751 F.2d 501 | 2d Cir. | 1984
Lead Opinion
This is an appeal from a judgment of the United States District Court for the Southern District of New York, Robert L. Carter, Judge, dismissing after a bench trial plaintiff-appellant Financial Information Inc.’s complaint charging defendant-appel-lee Moody’s Investors Service, Inc. with copyright infringement and unfair competition. The court first found that the information gathered was copyrightable under the “compilation” provision of the copyright law, 17 U.S.C. § 101 (1982). Then, in dismissing the action, the district court found that the fair use doctrine, codified in 17 U.S.C. § 107 (1982), shielded Moody’s from liability on the federal copyright infringement allegations, and that there could be no unfair competition under state law since the parties were not competitors and because plaintiff could show no injury.
One week after the judgment below was entered, the Supreme Court decided Sony Corp. of America v. Universal City Studios, Inc., — U.S. -, 104 S.Ct. 774, 78 L.Ed.2d 574 (1984), which now represents the leading exposition of the fair use doctrine. In light of Sony, and in view of other considerations discussed below, we reverse and hold that Moody’s copying was not protected by the fair use doctrine. This holding on the federal cause of action necessarily entails a correlative holding that the state cause of action is preempted under 17 U.S.C. § 301 (1982). But it also entails an examination of the copyrightability issue, as to which we wish further elaboration on the facts and reconsideration. We remand to the district court for such elaboration and reconsideration and a determination of damages, if that issue is reached.
Facts
Financial Information Inc. (“FII”) is a publisher of financial information. It produces five separate works, including its “Financial Daily Called Bond Service,” from which it alleges Moody’s stole copyrighted material.
FII’s Financial Daily Called Bond Service (“Daily Bond Cards”) consists literally of packets of 4" x 6" index cards on which are printed information regarding municipal bonds which the issuer has elected to redeem, or “call.” The information typically includes the identity of the issuing (and, of course, calling) municipal authority, the series of bonds being called, the date and price of the redemption, and the name of the trustee or paying agent to whom the bond should be presented for payment. FII obtains this information principally by having its employees read newspapers from all over the country, looking for and clipping the paid notices of calls that issuers place in the local press. Occasionally, FII employees telephone issuers for information, and sometimes other commercial bond redemption services — such as Moody’s — are consulted. There is apparently little, if any, editorial skill or creative discretion involved; the cards are essentially a compilation of financial facts collected from various sources, the key facts being those selected for publication.
The Daily Bond Cards seek to report all United States municipal redemptions. The packets are mailed each day to roughly 500 subscribers, who are generally large financial institutions who own numerous municipal bonds and who would find it difficult to monitor all relevant redemptions. The cost of the service is $279 per year, which includes an annual cumulative volume and a filing cabinet.
Defendant-appellee Moody’s Investors Service, Inc. (“Moody’s”) is also in the business of selling financial information. Among its products is its Municipal and Government News Reports (“News Reports”), a bi-weekly supplement to a yearly publication called the Municipal and Government Manual.
Like FII’s Daily Bond Cards, Moody’s News Reports provide institutional sub
According to FII, in 1980 it noticed a “coincidence of Moody’s errata publishing after FII,” and began to suspect that Moody’s was simply copying information from its cards. In December of that year, in an attempt to confirm its suspicions, FII fabricated and planted an error “redeeming” some bonds that had, in fact, been called a year earlier. The fake redemption was picked up and published by Moody’s. While FII did not plant any more errors — it was, after all, in jeopardy of sacrificing its reputation with its customers for credibility and reliability — FII did compare Moody’s reports with its own in cases where FII had made unintentional errors. It was ultimately determined — and Moody’s does not contest — that in 1980 seven of ten FII errors were reproduced by Moody’s, while in 1981, all eight FII errors were copied by Moody’s.
At trial, FII’s expert witness, Professor Herbert Robbins, Professor of Mathematical Statistics at Columbia University, testified that based on the incidence of FII errors which appeared in Moody’s News Reports it was more than 95% certain that Moody’s had copied at least 40-50% of FII’s information in the years 1980 and 1981. Professor Robbins further testified that in 1981, where eight of eight FII errors appeared in the News Reports, the probabilities were 51% that Moody’s copied 91% of FII’s Daily Bond Cards. At a “confidence level,” i.e., probability, of 95%, Professor Robbins put the percentage of Moody’s copying in 1981 at 68%.
Moody’s presented evidence at trial of what it calls its “independent creation” of its News Reports. According to the testimony of Moody’s employees, the research effort that went into the News Reports included the reading of over two dozen newspapers and the monitoring of four wire services. Moody’s maintained that nine employees were directly, though not necessarily exclusively, involved in the production of the News Reports, and that the “total accrued annual costs allocable” to the News Reports, including research effort costs, varied from between $700,000 to $1,000,000. In addition, Moody’s sought to counter the expert testimony of Professor Robbins by showing through reference to publication deadlines and schedules that more than half of its bond redemption notices simply could not have been copied from FII’s cards.
The Decision Below
In the post-trial opinion dismissing the complaint, the district court found, as a preliminary matter, that FII’s Daily Bond Cards were copyrightable, thus confirming the substance of the decision on appellee’s prior motion for summary judgment.
After stating that the fact of a commercial use of copyrighted material does not in and of itself defeat a fair use defense, the district court noted that in some instances, Moody’s publication served as legal notice of a bond’s redemption and that many institutions were “eager” to have this information. The court then found that Moody’s was performing a “public function” by “making available this needed financial information.” Thus, the fulfillment of this public function brought Moody’s “clearly ... within the ambit of the first requirement for fair use protection.”
The district court then focused on what it called “the simplest hurdle for Moody’s to overcome,” namely the nature of the copyrighted work. The court stated that since FII’s cards represented a mere compliation of factual material, the scope of permissible fair use was greater than it would be with respect to a “truly creative work,” and held that Moody’s thus met “the second yardstick.”
The third factor analyzed was the “amount and substantiality of the portions copied.” Despite Professor Robbin’s testimony, the court held that FII “failed to establish substantial use of its work” by Moody’s. Apparently, the basis for the rejection of the expert statistical evidence was the crediting of Moody’s evidence of significant simultaneous or even prior publication of the call data. The district court was also impressed by the fact that Moody’s received more newspapers than FII, and monitored four wire services. The judge then stated that he could not “make the leap that 8 of 8 proven instances shows copying by defendant 91% of the time,” and found instead that Moody’s had copied FII cards “in some 20-24 instances throughout a given year,” a finding which, according to FII, translates into a copying rate of just 1 to 2%.
Turning to the “final and most important factor,” the district court found no evidence that Moody’s use of FII’s materials tended to diminish or prejudice the potential sale of plaintiff’s work. The judge based this finding on his view that the parties “serviced separate, discrete markets,” and on the lack of any evidence showing a decline in the number of FII subscribers or an increase in the number of Moody’s subscribers resulting from the copying.
Having found that Moody’s cleared all four hurdles, the court held that the fair use doctrine operated as a complete defense to FII’s claims under the Copyright Act. The court then stated that the only New York state unfair competition action viable on the facts was that of misappropriation, but held that given the absence of a showing of competition and injury, the claim necessarily failed.
Discussion
Copyrightability
Moody’s made below, as it makes here, two basic arguments why the called bond data that FII publishes each day cannot receive copyright protection: first, that the daily data are facts and therefore not copyrightable and, second, that the daily data are not compilations and are therefore not copyrightable as such.
As the district court held, of course, Moody’s is quite accurate that facts are not copyrightable. See, e.g., Miller v. Universal City Studios, Inc., 650 F.2d
While facts have not been given copyright protection, as the district court pointed out, compilations of such facts traditionally have been. See, e.g., Schroeder v. William Morrow & Co., 566 F.2d 3 (7th Cir.1977) (directory of nurseries); Leon v. Pacific Telephone & Telegraph Co., 91 F.2d 484 (9th Cir.1937) (telephone directory); Hartfield v. Peterson, 91 F.2d 998 (2d Cir.1937) (cable and telegraphic code); Jeweler’s Circular Publishing Co. v. Keystone Publishing Co., 281 Fed. 83 (2d Cir.), cert. denied, 259 U.S. 581, 42 S.Ct. 464, 66 L.Ed. 1074 (1922) (jeweler’s directory); List Publishing Co. v. Keller, 30 Fed. 772 (C.C.S.D. N.Y.1887) (social register); see also 1 Nimmer § 2.04[B], at 2-40. Jeweler’s Circular Publishing Co. stated the rule:
“A man’s name, his occupation, his place of business, and his residence are none of them subjects of copyright.”
But, if a man compiles a book containing such information about the residents of a particular place, he may ... copyright it as a whole, notwithstanding the fact that the separate parts of which it is composed are not copyrightable.
281 Fed. at 87. See also Schroeder v. William Morrow & Co., supra, 566 F.2d at 5 (“An original compilation of names and addresses is copyrightable even though the individual names and addresses are in the public domain and not copyrightable”). The Copyright Act of 1976 incorporates the common law by expressly providing for the protection of compilations, 17 U.S.C. § 103(a), defining a “compilation” as follows:
a work formed by the collection and assembling of preexisting materials or of data that are selected, coordinated, or arranged in such a way that the resulting work as a whole constitutes an original work of authorship.
17 U.S.C. § 101.
Thus the law of copyrights defies the laws of logic, or, as one commentator puts it, “the dictates of algebra,” since it “affords to the summation of one hundred or one million [individual facts and their unadorned expression] a significant measure of protection” while affording none to the facts themselves. See Denicola, Copyright in Collection Facts: A Theory for the Protection of Nonfiction Literary Works, 81 Colum.L.Rev. 516, 527 (1981) (hereafter Denicola). The whole of a compilation is thus greater than the sum of its parts. It is not surprising, therefore, that there are opposing lines of cases and competing theories advanced by the commentators, or that this case has caused the district court and ourselves much difficulty.
The competing respectable lines of authority are exemplified by those that seemingly make copyrightability turn on the labor or effort expended in assembling the data, e.g., Leon v. Pacific Telephone & Telegraph Co., 91 F.2d 484, 486 (9th Cir. 1937) (citing Jeweler’s Circular Publishing Co., supra) (telephone directory copyrightable), and those that evidently look exclusively to arrangement of the data as the keystone of protection, e.g., Triangle Publications, Inc. v. Sports Eye, Inc., 415 F.Supp. 682 (E.D.Pa.1976); Simms v. Stanton, 75 Fed. 6, 13 (C.C.N.D.Cal.1896). While the commentators agree that there
we have been particularly restrictive in the protection of non-fiction works indicating, for example, that the fruits of another’s labor in lieu of independent research obtained through the sweat of a researcher’s brow, does not merit copyright protection absent, perhaps, wholesale appropriation____
Nevertheless, our cases do not hold that subjective selection and arrangement of information does not merit protection.
Id. at 862.
The statement of abstract principles, however, may assist in, but usually does not compel, the determination of concrete cases. See Soper, On the Relevance of Philosophy to Law: Reflection on Acker-man’s Private Property and the Constitution, 79 Colum.L.Rev. 44, 64-65 (1979); R. Dworkin, Taking Rights Seriously, ch. 4 (“Hard Cases”), at 81-130 (1977).
Thus, the “pivotal question” as the district court put it is “whether FII’s daily called bond data can be considered a compilation.” Recognizing Moody’s argument as, in essence, “that FII, in publishing its bond data index cards, records only public information about single public occurrences, i.e., bond calls, and does so without adding any ‘peculiar skill and judgment,’ ” the court nevertheless held the argument unavailing on the basis that “[i]n publishing ten or twelve index cards each day, fifty [or] sixty each week, and some 2,500 in each year’s cumulative volume, FII is unquestionably ‘assembling, connecting and categorizing disparate facts which in nature occurred in isolation,’ ” and “[t]he FII employees who put together the daily called bond service assemble into one handy compilation disparate information about thousands of calls by hundreds of municipalities and other governmental entities in dozens of states.”
Thus, “[i]n amassing its bond data, FII is not just assembling a handful of statistics about a single, solitary occurrence, but is instead weaving together into a handy, coherent entity thousands of facts that ‘occurred in isolation.’ ”
And the district court went on to conclude:
To accord copyright protection to the annual compilation but deny it to each daily component would negate the value of the protection accorded the yearly compilation. If a competitor were to be able to copy each day’s quantum of FII’s called bond cards with impunity, the copyright protection given to the annual volume would be rendered meaningless as to that copying competitor____
Just as an annual aggregation of called bond data should be considered a compilation, so, too, should each serial ingredient — be it the quarterly, monthly, weekly, or daily component — be considered a species of compilation. If a cumulative volume warrants copyright protection as a compilation, then so do the serial components that ultimately comprise the cumulative tone — so long as there is adequate copyright notice on each component.
While Judge Carter’s analysis has considerable appeal — why should the serial thief of the data being developed for a compilation go scot-free while the fellow who waits for the compilation to be complete before he steals it does not — and sup
Fair Use
Copyright protection “subsists ... in original works of authorship fixed in any tangible medium of expression____” 17 U.S.C. § 102(a). But, as the Supreme Court has recently reiterated, copyright protection “has never accorded the copyright owner complete control over all possible uses of his work.” Sony Corporation of America v. Universal City Studios, 104 S.Ct. at 784 (footnote omitted). One of the most significant exceptions to the copyright monopoly is the common law concept of “fair use,” codified by Congress in 1976. 17 U.S.C. § 107.
As we have stated, the fair use doctrine “offers a means of balancing the exclusive right of a copyright holder with the public’s interest in dissemination of infor
In Sony, Justice Stevens, writing for the majority, emphasized that a court must employ an “equitable rule of reason” analysis to determine whether fair use operates in a given case as a shield against claims of infringement. 104 S.Ct. at 792. The four factors which the fair use statute identifies as relevant to a determination of whether the doctrine applies — the purpose and character of the use; the nature of the copyrighted material; the amount of the portion used in relation to the copyrighted work as a whole; and the effect of the use on the potential market for or value of the copyrighted work — are equitable considerations to be assessed and weighed by the court;' they are not simply hurdles over which an accused infringer may leap to safety from liability. Rather than a sequence of four rigid tests, the fair use analysis consists of a “sensitive balancing of interests.” Id. at 795 n. 40. On this basis alone, it would not be inappropriate were we to remand for purposes of conducting the required balancing.
But the Supreme Court’s decision in Sony makes a remand on the fair use issue unnecessary. Sony repeatedly stresses that the “first factor” — whether the use is commercial or nonprofit — not only must necessarily be a part' of a court’s analysis, but must also, if the use proves commercial, give rise to a presumption of unfair use. The Court could hardly have been more lucid on this point, stating that “every commercial use of copyrighted material is presumptively an unfair exploitation of the monopoly privilege that belongs to the owner of the copyright____” Id. at 793.
There is no argument and of course can be no doubt but that Moody’s use is commercial, and thus presumptively unfair. The district court, of course, not having the benefit of the Supreme Court’s decision in Sony, noted with respect to Moody’s for-profit use of the copyrighted material only that that fact was not dispositive on the question of liability. To the degree that the district court's discussion of the “public function” of Moody’s use was intended to address the first factor, we must state our disagreement with its characterization. The fact that large financial institutions are willing to pay considerable sums of money to receive information hardly qualifies the selling of that information (again, for a substantial amount of money) as a public function. This information on bond redemptions is not, after all, information that is widely disseminated, and to the degree that it is circulated, it may reach more individuals through their newspapers, where the call announcements first appear, than through any publication that Moody’s produces. To classify Moody’s sales of its
Sony requires that we recognize a presumption of unfair use by Moody’s arising from its commercial use of the copyrighted material. As we engage in the required “rule of reason” analysis mandated by Sony, we are persuaded that Moody’s cannot rebut that presumption, and that if anything, the other specified equitable factors support, rather than undermine, the conclusion that Moody’s use was not protected by the fair use doctrine.
With respect to the second factor — the nature of the copyrighted material — it is true, as appellee argues, that FII’s Daily Bond Cards do require a minimum of editorial skill or judgment, and consequently may be more susceptible to fair use than would be a truly creative work. Harper & Row Publishers v. Nation Enterprises, 723 F.2d at 208. We are disinclined to place great importance on this factor, however, given that the use that Moody’s seeks to make of the material is similarly “non-ereative” and purely commercial. In other words, were a songwriter or playwright to copy from a compilation of information regarding municipal bond redemptions in order to, say, enhance the verisimilitude of his art, we might be more easily persuaded that the factual nature of FII’s copyrighted material should significantly cut against it in a fair use balancing. Given the equitable nature of the fair use analysis, we are somewhat unmoved by the pot here calling the kettle black.
Far more significant to the present case is the evidence of substantial, if not wholesale copying by Moody’s from FII, evidence which we fear the district court failed adequately to credit. From the district court’s opinion, it appears that it understood the expert testimony of Professor Robbins to be an assertion that Moody’s must have copied from FII “91% of the time,” a leap the court refused to make from the established fact that eight of eight errors were copied in one year. But, as we have indicated above, Professor Robbins did not testify that 91% was copied, he merely presented a continuum of what Moody’s might have copied from FII, expressed along with the probability or “confidence level” associated with each possible copying level. The 91% figure referred to was, as appellant readily concedes, at the “statistical fringe.” More to the point was Professor Robbins’s testimony that it was statistically certain (i.e., 95-99% probable) that Moody’s had copied at the 40-50% level.
In this connection, it is important to note that Moody’s never seriously attacked or challenged the expert statistical testimony; it offered no expert witness of its own, and instead sought to prove at trial that it could not have copied more than 50%, due to publication schedules and the like. But as FII correctly notes, Moody’s insistence that one half of what it published could not have been taken from FII is entirely consistent with FII’s proof that Moody’s did steal the other 50%, and FII is also correct to note that the 91% figure attacked by Moody’s is a “straw man.” At oral argument, Moody’s could neither defend nor explain the apparent misunderstanding of the statistical evidence on the part of the district court, nor offer any rebuttal as to why the 40-50% copying figure was not accurate. Accordingly, we simply cannot agree with the district court’s conclusion that Moody’s copying was “limited.”
Finally, with respect to the fourth fair use factor, that of the effect of Moody’s use upon the potential market for or value of FII’s Daily Bond Cards, we again must reject the analysis of the court below that the parties “are not competitors.” It is true that Moody’s News Reports and FII’s Daily Bond Cards differ somewhat in scope
In sum, on the record before us Moody’s does not make “fair use” of the material it copied from FII.
Preemption of State Claim
As noted above, we do not agree with the district court that the parties do not compete, a finding which served as the basis for the dismissal of the state claim. Nonetheless, the Copyright Act, by express terms, preempts state actions with respect to rights “equivalent to any of the exclusive rights within the general scope of copyright.” 17 U.S.C. § 301(a). FII has alleged in essence that Moody’s unlawfully misappropriated its copyrighted material. If FII does prevail on its federal claim, the New York state claim will be preempted; if it does not prevail, the state claim will have to be considered.
Remand for a calculation of damages.
. By stipulation at trial, FII limited its proof of copying to the years 1980 and 1981.
. There is a question whether the issue of the copyrightability of the Daily Bond Cards was preserved on this appeal. In a footnote in its brief on appeal Moody’s does note that it "argued below,” and states that it "maintains,” that the Daily Bond Cards are not copyrightable.
. Section 107 provides:
Notwithstanding the provisions of section 106, the fair use of a copyrighted work, including such use by reproduction in copies or phonore-cords 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 the 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.
. This point was repeated and emphasized throughout the opinion:
If the Betamax were used to make copies for a commercial or profit-making purpose, such use would presumptively be unfair.
104 S.Ct. at 792.
If the intended use is for commercial gain that likelihood [of harm to the copyright owner] may be presumed.
Id. at 793.
Copying for commercial gain has a much weaker claim to fair use than copying for personal enrichment.
Id. at 795 n. 40.
Concurrence Opinion
concurring:
Agreeing with all of what Judge Oakes has written on the fair use aspect of this appeal and with most of his observations on the issue of copyrightability, I concur in the decision to remand for further findings. The usefulness of the remand may be enhanced by some brief further elaboration of the issues.
The basic question on the copyright issue is whether the Daily Called Bond Cards, published by Financial Information, Inc. (“FII”), are a “compilation” within the meaning of the Copyright Act, 17 U.S.C. § 103(a) (1982). Preferring not to extend copyright protection to all collections of facts, Congress has defined a “compilation” as “a work formed by the collection and assembling of preexisting materials or of data that are selected, coordinated, or arranged in such a way that the resulting work as a whole constitutes an original work of authorship.” Id. § 101 (emphasis added). Thus, the primary task for the District Court is to determine whether the data appearing on the Daily Called Bond Cards have been “selected, coordinated, or arranged” so as to constitute an original work of authorship.
I do not share the view that Congress, in establishing these criteria, or this Court, in insisting that they be satisfied, has defined the laws of logic or of algebra. The “whole” of a copyrightable “compilation” is not greater than the sum of its “parts.” If its “parts” are only the discrete items of data that have been collected, the resulting work is not copyrightable. It may receive a valid copyright only if something has been added to the data: the “authorship” of the compiler in making the requisite selection, coordination, or arrangement of the data. The fact that some language in early cases, see Jeweler’s Circular Publishing Co. v. Keystone Publishing Co., 281 F. 83, 87 (2d Cir.), cert. denied, 259 U.S. 581, 42 S.Ct. 464, 66 L.Ed. 1074 (1922), and, perhaps, even some results reached in those cases, e.g., Leon v. Pacific Telephone & Telegraph Co., 91 F.2d 484 (9th Cir. 1937), have supported a view that copyright protection should be extended solely because of laborious effort is no reason for us to disregard the statutory criteria that Congress articulated in 1976 when it enacted the current statute. The “sweat of the brow” rationale is no substitute for meeting one of those statutory criteria.
A second aspect of the copyright issue arises in this litigation because of the fact that the Daily Called Bond Cards are initially issued daily as individual cards and subsequently issued annually in a bound volume. This fact poses issues of procedural compliance with copyright registration requirements and substantive entitle
If annual registration sufficed, the substantive issue is whether the statutory criteria of a protectable “compilation” must be satisfied by each card or only by the series of cards that constitute FII’s called bond “service” and are collected in the annual volume. I am not as sure as Judge Carter that the copyrightability of the annual volume (assuming for the moment that such can be established) necessarily requires copyright protection for the daily cards. The annual volume may satisfy the statutory criteria of a “compilation” even though the daily cards do not. Congress has recognized that the copyright in a compilation “extends only to the material contributed by the author of such work, as distinguished from the pre-existing material employed in the work____” 17 U.S.C. § 103(b). Just as the cards might be copyrightable though the data on them are not, so the annual volume might be copyrightable even if the daily cards are not.
It remains to be determined what copyright protection, if any, the copyright in the annual volume confers upon the daily cards. I doubt that protection of each card may arise simply because denial of protection would impair the value of the copyright in the annual volume. That argument seems to be boot-strapping. However, the daily components of a serial publication perhaps might be copyrightable because of their relationship to each other as issued, as distinguished from their relationship to each other as collected in an annual volume. For example, a mystery story published in the form of daily clues might secure protection for each clue because of its relationship to what has previously appeared and perhaps even to what will appear. The application of that approach to reported facts like those appearing on FII’s cards is arguable, though doubtful since these facts appear to have primary significance as issued each day.
By copying FII’s daily cards, Moody’s is not necessarily a copyright infringer or a “serial thief.” Though it would be liable for conversion if it had stolen from FII’s office the sheets on which the daily data are assembled for publication and might be liable for unfair competition if it appropriated for competitive purposes material not entitled to copyright,
Whether such protection exists cannot be determined until further facts have been found. Principal among these are the facts concerning whatever selection, coordination, or arrangement of the data on the cards may have occurred. I therefore concur in the decision to remand for further inquiry.
. If the Daily Called Bond Cards are not copyrightable compilations, FII’s unfair competition claim under state law might not be preempted either because FII would not be asserting rights "that are equivalent to any of the exclusive rights within the general scope of copyright” or because the cards are not "works of authorship,” 17 U.S.C. § 301. See 1 Nimmer on Copyright § 1.01 [B] at 1-22 (1984) (“if a work does not constitute a 'work of authorship,' ... as a compilation ... under Section 103, then there is no federal preemption”) (footnotes omitted).