Federal Trade Commission Received Documents July 1, 1996 P894219 B18354900195 Collier, Shannon, Rill & Scott, PLLC Attorneys at Law July 1, 1996 Mr. Donald S. Clark Secretary Federal Trade Commission 6th Street and Pennsylvania Avenue, N.W. Room 172 Washington, D.C. 20580 Re: Made in USA Policy Comment, FTC File No. P894219 Dear Secretary Clark: These comments are submitted on behalf of the Bicycle Manufacturers Association of America, Inc. ("BMA") to respond to specific questions posed by the Federal Trade Commission in its Federal Register notice of April 26, 1996.(1) BMA is a non-profit trade association representing the interests of U.S.-based bicycle manufacturers accounting for more than 90 percent of U.S. bicycle production and their suppliers. Previous comments submitted by BMA set out in detail the association's positions on the use of "Made in USA" claims.(2) Thanks in large part to its participation in the Commission's Workshop on this subject, BMA has come to appreciate the complexity of the various issues surrounding "Made in USA" claims and the varying perspectives of different U.S. manufacturing interests on this subject. Accordingly, while BMA continues to believe that certain criteria for "Made in USA" claims are most appropriate in the context of bicycles, the association now advocates the adoption of a "reasonable basis" approach in evaluating "Made in USA" claims generally. Attached to these comments is a paper expanding upon the presentation by Dr. Howard Beales, III, at the Commission's "Made in USA" Workshop. In this paper, Dr. Beales demonstrates that the available consumer perception data provide no basis for the Commission's longstanding "all, or virtually all" standard for assessing "Made in USA" claims and explains the appropriateness of the "reasonable basis" approach. Since the Workshop, BMA has been part of an ad hoc group of companies and industry associations with an interest in the development of a set of guidelines for assessing U.S. origin marketing claims that is modeled on the FTC's successful "Guides for the Use of Environmental Marketing Claims."(3) The product of that collaborative effort Ä a document entitled "Guidelines for Making U.S. Origin Advertising and/or Labeling Claims" Ä has been filed under separate cover on behalf of BMA and the other subscribing parties. Among other things, the proposed Guidelines establish various safe harbors for "Made in USA" claims, one of which is compliance with a modified version of the tariff preference rules under the North American Free Trade Agreement Ä the particular criterion favored by BMA in its prior submissions. BMA believes that these proposed Guidelines reflect a broad consensus of American industries concerning the proper use of U.S. origin claims and urges the Commission to give them serious consideration. Below, BMA has addressed several of the questions in section III of the Commission's April 26 Federal Register notice. 3. Substantial Transformation Standard A.ii. A number of participants at the Commission's Workshop suggested that the substantial transformation (or tariff shift) test should be adopted, but with minor alterations to assure that a product labeled "Made in USA" in fact had a meaningful amount of domestic content. Should the FTC adopt a modified version of the substantial transformation test applied by the U.S. Customs Service? As an integral component of the "reasonable basis" approach, BMA recommends that the Commission carve out one or more safe harbors for products that satisfy a hybrid standard that combines a modified version of the substantial transformation test with a standard that ensures the addition of significant U.S. value. By definition, a "Made in USA" claim for a product meeting this hybrid standard would not be deceptive. For products not meeting the safe harbors, the Commission would require that a manufacturer adequately substantiate its claim that the product is "Made in USA." At the Commission's Workshop, a number of participants argued forcefully that the Commission's "all, or virtually all" standard for "Made in USA" claims should be harmonized with the substantial transformation standard long used by the U.S. Customs Service (and its foreign counterparts) in connection with claims of foreign origin. To be sure, there are extremely powerful reasons to endorse such harmonization. For one (as will be discussed in more detail below), it is difficult (if not impossible) to contend that consumers reasonably could interpret a statement that a product is "Made in USA" as any different in basic meaning from the statement that a product is "Made in Japan." But more importantly, the available consumer perception data suggest that consumers understand "Made in USA" claims, in particular, to describe where a product "came into being" Ä an idea similar in concept, we believe, to the traditional notion of "substantial transformation" (i.e., the creation of a "new product" with a "new name, character and use"). At the same time, BMA believes that consumers recognize that the process of "making" an article (or "bringing it into being") generally involves a series of steps, and it is more the exception than the rule in today's complex economy for a single party to undertake all of them. While we believe there would be general agreement that the "making" or "coming into being" of a finished product most clearly occurs more toward the end of the overall process (e.g., with the final assembly of components), we believe that consumers would also generally recognize that a large portion of the total value of the article might well be attributable to earlier stages of the process (i.e., the fabrication of parts or components). The "substantial transformation" concept employed in customs law does not necessarily acknowledge this, as the process of final assembly alone Ä for example, the assembly of finished bike parts into a complete bicycle Ä is frequently treated as conferring origin for customs purposes. In BMA's view, a simple substantial transformation standard Ä for example, the simple tariff shift rules adopted for marking purposes under the NAFTA Ä is not suitable for use as a criterion for assessing "Made in USA" claims. While processing that counts as a "substantial transformation" often entails the addition of significant value to a product, many processes (e.g., the act of final assembly) often do not, despite "bringing a new product into being." BMA believes that consumers are entitled to expect that a claim that a product was "Made in USA" means not only Ä but most fundamentally Ä that the product "came into being" (i.e., was substantially transformed) here, but that substantial value was added in the United States. This would normally require the addition of U.S. material and labor value beyond that entailed in the final assembly process (e.g., the fabrication of components used in the product). In the proposed Guidelines developed by the ad hoc industry group, each of the three "safe harbors" acknowledge this principle, albeit in different ways. Certainly, it is made most explicit in the "Value Added" safe harbor, which requires, among other things, that the cost of U.S. processing account for at least 50 percent of the cost of goods sold. Similarly, the "NAFTA Preference Rules" safe harbor is predicated upon compliance with a modified version of the tariff shift and/or value-added requirements that determine eligibility for preferential duty rates under the NAFTA Ä rules that generally require significantly more processing than would ordinarily be required for a simple "substantial transformation." And finally, while the "Majority of Processing" safe harbor deliberately avoids resort to a quantitative value analysis, its essential requirement Ä that "a majority of all the processing that is normally undertaken to produce a product (including the last significant manufacturing process or processes) take place in the United States" Ä effectively ensures, we believe, that the processing performed in the United States will account for a substantial portion of the product's total cost of goods sold.(4) B. How does a substantial transformation standard in any of the variations discussed above relate to consumer perceptions of "Made in USA" claims? Does empirical evidence suggest that consumers think about the phrase "Made in USA" in terms of the process by which parts or materials are transformed into a finished product? Does empirical evidence suggest that consumers think the phrase "Made in USA" refers both to the transformation process and the origin of the parts and materials themselves? As discussed in detail in Dr. Beales's appended paper, the great majority of consumers appear to interpret "Made in USA" claims to mean that a product so labeled was "brought into being" in the United States. In fact, very few consumers thought the claim referred to parts until prompted to think in terms of parts by the questioner. Fundamentally, then, the empirical evidence suggests that consumers conceptualize "Made in USA" claims in terms of the process by which parts or materials are transformed into a "new and different" finished product Ä that is, "substantially transformed." This suggests that there is strong empirical support for the legal standard governing foreign origin claims (e.g., "Made in Japan"), but only scant empirical support for the "all, or virtually all" standard that has been employed by the Commission to assess claims of U.S. origin. The evidence suggests that only an insignificant minority (i.e., less than 10 percent) of consumers interpret "Made in USA" claims as meaning that "all, or virtually all" of the parts and materials in the product originated in the United States. Notwithstanding the foregoing, it must be admitted that the phrase "Made in" is inherently ambiguous, if for no other reason than what we characterize as "making" a product covers an enormously wide array of processes, some of which are quite simple (e.g., making cookies) and others of which are enormously complex (e.g., making a computer). There is no particular reason to believe that even individual consumers employ a single concept of "making" in these various contexts. To make cookies, one needn't grind wheat into flour, but one must certainly do more than spoon frozen cookie dough onto a baking sheet; the latter may arguably be termed "baking cookies," but it certainly would not strike most consumers as "making cookies." Similarly, one could reasonably claim to have "made" a computer without having manufactured the semiconductors or the chips oneself; but on the other hand, creating a finished computer in ten minutes simply by joining together half a dozen separate finished subassemblies is not "making" a computer. It might fairly be characterized as "assembling" a computer, but it is not "making" a computer in any meaningful sense. These distinctions have tremendous significance in the context of country of origin claims in general and "Made in USA" claims in particular. While a claim about where a product is made is fundamentally a claim about where certain processes occurred, rather than a claim relative to the origin of parts or materials, it would be wrong to say that the origin of parts or materials is wholly irrelevant to the issue. Indeed, BMA submits that there is something of an inverse relationship between the two; the more complex the processing, the smaller the bearing that the origin of parts or materials has upon the question of where the product was "made." In some cases, a product with virtually no U.S.-origin material in it (e.g., a hand tool manufactured entirely in the United States using imported steel billets as a raw material) would merit a "Made in USA" claim. In other cases (e.g., the assembly of a bicycle entirely from finished components of foreign origin), a "Made in USA" claim would be both misleading to consumers and unfair to any U.S. firm that actually "made" bicycles by engaging not merely in final assembly, but in part fabrication as well. If there is any common thread to these examples, it would seem to be that the foreign origin of fabricated parts may have a bearing on U.S. origin claims, but that foreign-origin materials do not to the extent that they are "substantially transformed" by the U.S. processing operation. But in neither event is there evidence that consumers necessarily interpret a "Made in" claim as meaning that "all, or virtually all" of either the materials or fabricated parts have the same origin as the finished article. C. Is there evidence as to whether consumers' understanding of "Made in USA" claims is the same or different than their understanding of foreign origin claims (e.g., "Made in Japan")? Is there evidence as to whether claims of foreign origin are as material to consumers across all or most products as are claims of domestic origin? Please provide any supporting documentary evidence or citations. As discussed above and in the appendix, the Commission's consumer survey suggests strongly that the majority of consumers interpret the phrase "Made in USA" as meaning that the product "came into being" in the United States Ä a notion quite similar to the "substantial transformation" concept that has long operated as the legal criterion for foreign country of origin marking claims. While it would be naive to assume that consumers have a complete understanding of the "substantial transformation" concept, this evidence suggests that there is no empirical basis for the Commission's "all, or virtually all" standard for "Made in USA" claims. Rather, the evidence suggests that consumers understand "Made in USA" claims in precisely the same way that they understand "Made in Japan" or "Made in Taiwan" claims. In neither case does an appreciable percentage of consumers understand these statements as meaning that "all, or virtually all" of the product Ä in the sense of not only the finished product, but all of its materials and parts Ä originated in the country in question. That the evidence should suggest as much should hardly be surprising, as it would be logically impossible for a "Made in USA" claim to mean something fundamentally different from a "Made in Japan" or a "Made in Taiwan" claim Ä at least in the absence of clear evidence that consumers understand the statements to mean different things despite their identical terminology. There is, of course, no such evidence; indeed, all of the evidence is to the contrary. The consumer perception data suggest that consumers understand all "Made in" claims in fundamentally the same way; whether the site of the "making" is America or some foreign country, this statement communicates where the product "came into being." This is not to say that the legal criteria governing statements of foreign origin should necessarily be identical to those governing claims of U.S. origin. It is well-established that the purpose of the United States' country of origin marking statute (19 U.S.C.  1304) is to give consumers a means of distinguishing foreign-made articles from those of U.S. manufacture, the assumption being that many consumers will prefer to buy the latter over the former. See United States v. Friedlander & Co., 27 CCPA 297, C.A.D. 104 (1940). Under this assumption, which foreign country is the legal country of origin is less important than the fact that some foreign country is identified as the source of the product. The relatively low threshold imposed by the judicially-created concept of "substantial transformation" can be seen as based in large part on the pragmatic need for a test that yields meaningful and consistent results without the need for too extensive a probe of the total process by which a product was created, given the legislative mandate that all imported articles be labeled as such. To some extent, a different set of considerations is relevant in assessing "Made in USA" claims. No provision of U.S. law requires American-made goods to be labeled as such. Because the impact of such claims on consumers is presumptively positive, it is reasonable to impose a somewhat more rigorous standard on such claims. In particular, it is appropriate to insist that such claims mean not only that a product "came into being" in the United States (i.e., was substantially transformed here), but that that "coming into being" entailed a substantial amount of processing, with the attendant contribution of significant U.S. labor and material value. It is one thing to permit a "screwdriver operation" in Hong Kong to confer Hong Kong origin on Chinese-made components; it is another thing for a "screwdriver operation" in the United States to be used as a means of "passing off" a largely Chinese-made product as "U.S. made." 4. Other Issues B. Are there terms that are, or can be, used to denote some lesser or different level of domestic content than a broad "Made in USA" claim, e.g., "Assembled in USA," "Product of USA," "Processed in USA," etc. What are the costs and benefits of using such alternative terms to label products that would not meet a standard for "Made in USA" claims but nonetheless involve some significant domestic inputs? A basic premise of the proposed Guidelines for U.S. origin claims developed by the ad hoc industry coalition is that marketers who wish to make a claim that is either stronger or weaker than an unqualified "Made in USA" claim should be entitled to do so by making an appropriately qualified claim. For those wanting to make a stronger claim because, for example, "all, or virtually all" of a product's materials or components are of U.S. origin should be free to communicate that message to consumers by using "Wholly Made in USA" or some equivalent claim. Similarly, marketers that conduct the final portion of the overall manufacturing process in the United States Ä but not enough to warrant an unqualified "Made in USA" claim Ä should be permitted to provide consumers with a factually accurate description of what it is they do in the United States short of "making" the product here. In BMA's view, statements such as "Assembled in USA" and "Processed in USA" communicate a message distinct from "Made in USA."(5) Most consumers, we believe, would interpret these statements as different from a "Made in USA" claim; the implication is that the U.S. operation is confined to the final assembly or final processing of mostly or entirely foreign-origin components. At the other end of the process is a "Designed in USA" claim, which means Ä literally, at least Ä that the product was merely designed in America but could very well have been manufactured entirely overseas. There is, however, a significant risk that these lesser claims may be misinterpreted by a significant number of consumers unless they are further distinguished in some way from unqualified "Made in USA" statements. Indeed, the Commission's consent decree in the Hyde case appears to view the terms "Made in" and "Assembled in" as totally interchangeable. See 59 Fed. Reg. 48,892 (1994). To treat them as such, we submit, is to devalue the "Made in" claim, putting bona fide U.S. manufacturing operations on a par with simple "screwdriver" plants that "make" products in the United States simply by assembling foreign-origin parts and components. While marketers wishing to boast of these relatively modest investments in the United States should be free to do so, BMA believes strongly that it is essential that their claims be accompanied by an express reference to components of foreign origin in order to reduce the risk that consumers will misinterpret them as the equivalent of a "Made in USA" claim. Similarly, any "Designed in USA" claim made in connection with a product that does not merit an unqualified "Made in USA" claim should be accompanied by an explicit statement that the product is of foreign origin (e.g., "Designed in USA. Made in China"). The primary benefit of these various qualified and unqualified claims should be obvious: they give consumers more Ä and more accurate Ä information about the nature and extent of a product's connection to the United States. In any particular product market, goods labeled "Made in USA," "Wholly Made in USA," or "Assembled in USA" may well be found. Use of such claims may, in turn, generate market forces encouraging greater investment in U.S. manufacturing and assembly operations. C.i. Do consumers generally believe that unlabeled products are domestic? Does consumer perception of the origin of unlabeled products vary by type of product? In BMA's view, consumer perception of the origin of unlabeled products varies, depending largely upon the extent to which foreign-made products (labeled with their countries of origin) are present in a particular market. Where foreign-made products are a small or nonexistent portion of a product market, consumers are somewhat more likely to infer that all unlabeled products are of American origin. This inference becomes weaker where a substantial portion of the market is composed of products labeled with foreign origins. Similarly, where both imported and U.S.-made products share some of the same foreign-origin components, consumers may infer incorrectly that both finished products are imported. In such a market, it becomes more important for domestic producers to distinguish their products affirmatively by labeling them as "Made in USA." BMA believes strongly that the "reasonable basis" approach reflected in the proposed "Guidelines for Making U.S. Origin Advertising and/or Labeling Claims" submitted by the ad hoc industry group will ensure that products labeled "Made in USA" reflect consumers' understanding of the term "Made in USA." Consumers understand "Made in USA" to mean where a product was "brought into being." At the same time, the Guidelines recognize that consumers justifiably expect "U.S.-made" products to be the result of significant U.S.-based processing that entails the addition of significant U.S. value. A "reasonable basis" approach to evaluating "Made in USA" claims will ensure that consumer expectations regarding the origin of products are satisfied. Respectfully submitted, MICHAEL R. KERSHOW JUDITH L. OLDHAM BRIAN A. DAHL Collier, Shannon, Rill & Scott, PLLC 3050 K Street, N.W. Suite 400 Washington, D.C. 20007 (202) 342-8400 Counsel to the Bicycle Manufacturers Association of America, Inc. Enclosure APPENDIX ANALYSIS OF CONSUMER PERCEPTION DATA RELATIVE TO "MADE IN USA" CLAIMS by Howard Beales, III, Ph.D.(1) 1. INTRODUCTION These comments are submitted on behalf of the Bicycle Manufacturers Association of America, Inc. ("BMA") to supplement the record following the Federal Trade Commission's public workshop on "Made in USA" advertising. The purpose of these comments is to document and expand upon my presentation at the workshop regarding the Commission's consumer perception data as well as a discussion of other similar data placed on the record by interested parties. The Commission's Copy Test data clearly indicate that only an insignificant minority of consumers understand "Made in USA" to mean that 100 percent of the parts and labor of a product bearing such a claim are of United States origin. The majority of consumers understand "Made in USA" as a reference to where the product "came into being." This is a concept requiring, at a minimum, that significant steps in creating the final product occur in the United States. Based on the results of the Commission's Copy Test and other evidence in the record, the Commission should not adhere to an "all, or virtually all" standard that protects only those few consumers who might interpret "Made in USA" claims in this manner. As it does in other areas, the Commission should follow a "reasonable basis" approach when determining whether "Made in USA" claims are adequately substantiated.(2) Such an approach would allow advertisers to make "Made in USA" claims that are substantiated if the advertiser has fulfilled the requirements of certain existing regulations relating to origin claims (e.g., the duty preference rules developed in connection with the North American Free Trade Agreement ("NAFTA")). 2. THE CORE MEANING OF "MADE IN USA" IS THAT THE PRODUCT WAS MANUFACTURED OR "BROUGHT INTO BEING" IN THE UNITED STATES a. The Copy Test Results Demonstrate the Importance of the Process of Making the Product When asked generally what "Made in USA" means, a substantial majority of Copy Test respondents (63.5 percent) either simply repeated the phrase "Made in USA" or responded with a virtually identical phrase.(3) See Chart 1. Presumably, these respondents interpreted the phrase using commonly understood meanings of the word "made," which focus on creating or causing to come into being. A good indication of what respondents meant when they provided the "Made in USA" answer reflected in Chart 1 can be found by consulting a dictionary. Included among the ordinary dictionary definitions of the word "made" are the following: 1. a. To cause to exist or happen; create. b. To bring into existence by forming or modifying materials: make a coat. c. To create by putting together component parts.(4) The dictionary meaning of the word "made" is important in this context not necessarily to explain what the phrase "Made in USA" itself means, but rather, to offer insight into what respondents meant by the word "made" when they parroted the phrase back to the interviewer. The example of making a coat is especially helpful in understanding how respondents likely were defining "made" when they repeated the phrase "Made in USA." A person who "makes" a coat brings it into existence by forming materials into a finished product. They generally do not, however, shear sheep to make cloth and thread, or mold plastic into buttons, or forge metal into zippers. On the other hand, most would agree that simply sewing a zipper into an otherwise fully assembled coat would not be "making a coat;" this would likely be regarded as "finishing a coat." The process of "making a coat" is commonly understood to entail more significant processing than this Ä for example, the cutting of cloth using a purchased commercial pattern. Another example more common in the experience of most consumers would be to "make a meal." Most consumers at some point have made a meal, but that does not mean they raised and slaughtered livestock, grew vegetables, or crushed grapes for wine. On the other hand, most dinner guests would feel severely cheated if the "homemade meal" they were served were simply a heat-and-serve frozen meal. But between these extremes, there is broad variation in the concept of "making a meal." Must the bread be baked from scratch and the bernaise sauce made from fresh eggs for the meal to have been "made" by the purported chef? We think most consumers would probably say no, but would nonetheless be more impressed if they learned this were the case. And this, we submit, is an important distinction that provides useful guidance for how the Commission should approach regulation of "Made in USA" claims. These examples help demonstrate that Copy Test respondents who explained the meaning of "Made in USA" by simply repeating the phrase "Made in USA" likely meant that the product bearing such a claim came into being in the United States as the result of some significant processing operation. Consumers who understand the phrase "Made in USA" in this way also will answer questions about how much of a product was made in the United States from a very different perspective than the interviewer might expect. If asked how much of the meal they made, for example, the consumer might well respond "all, or virtually all." They would mean, however, that they were responsible for "all, or virtually all" of the process of bringing the meal into existence; they would not mean that they raised and slaughtered the livestock, grew the vegetables, or crushed the grapes. The survey evidence offered to support the "all, or virtually all" standard suffers from precisely this problem. Consumers who understand "Made in USA" in the sense of bringing the product into being may naturally respond that it was "all, or virtually all" made in the United States; but this does not necessarily mean that consumers believe that "all, or virtually all" of the product's material content is of U.S. origin. In the Commission's 1991 copy test, for example, consumers were asked what the phrase suggests about "how much" of a product was made in the United States. Similarly, the Danaher Tool Group survey asked consumers "what percentage" of a "Made in USA" tool they assume was made in the United States. In both cases, responses of "all, or virtually all" are completely consistent with an understanding of "made" as confined to the process of bringing the product into being. When asked specifically whether "Made in USA" implied anything about where a product was assembled, almost one-half of respondents (49 percent) said yes.(5) See Chart 2. Presumably, these respondents thought that the process of "making" or "assembling" the final product was important. That expectation would be disappointed if the processing performed by the purported "maker" of the article were only an insignificant portion of the total production process. For example, a consumer who purchases a bicycle that requires some final assembly would hardly claim to have "made" the bicycle after inserting the handlebars, threading the pedals on to the crank, and affixing the saddle to the seat tube. So too, a company that simply collected all of the parts needed to make a complete bicycle and packaged them together, for final assembly by the retailer or the consumer, could not properly claim to have "made anything."(6) b. "Made in USA" Does Not Convey a Parts Claim to More Than an Insignificant Minority of Consumers Again, when asked generally what "Made in USA" means, only 28 Copy Test respondents (14 percent) mentioned meanings that had anything to do with parts. Of these parts mentions, only 14 respondents Ä seven percent of the total sample Ä said that the claim implied that all parts were made in the United States. The other 14 respondents recognized the likely presence of imported parts.(7) In response to closed-ended questions as well, most respondents did not see an implication about parts. When asked specifically whether "Made in USA" implied anything about where the product's parts were made, more than two-thirds of respondents (69.5 percent) said no.(8) Even when the question specifically referred to parts, less the one-third of respondents (27.5 percent) thought that "Made in USA" suggested anything about where the parts were made.(9) See Chart 3. Moreover, the fraction of respondents that saw an implication about parts essentially was the same for all three tested claims, (i.e., "Made in USA" (27.5 percent), "Assembled in USA" (25.5 percent), and "70% Made in USA" (29 percent)).(10) The three claims conveyed different messages about parts to those few respondents who received a parts message, but again, none of the claims conveyed a message about parts to even one-third of respondents. Even respondents who thought a "Made in USA" claim implied something about parts did not generally perceive a message about the quantity of parts. Among the relatively few respondents who saw any implication about parts, 25 (53 percent of those who were asked the question and 12.5 percent of the total sample) said "Made in USA" did not imply anything about how much of the parts were made in the United States. Only 22 respondents Ä 11 percent of the total sample Ä said the claim did have such an implication.(11) See Chart 4. Twenty of these 22 respondents Ä 10 percent of the total sample Ä said that "Made in USA" implied that all of the product's parts were made in the United States.(12) See Chart 5. This percentage is virtually identical to the seven percent of respondents who indicated that all parts were made in the United States in response to the general, open-ended question about the meaning of "Made in USA."(13) These two sets of results from the Copy Test clearly demonstrate that only an insignificant minority of consumers interpret "Made in USA" to imply that a product is made from parts manufactured entirely in the United States. Therefore, under the Commission's traditional deception analysis, qualifications of "Made in USA" claims are not necessary to prevent deception.(14) c. The Attitude Survey Results are Consistent With the Notion that Consumers Do Not Contemplate Parts When Defining "Made in USA" To understand the results of the Attitude Survey, the responses must be analyzed in light of the results obtained in the Copy Test. No Copy Test respondents mentioned anything about cost when asked generally what "Made in USA" meant, and fewer than a quarter (22.5 percent) saw any message about cost in the claim when asked specifically whether the claim implied anything about how much of the total cost of making a product was from the United States.(15) See Chart 6. Nevertheless, the design of the Attitude Survey assumed that cost was an important element of respondents' understanding of the phrase "Made in USA" and posed hypothetical situations based on this assumption. For most respondents, therefore, cost apparently was an element suggested by the interviewer, not the phrase "Made in USA" itself. The strong "order effects"(16) seen in the results of the Attitude Survey support this conclusion. If respondents' own interpretations of "Made in USA" had included an element of cost, the presentation of the Attitude Survey's cost scenarios in increasing or decreasing order would have had relatively little impact on the survey's results. The order effects, however, are quite pronounced. When the interviewer presented the scenario "90% US cost/10% foreign" and assembly in the United States first in the sequence, not quite two-thirds (65.5 percent) of respondents agreed with the "Made in USA" claim. When the interviewer presented the same scenario last in the sequence, almost nine of ten respondents (88.1 percent) agreed.(17) When the interviewer presented the scenario "10% US cost/90% foreign" and assembly in the United States first in the sequence, just over one-third (35.7 percent) agreed with the "Made in USA" claim. When the interviewer presented this scenario last, however, fewer than one in ten respondents (8.6 percent) agreed.(18) In any event, the Attitude Survey results consistently demonstrate that assembly was a crucial variable. Under the scenario "90% US cost/10% foreign" and assembly in the United States, three-quarters of respondents (75 percent) agreed with the "Made in USA" label. Barely a majority (54.6 percent), however, agreed if the same high percentage of U.S. content was present but the assembly occurred abroad. The difference was even more dramatic under the scenario "70% US cost/30% foreign." With foreign assembly, only 38.9 percent of respondents agreed with a "Made in USA" label. The percentage increased to 67 percent when assembly in the United States was specified. The same effect is apparent at the other end of the cost scale. Only 10.2 percent of respondents agreed with a "Made in USA" label under the scenario "10% US cost/90% foreign" and foreign assembly. That percentage doubled to 20 percent if the product was assembled in the United States. 3. OTHER CONSUMER PERCEPTION DATA ARE CONSISTENT WITH THAT CONTAINED IN THE COPY TEST Much of the private consumer perception data presented in connection with the Workshop is remarkably consistent with that contained in the Commission's Copy Test. In particular, the surveys submitted by New Balance and the Danaher Tool Group are consistent with the basic conclusions of the FTC's Copy Test. On the other hand, the contrary conclusions reached in the Commission's Attitude Survey and the National Consumer League survey are attributable to flaws in the design of the questions asked in those surveys. The consumer perception data submitted by New Balance Athletic Shoes confirms the above-discussed results of the Commission's Copy Test. Both New Balance's and the Commission's copy tests were designed to capture the most accurate picture of how consumers interpret the phrase "Made in USA" because they did not bias the results of their general, open-ended question by suggesting or assuming the importance of parts or labor. When New Balance asked its open-ended question about the meaning of "Made in USA" in the context of an athletic shoe labeled with the phrase, 40 percent of respondents simply repeated the phrase or responded with a virtually identical phrase. Applying the definitional logic discussed above, these respondents presumably meant that the shoe came into being in the United States. Only 14 percent provided responses making any mention of materials and only six percent stated that all of the materials were of United States origin. In results similar to those obtained by the Commission and New Balance, the Danaher Tool Group ("DTG") survey found that, when asked generally what "Made in USA" means in the context of a hand tool, 44.5 percent of respondents simply repeated the phrase or responded with a virtually identical phrase. Again, these respondents presumably meant that the tool came into being in the United States. When asked to estimate what percentage of a hand tool carrying a "Made in USA" label was made in the United States, 53 percent of respondents said 100 percent of the tool was made in the United States. As discussed above, however, they likely meant that 100 percent of the process of bringing the tool into being was performed in the United States. When asked specifically whether a "Made in USA" label pertained mostly to the labor or to the materials involved in making hand tools, respondents gave an answer, but may not have been thinking about either materials or labor until the question was asked. Moreover, these results are specific to the expectations of hand tool users meeting certain selection criteria. Unlike other surveys, these tool users reported placing higher importance on the "Made in USA" characteristic as related to tools than to other products.(19) A survey submitted by the National Consumers League ("NCL") purports to indicate that consumers understand "Made in USA" to mean that a high percentage of a product's parts are of United States origin. The NCL reported that 50 percent of the people responding to its informal survey said they assumed that a product advertised as "Made in the USA" contained 100 percent U.S. parts. One of the many problems with this survey, however, is that the very first question asked specifically what percentage of U.S. parts respondents assumed composed a product advertised or labelled as being "Made in the USA."(20) Therefore, like the Commission's Attitude Survey, the NCL survey begged the question by assuming that parts were an important element of respondents' understanding of the phrase "Made in the USA" and posing its question based on this assumption, rather asking a general, open-ended question Ä as did the Commission's Copy Test Ä and allowing respondents to indicate on their own what was important to their understanding of the phrase. It also is interesting to note that NCL mailed its questionnaire to all 4,997 of its members but received only 115 responses. Therefore, the results can hardly be viewed as representative of consumers generally. 4. "MADE IN USA" CLAIMS SHOULD BE ANALYZED UNDER THE "REASONABLE BASIS" STANDARD As discussed above, when asked the meaning of the phrase "Made in USA," the majority of respondents simply parroted the phrase back to the interviewer or responded with a virtually identical phrase. As also discussed, it is logical to conclude that when repeating the word "made," these respondents meant the ordinary dictionary meaning of the word. Thus, these respondents interpreted the phrase "Made in USA" to mean that the product came into being in the United States, and, presumably, that "coming into being" involved significant processing. As such, "Made in USA" claims could be substantiated in a variety of ways, depending on the particular circumstances. The claim could be substantiated by showing compliance with other regulatory definitions (e.g., the NAFTA tariff preference rules),(21) or by evidence that the product was manufactured with only minor foreign components and labor. These methods of substantiating a "Made in USA" claim are consistent with the consumer perception evidence on the meaning of the claim. Absent special circumstances, advertisers should be able to utilize these methods to substantiate "Made in USA" claims. Requiring that "all, or virtually all" component parts be manufactured in the United States as a condition of using a "Made in USA" claim imposes significant and unnecessary costs on advertisers and ultimately on consumers. Under such a regime, qualification of "Made in USA" claims would be necessary in the vast majority of instances. Qualifying "Made in USA" claims is unnecessary, however, for the substantial majority of consumers who do not receive a parts message from such a claim. Therefore, it is more efficient for those few companies whose products do satisfy the "all, or virtually all" standard to qualify their claims to consumers (e.g., "Made in USA of 100% U.S. materials") than to require the vast majority of manufacturers that do not satisfy this standard Ä but that would satisfy the definition attached to "Made in USA" by most consumers Ä to qualify their claims. Moreover, if 100 percent "Made in USA" is important to consumers, manufacturers satisfying the stricter standard will have the incentive to make such claims to distinguish their products. The cost of identifying these products then would be imposed on the small minority of consumers who seek 100 percent "Made in USA" products. This way, consumers who care about where a product's component parts are made would still be able to get what they want, but the majority of consumers would not have to pay for it. This allocation of cost not only lowers total cost, but also provides a market test of whether the "100% Made in USA" attribute is really important to consumers, rather than imposing the Commission's judgment.(22) The "reasonable basis" approach offers important advantages over the "all, or virtually all" standard. First, it preserves flexibility for manufacturers, rather than codifying a particular definition with specific standards and implicit or explicit accounting requirements to determine whether the standard is met. Furthermore, it would free the Commission of the burden of resolving the specific details of a codified definition. Second, the reasonable basis approach would also allow the Commission to consider the unique circumstances of different industries in evaluating whether "Made in USA" claims are adequately substantiated. Finally, it focuses the Commission's attention on the right question Ä namely, whether it is reasonable to identify the advertised product as "Made in USA." The "reasonable basis" approach will ensure that nondeceptive "Made in USA" claims are available to those consumers that truly place importance on such claims. 5. CONCLUSION The Commission should adopt the "reasonable basis" approach when evaluating "Made in USA" claims. Such an approach would allow information that is correctly understood by the vast majority of consumers to reach those consumers and assist them in their purchase decisions. As demonstrated by the consumer perception data, only an insignificant minority of consumers interpret unqualified "Made in USA" claims to mean that "all, or virtually all" of a product was manufactured in the United States. It is not helpful to consumers to burden the majority of consumers with advertising and product labels containing superfluous disclosures in order to reduce the likelihood of confusion for an insignificant minority of consumers. Those relatively few manufacturers that produce products that are "all, or virtually all" of U.S. material content will have strong economic incentives to label and advertise their products in such a way that consumers will be able to identify and buy those products if they choose. Endnotes 1) 61 Fed. Reg. 18,600 (1996). 2) See Comments of the Bicycle Manufacturers Association of America, Inc. Regarding "Made in USA" Claims with Respect to Bicycles, FTC File No. P894219 (Jan. 22, 1996); Comments of the Bicycle Manufacturers Association of America, Inc. Regarding the Commission's Proposed Consent Agreement with Hyde Athletic Industries, Inc. Concerning Hyde's Use of "Made in USA" Claims in Connection with the Sale of Footwear, FTC File No. 922 3236 (Nov. 22, 1994). 3) 16 C.F.R. pt. 260. 4) While this proposed safe harbor, by its terms, simply requires that the U.S. processing "result in a ratio of the cost of U.S. processing to the cost of goods sold that is not insignificant," BMA would prefer that this criterion be expressed in the affirmative (i.e., that the ratio of the cost of U.S. processing to the cost of goods sold be "significant." As used here, the term "significant" would not necessarily mean a majority, but would imply a percentage approaching 50 percent. 5) On the other hand, BMA believes that "Product of USA" and "Manufactured in USA" are reasonably interpreted as synonymous with "Made in USA," particularly in light of the longstanding construction of these terms by the U.S. Customs Service in the foreign country-of-origin marking context. APPENDIX Endnotes 1) Dr. Beales is Associate Professor of Strategic Management and Public Policy at George Washington University, a post he has held since 1988. Between 1983 and 1987, he was Associate Director for Policy and Evaluation in the FTC's Bureau of Consumer Protection. Between 1981 and 1986, he also served as Assistant to the Director and Acting Deputy Director of the Bureau. 2) The "reasonable basis" approach affords the Commission flexibility in determining whether a claim such as "Made in USA" is substantiated depending on the context in which the claim is made. Under this approach, the Commission takes into account a number of factors Ä namely, the type of product, the type of claim, the benefits of a truthful claim, the consequences of a false claim, the cost of developing substantiation for the claim, and the level of substantiation experts in the field would consider reasonable Ä in determining whether a claim is substantiated. 48 Fed. Reg. 10,471 (March 11, 1983), appended to Thompson Medical Co., 104 F.T.C. 648, 839 (1984), aff'd, 791 F.2d 189, (D.C. Cir. 1986), cert. denied, 479 U.S. 1086 (1987). For a discussion of the advantages of applying the "reasonable basis" approach to "Made in USA" claims, see infra at 13-14. 3) Table 3: Summary of Copy Test Results, at 4. 4) See The American Heritage Dictionary (Second College Edition) 758 (1991). 5) Table 3: Summary of Copy Test Results, at 8. 6) Cf. id. at 6 (only two respondents (1 percent) thought that "Made in USA" meant that the product was packaged in the United States). 7) A breakdown of the parts mentions is as follows: "all made in US" 10 respondents "most/more than 50% made in US" 2 respondents "made in US with some imported parts" 4 respondents "assembled in US with all US parts" 1 respondent "assembled in US with some imported parts" 7 respondents "all parts made in US" 3 respondents "some parts imported" 1 respondent TOTAL PARTS MENTIONS 28 respondents Id. at 4-7. The mentions indicating that all parts were made in the United States are highlighted. 8) Id. at 10. 9) Id 10) Id. 11) Id. at 11. Three respondents said something else. 12) Id. at 11. The summary tabulation, which reported that 21 respondents said all parts were made in the United States, appears to be in error on this point. The detailed tabulations reported only 20 respondents who stated that all parts were made in the United States in response to the last question in the following series of questions: 1. "When the ad/label says this (point to MUSA statement but do not read it) does that suggest or imply anything about where the parts that went into the product were manufactured?" 2. "What does that suggest or imply about where the parts that went into the product were manufactured? Anything else?" 3. "When the ad says this (point to MUSA statement, but do not read it) does that suggest or imply anything about how much of the parts that went into the product were manufactured in the United States?" 4. "What does that suggest or imply about how much of the parts that went into the product were manufactured in the United States? Anything else?" 13) See supra note 7. 14) Under Commission law, an advertisement is deceptive if it contains a material representation or omission that is likely to mislead consumers acting reasonably under the circumstances. See Letter from James C. Miller, III, Chairman, Federal Trade Commission, to Chairman, House Energy and Commerce Committee (Oct. 14, 1983), appended to Cliffdale Assocs., Inc., 103 F.T.C. 110, 174 (1984). Therefore, the Commission considers only the reasonable interpretations of a claim when determining whether the claim is deceptive. An interpretation shared by a significant minority of consumers is, by FTC definition, reasonable. Conversely, an interpretation shared by only an insignificant minority of consumers is not reasonable and not deceptive. Id. at 177 n.20; see also Kraft, Inc., 114 F.T.C. 40, 122 (1991), aff'd 970 F.2d 311 (7th Cir. 1992), cert. denied, 507 U.S. 909 (1993); 61 Fed. Reg. 18,600, 18,603 n.1 (April 26, 1996). 15) Table 3: Summary of Copy Test Results, at 13. Of the 45 respondents who saw a message about cost in the "Made in USA" claim, 26 (only 13% of the total sample) said the claim implied that 100% of the cost of making the product was from the United States. See Chart 7. 16) "Order effects" occur when answers respondents give depend on the order in which the questions are asked, or the order in which possible choices of answers are presented. 17) Attitude Survey Detailed Tables, Question 2a (based on combining responses for stereo and pen). 18) Attitude Survey Detailed Tables, Question 2e (based on combining responses for stereo and pen). Interviewers used descending order (i.e., "90% US costs/10% foreign" first) in 58 of the 100 interviews. 19) DTG excluded from the survey those persons who did not use hand tools on at least a casual, do-it-yourself basis. Therefore, the results of DTG's survey are not representative of the general public's opinion of "Made in USA" claims on other products. Danaher Tool Group, "Made in USA" Hand Tool Opinion Survey at 4 (Feb. 1996). 20) The NCL question asked: When you see a product advertisement or label stating "Made in the USA": a. What amount of U.S. parts (i.e., components) do you assume is in the product? State a minimum percentage. ( ) percent ( ) no opinion b. What amount of U.S. labor (i.e., production) do you assume is in the product? State a minimum percentage. ( ) percent ( ) no opinion 21) If a definition is appropriate or required for labeling purposes, it should not be regarded as deceptive to employ the same definition in advertising. In such circumstances, advertising simply reduces the cost of finding information that already is available (i.e., on the product label). On the other hand, inconsistent definitions for labeling and advertising are likely to create confusion among consumers. The FTC need look no further than its own Food Policy Statement for an example of the appropriateness of harmonizing advertising definitions to label definitions. This remains true even if a significant minority of consumers understands advertising terms differently from the common understanding. For example, the fact that some consumers may think "high in fiber" means more fiber than the FDA definition should not bar use of the claim in advertising when it meets the definition. If additional features or characteristics are important to consumers, sellers have every incentive to provide more explicit information about those characteristics. 22) The survey evidence indicates that the attribute of being "Made in USA" is important to many consumers, though not necessarily so important that consumers would pay much of a premium for this attribute. This raises questions about the materiality of a "Made in USA" claim. Methodologically, the ratings of importance that did not ask consumers to make any tradeoffs are highly suspect. Surveys that incorporated tradeoffs Ä such as those conducted by New Balance, the Danaher Tool Group, and the International Mass Retail Association Ä found lower importance ratings. When asked in the abstract whether or not a particular characteristic is important, many consumers are likely to agree that it is, even if it is the least important attribute of the product to them. This is especially the case if there is a "politically correct" answer. The relatively low notice of the "Made in USA" claim in the Copy Test Ä only 11.5% unaided Ä is not consistent with the generally high ratings of importance seen in other studies. Consumers tend to notice and remember messages that are important to them. Furthermore, there is no reason to presume that parts are a material component of a "Made in USA" claim, even for those consumers who receive a parts message. The data on importance do not indicate what facet of the meaning of the phrase is important. Among those respondents who thought "Made in USA" meant assembly, the mean importance rating (calculated from individual copy test responses) for those respondents who said the claim also meant something about parts (6.5) was not significantly different from the mean importance rating for those respondents who said the claim did not mean anything about parts (5.3). Therefore, even assuming the materiality of a "made in USA" claim generally, the "importance" data do not indicate that the use of domestically manufactured parts in a product so labled is material to consumers. 15