. The District Court properly emphasized that Texaco’s photocopying was not “transformative.” Af*923ter the District Court issued its opinion, the Supreme Court explicitly ruled that the concept of a “transformative use” is central to a proper analysis under the first factor, see Campbell, — U.S. at -, 114 S.Ct. at 1171-73. The Court explained that though a “transformative use is not absolutely necessary for a finding of fair use, ... the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use.” Id. at -, 114 S.Ct. at 1171.
The “transformative use” concept is pertinent to a court’s investigation under the first factor because it assesses the value generated by the secondary use and the means by which such value is generated. To the extent that the secondary use involves merely an untransformed duplication, the value generated by the secondary use is little or nothing more than the value that inheres in the original. Rather than making some contribution of new intellectual value and thereby fostering the advancement of the arts and sciences, an untransformed copy is likely to be used simply for the same intrinsic purpose as the original, thereby providing limited justification for a finding of fair use. See Weissmann v. Freeman, 868 F.2d 1313, 1324 (2d Cir.) (explaining that a use merely for the same “intrinsic purpose” as original “moves the balance of the calibration on the first factor against” secondary user and “seriously weakens a claimed fair use”), cert. denied, 493 U.S. 883, 110 S.Ct. 219, 107 L.Ed.2d 172 (1989).9
In contrast, to the extent that the secondary use “adds something new, with a further purpose or different character,” the value generated goes beyond the value that inheres in the original and “the goal of copyright, to promote science and the arts, is generally furthered.” Campbell, — U.S. at -, 114 S.Ct. at 1171; see also Pierre N. Leval, Toward a Fair Use Standard, 103 Harv. L.Rev. 1105, 1111 (1990) [hereinafter Leval, Toward a Fair Use Standard ]. It is therefore not surprising that the “preferred” uses illustrated in the preamble to section 107, such as criticism and comment, generally involve some transformative use of the original work. See 3 Nimmer on Copyright § 13.05[A][l][b], at 13-160.
Texaco suggests that its conversion of the individual Catalysis articles through photocopying into a form more easily used in a laboratory might constitute a transformative use. However, Texaco’s photocopying merely transforms the material object embodying the intangible article that is the copyrighted original work. See 17 U.S.C. §§ 101, 102 (explaining that copyright protection in literary works subsists in the original work of authorship “regardless of the nature of the material objects ... in which they are embodied”). Texaco’s making of copies cannot properly be regarded as a transformative use of the copyrighted material. See Steven D. Smit, “Make a Copy for the File ...”: Copyright Infringement by Attorneys, 46 Baylor L.Rev. 1, 15 & n. 58 (1994); see also Basic Books, 758 F.Supp. at 1530-31 (repackaging in anthology form of excerpts from copyrighted books not a transformative use).
Even though Texaco’s photocopying is not technically a transformative use of the copyrighted material, we should not overlook the significant independent value that can stem from conversion of original journal articles into a format different from their normal appearance. See generally Sony, 464 U.S. at 454, 455 n. 40, 104 S.Ct. at 795 n. 40 (acknowledging possible benefits from copying that might otherwise seem to serve “no productive purpose”); Weinreb, Fair’s Fair, at 1143 & n. 29 (discussing potential value from nontransformative copying). As previously explained, Texaco’s photocopying converts the individual Catalysis articles into a useful format. Before modern photocopying, Chickering probably would have converted the original article into a more serviceable form by taking notes, whether cursory or *924extended;10 today he can do so with a photocopying machine. Nevertheless, whatever independent value derives from the more usable format of the photocopy does not mean that every instance of photocopying wins on the first factor. In this case, the predominant archival purpose of the copying tips the first factor against the copier, despite the benefit of a more usable format.
3. Reasonable and Customary Practice. Texaco contends that Chickering’s photocopying constitutes a use that has historically been considered “reasonable and customary.” We agree with the District Court that whatever validity this argument might have had before the advent of the photocopying licensing arrangements discussed below in our consideration of the fourth fair use factor, the argument today is insubstantial. As the District Court observed, “To the extent the copying practice was ‘reasonable’ in 1973 [when Williams & Wilkins was decided], it has ceased to be ‘reasonable’ as the reasons that justified it before [photocopying licensing] have ceased to exist.” 802 F.Supp. at 25.
In amplification of Texaco’s arguments, our dissenting colleague makes two further points about the first factor analysis that merit a response. First, the dissent disputes our characterization of Chickering’s use as “archival” on the ground that such a use would occur in an institutional setting, whereas Chickering copied for his personal use. Second, the dissent contends that Chiekering’s use is transformative because it is an important step in the process of doing research. We think the proper response to these observations emerges from considering how they would fare if the Texaco library had sent around entire books, rather than issues of a journal. Clearly, Chickering (and all the other researchers at the Beacon facility) would be making archival use of the circulating books if they made photocopies of the books for their individual offices and thereby spared Texaco the expense of buying them all individual volumes. An individual copies for archival purposes even if the resulting archive remains in a private office. When a corporation invites such archival copying by circulating items likely to be worth copying (whether articles or entire books), any distinction between individual and institutional archiving loses all significance.
Moreover, the concept of a “transforma-tive” use would be extended beyond recognition if it was applied to Chiekering’s copying simply because he acted in the course of doing research. The purposes illustrated by the categories listed in section 107 refer primarily to the work of authorship alleged to be a fair use, not to the activity in which the alleged infringer is engaged. Texaco cannot gain fair use insulation for Chickering’s archival photocopying of articles (or books) simply because such copying is done by a company doing research. It would be equally extravagant for a newspaper to. contend that because its business is “news reporting” it may line the shelves of its reporters with photocopies of books on journalism or that schools engaged in “teaching” may supply its faculty members with personal photocopies of books on educational techniques or substantive fields. Whatever benefit copying and reading such books might contribute to the process of “teaching” would not for that reason satisfy the test of a “teaching” purpose.
On balance, we agree with the District Court that the first factor favors the publishers, primarily because the dominant purpose of the use is a systematic institutional policy of multiplying the available number of copies of pertinent copyrighted articles by circulating the journals among employed scientists for them to make copies, thereby serving the same purpose for which additional subserip-*925tions are normally sold, or, as will be discussed, for which photocopying licenses may be obtained.
B. Second Factor: Nature of Copyrighted Work
The second statutory fair use factor is “the nature of the copyrighted work.” 17 U.S.C. § 107(2). In assessing this factor, the District Court noted that the articles in Catalysis “are created for publication with the purpose and intention of benefiting from the protection of the copyright law,” and that copyright protection “is vitally necessary to the dissemination of scientific articles of the sort that are at issue.” 802 F.Supp. at 16. Nevertheless, the Court ultimately concluded that this factor favored Texaco because the photocopied articles were essentially factual in nature and the “‘scope of fair use is greater with respect to factual than nonfactual works.’ ” Id. at 16-17 (quoting New Era Publications International, ApS v. Carol Publishing Group, 904 F.2d 162, 167 (2d Cir.), cert. denied, 498 U.S. 921, 111 S.Ct. 297, 112 L.Ed.2d 251 (1990)).
On appeal, the publishers stress the District Court’s comments concerning the importance of broad copyright protection for journal publications in order to foster journal production. Further, citing Harper & Row for the proposition that the creativity of an original work weighs against finding fair use, see 471 U.S. at 563, 105 S.Ct. at 2232, the publishers also point out that “the journal articles are expressions of highly original, creative and imaginative thinking.”
Though a significant measure of creativity was undoubtedly used in the creation of the eight articles copied from Catalysis, even a glance at their content immediately reveals the predominantly factual nature of these works.11 Moreover, though we have previously recognized the importance of strong copyright protection to provide sufficient incentives for the creation of scientific works, see Weissmann, 868 F.2d at 1325, nearly every category of copyrightable works could plausibly assert that broad copyright protection was essential to the continued vitality of that category of works.
Ultimately, then, the manifestly factual character of the eight articles precludes us from considering the articles as “within the core of the copyright’s protective purposes,” Campbell, — U.S. at -, 114 S.Ct. at 1175; see also Harper & Row, 471 U.S. at 563, 105 S.Ct. at 2232 (“The law generally recognizes a greater need to disseminate factual works than works of fiction or fantasy.”). Thus, in agreement with the District Court, we conclude that the second factor favors Texaco.
C. Third Factor: Amount and Substantiality of Portion Used
The third statutory fair use factor is “the amount and substantiality of the portion used in relation to the copyrighted work as a whole.” 17 U.S.C. § 107(3). The District Court concluded that this factor clearly favors the publishers because Texaco copied the eight articles from Catalysis in their entirety.
Texaco makes various responses to the District Court’s straightforward conclusion. First, Texaco claims that this factor is significant only as a means to determine whether a copy unfairly supersedes demand for the original and should be considered “largely irrelevant” where, as here, a copy is not sold or distributed. Second, Texaco claims that, rather than focus on Texaco’s copying of entire articles, it is more appropriate to consider that Texaco copied only a very small portion of any particular issue or volume of Catalysis. Finally, Texaco cites Sony and Williams & Wilkins for the proposition that the copying of entire copyrighted works can still constitute fair use. See Sony, 464 U.S. at 449-50, 104 S.Ct. at 792-93; Williams & Wilkins, 487 F.2d at 1353.
Texaco’s suggestion that we consider that it copied only a small percentage of the *926total compendium of works encompassed within Catalysis is superficially intriguing, especially since Catalysis is traditionally marketed only as a periodical by issue or volume. However, as the District Court recognized, each of the eight articles in Catalysis was separately authored and constitutes a discrete “original work[] of authorship,” 17 U.S.C. § 102. As we emphasized at the outset, each article enjoys independent copyright protection, which the authors transferred to Academic Press, and what the publishers claim has been infringed is the copyright that subsists in each individual article— not the distinct copyright that may subsist in each journal issue or volume by virtue of the publishers’ original compilation of these articles. The only other appellate court to consider the propriety of photocopying articles from journals also recognized that each article constituted an entire work in the fair use analysis. See Williams & Wilkins, 487 F.2d at 1353.
Despite Texaco’s claims that we consider its amount of copying “minuscule” in relation to the entirety of Catalysis, we conclude, as did the District Court, that Texaco has copied entire works. Though this conclusion does not preclude a finding of fair use, it militates against such a finding, see Sony, 464 U.S. at 449-50, 104 S.Ct. at 792-93, and weights the third factor in favor of the publishers.
Finally, though we are sensitive to Texaco’s claim that the third factor serves merely as a proxy for determining whether a secondary use significantly interferes with demand for the original — a concern echoed by some commentators, see William W. Fisher III, Reconstructing the Fair Use Doctrine, 101 Harv.L.Rev. 1661, 1678 (1988) [hereinafter Fisher, Reconstructing Fair Use ] — we think this factor serves a further end that advances the fair use analysis. Specifically, by focuss-ing on the amount and substantiality of the original work used by the secondary user, we gain insight into the purpose and character of the use as we consider whether the quantity of the material used was “reasonable in relation to the purpose of the copying.” See Campbell, — U.S. at -, 114 S.Ct. at 1175. In this case, the fact that Texaco photocopied the eight Catalysis articles in their entirety weakens its assertion that the overriding-purpose and character of its use was to enable the immediate use of the article in the laboratory and strengthens our view that the predominant purpose and character of the use was to establish a personal library of pertinent articles for Chickering. Cf id. at -, 114 S.Ct. at 1176 (intimating that extent of copying can provide insight into primary purpose of copying).
D. Fourth Factor: Effect Upon Potential Market or Value
The fourth statutory fair use factor is “the effect of the use upon the potential market for or value of the copyrighted work.” 17 U.S.C. § 107(4). Assessing this factor, the District Court detailed the range of procedures Texaco could use to obtain authorized copies of the articles that it photocopied and found that “whatever combination of procedures Texaco used, the publishers’ revenues would grow significantly.” 802 F.Supp. at 19. The Court concluded that the publishers “powerfully demonstrated entitlement to prevail as to the fourth factor,” since they had shown “a substantial harm to the value of their copyrights” as the consequence of Texaco’s copying. See id. at 18-21.
Prior to Campbell, the Supreme Court had characterized the fourth factor as “the single most important element of fair use,” Harper & Row, 471 U.S. at 566, 105 S.Ct. at 2233; accord 3 Nimmer on Copyright § 13.05[A][4], at 13-183. However, Campbell ’s discussion of the fourth factor conspicuously omits this phrasing. Apparently abandoning the idea that any factor enjoys primacy, Campbell instructs that ‘[a]ll [four factors] are to be explored, and the results weighed together, in light of the purposes of copyright.’ [— U.S. at -] 114 S.Ct. at 1171.
In analyzing the fourth factor, it is important (1) to bear in mind the precise copyrighted works, namely the eight journal articles, and (2) to recognize the distinctive nature and history of “the potential market for *927or value of’ these particular works.12 Specifically, though there is a traditional market for, and hence a clearly defined value of, journal issues and volumes, in the form of per-issue purchases and journal subscriptions, there is neither a traditional market for, nor a clearly defined value of, individual journal articles. As a result, analysis of the fourth factor cannot proceed as simply as would have been the case if Texaco had copied a work that carries a stated or negotiated selling price in the market.
Like most authors, writers of journal articles do not directly seek to capture the potential financial rewards that stem from their copyrights by personally marketing copies of their writings. Rather, like other creators of literary works, the author of a journal article “commonly sells his rights to publishers who offer royalties in exchange for their services in producing and marketing the author’s work.” Harper & Row, 471 U.S. at 547, 105 S.Ct. at 2223. In the distinctive realm of academic and scientific articles, however, the only form of .royalty paid by a publisher is often just the reward of being published, publication being a key to professional advancement and prestige for the author, see Weissmann, 868 F.2d at 1324 (noting that “in an academic setting, profit is ill-measured in dollars. Instead, what is valuable is recognition because it so often influences professional advancement and academic tenure.”). The publishers in turn incur the costs and labor of producing and marketing authors’ articles, driven by the prospect of capturing the economic valué stemming from the copyrights in the original works, which the authors have transferred to them. Ultimately, the monopoly privileges conferred by copyright protection and the potential financial rewards therefrom are not directly serving to motivate authors to write individual articles; rather, they serve to motivate publishers to produce journals, which provide the conventional and often exclusive means for disseminating these individual articles. It is the prospect of such dissemination that contributes to the motivation of these authors.
Significantly, publishers have traditionally produced and marketed authors’ individual articles only in a journal format, i.e., in periodical compilations of numerous articles. In other words, publishers have conventionally sought to capture the economic value from the “exclusive rights” to “reproduce” and “distribute copies” of the individual articles, see 17 U.S.C. § 106(1) & (3), solely by compiling many such articles together in a periodical journal and then charging a fee to subscribe. Publishers have not traditionally provided a simple or efficient means to obtain single copies of individual articles; reprints are usually available from publishers only in bulk quantities and with some delay.
This marketing pattern has various consequences for our analysis of the fourth factor. First, evidence concerning the effect that photocopying individual journal articles has on the traditional market for journal subscriptions is of somewhat less significance than if a market existed for the sale of individual copies of articles. Second, this distinctive arrangement raises novel questions concerning the significance of the publishers’ establishment of an innovative licensing scheme for the photocopying of individual journal articles.
1. Sales of Additional Journal Subscriptions, Back Issues, and Back Volumes. Since we are concerned with the claim of fair use in copying the eight individual articles from Catalysis, the analysis under the fourth factor must focus on the effect of Texaco’s photocopying upon the potential market for or value of these individual articles. Yet, in their respective discussions of the fourth statutory factor, the parties initially focus on the impact of Texaco’s photocopying of indi*928vidual journal articles upon the market for Catalysis journals through sales of Catalysis subscriptions, back issues, or back volumes.
As a general matter, examining the effect on the marketability of the composite work containing a particular individual copyrighted work serves as a useful means to gauge the impact of a secondary use “upon the potential market for or value of’ that individual work, since the effect on the marketability of the composite work will frequently be directly relevant to the effect on the market for or value of that individual work.13 Quite significantly, though, in the unique world of academic and scientific articles, the effect on the marketability of the composite work in which individual articles appear is not obviously related to the effect on the market for or value of the individual articles. Since (1) articles are submitted unsolicited to journals, (2) publishers do not make any payment to authors for the right to publish their articles or to acquire their copyrights, and (3) there is no evidence in the record suggesting that publishers seek to reprint particular articles in new composite works, we cannot readily conclude that evidence concerning the effect of Texaco’s use on the marketability of journals provides an effective means to appraise the effect of Texaco’s use on the market for or value of individual journal articles.
These considerations persuade us that evidence concerning the effect of Texaco’s photocopying of individual articles within Catalysis on the traditional market for Catalysis subscriptions is of somewhat limited significance in determining and evaluating the effect of Texaco’s photocopying “upon the potential market for or value of’ the individual articles. We do not mean to suggest that we believe the effect on the marketability of journal subscriptions is completely irrelevant to gauging the effect on the market for and value of individual articles. Were the publishers able to demonstrate that Texaco’s type of photocopying, if widespread,14 would impair the marketability of journals, then they might have a strong claim under the fourth factor. Likewise, were Texaco able to demonstrate that its type of photocopying, even if widespread, would have virtually no effect on the marketability of journals, then it might have a strong claim under this fourth factor.
On this record, however, the evidence is not resounding for either side. The District Court specifically found that, in the absence of photocopying, (1) “Texaco would not ordinarily fill the need now being supplied by photocopies through the purchase of back issues or back volumes ... [or] by enormously enlarging the number of its subscriptions,” but (2) Texaco still “would increase the number of subscriptions somewhat.” 802 F.Supp. at 19.15 This moderate conclusion concem-*929ing the actual effect on the marketability of journals, combined with the uncertain relationship between the market for journals and the market for and value of individual articles, leads us to conclude that the evidence concerning sales of additional journal subscriptions, back issues, and back volumes does not strongly support either side with regard to the fourth factor. Cf. Sony, 464 U.S. at 451-55, 104 S.Ct. at 793-95 (rejecting various predictions of harm to value of copyrighted work based on speculation about possible consequences of secondary use). At best, the loss of a few journal subscriptions tips the fourth factor only slightly toward the publishers because evidence of such loss is weak evidence that the copied articles themselves have lost any value.
2. Licensing Revenues and Fees. The District Court, however, went beyond discussing the sales of additional journal subscriptions in holding that Texaco’s photocopying affected the value of the publishers’ copyrights. Specifically, the Court pointed out that, if Texaco’s unauthorized photocopying was not permitted as fair use, the publishers’ revenues would increase significantly since Texaco would (1) obtain articles from document delivery services (which pay royalties to publishers for the right to photocopy articles), (2) negotiate photocopying licenses directly with individual publishers, and/or (3) acquire some form of photocopying license from the Copyright Clearance Center Inc. (“CCC”).16 See 802 F.Supp. at 19. Texaco claims that the District Court’s reasoning is faulty because, in determining that the value of the publishers’ copyrights was affected, the Court assumed that the publishers were entitled to demand and receive licensing royalties and fees for photocopying. Yet, continues Texaco, whether the publishers can demand a fee for permission to make photocopies is the very question that the fair use trial is supposed to answer.
It is indisputable that, as a general matter, a copyright holder is entitled to demand a royalty for licensing others to use its copyrighted work, see 17 U.S.C. § 106 (copyright owner has exclusive right “to authorize” certain uses), and that the impact on potential licensing revenues is a proper subject for consideration in assessing the fourth factor, see, e.g., Campbell, — U.S. at -, 114 S.Ct. at 1178; Harper & Row, 471 U.S. at 568-69, 105 S.Ct. at 2234-35; Twin Peaks, 996 F.2d at 1377; DC Comics, Inc. v. Reel Fantasy, Inc., 696 F.2d 24, 28 (2d Cir.1982); United Telephone Co. of Missouri v. Johnson Publishing Co., Inc., 855 F.2d 604, 610 (8th Cir.1988).
However, not every effect on potential licensing revenues enters the analysis under the fourth factor.17 Specifically, courts have *930recognized limits on the concept of “potential licensing revenues” by considering only traditional, reasonable, or likely to be developed markets when examining and assessing a secondary use’s “effect upon the potential market for or value of the copyrighted work.” See Campbell, — U.S. at -, 114 S.Ct. at 1178 (“The market for potential derivative uses includes only those that creators of original works would in general develop or license others to develop.”); Harper & Row, 471 U.S. at 568, 105 S.Ct. at 2235 (fourth factor concerned with “use that supplants any part of the normal market for a copyrighted work”) (emphasis added) (quoting S.Rep. No. 473, 94th Cong., 1st Sess. 65 (1975)); see also Mathieson v. Associated Press, 23 U.S.P.Q.2d 1685, 1690-91, 1992 WL 164447 (S.D.N.Y.1992) (refusing to find fourth factor in favor of copyright holder because secondary use did not affect any aspect of the normal market for copyrighted work).
For example, the Supreme Court recently explained that because of the “unlikelihood that creators of imaginative works will license critical reviews or lampoons” of their works, “the law recognizes no derivative market for critical works,” Campbell, — U.S. at -, 114 S.Ct. at 1178. Similarly, other courts have found that the fourth factor will favor the secondary user when the only possible adverse effect occasioned by the secondary use would be to a potential market or value that the copyright holder has not typically sought to, or reasonably been able to, obtain or capture. See Twin Peaks, 996 F.2d at 1377 (noting that fourth factor will favor secondary user when use “filled a market niche that the [copyright owner] simply had no interest in occupying”); Pacific and Southern Co. v. Duncan, 744 F.2d 1490, 1496 (11th Cir.1984) cert. denied, 471 U.S. 1004, 105 S.Ct. 1867, 85 L.Ed.2d 161 (1985) (noting that the fourth factor may not favor copyright owner when the secondary user “profits from an activity that the owner could not possibly take advantage of’).18
Thus, Texaco is correct, at least as a general matter, when it contends that it is not always appropriate for a court to be swayed on the fourth factor by the effects on potential licensing revenues. Only an impact on potential licensing revenues for traditional, reasonable, or likely to be developed markets should be legally cognizable when evaluating a secondary use’s “effect upon the potential market for or value of the copyrighted work.”
Though the publishers still have not established a conventional market for the direct sale and distribution of individual articles, they have created, primarily through the CCC, a workable market for institutional users to obtain licenses for the right to produce their own copies of individual articles via photocopying. The District Court found that many major corporations now subscribe to the CCC systems for photocopying licenses. 802 F.Supp. at 25. Indeed, it appears from the pleadings, especially Texaco’s counterclaim, that Texaco itself has been paying royalties to the CCC. See Complaint ¶38; First Counterclaim ¶ 71. Since the Copyright Act explicitly provides that copyright holders have the “exclusive rights” to “reproduce” and “distribute copies” of their works, see 17 U.S.C. § 106(1) & (3), and since there currently exists a viable market for licensing these rights for individual journal articles, it is appropriate that potential licensing revenues for photocopying be considered in a fan-use analysis.
Despite Texaco’s claims to the contrary, it is not unsound to conclude that the right to seek payment for a particular use tends to *931become legally cognizable under the fourth fair use factor when the means for paying for such a use is made easier. This notion is not inherently troubling: it is sensible that a particular unauthorized use should be considered “more fair” when there is no ready market or means to pay for the use, while such an unauthorized use should be considered “less fair” when there is a ready market or means to pay for the use. The vice of circular reasoning arises only if the availability of payment is conclusive against fair use. Whatever the situation may have been previously, before the development of a market for institutional users to obtain licenses to photocopy articles, see Williams & Wilkins, 487 F.2d at 1357-59, it is now appropriate to consider the loss of licensing revenues in evaluating “the effect of the use upon the potential market for or value of’ journal articles. It is especially appropriate to do so with respect to copying of articles from Catar lysis, a publication as to which a photocopying license is now available. We do not decide how the fair use balance would be resolved if a photocopying license for Catalysis articles were not currently available.
In two'ways, Congress has impliedly suggested that the law should recognize licensing fees for photocopying as part of the “potential market for or value of’ journal articles. First, section 108 of the Copyright Act narrowly circumscribes the conditions under which libraries are permitted to make copies of copyrighted works. See 17 U.S.C. § 108. Though this section states that it does not in any way affect the right of fair use, see id. § 108(f)(4), the very fact that Congress restricted the rights of libraries to make copies implicitly suggests that Congress views journal publishers as possessing the right to restrict photocopying, or at least the right to demand a licensing royalty from nonpublic institutions that engage in photocopying. Second, Congress apparently prompted the development of CCC by suggesting that an efficient mechanism be established to license photocopying, see S.Rep. No. 983, 93d Cong., 2d Sess. 122 (1974); S.Rep. No. 473, 94th Cong., 1st Sess. 70-71 (1975); H.R.Rep. No. 83, 90th Cong., 1st Sess. 33 (1968). It is difficult to understand why Congress would recommend establishing such a mechanism if it did not believe that fees for photocopying should be legally recognized as part of the potential market for journal articles.
Primarily because of lost licensing revenue, and to a minor extent because of lost subscription revenue, we agree with the District Court that “the publishers have demonstrated a substantial harm to the value of their copyrights through [Texaco’s] copying,” 802 F.Supp. at 21, and thus conclude that the fourth statutory factor favors the publishers.
E. Aggregate Assessment
We conclude that three of the four statutory factors, including the important first and the fourth factors, favor the publishers. We recognize that the statutory factors provide a nonexclusive guide to analysis, see Harper & Row, 471 U.S. at 560, 105 S.Ct. at 2230, but to whatever extent more generalized equitable considerations are relevant, we are in agreement with the District Court’s analysis of them. See 802 F.Supp. at 21-27. We therefore agree with the District Court’s conclusion that Texaco’s photocopying of eight particular articles from the Journal of Catalysis was not fair use.
Though we recognize the force of many observations made in Judge Jacobs’s dissenting opinion, we are not dissuaded by his dire predictions that our ruling in this ease “has ended fair-use photocopying with respect to a large population of journals,” 60 F.3d at 938-39, or, to the extent that the transactional licensing scheme is used, “would seem to require that an intellectual property lawyer be posted at each copy machine,” id. at 937-38. Our ruling does not consider photocopying for personal use by an individual. Our ruling is confined to the institutional, systematic, archival multiplication of copies revealed by the record — the precise copying that the parties stipulated should be the basis for the District Court’s decision now on appeal and for which licenses are in fact available. And the claim that lawyers need to be stationed at copy machines is belied by the ease with which music royalties have been collected and distributed for performances at thousands of cabarets, without the attendance of *932intellectual property lawyers in any capacity other than as customers. If Texaco wants to continue the precise copying we hold not to be a fair use, it can either use the licensing schemes now existing or some variant of them, or, if all else fails, purchase one more subscription for each of its researchers who wish to keep issues of Catalysis on the office shelf.
Conclusion
The order of the District Court is affirmed.19
. Texaco, for example, uses a significant portion of its initial brief to expound on photocopying activities in various industries. Similarly, a large part of the publishers’ statement of facts is devoted to a broad discussion of Texaco's photocopying practices, the social importance of academic and scientific journals, and the economics of journal publication and photocopying. These and other details presented by the parties are discussed in the District Court's opinion, 802 F.Supp. at 4-9.
. In full, 17 U.S.C. § 107 reads:
§ 107. Limitation on exclusive rights: Fair Use
Notwithstanding the provisions of sections 106 and 106A, the fair use of a copyrighted work, including such use hy reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include—
(1)the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
(2) the nature of copyrighted work;
(3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
(4) the effect of the use upon the potential market for or value of the copyrighted work.
The fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors.
. See Agreement on Guidelines for Classroom Copying in Not-For-Profit Educational Institutions, quoted in Patiy, The Fair Use Privilege, at 308, discussed infra, note 5.
. For various reasons, for example, because certain articles are the work of the United States Government (which makes copyright protection unavailable, see 17 U.S.C. § 105), the publishers do not always possess the copyrights for all articles within each journal.
. These guidelines were included in the legislative history of the 1976 revision of the Copyright Act, see H.R.Rep. No. 1476, 94th Cong., 2d Sess. 68-71 (1976), U.S.Code Cong. & Admin.News 1976, p. 5659, and were endorsed by the House Judiciary Committee as “a reasonable interpretation of the minimum standards of fair use.” Id. at 72, U.S.Code Cong. & Admin.News 1976, at 5686. Though these guidelines are not considered necessarily binding on courts, see Marcus v. Rowley, 695 F.2d 1171, 1178 (9th Cir.1983), they exist as a persuasive authority marking out certain minimum standards for educational fair uses, see Basic Books, Inc. v. Kinko’s Graphics Corp., 758 F.Supp. 1522-36 (S.D.N.Y.1991). See generally 3 Nimmer on Copyright § 13.05[E][3][a], at 13-226.1 to 13-226.2 (discussing nature and impact of guidelines); Patry, The Fair Use Privilege, at 307-09, 404-07 (same).
. In this regard, the District Court's conclusion that the "primary aspect" of Texaco's copying was to multiply copies is accurate, see 802 F.Supp. at 14-15, irrespective of the evidence (or lack of evidence) concerning the nature and scope of Texaco’s photocopying activity for its entire population of scientists. Even if the photocopies of the Catalysis articles in Chickering's files were the only copies ever made by Texaco— which, as Texaco stresses, is all that the evidence developed below conclusively showed — the primary objective in making these single copies was to provide Chickering with his own, additional, readily accessible copy of the original article. As the District Court noted, "[I]f Chickering were the subscriber and sole user of the subscription to Catalysis, and he made an extra copy of an article for use in the lab or for marking with scratch notes, the argument [for a transformative *920fair use] might have considerable force.” 802 F.Supp. at 14.
. Though Texaco claims that its copying is for "research" as that term is used in the preamble of section 107, this characterization might somewhat overstate the matter. Chickering has not used portions of articles from Catalysis in his own published piece of research, nor has he had to duplicate some portion of copyrighted material directly in the course of conducting an experiment or investigation. Rather, entire articles were copied as an intermediate step that might abet Chickering's research.
. See Patry, The Fair Use Privilege, at 416-17 (noting that the nature of person or entity engaging in use affects the character of the use); Report of the Register of Copyrights — Library Reproduction of Copyrighted Works (17 U.S.C. 108) 85 (1983) (explaining that though a scientist in a for-profit firm and a university student may engage in the same photocopying of scholarly articles to facilitate their research, "the copyright consequences are different; [the scientist's] copying is of a clearly commercial nature, and less likely to be fair use”) quoted in Patry, The Fair Use Privilege, at 417 n. 307.
. See also Marcus v. Rowley, 695 F.2d at 1175 (emphasizing that “a finding that the alleged infringers copied the material to use it for the same intrinsic purpose for which the copyright owner intended it to be used is strong indicia of no fair use."). See generally Leon E. Seltzer, Exemptions and Fair Use in Copyright 24 (1978) (noting traditional limit on applicability of fair use doctrine when reproduction of original work is done “in order to use it for its intrinsic purpose — to make what might be called the 'ordinary' use of it”).
. In stating that a handwritten copy would have been made, we do not mean to imply that such copying would necessarily have been a fair use. Despite the 1973 dictum in Williams & Wilkins asserting that "it is almost unanimously accepted that a scholar can make a handwritten copy of an entire copyrighted article for his own use ..487 F.2d at 1350, the current edition of the Nimmer treatise reports that "[t]here is no reported case on the question of whether a single handwritten copy of all or substantially all of a book or other protected work made for the copier's own private use is an infringement or fair use.” 3 Nimmer on Copyright § 1305[E][4][a], at 13-229.
. Not only are the Catalysis articles essentially factual in nature, but the evidence suggests that Chickering was interested exclusively in the facts, ideas, concepts, or principles contained within the articles. Though scientists surely employ creativity and originality to develop ideas and obtain facts and thereafter to convey the ideas and facts in scholarly articles, it is primarily the ideas and facts themselves that are of value to other scientists in their research.
. We focus on the eight articles to emphasize the special characteristics of articles as distinguished from journal issues or bound volumes. In doing so, we recognize, as did the District Court, see 802 F.Supp. at 18 n. 15, that the fourth factor is concerned with the category of a defendant's conduct, not merely the specific instances of copying. See 3 Nimmer on Copyright § 13.05[A][4], at 13-183 to 13-184 ("[I]t is a mistake to view [the fourth] factor ... as merely raising the question of the extent of damages to plaintiff caused by the particular activities of the defendant. This factor rather poses the issue of whether unrestricted and widespread conduct of the sort engaged in by the defendant ... would result in a substantially adverse impact on the potential market for or value of the plaintiff's present work.") (emphasis added).
. One reason that the effect on the marketability of the composite work is typically relevant is because the strength of the market for the composite work will influence the payment producers will be willing to give to the author of the individual work for permission to include that individual work. For example, if a secondary use of a copyrighted story adversely affects purchases of a collection of short stories in which this story appears, then other producers of short story collections will less likely seek to have, or will pay less to have, that story as part of their collections. In this way, the market for or value of the story has clearly been affected by the secondary use.
. Properly applied, the fourth factor requires a court to consider "not only ... particular actions of the alleged infringer, but also 'whether unrestricted and widespread conduct of the sort engaged in by the defendant ... would result in a substantially adverse impact on the potential market' for the original.” Campbell, - U.S. at -, 114 S.Ct. at 1177 (quoting 3 Nimmer on Copyright § 13.05[A][4] ). Accord Harper & Row, 471 U.S. at 568-69, 105 S.Ct. at 2234-35; Rogers, 960 F.2d at 312.
.Texaco assails the conclusion that, without photocopying, it would increase subscriptions "somewhat" as an improper inference unsupported by the evidence. Though we accept Texaco's assertion that additional subscriptions provide an imperfect substitute for the copies of individual articles that scientists need and prefer, we cannot conclude that the District Court's factual finding that "Texaco would add at least a modest number of subscriptions," 802 F.Supp. at 19, is clearly erroneous.
First, though Texaco claims that there is no reliable evidence suggesting that photocopying served to facilitate journal circulation, the evidence concerning Texaco's routing practices supports the District Court’s inference that, without photocopying, Texaco will need a greater number of subscriptions to insure the prompt circulation of journals. Second, as discussed in connection with the first statutory factor, the domi*929nant reason for, and value derived from, the copying of the eight particular Catalysis articles was to make them available on Chickering’s shelf for ready reference when he needed to look at them. Thus, it is reasonable to conclude that Texaco would purchase at least a few additional subscriptions to serve this purpose, i.e., to provide certain researchers with personal copies of particular articles in their own offices.
. The CCC is a central clearing-house established in 1977 primarily by publishers to license photocopying. The CCC offers a variety of licensing schemes; fees can be paid on a per copy basis or through blanket license arrangements. Most publishers are registered with the CCC, but the participation of for-profit institutions that engage in photocopying has been limited, largely because of uncertainty concerning the legal questions at issue in this lawsuit. The CCC is fully described in the District Court's opinion. 802 F.Supp. at 7-9. A more extended discussion of the formation, development, and effectiveness of the CCC and its licensing schemes is contained in Stanley M. Besen & Sheila Nataraj Kirby, Compensating Creators of Intellectual Property: Collectives that Collect (1989).
. As Texaco notes and others have recognized, a copyright holder can always assert some degree of adverse affect on its potential licensing revenues as a consequence of the secondary use at issue simply because the copyright holder has not been paid a fee to permit that particular use. See Leval, Toward a Fair Use Standard, at 1124 ("By definition every fair use involves some loss of royalty revenue because the secondary user has not paid royalties.”); Fisher, Reconstructing Fair Use, at 1671 (noting that in almost every ' case "there will be some material adverse impact on a 'potential market' ” since the secondary user has not paid for the use). Thus, were a court automatically to conclude in every case that potential licensing revenues were impermis-sibly impaired simply because the secondary user did not pay a fee for the right to engage in the use, the fourth fair use factor would always favor the copyright holder. See Leval, Toward a Fair Use Standard, at 1125; Fisher, Reconstructing Fair Use, at 1672.
. The Supreme Court's holding in Sony implicitly recognizes limits on the concept of "potential market for or value of the copyrighted work.” Despite Justice Blackmun's dissenting view that the copying of television programs to enable private viewing at a more convenient time, i.e., "time-shifting,” deprived copyright holders of the ability to exploit the “sizable market” of persons who "would be willing to pay some kind of royalty” for the "privilege of watching copyrighted work at their convenience," Sony, 464 U.S. at 485, 104 S.Ct. at 811, the majority found that the copyright holders "failed to demonstrate that time-shifting would cause any likelihood of non-minimal harm to the potential market for, or the value of, their copyrighted works.” Id. at 456, 104 S.Ct. at 796. The Court thus implicitly ruled that the potential market in licensing royalties enunciated by Justice Blackmun should be considered too insubstantial to tilt the fourth fair use factor in favor of the copyright holder.
. Though neither the limited trial nor this appeal requires consideration of the publishers’ remedy if infringement is ultimately found, we note that the context of this dispute appears to make ill-advised an injunction, which, in any event, has not been sought. If the dispute is not now settled, this appears to be an appropriate case for exploration of the possibility of a court-imposed compulsory license. See Campbell, - U.S. at - n. 10, 114 S.Ct. at 1171 n. 10; 3 Nimmer on Copyright § 13.05[E][4][e], at 13-241 to 13-242.