By James Grimmelmann, Associate Professor, New York Law School.
David Balto's reply to my Issue Brief on the proposed Google Book Search settlement is careful and thoughtful. Unfortunately, it gets some of its analysis of the settlement's anticompetitive effects wrong. I'll respond to three of the points on which I believe he's mistaken.
First, Balto's discussion of barriers to entry makes an unwarranted leap. He writes, "First, Google will affirmatively not obtain a monopoly over orphan works because the settlement does nothing to make entry more difficult for a second entrant."
He's right about making things more difficult, but wrong about the monopoly. The settlement gives Google a monopoly not by raising barriers for other entrants, but by preferentially lowering them for Google. The barriers to entry for the large-scale market in selling orphan works are currently prohibitive; no one at all can legally compete in it. The settlement opens it up only to Google; and the result will be a monopoly. (I was sloppy about my phrasing in the Issue Brief here -- I said the settlement "creates" a barrier to entry, which may be the source of Balto's more substantive error.)
Here's an analogy to make this distinction clear. Imagine that it were generally illegal to buy or sell milk. A company -- call it Moogle -- obtains from the government a special authorization to start selling milk. Moogle hasn't done anything to raise entry barriers for its competitors; the barriers were forbiddingly high to start with. But as a result of this special dispensation, Moogle now has the milk market all to itself. That's a monopoly.
This leads into a second way in which Balto's argument is questionable. He believes that Google's success in obtaining this settlement means that other potential book-scanners will be similarly able to obtain settlements authorizing them to sell books. He focuses on class certification, which he argues would be readily achievable for a hypothetical second scanner (let's call it "Two-gle"), once Google's precedent is on the books.
I have my doubts. There are serious grounds to question whether Two- gle's motion for class certification would be as trivial as Balto suggests. The Google case is already pushing at the limits of class action law. The parties are spending tens of millions of dollars in legal fees getting this class certified -- and that's with cooperative plaintiffs. Two-gle could quite plausibly face plaintiffs who like the Google deal well enough not to welcome competition. They could sabotage Two-gle's efforts in all sorts of ways: drafting idiosyncratic and fact-specific complaints, fighting any motion to force a class on them, or flatly refusing to negotate for the kind of settlement Google got.
In any event, if, as Balto argues, future Two-gles would be able to obtain similar settlements so readily, then presumably there should be no objection to offering them the same deal now, without having to go through the expense and waste of a lawsuit. That's what the non- exclusivity I recommended in my Issue Brief would consist of. No one has convincingly explained why it would pose any sort of problem.
Third, Balto's analysis of the leading antitrust case, Broadcast Music, Inc. (BMI), is misleadingly incomplete. There, BMI (along with ASCAP) offered public performance licenses for copyrighted music. It had a huge catalog of songs licensed from copyright owners, and it would sell "blanket" licenses to television networks, giving them the right to broadcast any song in its catalog. CBS sued, arguing that blanket licensing was a form of price-fixing. In the well-known passage that Balto quotes, the Supreme Court held that BMI's decision to offer only blanket license was pro-rather than anti-competitive.
It's true that BMI blessed blanket licenses, but under the settlement only the library subscriptions will involve blanket licenses. The sales of individual books to readers are akin to traditional book sales by individual copyright owners. That's exactly the distinction the Court drew in a later section of the opinion, explaining that the blanket license "is "quite different from anything any individual owner could issue." Thus, BMI itself calls the individual purchase part of the settlement into question.
As for the Book Rights Registry the settlement will create, Balto is puzzled that I both criticize its monopoly position and want to expand its role. No mystery: I'm merely proposing to treat it as a regulated monopoly, a familiar beast from public utility law. The Court in BMI accepted this framing when it pointed to the "substantial restraints"
of the consent decree BMI operated under as a justification for allowing BMI a free hand.
Balto cites the Copyright Clearance Center (CCC) as a similar but harmless entity offering nonexclusive licenses to copyrighted works. Unlike the CCC, however, the Registry will have the rights to copyrights no copyright holder ever agreed to assign it. For orphan works, those rights are effectively exclusive. The Court in BMI pointed out that CBS had a "real choice" of where to obtain rights to individual songs; for orphan works under the settlement, there is no alternative licensor.
Balto's Columbus metaphor is an ironic choice. He writes, "Like 'the earth is flat' critics of Columbus' journey, these critics lack the vision to see the truly beneficial aspects of the settlement." But Columbus's critics weren't flat-earthers; they knew that the earth was roughly 25,000 miles around. Columbus was the mistaken one; a combination of calculation and translation errors apparently led him to believe that the true circumference was much smaller and the Indies much closer. He got lucky, though; there turned out to be a continent in the way of his badly planned circumnavigation. What Balto calls lack of vision, I prefer to think of as drawing sound conclusions on the basis of all available evidence.
In the end, Balto says that the settlement should "unquestionably" be approved. Beware all such absolute confidence. The settlement is immensely intricate and amazingly ambitious. It offers obvious and substantial public benefits, but also comes with subtle and wide-ranging risks. Questions, and plenty of them, are entirely appropriate.