Apr. 2003 to
Letter from the American Bar Association Opposing UCITA
(from the Working Group on Consumer Protection, Business Law Section, Committee on the Law of Cyberspace, Subcommittee on electronic Commerce)
June 10, 1999
President Gene Lebrun and Other Commissioners
National Conference of Commissioners on Uniform State Laws
211 e. Ontario St., Suite 1300
Chicago, Illinois 60611
Dear President Lebrun and interested commissioners:
We are writing on behalf of ourselves and the other members of the Working Group on Consumer Protection of the ABA's Business Law Section, Committee on the Law of Cyberspace, Subcommittee on electronic Commerce, to express our opposition to the proposed Uniform Computer Information Transactions Act (UCITA). Our group includes academics who specialize in the law of consumer protection, current and former legal services lawyers, representatives of consumer advocacy organizations, and lawyers in private practice who represent consumers and small customer interests.
1. Overview of Our Criticisms.
Although protesting in its prefatory note that it is a commercial contract law statute, UCITA's primary purpose is to shift the balance of power in mass market software transactions and strengthen the ability of vendors to dictate terms in adhesive form contracts. The effect would be reduced competition among software vendors concerning the products and terms they offer and less incentive to increase quality, avoid interference with fair use rights, and be responsive to customer dissatisfaction.
Our group's members have been involved in commenting upon other recent NCCUSL projects, including Revised UCC Articles 9, 2 and 2A, and the Uniform electronic Transactions Act. Although we did not get everything we wanted in any of these projects, we believe these statutes avoided gross imbalance and shoddy drafting. We cannot say the same about UCITA. A vendor point of view has dominated the drafting process and permeates this statute. It would take a treatise to describe every flaw in UCITA, so we will detail only three points here: contrary to claims, UCITA does not provide new consumer protection; furthermore, its expansive "opt-in" provision threatens the integrity of the law of goods; and finally, UCITA's complex formation provisions provide a good example both of the statute's weak drafting and its policy tilt delegating regulatory power to vendors.
2. UCITA does not provide significant "new consumer protections."
Proponents of UCITA have claimed that it provides consumer protection not available under current law. For example, Reporter's Note 4, UCITA Sec. 105, states that the proposed act contains "new consumer protections." We disagree strongly. Consumers are much better off under current law. Currently courts apply common law and Article 2, either directly or by analogy, to purchases of off-the-shelf software.
A. Right of return. The usual first example of supposedly new consumer protection in UCITA is the right of return in Section 211(b). This provision states that where a vendor first provides a mass market customer with terms after payment, the customer can reject the terms by returning the product for a refund. This is not a new consumer protection. Currently the case law concerning the enforceability of post-payment terms is split, but even the leading case enforcing such terms relies upon the right to reject the terms and return the product for a refund. In ProCD, Inc. v. Zeidenberg, 86 F. 2d 1447 (7th Cir. 1996), the court found assent to terms inside a box based on "[n]otice on the outside, terms on the inside, and a right to return the software for a refund if the terms are unacceptable."
ProCD is the most unfavorable reported case from a consumer protection standpoint, but it is notable that it is more favorable to consumers than UCITA because it makes necessary a notice or disclosure that additional terms will be presented after payment. UCITA does not require this.
The favorable current case law holds that a contract is formed by order, payment and shipment, so that terms that arrive with the product are not binding. See Arizona Retail Systems v. The Software Link, Inc., 831 F. Supp. 759 (D. Ariz. 1993) and Step-Saver Data Systems v. Wyse Technology, 939 F. 2d 91 (3d Cir. 1991). Under this conventional approach, a purchaser of software who was not presented with any terms prior to payment gets all the background rights of Article 2, including an implied warranty of merchantability and a more meaningful right of refund upon rejection. The right of rejection allows the purchaser to inspect the product itself, not just the terms, and obtain a refund if the product does not conform in every respect to the terms of the contract, including the implied warranty. Terms shipped with the product, trying to cut back on the customer's Article 2 rights, do not become part of the contract.
B. Electronic error. Section 217 is the other section usually mentioned as a new consumer protection. It supposedly provides a way for a consumer to extricate herself from being bound by a mistaken on-line order, but the protection from error disappears if the on-line vendor provided "a reasonable method to detect and correct or avoid the error." Thus, using Illustration 3 in Reporter's Note 3 on this section, if a consumer mistakenly types "110" instead of "10" for the quantity in an electronic order, the consumer is bound if the vendor's computer system responded with a confirmation of an order for 110 copies but the consumer did not notice the mistaken quantity in the confirmation.
No reputable vendor would try to make a customer pay for an extra 100, unwanted copies (say of an expensive CD-ROM product) if the customer discovered the error before shipment and gave notice to the vendor or was willing to send back unopened copies. If a vendor was so careless of its own good name as to try to enforce a contract in these circumstances, most courts under current law would be quick to find lack of agreement or misunderstanding (reasoning that the vendor should have realized that an order for 110 copies was not intended by an individual purchaser). The UCITA provision does not qualify as a "new consumer protection."
C. Terms against public policy. Of great importance to the public interest and thus to consumers is the right of fair use of software and works in digital form. UCITA recognizes the public interest in free speech as a reason for limiting the warranty liability of licensors. See Section 402(c) and its Reporter's Note 8 and Section 403(c). But UCITA's concern about free speech seems to only operate in favor of licensors. Users are not adequately protected against boilerplate terms purporting to limit their ability to publish results of tests of software or other criticism (found in several licenses) or to use software for "immoral purposes" (as a leading word processing product's license provides). When copies of software are sold, intellectual property law protects these user freedoms. It also makes unenforceable sweeping contract restrictions on quotation, communication of non-copyrightable information, or reverse engineering to fix problems with software or make products that work with other products. But UCITA raises a question whether "licensing" of software and information permits the licensor to restrict fair use by contract.
To address these concerns, NCCUSL's membership overwhelmingly approved a motion last summer endorsing a provision to authorize courts to refuse to enforce terms "contrary to public policies relating to innovation, competition, and free expression." The drafting committee has since watered down this idea, eliminating the references to particular public policies and adding two new tests, that the public policy must be "fundamental" and the interest in enforcement of the contract term must be "clearly outweighed" by the public policy. These tests create difficult burdens for users to meet in litigation and would chill exercise of user freedoms now taken for granted.
D. Other consumer provisions. Reporter's Note 4 to Section 105 lists a number of other consumer rules, such as in Section 109 (choice of law), 303 (limit on no-oral modification clause), 304 (limit on "modification" of "continuing contract"), and 704 (perfect tender). These provisions at most preserve existing law, often while cutting back on customer rights for non-consumer transactions.
Choice of law and forum. The UCITA choice of law provision for commercial contracts is unlimited and does not require a reasonable relationship between the law chosen and the transaction, as currently required by the UCC. UCITA makes a bow to existing common law authority by codifying a consumer exception making a choice of law unenforceable if it would vary a mandatory consumer rule that would otherwise apply. This is not a new consumer protection, and any benefit from this consumer rule can be eliminated by means of a choice of forum clause. Section 110 on choice of forum has no consumer rule and permits a vendor to choose any state as the exclusive judicial forum if it has a commercial purpose for doing so. See Reporter's Note 3, Section 110. Presumably a vendor's desire to minimize its own litigation costs would be a valid commercial purpose under this section, so that the cost of litigating in a remote forum would be shifted to the consumer purchaser. As a practical matter, most consumers would be denied any relief by choices of remote jurisdictions. Reporter's Note 3 shows no sympathy for this problem, stating, "a contractual choice of forum that responds to a valid commercial purpose is not invalid simply because it has an adverse effect on a party, even if bargaining power is unequal."
Oral modifications. The provision on no-oral modification clauses requires a manifestation of assent by a consumer to a form term prohibiting oral modification. Section 303. Both current and Revised Article 2, in Section 2-209, state a broader customer protection. The current language, in Section 2-209(2), requires that in a form contract between a merchant and a non-merchant, the non-merchant must separately sign the no-oral modification clause. (The UCITA concept of "manifesting assent," discussed below, requires considerably less than a separate signature.) Revised Article 2, Section 2-209(c), expands upon customer protection by precluding a seller from unjustly relying on a term supposedly excluding oral changes if the seller induces reliance by making oral promises. UCITA does not adopt this approach even for consumer transactions.
"Continuing contracts." The UCITA "consumer protection" provision on "modification" of "continuing contracts" is a good example of a rule that is not even reasonable for non-consumer contracts. It provides that if a vendor announces a unilateral change (for example, by cancelling part of what is available in an on-line database), a consumer customer has no right to cancel as to future performance without meeting the burden of showing that the change was "material." Section 304(b)(2). Non-consumers are bound even if material changes are made by the vendor, so that apparently the vendor could double the price in the middle of the contract period. Current and Revised Article 2 have no analogue to Section 304 and require agreement of both parties for a change to be a modification. See Section 2-209, current and Revised Article 2.
Perfect tender. Article 2 has a perfect tender rule for consumer and non-consumer transactions. See current Section 2-601 and revised Section 2-703(a)(2). UCITA restricts this rule to mass-market transactions for a single copy. Section 704(b).
Unconscionability. Finally, UCITA does not adopt the modest new codified consumer protections in Revised Article 2, which are based on existing case law. The most important of these is a recognition of substantive unconscionability in limited situations. In light of current practices, this provision is as much or more needed in software transactions as in sales of goods. Effective consumer protection requires a means of oversight of nonnegotiated forms terms that impose "manifestly unreasonable risk or cost" on consumers, as Revised Article 2 provides in Section 2-105(b)(3).
3. The opt-in scope provision is unacceptable.
Many persons have commented on problems with UCITA's scope provision, Section 103. From a consumer perspective, the worst part of the scope section is its "opt-in" provision, Section 103(e). This provision threatens to swallow up the law of goods and undermine Revised Article 2.
Section 103(e) allows a seller of goods to opt into UCITA if the transaction involves "information." "Information" is a term defined in Section 102(a)(37) to include data, text and images. Any goods with text or images, from books and magazines to t-shirts and plates, would qualify under this language, permitting opt-in to UCITA. When asked about use of the term "information" in this subsection on the floor of the American Law Institute annual meeting in May, the drafting committee chair and reporter indicated that "computer information" was the term intended. But the NCCUSL annual meeting draft continues to use the term "information."
Assuming that "computer information" is in fact the term intended, Section 103(e) allows opt-in to UCITA for a transaction if any part of it, no matter how trivial, involves computer information. Computer sellers who include an operating system (as nearly all do) could opt into UCITA. Many other goods also contain software–cars, television sets, cameras, and a number of appliances. In fact, as the opt-in provision is drafted, a loophole exists to allow any goods transaction to be put under UCITA–by throwing a CD-ROM into the deal. For example, furniture could be sold with a CD-ROM giving cleaning or assembly instructions, allowing opt-in to UCITA.
Furthermore, the seller could opt into UCITA by an inconspicuous term in a standard form first disclosed after payment. The opt-in provision explicitly allows a form drafter to opt into UCITA's contract formation rules. For example, a seller of a car could opt into UCITA by standard terms put in the glove compartment or on the dash-board computer screen. Consumers of course would have no idea what opt-in meant.
It would be impossible even for lawyers to feel confident about the full implications of opt-in, so that doing so would be a possible strategy to make consumer redress more expensive and difficult to obtain. Section 103(e)(1) provides that an "agreement" to opt into UCITA "does not alter an otherwise applicable rule that may not be varied by agreement." Under this language, what parts of Article 2 would still apply to a sale of goods where the seller opted into UCITA (on the basis of an incidental piece of software)? UCITA provides a lower standard of merchantability ("reasonably fit for the ordinary purpose for which it is distributed") than Article 2 ("fit for the ordinary purposes for which such goods are used"), both by using the qualifier "reasonably" and by taking the vendor's perspective at distribution, rather than focusing on the ordinary uses by customers. The Article 2 warranty, like the UCITA warranty, is variable–it can be disclaimed. Could a seller of goods lower the standard of merchantability by opting into UCITA? If a seller of consumer goods opted into UCITA, what impact would this have on the federal Magnuson-Moss Warranty Act, which applies when a warrantor provides a written warranty for a consumer product? Magnuson-Moss requires that such a warrantor at least provide an implied warranty of merchantability. Would opt-in to UCITA by a seller of goods mean that the required merchantability warranty under Magnuson-Moss is downgraded to the UCITA standard? This is arguable because the warranty of merchantability under Magnuson-Moss is defined by reference to state law. 15 U.S.C. section 2301(7).
Opt-out is also problematic. A partial opt-out is permitted: "the parties may agree that this Act ... governs the transaction in whole or in part." A few sections are listed in Section 103(e)(2) as unalterable by agreement, but not included in this unalterable list is Section 211, the refund right. Would it be permissible to opt out of the refund right in Section 211(b)?
4. The technical quality of UCITA is poor (formation provisions as an example).
The technical quality of UCITA, in its structure and in many individual provisions, is so poor that it would create great uncertainty and resulting cost. Consumers would pay this cost in at least two ways–in the price of products and in the price of obtaining legal services to resolve disputes.
The formation provisions illustrate the problem. UCITA contains a complex section defining "manifesting assent." Section 112. This multi-subsectioned provision, also elaborated upon in a bewildering maze of 16 Reporter's notes, uses many defined terms and terms of art, including "opportunity to review," "authenticates," "adopt," "accept," "assent," and "electronically reaffirms assent," without communicating why there is a need for a definition. Neither current or Revised Article 2 uses this term, instead simply saying that a contract "may be made in any manner sufficient to show agreement, including by offer and acceptance, or conduct of both parties which recognizes the existence of" a contract. Current Section 2-204(1) and Revised Section 2-203(a). Indeed, UCITA also uses this much simpler formulation in Section 202(a). Why not junk the messy "manifesting assent" section and just use the phrase "show agreement"?
Claims are made in Reporter's Note 2 to Section 112 that the term "manifesting assent" comes from the Restatement (Second) of Contracts, Section 19, and that the UCITA definition "corresponds to the Restatement." But the comments to the Restatement section, on "Conduct as Manifestation of Assent," make clear that it has very different contexts in mind than UCITA. UCITA treats post-payment actions such as a mouse click during installation or opening a package as manifestations of assent, while the Restatement's examples involve requesting services, in the context of a long-term relationship, or initiating communication of an offer. Not one of the Restatement's examples involves conduct by a customer in response to form terms drafted by a commercial vendor, let alone customer conduct after form terms are supplied post-payment.
The use of the phrase "opportunity to review" in Section 112 is a rhetorical maneuver that evokes the controversial, counter-factual standard of a "duty to read" that most consumers do not meet when entering into form contracts of any length. But the true import of the phrase only becomes clear when one finds out in subsection 112(e)(3) and later in Sections 210 and 211, that an "opportunity to review" exists even though a vendor withholds already drafted terms until after payment.
Reporter's Note 3 to Section 210 explains this conceptual maze in terms of "layered contracting," stating that some contracts involve "a rolling or layering process. An agreement exists, but terms are clarified or created over time." This is indeed a valid commercial contracting practice, but it applies to situations where it is not possible or desirable to attempt to spell out everything in advance, for example in construction or long-term supply contracts. UCITA hijacks this idea and applies it in Section 211 to one-shot purchases of mass-market products where the vendor knows the terms it wants and has them pre-printed or saved in computer memory, ready to go, but holds them back until after the purchaser pays and feels psychologically committed. The vendor's objective is not to work things out over time; rather, holding back terms has the effect of minimizing attention to unfavorable terms and making it difficult to shop for better ones. A customer would have to engage in repetitive purchases to compare the deals offered by various vendors.
There is a need for pre-transaction availability of terms in order for a competitive market to operate. Even if only a fraction of customers comparison shop, this can introduce some market discipline. In the face of this unassailable point and after repeated urgings during the drafting process, UCITA's great concession is a safe harbor for Internet transactions. Section 213 provides that if the vendor makes the terms available before payment, then an opportunity to review has been provided. Would anyone doubt this statement of the obvious? Buried in Section 213(1)(B), however, one finds a big loophole: an Internet vendor can satisfy the safe-harbor provision not only by posting the terms, but also by disclosing their availability and providing them on request. Thus, the language permits the safe harbor to be met by posting a notice saying the terms are available by mail (including snail-mail). The terms do not have to be posted electronically, on a Web site. This is the Internet safe harbor; under UCITA, an Internet vendor still could choose not to disclose even the availability of terms in on-line transactions and instead send them with the product.
Apart from bad policy choices, UCITA's implementation of these choices is unclear. Not only does the definition of "manifesting assent" attempt to coexist with the clearer language "show agreement," but UCITA's authorizations of post-transaction disclosure of terms in Sections 112, 210 and 211 sit uncomfortably alongside a more generally accepted version of contract formation found in Section 203. This UCITA section picks up language from Article 2, current Section 2-206(1) and Revised Section 2-205(a), and provides that a contract is formed by order and shipment "unless otherwise unambiguously indicated." In litigation, customers can invoke this familiar idea, and courts baffled by the concept of "layered contracting" may choose the entirely sensible course of ignoring it, or more elegantly, finding the sections based on it inapplicable in cases involving formation by order and shipment.
Because the problems with UCITA are profound and structural, they cannot be resolved by tinkering with language here and there. We urge you to disapprove the draft. That would leave the common law to further explore the issues involved in software and other "computer information" transactions before codification. More than 75 years ago, Karl Llewellyn, the principal architect of the Uniform Commercial Code, wrote in his book, The Case Law System in America, that it is necessary to have a period of case law experimentation before a new development can be properly addressed in a statute: "At the development's start, both the insight and experience necessary to create a statute are lacking. A statute passed under such circumstances is a far greater misfortune than any misstep taken by a case law court."
Some have suggested that this flawed statute should be approved by NCCUSL and fixed later. But it will be more difficult, not easier, to get consensus for sound and balanced legislation if UCITA becomes a standard held up by vendor advocates who have resisted compromise to date.
Members of the Working Group on Consumer Protection,
American Bar Association Business Law Section,
Committee on the Law of Cyberspace,
Subcommittee on electronic Commerce
||Mark Budnitz, Co-chairs
UCITA Pages On
James S. Huggins' Refrigerator Door