Trans Union Corporation

Text (OCR of the scan at left; may contain errors)

IN THE MATTER OF TRANS UNION CORPORATION OPINION OF THE COMMISSION, FINAL ORDER, AND INITIAL DECISION IN REGARD TO ALLEGED VIOLATIONS OF SECTIONS 604 AND 607 OF THE FAIR CREDIT REPORTING ACT Docket No. 9255; File No. 922 3036 Complaint, December 15, 1992 BOpinion and Final Order, March 1, 2000 In a unanimous Opinion, the Commission concluded that Trans Union Corporation (ATrans Union@), one of the three national credit bureaus, violated the Fair Credit Reporting Act (FCRA) by selling its consumer reports to target marketers who in turn solicit the consumers to purchase goods and services. As a consumer reporting agency, Trans Union received detailed credit information for over 160 million consumers from numerous credit grantors, including banks, mortgage companies, credit unions, auto dealers and others, and then compiled this information into consumer reports and sold these reports to target marketers. The Commission held that Trans Union=s disclosure of this information to entities that lacked a statutorily-defined permissible purpose for obtaining them, violated the FCRA, which protects the privacy of credit information by limiting the circumstances under which a consumer reporting agency can disclose a consumer report. In the Final Order, Trans Union is prohibited from selling consumer reports as target marketing lists to marketers lacking an authorized purpose for receiving them under the FCRA. The Final Order applies to a number of Trans Union=s target marketing list products, including its Master File/Selects products, its modeled products and its TransLink/reverse append products.

Participants For the Commission: Kellie A. Cosgrove, Annemarie Scanlon Harthun, Christopher W. Keller, Lucy Morris, and Jonathan A. Smollen.

For the Respondents: Stephen L. Agin, Elizabeth R. Bacon Ehlers, Roger L. Longtin, and Colleen E. McManus, Rudnick & Wolfe; Steven B. Feirman, Rudnick, Wolfe, Epstein & Zeidman. VOLUME 129 Complaint COMPLAINT The Federal Trade Commission, having reason to believe that Trans Union Corporation, a corporation, hereinafter sometimes referred to as respondent, has violated the Fair Credit Reporting Act, 15 U.S.C. ' 1681 et seq., and it appearing to the Commission that a proceeding by it in respect thereof would be in the public interest, hereby issues this complaint, and alleges as follows: DEFINITIONS For the purposes of this complaint, the terms, "person," "consumer," "consumer report," and "consumer reporting agency" are defined as set forth in '' 603(b), (c), (d), and (f), respectively, of the Fair Credit Reporting Act, 15 U.S.C. '' 168la(b), 1681a(c), 1681a(d) and 1681a(f).

"Credit information" means the information that respondent maintains bearing on any of the characteristics listed in ' 603(d) of the Fair Credit Reporting Act, 15 U.S.C. ' 1681a, as amended, with respect to any consumer that respondent: obtains from subscribers, court records or any other source and from which respondent creates consumer reports.

"Permissible purpose" means any of the purposes listed in Section 604 of the Fair Credit Reporting Act, 15 U.S.C. ' 1681b, as amended, for which a consumer reporting agency may lawfully furnish a consumer report.

"Prescreening" means the process whereby respondent, utilizing credit information, compiles or edits for a client a list of consumers who meet specific criteria and provides this list to the client or a third party (such as a mailing service) on behalf of the client for use in soliciting those consumers for an offer of credit. ASubscriber@ means any person who furnishes credit information to respondent or who requests or obtains a consumer report from respondent, excluding consumers. TRANS UNION CORPORATION 473 Complaint PARAGRAPH ONE: Respondent, Trans Union Corporation is a corporation organized, existing and doing business under and by virtue of the laws of the state of Delaware, with its office and principal place of business located at 555 West Adams Street, Chicago, Illinois, 60661.

PARAGRAPH TWO: Respondent is, and has been, regularly engaged in the practice of procuring and assembling information on consumers for the purpose of furnishing for monetary fees, Consumer reports to subscribers and consumers. Respondent furnishes these consumer reports through the means and facilities of interstate commerce. Hence, respondent is a consumer reporting agency, as defined in Section 603(f) of the Fair Credit Reporting Act.

PARAGRAPH THREE: Respondent regularly provides consumer reports in the form of prescreened lists to credit grantors and fails to require or monitor that credit grantors that receive such lists make a firm offer of credit to each person on the list.

PARAGRAPH FOUR: By and through the acts and practices alleged in Paragraphs Two and Three, and others not specifically set forth herein, Respondent has violated Sections 604 and 607 of the Fair Credit Reporting Act by furnishing consumer reports to persons whom Respondent did not have reason to believe intended to use the information for a Permissible Purpose under Section 604.

PARAGRAPH FIVE: Respondent regularly compiles, for sale to clients, lists of consumers, based in whole or in part on information contained in its consumer reporting database bearing on the characteristics enumerated in Section 603, thereby creating consumer reports, and provides such consumer reports in the form of target marketing lists to persons that do not intend to make a VOLUME 129 Complaint firm offer of credit to all those consumers on the list and who intend to use the information for purposes not authorized under Section 604 of the Fair Credit Reporting Act. PARAGRAPH SIX: By and through the acts and practices alleged in Paragraphs Two and Five, and others not specifically set forth herein, Respondent has violated Sections 604 and 607 of the Fair Credit Reporting Act by furnishing consumer reports to persons whom Respondent did not have reason to believe intended to use the information for a Permissible Purpose under Section 604.

NOTICE Notice is hereby given to the respondent herein before named that the 16th day of March, 1993, at 10:00 o=clock is hereby fixed as the time and Federal Trade Commission Offices, Rm: 532, 6th and Pennsylvania Avenues, N.W., Washington, D.C. 20580, as the place when and where a hearing will be had before an Administrative Law Judge of the Federal Trade Commission, on the charges set forth in this complaint, at which time and place you will have the right under said Act to appear and show cause why an order should not be entered requiring you to cease and desist from the violations of the law charged in this complaint. You are notified that the opportunity is afforded you to file with the Commission an answer to this complaint on or before the thirtieth (30) day after service of it upon you. An answer in which the allegations of the complaint are contested shall contain a concise statement of the facts constituting each ground of defense; and specific admission, denial, or, explanation of each fact alleged in the complaint or, if you are without knowledge thereof, a statement to that effect. Allegations of the complaint not thus answered shall be deemed to have been admitted. If you elect not to contest the allegations of fact set forth in the complaint, the answer shall consist of a statement that you admit all of the material allegations to be true. Such an answer shall TRANS UNION CORPORATION 475 Complaint constitute a waiver of hearings as to the facts alleged in the complaint, and together with the complaint will provide a record basis on which the Administrative Law Judge shall file an initial decision containing appropriate findings and conclusions ,and an appropriate order disposing of the proceeding. In such answer you may, however, reserve the right to submit proposed findings and conclusions and the right to appeal the initial decision to the Commission under Section 3.52 of the Commission's Rules of Practice for Adjudicative Proceedings.

Failure to answer within the time above provided shall be deemed to constitute a waiver of your right to appear and contest the allegations of the complaint and shall authorize the Administrative Law Judge, without further notice to you, to find the facts to be as alleged in the complaint and to enter an initial decision containing such findings, appropriate conclusions and order.

NOTICE OF CONTEMPLATED RELIEF Should the Commission conclude from the record developed in any adjudicative proceeding in this matter that the Respondent, Trans Union Inc., is in violation of Sections 604 and 607 of the Fair Credit Reporting Act, as alleged in the complaint, the Commission may order such relief as is supported by the record and is necessary and appropriate including, but not limited to, ordering that Respondent:

1. Cease and desist from providing consumer reports in the form of prescreened lists to credit grantors and failing to require and monitor to ensure that credit grantors who receive such lists make a firm offer of credit to each person on the list; 2. Cease and desist from compiling and/or selling consumer reports in the form of target marketing lists to any person unless Respondent has reason to believe that such person either intends VOLUME 129 Complaint to make a firm offer of credit to all consumers on such lists or to use such lists for purposes authorized under Section 604 of the Fair Credit Reporting Act.

3. Maintain for at least five (5) years from the date of service of this order and upon request, make available to the Federal Trade Commission for inspection and copying, all records and documents necessary to demonstrate fully its compliance with this Order.

4. Deliver a copy of this Order to all present and future management officials having administrative, sales, advertising, or policy responsibilities with respect to the subject matter of this Order.

5. For the five (5) year period following the entry of this Order, notify the Commission at least thirty (30) days prior to any proposed change in Respondent such as dissolution, assignment, or sale resulting in the emergence of a successor corporation, the creation or dissolution of subsidiaries, or any other change in the corporation that might affect compliance obligations arising out of this Order.

6. Within one hundred and eighty (180) days of service of this order, deliver to the Commission a report, in writing, setting forth the manner and form in which it has complied with this Order as of that date.

IN WITNESS WHEREOF, the Federal Trade Commission has caused this complaint to be signed by its Secretary and its official seal to be hereto affixed at Washington, D.C. this fifteenth day of December, A.D. 1992.

By the Commission.

TRANS UNION CORPORATION 477 Opinion of the Commission OPINION OF THE COMMISSION By Thompson, Commissioner:

I. INTRODUCTION In this information age, technological advances in information gathering and dissemination have generated substantial benefits for American consumers by providing them with, among other things, the strongest and most efficient credit markets in the world. In 1970, Congress recognized the importance of personal financial data to these markets when it enacted the Fair Credit Reporting Act (AFCRA@ or AAct@). Congress expressly noted in the Act=s findings and statement of purpose that the Abanking system is dependent upon fair and accurate credit reporting@ and acknowledged the Avital role@ of credit bureaus (called Aconsumer reporting agencies@ under the Act) Ain assembling and evaluating consumer credit and other information on consumers.@ 15 U.S.C. ' 1681(a)(1) and (3).

Under the U.S. credit reporting system, consumer reporting agencies (hereinafter ACRAs@) collect consumer credit information from credit grantors and other sources, compile the information into credit reports, and then sell the reports to banks and other lenders, as well as to employers and insurance companies. Credit grantors have an incentive to provide data to CRAs because they benefit from the credit reporting system as well. The effectiveness of this system depends upon a constant flow of consumers= credit information into large databases maintained by CRAs. It also depends on accuracy and timeliness. As a result, CRAs, unlike other data providers, have access to a broad range of continually-updated, detailed information about millions of consumers= personal credit histories. This information includes, for example, consumers= delinquencies and defaults, the types of credit accounts they have, when they obtained credit, and VOLUME 129 Opinion of the Commission additional information that banks and other lenders often use in determining whether to extend credit.

Although Congress understood the importance of CRAs= access to such information regarding millions of consumers, it also recognized the importance of protecting consumers= financial privacy. In fact, legislative history reveals that one of the FCRA=s principal goals was to protect the privacy of individuals whose sensitive credit and financial data are collected, used, reviewed and transmitted by CRAs1. Thus, in enacting the FCRA, Congress struck a balance between these competing interests. While Congress did not disturb the ability of CRAs to collect personal credit information, it did provide safeguards designed to protect the confidentiality of these data. Specifically, Section 604 of the FCRA limits the circumstances under which a CRA may disclose a Aconsumer report@2 - - the statutory term for information commonly referred to as a credit report. For instance, Section 604 allows a CRA to furnish consumer reports to, inter alia, persons with certain Apermissible purposes.@ These permissible purposes include: (1) the extension of credit; (2) employment purposes; (3) underwriting of insurance; (4) determination of license eligibility; (5) risk assessment for an existing credit obligation; and (6) legitimate business need for the information. 15 U.S.C. ' 1681b. Section 607 of the Act also requires CRAs to maintain reasonable procedures to ensure that they only furnish consumer reports for the purposes set forth in Section 604. See 15 U.S.C. ' 1681e(a). 1 Report of the Committee on Banking and Currency, S. Rep. No. 91- 517 (1969).

2 Section 603(d) of the FCRA defines Aconsumer report@ as: A[a]ny written, oral, or other communication of any information by a consumer reporting agency bearing on a consumer=s credit worthiness, credit standing, credit capacity, character, general reputation, personal characteristics, or mode of living which is used or expected to be used or collected in whole or in part for the purpose of serving as a factor in establishing a consumer=s eligibility for . . . credit or insurance . . . [or] employment . . .@ 15 U.S.C. ' 1681a(d). TRANS UNION CORPORATION 479 Opinion of the Commission After careful consideration of the parties= arguments and thorough review of the substantial record in this case, the Commission concludes that Trans Union Corporation (ATrans Union@), a CRA, violates or has violated Sections 604 and 607 of the FCRA through the activities of its target marketing business3. In connection with its consumer reporting business, Trans Union receives various types of personal, credit information about consumers. Much of this information constitutes a Aconsumer report@ as that term is defined by Section 603(d). Trans Union=s sale of consumer reports to target marketers without a Apermissible purpose@ under the FCRA is a violation of the Act. II. PROCEDURAL HISTORY On December 15, 1992, the Commission filed an administrative complaint alleging, in pertinent part, that Trans Union violated Sections 604 and 607(a) of the FCRA by compil[ing], for sale to clients, lists of consumers, based in whole or in part on information contained in its consumer reporting database bearing on the characteristics enumerated in Section 603, thereby creating consumer reports, and provid[ing] such consumer reports in the form of target marketing lists to persons that do not intend to make a firm offer of credit to all those consumers on the list and who intend to use the information for purposes not authorized under [the FCRA].

In re Trans Union Corporation, 116 F.T.C. 1334, 1336 (1993). 3 As described infra, Trans Union may have discontinued some of the practices at issue in this matter. To the extent it continues to engage in certain other of the activities at issue, however, Trans Union=s FCRA violations are ongoing.

VOLUME 129 Opinion of the Commission On September 20, 1993, Administrative Law Judge (AALJ@) Parker entered a summary decision in favor of Complaint Counsel. The Commission upheld that decision, ruling specifically that Trans Union=s target marketing lists were Aconsumer reports@ because the minimum criteria for a consumer file appearing on any of the target marketing lists - - that the consumer had at least two open credit accounts - - satisfied the definition of Aconsumer report@ under Section 603(d) of the Act. In re Trans Union Corporation, 118 F.T.C. 821, 869-70 (1994). A key part of the Commission=s determination was its finding that the mere existence of two credit accounts, or Atradelines,@4 constituted information Acollected in whole or in part by [Trans Union] with the expectation that it would be used by credit grantors for the purpose of serving as a factor in establishing the consumer=s eligibility [for credit].@ Id. at 861. The Commission also held that target marketing is not a permissible purpose under the FCRA. Therefore, according to the Commission, Trans Union violated the FCRA by disclosing consumer reports to persons lacking any of the required permissible purposes. In ruling on Trans Union=s appeal of the Commission=s decision, the United States Court of Appeals for the District of Columbia Circuit agreed that target marketing was not a 4 A Atradeline@ is a segment of a consumer report that reflects a credit relationship between a consumer and a creditor - - usually a debt or a potential debt owed by the consumer to the credit grantor. An example of such an account relationship is a consumer=s Visa, American Express or other credit card account. A typical consumer report contains multiple tradelines, and each reveals specific information about the account relationship, including: the account holder=s account number, name, address, telephone number, date of birth, social security number, any generational suffix; the name and subscriber code of the credit grantor and its kind of business; the open date of the account; the verified date on the account; the type of loan; the credit limit assigned by the credit grantor; the payment patterns and history; the present status of the account; and the closed date of the account. Public record information such as bankruptcies, tax liens, foreclosures and civil judgments as well as collection accounts are also considered tradelines. See Stockdale 872, 875/23--876/2, 888/5-24, 893/6-15, 894/4-12, 895/16--896/1, 896/19-23, 897/13--898/2; Botruff 2049/1-6; Weith 1844/18-22; Smith 3372/15--3373/15. TRANS UNION CORPORATION 481 Opinion of the Commission permissible purpose under the Act. Trans Union Corp. v. F.T.C., 81 F.3d 228 (D.C. Cir. 1996). The Court also held, however, that it was inappropriate for the Commission to use summary procedures to decide whether Trans Union=s target marketing lists were consumer reports because the question presented a genuine issue of material fact. Consequently, the Court remanded the case to the Commission to resolve two primary questions. The first is factual - - whether there is sufficient evidence to support the finding that Trans Union=s target marketing lists are consumer reports. The second question is a legal one - - if we find that Trans Union=s target marketing lists are consumer reports, does the FCRA pass constitutional muster? On July 31, 1998, Administrative Law Judge James Timony issued an Initial Decision and Order on remand holding that Complaint Counsel provided sufficient evidence to show that Trans Union=s lists are Aconsumer reports@ under the Act and that Trans Union disclosed them to entities who lacked a permissible purpose. This disclosure violated Sections 604 and 607(a) of the FCRA. Judge Timony also held that the FCRA, as applied to Trans Union=s practices, is constitutional. Trans Union appealed both rulings5.

5 References to the record are abbreviated as follows, using the following hypothetical examples:

Initial Decision ID at 200. Initial Decision Finding IDF-500. Complaint Counsel Exhibit CX-500. Trans Union Exhibit TU-500. Trial Transcript testimony Jones 1234/56-78. Deposition Transcript testimony Jones CX-100 at 123/45-46. Trans Union=s Appellant Brief TUAB at 200. Complaint Counsel=s Answering Brief CCAB at 200. Trans Union=s Reply Brief TURB at 200. Complaint Counsel=s Proposed Findings CCPF at 200. Trans Union=s Proposed Findings TUPF at 200. VOLUME 129 Opinion of the Commission After reviewing the full record in this case, including the extensive arguments of the parties, we adopt the ALJ=s July 1998 findings and conclusions to the extent that they are consistent with those set forth in this opinion.

III. STANDARD OF REVIEW The Commission reviews the decision of the ALJ under a de novo standard. FTC Rules of Practice, Rule 3.54(a). The Commission can, however, give some deference to the ALJ=s credibility determinations because, as the trier of fact, the ALJ had the opportunity to Aclosely scrutinize witnesses= overall demeanor and to judge their credibility.@ In the Matter of Horizon Corp., 97 F.T.C. 464, 857 n.77 (1981).

IV. FACTUAL BACKGROUND A. Trans Union=s Business Trans Union is a Delaware corporation whose principal place of business is located at 555 West Adams Street, Chicago, IL 60661. Trans Union=s primary business is credit reporting and it is a CRA under Section 603(f) of the Act. (Rodgers CX-191 at 27/3-7). As a CRA, Trans Union collects credit information about millions of American consumers from numerous credit grantors and others, compiles this information into credit reports and sells the reports to credit grantors nationwide. (Connelly 2588/19B2590/18; Pendleton 404/12--405/9; Johnson 1206/16-- 1209/7). Trans Union=s main competitors in the credit reporting business are Experian (formerly TRW) and Equifax. (Rodgers CX-191 at 47/10-12). These companies are also CRAs. The millions of pieces of consumer information Trans Union receives every month are maintained in an extensive database called CRONUS. (Weith 1867/19--1870/9; Botruff CX 181 at TRANS UNION CORPORATION 483 Opinion of the Commission 19/14-22)6. The information in CRONUS comes from credit grantors - - including banks, mortgage companies, credit unions and auto dealers - - collection agencies, public records and others. (Stockdale 873/22-25). The information is very current as Trans Union receives new data every day and updates CRONUS weekly. (Botruff CX 181 at 30/18--31/8). Information compiled on a specific consumer within CRONUS is called a consumer file. In addition to its credit reporting business, Trans Union also sells a variety of target marketing products through its subsidiary, Performance Data (formerly Trans Mark and Trans Union Lists). Performance Data creates lists of the names and addresses of specific classes of consumers and sells them to target marketers who in turn solicit the consumers to purchase goods and services. Performance Data employs 46 people, including 10 salespersons. (Davis 37/25--38/4). At the beginning of 1998, Performance Data had 440 customers; during 1997, it generated over $34 million in sales. (Davis 48/8-10, 141/13-14). Performance Data=s sales comprise 2% of the target marketing industry. (Davis 3322/15- 18). Hereinafter, unless otherwise noted, our references to Trans Union=s target marketing business include Performance Data=s activities.

As a CRA, Trans Union is in a special position. Trans Union has access to a vast array of very current and detailed consumer information from its credit reporting business which affords it a distinct advantage as a target marketer. Trans Union takes consumer information from CRONUS to create two primary databases called the Master File and the Standard Characteristics database. (Cabigon 1365/13-18; Kinsinger 2017/19-23; Weith CX-196 at 179/11-13). Trans Union offers different target marketing products based upon the information gathered in these two databases as well as data taken directly from CRONUS. See 6 Each month CRONUS takes in 85,000 updates from credit grantors and data providers and 1.8 billion tradelines. (Stockdale 874/4-10, 908/1-19). VOLUME 129 Opinion of the Commission chart detailing Trans Union=s various target marketing products, appended hereto as Figure 1. For the reasons explained in detail infra p. 13, the fact that Trans Union uses CRONUS information in its target marketing business is significant because CRONUS information is far richer and more detailed than the data collected and used by non-CRA competitors who sell target marketing lists. Trans Union is also the only CRA that sells to target marketers an array of personal credit information obtained from its credit reporting database.

1. The Master File The CRONUS-derived Master File is one of the databases Trans Union uses for target marketing. (CX-72-C). It contains information on 160 million people and 105-110 million households. (Weith 1859/8-18; CX-333). Trans Union updates the Master File three times per year. (Cabigon 1366/10-12; Davis 62/16--63/7).

In order for Trans Union to include a CRONUS consumer file in the Master File, thereby making the consumer=s name and address available for target marketing purposes, the consumer file must satisfy several minimum criteria. These criteria have changed over time. Prior to January 1998, each CRONUS consumer file had to show at least two open tradelines with one of the tradelines verified - - i.e., that some reported activity took place - - during the preceding 12 months. (Cabigon 1372/18-- 1373/7; CX-329-A; Weith CX-196 at 197/24--198/14). In addition, a qualifying tradeline could not be closed or an account about which there was a consumer dispute, and could not be a collection record or public record. These criteria are hereinafter referred to as the Apre-1998 Minimum Criteria.@ (Weith CX-196 at 191/7-15, 227/1-5; Cabigon 1374/5-22). In January 1998, in order to be included in the Master File, Trans Union began to require CRONUS consumer files to contain two tradelines active within the last six months or one tradeline active in the last six months with an address confirmed by an outside source. We refer to these later criteria as the Apost-1997 TRANS UNION CORPORATION 485 Opinion of the Commission Minimum Criteria@ and both sets jointly as the AMinimum Criteria@. (Weith 1830/23--1831/4; Cabigon 1386/14--1388/7; CX-332-A; CX-339-A). As with the pre-1998 Minimum Criteria, the qualifying tradeline could not be a collection record or a public record. (Cabigon 1374/12-21; CX-332-A; CX-340-A). Trans Union claims that the two tradeline, pre-1998 Minimum Criteria did not reveal consumer credit information and that the two tradeline minimum was only important because it confirmed, by two sources, the subject=s current name and address. TUAB at 11. Statements made by Trans Union during the relevant time and in its regular course of business, however, belie this simple characterization. For instance, Trans Union=s promotions boasted that the Master File is a list of A135 million financially active individuals@ (emphasis added), that A[a]ny adult with at least two active tradelines is represented,@ and that a person with no activity in a 12 month period - - i.e., making payments or establishing credit - - is dropped from the Master File. (CX-70-A; CX-69-A; CX-58-C). We agree with Trans Union=s written characterizations and find that the Atwo-tradeline minimum@ criterion indicates more than just a confirmed address. It instead reveals a significant fact about consumers in the Master File, i.e., that they are current, at least somewhat active users of credit. 2. Trans Union=s AMaster File / Selects@ Product While the Master File contains names, addresses and other demographic information on people who meet the Minimum Criteria discussed above, it also is frequently enhanced with the addition of other personal, often credit-related, information on each individual. This enhancement enables Trans Union to offer its target marketing customers the opportunity to select, from the 160 million consumer files in the Master File, names and addresses of a smaller set of consumers who meet certain criteria specified by the target marketing customer. The criteria Trans Union uses to create these subsets are called Aindicators@ or VOLUME 129 Opinion of the Commission Aselects,@ and Trans Union generates half of them from its consumer reporting database CRONUS. (Cabigon 1438/12-25). Trans Union=s target marketing customers use the Master File / Selects product in two ways. Some customers provide a list of consumers to Trans Union and purchase Master File select information regarding those customers. (Davis 33/22-25). Other customers request that Trans Union extract from the Master File names and addresses of those consumers who satisfy criteria selected by the customer. (Davis 34/1-5). In other words, Trans Union=s target marketing customers can choose from a menu of selects and ask for a tailored list of consumers= names and addresses who, for example, have a bank card, an open mortgage, but never have obtained short term (30/60/90 day) financing. Trans Union sells these lists for one-time use by its customers either by rental or by license and charges a Abase price@ per thousand names, with additional charges per thousand based on the selects that the customer has chosen. (Davis 44/6-24, 64/6-22, 65/3-14).

Prior to October 1997, when it made certain changes in its business practices (see infra pp. 10-11), Trans Union permitted its target marketing customers to order from the Master File lists of the names and addresses of consumers who had the following types of credit accounts:  Automobile - - indicating whether the consumer has an auto loan or lease not more than five years old; a second auto loan or lease not more than five years old; and for the most recent first and second loan or lease, the open date, expiration, and loan type, and range indicating high credit value (i.e., highest amount ever owed);

 Bank Card - - indicating whether the consumer has an open bank card, including the open date of the most recent bank card account;

 Premium Bank Card - - indicating whether the consumer has an open premium bank card, defined as a bank card with a credit limit that exceeds $9,999, and the open date of the most recent premium bank card account;

TRANS UNION CORPORATION 487 Opinion of the Commission  Department Store Card - - indicating whether the consumer has an open department store card account, including the open date of the most recent department store card account;  Finance Tradeline - - indicating whether the consumer has an open account with a finance company, the open date of the most recent account with a finance company, and whether the account type is a mortgage or auto finance loan;7  A30/60/90 day@ Finance Tradeline - - indicating whether the consumer has a open account with a finance company with a 30, 60, or 90 day loan term;

 Mail Order - - indicating whether the consumer has an open account with any of a number of mail order companies;  Mortgage - - indicating whether the consumer has a first mortgage and/or second mortgage; for the most recent first and second mortgage, the open date, closed date, loan type (refinance, secured mortgage, secured home improvement loan); and range indicating high credit value;  Student Loan - - indicating whether the consumer has a student loan, the type of loan, the open date of the most recent student loan, whether it is closed, and the high credit amount (range); and  Upscale Retail Card - - indicating whether the consumer has an upscale retail card, based upon the National Retail Federation=s listing of Aprestigious@ stores, and the open date of the most recent upscale retail card.

7 In the lending industry, having a finance loan indicates that the consumer has approached a lender of Alast resort@ and is more likely to need credit. (Rapaport 792/17--793/21). Trans Union expressly advised its mortgage lender / customers to use the homeowner and finance tradeline selects because the finance tradeline select provides names of consumers who have Agenerally had trouble with their credit in the past and are highly responsive to credit offers.@ (CX-33; CX-68-A).

VOLUME 129 Opinion of the Commission Trans Union also offered its target marketing customers the option to purchase other types of Ainferential@ selects, including:  Head of Household - - identifying the person in the household with greatest number of tradelines;

 Length of Residence - - identifying people who have maintained their residence for more than a certain period of time based on duration that credit grantors report on person at that residence or based on mortgage open dates;  Singles - - identifying people without joint credit accounts; and  Drivers - - identifying individuals with either an auto loan or a tradeline with a business that issues gas cards and thus presumably own or lease a car.8 The record contains ample evidence of how Trans Union=s customers used the Master File / Selects product. For example, Mercantile Mortgage Co. obtained information from Trans Union to advance its telemarketing promotion which offered homeowners who had been denied credit elsewhere the opportunity to reduce their monthly mortgage rates by refinancing their mortgage, thereby freeing up funds for Ahome improvements,@ a Anew car,@ or a Adream vacation.@ (CX-18). Mercantile purchased from Trans Union a list of consumers in Mercantile=s area of business (Ohio), with telephone numbers (necessary for telemarketing promotion), who also had single or multiple mortgages (an important minimum eligibility factor) and credit with a finance company. Id.

Ramsay Mortgage purchased a target marketing list from Trans Union for its mail offer to lower consumer debt payments, clean up credit, consolidate debt, and/or refinance a mortgage. Ramsay obtained for the Spotsylvania, Virginia area a list of consumers with a mortgage, a bank card, and a retail card. (CX- 25). Another lender, the Mortgage Banc, purchased from Trans Union, for certain counties, lists of homeowners, with phone 8 IDF-37; TUPF at 186; CX-1; Cabigon 1378/12-19. TRANS UNION CORPORATION 489 Opinion of the Commission numbers, who used finance companies, who had been at their residence for 6-15 years, and who also had a bank card. (CX-23- C). In another example, Trans Union sold a target marketing list to Rubinstein Bros., a retail clothing store offering a no-fee charge account to promote its new ALadies Department.@ Rubinstein Bros. purchased a list of females from certain geographical areas, who were between age 25 and 75 and had upscale retail cards and phone numbers. (CX-35).9 3. Trans Union=s Standard Characteristics / Model Products As previously described, Trans Union also maintains a second database, called the AStandard Characteristics@ or AAttribute@ file. This file contains 313 attributes on each CRONUS consumer who meets the Master File Minimum Criteria. (Cabigon 1373/23-- 1374/4; CX-329-A). Trans Union used this personal credit information to create certain proprietary models that it offered to target marketers until October 1997. These proprietary models assign a value, or Ascore,@ to each consumer file in the following ways through the following products:

 E-Val. A scoring system that, using information in the Standard Characteristics file, estimates the amount of equity available in a consumer=s home. A Trans Union customer can purchase a consumer=s E-Val Ascore@ showing: (1) the estimated actual amount of equity in the consumer=s home; (2) the percentage of equity over home value; and (3) the home value range. (CX-1-I-J; CX-118-B; Davis 134/12--135/11).  TIE. The TIE scoring system provides a consumer=s estimated income within a $5,000 range (culminating in an over $100,000 category). (CX-1-X). TIE estimates income by 9 Most of these lenders sought lists of consumers with some type of finance tradeline. See supra n.7.

VOLUME 129 Opinion of the Commission modeling 23 attributes in the Standard Characteristics file. (CX-120; CX-119-I; Wiermanski 1803/17-24).  SOLO. The Solo model places consumers into one of 40 Aclusters,@ based on a modeling of 35 attributes in the Standard Characteristics file. (Wiermanski 1736/23-25; CX- 114; Davis 67/2-13). SOLO evaluates individual behavior and describes tendencies based on how individuals are using credit. (IDF-85). Examples of SOLO cluster categories are: AUrban Ethnics,@ AUrban Upscale,@ AEmpty Nesters,@ ASingle Strugglers,@ AKids and Cars.@ (CX-114-F).  P$YCLE. This model also assigns people to one of 60 Abuckets@ that are intended to estimate a consumer=s income producing assets. (Pellizzon 3446/16--3447/3, 3461/12-15; Davis 109/16--110/24, 214/1-9). Categories of buckets include, AThe Wealth Market,@ AUpscale Retired,@ ADownscale Retired.@ (TU-22-B).

 PIC. The PIC product uses the Standard Characteristics file to model the likelihood that a person owns financial service products. (TU-20; Weith 1864/6-10). On the Master File, the PIC option will indicate whether there is a negative or positive propensity to purchase, among other things, a home equity loan, a mutual fund, an installment loan or term life insurance. (CX-1-S).

4. Trans Union=s Other Target Marketing Products In addition to its Master File / Selects product and the Standard Characteristics models, Trans Union offers several other products derived from CRONUS, including:  TransLink / Reverse Append. This product provides merchants with names and addresses of bank card holders. The merchant submits to Trans Union a list of bank card numbers that were used to make purchases from the merchant. Trans Union then retrieves from CRONUS the name and TRANS UNION CORPORATION 491 Opinion of the Commission address of the primary cardholder10. (Weith 1823/22-- 1824/14; Dexter 1305/24--1307/6; Davis 89/25--90/10; CX- 126; CX-132-D; CX-133-B; CX-266). While a customer name is presumably already available to the merchant,11 the address is not. TUAB at 4. By purchasing TransLink, merchants can obtain a useful list of names and addresses without asking their customers for this information. TransLink is among Trans Union=s largest selling target marketing services and Trans Union is the only CRA that provides this type of Areverse append@ service. Until September 30, 1997, Trans Union appended SOLO, TIE and age data to TransLink lists; it currently only appends age data. (Dexter 1236/22-- 1237/25; Smith 1488/23--1489/5; CX-125-E; CX-129).  New Issues File. This file contains names and addresses of individuals who received credit within the last 90 days. It also discloses when an individual obtained the credit and the type of credit issued. (CX-4; IDF-106; Davis 42/16--43/1).  Emerging Consumers File. This file included individuals with only one tradeline from the prior twelve months. (Cabigon 1373/12-22; CX 329-F). To qualify, the tradeline must be open. (IDF-107). Trans Union discontinued the Emerging Consumers File in part because it feared that it might be Acommunicating information that we shouldn=t be communicating.@ (Davis 89/18-20).

10 Citibank does not permit Trans Union to use its credit card account numbers for reverse-append disclosure of names and addresses through Trans Link. (Marquis CX-188 at 147/20--148/1). 11 This information may not reflect the person who actually used the card with the merchant if the account is a joint account. (Weith 1824/16--1825/17, 1827/4-15).

VOLUME 129 Opinion of the Commission 5. Changes in Trans Union=s Practices The target marketing practices described above led the Commission to issue its complaint in 1992. In October 1997, contemporaneous with the effective date of the 1996 FCRA amendments, Trans Union discontinued some of the practices that were the most problematic under the FCRA. Specifically, the company stopped providing certain information about open dates of loans, high credit amounts, most loan types, and whether a student loan was closed. Trans Union also ceased providing to target marketing customers its modeled products (e.g., E-Val, PIC), its New Issues File, and its Emerging Consumers File.12 Trans Union changed its practices shortly after the 1996 FCRA amendments authorized for the first time civil penalties of $2,500 per FCRA violation (i.e., $2,500 per prohibited disclosure of consumer financial information). 15 U.S.C. ' 1681s(2)(A). Trans Union=s General Counsel Oscar Marquis stated that the company stopped providing certain lists in light of the new statute=s provision for civil penalties. (Marquis CX-188 at 174/23--175/6, 22-25). In the words of Stephen Dexter, a senior account manager with Performance Data, A[a]s of 10/1/97, the risk outweighed the reward for violating the FCRA.@ (Dexter 1280/19--1281/10). Jan Davis, Vice President and General Manager at Performance Data, also testified that Trans Union Ahad gone from an environment where the worst thing that could happen is that we would have to stop selling certain lists to a world where there were significant financial penalties.@ (Davis 142/21-25). A[B]efore it was a cease and desist penalty, it now became a $2,500 per occurrence penalty.@ (Dexter 1280/19-- 1281/10).

12 Trans Union continues to offer these products to entities extending so called Afirm offers@ of credit, a practice allowed under the prescreening provisions of the FCRA, described infra p. 18. TRANS UNION CORPORATION 493 Opinion of the Commission In December 1997, however, Trans Union reintroduced the practice of selling Atype of tradeline@ information - - e.g., information reflecting a specific type of account relationship between a credit grantor and a customer. Thus, Trans Union currently offers its customers access to the following information about consumers in its Master File: whether the consumer has an auto loan; a second auto loan; a bank card; a department store card; a finance loan; a 30/60/90 day finance loan; a mortgage; two or more mortgages; a gold, platinum or optima card; a student loan; an upscale retail card; seven kinds of business tradelines; a mail order trade; and auto loans. (CX-342; CX-315-D, E, G-M, Q-W; CX-332-B; Cabigon 1426/9-23, 1427/18--1428/3, 1429/9- -1430/2; Weith 1832/2--1833/6).

B. Trans Union and Its Competitors Trans Union has both CRA and non-CRA competitors in the target marketing industry. But Trans Union differs from its CRA rivals - - Experian and Equifax - - in at least two significant respects. First, Trans Union bases its target marketing lists on a minimum requirement of some tradeline activity. Although Experian, like Trans Union, also derives its target marketing database from its consumer reporting database, it does not require that a tradeline exist. (Smith 3428/18--3429/18). Similarly, Equifax also does not apply a minimum tradeline criterion. Its target marketing activities are limited to providing certain data to Claritas, Inc., which then offers target marketing products to customers. (IDF-40, 162-163).

Second, Trans Union is unique among CRAs because it provides credit data on individuals. By contrast, the other CRAs provide consumer credit information on an Aaggregated@ basis, i.e., information about a group of people. Both Experian and Equifax aggregate information about individuals= credit characteristics on a zip code or Azip-plus-four@ geographic VOLUME 129 Opinion of the Commission basis13. With zip-plus-four aggregation, a company essentially pulls all the credit reports of individuals within a geographical area covering 5-15 households (the zip-plus-four geographical area), adds all the credit data together, and then divides by the number of people in the area who have credit reports. (Smith 3290/11-24). This aggregation Ashows what a typical consumer looks like in that area as opposed to the specific consumer in the area.@ (Smith 3290/14-18; TU-112; TU-113). Experian does provide some Aindividual-level data,@ but it is limited by a consent agreement that the company entered into with the Commission in 1993 (hereinafter ATRW Consent@). Pursuant to the TRW Consent, Experian can disclose from its consumer reporting database only the following information about individuals: name, address, telephone number, mother=s maiden name, zip code, year of birth, age, any generational designation, social security number, or substantially similar identifier. (TU- 109; Smith 3287/11--3294/11). This information is commonly referred to as Aabove the line@ information because of its physical location on most consumer reports. See, e.g., TU-61(a). The TRW Consent prohibits the disclosure of Abelow-the-line@ information, i.e., most tradeline data including credit performance information14. TRW / Experian previously offered a reverse append product, but apparently discontinued the practice based on the TRW Consent=s provisions. (Smith 3295/9-17). The TRW Consent agreement does not address the legality of Experian=s current practice of disclosing credit information on an aggregated basis.

13 AZip-plus-four@ is the Postal Service=s more refined zip code system which adds four additional digits to identify a specific area within a zip code location.

14 One of Trans Union=s promotional letters states that AExperian comes closest as a competitor, but since they cannot provide to you any credit based data only the demographic data obtained from the credit reports (abiding by the Consent Decree with the FTC) . . . our data far outweighs their strength.@ (CX- 70-B).

TRANS UNION CORPORATION 495 Opinion of the Commission Equifax, Trans Union=s other CRA competitor, does not offer any individual credit data in its target marketing business. Prior to 1997, Equifax=s subsidiary, National Decision Systems (ANDS@), offered AAce Indicators,@ a product which disclosed information based on 39 credit performance characteristics aggregated at the zip-plus-four level. In 1997, however, Equifax sold NDS to Claritas, Inc., which now continues to use Equifax data at zipplus-four-level to offer the Ace Indicators product. (TU-103; TU- 114; TU-177(c); Pellizzon 3440--3446). Claritas edits its ACE Indicators data to ensure that data are not released that describe one household, one record, or one individual. This Aconfidentiality edit@ is applied where there are too few records in a zip-plus-four area. (Pellizzon 3471/12--3472/4).

Trans Union also faces competition from various non-CRAs. The industry leaders in this category are R.L. Polk & Company (APolk@), ACXIOM Corporation (AACXIOM@), Metromail Corporation (AMetromail@), and First Data Solutions (formerly Donnelly Marketing)(AFirst Data@). (Davis 161/5-16; Cleary 2942/4-18; Hinman 2199/19--2200/17; M. Smith 3299/22-- 3300/8). These competitors also furnish consumer information on an aggregated basis, e.g., at the household level or broader. (IDF- 157). While these companies obtain data from a host of sources, including state motor vehicle departments, county records, telephone directory white pages, census data, and self-reported data from surveys or product registration cards, such sources do not compare with the vast scope of information in Trans Union=s credit reporting database. CRONUS information covers a wider population and includes a more comprehensive range of instantly available information on individuals. CRONUS data are also significantly more accurate and timely.

The difference between Trans Union=s target marketing products and those its competitors sell is perhaps best described in Trans Union=s own words: Trans Union states that its Master File contains Athe freshest@ and Amost comprehensive@ data due to its VOLUME 129 Opinion of the Commission Arobust and extensive source of the original credit based information@ and that Trans Union has the largest data file of consumer credit information in the United States. (CX-268-A; CX-264-A; CX-75-B). Trans Union further describes the Master File as the Arichest source of individual-level data available@ (CX- 321-J), and asserts that its database is Akept fresh and current by nearly two billion updates supplied by credit grantors every month.@ (CX-72-B). Finally, Trans Union touts its advantage over other target marketing list providers, due to its ability to capitalize on the information in its credit reporting database. Trans Union boasts that it is:

Aa unique provider of credit-based marketing information. Our database is unmatched when compared to traditional direct marketing vehicles on the market today.@ (CX-260- B).

A[N]o one offers you a greater source of true individuallevel data than we do . . . . This unique resource includes financial and behavioral data on over 140 million consumers . . . This information is not only current, it is also highly accurate . . . . All information is based on actual behavior - - not self-reported or neighborhood values. Even our estimates - - of income, net worth, income producing assets, and home market value - - are modeled from actual observations for each individual in our file.

(CX-83-C). Such statements by Trans Union provide insight into the nature of the data it collects as a CRA and sells to target marketers.

TRANS UNION CORPORATION 497 Opinion of the Commission V. FCRA ANALYSIS:

ARE TRANS UNION=S TARGET MARKETING LISTS ACONSUMER REPORTS@ UNDER THE FCRA? A. Introduction As discussed above, Trans Union sells lists of consumer names and addresses to its target marketing customers. In creating these target marketing products, Trans Union applies various criteria to identify those consumers in its large database, CRONUS, who possess specific credit-related characteristics. The resulting lists thus communicate far more information to target marketers than simply names and addresses. A purchaser of a Trans Union target marketing list knows that every consumer included has at least one tradeline and possesses whatever additional characteristics the purchaser has specified. A key question in this case is whether Trans Union=s target marketing lists fall within the Act=s definition of Aconsumer report.@ If they do, then Section 604 requires that Trans Union sell them only to entities who have a Apermissible purpose@ as defined by the Act. According to the court of appeals, target marketing is not a permissible purpose. Trans Union, 81 F.3d at 230. Thus, if Trans Union=s lists are consumer reports then Trans Union has violated the FCRA by disseminating those lists for target marketing purposes.

Trans Union=s target marketing lists qualify as consumer reports if they communicate information that: (1) bears on a consumer=s Acredit worthiness, credit standing, credit capacity, character, general reputation, personal characteristics, or mode of living@ and (2) is Aused or expected to be used or collected in whole or in part@ to serve as a factor in determining credit eligibility. 15 U.S.C. ' 1681a(d)(1) (ASection 603(d)@). Our determination of whether Trans Union=s lists are consumer reports VOLUME 129 Opinion of the Commission does not require a mere application of fact, but instead requires a close examination and interpretation of Section 603(d).15 The court of appeals determined that the tradeline information in Trans Union=s lists meets the first prong of the consumer report definition - - i.e., it bears on one or more or the seven enumerated factors16. With respect to the second prong, however, the Court held there was insufficient evidence to support the Commission=s 1994 finding and remanded the case back to the Commission, stating:

On remand, if the FTC wishes to classify existence-oftradeline17 information as a consumer report, it must gather evidence that indicates that Trans Union intended the mere existence of a tradeline, as distinguished from payment history . . . to serve as a factor in credit-granting decisions, or, of course, that someone used or expected it to be used for that purpose. Evidence - - lacking here - - that credit decisions could be made, even in part, on such Aexistence@ information might be probative of Trans Union=s intent. If under this standard, tradeline-existence information is found not to [be covered by the definition of consumer report], the FTC may of course embark on a similar inquiry about any individual list criterion to which it objects. 15 We are mindful that, to the extent that Section 603(d) raises constitutional questions, we must construe the statute where fairly possible to eliminate such questions as long as such construction is not plainly contrary to Congress= intent. United States v. X-Citement Video, Inc., 513 U.S. 64, 78 (1994).

16 AThe first element does not seem very demanding, and we do not understand Trans Union to even contest the proposition that a person=s having two tradelines >bear[s]= on one or more of the seven enumerated factors.@ Trans Union, 81 F.3d at 231.

17 AExistence of a tradeline@ refers to the mere existence of tradeline information as determinative of whether the information disclosed is a consumer report under the FCRA. This term is distinct from the term Atype of tradeline@ which refers to the character or type of information that is disclosed. See discussion infra pp. 20-26.

TRANS UNION CORPORATION 499 Opinion of the Commission Trans Union, 81 F.3d at 233 (footnotes omitted)(emphasis added). With this direction in mind, we have reviewed what is now a full record in this case and find that the existence-of-tradeline information, as well as other information Trans Union disclosed in its target marketing lists, meets the Section 603(d) definition of a consumer report. We therefore conclude that Trans Union violated the FCRA by selling consumer reports to target marketers who lacked a statutorily permissible purpose. In reaching this conclusion, we examined Trans Union=s various target marketing lists - - the Master File / Selects, proprietary models, and reverse append products - - and find that information disclosed through these products is the type of information that is Aused@ and/or Aexpected to be used@ in whole or in part18 for the purpose of serving as a factor in establishing a consumer=s eligibility for credit19. Accordingly, these products are consumer reports and Trans Union cannot lawfully sell them for target marketing purposes.

18 Under Section 603(d), it is not necessary to show that the information communicated by the target marketing lists, standing alone, could be used to make a credit-related decision. We need merely determine that the information is used or expected to be used as Aa factor@ in such a decision. Trans Union, 81 F.3d at 233. Something serves Aas a factor@ if it A>contributes to the production of a result.=@ United States v. Wilson, 896 F.2d 856, 858 n.3 (4th Cir. 1990), citing Webster=s 3rd International Dictionary, 1971. 19 In 1995, the Commission took the position before the court of appeals that Trans Union=s lists, based on the Aexistence of two tradelines@ feature, were Acollected for the purpose of@ serving as a factor in credit eligibility decisions. The court of appeals rejected this argument on the grounds that the A>existence of a tradeline= seems not so much >collected= by Trans Union as created by it for organizing the nuts-and-bolts payment data upon which credit decisions are made.@ Trans Union, 81 F.3d at 232. On remand, Complaint Counsel and Trans Union have focused their argument on the used and expected to be used elements of the definition.

VOLUME 129 Opinion of the Commission We also analyzed the demographic information that Trans Union maintains in CRONUS and find that, based on the record before us, most of that information - - including name, mother=s maiden name, generational designator, address, zip code, telephone number, and social security number - - does not constitute a consumer report because there is no showing that it is used or expected to be used as a factor in determining credit eligibility.20 We conclude, however, that Trans Union, as a CRA, cannot lawfully disclose age information to target marketers because the record in this case shows that lenders use age as a credit factor and age bears on credit capacity21. Accordingly, products that Trans Union creates by way of its consumer reporting business that are based upon, or contain, references to age are consumer reports under Section 603(d) and their disclosure for target marketing purposes violates Section 604 of the Act.22 B. Analysis of Target Marketing Products 1. Background To determine whether the information communicated through Trans Union=s target marketing lists is Aused or expected to be 20 The Commission=s argument before the court of appeals focused on the relevance of Trans Union=s data to consumer=s eligibility for credit, and not to insurance, employment, or other items set forth in Section 603(d). The court of appeals followed suit as did the parties following remand. Accordingly, we limit our analysis to credit eligibility. 21 We also stress that, although the FCRA does not prohibit Trans Union from disclosing most demographic information, disclosure of such information may raise significant privacy concerns and may facilitate misuses including identity theft.

22 As noted, the TRW Consent permits Experian to use age information from its consumer reporting business for target marketing purposes. The TRW Consent is not before us in this matter and it is without precedential effect to this opinion.

TRANS UNION CORPORATION 501 Opinion of the Commission used@ in credit eligibility decisions, we reviewed record evidence detailing the various factors lenders use in evaluating credit eligibility. We focused in particular on the factors that are important in calculating credit scores - - a tool that many lenders use in evaluating credit eligibility. We also examined the factors that are important to lenders offering credit in prescreening promotions.23 Credit scoring systems use past credit information and other data to build models that predict a consumer=s likely future credit performance. (Rapaport 673/15-23). Credit grantors - - such as credit card issuers, retailers and finance companies - - use credit scores in deciding whether to grant an applicant credit, to make a preapproved credit offer, to reissue, increase or decrease a credit line, or for over-limit authorizations. (Rapaport 675/1-8, 680/23- -682/16). Most of the data used for credit scoring comes from CRAs. (Coffman 3825/18--3826/2).

Mr. Michael Rapaport of the Fair Isaac Company (AFICO@), the leading developer of credit scoring models, testified that credit scoring combines similar consumer credit files and then isolates the key 10 or 15 factors that are predictive of future credit performance for that group. (Rapaport 686/25--687/9, 779/20- 25)24. The record demonstrates that Trans Union was aware of the 23 Through the 1996 amendments to the FCRA (effective September 30, 1997), Congress included an additional statutory permissible purpose - - Aprescreening.@ The FCRA amendments allow consumer reporting agencies to provide to a credit grantor names and addresses of consumers meeting certain credit-related criteria so long as the credit grantor makes a firm offer of credit or insurance to the recipient. Furthermore, to afford consumers privacy protection, individuals receiving a prescreened offer must be told that they were chosen because they met certain criteria, that they have the right to opt out of appearing on future prescreened offer lists, and the procedures for opting out. 15 U.S.C. ' 1681m(d)(1).

24 The first step is called Ascorecard segmentation@ and is useful because isolating a group with similar traits within a population can improve the VOLUME 129 Opinion of the Commission factors that credit grantors use to predict future credit performance because Trans Union partnered with FICO to create its own models25. (Rapaport 672/25--673/6, 680/8-21). In fact, Trans Union and FICO together created scoring models to predict future credit risk generally (EMPIRICA), the likelihood a consumer will go bankrupt (Horizon), and the likelihood a mortgage account will become delinquent (Uniquote)26. (Rapaport 690/15--691/7, 692/21--693/7, 799/11--803/15). By working with FICO, Trans Union knew the categories of information in a consumer=s credit file that lenders used as predictive characteristics in credit scoring, and hence in credit eligibility decisions. Prescreening provides another way to determine the factors that bear on credit granting decisions. Trans Union was similarly aware of the prescreen criteria credit grantors use to make firm offers of credit.27 In prescreening, the credit grantor mails a firm predictive quality of the scoring model. (Rapaport 685/1--686/11, 767/13 -- 768/24; CX-88-E). After determining the predictive characteristics, FICO assigns values to Aattributes@ within each predictive characteristic - - e.g., four bank cards within Anumber of bankcards@ characteristic. The sum of the values of the attributes is the credit score. (Rapaport 687/16--688/6, 769/4-23, 851/2- 17). Trans Union=s credit scoring witness, Dr. John Coffman, flatly contradicted Mr. Rapaport by testifying that individual attributes have no meaning in credit scoring and that it is the combination taken as a whole that has value. (TUPF at 92). Having seen both witnesses testify, the ALJ found Mr. Rapaport more credible and we give deference to this determination. See In the Matter of Horizon Corp., 97 F.T.C. at 857 n.77. Moreover, based upon our review of the record, we find Mr. Rapaport=s testimony to be more persuasive because it was based on Trans Union=s own credit scoring models. 25 FICO has partnered with each of the three major credit bureaus to develop credit risk scoring products. (Rapaport 680/8-21). 26 Industry Options, refinements of the EMPIRICA model, offer scores for the bank card, personal finance, installment and auto loan industries. (Rapaport 692/1-18).

27 Trans Union=s subsidiary, Marketing Services, Inc., is engaged in the business of prescreening and reviews approximately four to five billion consumer files per month against criteria provided by approximately 100 to 150 prescreening clients per month. (Rock 2115/24--2116/13). TRANS UNION CORPORATION 503 Opinion of the Commission offer of credit to consumers who meet certain specifications or criteria. (Koppin 482/21-23, 488/20-23; Pendleton 357/22-- 359/5). CRAs like Trans Union generate and sell lists of consumers meeting the specified criteria. (Koppin 583/2-9; Zancola 668/22--669/1). For example, Chase Manhattan Bank sends firm offers of credit to consumers who meet its prescreen criteria, e.g., three open tradelines, no charge-offs, no payments 60 days past due. (CX-280-L). The specifications or criteria that credit grantors use in prescreen offers are based on statistical analyses of elements to predict credit behavior. (Koppin 489/19-- 490/11, 511/3-14; Pendleton 360/5-8; Zancola 669/21--670/10; McCoy 599/7-18, 606/21--607/7).

The record in this case includes substantial evidence of factors important to credit scoring and prescreening criteria. The record demonstrates that much of the information Trans Union discloses in its target marketing lists - - including the Master File / Selects, proprietary models, and TransLink / reverse append products - - is the same information that credit grantors, such as Wachovia Bank Card Services, Inc. (AWachovia@), First Card First Chicago NBD (AFirst Card@), the Northern Trust Company (ANorthern Trust@), Discover Card Brand, Novus Services, Inc. (ADiscover@), and Chase Manhattan Bank (AChase Manhattan@), use in credit eligibility determinations. Moreover, the record shows that Trans Union expected its credit grantor customers to use the information as factors in such determinations.

2. Master File / Selects Target marketers use Trans Union=s Master File / Selects to obtain a variety of information about consumers. See discussion supra pp. 6-8. As detailed below, the record shows that credit grantors use the same types of information as factors in credit granting decisions. The record also demonstrates that, in many instances, Trans Union expected credit grantors to use such information for credit granting decisions. Accordingly, the Master VOLUME 129 Opinion of the Commission File / Selects product falls under the FCRA=s definition of consumer report and Trans Union=s disclosure of it for target marketing purposes violates the Act.

a. Credit Limits Trans Union does not contest the fact that information about a consumer=s credit payment history, balance, and credit limit, is used by credit grantors in credit eligibility decisions; is covered by the definition of consumer report; and cannot be disclosed in target marketing. Indeed, the record confirms that credit limits, like payment history and balance, are pieces of information commonly used in credit scoring models. (Coffman 3848/16B3850/8, 3882/7B3884/4).

Trans Union instead argues that its target marketing lists did not provide any information about the credit limit on a consumer account. (TUPF at 189, 229). The record contradicts this statement. Evidence demonstrates that Trans Union did provide such information by selling lists of consumers who hold a premium bank card, which, as Trans Union expressly informed its target marketing customers, is defined as having a credit limit of over $9,999. (CX-64-A; Dexter 1271/17-20; Weith 1867/5-13). Accordingly, we find that where Trans Union has disclosed credit limit information to target marketers, it violated the FCRA by disclosing a Aconsumer report@ without a Apermissible purpose.@ b. Open Dates of Loans Until October 1997, Trans Union routinely provided its target marketing customers with information, obtained from CRONUS, about the open date of loans. (Cabigon 1377/10B1378/11). The record sufficiently documents that the open date of a loan is a piece of information regularly used by credit grantors. How long credit has been established and how recently a consumer has pursued such credit are each strong predictors of future risk. (Rapaport 774/6-19, 793/22--794/12). A[M]ost recent date opened indicates a pursuit of new credit, which is one of the types of characteristics that are indicative of future credit risk.@ TRANS UNION CORPORATION 505 Opinion of the Commission (Rapaport 792/12-15, see also 774/6-19, 793/22--794/12). It is, therefore, not surprising that scoring models look to the open date of tradelines to determine how long the consumer has had credit generally, and how long the consumer has had particular types of credit. (Coffman 3847/12-24, 3876/14--3877/20)28. Importantly, Trans Union=s own EMPIRICA and Uniquote models include the open date of loans among their predictive characteristics of credit risk. (CX-93-P; CX-93-Z-4)29.

Because this evidence clearly demonstrates that credit grantors use open date information to make credit eligibility decisions, we find that the FCRA=s definition of consumer report covers information on the open date of loans and that Trans Union violated the FCRA by disclosing such information in its target marketing lists.

c. Number of Tradelines The record further shows that the number of tradelines in a consumer=s credit file is also a predictive characteristic in [redacted] scoring models. (CX-93). Lenders like Chase Manhattan also use a Anumber of tradelines@ criterion in evaluating whether to grant credit30. We therefore conclude that 28 Mr. Rapaport testified that he has seen scoring models that score the open date of newest tradeline, open date of oldest tradeline, open date of newest finance loan, and open date of newest auto loan. (Rapaport 772/9B774/10; see also Stormoen 3154, 3155/1-6, 3210/5-21). 29 An open date is particularly significant in the mortgage context because it enables target marketers to determine the date a mortgage was taken out and the interest rate. The Mortgage Banc ordered a list of consumers with FHA mortgages opened between January 1994 and October 1995 with initial loan values between $75,000-$99,999 and $100,000-$150,000. (CX-17-A-B). 30 Chase Manhattan=s prescreen criteria require that credit eligible consumers have at least three lines of credit; its analysis of existing account holders showed that consumers with fewer than three tradelines had a higher incidence of failing to repay their accounts. (CX-280-L; Zancola 736/10-23). VOLUME 129 Opinion of the Commission credit grantors use such information in credit eligibility determinations.

Trans Union concedes that lenders use the number of tradelines as a factor in credit granting decisions31. It argues, however, that it does not disclose such information in its target marketing lists. Trans Union instead claims that it merely discloses the existence of a tradeline and the existence of particular types of tradelines. TUAB at 50. We find that the record shows otherwise.

In many cases, Trans Union does reveal the number of tradelines a consumer has by permitting its target marketing customers to order, for example, lists of people who have a bank card and a retail card and an auto loan - - in other words, three tradelines, the minimum requirement in Chase Manhattan=s prescreen. In addition, the record shows that even the Amere@ existence of a tradeline counts as a meaningful number (i.e., one) in credit scoring. See infra p. 25. Trans Union, therefore, violated the FCRA by disclosing in its target marketing lists information concerning an individual=s number of tradelines. d. Type of Tradeline Based on our review of the factors that credit grantors use in credit scoring and prescreening, we also find that type of tradeline information is itself a factor in credit eligibility decisions, regardless of performance on that tradeline. Consequently, this category of information also constitutes a consumer report. Type of tradeline information is particularly important in this case because it constitutes the lion=s share of Trans Union=s target marketing business. This fact is demonstrated by the list of selects 31 Although Trans Union argued in the first proceeding before the Commission that the number of tradelines is not information that credit grantors use in establishing a consumer=s eligibility for credit, Trans Union appears to have changed its position. Trans Union now argues that credit scoring models treat the characteristic Anumber of tradelines@ (but not the Aexistence of a tradeline@) as a predictive characteristic. TUAB at 16, 33--34. TRANS UNION CORPORATION 507 Opinion of the Commission that Trans Union offers and sells to its customers. (CX-1; CX- 342). As noted, Trans Union offers and sells target marketing lists that provide the names of consumers who have a bank card or a mortgage or an auto loan, among other type-specific credit relationships. See supra pp. 7-8.

Evidence in the record also indicates that type of tradeline information is used as one of possibly a dozen predictors of future risk in credit scoring. For example, the existence of a bank card is given weight in Trans Union=s own Uniquote and Horizon scoring models and other scoring models. (Rapaport 785/4--786/7; Coffman 3869/16--3870/9)32. One of Wachovia=s scorecards also assigns points for the presence of a bank card. (CX 275-R; Pendleton 400/22--401/4)33. The existence of a finance company tradeline is also scored in Trans Union=s EMPIRICA and Horizon models. (CX-93-H; Rapaport 789/15--790/15). According to Mr. Rapaport, this factor is scored because pursuit of new credit, particularly with a finance company, tends to be more indicative of future credit risk. (Rapaport 792/17--793/21). For example, finance company users are people who have had credit problems in the past, and quite likely, have had a bankruptcy. (Scott 32 Section 701(d) of the Equal Credit Opportunity Act (AECOA@), 15 U.S.C. ' 1691(d), requires that when lenders deny credit applications, they must provide reasons for the denial. Credit scoring models generate descriptions of reasons why an applicant=s score deviates from an optimal score which can be used by lenders to comply with ECOA obligations. One of EMPIRICA=s reasons for denial is Alack of bank cards.@ (CX-87; Rapaport 851/23--853/6).

33 Trans Union has crafted a novel, but unsupportable argument that the only reason Wachovia Ascored@ a bank card reference was not as a factor relevant to establishing credit eligibility, but to establish whether the applicant was willing to list the presence of a bank card on his or her application - - a test of character if you will or, as described by Complaint Counsel, a lie-detector test. We agree with Complaint Counsel that this interpretation is not supported anywhere in the record, even though Trans Union had the opportunity to question Wachovia=s witness about it.

VOLUME 129 Opinion of the Commission 2855/23--2856/7). The existence of one mortgage tradeline, again without regard to performance on that account, also is used as a predictive attribute in credit scoring. (Coffman 3862/9--3864/5). In Discover=s scoring model, for example, an applicant receives points for an open mortgage tradeline, regardless of the payment status of that mortgage. (Stormoen 3153/8--3154/2, 3204/5-17). Discover also assigns points for the existence of a retail tradeline and a bank card. (Stormoen 3150/16--3151/3). Indeed, Mr. Rapaport testified that he has seen as predictive characteristics in scoring models many of the types of tradelines disclosed by Trans Union=s target marketing lists, including the existence of a bank card, retail account, finance loan, auto loan, and mortgage loan. (Rapaport 772/24--774/19).

Trans Union argues that credit scoring does not take into account particular types of tradelines but instead the number of types of tradelines. TUAB at 46--50. The record, however, shows that this claim is not true. In fact, Trans Union=s own credit models score those who have [redacted] differently from those who have [redacted]; and they score consumers who have a [redacted] differently from those who have a [redacted]. (CX-93; Rapaport 785/4--786/15, 789/15--790/15). Furthermore, the testimony of Mr. Rapaport and other witnesses, as well as the documentary evidence, confirms that the existence of Amere@ types of tradelines - - e.g., a bank card, a finance tradeline or a mortgage tradeline - - without regard to performance on those accounts, conveys to credit grantors useful information about an individual=s creditworthiness. Also, such information is in fact used in the credit scoring systems credit grantors employ34. 34 Trans Union also claims that the ALJ improperly ignored the testimony of Mr. Connelly who stated that the credit risk model requires all information on all tradelines to run - - and that it could not operate using only the information from the Master File. TUAB at 33. This argument also misses the point. The Commission is not finding, and need not find, that the information in the Master File is all a credit grantor needs to make a credit decision. We are simply required to determine what information is used or expected to be used, in whole or in part, by credit grantors as a factor in determining a consumer=s eligibility for credit. See Section 603(d) of the FCRA.

TRANS UNION CORPORATION 509 Opinion of the Commission The importance of type of tradeline as a factor in credit eligibility is further illustrated in the context of prescreening. One Wachovia prescreening model considers the existence of a bank card so significant that it sets a lower minimum credit score for persons with a bank card as compared to those without. (CX-275- I,J; Pendleton 396/2-13). Wachovia explained that Aindividuals who do not have any bank card experience are significantly riskier.@ (Pendleton 395/17--396/1). First Card finds the existence of a finance tradeline so significant that it excludes from one of its prescreen offers consumers with a small finance company tradeline. (CX-278-B; Koppin 517/9-14). Similarly, Northern Trust=s 1996 home equity prescreen offer rejected files without at least one open mortgage. (CX-283-A). Chase Manhattan=s prescreen requires at least two qualifying tradelines, one of which cannot be a refinanced loan or student loan. (CX- 280-O; Zancola 712/20--713/3).

To rebut the significance of type of tradeline in the prescreening context, Trans Union argues, based on the testimony of Ms. Judy Pendleton of Wachovia, that the ALJ failed to understand how prescreening works and overestimated the importance of type of tradeline in prescreen criteria. TUAB at 55- 57. According to Trans Union, prescreening models first apply Aexclusionary@ factors, eliminating consumers whose credit files show, among other things, certain derogatory credit information. In Trans Union=s view, when Wachovia looks to see if a consumer has a bank card, it is actually looking to see if the consumer has a near perfect bank card. (Pendleton 439--441). Here again, however, the record does not support Trans Union=s claim. First Card=s prescreening model rejects a consumer with a small finance company tradeline, even a tradeline that has met Agood performance@ criteria. (CX-278-B). Similarly, Chase Manhattan=s prescreen would reject a consumer whose only credit account is a student loan or a refinanced loan, even if such a VOLUME 129 Opinion of the Commission tradeline passed the Agood performance@ test. (CX-280-Z-34). Even in the Wachovia example, for a consumer that passes through a gauntlet of exclusionary criteria, the presence or absence of a bank card may determine the range of acceptable credit scores for a consumer to receive credit. (CX-275-I, J; Pendleton 396/2-13).

These examples demonstrate that, even when all relevant consumer tradelines pass the exclusionary criteria, credit grantors value specific types of tradelines differently - - e.g., a bank card tradeline is generally more highly valued than a finance tradeline. This is confirmed by Ms. Pendleton=s testimony about the risk associated with individuals who do not have any bank card experience, and Mr. Rapaport=s and Mr. Scott=s testimony that people with a finance tradeline are riskier and are likely to have had a bankruptcy. (Pendleton 395/17--396/1; Scott 2855/23-- 2856/7; Rapaport 792/17--793/21). In addition, Trans Union=s argument that each individual criterion for a prescreen must be examined against other criteria that impose additional requirements simply fails to address the plain language of the FCRA=s Section 603(d) definition of a consumer report - information that is used Ain whole or in part@ in credit eligibility determinations35.

35 Trans Union also argues, based on the testimony of First Card=s Mr. Koppin and Discover=s Mr. Stormoen, that the existence of a type of tradeline is not relevant to determining credit eligibility; rather, it is performance information found in that tradeline that counts. TUAB 13-14. (Koppin 547B548; Stormoen 3180/6-24). Notwithstanding the selected statements of Mr. Koppin and Mr. Stormoen, the weight of the evidence indicates that the existence of a type of tradeline is used as a factor in determining credit eligibility. The portion of Mr. Stormoen=s testimony that Trans Union highlights is belied by the remainder of his testimony describing predictive characteristics that have nothing to do with credit performance, such as number of retail and bank card tradelines, existence of a mortgage, age of oldest tradeline, and even existence of a tradeline. (Stormoen 3150/3--3151/3, 3153/8--3154/2, 3204/5-17, 3155/11--3156/4). Similarly, Mr. Koppin=s testimony must be viewed against First Card=s documentary evidence, described above, that requires the rejection of consumers who have a small company finance tradeline, even if they meet the other good performance criteria of no derogatory or adverse file flags, no trades currently 30 days past due, no trades historically 90 days past due, among others. (CX-278-B). TRANS UNION CORPORATION 511 Opinion of the Commission Finally, Trans Union=s own promotions clearly indicate that it Aexpected,@ within the meaning of the FCRA Section 603(d), type of tradeline information to be used in credit granting decisions. Indeed, it boasted, Asince credit has been established [for individuals on the student loan list], one could argue that this list would have higher pass rates through the credit bureaus.@ (CX- 136). According to Trans Union, the premium bank card target marketing list identifies individuals Awho have been approved for this high amount of credit in the past.@ (CX-64-A). In light of these facts, the Commission finds that a type of tradeline, even without regard to performance on that account, is valuable information used by credit grantors to decide whether to extend credit. Therefore, because this information is both used and expected to be used in credit granting decisions, such information is covered by the FCRA=s definition of consumer report, and Trans Union=s disclosure of type of tradeline information to target marketers violates the Act. e. Existence of a Tradeline As discussed, the court of appeals remanded this case so that the Commission could determine whether there was sufficient evidence to show that the mere Aexistence of a tradeline@ is information used, expected to be used, or collected for the purpose of establishing an individual=s eligibility for credit. Trans Union, 81 F.3d at 233. The consumer names and addresses that Trans Union sells in its target marketing lists have met the Minimum Criteria, 36 including that his or her CRONUS (i.e. 36 The exception is the lists of consumers Trans Union disclosed or discloses through its TransLink, New Issues and Emerging Consumer products. These names come directly from CRONUS rather than the Master File and thus do not necessarily meet all the elements of the Minimum Criteria. See VOLUME 129 Opinion of the Commission Acredit@) file has at least one tradeline or, for lists generated prior to January 1998, at least two tradelines. See discussion supra p. 6. The record in this case, however, shows that Trans Union=s customers do not purchase only the names and addresses of consumers with a tradeline meeting the Minimum Criteria. Instead, they purchase the names and addresses of individuals who also meet other criteria, e.g., consumers who also have an upscale retail card and an auto loan. Indeed, Trans Union=s promotional materials recognize that customers do not simply request the Master File list and the materials encourage them to narrow down that list based on additional criteria that meet their needs.

You=ll find . . . in our Master File . . . over 140 million consumers. Of course, you won=t want to reach all of them. That=s why each consumer record includes more than 350 variables that allow you to segment, select, target, and sell with unmatched precision.

(CX-79-B). In other words, Trans Union does not sell lists of people who just have one tradeline. Instead, Trans Union sells lists of people with a tradeline who meet other specified criteria. Nonetheless, because the Minimum Criteria apply to virtually all of Trans Union target marketing lists, we make the following determinations based on our review of the record. First, the Minimum Criteria for appearing on Trans Union=s base marketing lists are not the Amere existence of tradeline.@ Rather, the Minimum Criteria also reveal, among other things, the existence discussion supra p. 10. Every consumer identified through these products does, however, have at least one tradeline, the existence of which bears on credit eligibility and is used in credit decisions. Specifically, Trans Link / Reverse Append discloses the names and addresses of consumers with a bank card (and in some instances age and other data) and the New Issues File and Emerging Consumers File disclose (or disclosed) consumers with open credit tradelines. Id. Accordingly, the lists Trans Union sells through these products constitute consumer reports.

TRANS UNION CORPORATION 513 Opinion of the Commission of a recently active and current credit relationship. Specifically, the prerequisite for appearing on a list is: (1) the existence of either two tradelines active within the last six months or one tradeline active within the last six months with an address matched to an outside vendor file, and (2) that the tradeline must have no closed date, must not be disputed, and cannot be a collection tradeline or a public record tradeline. Second, the record shows that even these Minimum Criteria are more important than they initially appear. Interestingly, Trans Union=s Minimum Criteria are substantially similar to FICO=s minimum criteria for every credit scoring model that the three national CRAs use. Like the names in the Master File, for example, FICO requires initially that a consumer have at least one open line of credit updated within six months that is not the subject of a dispute and that gives no indication that the subject is deceased. (Rapaport 763/4-23; CX-89-S; Wiermanski 1795/21- -1796/20).37 The importance of the existence of a tradeline is further revealed through the scorecard segmentation process - - a mechanism for grouping like people together to better determine future risk. See supra n.24. Significantly, Mr. Rapaport testified that each of the three national credit reporting agencies uses scorecard segmentation and that they each have a scorecard for consumers who have only one tradeline; consumers who have either zero or two or more tradelines are evaluated through different scorecards. (Rapaport 770/15--771/5). 37 When a loan applicant does not have a tradeline, Trans Union=s EMPIRICA model cannot calculate a score and returns a message AEMPIRICA Not Scored B Insufficient Credit.@ (Rapaport 764/12-15; CX-87-A). In fact, most credit grantors will not approve an applicant where there is no score due to the absence of a tradeline, although some will build custom scorecards for those who have no tradeline. (Rapaport 766/1-19). Discover Card, for instance, declines credit applicants whose credit reports indicate no tradeline. (Stormoen 3155/24B3156/4).

VOLUME 129 Opinion of the Commission Prescreening criteria similarly illustrate the significance to credit grantors of having a tradeline. The Chase Manhattan prescreen criteria require that a person have at least one tradeline verified within the last six months. (CX-280-L; Zancola 723/3- 6). Similarly, in the Wachovia prescreen, the first factor applied to a consumer=s credit file, without reference to any performance information, is whether the consumer has a line of credit which has been open for a specified duration. (Pendleton 393/3-23; CX- 275-F; CX-276-A).38 Further, the record demonstrates that Trans Union Aexpected,@ within the meaning of the FCRA, that information regarding the existence of a tradeline (or two tradelines) would be used in credit eligibility decisions. Trans Union promoted: AAny adult with at least two active tradelines of credit is represented on the Masterfile.@ (CX-33-A; CX-69-A). AAny individual with at least two lines of credit is included in the [Master File=s]140 million plus names and addresses.@ (CX-61-A). Trans Union=s lists are Anot just ordinary lists but lists of people who are active users of credit.@ Trans Union, 118 F.T.C. at 845. These record examples establish that the information Trans Union routinely discloses through its Master File / Selects product is used and/or expected to be used by credit grantors in eligibility decisions. Therefore, the target marketing lists created from this product are consumer reports and Trans Union violates the FCRA by disclosing them to target marketers without a permissible purpose.

38 One of Wachovia=s prescreens requires one open tradeline for two years; another Wachovia prescreen requires at least one open tradeline for one year. (CX-275-C, F; CX-276-A, C; Pendleton 393/3-23, 414/17--415/21). In addition, First Card=s prescreening criteria also reject consumers with no tradeline, although First Card extends credit to consumers without tradelines under special circumstances. (CX-278-A; Koppin 515/17-24, 516/12-20, 526/9-528/4). Also, Northern Trust=s 1993 and 1996 home equity prescreens look to tradeline activity within the last year. (CX-281-A; CX-283-A; McCoy 603/9-25, 611/8-12).

TRANS UNION CORPORATION 515 Opinion of the Commission 3. Proprietary Models39 The lists Trans Union generated through its proprietary models -- E-Val, TIE, SOLO, P$YCLE, and PIC -- similarly convey information that is used or expected to be used in credit eligibility determinations and Trans Union=s disclosure of them to target marketers similarly violated the FCRA40. Trans Union=s promotion of each of these products41 reveals its expectation that they would be used as factors in establishing credit eligibility. Trans Union marketed its E-Val product -- its scoring system that estimates the amount of equity available in a consumer=s home -- to lenders as:

39 Trans Union argues that Complaint Counsel Aessentially@ concedes that information disclosed in its proprietary target marketing models and its reverse append product is not a consumer report. TURB at 1. We find to the contrary. Complaint Counsel=s brief expressly states that the modeled products Aare not only derived from and disclose credit eligibility factors (IDF-82, 85, 92, 108), they are specifically marketed by Trans Union for both target marketing and credit eligibility uses (IDF-87, 89, 93-94).@ CCAB at 88, n.124. Complaint Counsel=s brief also discusses the privacy-intrusive aspects of reverse append and Trans Union=s use of the product without a permissible purpose. Id. at 10.

40 Trans Union objects that the ALJ Aleaped@ without analysis to the conclusion that its proprietary models are Aconsumer reports,@ arguing that the ALJ made no finding about whether the models were used or expected to be used as factors in establishing credit eligibility. TUAB at 3--4. We agree that such additional findings must be made before Trans Union can be held responsible for FCRA violations, and based on the record now before us, we make such findings here.

41 Trans Union openly characterized its AStandard Characteristics,@ upon which all of its five proprietary models were based, as Acorrelat[ing] highly with lending activity.@ (CX-263-A).

VOLUME 129 Opinion of the Commission the ideal tool for marketers of home equity lines of credit or other secured loans. It clearly identifies homeowners who have both equity available and an interest in securing credit. . . E-VAL can be of significant value to a wide range of marketers: Banks, Credit Unions, Brokerages, Mortgage Brokers, Mortgage Guarantors, Fannie Mae/Ginnie Mae Agencies, Ad Agencies, Modelers, [and] Catalogers.

(CX-118). The record also shows that Trans Union=s lender/customers requested E-Val home values on individuals in deciding whether to make loan offers. (CX-23; CX-24; CX-38). In its seller=s guide, Trans Union describes the following uses for its income estimator model (ATIE@): Ain credit risk scoring for new or existing accounts . . . in existing prescreen criteria . . . as a supplement to credit application data . . . to set initial credit limits.@ (CX-119). The guide also states that A[t]he most prominent markets for TIE are: credit grantors (including bank card issuers, finance companies, retailers, and auto finance companies) and other lenders (retail banks, savings & loans, and credit unions).@ Id42. Further, Trans Union=s Vice President Chester Wiermanski testified that TIE was intended for use in approve/decline decisions. (Wiermanski 1719/25--1720/20)43. 42 A product brochure for TIE states that customers can use TIE Awith confidence@ to A[f]ine tune credit limits and loan conditions on credit applications . . . >[r]ed flag= applicants whose low income estimate may indicate the need for additional verification,@ and A[f]lag accounts to increase/decrease lines of credit.@ (CX-120-B, C).

43 Trans Union argues that the ALJ=s finding that it A>uses TIE in credit granting= (F 93)@ is not supported by the record and that the record shows it merely Acontemplates@ using TIE in credit granting. TUAB at 4. Trans Union=s characterization of the record is correct. Still, Trans Union=s Acontemplation@ (the actual term used was Aenvisioning@) of TIE for use in credit approve/decline decisions demonstrates that Trans Union expected information in TIE to be used in credit eligibility decisions. Because Section 603(d) of the FCRA covers such expected use, information in TIE is a consumer report.

TRANS UNION CORPORATION 517 Opinion of the Commission Clearly, Trans Union expected lenders to use information in TIE in credit granting decisions and knew that they did, in fact, use such information in these decisions.

Trans Union=s internal seller=s guide for SOLO also notes that ASOLO is most often used by credit grantors for non-preapproved offers, such as home equity offers or secured card offers.@ (CX- 115-O). It also discusses using SOLO in preapproved offers of credit. (CX-115-Z-2). Trans Union, therefore, expected this product also to be used as a factor in credit granting, and conceded this point when it stated in oral argument that its target marketing customers were using SOLO in credit eligibility decisions. (Oral Arg. Tr. 100/10-13; Davis 67/19--68/4). The AP$YCLE@ model uses CRONUS data to estimate a consumer=s income producing assets. Trans Union=s promotional brochure for P$YCLE states:

P$YCLE allows marketers to segment consumers according to affluence, financial product and service usage, and account balances . . . P$YCLE, designed for financial service companies.

(TU-56).

Similarly, PIC, Trans Union=s model that predicts the likelihood that an individual owns financial service products, is promoted as follows:

It=s easier to acquire individuals= money if you know where they keep it . . . Imagine the benefit of knowing which financial vehicles an individual investor will choose. That=s exactly what PIC (Prospect Identification and Classification) offers to marketers of mutual funds, money market accounts, insurance, annuities and home equity credit lines . . . Tap into the richest source of VOLUME 129 Opinion of the Commission individual-level financial data in America . . . a new, higher level of predictive behavior . . . and a profitable way for your company to acquire new business . . .. (emphasis added) (TU-20).

All of this evidence plainly shows that Trans Union fully expected lenders to use information in Trans Union=s proprietary models to find the most eligible and profitable targets for the lenders= promotions44. In addition, each of the models provides information about the consumer=s income - - a significant factor Aused@ in credit eligibility decisions. For E-Val and SOLO in particular, the record shows that lenders used the model scores or categories to make such decisions.

Consequently, we find that Trans Union=s proprietary models were Aused or expected to be used@ in credit eligibility decisions, and thus constitute consumer reports within Section 603(d) of the FCRA. By disclosing these reports to target marketers which do not have a permissible purpose under the Act, Trans Union has violated the FCRA.

44 Trans Union=s promotions also disclose in general terms what inputs it uses to generate its model results. For E-Val, Trans Union announced, AOpen mortgage dates, initial mortgage amount, presence of additional mortgages - - [w]ith this information, we can create reliable estimates of the length of residence and the actual equity ratio and equity amount available to homeowners. By deducting existing mortgage balances from the estimated home value, and applying an adjustment factor of 75%, you now possess invaluable data on 62 million U.S. homeowners.@ (CX-118-B). For SOLO, Trans Union stated: A[G]roups individual consumers with similar lifestyle, spending and payment behaviors into clusters.@ (CX-114-B). For P$YCLE, AThe P$YCLE model draws on the five economic and demographic factors that have the greatest effect on consumers= financial behavior: income producing assets, total household income, age of household head, home ownership, urbanization.@ (TU-56). With TIE, the sellers guide tells Trans Union=s salespeople that the model uses 23 key characteristics that predict income - - Aage and type of accounts, amount of available credit, amount of credit used, number and type of new accounts.@ (CX-119-I). TRANS UNION CORPORATION 519 Opinion of the Commission 4. TransLink - - the Reverse Append Product TransLink is a special service through which Trans Union provides its merchant customers with names and addresses of the people who have used their bank cards to make purchases from a particular merchant. TransLink differs from Trans Union=s other products because the merchant already has access to some of the information contained in the reverse append list, i.e., the name of a purchaser and the account number based on the customer bank card transaction record. Trans Union does, however, communicate a variety of information that the merchant does not already have. Specifically, by matching the merchant=s consumer information with the information in CRONUS, Trans Union confirms the accuracy of the merchant=s data at the time it generates the list. Moreover, Trans Union communicates the consumer=s address - a valuable asset - - and, as previously noted (see supra p. 10), can also append age data45 to its reverse append lists. Despite the fact that the merchant purchasing a reverse append list already has a name and account number, the FCRA analysis for TransLink is the same as for Trans Union=s other products. This analysis requires us to determine whether the information Trans Union sells through reverse append is a consumer report and whether the recipient of the information has a permissible purpose under the Act. Trans Union accesses its consumer reporting database to obtain, match and disclose names and addresses of consumers with a certain type of credit card, in this case an active bank card. The matching of a bank card number with a consumer=s name and address, and the communication of that matched information to a merchant constitutes a consumer report under the Act. As discussed above, such type of tradeline information is used, or expected to be used, in determining credit eligibility. Accordingly, reverse append lists are consumer reports 45 This is significant because we find that age data meets the FCRA=s definition of a consumer report. See discussion infra pp. 30-31. VOLUME 129 Opinion of the Commission and, because target marketing is not a permissible purpose under the Act, Trans Union cannot disclose these lists to its target marketing customers.

C. Analysis of Demographic Information Section 603(d)=s definition of a consumer report requires not only that the information be Aused@ or Aexpected to be used@ in a credit decision, but also that the information bear on a consumer=s Acredit worthiness, credit standing, credit capacity, character, general reputation, personal characteristics, or mode of living.@46 When viewed against the FCRA=s statutory purpose of protecting the privacy of personal credit information,47 we find that the Abearing on@ limitation, set forth in Section 603(d) excludes from the FCRA=s definition of consumer report certain predominantly identifying information including: name, mother=s maiden name, generational designator, telephone number, and social security number. Although the record shows that certain lenders exclude from prescreening offers consumers who have a generational designator (e.g., AJr.,@ ASr.,@ etc.), or do not have a social security number, they do so only based on concern about identity, i.e., accessing the file of the correct individual. This information does not, however, bear on creditworthiness, credit capacity, credit standing, character, general reputation, personal characteristics, or mode of living, unless such terms are given an impermissibly broad meaning.48 46 See supra n.2 and p. 14. Although the court of appeals viewed this part of the definition as Anot very demanding,@ it did so in the context of examining the impact of the existence of two tradelines. Trans Union, 81 F.3d at 231. This part of the definition is not as easily met for other information considered in credit decisions.

47 See discussion infra pp. 37-39.

48 See also supra n.21.

TRANS UNION CORPORATION 521 Opinion of the Commission The treatment of two other categories of demographic information - - (1) address and (2) age - - also merits additional analysis. With respect to the address of an individual, the court of appeals noted the ease with which zip codes, a component of an address, could be used in lending decisions to ensure that only the wealthy - - for example people living in the Beverly Hills, California zip code 90210 - - would be eligible for loans. Trans Union, 81 F.3d at 232. Regardless of whether this information might bear on credit worthiness, nothing in the record before us establishes that zip codes are actually used, or expected to be used as a credit eligibility factor in scoring or as a credit criterion in prescreening49. Absent such evidence, the FCRA does not prohibit Trans Union=s disclosure of simple address information to target marketers.50 On the other hand, the record shows that an individual=s age does bear on their credit capacity and is used in credit granting decisions. Witnesses from both Northern Trust and Chase Manhattan testified that their companies do not offer credit to consumers who are younger than the legal age. (McCoy 631/19- 24; Zancola 711/9-16). In addition, Discover Card looks at Alongevity@ of Aeconomic dealings people have,@ which may be determined by a consumer=s age. (Stormoen 3190/20-- 3191/7). Mr. Rapaport also testified that some scorecards use age as a factor. (Rapaport 847/8-16). The record, therefore, demonstrates that lenders use age information as a factor in credit granting 49 Mr. Rapaport testified that zip codes are not used in credit bureau scoring. (Rapaport 847/17-21). Mr. Koppin stated that zip codes are used in extracts to narrow the geographic area of First Card=s prescreen list but zip codes are not used as a credit criterion. (Koppin 582/1-16, 583/25--584/18). 50 Although some lenders will not extend credit to consumers with a P.O. Box address, we do not find that the P.O. Box feature bears on Acredit worthiness, credit standing, credit capacity, character, general reputation, personal characteristics or mode of living.@ VOLUME 129 Opinion of the Commission decisions51. Further, age clearly bears on credit capacity where state laws restrict contracting with minors. Therefore, age information falls within the definition of a consumer report and its disclosure by a CRA to target marketers violates the FCRA. D. Trans Union=s Remaining FCRA Arguments Trans Union contends that the ALJ=s decision is unsupportable because the ALJ ignored the expert testimony of Dr. John Coffman, Mr. Kenneth Scott, and Mr. Barry Connelly, each of whom testified that the existence of a tradeline does not factor into credit eligibility decisions. TUAB at 31-33. Although Trans Union called these witnesses as expert witnesses, it is not clear that the ALJ found them qualified as Aexperts.@ Indeed, the ALJ stated that two of the purported experts were Anot credible on this issue@ and found that Dr. Coffman showed bias through his inconsistent testimony and that Mr. Scott had no relevant experience to support his testimony. IDF at 86, n.183. We agree with the ALJ.

The record indicates that Dr. Coffman made internally inconsistent statements on direct and cross examination. Dr. Coffman stated on direct that none of the information sold by Trans Union was used as a factor in determining credit eligibility, with the exception of P.O. Box information, which was used to exclude certain consumers from prescreened offers of credit. (Coffman 3840/5-21). On cross examination, however, Dr. Coffman admitted that information on the existence and number of mortgages, auto loans, and open bank cards has been used as a predictive attribute in some scoring models. (Coffman 3862/5-- 3863/22, 3868/16--3870/9).

51 We recognize that the Equal Credit Opportunity Act generally prohibits credit decisions based on age. 15 U.S.C. ' 1691(a). There are exceptions, however: a lender can favor applicants who are age 62 or older. A lender also can consider age if it bears on other elements of creditworthiness. For example, a lender can consider whether an applicant is close to retirement age, which could impact future income. Section 202.6 of Regulation B (the implementing regulation of the ECOA), 12 C.F.R. ' 202.6. TRANS UNION CORPORATION 523 Opinion of the Commission We agree with the ALJ=s finding that Mr. Scott=s experience was not sufficient to support giving particular weight to his testimony. Mr. Scott=s experience was in marketing credit cards and not in making credit eligibility decisions. While Trans Union argued that Mr. Scott testified extensively regarding his experience in credit eligibility, it offered no evidence to support this assertion. TUAB at 22, 33. We find that Mr. Scott=s testimony, viewed in the best light for Trans Union, supports only that he attended meetings at American Express where credit eligibility criteria were discussed. (Scott 2616/22--2617/16). This fact only demonstrates that he had indirect knowledge of the subject matter. We thus find that the testimony of credit grantor witnesses, with far more intimate knowledge of the complex array of factors that influence credit eligibility decisions, substantially outweighs Mr. Scott=s testimony.

As President of Associated Credit Bureaus (AACB@), the primary trade association for credit bureaus, Mr. Connelly serves the interests of ACB=s members and Trans Union is one of the three main dues paying members. (Connelly 2565/4-12, 2566/4-11)52. Although the ALJ did not specifically comment on Mr. Connelly=s testimony, we have throughly considered it and determined that it is also entitled to little weight. Further, Mr. Connelly testified that he had no experience as a credit grantor or credit scorer and that he did not know how Trans Union=s credit scoring model worked. (Connelly 2560/11--2561/18, 2601/20-- 2602/1).

Trans Union further notes that Complaint Counsel produced no expert testimony showing that the information disclosed in Trans Union lists is used by credit grantors in credit granting. 52 In addition, Trans Union=s CEO and a Senior Vice President sit on ACB=s Board of Directors and thereby control Mr. Connelly=s budget as well as his salary. (Connelly 2570/20B2571/8, 2572/5-11). VOLUME 129 Opinion of the Commission Complaint Counsel did, however, provide sufficient factual evidence - - both contemporaneous documentary evidence and non-expert testimony - - that Trans Union=s target marketing lists disclose information used in credit granting decisions and constitute consumer reports that cannot be disclosed for target marketing purposes53. Hence, Complaint Counsel was not required to present expert testimony to support the complaint allegations.

E. Conclusion Based on a thorough review of the record, including the testimony, we find that Trans Union=s target marketing lists are indeed consumer reports under the FCRA because they contain information that bears on the factors set forth in Section 603(d)(1) and is used or expected to be used as a factor in determining a consumer=s eligibility for credit. By selling these lists to target marketers without a permissible purpose, Trans Union violates the FCRA. This conclusion applies to Trans Union=s Master File / Selects; proprietary models; and TransLink / reverse append products.

Trans Union=s disclosure to target marketers of information on the existence of a tradeline violates the FCRA. Further, Trans Union=s disclosure in its target marketing products of other 53 Trans Union=s remaining arguments are, at best, splitting hairs as they have little impact on the core of Trans Union=s practices and, thus, our analysis. For example, when Complaint Counsel showed that Wachovia=s PCL prescreen requires at least one tradeline open for a year, Trans Union argues that such tradeline must also have a balance update, not just an open date. TURB at 18. In addition, responding to Complaint Counsel=s showing that First Card=s prescreen rejects any file showing a finance tradeline, Trans Union points out that the precise criteria are A>small company finance trade[line] with a current balance of $1.00 or more, excluding student loans.=@ TURB at 19. The inquiry at issue, however, is whether the information disclosed by Trans Union, including the existence of a credit account and specific types of credit accounts, are used, in whole or in part, as factors in credit eligibility determinations. The more detailed description by Trans Union of Wachovia and First Card=s prescreen criteria support the same answer of Ayes.@ TRANS UNION CORPORATION 525 Opinion of the Commission information, such as the existence of a type of tradeline, open date of tradeline, home equity information, and income estimations, among other list criteria described above, also violates the Act. Finally, the record in this case supports, with one exception, the lawful disclosure of most demographic information. The one exception, however, is age information which the record here shows is used in credit decisions, bears on credit capacity, and is accordingly a consumer report that cannot be disclosed in target marketing.54 VI. CONSTITUTIONAL ANALYSIS Trans Union raises two constitutional defenses in this matter. Trans Union first asserts that, by barring it from selling target marketing lists, the FCRA violates the First Amendment of the United States Constitution. Second, Trans Union claims that the FCRA=s definition of consumer report is unconstitutionally vague under the Fifth Amendment. We disagree with both arguments. A. The FCRA Is a Constitutionally Permissible Restriction on Speech The First Amendment states that ACongress shall make no law . . . abridging the freedom of speech . . . .@ The right to free speech, however, is not unfettered and it is well settled that different types of speech merit different levels of constitutional protection. Specifically, courts apply the highest degree of 54 We note here that our conclusions are consistent with the Commission=s Statements of General Policy or Interpretations under the Fair Credit Reporting Act, 16 C.F.R. Part 600, et seq., which offer general guidance on the FCRA and are not regulations and do not have the force of statutory provisions. 16 C.F.R. ' 600.2(a) Further, the Statements appear to be of marginal relevance to the issues here, as neither the parties nor the ALJ based their arguments, conclusions or findings on these Statements. VOLUME 129 Opinion of the Commission protection to speech related to issues of public concern such as political or social change or artistic or scientific expression. See, e.g., Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U.S. 749, 758-59 (1985) (plurality opinion). Such fully protected speech may be called Apure@ speech. American Future Systems, Inc. v. Pennsylvania State Univ., 752 F.2d 854, 861 (3rd Cir. 1984). By contrast, courts apply a reduced or intermediate level of protection to Acommercial@ speech - - speech, such as advertising, that is related to a commercial transaction. See, e.g., Central Hudson Gas & Electric Corp. v. Public Serv. Comm=n of New York, 447 U.S. 557, 561-63 (1980). Courts have also recognized that the First Amendment does not protect certain types of speech, such as obscenity and Afighting words,@ Dun & Bradstreet, 472 U.S. at 758-59, n.5, or conduct that does not constitute speech, Michael Barnes, et al. v. Glen Theatre, Inc., et al., 501 U.S. 560, 570 (1991). For the reasons discussed below, we find that Trans Union=s consumer reports are entitled to intermediate First Amendment protection. Accordingly, we analyze the FCRA under the standard established by the Supreme Court in Central Hudson and its progeny and conclude that the Act does not violate the First Amendment by prohibiting Trans Union from selling consumer reports to target marketers.

1. Type of Speech In order to apply the appropriate First Amendment analysis to Trans Union=s challenge to the FCRA, we must initially determine what type of expression or conduct the Act regulates in this case. a. Pure Speech Trans Union contends that its target marketing lists are pure speech and, as such, should receive the highest degree of constitutional protection. TUAB at 67--68. We are not persuaded, however, that Trans Union=s lists rise to the level of such fully protected pure speech. The Supreme Court has held that speech on public issues deserves the highest degree of protection because the First Amendment A>was fashioned to assure the unfettered interchange of ideas for the bringing about of political and social TRANS UNION CORPORATION 527 Opinion of the Commission changes desired by the people.=@ Dun & Bradstreet, 472 U.S. at 759, quoting Roth v. United States, 354 U.S. 476, 484 (1957). Speech related to matters of purely private concern, however, merits less First Amendment protection because regulation of such speech has less of an impact on the exchange of ideas on public issues. Id.

Here, the record clearly establishes that Trans Union=s target marketing lists do not concern the types of lofty or important public issues or themes traditionally recognized as central to the First Amendment=s guarantee of freedom of expression. See, e.g., McIntyre v. Ohio Elections Comm=n., 514 U.S. 334 (1995) (distribution of anonymous political campaign literature); City of Ladue v. Gilleo, 512 U.S. 43 (1994) (display of sign opposing Persian Gulf War); United States v. Eichman, 496 U.S. 310 (1990) (burning American flag); Hustler Magazine, Inc. v. Falwell, 485 U.S. 46 (1988) (vulgar parody of public figure); New York Times Co. v. Sullivan, 376 U.S. 254 (1964) (allegedly libelous newspaper editorial advertisement about public official and civil rights movement); Kingsley International Pictures Corp. v. Regents of the University of the State of New York, 360 U.S. 684 (1959) (exhibition of film depicting and expressing approval of adultery). Rather, the lists concern private information about individual consumers= credit history and other confidential, personal financial data. Because the lists do not possess the type of public component that compels full First Amendment protection, we conclude that Trans Union=s lists are not pure speech and, consequently, we do not apply the strict scrutiny analysis to the FCRA=s restriction on the dissemination of these lists.

b. Nonspeech We also reject Complaint Counsel=s position that the lists are not speech at all and thus fall outside the scope of First Amendment protection. Complaint Counsel argues that Trans VOLUME 129 Opinion of the Commission Union=s lists are not expression but, rather, are simply Acommercial products@ that Trans Union sells to its customers. CCAB at 58-6155. Although Trans Union=s lists are products, we find that they do possess a quality of speech because they communicate substantial consumer information to Trans Union=s target marketer customers. In other words, the lists Trans Union sells to its clients are more than simply a collection of names and addresses. Instead, these lists reflect Trans Union=s complex analysis and qualitative judgment regarding which consumers meet various credit and financial-related criteria. Moreover, although courts have accorded them varying levels of protection, they have also treated consumer reports as speech56. Indeed, by questioning the application of the First Amendment here, the court of appeals in this case has necessarily assumed that a consumer report is some form of speech. See Trans Union, 81 F.3d at 235.

c. Commercial Speech The ALJ held that Trans Union=s lists constitute commercial speech and, as such, applied intermediate constitutional scrutiny to the FCRA. ID at 88-89. We find that, although the target marketing lists do not possess all the elements typically associated with commercial speech, the lists have sufficient commercial speech qualities (without rising to the level of fully protected pure 55 Of course, both Trans Union and Complaint Counsel also assert in the alternative that the target marketing lists are commercial speech and that we should therefore apply intermediate level scrutiny to the FCRA. We analyze commercial speech infra.

56 See Dun & Bradstreet, Inc. v. Greenmoss Builders, 472 U.S. 749, 762 (1985) (credit report Awas speech solely in the individual interest of the speaker and its specific business audience@); Millstone v. O=Hanlon Reports, Inc., 528 F.2d 829, 833 (8th Cir. 1976) (treating consumer credit reports as commercial speech); U.D. Registry, Inc. v. State, 40 Cal.Rptr.2d 228, 233 (Cal. 1995) (much of the information contained in the credit reports is Ahighly protected noncommercial speech@); Equifax v. Cohen, 420 A.2d 189 (Me. 1980) (rejecting appeal of lower court determination that credit reports are commercial speech), cert. denied, 450 U.S. 916 (1981). TRANS UNION CORPORATION 529 Opinion of the Commission speech) to warrant intermediate First Amendment Protection. Our conclusion is supported by the full record here as well as Supreme Court precedent.

The Supreme Court has defined the Acore notion of commercial speech@ as an expression that does no more than propose a commercial transaction. Bolger v. Young Drug Products Corp., 463 U.S. 60, 66 (1983) citing Virginia Pharmacy Bd. v. Virginia Consumer Council, 425 U.S. 748, 762 (1976). Recognizing that the line between commercial and other types of speech is not always distinct, the Court expanded upon the concept by identifying three factors relevant to the determination of whether speech is Acommercial@: (1) whether the speech is an advertisement; (2) whether it mentions a specific product by name; and (3) whether it is economically motivated. Bolger, 463 U.S. at 66-67.

Trans Union=s lists do not fall neatly into this core notion of commercial speech as articulated by the Court in Bolger. The lists are not advertisements but instead are antecedent to advertisements - - i.e., the solicitations that Trans Union=s target marketing customers send to the consumers identified in the target marketing lists. Trans Union=s lists also do not mention a product by name; instead, as asserted by Complaint Counsel, they are the actual product. Finally, although Trans Union=s marketing list business is certainly motivated by economic considerations, that fact alone does not confer commercial speech status. See Bolger, 463 U.S. at 67 (fact that party had an economic motivation for mailing pamphlets at issue was insufficient by itself to turn materials into commercial speech).

Still, Bolger does not establish a bright line test for commercial speech and the Supreme Court has also regarded Aexpression related solely to the economic interests of the speaker and its audience@ as commercial speech. Central Hudson, 447 U.S. at 561. Similarly, Trans Union creates and sells its lists for VOLUME 129 Opinion of the Commission its own economic benefit as well as the benefit of its target marketing customers. In other words, while the ultimate consumers who are the subject of Trans Union=s lists have an interest in protecting their credit and financial privacy, Trans Union=s sale of its target marketing lists is a commercial transaction motivated by the economic interests of the list seller and the list purchaser.

Moreover, in Dun & Bradstreet, which concerned a consumer report containing false information, a plurality of Justices found that the consumer report at issue deserved reduced First Amendment protection. The plurality opinion explained that speech related strictly to private concerns has less First Amendment value and merits less stringent protection than speech on matters of public concern. Thus, the level of protection the Court should give to a consumer report turned on whether the report concerned public or private matters. The plurality concluded that, based upon an examination of the content, form and context of the report, it involved Aspeech solely in the individual interest of the speaker and its specific business audience@ and deserved reduced First Amendment protection. Dun & Bradstreet, 472 U.S. at 762 (plurality).57 Even though Trans Union=s lists do not embody all of the characteristics of core commercial speech outlined by the Court in Bolger, the lists concern private matters primarily concerning the economic interests of the speaker and its specific business audience. As such, we find that the best fit here is to grant Trans Union=s lists the same degree of First Amendment protection 57 Although Trans Union correctly notes that Dun & Bradstreet, unlike the instant matter, concerned the distribution of a false credit report that injured the reputation of the report=s subject, the falsity of the credit report was only one of the several considerations that led the Court to conclude that its distribution was not entitled to full First Amendment protection. The fact that the consumer report was of limited distribution and, like advertising, was hardy and unlikely to be deterred by incidental state regulation supported this conclusion. Id. at 762.

TRANS UNION CORPORATION 531 Opinion of the Commission accorded to commercial speech58. Consequently, we apply intermediate constitutional scrutiny to the FCRA=s restriction of Trans Union=s sale of its target marketing lists.59 2. The FCRA Passes Intermediate Constitutional Scrutiny The Central Hudson case and its progeny set forth the analysis appropriate for intermediate level scrutiny in a First Amendment context. Under this test, a court must examine the following: (1) whether the expression at issue concerns lawful activity and is not misleading; (2) whether the asserted governmental interest supporting the restriction is substantial; (3) whether the regulation directly and materially advances the governmental interest asserted; and (4) whether the regulation is narrowly drawn to advance the government interest. Central Hudson, 447 U.S. at 566. Because it is undisputed that the expression at issue here 58 Our conclusion is supported by a recent ruling of the U.S. Court of Appeals for the Ninth Circuit in which the court held that a private party=s sale of the names and addresses of arrestees was a Apure economic transaction@ that constituted commercial speech entitled to intermediate First Amendment protection. United Reporting Pub. Corp. v. California Highway Patrol, 146 F.3d 1133 (9th Cir. 1998), rev=d on other grounds, Los Angeles Police Dept. v. United Reporting Pub. Corp., 120 S. Ct. 483 (1999) (Court reversed facial invalidation of the statute and did not reach the issue of whether the information constituted commercial speech); see also Lanphere & Urbaniak v. State of Colorado, 21 F.3d 1508, 1513 (10th Cir. 1994) (state statute restricting release, for commercial use, of criminal justice records containing personal information was subject to intermediate First Amendment scrutiny). 59 Because the precise nature of Trans Union=s lists was unclear, in an abundance of caution, the Commission formerly applied both the commercial speech and fully protected pure speech analysis when it first examined Trans Union=s target marketing lists. See In re: Trans Union, 118 F.T.C. at 881-89. The Commission ruled that the FCRA did not violate the First Amendment under either standard. Our review of this matter is de novo and, based on our evaluation of a full record that was not previously before the Commission, we have determined that Trans Union=s lists are not fully protected speech and thus we decline to apply strict constitutional scrutiny to the FCRA. VOLUME 129 Opinion of the Commission concerns truthful, non-misleading factual information, we will focus our attention on the other three prongs of the Central Hudson test.

a. The Government Has a Substantial Interest in Protecting the Privacy of Consumers= Personal Credit Information The FCRA and its legislative history indicate that the government=s interest in restricting CRAs= dissemination of consumer reports is to protect consumers= privacy of their personal credit information. Congress expressly found that: A[t]here is a need to insure that consumer reporting agencies exercise their grave responsibilities with fairness, impartiality, and a respect for the consumer=s right to privacy.@ 15 U.S.C. ' 1681(a)(4)(emphasis added). Congress based this finding on the record at the time of the Act=s genesis, which demonstrated significant concerns in the area of consumer privacy. The record included, for instance, both media accounts as well as examples cited by Columbia University Professor Alan Westin regarding CRAs= disclosure of personal information for non-credit related purposes. (Reidenberg 961/22--963/19).60 60 During his testimony at Congressional hearings on the bill that became the FCRA, Professor Westin provided various examples of non-credit grantors easily obtaining consumer credit files. These examples include Professor Westin=s own success in securing, without any credit-related purpose, a co-worker=s consumer report as well as the ability of police agencies and federal investigators to obtain and use consumer reports in connection with non-credit related investigations. Fair Credit Reporting: Hearings on S. 823 Before the Subcomm. on Financial Institutions of the Senate Comm. on Banking and Currency, 91st Cong. 73-97, pp. 92-93 (1969) (testimony of Alan Westin).

TRANS UNION CORPORATION 533 Opinion of the Commission Senator Proxmire, in introducing the original legislation, stated that his bill Aseeks to prevent an undue invasion of the individual=s right to privacy in the collection and dissemination of credit information.@61 He also noted that A[t]he consumer has . . . a right to see that the information is kept confidential and . . . he has a right to be free from unwarranted invasions of his personal privacy. The Fair Credit Reporting Act seeks to secure these rights.@62 In light of these concerns, Congress drafted the Act to limit CRAs= disclosure of credit reports to people with a Apermissible purpose.@ Congressional interest in protecting consumers= privacy is further illustrated by the 1996 amendments to the FCRA, in which Congress added to the permissible purposes of consumer reports 61 Statement of Senator Proxmire, Committee on Banking and Currency, Subcommittee on Financial Institutions, S. REP. No. 91-517 (November 5, 1969). Further, A[a] fourth problem is that the information in a person=s credit file is not always kept strictly confidential.@ S. REP. No. 91-517, at 4 (1969). 62 Fair Credit Reporting: Hearings on S. 823 Before the Subcomm. on Financial Institutions of the Senate Comm. on Banking and Currency, 91st Cong. 2 (1969) (statement of Senator Proxmire). In addition to the interest in protecting the confidentiality and privacy of personal credit information, the FCRA=s legislative history also makes clear the importance - - to consumers and CRAs alike - - of the free flow of accurate and reliable consumer credit information between consumers and credit grantors. The Committee Report cites the Avital role@ of CRAs in our economy and states that credit grantors have the right to the facts necessary to make sound decisions on whether to grant credit. S. REP. No. 91-517, at 2 (1969). The report also stresses that consumers have the right to correct erroneous information in their credit files. Id. Professor Westin also referenced the importance of accurate credit information, stating that CRAs exist to help credit grantors avoid loss through fraud or misuse of credit and to keep the costs of such losses from falling on the average consumer. Fair Credit Reporting: Hearings on S. 823 Before the Subcomm. on Financial Institutions of the Senate Comm. on Banking and Currency, 91st Cong. 73-97, p.91 (1969) (testimony of Alan Westin). Consumers benefit from strong credit markets, which in turn require accurate, current and reliable data.

VOLUME 129 Opinion of the Commission prescreening for certain defined firm offers of credit and insurance. The Committee Report to the amendments notes an effort Ato balance any privacy concerns created by prescreening with the benefit of a firm offer of credit or insurance for all consumers who meet the criteria for the credit or insurance being offered.@63 In striking this balance, however, Congress ensured significant privacy protections for consumers, requiring that they receive notice that their personal credit information is being used for such purposes, and that they have the right to Aopt out@ of such use. See 15 U.S.C. ' ' 1681b(c), (e) and 1681m(d)(1). Trans Union=s practices at issue here do not provide for such safeguards. Courts have also recognized that privacy protection of creditrelated data is among the important purposes of the FCRA. The court of appeals in this matter found that Aa major purpose of the Act is the privacy of a consumer=s credit-related data.@ Trans Union, 81 F.3d at 234. See also St. Paul Guardian Ins. Co. v. Johnson, 884 F.2d 881, 884 (5th Cir. 1989); Zamora v. Valley Fed. Sav. & Loan Ass=n, 811 F.2d 1368, 1370 (10th Cir. 1987); Heath v. Credit Bureau of Sheridan Inc., 618 F.2d 693, 696 (10th Cir. 1980).

Although enacted congressional policy does not necessarily constitute substantial governmental interest for purposes of the Central Hudson analysis, Greater New Orleans Broadcasting Ass=n, Inc. v. United States, 119 S. Ct. 1923, 1932 (1999), we are satisfied that the interest here is sufficient. First, the FCRA=s legislative history is consistent with other congressional enactments related to personal privacy and the concerns raised by compilations of personal information in large databases. In United States Dep=t of Justice v. Reporters Committee for Freedom of the Press, 489 U.S. 749, 762-67 (1989), the Supreme Court reviewed in detail the terms and history of the Privacy Act, 5 U.S.C. ' 552a, as well as the privacy exemptions of the Freedom of Information Act (AFOIA@), 5 U.S.C. ' 552. The Court recognized, for 63 Economic Growth and Regulatory Paperwork Reduction Act of 1995 (12/14/95), S. REP. No. 104-185, at 36 (1995). TRANS UNION CORPORATION 535 Opinion of the Commission example, that A[t]he Privacy Act was passed largely out of concern over >the impact of computer data banks on individual privacy.=@ Id. at 766 (quoting H.R. REP. No. 93-1416, at p. 7 (1974)). Additionally, the Court concluded that the essence of the Aprivacy@ interest Congress sought to protect under the FOIA was the individual=s Acontrol of information concerning his or her person,@ by deciding for him or herself the Adegree of dissemination@ of personal information. Id. Such Congressional consistency supports our conclusion that the government=s interest is substantial.

Furthermore, case law indicates well-settled privacy interests in personal information generally,64 and financial and credit information in particular. See, e.g., Barry v. City of New York, 712 F.2d 1554, 1561 (2d Cir.) (A[w]e recognize that public disclosure of financial information may be personally embarrassing and highly intrusive.@) cert. denied, 464 U.S. 1017 (1983); Millstone, 528 F.2d at 833 (recognizing, in a FCRA case, that the right to privacy is Aa significant personal right@). But see U.S. West, Inc. v. Federal Communication Commission, 182 F.3d 1224, 1228 (10th Cir. 1999) (vacating a Federal Communication Commission (AFCC@) regulation on First Amendment grounds).65 64 See Edenfield v. Fane, 507 U.S. 761, 769 (1993) (for purposes of Central Hudson analysis, state has substantial interest in protecting privacy of potential clients of certified public accountants); Whalen v. Roe, 429 U.S. 589, 599 (1976) (although the Supreme Court upheld a statute authorizing New York state to record names and addresses of consumers receiving prescriptions for certain drugs, the Court acknowledged that individuals have a protectable Ainterest in avoiding disclosure of personal matters@). 65 U.S. West involved review of an FCC regulation implementing a section of the Telecommunications Act of 1996. The petitioner argued that the regulation violated the First Amendment by restricting its ability to engage in commercial speech with its customers. A majority of the panel applied a Central Hudson analysis and expressed Areservations@ about whether the FCC had Aasserted a substantial state interest in protecting people from the disclosure of sensitive and potentially embarrassing personal information.@ U.S. West, 182 F.3d at 1235-36. The majority=s skepticism was based upon its VOLUME 129 Opinion of the Commission Trans Union argues that any privacy interest in the existence of a consumer=s credit relationship is de minimis because only 1% of consumer files in CRONUS do not have a tradeline. TUAB 39- -40, 72. (Stockdale 906/1-8, 21-23, 904/15-18; CX-358-G). This argument is not compelling. First, the mere fact that 99% of the consumer records in CRONUS have at least one tradeline is not indicative of whether there is a privacy interest worthy of FCRA protection66. Section 603(d) of the Act focuses on the nature of the information disclosed and not the amount of information worthy of protection. Second, Trans Union itself asserts that CRONUS data relate to a subset of people. The fact that this number is clearly less than the total number of adults in the United States, demonstrates just how effective tradeline information is in restricting to credit worthy individuals the pool of consumers eligible to be included in Trans Union=s target marketing lists.

Also, we believe that Trans Union=s argument ignores the full range of CRONUS information that Trans Union actually discloses or has disclosed about individuals. As discussed, Trans Union not only discloses information about the existence of a consumer=s credit relationship, but also open dates, credit limits, number of tradelines, type of tradelines, among other information. Finally, Trans Union asserts that consumers= routine disclosure of credit relationships, through the use of credit cards or mortgage applications that appear on the public record, concern about the lack of evidence that carriers such as U.S. West would actually disclose the regulated information to outside parties and thereby breach consumers= privacy. The case before us is distinguishable from U.S. West as the facts here differ significantly. It is undisputed that Trans Union discloses consumer information to third party target marketers; indeed, the record demonstrates that Trans Union actively promotes the value of its lists to third party purchasers.

66 Moreover, as a practical matter, see supra p. 24, Trans Union=s customers do not purchase lists of people with one tradeline. TRANS UNION CORPORATION 537 Opinion of the Commission demonstrates that consumers do not view credit relationships as private. TUAB at 73. The examples Trans Union cites, however, involve situations where the consumer knowingly relinquishes his or her privacy in return for a direct and known benefit that is also sought by the consumer. In those cases, the consumers were exercising their right to control the dissemination of their own personal information. See Reporters Committee, 489 U.S. at 763. By contrast, Trans Union=s disclosure of consumers= personal credit information - - including the fact that a consumer has a recently used credit account that is not the subject of a credit dispute - - shares neither of these important attributes. In fact, such disclosures are made without the consumer=s knowledge. For all these reasons, we find that the government has a substantial interest in protecting the privacy of consumers= personal credit information, and we reject Trans Union=s arguments to the contrary.

b. The Restriction Directly and Materially Advances the Government=s Interest The next question in the Central Hudson analysis is whether the FCRA=s speech restriction directly and materially advances the government=s interest. To meet this burden, the government may not rely on Amere speculation or conjecture@ but must instead demonstrate that the restriction at issue will alleviate real harms to a Amaterial degree.@ Edenfield, 507 U.S. at 770-71. The Supreme Court has struck down regulations of commercial speech where the government failed to offer sufficient evidence that the restriction at issue would advance its interests. See 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484, 505 (1996) (no Afindings of fact, or indeed any evidentiary support whatsoever@); Edenfield, 507 U.S. at 771 (no studies or anecdotal evidence presented); see also Rubin, 514 U.S. at 490 (1994) (government submits only Aanecdotal evidence and educated guesses@); cf. Edge Broadcasting, 509 U.S. at 434 (upholding federal ban on lottery VOLUME 129 Opinion of the Commission advertising based solely on Acommon sense judgment@ that the ban would advance governmental interest in supporting the state=s anti-gambling policies).

We rely upon substantial record evidence rather than mere conjecture or speculation to conclude that the FCRA, by limiting CRAs= disclosure of personal credit information, directly and materially advances the substantial government interest in protecting the privacy of personal credit information. Indeed, it is almost tautological - - because the unauthorized disclosure of personal credit information causes the privacy harm, restricting the unauthorized disclosure of the information directly limits the infringement on privacy.

In attacking the FCRA, Trans Union argues that the statute is underinclusive because it elsewhere allows practices that undermine the consumer privacy interest in the information that Trans Union=s target marketing lists communicate67. TUAB at 76- -82. The court of appeals expressed an underinclusiveness concern as well, remarking on the apparent freedom of Trans Union=s non-CRA competitors to gather and distribute the same information that Trans Union discloses in its marketing list business68.

67 Trans Union cites United Reporting, 146 F.3d at 1135, as an example of a statute struck down on First Amendment grounds. As noted, supra n.58, the Supreme Court recently reversed this decision. Furthermore, the FCRA is factually different. The statute at issue in United Reporting prohibited the release of arrest information for commercial purposes, but permitted it for Ajournalistic, scholarly, political, governmental, or investigative purposes.@ Prior to reversal, the Ninth Circuit had found that the governmental interest in protecting the privacy of arrestees was substantial but that the exceptions to the statute - - which include the right to broadly publish this information - precluded it from advancing the privacy interest in a direct and material way. Unlike the regulation at issue in United Reporting, however, in this case none of the FCRA=s permissible purposes allows broad public disclosure of consumer report information.

68 In particular, the court criticized the position that Trans Union could Aseparately obtain@ and distribute consumer information - - i.e., gather the information at issue from sources other than its credit reporting database - without violating the FCRA. Such a requirement, the court suggested, would TRANS UNION CORPORATION 539 Opinion of the Commission As a general rule, however, a regulation=s underinclusiveness is fatal only where it is material, substantial or significant. Bad Frog Brewery v. New York State Liquor Auth., 134 F.3d 87, 98-99 (2d Cir. 1998). The Supreme Court has offered the following guidance:

ANor do we require that the Government make progress on every front before it can make progress on any front . . . [T]he Government may be said to advance its purpose by substantially reducing [the proscribed conduct], even where it is not wholly eradicated.@ Edge Broadcasting, 509 U.S. at 434; See also R.A.V., 505 U.S. at 387, Moser v. F.C.C., 46 F.3d 970, 974 (9th Cir. 1995). Based on the full record developed on remand from the court of appeals, as well as the above-cited cases, we conclude that the FCRA=s restrictions are not materially, substantially, or significantly underinclusive. Instead, we find that any disparity between Trans Union and its non-CRA competitors is reasonable given Trans Union=s position as a CRA and the nature of the information it discloses in its target marketing products. In addition, neither the disclosure of information by credit grantors, nor the practice of prescreening, significantly undermines the Act=s protection of privacy. Furthermore, the FCRA=s restrictions on the dissemination of private, credit-related information are not, in result in a waste of time and resources. Although it did not rule on the issue, the court indicated that the Adisparity@ between Trans Union (as a CRA) and its competitors raised constitutional concerns. Trans Union, 81 F. 3d at 235. The court of appeals= concern presupposed that Trans Union, wholly independent of its status as a CRA, could gather for target marketing purposes the same type of rich consumer information that it gathers by way of its consumer reporting business. As explained infra, the record now before us clearly establishes that this is not the case; rather, the high quality and comprehensiveness of the underlying data in Trans Union=s target marketing products stem from its special position as a CRA.

VOLUME 129 Opinion of the Commission fact, restricted to CRAs. Section 1681e(e) imposes restrictions on the resale by a CRA=s customer of a credit report that are similar to the restrictions on the CRA itself. Therefore, the FCRA does substantially reduce the harm to consumers of intrusion on the privacy of their personal credit information. (i) The FCRA Regulates the Activities of Trans Union and Other CRAs Because They Have Access to Vastly Superior Information The fact that the FCRA applies to CRAs, but not to other target marketing or data compilation companies, does not render the Act constitutionally infirm. As the record in this case demonstrates, CRAs are able to quickly obtain a broad array of current, accurate, detailed and highly personal credit information about consumers. Balancing this unique ability with safeguards against abuses, the FCRA requires that CRAs disclose such information only to persons with a permissible purpose as set forth in Section 604 of the Act. A comparison of Trans Union=s operations to those of its non-CRA competitors demonstrates that Congress acted properly in treating CRAs differently than other information gatherers.

Trans Union=s primary non-CRA competitors are Polk, Metromail, First Data, and ACXIOM. These companies obtain most of their data from state departments of motor vehicles (ADMVs@), census data, telephone directory white pages, county registrar and tax assessor records, self-reported surveys, and product registration or warranty cards. Polk, First Data, and Metromail=s lists are compiled from two primary sources - - DMV data and white pages. (Cleary 3085/9-20, 3114/6-19; TU-119-3; Litz 2969/16--2970/4; TU-115-p.158; Nusbaum 2880/8-12). As a CRA, Trans Union=s data sources are far superior and, as a result, the information it obtains through its credit reporting business has considerable advantages over the information of its non-CRA competitors. The quality of Trans Union=s data is superior in terms of detail and accuracy as well as availability and comprehensiveness. Credit grantors and other information TRANS UNION CORPORATION 541 Opinion of the Commission providers are responsible for providing CRAs with accurate, complete and up-to-date information and/or providing supplemental information to correct errors69. Because Trans Union obtains its information from third parties, its information is also less biased and thus more reliable than the self-reported information many non-CRA information brokers receive. Trans Union=s own credit scoring witness testified that data from CRAs are objective and better predictors of future credit performance than information provided by a consumer filling out an application. (Coffman 3806/2-14, 3857/24--3858/4, 3858/17-- 3859/3; CX-122-P)70. Finally, because Trans Union=s information is reported and updated on an ongoing basis,71 it is far more current than reports by, for example, census bureaus and state DMVs.

By virtue of its status as a CRA, Trans Union also has the advantage of being able to provide an instant compilation of nearly all relevant information. Moreover, some of the more specialized information that Trans Union has access to and discloses in target marketing lists simply may not be available to other information brokers. Examples include the existence of 30/60/90 day finance trade; an upscale retail card; a student loan; a premium bank card; and the open dates of bank cards. (See e.g., TU-130-4; CCPF at 76; TU-117-2; TU-120-2; Schultea 3928/2- 4; CX-310-D). Although other types of information that Trans 69 Section 623 of the FCRA, 15 U.S.C. ' 1681s-2. 70 Trans Union=s competitors= modeled products are not as Apredictive@ as Trans Union=s own CRONUS-derived products. (Hinman 2270/13-- 2271/11).

71 Trans Union receives information regarding 1.8 billion tradelines per month. (Stockdale 908/1-19). Some customers report information on a daily basis to TU; the majority report on a monthly basis or according to their billing cycles. (Stockdale 904/8-14; Frank CX-186 at 19/3-7). VOLUME 129 Opinion of the Commission Union discloses may be separately available from a range of sources, only CRAs have instant access to them all. Recent legislation and case law have recognized and, indeed, expanded the disparity between Trans Union and its non-CRA competitors. As noted, Polk, Donnelly, and First Data use DMV data as a primary source of information and First Data also offers automobile data from state DMVs. However, the 1994 enactment of the Federal Drivers Privacy Protection Act (ADPPA@), 18 U.S.C. ' 2721, generally restricts state DMVs from disclosing, without the licensee=s permission, personal identifying information contained in state DMV records. Reversing the holding of the U.S. Court of Appeals for the Fourth Circuit, the Supreme Court in Reno v. Condon, 2000 WL 16317 (January 12, 2000), unanimously upheld the constitutionality of the DPPA. Thus, there is no question that the DPPA drastically limits the personal data that these information brokers can obtain72. Moreover, the DPPA aside, information brokers have never been able to obtain driver data on a nationwide basis. By contrast, Trans Union=s coverage provides continual access to current information on consumers= auto loans in all 50 states.73 Trans Union enjoys profound advantages with respect to other types of data as well. For instance, information brokers obtain consumer mortgage information from county records. (Litz 2972/6-9, 2975/5--2976/7; M. Smith 3373/18-23, 3390/18-- 3392/9; Nusbaum 2889/1-8, 2914/22--2915/3, 2933/22--2934/15; 72 The Court=s holding in Condon v. Reno, that Congress may regulate the sale or release of personal identifying information, implicitly supports the notion that individuals have a right to personal data privacy. The same notion is presented here through our determination that the government, through the FCRA, has a substantial interest in protecting the privacy of consumers= personal credit information. See supra pp. 37-41. 73 Trans Union understands its superiority as a source for consumer automobile data and has used this fact as a selling point. One of Trans Union=s promotional letters notes that its coverage for automobile loan information encompasses all states and is not limited by the commercial restrictions that some states have imposed upon access to similar information. (CX-66-A). TRANS UNION CORPORATION 543 Opinion of the Commission Cleary 3099/16-23; Hinman 2250/5-14). Such information gathering can be quite burdensome; in Texas alone, for example, information brokers may need to consult over 240 counties to fully cover the state=s mortgages74. Trans Union has the advantage of having national coverage of this information through its single database. This allows Trans Union to offer more current mortgage information than the public record information non-CRAs sell. It was therefore not surprising when Polk=s Vice President for Operational Planning & Analysis conceded that Polk=s consumer model assigns greater reliability weight to mortgage data coming from Trans Union, as compared to the data coming from county records. (Nusbaum 2888/2--2890/13, 2927/6B2928/11).75 Another source of information for Trans Union=s non-CRA information broker competitors is from consumer surveys and warranty cards where consumers are obviously under no duty to provide accurate or complete information. As a CRA, however, Trans Union must Aassure maximum possible accuracy@ of all the information it gathers and disseminates. Section 607(b), 15 U.S.C. ' 1681e(b). Trans Union has stated that self-reported data are Ainevitably biased@ (CX-115-Z-6) and has promoted its Master File as a unique source for individual-level observed behavior data - - Awithout equal@ that is Abased on actual behavior - - not self-reported or neighborhood values.@ (emphasis added) (CX- 74 Mr. Kenneth Scott, witness for Trans Union, described the difficulty in obtaining such data, noting that information brokers send their employees with laptop computers to county courthouses to input public data and modem it to the front office. (Scott 2659/10-14). 75 Even the legislative history reveals a concern that public information may not be as current as consumer reporting information. AUnfortunately, the [public record] information cannot always be kept up to date either because it is costly or because the correct information is simply not available.@ Report of the Committee on Banking and Currency, S. REP. 91-517, at 4 (1969). VOLUME 129 Opinion of the Commission 83-C)76. Consumer surveys and warranty cards are also weak with respect to coverage. The record shows that 20-40 million consumers respond to surveys or provide warranty cards. (Nusbaum 2879/6-23; Cleary 3088/20--3089/6). By contrast, Trans Union provides information on over 140 million people. Additionally, several information brokers use census data to estimate income. (CX-119-Z-7; Cleary 3123). These data are significantly less timely because they are only reported every ten years whereas Trans Union=s income estimator, TIE, is updated every 7 days. (Wiermanski 1723/10-24; CX-120-B). In light of the full record here, we find that Trans Union=s status as a CRA allows it to collect a much wider array of consumer information that is richer, more detailed and more current than the information available to its non-CRA competitors. Trans Union could not obtain the same type and quality of information outside the scope of its consumer reporting business. Accordingly, we are not persuaded that the FCRA is unconstitutionally underinclusive because it treats CRAs and non- CRAs differently.77 Our conclusion is consistent with relevant case law. The Supreme Court has recognized the special threats to privacy that compilations of information pose, even though each constituent bit of information may be publicly available elsewhere. In D.O.J. 76 Elizabeth Dixon, Account Manager at Performance Data, and Patricia Porretto, Senior Account Executive at Performance Data, testified to the fact that the non-self-reported aspect of the Master File is a distinct advantage over other target marketers. (Dixon 292/16--293/4; Porretto 1621/16-20). 77 Trans Union also differs from other CRAs in terms of the type of information it discloses. Although Experian and Equifax disclose credit information to target marketers - - either directly or through third parties - they do so only on an aggregated, zip-plus-four basis. Such aggregated credit information relates to the typical consumer in a geographic area. By contrast, Trans Union=s information concerns specific, identifiable individuals. Thus, it intrudes more acutely on individuals= privacy. In any event, the lawfulness of zip-plus-four aggregation is not an issue in this proceeding and we decline to rule on it here.

TRANS UNION CORPORATION 545 Opinion of the Commission v. Reporters Comm. for Freedom of the Press, 489 U.S. 749 (1989), the court considered whether a privacy-related Freedom of Information Act exemption applied to a request for a Arap sheet@ (a compiled database of publicly available information Abits@). The Court found:

[T]he issue here is whether the compilation of otherwise hardto-obtain information alters the privacy interest implicated by disclosure of that information. Plainly there is a vast difference between the public records that might be found after a diligent search of courthouse files, county archives, and local police stations throughout the country and a computerized summary located in a single clearinghouse of information.

Id. at 764. The Court acknowledged the A. . . power of compilations to affect personal privacy that outstrips the combined power of the bits of information contained within.@78 Id. at 765.

(ii) Credit Grantors Do Not Disclose the Same Information As CRAs To further support its underinclusiveness argument, Trans Union points to the Atransactions or experiences@ exception in Section 603(d)(2)(A)(i) and asserts that this provision undermines 78 Trans Union argues that its witness, Kenneth Scott, testified that all of the selects available from Trans Union were also available from others in the marketplace. (Scott 2711--2730). We are unpersuaded. In reaching his conclusion, we find that Mr. Scott performed only a superficial analysis. Mr. Scott examined only what is currently available - - not Trans Union=s past practices. Importantly, Mr. Scott did not examine the difference in the source of the information for each select. Thus, he did not take into account that Trans Union, using CRONUS as its primary source of information, uniquely and instantly has a full range of richer and more comprehensive information available to it.

VOLUME 129 Opinion of the Commission privacy protection under the Act because it allows credit grantors to disclose, for target marketing purposes, substantial information about their own customers79. We disagree. Trans Union introduced evidence demonstrating that Wachovia, First Card, Discover, First USA Bank, and American Express provide credit information on their credit card customers for target marketing. (Eulie 2376B78, 2381; Koppin 588B589; Stormoen 3165B66; Scott 2614--2622, 2628--2630). The information available to a single credit grantor, however, is far more limited than Trans Union=s CRONUS database, which compiles information from hundreds of creditors. Also, the record shows that credit grantors generally do not disclose particulars about credit accounts beyond the name and address of account holders. They do not, for example, disclose open dates or credit limits. (Pendleton 405/15B406/3; Koppin 588/6B589/6, 596/24B597/18; Stormoen 3165/3B3168/4; Eulie 2376/7B2377/23, 2380/4-10). Further, the ability of creditors and other merchants to collect customer information and disclose it may be limited by state law. New Jersey, Pennsylvania and Massachusetts, for example, prohibit merchants from collecting personal identification information that is not required for the transaction80. Other states prohibit the disclosure of information by merchants and financial institutions81. Because it is a CRA, Trans Union is not subject to these restrictions.

79 Under this provision, information related solely Ato transactions or experiences between the consumer and the person making the report@ is not a consumer report. Section 603(d)(2)(A)(i) of the FCRA, 15 U.S.C. ' 1681a. 80 See, N.J. STAT. ANN. ' 56:11-17; PA. STAT. ANN. tit. 69, ' 2602; MASS. ANN. LAWS ch. 93, ' 105.

81 California and Virginia prohibit credit card issuers and merchants respectively from selling personal information to third parties without notifying the individual and providing him/her with the ability to opt out; New Jersey prohibits the disclosure of electronic fund transfer transactions for marketing purposes without consent. See CAL. CIV. CODE ' 1748.12; N.J. STAT. ANN. ' 17:16K-3; VA. CODE ANN.' 59.1-442. Connecticut prohibits financial institutions from selling the names of card holders or disclosing financial records without written consent. See CONN. GEN. STAT. Ann. ' 42-133gg; CONN. GEN. STAT. ' 36a-42 (1997). TRANS UNION CORPORATION 547 Opinion of the Commission Trans Union also cites the National Marketing File (ANMF@) created by Business Development Inc. - - a target marketing list provider - - as additional evidence that the FCRA=s Acredit grantor@ exception can be used to intrude upon a consumer=s financial privacy. This argument fails because NMF information is far less complete, detailed or timely than the data in CRONUS.82 Similarly, the existence of cooperatives that share data and reveal consumer credit information also fails to support Trans Union=s underinclusiveness argument. One such cooperative, Abacus, collects information regarding consumers= mail order buying behavior from 700 catalogers and shares the information among its members. Abacus discloses how many credit transactions a consumer has had over his/her lifetime, how much money a consumer has spent using credit cards over the last 12 months, and other information about a person=s history of buying by credit card. (TU-206). Like NMF=s information, however, this type of data also does not share the same level of 82 The NMF discloses two types of information in two Afields@: (1) a counter field that shows the number of credit cards a consumer has up to nine, and (2) a bank card field that indicates whether there is one bank card or not. (Schultea 3911/3B3913/7, 3943/6-23). The NMF is not kept current and a record is deleted only when the person moves or dies. (Schultea 3912/4-12, 3918/24--3919/7, 3920/2-11). The NMF does not reveal whether the person has obtained or used credit within a specified time period; instead, it reveals only whether a consumer has ever had a bank card and how many, up to nine, accounts the consumer has ever had. Mr. Schultea, President of Business Development, testified that in all likelihood, the NMF contains references to credit card accounts that have been closed and/or are currently inactive. (Schultea 3922/3-14). The NMF also does not indicate open dates of any of the accounts, or the type of credit a consumer has obtained (other than the existence of a bank card). (TU-130 p. 4; TU-117 p. 2; TU-120 p. 2; Schultea 3928/2-4; CX-310-D). Finally, as for coverage, the information in the NMF comes from only 20-25 retail companies and bank clients with credit card customers. (Schultea 3915/2B3916/21).

VOLUME 129 Opinion of the Commission comprehensiveness as the Trans Union information. Moreover, several of the state laws previously described may limit the disclosure of personal credit information to such cooperatives.83 (iii)Prescreening Finally, Trans Union argues that the FCRA is underinclusive because the Act, as amended in 1996, allows the practice of prescreening for so-called Afirm offers@ of credit and insurance. We find that Congress= decision to allow prescreening does not constitutionally undermine the FCRA. Any privacy intrusions that result from prescreening are significantly less harmful than the privacy intrusion at issue in Trans Union=s target marketing business. In prescreening, the types of consumer report information that can be used are restricted - - Section 604(c)(2) - and prescreening itself may provide a concrete benefit to consumers, i.e., a Afirm offer of credit,@ that they might not otherwise have. Equally important is the FCRA=s requirement that those seeking to use the prescreening mechanisms notify consumers that they may opt-out of future, prescreened solicitations. 15 U.S.C. '' 1681b(c) and 1681m(d). While a central concern of legislators at the time the FCRA was enacted was to protect the privacy of consumers= personal credit information, a related concern was to limit the disclosure of consumers= credit information. This latter concern and the means to address it were raised throughout the 1969 legislative hearings. 83 We understand that substantial development of broad-based information-sharing agreements, in the presence of an ever-growing electronic information-handling medium, may advance the quality of information that such cooperatives offer. It is possible that over time, the disparity between CRAs and non-CRAs may narrow. If so, Congress may find it appropriate to respond to new threats to financial privacy with new legislation as they arise. Cf. Gramm-Leach-Bliley Act of 1999, Pub. L. 106-102 (limiting, inter alia, disclosure of nonpublic personal information to nonaffiliates). But that does not change the legal obligations the Act imposes upon CRAs like Trans Union as a result of their unique status and the benefits they receive. At this time, the range and detail of information provided by CRAs far surpasses that of other information brokers and supports the legislative scheme. TRANS UNION CORPORATION 549 Opinion of the Commission Senator Proxmire, in introducing the original legislation, noted that A[t]he consumer has . . . a right to see that the information [is] . . . used for the purposes for which it is collected . . .@84 Professor Alan Westin also testified that A[t]he central issue of privacy is the release of personal credit information to other than credit grantors.@85 Even a representative from the credit reporting industry testified that information gathered specifically for creditgranting purposes should not be made available for other purposes.86 Congress=s approach to prescreening, in particular its requirement of notice and opt out rights for the consumer, is consistent with the twin goals of protecting the privacy of consumers= personal credit information and ensuring that consumer credit information not be used for inappropriate purposes. Permitting the disclosure of certain consumer credit information for prescreening, as tailored by statutory limitations, does not undermine the FCRA.

84 Statement of Senator Proxmire, Fair Credit Reporting: Hearings on S. 823 Before the Subcomm. on Financial Institutions of the Senate Comm. on Banking and Currency, 91st Cong. 92 (1969). He also stated that A[a] second aspect to the problem of confidentiality is the use of information inconsistent with the purposes indicated when the information was collected.@ 115 CONG. REC. S2340, 2410-16 (1969).

85 Fair Credit Reporting: Hearings on S. 823 Before the Subcomm. on Financial Institutions of the Senate Comm. on Banking and Currency, 91st Cong. 73-97, at 92 (1969) (testimony of Alan Westin). 86 Fair Credit Reporting: Hearings on S. 823 Before the Subcomm. on Financial Institutions of the Senate Comm. on Banking and Currency, 91st Cong. 224, 228 (1969) (statement of Harry C. Jordan, Chairman of the Board, Credit Data Corp., A[Credit Data=s] rules can be stated as follows: . . . Credit information is available to credit grantors only for credit purposes. . . Credit Data, as a matter of policy sells information only to credit grantors.@). VOLUME 129 Opinion of the Commission Consequently we find that the FCRA is not underinclusive. Congress recognized that CRAs like Trans Union, by virtue of their credit reporting business, are uniquely positioned to obtain the most up to date, detailed and comprehensive set of personal credit information on an individual, observed basis. Mishandling such information poses a special threat to consumer privacy. Therefore, by limiting the disclosure of this information, the FCRA directly and materially advances the government=s interest in protecting the privacy of personal credit-related information. To find otherwise would allow Trans Union to have it both ways - - i.e., to enjoy unique access to the widest array of the best consumer credit information available, without following the restrictions Congress imposed in order to protect consumer privacy. Further, the fact that the FCRA allows credit grantors to disclose limited information and permits prescreening does not make the Act underinclusive.

c. The Restriction Is Narrowly Tailored The final prong of the Central Hudson test requires a reasonable fit between the goals of the statute and the statute itself. The proper fit Ais not necessarily perfect, but reasonable@; it must Arepresent[] not necessarily the single best disposition but one whose scope is >in proportion to the interest served.=@ Board of Trustees of the State University of New York v. Fox, 492 U.S. 469, 480 (1989). With these words in mind, it is appropriate to consider whether there are alternative means of accomplishing the government=s stated interest with a lesser intrusion into speech. In doing so, we recognize that a commercial speech restriction may be unconstitutional if there is a Afar less restrictive@ alternative. Pearson v. Shalala, 164 F.3d 650, 658 (D.C. Cir. 1999). We believe that the FCRA=s restriction on the disclosure of consumer reports, including Trans Union=s target marketing lists, is narrowly tailored to protect the privacy of consumers= personal credit and other financial information. Moreover, we do not believe that alternative restrictions proposed by Trans Union are Afar less restrictive,@ nor would they afford sufficient privacy protection.

TRANS UNION CORPORATION 551 Opinion of the Commission Congress established a three-tiered system for disclosure and privacy protection:87  No Consumer Permission Required. Where a consumer has initiated a transaction involving credit, employment, or insurance, for example, CRAs may provide a consumer report for purposes of that specific transaction without the consumer=s permission.

 Notice and Opt Out. Where a consumer has not initiated such a transaction, but where a creditor or insurance company seeks to make a Afirm offer of credit or insurance@ (i.e., prescreening), a CRA may provide certain consumer report information as long as the consumer is provided notice that his or her name was provided by a CRA and the opportunity to opt out of appearing on such lists in the future, i.e., notice and opt out rights.  Opt In. Where a consumer has not initiated a transaction, and where the purpose of the credit report is not for a permissible purpose under the Act, a consumer report may only be disclosed with a consumer=s express consent, i.e., an Aopt in@ system.

This three-tier scheme is sufficiently tailored to achieve Congress=s goal. Congress=s determination that consumers would not be adequately protected from privacy intrusions by target marketers through a Anotice and opt out@ system is reasonable. 87 Congress has used a Atiered@ privacy-protecting approach in other areas as well. In the Cable Communications Policy Act of 1984, 47 U.S.C. ' 551, the Electronic Communications Privacy Act of 1986, 47 U.S.C. ' 222(c), and the Video Privacy Protection Act of 1988, 18 U.S.C. ' 2710, Congress afforded different levels of privacy protection - - from disclosure without consumer=s permission, to notice and opt out, to opt in - - as the uses of such information become less related to the purposes for which the information was collected.

VOLUME 129 Opinion of the Commission Specifically, Congress=s decision to favor the more privacy protective Aopt in@ is a sound system in light of documented problems of credit reports being widely disclosed for purposes unrelated to credit.

Trans Union contends that a simple opt out procedure would adequately protect consumer privacy without compromising Trans Union=s speech. TUAB at 82--85. However, this proposal is untenable and is based upon a misstatement of the Equifax/Harris findings. According to Trans Union, the 1996 Equifax/Harris survey revealed that 80% of consumers surveyed who object to the use of credit reporting information, change their minds when they are told of the right to opt out. (TU-88; Beales 3656--3665). Those survey findings, however, examined consumer attitudes about the use of credit reporting information to provide preapproved offers of automobile insurance or life insurance. The survey did not examine consumer attitudes toward the use of such information for target marketing (at issue here).88 Additionally, to the extent that Trans Union contends that opt out rights alone would adequately protect consumers in the target marketing context, Trans Union ignores the fact that, in the prescreening context, Congress authorized a notice and opt out system. The notice segment of the system is essential because it 88 Survey evidence introduced by Complaint Counsel indicates that consumers view credit relationships as private and that they experience a privacy invasion from the disclosure of the existence of types of credit accounts. (See CX-274; Mazis 1109/20-25). The survey, conducted by Dr. Michael Mazis, assessed the attitudes of 1,002 consumers regarding the use of information derived from CRONUS and from credit reports to compile marketing lists. (Mazis 1080/10-18; CX-354-A; Waldeck 1060/12-16). A total of 68.1% of the respondents found the use of credit report information for the compilation of marketing lists to be unacceptable. (Mazis 1105/13-20; CX-354-B). Based on these results, Complaint Counsel=s expert opined that consumers have a strong privacy interest in the use of information from their credit reports. (Mazis 1107/23-25).

TRANS UNION CORPORATION 553 Opinion of the Commission provides consumers the information to allow them to exercise opt out rights89.

Further, Trans Union argues that S. 650, 104th Cong. (1995) (a bill that was not enacted) could serve as a less restrictive alternative to the FCRA. (TU-214). Before a less restrictive alternative can be considered, however, it is necessary to determine whether the alternate approach furthers the government=s interest to the same extent as the FCRA. Because S. 650 would have provided considerably less data privacy,90 we cannot conclude that Congress acted unreasonably or disproportionately in the balance it struck between consumer privacy and commercial speech. Accordingly, the question of whether S. 650 would have had less of an impact on speech than the FCRA is irrelevant.

Based on this analysis, it is evident that the FCRA=s restriction on Trans Union=s target marketing lists is sufficiently narrowly tailored to achieve the goal of protecting the privacy of consumers= personal credit information. We therefore conclude 89 Indeed, although the Direct Marketing Association offers consumers the right to opt out of appearing on its members= marketing lists, most consumers are unaware that they can opt out of such lists. (Beales 3669/13-- 3670/10; Westin 3639/8-19; TU-88-2-58). It is also significant that, when faced with a question at oral argument about whether Trans Union would agree to a notice and opt out system in the target marketing area, counsel for Trans Union evinced a strong reluctance to do so unless non-CRA competitors were similarly asked to abide by a notice and opt out system. (Oral Arg. 54/3-7). Trans Union=s pursuit of equal treatment vis-a-vis non-CRA competitors is again based on its failure to recognize the special privacy concerns that CRAs= databases create.

90 S. 650 would have permitted CRAs to sell target marketing lists that disclosed a wide variety of information, including information that is used in credit eligibility decisions and that bears on consumers= credit worthiness, such as number of tradelines, open dates of loans, and types of tradelines. VOLUME 129 Opinion of the Commission that the provisions of the FCRA at issue here do not violate the First Amendment.

B. The FCRA Is Not AVoid for Vagueness@ Under the Fifth Amendment Due Process Clause As a final argument, Trans Union claims that the FCRA is unconstitutional because the term Aeligibility for credit@ is too vague a concept under the Fifth Amendment. As previously discussed, the definition of consumer report in the FCRA is designed to protect personal credit-related information that is Aused, expected to be used, or collected in whole or in part for the purpose of serving as a factor in establishing the consumer=s eligibility for credit.@ 15 U.S.C. ' 1681a. Trans Union argues that defining consumer reports based on the ultimate purpose to which information is put makes it impossible to know what information is covered and what is not. In this case, we disagree for the following reasons.

A statute is void for vagueness if its prohibitions are not clearly defined so as (1) to give regulated parties adequate notice and (2) to prohibit arbitrary and discriminatory law enforcement. Grayned v. City of Rockford, 408 U.S. 104, 108 (1972); United States v. Thomas, 864 F.2d 188, 194 (D.C. Cir. 1988). The level of vagueness that the Constitution will tolerate depends upon the type of regulation at issue.

Thus, economic regulation is subject to a less strict vagueness test because its subject matter is often more narrow, and because businesses, which face economic demands to plan behavior carefully, can be expected to consult relevant legislation in advance of action. Indeed, the regulated enterprise may have the ability to clarify the meaning of the regulation by its own inquiry, or by resort to an administrative process. The Court has also expressed greater tolerance of enactments with civil rather than criminal penalties because the consequences of imprecision are qualitatively less severe. And the Court has recognized that a scienter requirement may mitigate a TRANS UNION CORPORATION 555 Opinion of the Commission law=s vagueness, especially with respect to the adequacy of notice to the complainant that his conduct is proscribed. Finally, perhaps the most important factor affecting the clarity that the Constitution demands of a law is whether it threatens to inhibit the exercise of constitutionally protected rights. If, for example, the law interfered with the right of free speech or of association, a more stringent vagueness test should apply.

Village of Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489, 498-499 (1982).91 91 Trans Union contends that the level of scrutiny should be especially demanding because the FCRA is a content-based (as opposed to contentneutral) regulation of speech. We disagree. Stringent scrutiny of a contentbased regulation, assuming this is a content-based regulation, is necessary only where the regulation will impose an Aobvious chilling effect on free speech.@ See Reno v. ACLU, 521 U.S. 844, 871-72 (1997). Here, any Achilling@ impact of the regulation is mitigated by the fact that the regulation is a civil regulation, and it affects the conduct of sophisticated businesses who have a substantial incentive, and the ability, to determine the reach of the statute. Indeed, the D.C. Circuit has indicated that in the context of an administratively enforced regulation of commercial speech, all the Fifth Amendment requires is that Ait must be possible for the regulated class to perceive the principles which are guiding agency action.@ Pearson, 164 F.3d at 661. An agency can met this requirement by Acase by case@ adjudication rather than through Aa comprehensive definition all at once.@ Id. That standard is met here. Trans Union also claims that stricter scrutiny is appropriate because Section 620 of the FCRA provides criminal sanctions in certain situations. However, this action is civil (indeed, the Commission lacks the authority to enforce the criminal provision of the FCRA) and the only issue here is whether the FCRA is sufficiently precise to support the Commission=s civil enforcement action, not whether the FCRA would be unduly vague when enforced criminally. Nonetheless, even if judged as a criminal statute, the scienter requirement - - Section 620 covers a Aknowing@ or Awilling@ unauthorized disclosure - - Amay mitigate a law=s vagueness, especially with respect to the adequacy of notice to the complainant that his conduct is proscribed.@ Hoffman Estates, 455 U.S. at 499.

VOLUME 129 Opinion of the Commission Based on this guidance and the facts contained in the record, we conclude that the term Aeligibility for credit@ in the FCRA=s definition of a Aconsumer report@ is not too vague to provide adequate notice to Trans Union of the conduct proscribed under the FCRA. We also believe that the term is sufficiently clear to prevent arbitrary and discriminatory enforcement. This is true even though the Act has some impact upon Trans Union=s First Amendment right to freedom of expression. The record here amply demonstrates that information that indicates the existence of credit relationships, and other information about such credit relationships, is information that is used and expected to be used in establishing a consumer=s eligibility for credit. We therefore disagree with Trans Union=s contention that linking the information protected by the Act to the purpose for which the information is used is impermissibly vague. Our conclusion is buttressed by the Supreme Court=s instruction to examine whether the meaning of a regulation is clear from an industry member=s vantage point. Hoffman Estates, 455 U.S. at 501 n.18. Trans Union is a CRA that assists in the development of credit scoring models and has a substantial prescreening business. See discussion supra pp. 4 and 17-18. Its business also requires that it know what information is used in establishing a consumer=s credit eligibility. Accordingly, Trans Union cannot credibly argue that it had insufficient notice as to the information that falls under the consumer report definition. Indeed, Trans Union is statutorily obligated to know how its information is used. See Sections 604(a) and 607 of FCRA. We can also infer knowledge through Trans Union=s termination of many of the practices now challenged by the Commission, following the statutory amendments making clear that such conduct could lead to monetary penalties. See discussion supra p. 11. Finally, Trans Union asserts that it is uncertain whether it can disclose certain information - - including name, address, social security number, and credit performance data - - on a zip-plusfour basis. Trans Union asserts that this uncertainty renders the FCRA unconstitutionally vague. We disagree because any TRANS UNION CORPORATION 557 Opinion of the Commission question pertaining to the disclosure of these particular pieces of information is irrelevant to Trans Union=s use of core consumer information which is of concern in this case. Moreover, any claim of Avagueness@ is without merit as the Commission has never condoned the disclosure of credit performance information aggregated on a zip-plus four basis and pursuant to Pearson, we need not address this issue here where the question is not before us.92 Accordingly, we conclude that the definition of Aconsumer report,@ including the term Aeligibility,@ under Section 603(d) of the FCRA gives regulated parties like Trans Union adequate notice of what conduct is proscribed and is sufficiently clear to avoid risk of discriminatory enforcement93. For these reasons, the FCRA is not unconstitutionally vague.

92 Trans Union asserts that this case is analogous to the recent Supreme Court case Reno v. A.C.L.U., 521 U.S. 844, which struck down portions of the Communications Decency Act (ACDA@). The regulation in Reno defined prohibited speech Aby contemporary community standards@ and the Court held the speech restriction unconstitutional. Id. Here, Trans Union argues that the Commission=s application of the FCRA is similarly dependent upon the views of the community receiving the message, implying that the FCRA is also constitutionally flawed. This analogy fails for several reasons. First, the Reno Court expressly declined to make any finding of constitutionality under the Fifth Amendment=s void for vagueness doctrine, deciding the case on First Amendment grounds only. Id. at 864. Second, Reno involved a criminal statute and a complete ban on pure speech and therefore was evaluated under a stricter standard. Third, applying the definition of Aconsumer report@ and the term Aeligibility for credit@ in this case does not depend on the views of the Acommunity@ recipients of the information. It depends on the use to which such recipients put the information, a use which Trans Union could easily ascertain. Finally, Trans Union=s liability is also based in part on our finding that Trans Union provided information that it expected to be used in credit granting decisions. See discussion supra p. 33.

93 See Grayned, 408 U.S. at 114 (AAs always, enforcement requires the exercise of some degree of police judgment, but, as confined, that degree of judgment here is permissible.@).

VOLUME 129 Final Order VII. CONCLUSION Based on the foregoing, as well as the thorough and substantial record in this case, we find that Trans Union violated Sections 604 and 607(a) of the FCRA because its target marketing lists are Aconsumer reports@ that were disclosed without a Apermissible purpose.@ We also find that the FCRA, as applied in this case, passes constitutional muster. FINAL ORDER This matter has been heard by the Commission upon the appeal of respondent Trans Union Corporation from the Initial Decision and Order on remand, and upon briefs and oral argument in support of and in opposition to the appeal. For the reasons stated in the accompanying Opinion, the Commission has determined to adopt the Administrative Law Judge=s findings and conclusions to the extent that they are consistent with those set forth in the accompanying Opinion. Accordingly, the Commission enters the following order:

IT IS HEREBY ORDERED that, consistent with the terms of this opinion, respondent Trans Union Corporation, and its successors and assigns:

a) Cease and desist from distributing or selling consumer reports, including those in the form of target marketing lists, to any person unless respondent has reason to believe that such person intends to use the consumer report for purposes authorized under Section 604 of the FCRA.

TRANS UNION CORPORATION 559 Final Order b) Maintain for at least five (5) years from the date of service of this Order and upon request make available to the Federal Trade Commission for inspection and copying, all records and documents necessary to demonstrate fully its compliance with this Order. c) Deliver a copy of this Order to all present and future management officials having administrative, sales, advertising, or policy responsibilities with respect to the subject matter of this Order.

d) For the five (5) year period following the entry of this Order, notify the Commission at least thirty (30) days prior to any proposed change in respondent such as dissolution, assignment, sale or change in control resulting in the emergence of a successor corporation, the creation or dissolution of subsidiaries, or any other change in the corporation that might affect compliance obligations arising out of this Order.

e) Within one hundred and eighty (180) days of service of this Order, deliver to the Commission a report, in writing, setting forth the manner and form in which it has complied with this Order as of that date. By the Commission, Commissioner Leary not participating. VOLUME 129 Initial Decision INITIAL DECISION By Timony, Administrative Law Judge:

I. INTRODUCTION The complaint, filed on December 15, 1992, alleges that Trans Union's sale of target marketing lists, with consumer information from its credit reporting files, was the sale of consumer reports to persons who did not have a permissible purpose to receive them, in violation of the FCRA, 15 U.S.C. ' 1681.1 Administrative Law Judge Parker entered summary decision on September 20, 1993, upholding the complaint.

Trans Union appealed to the Commission, which upheld Judge Parker on September 28, 1994.2 On appeal, the United States Court of Appeals for the District of Columbia Circuit held that there was a genuine issue of material fact and that summary decision was inappropriate and remanded the case to the Commission. Trans Union Corp. v. FTC, 81 F.3d 228 (D.C. Cir. 1996).

By Commission Order of July 9, 1996, the matter was remanded for trial. I was assigned to preside on December 22, 1997. The trial commenced on February 17, 1998. The transcript is 3,962 pages. About 500 exhibits were received. The record closed on March 27, 1998.3 1 116 F.T.C. 1334 (1993). The original complaint in this matter also alleged violations of the FCRA by Respondent in connection with its prescreening activities. Those allegations were settled by consent agreement on November 18, 1993. Id. at 1357-61. Respondent's prescreening is not, therefore, at issue.

2 118 F.T.C. 821 at 838-95 (1994).

3 Judge Parker ruled that the time limits required by recent amendment were impossible in this case. Order of September 30, 1996. TRANS UNION CORPORATION 561 Initial Decision II. FINDINGS OF FACT A. Respondent 1. Trans Union Corporation (ATrans Union@) is a Delaware corporation, with its office at 555 West Adams Street, Chicago, Illinois 60661.

2. Trans Union gathers information on consumers and sells consumer reports in interstate commerce. Its main competitors are Experian and Equifax. (Rodgers CX 191 at 47/3-13.)4 3. Trans Union is a consumer reporting agency, Section 603(f), Fair Credit Reporting Act (AFCRA@). It sells data about the credit of millions of Americans. Buyers use this information to evaluate consumers' credit. (Rodgers CX 191 at 27/3-7.) 4 References to the record are made using the following abbreviations: F Finding of Fact CX Commission Exhibit TU Respondent's Exhibit References to trial transcript are usually made using witness name, page and lines:

Dexter 1231/4-21.

References to exhibits include number and page: CX 125-3 TU-115 p.4.

Reference to deposition exhibits include witness name, exhibit number, and transcript page and lines:

Marquis CX 188 at 147/20 -- 148/1.

VOLUME 129 Initial Decision 4. PerformanceData is a division of Trans Union engaged in Atarget marketing@ since 1987. Previously known as ATransMark@ then as ATrans Union Lists,@ it became APerformanceData@ in 1997. (CX 84-A.) 5. Trans Union's Atarget marketing@ uses information from its consumer reports. It prepares a list of consumers who meet criteria specified by the client and sells this list for use in soliciting those consumers. Target marketing aims sales efforts to consumers most likely to respond to an offer. (Admitted.) B. Consumer Reporting Agencies 1. Credit Reporting 6. In the United States, information on credit-worthiness of individuals for use by credit grantors is gathered by consumer reporting agencies from credit grantors. (Connelly 2508/10-14.) 7. Consumer reporting agencies do not pay for the data they receive from credit grantors, but charge those credit grantors for edited credit data. (Connelly 2590/8-17.) 8. Credit grantors supply information to consumer reporting agencies so they can later determine credit eligibility based on information they receive from consumer reporting agencies. (Pendleton 405/1-5.) 9. Credit grantors need accurate data about the creditworthiness of consumers. (Johnson 1206/17 -- 1209/7.) TRANS UNION CORPORATION 563 Initial Decision 2. Evolution of the Credit Reporting Industry 10. The credit reporting industry has changed over time. Initially, retailers and other credit grantors relied on their own experience with a consumer to grant credit. Credit grantors started to exchange information about consumers. Credit bureaus developed to pool information about consumers. (Johnson 1206/16 -- 1208/1.) 11. After World War II, consumer credit grew. Computers enabled consumer reporting agencies to store more information. (Johnson 1205/24 -- 1206/11; 1210/7-16.) This led to a uniform system of reporting credit information. (Johnson 1206/7-11; 1210/7-16.) The industry consolidated with large regional and national privately held consumer reporting agencies. (Connelly 2494/19 -- 2496/21.) 12. From 1975 to 1980, the large national consumer reporting agencies offered Aaffiliations@ whereby local bureaus own data housed in the computers of the national consumer reporting agencies which supply information in their computers to credit grantors. (Connelly 2498/4 -- 2499/2, 2499/7-15.) In the late 1980s, a final wave of consolidation took place, resulting in three consumer reporting agencies -- Equifax, Experian, and Trans Union. (Connelly 2506/2 -- 2508/14.) 3. Credit Scoring 13. The information credit grantors use to evaluate consumers has evolved. At first, credit grantors pooled only negative information. Consumer reporting agencies started gathering both negative and positive information. (Johnson 1209/12-25.) 14. Computers enabled credit grantors to analyze information about a consumer through Acredit scoring.@ (Johnson 1209/15 -- 1213/11.) VOLUME 129 Initial Decision 15. Credit scoring uses credit information to build models to predict a consumer's likely future credit performance. Most of the data used for credit scoring comes from the consumer reporting agencies. (Coffman 3825/18 -- 3826/2; Rapaport 673/15-25.) This data is objective and is a better predictor than information provided by the consumer filling out an application. (Coffman 3806/2-14; 3857/24 -- 3858/4; 3858/17 -- 3859/3; CX 122-P.) C. Trans Union Consumer Reporting 16. Trans Union has CRONUS, its system for supplying consumer reports. (Botruff 2043/3-6; Rock 2086/15-16.) CRONUS contains 600 million records. (Weith 1867/21-23.) 17. Trans Union receives consumer accounts receivable information from credit grantors such as banks, mortgage companies, credit unions, automobile dealers, and collection agencies, as well as public record information. (Stockdale 873/22- 25.) Trans Union receives 16,000 magnetic tapes or computer disks monthly, with information on 85,000 customers. (Stockdale 874/4-23.) 18. Trans Union receives information on 1.8 billion tradelines5 a month, including public records, collections, student loans, and child support information. (Stockdale 908/1-19.) Some customers report information daily to Trans Union; most report monthly or according to their billing cycles. (Stockdale 904/8-11; Frank CX 186 at 19/3-6.) Most subscribers to Trans Union (credit grantors) report customer account information in the Metro Format, a form designed by credit grantors. (Stockdale 901/11-18; Botruff CX 181 at 26/15 -- 27/17; 65/3-12.) 5 A Atradeline@ is a consumer account. F 22, 23. TRANS UNION CORPORATION 565 Initial Decision 19. Information submitted by credit grantors to Trans Union is added to CRONUS. (Stockdale 879/14-19; 880/3 -- 881/8.) Accounts receivable information is separated geographically. (Stockdale 879/7-16.) Due to large volume, only some files on CRONUS can be updated each day. (Stockdale 879/13-16.) CRONUS is updated each week. (Botruff CX 181 at 30/18 -- 32/7.) 20. To match incoming information to that in CRONUS, Trans Union will use account number, name, account type and house number of address. If no match occurs, a new record is created on CRONUS. (Stockdale 909/13 -- 910/8; 911/17-25.) 21. Information is also added to CRONUS by the ANCI file. (Stockdale 881/10 -- 882/9; 884/6-11.) The ANCI file gets information from consumers and credit grantors. (Stockdale 882/10-24.) Each day the ANCI file is read into CRONUS. (Stockdale 884/6-11.) 22. For CRONUS, tradeline information includes: a customer's account number, name, address, telephone number, social security number, and any generational suffix; the open date of the account; the subscriber's (credit grantor) name and code, and its kind of business (AKOB@); the verified date on the account; the type of loan; the credit limit assigned by the credit grantor; and the closed date of the account. (Stockdale 893/6-15; 894/4-12; 895/16 -- 896/1; 896/19-22; 897/13 -- 898/2; Botruff 2049/1-6; Weith 1844/18-22; Tr. 872, 875-76, 3372-73.) 23. On CRONUS, public records and collection accounts are considered Atradelines.@ CRONUS contains records that have only public record information and records that contain inquiries only - - that is, records with no trade information and no public record information; less than 1% of the records in CRONUS do not have tradelines. (Stockdale 906/1-8, 21-23; 904/15-18; Botruff CX 181 at 62/15-24; Tr. 904-05; CX 358-G.) VOLUME 129 Initial Decision D. Trans Union's Target Marketing 24. Trans Union uses credit information from CRONUS, along with other information, to create products for its target marketing business, using the AMaster File@ database (F 25-36; 112-116), and AStandard Characteristics@ (F 108-111). Trans Union's target marketing products are: Aindicator@ or Aselect@ (F 37-76); modeled products (F 77-98); and TransLink and the other databases (F 99-107).

1. Trans Union's Master File 25. Trans Union's data for target marketing is its Master File. (CX 72-C.) The source for the Master File is CRONUS and the TransMark file which is a Asnapshot@ of CRONUS. (Cabigon 1365/13-18; Kinsinger 2017/19-23; Weith CX 196 at 179/11-13.) The TransMark database is created each month of records from CRONUS. (Weith CX 196 at 179/12-13, 21-22; 180/11-12.) It combines all CRONUS records on the same name and address. (Weith CX 196 at 186/12-19; 188/7-14.) 26. The Master File is rebuilt three times a year (in 1997, in April, September and December). (Admitted; Cabigon 1366/6-12; Davis 62/16 -- 63/8.) The Master File involves 160 million persons and 105-110 million households. (Weith 1859/8-18; CX 333.) 27. The Master File excludes persons based on the number of and qualifications of their tradelines. (Weith CX 196 at 189/19 -- 190/11; 191/7-15; CX 101-A - C; Cabigon 1372/7 -- 1374/8; CX 329-A; CX 341.) 28. The September 1997 Master File initially verifies the existence of at least one open and qualifying trade in the individual's CRONUS record. AVerified@ means that some activity occurred within the last 12 months, and Aactivity@ means that something was reported, such as the opening of the tradeline or TRANS UNION CORPORATION 567 Initial Decision the receipt of a payment. (Cabigon 1372/18 -- 1373/7; CX 329-A; Weith CX 196 at 197/24 -- 198/14.) 29. For a single trade to qualify for the Master File, there must be no date closed present on the record, the trade must not be disputed, and the trade cannot be a collection set or a public record set. (Weith CX 196 at 191/7-15; 227/1-5; Cabigon 1374/5- 21.) 30. A single tradeline with an account type AI@ (installment) or AM@ (mortgage) and a current balance of zero will not count as the single qualifying trade for the Master File because those indicators signify that the tradeline is closed. (Weith CX 196 at 194/13 -- 195/3.) 31. The Master File Alist exclusion@ is a second qualification requiring that a consumer have at least two open trades, one verified within 12 months. Prior to January 1998, consumers with less than two open trades but still meeting the one qualified tradeline requirement were placed in the Emerging Consumers file. Consumers who have at least two open trades, one verified within 12 months, were than included in the Master File and the Attribute or Standard Characteristics File. (Cabigon 1373/8 -- 1374/4; CX 329-A; Weith CX 196 at 224/23 -- 225/10; CX 104- B.) 32. Trans Union withdrew some information in the Master File for sale after September 30, 1997, but the minimum requirement for the Master File did not change until January 1998. (Weith 1831/23 -- 1832/6.) 33. The minimum requirement for the Master File now is either two trades reported in the last six months, or one trade reported in the last six months matched to an outside vendor file. (Weith 1830/23 -- 1831/4; Cabigon 1386/14 -- 1388/7; CX 332-A; CX 339-A.) VOLUME 129 Initial Decision 34. The Master File obtains a subscriber number from a person's tradeline in CRONUS. (Cabigon 1393/11-16; Weith 1847/13 -- 1848/19.) 35. Employees of ACXIOM, a corporation whose single largest stockholder and second largest customer is Trans Union, with Trans Union employees, build and sell information in the Master File. ACXIOM ensures that addresses are correct. (Davis 85/12-17; Weith 1842/4-15; Hinman 2338/7 -- 2339/11.) 36. The Master File is verified through the U.S. Postal Service's National Change of Address file (ANCOA@). (Weith 1838/10-21; Davis 87/1-10.) Fraudulent addresses are excluded. (Weith 1833/19 -- 1834/24.) 2. Indicators from the September 1997 Master File 37. The Master File contains Aindicators@ (Aselects@) (such as the fact that a person has a bank card) derived from CRONUS and other sources. CX 1-A -- Z-6 shows the selects available to a customer assembling a target marketing list. (Davis 91/2-21.) Half of this information is from CRONUS. (Cabigon 1438/12-25.) a. Automobile 38. Trans Union offers data showing open automobile loans, loan type (lease, refinanced loan, equity transfer loan, automobile loan), open dates and expiration dates for the lease or loan, and the high credit amount of most current loans/leases. Trans Union also offers Adriver@ lists. (CX 1-A -- E.) 39. The source for the Aauto expiration date@ (CX 1-A), Aauto high credit@ (CX 1-B), Aauto loan type@ (CX 1-B), and Aauto open date@ (CX 1-C), is CRONUS. (Cabigon 1377/10-24.) A consumer is a Adriver@ (CX 1-E) if CRONUS shows an auto loan or a business that issues gasoline cards. (Cabigon 1378/12-19.) When the September 1997 Master File examines CRONUS to determine TRANS UNION CORPORATION 569 Initial Decision whether a person has an auto loan, it will not consider the loan if the open date is more than five years old. (Weith CX 196 at 211/13-16; 212/3-4.) The availability of loan and lease dates permits targeted promotions. (CX 45; CX 261-E -- G, J.) 40. [redacted] 41. [redacted] 42. [redacted] 43. [redacted] b. Bank Card 44. Trans Union offers a list of persons with an open bank card (including the open date of the most recent bank card) and an open premium bank card (including the open date of the most recent premium bank card). (CX 1-C, R, T.) Trans Union's September 1997 Master List defines and calculates its premium bank card select as a bank card with a credit limit of more than $ 9,999. (Dexter 1271/17-20; CX 64-A; Weith 1867/5-13.) 45. To generate the indicator Apresence of an open bank card@ (CX 1-C), the September 1997 Master File program looks to the CRONUS record. (Cabigon 1377/21 -- 1378/4.) For a person to qualify for a premium bank card, CRONUS must indicate a tradeline with a type of business AB,@ an account type equal AR,@ and a credit limit field indicating greater than or equal to $ 9,999. (Weith CX 196 at 230/22 -- 231/8; CX 102-E; Cabigon 1427/9- 17.) 46. Trans Union's target marketing competitors offer bank card and premium bank card selects obtained from self-reported survey responses or from files of credit card information obtained VOLUME 129 Initial Decision from Business Development, Incorporated (ABusiness Development@) or Dresden Direct. (F 47-50; 148-154.) 47. [redacted] 48. First Data's active bank card select is an open bank card in a household. (Cleary 3094/17-22; TU-119 at 6.) [redacted] Nothing in the self-reported credit card information nor in the information provided by Business Development reveals the open date of the bank cards or premium bank cards. (TU-130 p.4; TU- 117 p.2; TU-120 p.2; Schultea 3928/2-4; CX 310-D.) 49. [redacted] 50. In addition to information from self-reported questionnaire responses and Dresden Direct, ACXIOM obtains credit card information from other sources, including information that a consumer used a credit card at one time, obtained from companies that compile that information. (Hinman 2247/18 -- 2248/7, 14-16.) 51. JAMI Marketing (AJAMI@) manages several lists that purport to provide credit card information. The source of the credit card data in those lists is neither credit grantors nor consumer reporting agencies. (Miller 3017/6-11.) The source is a list of the list owners' customers who used a credit card to make a purchase from the list owner. (Miller 3018/22 -- 3019/11; 3058/5- 17; 3060/20 -- 3061/13; 3065/13 -- 3067/25.) JAMI-managed lists do not have premium bank card selects. (Miller 3019/22 -- 3020/5.) c. Department Store Card 52. Trans Union's September 1997 Master File offers open department store trades (including open date of the most recent department store trade). (CX 1-E, R.) To generate the open department store card indicator (CX 1-E), the Master File looks to CRONUS. (Cabigon 1378/5-11.) TRANS UNION CORPORATION 571 Initial Decision 53. Trans Union's target marketing competitors offer similar department store card selects obtained from self-reported survey responses or from files of credit card information obtained from Business Development, or Dresden Direct. (F 54; 47-48; 148- 154.) 54. [redacted] d. Finance Trade 55. Trans Union's September 1997 Master File List offers several finance loan lists including an open finance trade (with the open date of the most recent finance loan), a A30/60/90 day@ finance trade, a mortgage or auto loan, and a closed finance loan trade. (CX 1-F, R.) A A30/60/90 day@ trade is due in 30/60/90 days. (Davis 154/24 -- 155/7; Cabigon 1412/7-11.) 56. To generate the open finance trade indicator (CX 1-F), the September 1997 Master File examines CRONUS to determine if the subscriber's business is Afinance,@ if the account type is installment, and if the finance trade has an open date. (Cabigon 1378/20-23; Weith CX 196 at 228/14-25; CX 102-A -- B.) 57. To generate the indicator A30/60/90 day finance trade@ (CX 1-F), the September 1997 Master File determines if the individual has a tradeline with a finance company that issues 30/60/90 day loans by examining from CRONUS the subscriber's business and the account. (Cabigon 1378/24 -- 1379/16.) To generate the Afinance loan closed@ indicator (CX 1-F), the September 1997 Master File uses CRONUS. (Cabigon 1379/20- 24.) VOLUME 129 Initial Decision e. Head of Household 58. Trans Union offers a select for head of household. (CX 1- H.) The Ahead of household@ (CX 1-H) is derived from CRONUS; the person with the highest number of trades is the head of household. (Cabigon 1380/1-19; Weith 1832/11-14.) If a household contains two CRONUS records with the same number of trades, the head is the first record on file. (Weith 1832/17 -- 1833/6.) f. Mortgage 59. Trans Union offers mortgage lists from the September 1997 Master File List including: an open mortgage; a second open mortgage; and the open and closed dates and high credit amounts of both mortgages. An additional list for both mortgages is the loan type (including indicators for refinanced and secured mortgages and secured home improvement loans). (CX 1-N -- P.) 60. The September 1997 Master File lists Amortgage one@ and Amortgage two@ (CX 1-N), mortgage high credit (CX 1-O), mortgage closed dates (CX 1-O), and mortgage type (CX 1-P) are derived from CRONUS. (Cabigon 1381/18 -- 1383/1.) 61. The September 1997 Master File list AHome Value Range@ uses E-Val's model to calculate an individual's home value. (CX 1-J.) The following are counted: (i) mortgage; or (ii) nonautomobile installment loan with a high credit amount greater than $ 49,000 or current balance greater than $ 50,000; or (iii) real estate installment loan; or (iv) non-automobile secured installment loan with a high credit or current balance greater than $ 16,000. (Weith CX 196 at 205/21 -- 207/3.) 62. CX 326-B specifies the Trans Union attributes (F 108- 111) used for the home value calculation in September 1997. The following attributes were used to calculate home value: RE20 (months since oldest revolving trade opened, CX 327-B); MT28 (total mortgage high credit/credit limit, CX 327-B); MT32 TRANS UNION CORPORATION 573 Initial Decision (maximum balance owed on all mortgage trades, CX 327-B); MT33 (total balance of all mortgage trades, CX 327-B); PB14 (number of premium bank card trades verified last 24 months, CX 327-D); and G051 (percent of trades never delinquent, CX 327- E). (Cabigon 1370/19 -- 1371/8; CX 3-S.) 63. Trans Union uses age, mortgage, the original mortgage amount, and the opening balance to calculate the current home market value for a house. (Weith 1848/24 -- 1849/25.) The calculation uses demographic information from CRONUS and from the census. (Weith 1850/17 -- 1851/3.) 64. The Master File list Ahome equity actual@ (CX 1-I) (the dollar amount of equity in a person's home) and Ahome equity range@ (CX 1-I) are calculated by using mortgage high credit and mortgage balance information from CRONUS and subtracting them from the estimated home value. (Weith 1846/17 -- 1847/5; 1850/1-3; Cabigon 1380/20 -- 1381/11.) 65. [redacted] 66. Metromail obtains information regarding mortgages -including rate type, loan type (FHA, VA, Conventional), mortgage amount, and purchase price -- from deeds in the public record. (Litz 2972/6-9; 2975/5 -- 2976/7.) Metromail also offers its clients the AEquity Spenders@ list which identifies households, from public record information and from self-reported information, that have recently refinanced or purchased a home. (Litz 2973/6-19; TU-115 p. 113; TU-116 p. 16.) 67. [redacted] 68. [redacted] 69. [redacted] VOLUME 129 Initial Decision g. Length of Residence 70. Trans Union offers a length of residence select showing the number of years a person has resided at an address. (CX 1-J.) The Master File length of residence indicator uses data from CRONUS, including mortgage open dates. (Cabigon 1371/9 -- 1372/3; 1381/12-17; 1415/14-19.) h. Mail Order Buyer 71. Trans Union lists mail order buyers, including its AMOB1" indicator (CX 1-J) based on identifying tradelines (such as Spiegel's, L.L. Bean, Eddie Bauer) on CRONUS and identifying as a mail-order buyer on the Master File persons with those tradelines. (Weith 1847/13 -- 1848/19.) i. Singles Select 72. Trans Union offers a Asingles@ list showing single persons. (CX 1-U.) This Master File indicator is from tradelines on CRONUS showing that credit was granted to a person (rather than a joint account). (Cabigon 1419/21 -- 1420/4; Weith CX 196 at 219/2-4; CX 99-O.) j. Student Loan 73. Trans Union offers several student loan selects showing a student loan (with the open date of the most recent student loan), the aggregate high credit amount of all of a person's student loans, and a closed student loan. (CX 1-W.) A CRONUS record qualifies as a student loan on the September 1997 Master File if the loan type is Astudent@ and the date closed is not equal to zero. (Weith CX 196 at 212/10-21; CX 100-N.) TRANS UNION CORPORATION 575 Initial Decision k. Upscale Retail Card Selects 74. Trans Union's September 1997 Master File offers a select for an open upscale retail card (including the open date). In its promotional materials and internally, Trans Union indicates that its upscale store select is based upon the National Retail Federation's listing of Aprestigious@ stores. (CX 1-Q, Y; CX 72-C; Davis 226/24 -- 227/3.) 75. For the upscale retail card and open date indicators, the September 1997 Master File searches CRONUS records to find an open upscale retail card and an open date for the tradeline. (Cabigon 1383/2-12; 1393/11-16; Weith CX 196 at 229/7 -- 230/6; CX 102-D.) 76. [redacted] 3. The September 1997 Master File a. Trans Union's AE-Val@ 77. Trans Union offers a product called E-VAL, a scoring system that estimates the amount of equity available in a consumer's home, on its Master File in the form of its AHome Value Ranges,@ AHome Equity (Actual),@ and AHome Equity Range@ selects (F 61-64). (CX 1-I, J; Davis 134/12 -- 135/13; CX 118-B.) E-Val selects were available for sale to target marketing customers prior to October 1, 1997; they are now available only for firm offers of credit. (Davis 66/8-10, 73/4-14.) b. PIC 78. PIC, a product created by Trans Union and LOCUS, predicts, inter alia, the likelihood that a person owns financial service products. PIC is derived from characteristics in CRONUS. (Tr. 1863-64.) Trans Union used its Standard Characteristics (F VOLUME 129 Initial Decision 108-111) in developing PIC and appends individual-level credit data to survey data to model the likelihood of a person owning particular financial products. Until October 1, 1997, Trans Union offered PIC as a Master File select; PIC is currently available only for firm offers of credit. (Davis 111/25 -- 112/24; 136/2-6; 214/1- 5; 234/17-19; Weith 3749/4 -- 3751/1.) c. P$ YCLE 79. [redacted] 80. [redacted] d. SOLO 81. [redacted] 82. [redacted] 83. According to the SILHOUETTE User's Guide, Aassignment to a cluster is based on a credit view of consumers . . . . Cluster 8, because of its generally high level of credit activity, might contain consumers who are good candidates for credit offers.@ SILHOUETTE distinguishes its clusters based on some of the same factors used by Trans Union in its target marketing list products -- e.g., finance accounts, bank revolving, department store, retail -- and the variables in clusters can be Azero,@ Aone,@ and higher numbers. (CX 122.) 84. [redacted] 85. [redacted] 86. Until October 1, 1997, Trans Union offered SOLO to its target marketing customers; currently SOLO is available only in the context of firm offers of credit. (Davis 68/19-22.) Trans Union withdrew SOLO because of the FCRA amendments and the fact TRANS UNION CORPORATION 577 Initial Decision that customers had learned that SOLO clusters correlated with credit performance characteristics. (Davis 67/23 -- 68/4.) 87. Prior to October 1, 1997, Trans Union attached SOLO codes to a customer's own client lists, allowing clients to analyze their existing customer base by SOLO codes. (Dexter 1312/17 -- 1315/4; Davis 70/6-18; CX 14.) This good credit risk list constitutes a consumer report. (Marquis CX 188 at 128/2-9.) [redacted] 88. Until September 30, 1997, Trans Union offered to append SOLO clusters to TransLink lists. (Dexter 1236/22 -- 1237/25; Davis 67/19 -- 68/4; CX 125-E.) 89. [redacted] 90. Products similar to Trans Union's SOLO and P$ YCLE in the target marketing list industry are not as strong as the Trans Union products. (Hinman 2271/9-11.) ACXIOM received SOLO and P$ YCLE codes from Trans Union until October 1, 1997. (Hinman 2269/15 -- 2270/12.) The models that ACXIOM used in place of P$ YCLE and SOLO (Claritas' Affluence and WealthWise and Experian's P$ YCLE) do not provide as strong predictive results as the Trans Union products. (Hinman 2270/13 - - 2271/15.) e. TIE 91. [redacted] 92. [redacted] 93. [redacted] 94. [redacted] VOLUME 129 Initial Decision 95. [redacted] 96. [redacted] 97. [redacted] 98. Metromail contributes income information, based on census data and self-reported consumer survey responses, to IN- SOURCE, a joint product with Experian. (Litz 2983/14 -- 2984/11.) ACXIOM also has income information that it receives from other companies' self-reported survey or product registration questionnaires and the estimated income models available from other list providers such as Polk, Experian, and Metromail. (Hinman 2251/17 -- 2252/9.) 4. Trans Union's Other Target Marketing Files a. TransLink 99. TransLink is Trans Union's reverse append product which matches a name and address with a bank card number. (Smith 1487/20-25.) TransLink works as follows: a merchant gets a bank card number from its customer, sends it to Trans Union, which returns a name and address (Areverse append@). The source of the account number and name and address information is CRONUS. (Weith 1823/22 -- 1824/14; Dexter 1305/24 -- 1307/6; Davis 89/25 -- 90/10; CX 126; CX 132-D; CX 133-B; CX 266.) 100. When a merchant supplies Trans Union with an account number of a joint account, Trans Union returns only one name -the first name on that account. (Weith 1824/16 -- 1825/17; 1827/4-15.) The name of the owner of the card is returned to the merchant, regardless of who used the card. (Weith 1827/19-24.) A TransLink customer complained that the TransLink product returned names differing from the consumers. (Porretto 1628/7 -- 1630/9; CX 273-A -- K.) TRANS UNION CORPORATION 579 Initial Decision 101. When Trans Union returns to the merchant the name and address of the person on the credit card, the merchant prints a mailing label and sends the consumer a piece of junk mail. The merchant can use the consumer's purchase information and know what the consumer bought. (Marquis CX 188 at 144/7-10; 151/10- 12.) 102. Trans Union is the only consumer reporting agency that owns and markets this reverse append product. (Davis 141/6-9; CX 78-Z-29.) Direct Tech, currently owned by Experian, licenses the reverse append product from Trans Union. (Smith 3298/16-23; 3303/5-12.) 103. Citibank does not permit Trans Union to use their credit card account numbers for reverse-append disclosure of names and addresses in the TransLink product. (Marquis CX 188 at 147/20 -- 148/1.) 104. TransLink is among Trans Union's largest selling target marketing products. (Admission.) 105. Until September 30, 1997, Trans Union appended SOLO, TIE, age data, and other Master File elements to TransLink lists. (Dexter 1236/22 -- 1237/25; Smith 1488/23 -- 1489/5; CX 125-E; CX 129; CX 279-B.) b. New Issues File 106. Trans Union offers the New Issues File whereby customers may rent a list of persons who received credit within the last 90 days. The file is created by selecting from CRONUS names and addresses of consumers with at least two tradelines, one of which has an open date within the last 90 days. (CX 4; Respondent's Admissions.) A customer sees the time (30/60/90 days) and the type of credit (retail, finance, mortgage or auto loan trades). (Davis 42/16 -- 43/1; CX 4.) Trans Union discontinued VOLUME 129 Initial Decision the sale of the New Issues File for target marketing customers and now the file is available only for firm offers of credit. (Davis 81/24 -- 82/4.) c. Emerging Consumers File 107. The Emerging Consumers File, based on data from CRONUS, included persons with one tradeline within the prior twelve months. (Cabigon 1373/13-23; CX 329-F.) The tradeline qualified if it was not closed. (Trans Union Corporation's Answers to Complaint Counsel's Third Set of Interrogatories, Answer to Interrogatory No. 3.) Trans Union discontinued the Emerging Consumer File in part because it might be Acommunicating information that we shouldn't be communicating.@ (Davis 89/18- 20.) 5. Characteristics and Attributes 108. [redacted] 109. The source for the Attribute File is CRONUS. (Cabigon 1368/8-22; Davis 35/19-21.) To be included in the Attribute File, a person must have at least two open tradelines, one within the last 12 months. (Cabigon 1373/23 -- 1374/4; CX 329-A.) 110. Trans Union released to its target marketing clients its Standard Characteristics (AMarketing Variables@ (CX 78-Z-20 -- 22)) with names and addresses. (Davis 58/22 -- 59/14.) Among these characteristics are: months since oldest trade opened, and the number of personal finance inquiries, finance installment trades opened in 24 months, and bank card trades. (CX 78-Z-20 -- Z-22.) Trans Union discontinued the disclosure of these characteristics for target marketing on October 1, 1997. (Davis 59/15-19.) TRANS UNION CORPORATION 581 Initial Decision 111. The Standard Characteristics File is available for use in the PerformanceBase/RelationBase File only if a customer is extending a firm offer of credit. (Davis 40/10-16; 120/17 -- 121/4; Dexter 1281/11-16; CX 317.) 6. Changes in the Trans Union's Master File and Target Marketing Products 112. The Master File's content remained unchanged until December 1997. On October 1, 1997, Trans Union discontinued certain files for target marketing. CX 2 lists the files in September 1997 that remained available after October 1, 1997. Trans Union discontinued: all open dates; high credit and loan type (although a finance mortgage trade remained available); student loan closed; and PIC, P$ YCLE, E-Val, SOLO and TIE. (CX 1-A -- Z-6; CX 2-A -- T; Dexter 1281/11 -- 1282/8; Davis 98/17 -- 99/7; 142/5- 12.) 113. Trans Union discontinued the Master File items after October 1, 1997, which were reintroduced in December 1997: 30/60/90 day finance loan; premium bank card; length of residence; and income and home value (both purchased from an outside source). (Davis 165/7 -- 166/23; CX 342-E, H -- K, M -- N.) 114. Trans Union stopped offering the data in Findings 112- 113 to its target marketing customers as of October 1, 1997, because Abefore it was a cease and desist penalty, it now became a $ 2500 per occurrence penalty@ (Dexter 1280/19 -- 1281/10; Marquis CX 188 at 174/23 -- 175/6, 22-25); Trans Union Ahad gone from an environment where the worst thing that could happen is that we would have to stop selling lists to a world where there were significant financial penalties . . . .@ (Davis 142/22-25; CX 298-B.) VOLUME 129 Initial Decision 115. CX 342 contains the Master File items available after December 1997, derived from CRONUS, showing presence of: an auto trade; second auto trade; bank card trade; department store card; driver data; finance loan; 30/60/90 day finance trade; head of household; seven KOB (kind of business) items; mail order trade; one mortgage; two or more mortgages; gold, platinum or optima card; student loan; and upscale retail trade. (CX 315-D, E, G -- M, Q -- W; CX 332-B; Cabigon 1426/9-23; 1427/18 -- 1428/3; 1429/9-17; 1429/22 -- 1430/2; Weith 13:1832/2 -- 1833/6.) 116. A file (APreapproved File@) similar to the September 1997 Master File is now available.

117. This file includes the Standard Characteristics and all of the data are only available for firm offers of credit. (Cabigon 1384/6 -- 1385/18; CX 340-B; CX 333; CX 334; CX 335.) 7. Promotion of Trans Union's Target Marketing a. Sales and Revenue 118. Trans Union sells its target marketing lists nationwide. Salespeople look to firms using mail solicitations or telemarketing. Trans Union also sells lists to brokers, list managers, and wholesalers, who sell the information from Trans Union lists with their own data, for sale or lease to direct marketers. (Davis 45/3 -- 47/24; CX 15; CX 34; CX 41; CX 54; CX 307; CX 42; CX 51; Dexter 1231/4-21.) 119. Trans Union sells target marketing lists for one-time use by its customers either by rental or by license. Trans Union charges a Abase price@ per thousand names, with additional charges per thousand based on the Aselects@ that the customer has chosen. (Davis 44/6-24; 64/6-22; 65/3-14.) TRANS UNION CORPORATION 583 Initial Decision 120. Trans Union and its competitors in the target marketing list industry advertise their lists in the Direct Marketing List Source which includes over 19,000 entries, including consumer, business, and international lists, that are available in 212 markets. It is published every other month by Standard Rate and Data Service (ASRDS@), the leading publisher of advertising rate and direct marketing information, and is considered to be the ABible@ of the target marketing list industry. (Markowski 2427/6-10; 2429/10-20; 2439/1-8; 2430/5-10; 2431/2-4; Dixon 298/9-16; Hinman 2256/17 -- 2257/22.) 121. The format for the Direct Marketing List Source has eleven segments, describing the list, selects and the list source. (TU-163 at A-11.) The information provided is obtained by SRDS from the list owner/manager. Numerous listings in SRDS, however, do not contain information for every segment, including source. The only segment that SRDS confirms independently involves the five most recent sales of the list. (Markowski 2440/3- 11; 2441/6-25.) 122. Trans Union's PerformanceData division employs 46 people including 10 salespersons. PerformanceData had 440 target marketing customers and $ 34 million in revenue in 1997. (Davis 27/18-21; 37/25 -- 38/4; 48/8-10; 141/10-14.) b. Credit-Based Marketing Information 123. Trans Union promotes its Master File as a unique source of credit-based marketing information from CRONUS. (CX 57; CX 58-C; CX 60-A; CX 69; CX 81; CX 264-A; CX 297-A.) A promotional letter states: ATrans Union is a unique provider of credit-based marketing information. Our database is unmatched when compared to traditional direct marketing vehicles on the market today.@ (CX 260-B.) VOLUME 129 Initial Decision 124. Trans Union promotes target marketing lists with credit information:

a. Trans Union's Standard Characteristics correlate highly with Alending activity.@ (CX 263-A.) b. Trans Union data is Ahighly predictive in response modeling and profiling, especially with financial offers.@ (CX 265-A.) c. Trans Union's finance trade select provides consumers who have Agenerally had trouble with their credit in the past and are highly responsive to credit offers.@ (CX 68-A.) d. Trans Union information contains Aspending, payment and demographic data@ that is Ahighly predictive and cost effective.@ (CX 57.) e. ASince credit has been established [for consumers on the student loan list], one could argue that this list would have higher pass rates through the credit bureaus.@ (CX 136.) f. A[The demographic variables that Trans Union can provide] are derived from approved application information and financial information provided by subscribers.@ (CX 70-A.) g. Trans Union's premium bank card select indicates individuals Awho have been approved for this high credit amount in the past.@ (CX 64-A.) 125. Trans Union distinguishes itself from Aits traditional competitors within the credit reporting industry,@ who have not pursued Trans Union's Aopen policy@ regarding the use of credit information by target marketers. (CX 61-A; CX 78-G; CX 268- A.) Although Experian Acomes closest as a competitor,@ Experian TRANS UNION CORPORATION 585 Initial Decision Acannot provide . . . credit based data . . . our data far outweighs their strength.@ (CX 70-B; Dixon 267/11 -- 268/1.) 126. Trans Union promotes the fact that each person in the Master File must have at least two active lines of credit. (CX 7; CX 33-A; CX 61-A; CX 78-G; CX 268-A.) Trans Union also publicizes that a person with no activity in a 12 month period, such as payments or credit, is dropped from the Master File. (CX 69-A.) Trans Union also publicizes the criteria for other lists in its Master File, such as the premium bank card with a credit limit above $ 9,999. (CX 64-A.) 127. The Master File lists chosen by Trans Union's target marketing clients show individual credit information. (CX 33-A; CX 38-C, D; CX 62; CX 68; CX 79; CX 256-A, B; CX 257.) In response to a client offering a $ 5,000 unsecured loan whose Aideal candidate@ had $ 10,000-$ 15,000 in debt, and $ 20,000+ in home equity, a salesperson recommended persons with open mortgage trades, open finance trades, $ 30,000+ in home equity and a home market value below $ 250,000. (CX 67.) Home Mortgage Funding, a mortgage refinancing lender, selected persons with open finance trades and multiple mortgages for a telemarketing offer to assist those whose property Acame up as having a high interest@ mortgage. (CX 19-A -- F.) 128. Mortgage refinance companies and home equity lenders use the Master File lists for finance loans, first and second mortgages, length of residence, high credit amounts or initial loan values for mortgages, and home equity or home value. (CX 18-A, B, J; CX 23-B, C, F, G; CX 24-A, B, L; CX 39-B -- D.) Client mail or telemarketing shows that those solicited have high interest loans or other debt to consolidate. (CX 19-F; CX 26-H; CX 28-B; CX 32-B; CX 39-D.) Clients, such as Mainstreet Mortgage, use a select identifying individuals with open mortgages issued from finance companies. (CX 40-C, D.) Respondent's witness, Kenneth Scott, stated that finance company users are people who have had VOLUME 129 Initial Decision credit problems in the past, and, quite likely, have had a bankruptcy. (Scott 2855/23 -- 2856/7.) 129. Trans Union's mortgage open date lists permit a target marketing client to determine the date a mortgage was taken out and the interest rate. The Mortgage Banc ordered a list of persons with FHA mortgages opened between January 1994 and October 1995 with initial loan values between $ 75,000-$ 99,999 and between $ 100,000-$ 150,000. (CX 17-A-B.) Another ordered a list of persons with FHA mortgages opened between January and June 1995 with an opening balance greater than $ 100,000. (CX 21-A.) 130. Trans Union's lists allow target marketing customers to obtain lists based on detailed credit information, such as whether the person has at least three open tradelines and the types of tradelines (CX 25-A) or whether a person has two open tradelines, one of which has been open for more than 36 months. (CX 23-B - - C; CX 28-A.) 131. Clients may copy information from the Master File. (Davis 162/2-17; Dexter 1249 -- 1250/10.) For the September 1997 Master File, Aprintable@ information includes: open dates for the first and second most current auto loans/leases and mortgages; home value ranges; and length of residence. (CX 1-C, J, N, P.) Other information is Aprintable,@ including high credit amounts and loan type for auto loans and mortgages and the aggregate high credit for all student loans (CX 1-B, C, O, P, W.) 132. Trans Union's Master File showing the presence of an upscale retail card was used as part of the credit-based decision for a mailing of a Apre-approved Rubenstein Bros@ credit offer. (CX 35-A; Tr. 1938-40, 1945-46, 1957-59, 1962-66, 1973-74.) TRANS UNION CORPORATION 587 Initial Decision c. Depth and Timeliness 133. Trans Union emphasizes that the Master File is Athe freshest and most comprehensive@ data due to its Arobust and extensive source of original credit based information@ (CX 268-A; CX 264-A), and that Trans Union has the largest data of consumer credit information in the United States. (CX 75-B.) Trans Union describes the Master File as the Arichest source of individual-level financial data available@ (CX 321), and that its database is Akept fresh and current by nearly two billion updates supplied by credit grantors every month, and is maintained for accuracy and quality.@ (CX 72-B.) 134. Trans Union's Aprimary database@ is updated every 6.5 days from accounts receivable information from the nation's credit grantors. (CX 78-G; CX 268-B.) Trans Union has automobile loan information for all states. (CX 66-A.) d. Individual Observed Behavior 135. Trans Union promotes the Master File as a unique source for individual-level observed behavior data. According to CX 83- C, a PerformanceData brochure, the Master File is Awithout equal@ and its information is Ahighly accurate@ and is Abased on actual behavior -- not self-reported or neighborhood values.@ (CX 83-C.) The Master File is Aliving and breathing data,@ Athe most comprehensive available in terms of observed behavior (not selfreported)@ (CX 264-A) and as the Aonly source of individual-level financial data@ that is Aindividual and behavioral.@ (CX 78-Z-36.) 136. The Master File's data has an advantage over target marketers like Polk, First Data, and Metromail, who rely on survey data directly from individuals, who may provide inaccurate information, while Trans Union's data is reported by third parties, making it more valid. (Dixon 292/14 -- 293/5; Porretto 1621/16- 20.) VOLUME 129 Initial Decision 8. Individual Observed Credit Information for Target Marketing a. Trans Union's Competitors' Target Marketing Lists 137. Trans Union's competitors in the target marketing list industry include: First Data Solutions (formerly Donnelley Marketing) (AFirst Data@); R.L. Polk & Company (APolk@); Metromail Corporation (AMetromail@); ACXIOM Corporation (AACXIOM@); and Experian (formerly TRW)(AExperian@). (Davis 161/5-16; Cleary 2942/4-18; Hinman 2199/19 -- 2200/17; M.Smith 3299/22 -- 3300/8.) (1) First Data, Metromail, and Polk 138. First Data, Metromail, and Polk provide their customers with lists of consumers for use in target marketing. (Hinman 2213/8 -- 2214/14; Scott 2686/15 -- 2687/8.) First Data's DQI<2> contains 97 million households and the names of 185 million persons (Cleary 3083/25 -- 3084/7); Metromail's National Consumer Database (ANCDB@) contains 100 million households and the names of 155 million persons (Litz 2968/12-23; 2969/5- 10); [redacted] 139. [redacted] 140. [redacted] 141. Metromail's NCDB gets information from: public records of real estate transactions from county clerks, courthouses and tax assessor files maintained in Metromail's Realty Database, a separate file of 30 million homeowners; information that Metromail obtains from newspapers; and information from catalogers and magazines that Metromail maintains in its Mail Order Responder (MOR) Bank. (TU-115 pp. 158 and 160; Litz 2970/5 -- 2971/7; 2972/10 -- 2973/3.) TRANS UNION CORPORATION 589 Initial Decision 142. [redacted] 143. First Data, Metromail, and Polk have self-reported information. Metromail's BehaviorBank includes 28 million households and 40 million persons (TU-115 p. 166); Polk's Lifestyle Selector includes 36 million persons (Nusbaum 2879/11- 19); [redacted] Self-reported information is obtained from responses to consumer surveys and from product registration cards. (TU-117; TU-120; TU-130.) Metromail's BehaviorBank data is stored separately from the NCDB information; some of the names in the two files overlap. (Litz 2976/16 -- 2977/17.) [redacted] (2) ACXIOM 144. ACXIOM is developing a list database called InfoBase Prospects, on 117 million households and names of 160 million persons. (Hinman 2340/4-10; 2261/12-19) ACXIOM compiles Consumer InfoBase from data it obtains from credit card information from Trans Union and from Dresden Direct (Hinman 2247/23 -- 2248/21); property transaction data, including mortgage data from deeds obtained from county tax assessors registrars (Hinman 2250/5-14); and income information from selfreported survey data and estimated income models obtained from Polk, Experian and Metromail, and, prior to September 30, 1997, from Trans Union. (Hinman 2251/11 -- 2252/21.) Data in Consumer InfoBase is in InfoBase Prospects. ACXIOM is prohibited by contract from using in its target marketing list business any data supplied by Trans Union for Consumer InfoBase. (Hinman 2308/21 -- 2309/3.) (3) Experian 145. [redacted] VOLUME 129 Initial Decision 146. Experian uses individual information in its consumer reporting database (FileOne) to compile PDS allowed by the consent order Experian (at that time ATRW@) entered with the FTC: name; telephone number; address; zip code; year of birth; age; generational designation; and social security number. The consent order permits Experian to also use mother's maiden name but it does not. (TU-109; M.Smith 3293/24 -- 3294/11.) [redacted] Prior to the consent order, Experian had used other information from its consumer reporting database in its target marketing business, including presence of a bank card, retail card, or a mortgage loan, and account numbers for the reverse append process, the use of which for target marketing purposes is now prohibited under the terms of the consent order. (M.Smith 3287/17 -- 3288/7.) 147. Every name that appears in Experian's consumer credit reporting database also appears in PDS, regardless of the number of tradelines the consumer does or does not have. To be included in Experian's FileOne, a consumer does not need any tradelines. (M.Smith 3428/18 -- 3429/8-18.) (4) Business Development 148. To the extent Trans Union's competitors obtain any information provided by credit grantors for use in their target marketing list business, they generally do so indirectly from Business Development, a small company (4 1/2 employees, 1997 revenue of $ 200,000, no profit in 1997) whose primary business is providing copies of its National Marketing File. Business Development is not a consumer reporting agency, but rather a business that collects information from credit grantors for use in their customer mailing lists. (Schultea 3946/23 -- 3947/12; 1028/5 -- 1029/2.) 149. The National Marketing File is a list of 90 million consumers who are credit users or who have used credit that Business Development provides to its contract customers -including Dresden Direct and First Data. (Schultea 3905/10-16; TRANS UNION CORPORATION 591 Initial Decision 3907/18 -- 3908/1; 3909/1-7; 3910/20 -- 3911/1.) The National Marketing File contains two credit card related fields: a counter field that indicates number of tradelines including credit cards, up to nine; and a bank card indicator field that is either blank or contains a single AB@. (Schultea 3911/3 -- 3913/7; Tr. 3943/6-23.) The National Marketing File contains no names of credit grantors and does not differentiate any type of credit card except bank card. (Schultea 3911/22-24; 3913/2-7.) 150. The National Marketing File was built in 1989-1990 by merging the customer lists of 20-25 of Business Development's major retail company and bank clients who had credit card customers. The counter was increased by one each time a consumer's name appeared on one of the lists and a AB@ was placed in the bank card field if the customer's name was on a list from a bank. (Schultea 3915/2 -- 3916/21.) These lists were obtained in connection with Business Development's providing target marketing services, such as developing target marketing lists. (Schultea 3913/12 -- 3914/23.) 151. Business Development updates the National Marketing File whenever it receives a list of consumers from a client. If a consumer is already in the National Marketing File, the counter is increased by one and, if the client is a bank and there is no AB@ in the bank card field, a AB@ is added. The bank card field contains only one AB@ even if the consumer has appeared on more than one bank's list. (Schultea 3913/2-4.) If there is a consumer name on the client's list that is not already in the National Marketing File, the name is added to the file. (Schultea 3917/20 -- 3918/18.) 152. A record is deleted from the National Marketing File only when the U.S. Postal Service's National Change of Address (ANCOA@) file indicates that the address is no longer viable or the consumer is deceased, in which case the entire record is removed from the National Marketing File. (Schultea 3912/4-12; 3918/24 - - 3919/7; 3920/2-11.) Otherwise, the information from the VOLUME 129 Initial Decision original file build in 1989-1990 is still contained in the National Marketing File, as is all information added later. The National Marketing File contains references to credit card accounts that have been closed or are inactive. (Schultea 3922/3-14.) 153. Business Development does not know that the consumers in the National Marketing File use one or more credit cards. Business Development's lists show active bank card holders, active retail card holders, and active finance users. (Schultea 3928-34, 3940.) 154. Business Development knows only that consumers have, or had, the number of credit cards -- but not that they are using that credit or that they currently have the number of cards designated. (CX 310-C, D; Schultea 3923/8-23; 3924/24 -- 3925/5.) A AB@ indicates only that consumers have, or had, a bank card, not that they currently have or use a bank card. (Schultea 3926/16-20.) The National Marketing File does not contain salary ranges, age ranges, dates accounts were opened, credit limits on accounts, loan amounts, whether a consumer file has been closed, or whether a consumer has a new bank card. (Schultea 3927/15 -- 3928/13.) b. Trans Union's Exclusively Individual-Level Data 155. Consumer information in target marketing lists can be individual, household, census block, zip-plus-four, and zip code. AZip-plus-four@ refers to a geographic area defined by a nine digit postal code and, in target marketing terms, Azip-plus-four@ level data refers to aggregated data of consumers or households from a particular Azip-plus-four@; the information within the Azip-plusfour@ is averaged. (Davis 156/1 -- 157/4.) [redacted] 156. Individual-level data is better for target marketing response rates and predicting buying behavior than data at broader levels, such as household or zip-plus-four. Respondent's witness, Donald Hinman, described the difference between the levels of data as follows: AThe broader the geography, the less powerful the TRANS UNION CORPORATION 593 Initial Decision data tends to be. And when you go from a broad geography to a narrow geography, it's much more powerful. When you go from a household to an individual it gets even more powerful.@ APowerful@ in this context refers to the Aability to accomplish your market objective. To achieve your response rate or to predict a buying behavior.@ (Hinman 2259/6 -- 2260/2.) 157. Trans Union's competitors in the target marketing list industry furnish information in their target marketing lists primarily at the household level or broader. 158. [redacted] 159. [redacted] 160. ACXIOM collects information on an individual level and records it on a household level. (Hinman 2260/15-16.) Within ACXIOM's household-level records, estimated income is household-level, but age and date of birth is individual-level data. (Hinman 2262/4-11.) 161. [redacted] 162. [redacted] 163. [redacted] E. Use of Trans Union's Target Marketing Information in Credit Eligibility The Court of Appeals remanded this case on the issue of whether the existence of a tradeline constitutes a consumer report. The Court stated that complaint counsel might satisfy the factual burden with evidence indicating that Trans Union intended the existence of a tradeline to serve as a factor in credit-granting decisions. The Court added that evidence Athat credit decisions VOLUME 129 Initial Decision could be made, even in part, on such 'existence' information might be probative of Trans Union's intent,@ and that the Commission might embark on a similar inquiry with respect to any list criterion employed by Trans Union in its target marketing lists. Trans Union, 81 F.3d at 233.

164. The existence of a tradeline, sold by Trans Union in its target marketing business, is used in credit scoring models by credit grantors in their eligibility determinations. 1. Credit Scoring 165. Predictive decision tools use data and statistical analyses to develop ways of using past information to predict future outcomes. Credit scoring uses past information on credit to predict likely outcomes. (Rapaport 673/15-25; Coffman 379/16- 22.) 166. Most credit risk models use credit bureau data to observe consumer behavior. (Coffman 3825/18 -- 3826/2.) 167. Established in 1956, Fair Isaac Company (AFICO@) is the leading developer of tools used by the credit industry to determine credit risk. (Rapaport 672/25 -- 673/6.) FICO has scorecards using self-reported information from consumers' applications and credit bureau information to help credit grantors predict credit risk. (Rapaport 674/15-20; 675/18 -- 676/2.) 168. FICO collaborated with each major credit bureau to develop credit risk scoring offered by each bureau under different names: EMPIRICA at Trans Union; Experian/Fair Isaac model at Experian; and Beacon at Equifax. (Rapaport 680/8-21.) 169. A credit grantor can purchase a credit score along with a credit report from the three major credit bureaus. (Rapaport 681/19-25.) Credit bureau scores use information only available on a credit report and predict credit risk involving the general population, not just the individual credit grantor. Credit bureau TRANS UNION CORPORATION 595 Initial Decision scores can be used for decisions to grant an applicant credit, to make a preapproved credit offer, to reissue, increase or decrease a credit line, or for over-limit authorizations. (Rapaport 675/1-8; 679/11 -- 680/7; 680/25 -- 681/9.) Credit card issuers, retailers, finance companies, auto lenders, installment lenders, utilities, and student loan lenders use bureau scores. (Rapaport 682/9-16.) 170. The credit score is returned by the credit bureau from which it is purchased. FICO products return a score and four reason codes showing why an individual's score deviated from the optimal score. Credit grantors use reason codes to provide reasons for adverse action when denying credit. (Rapaport 681/21 -- 682/7; 689/14 -- 690/1.) a. FICO's Scoring Models 171. Using Trans Union's credit bureau information, FICO and Trans Union offer: EMPIRICA, a generic risk score which predicts the likelihood of delinquency on a loan or credit transaction by a consumer having at least one tradeline that was delinquent or worse in a two year period; Horizon, which predicts the likelihood a consumer will go bankrupt; and UniQuote, which predicts the likelihood a mortgage account will become delinquent. (Rapaport 690/15 -- 691/7; 692/21 -- 693/7; Tr. 800- 02.) Industry Options, a refinement of EMPIRICA, offers scores for the bank card, personal finance, installment and auto loan industries. (Rapaport 692/1-18.) 172. CX 87 is a list of all the scoring factors or adverse action reason codes that are used by EMPIRICA and its Industry Options. CX 309 contains reason codes for the Horizon bankruptcy score. (Rapaport 693/17-22.) 173. [redacted] VOLUME 129 Initial Decision b. Scoring Model Design and Terminology 174. A credit scoring model may use scorecards, variables, and score values. A scorecard contains characteristics, ranges of attributes and score values of each attribute. (Rapaport 683/20 -- 684/18.) 175. A characteristic is a piece of information describing a consumer such as the number of bank cards in a consumer's credit report. (Coffman 3828/14-23.) 176. Each characteristic can be broken into attributes; for example, age of an applicant is a characteristic, and the attribute might be that the applicant's age is between 25 and 30 years old. (Coffman 3829/7-11.) An attribute is predictive if it adds statistically significant information to the forecast. (Coffman 3830/8-24.) The statistical model will look at each attribute within each characteristic. (Coffman 3829/12-13.) 177. [redacted] c. Scorecard Segmentation 178. The first step in scoring divides a consumer's credit file into scorecards of characteristics. This applies the model to homogeneous groups of people, and is called segmentation analysis. (Rapaport 685/1-5, 17-21; 686/10-11.) 179. [redacted] 180. [redacted] 181. [redacted] 182. [redacted] 183. [redacted] TRANS UNION CORPORATION 597 Initial Decision 2. Information Used in Scoring 184. Except for owning property in the case of a mortgage loan, there is no one factor that would make an applicant eligible for credit. (Coffman 3860/20 -- 3861/25.) [redacted] a. Tradelines 185. [redacted] 186. [redacted] 187. [redacted] 188. [redacted] AEMPIRICA NOT SCORED: INSUFFICIENT CREDIT message occurs when a credit file does not contain a tradeline opened for at least six months and a tradeline updated within the last six months.@ (CX 87-A.) 189. [redacted] b. Characteristics and Attributes 190. [redacted] 191. [redacted] (1) Bank Card 192. [redacted] 193. [redacted] 194. [redacted] VOLUME 129 Initial Decision 195. Number of open bank card trades is a scorecard attribute. (Coffman 3868/16 -- 3869/10.) The predictive attribute Anumber of open bank card trades@ may be qualified by the condition that the trades must have been updated within the last six or twelve months. (Coffman 3869/6-15.) In some models the weight given to the attribute Azero@ open bank card trades and the weight given to Aone@ bank card trade differs. (Coffman 3869/16 -- 3870/9.) (2) Open Date of Tradelines 196. Scoring models look to the open date of tradelines to determine the age of the oldest trade, how long the consumer has had credit, and how long the consumer has had certain types of credit. (Coffman 3847/13-17, 23-24; 3876/14 -- 3877/20.) Age of oldest trade is an important indicator of credit performance. (Rapaport 774/6-19.) 197. [redacted] 198. [redacted] 199. [redacted] (3) Collection Account 200. [redacted] 201. [redacted] (4) Finance Trade 202. [redacted] 203. [redacted] TRANS UNION CORPORATION 599 Initial Decision (5) Mortgage 204. The number and amount of mortgages are used as predictive attributes. (Coffman 3862/5-17.) Some scoring models accord different scores for mortgage trades. (Coffman 3860/20 -- 3861/25.) 205. Some scoring models use the attribute Amortgage exact amount@ and verified mortgage trades with different loan amounts are given different scores. (Coffman 3870/10 -- 3871/4.) (6) Automobile Loan 206. Some scoring models use as predictive variables the number of auto loans, updated and verified within a certain period as an open loan. (Coffman 3865/10 -- 3866/2; 3867/13-16; 3866/17 -- 3867/12.) (7) Credit Limit and Date Verified 207. To calculate credit use, a scoring model examines the credit limit in the consumer's credit report; for the calculation, a trade would be open, timely verified, and undisputed. (Coffman 3849/12 -- 3850/8; 3882/7 -- 3884/4.) A recent analysis of a consumer's credit report is important. (Coffman 3850/10-23.) (8) Zip, Age & SSN 208. [redacted] 209. [redacted] 210. [redacted] VOLUME 129 Initial Decision 3. Information Sold by Trans Union in its Target Marketing Business Used in Credit Scoring 211. The presence or absence of a tradeline, bank card, retail account, finance loan, auto loan, collection account or a mortgage is used in credit scoring; so are the amount of a mortgage, the open dates, credit limits, and date of a tradeline. (F 181, 182, 185- 187, 189, 191, 192-207.) 4. Credit Grantors Eligibility 212. [redacted] a. Invitation to Apply 213. An invitation to apply is an application for credit, mailed to persons who may return the application to the credit grantor. (Koppin 483/25 -- 484/2; 488/24 -- 489/2; Zancola 666/2-8; 666/25 -- 667/2.) Credit grantors purchase lists of names for mailing the invitation to apply. (Koppin 486/18-20; 487/2-4.) This mailing is not a guaranteed offer of credit. (Pendleton 360/13-18; Zancola 666/2-6; Koppin 482/24 -- 483/2; 489/2-4.) The credit grantor decides to grant credit only after receipt of the completed application. (Pendleton 360/13-25; Koppin 486/23 -- 487/1.) (1) Wachovia's Invitation to Apply 214. In evaluating applications responding to an invitation to apply, Wachovia uses a credit scoring process by Wachovia and FICO. (Pendleton 360/20 -- 361/9.) [redacted]. 215. [redacted] 216. [redacted] 217. [redacted] 218. [redacted] TRANS UNION CORPORATION 601 Initial Decision 219. [redacted] 220. [redacted] (2) Discover's Invitation to Apply 221. [redacted] 222. [redacted] 223. [redacted] 224. [redacted] 225. [redacted] 226. [redacted] 227. [redacted] 228. [redacted] 229. [redacted] 230. [redacted] 231. Discover formerly used the existence of a student loan as one of the factors in deciding to extend credit but that was stopped. (Stormoen 2959/22 -- 2960/24.) VOLUME 129 Initial Decision b. Information Sold by Trans Union in its Target Marketing Business is Used in Credit Grantors' Scoring 232. In deciding credit eligibility, credit grantors use the presence of a tradeline, bank card, or mortgage. Credit grantors also use: FICO scores, the number of retail cards, and the open dates and credit limits for tradelines. (F 216, 218, 220, 224, 226- 229.) c. Prescreened Offers of Credit 233. In a prescreened offer of credit, a credit grantor mails a firm offer of credit to a person. (Koppin 482/21-23, 488/20-23; Pendleton 357/23 -- 358/3; 359/1-6.) 234. Credit grantors obtain names of persons to consider for the prescreening from an outside list or by an Aextract@ of names from a consumer reporting agency. (Koppin 484/14-19; 485/2-7; Pendleton 357/23 -- 358/3; McCoy 495/14-24.) 235. [redacted] 236. [redacted] 237. Whether from an outside list or from an extract, the names are processed by a consumer reporting agency so that credit criteria may be applied. (Koppin 489/5-9; Pendleton 358/19 -- 359/6.) Credit grantors use the three national consumer reporting agencies to conduct prescreening. (Koppin 483/14-16; Pendleton 359/24 -- 360/2; Zancola 668/2-6; McCoy 495/14-20.) 238. [redacted] 239. [redacted] TRANS UNION CORPORATION 603 Initial Decision (1) Wachovia's Prescreen 240. [redacted] 241. [redacted] 242. [redacted] 243. [redacted] 244. [redacted] 245. [redacted] 246. [redacted] 247. [redacted] 248. [redacted] 249. [redacted] 250. [redacted] 251. [redacted] 252. [redacted] 253. [redacted] 254. [redacted] 255. [redacted] VOLUME 129 Initial Decision (2) First Card's Prescreen 256. [redacted] 257. [redacted] 258. [redacted] 259. [redacted] 260. [redacted] 261. [redacted] 262. [redacted] 263. [redacted] 264. [redacted] 265. [redacted] 266. [redacted] 267. [redacted] 268. [redacted] 269. [redacted] 270. [redacted] 271. [redacted] TRANS UNION CORPORATION 605 Initial Decision (3) Chase's Prescreen 272. [redacted] 273. [redacted] 274. [redacted] 275. [redacted] 276. [redacted] 277. [redacted] 278. [redacted] 279. [redacted] 280. [redacted] 281. [redacted] 282. [redacted] 283. [redacted] 284. [redacted] 285. [redacted] 286. [redacted] (4) Northern Trust's Prescreen Process 287. [redacted] VOLUME 129 Initial Decision 288. [redacted] 289. [redacted] 290. [redacted] 291. [redacted] 292. [redacted] 293. [redacted] 294. [redacted] d. Information Sold by Trans Union in its Target Marketing Business is Used in Credit Grantor's Prescreening 295. Credit grantors use a tradeline, bank card, finance company trade, mortgage, refinanced trade, and a student loan in prescreening. (F 243-245, 248-252, 258, 259, 269, 278-282, 288, 290, 291.) Credit grantors also use FICO scores, type of tradeline, and credit limit and open date of tradelines in prescreening. (F 242, 247, 253, 254, 264, 268, 275, 277, 283, 289.) F. Consumers' Privacy and Disclosure of Consumer Reports 1. Survey of Consumer Privacy 296. Complaint counsel offered a survey to assess consumer attitudes regarding the use of information derived from CRONUS and from credit reports to compile marketing lists. (Mazis 1080/10-18.) 297. Dr. Michael Mazis wrote the questionnaire and determined the methodology for the privacy survey. (Mazis 1081/21-24.) International Communications Research (AICR@) TRANS UNION CORPORATION 607 Initial Decision conducted the survey using Dr. Mazis' questionnaire and methodology. (Mazis 1082/25 -- 1083/3; Waldeck 1057/1-8; 1057/14-20; 1068/8-23.) ICR is qualified to conduct such a consumer survey. (Waldeck 1051/24 -- 1052/3.) 298. Ms. Karen Waldeck collected and reported the data for Dr. Mazis' survey (CX 274) and prepared a final report. (CX 274; Waldeck 1054/3-5; 1054/20 -- 1055/1; 1063/1-9; 1064/16 -- 1065/6.) 299. Dr. Mazis used a telephone survey to obtain a nationally representative sample. (Mazis 1082/22 -- 1084/10.) ICR conducted the survey by telephone. (Waldeck 1057/3-7.) 300. ICR selected the households to be surveyed through random digit dialing. (Waldeck 1059/9-12; Mazis 1084/5-6, 24- 25; CX 274-C; CX 354-A.) 301. In order to ensure a random sampling within each household, ICR used the Alast birthday@ method requiring interviewers to request to speak with the individual over the age of 18 who had the most recent birthday in each household. (Waldeck 1059/25 -- 1060/8; Mazis 1088/16-18; CX 274-C; CX 354-A.) 302. A pre-test of the questionnaire was conducted prior to conducting the survey. (Waldeck 1058/24 -- 1059/1; CX 354-A.) The pre-test resulted in only minor changes to the questionnaire, and demonstrated significant variation in responses indicating that the questions were understood. (Mazis 1087/3-14; Waldeck 1068/16-23.) The interviewers conducting the pre-test and the survey had no knowledge that the FTC was the client for whom the survey was conducted nor were they aware of the purpose for which it was conducted. (Waldeck 1061/12-15.) VOLUME 129 Initial Decision 303. After collecting data, ICR conducted a test to verify the data and that the interview was properly conducted. (Waldeck 1064/4-15; Mazis 1089/11-16; CX 354-A.) ICR's validation test requires calling back at least 20% of the individuals surveyed and confirming their responses to selected questions. (Waldeck 1064/4-16.) A 20% validation test is considered standard within the survey industry. (Mazis 1089/14-16.) 304. The sample size of the survey ICR conducted was 1,002. (Waldeck 1060/12-16; CX 354-A.) This was projectable to United States households with a margin of error of plus or minus 3%. (Mazis 1089/17-19; CX 354-A.) 305. The questionnaire stated that the survey concerned marketing lists. (Mazis 1091/6-10; CX 274-Z-34.) Participants could indicate that they had Ano opinion@ in response to a question. (Mazis 1092/11-19; CX 274-Z-34.) 306. Participants were provided with potential responses from Avery acceptable@ to Asomewhat acceptable@ and Asomewhat unacceptable@ to Avery unacceptable,@ and definitions of each of these terms. (Mazis 1092/2; 1092/24 -- 1093/5; CX 274-Z-34.) 307. Interviewers asked questions regarding types of consumer information used to create marketing lists. (Mazis 1095/14 -- 1096/1; 1097/10-21; CX 274-Z-34 -- Z-35.) For example: ADo you think that it is very acceptable, somewhat acceptable, somewhat unacceptable, very unacceptable to sell to companies a list of individuals based on whether or not they have a second mortgage?@ (CX 274-Z-35.) 308. Fifty percent of the respondents were first asked: ADo you think that it is very acceptable, somewhat acceptable, somewhat unacceptable, very unacceptable to sell to companies a list of individuals based on whether or not they have a mortgage?@ followed by the question: ADo you think that it is very acceptable, somewhat acceptable, somewhat unacceptable, very unacceptable to sell to companies a list of individuals based on the state that TRANS UNION CORPORATION 609 Initial Decision they live in?@ (Mazis 1095/13-20; CX 274-Z-34.) For the other 50% of the interviewees the order of these questions was reversed. Consecutive questions on a similar topic are rotated to prevent Aorder bias.@ (Mazis 1094/13 -- 1095/10; CX 274-Z-34.) 309. Interviewers asked similar questions on a random basis, with regard to selling lists based on the following: your estimated income whether or not you have an auto loan or lease whether or not you have a second mortgage type of mortgage you have the approximate amount of your mortgage estimated amount of your home equity whether or not you live in an apartment whether or not you have a credit card whether or not you have a premium bankcard (Mazis 1097/23 -- 1098/8; CX 274-Z-35.) 310. Dr. Mazis prepared charts, CX 354-A -- D, summarizing his findings of the survey. (Mazis 1104/2-4; 1105/24 -- 1106/6; CX 354-A-D.) 311. The results were as follows (stated in terms of the combined percentage of respondents indicating somewhat or very unacceptable to sell marketing list based on): VOLUME 129 Initial Decision Approximate amount of mortgage 72.8% whether or not you have a second mortgage 72.5% estimated amount of home equity 70.1% estimated income 68.6% type of mortgage 66.1% whether or not you have a mortgage 65.4% whether or not you have an auto loan or lease 64.1% whether or not you have a premium bank card 64.0% whether or not you have a credit card 60.9% whether or not you live in an apartment 55.5% state of residence 46.1% (CX 354-C; CX 274-Z-25 -- Z-26.) 312. These results extend from a high of 72.8% (regarding the approximate amount of a mortgage) to a low of 46.1% (regarding state of residence). (Mazis 1109/2-4; 1106/25 -- 1107/3.) 313. Next, the interviewers asked questions to determine consumer attitudes towards credit report data as a source of information for marketing lists. (Mazis 1080/15-24; 1099/17 -- 1100/3.) The next question in the survey stated: AWhen companies put together mailing or telemarketing lists to sell products to consumers they sometimes use information from consumers' credit reports. Have you heard of a credit report?@ (Mazis 1099/22 -- 1100/3; CX 274-Z-35.) TRANS UNION CORPORATION 611 Initial Decision 314. This filtering question insured that only respondents familiar with a credit report answered the next question. (Mazis 1100/13-17.) Eighty-nine point five percent of the respondents had heard of a credit report. (Mazis 1104/9-11; CX 274-Z-31; CX 354-B.) Ten point one percent of the respondents had not heard of a credit report and were not asked the next question. (Mazis 1100/10-12; 1101/3-5; CX 274-Z-31; CX 354-B.) 315. The following question stated: AThinking about information from consumers' credit reports, do you think it is very acceptable, somewhat acceptable, somewhat unacceptable, or very unacceptable to use information from credit reports to put together marketing lists to sell products to consumers?@ (CX 274-Z-36.) 316. Of those answering this question, 76% found this practice unacceptable. (Mazis 1104/22 -- 1105/3; CX 274-Z-32; CX 354- B.) This means that 68.1% of the survey's respondents found the use of credit report information for the compilation of marketing lists to be unacceptable. (Mazis 1105/13-19; CX 354-B.) 317. Based on these results, Dr. Mazis determined that consumers have a strong privacy interest in the use of information from their credit reports. (Mazis 1107/23-25.) 318. Consumers also have a strong privacy interest regarding the use of specific types of information derived from CRONUS for the purpose of compiling marketing lists. (Mazis 1109/20-25.) 319. The conclusion that the Fair Credit Reporting Act protects consumers' privacy interests by prohibiting the unauthorized dissemination of their credit histories to third-party marketers is supported by the results of the consumer survey performed by ICR.

VOLUME 129 Initial Decision 2. Government Interest in Consumers' Privacy 320. Professor Joel Reidenberg testified to the privacy protections afforded to consumers' credit information under the Fair Credit Reporting Act. (Reidenberg 964/5-14.) 321. Fair information practices are a set of standards that are applied to the treatment of personal information. These standards focus on ensuring fairness to the individual in the treatment of his or her personal information. (Reidenberg 947/7-13.) Personal information is Ainformation that relates to an identified or identifiable individual.@ (Reidenberg 949/20-21.) 322. Fair information practices espouse the principle that personal information should be collected openly for use for a specified purpose. (Reidenberg 947/14-19.) Once collected, personal information should only be used for the purpose for which it was collected, unless the individual has the opportunity to participate in the decision to allow other uses. (Reidenberg 947/18-24.) 323. Congress has enacted legislation regarding the use of personal information. (Reidenberg 953/24 -- 954/11.) The Fair Credit Reporting Act imposes legal obligations upon the credit reporting sector that are consistent with fair information practices standards. (Reidenberg 955/22 -- 956/18.) 324. Professor Reidenberg analyzed the legislative history of the FCRA. (Reidenberg 974/8-20; 986/10-13.) 325. In enacting the FCRA, Congress intended to ensure confidence in the credit reporting system. Congressional concern was due to the Arampant@ disclosures unrelated to the extension of credit of information held by consumer reporting agencies. (Reidenberg 958/2-15.) TRANS UNION CORPORATION 613 Initial Decision 326. Prior to the FCRA, the media publicized stories regarding non-credit related disclosures of information by consumer reporting agencies. These news accounts were considered by Congress in formulating the FCRA. (Reidenberg 960/19 -- 961/5.) Congress also considered testimony by Professor Alan Westin regarding examples of such disclosures. (Reidenberg 961/22 -- 962/5.) 327. The FCRA addressed non-credit disclosures by requiring Apermissible purposes@ for the disclosure of information held by consumer reporting agencies. (Reidenberg 955/22 -- 956/4; 956/9- 11.) 328. Under the FCRA, a permissible purpose for disclosure exists if a consumer authorizes the disclosure. (Reidenberg 975/3- 8.) 329. The 1996 amendments to the FCRA authorize disclosure of credit report information for prescreening if: (1) a firm offer of credit is made with limited post-screening to verify identity and continued compliance with the grantor's selection criteria; (2) the credit grantor's selection criteria are identified; (3) the prescreening has an opt-out system; and (4) the credit grantor provides notice to the individual. (Reidenberg 967/2-23; 971/5- 11.) 330. Under the amended FCRA, the permissible purpose for disclosure of credit reports in prescreening extends to firm offers of credit or insurance. Oscar Marquis, Trans Union's general counsel, rejected proposed prescreen mailings of a sweepstakes/magazine subscription promotion and other noncredit offers, that would have used Acredit criteria@ because he did not consider them to be offers of credit. Prior to October 1, 1997 such offers would have been target marketing. (Marquis CX 188 at 167/24 -- 171/22.) VOLUME 129 Initial Decision 331. In the amended FCRA, the right to opt-out of prescreening provides consumers with a right to participate in the decision to use their information for firm offers of credit. (Reidenberg 976/11-16.) Use of credit report data for a prescreen firm offer of credit is consistent with the purpose of collecting the information. (Reidenberg 966/19 -- 967/1.) 332. Professor Reidenberg testified that the maintenance of documentation of criteria is unique to the FCRA (Reidenberg 971/2-16), and that the opt-out requirements are the Aonly situation in the private sector where . . . a third party has to tell you where they got your name.@ (Reidenberg 967/23 -- 968/1.) 333. Trans Union does not provide target marketing lists on New Hampshire residents because the state credit reporting statute requires prior consent. (Marquis CX 188 at 148/11-19; N.H. Rev. Stat. Ann. ' 359.) 3. Opt-Out 334. AOpt-out@ refers to the procedure whereby consumers can request that their names be removed from target marketing lists, including direct mail and telemarketing lists. (Davis 210/20-25.) 335. The target marketing list industry consists of list providers, list managers, list brokers, and list users/renters/purchasers. (Miller 3008/5-25; 3034/1-9, 17-24; 3034/25 -- 3035/1-25.) Consumers in general are unaware that their names and addresses are used to compile lists since the majority of all lists are developed as a result of secondary uses of information. (CX 151-A.) 336. Most consumers are unaware that they can opt-out of target marketing lists. (Beales 3669 -- 3670/14; Westin 3639/8- 19.) Of consumers surveyed in 1991 and 1996, 44% stated that they were aware of procedures that would allow them to remove their name from direct mail lists. (TU-88 at FTC B0003194.) TRANS UNION CORPORATION 615 Initial Decision 337. The Direct Marketing Association (ADMA@) maintains two opt-out files for use by its member companies. Consumers who request that DMA place their names on the opt-out lists must reregister after five years. (Cleary 3081/15 -- 3082/14.) 338. DMA member companies are not required to use the DMA opt-out files, and can nevertheless make available on their mailing and telemarketing lists the names of those consumers who have requested to opt-out at the DMA. (Cleary 3082/17 -- 3083/6; Nusbaum 2905/24 -- 2906/25.) 339. DMA has a small advertising budget to notify consumers of their right to opt-out and relies on consumer affairs columnists, government, and interview news programming to notify consumers of DMA opt-out availability. (Cleary 3142/10-20.) 340. List brokers and list managers do not notify consumers about the opt-out process. They rely on the opt-out service provided by the DMA. (Miller 3073/15-19.) 341. Respondent's own witness was unaware of any list providers that require their clients to notify consumers of the optout process, except where required by law in the context of prescreening. (Hinman 2238/1-17.) 342. Trans Union's opt-out program complies with Section 604(e)(5) of the Fair Credit Reporting Act, which requires consumer reporting agencies to notify consumers of their right to opt-out of firm offers of credit. (Botruff 2065/22 -- 2066/9; TU- 203.) Trans Union provides consumers with the choice of opting out for a two-year period or opting out permanently -- the percentage of those who choose to opt-out permanently is in the high nineties. (Botruff 2063/1-7.) VOLUME 129 Initial Decision 343. Trans Union does not require its list clients to notify consumers of their right to opt-out of target marketing lists other than on prescreen. Trans Union sales people who review nonprescreen promotional pieces are not aware of any such mailings that carry an opt-out notification. None of the non-prescreen promotional mail in evidence contains an opt-out notification that informs the consumers that their names are on a Trans Union list nor how to opt-out of Trans Union's target marketing lists. (Dexter 1276/15 -- 1277/1; Clifton 1916/25 -- 1917/23; CX 11; CX 18-A-J; CX 19-A-G; CX 20; CX 24-A-M; CX 25; CX 26-A-I; CX 26-A-E; CX 32-A-B; CX 35-A-D; CX 36 A-D; CX 39-A-D; CX 40-A-H; CX 256; TU-175; TU-176.) There is no direct, credible evidence of the success rate of the opt-out actually stopping direct mail or telemarketing calls. 344. DMA's Mail Preference Service consists of names of 3.5 million consumers who do not want their names to be included on any target marketing mailing lists. (Tr. 3081-82, 2203-04.) 345. DMA's Telephone Preference Service includes the names of 750,000 consumers who do not want their names included on any telemarketing lists. (Tr. 3081-82, 2204.) 346. Trans Union's APrivacy Protocol,@ which appears on Trans Union's website, states that Trans Union Abelieve[s] consumers should have the right to make informed decisions about the use of their personal data, including the right to be removed from direct marketing lists.@ (TU-50; TU-51.) PerformanceData employees, including Jay Frank, Senior Vice President in charge of PerformanceData, indicated that they had either never seen the protocol, or were unfamiliar with its substance. (Porretto 1620/2-13; Dixon 293/14-16; Dexter 1286/2- 14; Frank CX 186 at 64/6 -- 65/11.) 347. CRONUS indicates whether a consumer wants to be excluded from any direct mail or telemarketing list. (Tr. 3677- 78.). In January 1997, CRONUS contained the names of 5.1 million consumers who opted out. (Tr. 2060.) TRANS UNION CORPORATION 617 Initial Decision 348. Trans Union placed an advertisement in the September 22, 1997 edition of the USA Today praising the benefits of direct mail advertisements and notifying consumers that information in Trans Union's consumer files may be used in connection with credit or insurance transactions that are not initiated by the consumer and of the address and toll-free telephone number for consumers to use to opt-out, but warning that the opt-out does not guarantee that the consumer won't receive direct mail offers from other sources. (TU-203.) 349. In February 1998, CRONUS contained the names of 6.1 million consumers who opted out. (Tr. 2066.) 350. PerformanceData does not include in the List Master File the name of any consumer in CRONUS who has opted out. (Tr. 3677-78.) 351. PerformanceData obtains from ACXIOM a list of telephone numbers to be included in the List Master File. (Tr. 3678.) The list of telephone numbers which PerformanceData receives from ACXIOM does not include unlisted telephone numbers or the telephone number of consumers whose names appear on DMA's Telephone Preference Service. (Tr. 3678-79.) 352. PerformanceData and the list providers who testified voluntarily do not include on any direct mailing or target marketing list the names of consumers who opt-out. (Tr. 3677-79, 2205, 2905-07, 2996-98, 3080-83, 3379-80.) G. The Direct Marketing Industry 353. There are 15,000 companies engaged in supplying consumer direct marketing lists. (Tr. 3337.) VOLUME 129 Initial Decision 354. The total gross annual revenue of companies supplying consumer direct marketing lists and file enhancement data may be $ 1.5 billion. (Tr. 3320-21.) Five companies' revenues represent 40% of this $ 1.5 billion market: R.L. Polk & Company, ACXIOM, Metromail Corporation, First Data Solutions and ABI/DBA. (Tr. 3321-22.) PerformanceData has a 2% share of this market. (Tr. 3322.) 355. There are 30,000 direct marketing lists. (Tr. 2213.) AThe Direct Marketing List Source@ identifies more than 19,000 lists in 212 markets. (Tr. 2429, 2432.) H. Consent Decree With TRW 356. Prior to January 1993, Experian's predecessor, TRW Information Systems and Services (ATRW@), conducted its target marketing business in the same manner as PerformanceData's business which is at issue here. (Tr. 3287-88.) 357. On January 14, 1993, the Commission and TRW entered into an amendment to a previously-entered consent decree which provides that TRW is allowed to extract the following information from its consumer reporting database for target marketing: name, telephone number, mother's maiden name, address, zip code, year of birth, age, any generational designation, social security number, or substantially similar identifiers, or any combination thereof. (TU-109; Tr. 3286-87, 3293.) Under the consent decree, Experian (TRW), unlike Trans Union, does not extract open dates, high credit amounts, auto loan expiration dates, and loan dates from its consumer reporting data base for use in target marketing lists. (F 146.) 358. Experian's Fall 1997 Catalog of Consumer Lists and Processing Services advertises a AConsumer Database@ composed of records on 161,235,677 consumers. (TU-112.) [redacted] TRANS UNION CORPORATION 619 Initial Decision 359. Experian's Fall 1997 Catalog of Consumer Lists and Processing Services advertises a 30-day new mover list totaling 950,000 consumers a month and a 14-day new mover list totaling 550,000 consumers a month with a 12-month total of 11,406,804 consumers. (TU-112.) 360. At the time that TRW entered into the consent decree with the Commission, TRW was using credit attributes from its consumer reporting database, aggregated to a geographic area defined by a nine digit postal zip code (Azip-plus-four@), in connection with its target marketing business without objection by FTC representatives. (Tr. 3290-93, 3305-07.) This information was made available by TRW for statistical modeling and as a data attribute. (Tr. 3306.) 361. There are 20 million zip-plus-four's in the U.S. A zipplus-four includes 5-15 households. (Tr. 2688.) I. The 1996 Amendments to the FCRA 362. On September 30, 1996, Congress passed amendments to the FCRA effective September 30, 1997. These amendments were included in the Omnibus Consolidated Appropriations Act of 1997. (CX-167-A.) 363. Effective September 30, 1997, ' 603 of the FCRA was amended to include a definition of Afirm offer of credit or insurance.@ FCRA, ' 603(1)(1997).

364. Section 604(c) of the amended FCRA created a new permissible purpose for transactions not initiated by consumers, i.e., prescreening; a provision addressing the information a credit grantor could receive on consumers in prescreening; and provisions for Aopt-outs.@ FCRA, '' 604(c) and (e)(1997). VOLUME 129 Initial Decision 365. Under the amended FCRA, when a consumer responds and accepts a firm offer of credit, a credit grantor can then review the consumer's credit report and deny the consumer the credit offered if the consumer's credit report reflects something negative about the consumer's credit performance. FCRA, ' 603(1)(1997). 366. The Senate passed S. 650, a bill amending the FCRA. (TU-214.) The bill passed by the Senate included the amendments to '' 603 and 604 discussed above; however, the Senate bill also included a provision affirmatively authorizing the use of information from consumer files in connection with direct marketing transactions by defining the information the Senate considered to be outside the scope of Aeligibility information@ covered by the FCRA as Aname and address of consumer and other information that would not disclose the credit payment history, credit limit, credit balance, or any negative information pertaining to the consumer.@ (TU-214; Tr. 2550-56, 3513-17.) 367. The FCRA amendments passed by Congress do not include the provision in the bill passed by the Senate which affirmatively authorized the use of information from consumer files in connection with direct marketing transactions. (CX-167- A[B] TU-213.) On September 30, 1996, the following colloquy took place between Senator Bryan and Chairman Alfonse D'Amato (R-NY):

Mr. BRYAN. Mr. President, I wish to engage my esteemed colleague Chairman D'Amato in a brief colloquy to clarify two items pertaining to the Fair Credit Reporting Act (FCRA) amendments contained in the H.R. 4278, the Omnibus Consolidated Appropriations Act of 1997. First, the House of Representatives in negotiations over the weekend deleted a Senate-approved measure which would have codified the permissibility of direct marketing under the FCRA. The deletion leaves the law silent on this issue, retaining the status quo. The House action does not reflect any congressional intent regarding the extent to which direct marketing is permissible under FCRA.

TRANS UNION CORPORATION 621 Initial Decision The second item relates to a requirement imposed under section 609 of the FCRA for personnel being accessible to consumers. The requirement that personnel be available under normal business hours is not intended in any manner to interfere with the use of automated menu telephone systems which provide the consumers with a range of options. The standard is satisfied as long as the system provides a consumer the option to speak to a live operator at some point in the audio menu. Does the chairman confirm these understandings? Mr. D'AMATO. Yes, Senator Bryan. I agree with your assessment on these points.

(TU-213; Tr. 2555-56.) III. ANALYSIS A. Target Marketing Target marketing involves selling goods and services directly to consumers by mail or telephone. Consumers are picked by geographic, credit-related, demographic traits. (F 5.) Marketers want to limit the number and type of persons solicited to improve the response rate in order to increase profits. Trans Union's target marketing lists6 use consumer information from CRONUS. On October 1, 1997, Trans Union stopped selling some target marketing lists and data in nonprescreen promotions.7 On that day amendments to the FCRA 6 Trans Union sold its lists under different names -- TransMark, Trans Union Lists and PerformanceData. (F 4.) 7 F 112 - 113.

VOLUME 129 Initial Decision became effective, providing for civil penalty liability. 15 U.S.C. ' 1681s(a)(2). Trans Union ceased selling those products for use other than firm offers of credit because of potential civil penalty liability for violations of the FCRA under the new amendments.8 Trans Union's caution there seems to conflict with its perseverance in this case. There would be no such risk if the products being sold to target marketers were not consumer reports.

Four weeks prior to trial in this matter, Trans Union revised its target marketing database,9 involving data on bank cards, finance trades, dollar amounts and open dates. In late-January 1998 Trans Union created a separate database (known as PerformanceBase or RelationBase) for dollar amounts, open and closed dates, loan types, home equity, and SOLO.10 The data on PerformanceBase is now available only for firm offers of credit. These revisions do not vitiate the charges in the complaint.11 Trans Union's target marketing product is the Master File.12 Until January 1998, when Master File was changed to comply with the October 1, 1997 changes to the FCRA, the Master File required at least two open tradelines, one of which must have 8 F 114.

9 F 32, 33, 112.

10 F 116. CX 333, 334 and 335 set out the contents and quantities for the List Master file (CX 333), the Apre-approved file (in RelationBase)@ (CX 334), and a comparison of the two (CX 335).

11 Trans Union continues to sell data for target marketing from its Master File (F 25, 115), and has not argued abandonment. Because of the possibility of recurrence, abandonment of a practice is not a basis to avoid entry of a Commission order. Official Airline Guides v. FTC, 630 F.2d 920, 928 (2d Cir. 1980).

12 This database was referred to by the Court of Appeals as the Abase list.@ Trans Union, 81 F.3d at 229, 232, 234. TRANS UNION CORPORATION 623 Initial Decision been verified within 12 months.13 The January 1998 Master File requires one tradeline.14 Consumer files in the Master File may have data on a bank card, auto loan, or mortgage, but every file must have at least two (or, as of January 1998, one) qualifying tradelines. The individual consumer information that comes from CRONUS may include the presence, amount, dates, and type of: auto loan, bank card, premium bank card, department (retail) card, upscale retail card, finance loan (mortgage, auto), student loan, and head of household.

In addition to its Master File, Trans Union sells other target marketing products that use consumer credit information from CRONUS. Trans Union sells monthly and weekly lists of persons who have been issued new credit -- in a bank card, mortgage, auto, retail card, upscale retail card, and finance company -within the last 30 to 90 days. Trans Union also sells its income estimator (ATIE@) showing a person's income. SOLO shows 35 credit characteristics; EVAL provides information about home value and home equity; PIC uses data from CRONUS, and external survey data, to predict the likelihood of purchase of such products as mutual funds and IRA accounts. Trans Union also sells TransLink, a product that Areverse appends@ a consumer's name and address to a third-party credit card account number obtained by a merchant at the point of sale and supplied to Trans Union for the purpose of discovering the card account holder's identity and address. Although TIE, E-Val, SOLO and PIC are now available only for firm offers of credit, 13 F 28, 31, 126. ATradeline@ refers to information supplied by a credit grantor about a consumer's account: the account number, type of account, date opened, high credit, current balance, payment history. F 22, 28. 14 F 33.

VOLUME 129 Initial Decision Trans Union continues to sell its TransLink product to those without a permissible purpose to receive consumer reports. Each of Trans Union's target marketing products is a consumer report because it discloses information from Trans Union's consumer reporting database that is also used by credit grantors for credit eligibility determinations. This is true of the element common to all Trans Union target marketing products -existence of tradeline -- as well as the specific elements included in list selects and in Trans Union's specialized products such as SOLO and TIE.

B. Credit AEligibility@ Information Credit grantors make eligibility determinations on applications submitted by consumers. Credit grantors also seek to increase their business by extending credit to new account holders through their own marketing efforts, including: prescreened offers of credit; non-prescreened offers, known as Ainvitations to apply@; and in-branch or take-one applications for credit.15 In each case, credit grantors use their own standards to evaluate the creditworthiness of applicants and other consumers to whom the credit grantor wishes to extend an unsolicited offer of credit. In many cases, credit grantors make use of credit scoring models to assist them in their eligibility determinations. Credit scoring is the most direct evidence that Trans Union itself intended the same elements that it uses in its target marketing lists also to be collected and used for eligibility determinations. 1. Credit Scoring The Court of Appeals stated that, Aon remand, if the FTC wishes to classify existence-of-tradeline information as a consumer report, it must gather evidence that indicates that Trans Union intended the mere existence of a tradeline . . . to serve as a 15 F 212.

TRANS UNION CORPORATION 625 Initial Decision factor in credit-granting decisions . . .@ 81 F.3d at 233. The trial record proves Trans Union's intent because it establishes that credit scoring products developed by Trans Union use number of tradelines in credit scoring based on information from CRONUS.16 Credit scoring, including Trans Union's own scoring, use factors -- such as presence of bank cards, auto loans and mortgages -- also used by Trans Union in target marketing lists.17 Credit scoring weighs factors in a credit grantor's decision to approve credit, evaluating the risk posed by each applicant. With computers, credit scoring has become more prevalent. It uses information (from loan applications or credit bureau records) to predict a person's future debt repayment performance.18 ACustom@ scoring helps credit grantors by identifying the Agood@ and Abad@ performing consumer accounts of that company by statistical analysis. Custom scoring also uses information from consumers' applications and credit bureau information.19 The three national credit bureaus sell credit history scores to help lenders assess risk on loans.20 Credit bureau scoring systems use only credit bureau data.21 Credit grantors receive a numerical score for each applicant, based on the individual's credit history. The higher the score, the better the credit risk.22 16 F 186.

17 F 190-191, 211.

18 F 167.

19 F 167.

20 F 168-169.

21 F 169.

22 F 170.

VOLUME 129 Initial Decision Trans Union, in partnership with Fair Isaac Co. (FICO), the leading credit scoring company, developed: EMPIRICA, Trans Union's credit scoring predicting delinquency on a loan or credit; Horizon, predicting bankruptcy; and UniQuote, predicting the likelihood that a mortgage will become delinquent.23 a. Existence of Tradelines Trans Union's EMPIRICA, Horizon, and UniQuote scoring models will not score an applicant whose credit file has no tradeline or at least one tradeline updated in the past six months.24 If there is no tradeline, the model cannot calculate a credit score.25 Most credit grantors will not approve an applicant for credit where there is no score at the credit bureau.26 This evidence alone establishes Trans Union's intent that a tradeline is a factor in credit eligibility.

b. Number of Tradelines Credit scoring involves identifying, by statistical analysis, characteristics in consumers' credit histories that predict credit performance.27 A Acharacteristic@ is information from a 23 F 167-168, 171. Industry Options, a refinement of EMPIRICA, offers scores for the bank card, personal finance, installment and auto loan industries. Trans Union also offers DELPHI, a bankruptcy model, developed with another modeling company, MDS. F 173.

24 F 186-187; CX 89-S. DELPHI requires at least one qualifying tradeline to score a consumer file. F 185. 25 F 188. An AEMPIRICA NOT SCORED: INSUFFICIENT CREDIT message occurs when a credit file does not contain a tradeline opened for at least six months and a tradeline updated within the last six months.@ (CX 87- A.) 26 F 189.

27 F 174-177, 179.

TRANS UNION CORPORATION 627 Initial Decision consumer's credit history, such as the number of bank credit cards in a consumer's credit report.28 Variables (Aattributes@) are assigned weights, based on credit history and statistical analysis.29 For the characteristic Aopen bank cards,@ attributes are the number of bank cards, such as Azero,@ Aone,@ A2-3,@ A4-6,@ A7 or more.@ The first step in scoring puts a consumer's credit file into a scorecard based on information in the consumer's file. A scorecard is tailored to homogenous people.30 EMPIRICA uses ten scorecards. Each scorecard contains characteristics, attributes and weights.31 Scoring assesses the consumer's credit history to gauge the number of tradelines. EMPIRICA's ten scorecards are Aeach tailored to a distinct consumer group based on their credit behavior.@32 As stated in Trans Union's EMPIRICA brochure:33 EMPIRICA selects one of 10 different scorecards which best reflects the consumer's credit history to calculate the EMPIRICA score. Scorecard selection is based upon such credit information as:

* Number of tradelines * Age of oldest tradeline * Age of newest tradeline 28 F 175.

29 F 176, 180.

30 F 178.

31 F 181, 183.

32 Id.; CX 88-B.

33 CX 88-E.

VOLUME 129 Initial Decision Of the ten scorecards, two are for applicants with serious delinquency, and one is for applicants with a single tradeline; other applicants have two or more tradelines.34 EMPIRICA distinguishes between consumers who have one tradeline and those who have two or more. It sends the credit files of those consumers who have no serious delinquency and a single tradeline to a different scorecard than consumers who have two or more tradelines. A consumer's credit is scored differently based on number of tradelines and on the age of the oldest and newest tradelines. EMPIRICA will not score an individual with no tradeline. The use of number and age of tradelines in Trans Union scoring products is evidence that Trans Union intended that these elements serve as factors in credit decisions.35 c. Other Characteristics The characteristics on a scorecard determine the applicant's score. Scorecard predictive characteristics use the same information that Trans Union uses to assemble its target marketing lists.36 EMPIRICA (for auto loans, bank cards, finance and installment loans), Horizon (bankruptcy), and AUniQuote@ (mortgage loan) use as predictive characteristics scores: * [redacted]37 * [redacted]38 34 F 181.

35 Trans Union promotes EMPIRICA for credit decisions, setting credit limits, assigning interest rates, repricing accounts and improving collections. CX 88-C.

36 CX 93, F 190.

37 F 191-194.

TRANS UNION CORPORATION 629 Initial Decision * [redacted]39 * [redacted]40 EMPIRICA provides adverse action codes that explain the score. The credit grantor can then supply the reasons to those applicants who are declined for credit. Among the EMPIRICA explanations are: ALack of recent information on bank card accounts or lack of bankcards@ and Alack of recent information on auto loans or lack of auto loans.@41 For Horizon, reason codes include: ANo mortgage loans reported.@42 Factors used by Trans Union in its target marketing lists are also used in scoring models. Michael Rapaport of FICO and John Coffman of May & Speh, another credit scoring firm,43 testified that they use information that Trans Union makes available in its target marketing lists. Mr. Rapaport testified that these factors include: presence or absence of a retail account, an auto loan, a mortgage loan, as well as a consumer's homeowner status. Scoring models also consider the open date of the newest tradeline, finance loan, and auto loan.44 Dr. Coffman testified that 38 F 191, 202-203.

39 F 191, 196-199.

40 F 200-201.

41 CX 87.

42 CX 309.

43 Rapaport 672/11-22, 674/12-14; Coffman 3795/20 -- 3796/23, 3798/1- 18, 3800/4, 6, 13, 20-21, 3803/13 -- 3804/22. 44 F 191.

VOLUME 129 Initial Decision number and amount of mortgages are used as predictive attributes in credit risk scoring. He also testified that verified mortgage trades in different loan amounts are given different scores.45 Dr. Coffman has seen auto loans used as a predictive characteristic.46 He testified that, to calculate credit, scoring models examine the credit limit in a consumer's credit report.47 The use of these factors in target marketing lists and in Trans Union credit scoring establishes Trans Union's intent.48 In its opinion, the Court of Appeals stated that On remand, if the FTC wishes to classify existence-oftradeline information as a consumer report, it must gather evidence that indicates that Trans Union intended . . . or, of course, that someone used or expected [existence-oftradeline information] to be used [as a factor in creditgranting decisions].

Trans Union, 81 F.3d at 233. The evidence proves that credit grantors consider existence of tradeline (and other factors that Trans Union uses in its target marketing lists) to extend consumer credit. Credit grantors extend credit to new accounts by prescreened offers of credit and invitations to apply. 2. Prescreening The FCRA requires that consumer reporting agencies furnish consumer reports only to those with a Apermissible purpose@ to obtain a report; prescreen offers are an exception to the requirement that, for a permissible purpose to exist, the consumer 45 F 204-205.

46 F 206.

47 F 207.

48 [redacted] TRANS UNION CORPORATION 631 Initial Decision must have initiated the transaction. Trans Union, 81 F.3d at 234. Prescreening is the process whereby a credit grantor extends credit to consumers who meet credit standards. A prescreened list is a list of consumer reports because it shows that each consumer meets criteria for creditworthiness. The Commission interpreted the FCRA to permit prescreening if the credit grantor agrees in advance that each consumer whose name is on the list after prescreening will receive an offer of credit.49 The Court of Appeals, in its opinion remanding this case, indicated that prescreening was reasonable and noted that Aprescreening and the guaranteed offers of credit it spawns can only take place through the use of consumer reports . . . .@ Trans Union, 81 F.3d at 234 (emphasis in original). Prescreening is now sanctioned by the FCRA in amendments effective October 1, 1997. Section 604(c), 15 U.S.C. ' 1681b(c). The mail offer of a pre-approved credit card is an unsolicited offer to consumers; the consumer has not applied for credit, and the credit grantor does not have application information. The credit grantor decides upon credit criteria to extend the Afirm offer@ of credit to creditworthy consumers.50 These criteria result in an offer of credit and are Aintended . . . to serve as a factor in credit-granting decisions . . . .@ Trans Union, 81 F.3d at 233. Credit grantors deciding to extend credit in prescreen offers use tradeline information from consumer reporting agencies, including Trans Union.51 The presence of one or more tradelines 49 Commentary on the Fair Credit Reporting Act, 55 Fed. Reg. 18,804, 18,815 (1990).

50 F 233.

51 F 237, 243-245, 258, 278-281, 288, 290, 295. VOLUME 129 Initial Decision is a threshold criterion for prescreens. Consumers without minimum tradelines are not offered credit.52 [redacted]53 [redacted]54 [redacted]55 [redacted]56 [redacted]57 [redacted]58 [redacted]59 [redacted]60 [redacted]61 [redacted]62 [redacted]63 [redacted]64 [redacted]65 [redacted]66 [redacted]67 [redacted]68 [redacted]69 [redacted]70 [redacted]71 [redacted].72 52 F 238, 244-245, 258, 276, 287.

53 Northern Trust's Personal Financial Services Marketing division issues consumer loans, including debt consolidation loans, preapproved home equity loans, and automobile loans. (McCoy 493/25-494/6.). 54 F 288-290.

55 Wachovia Bank Card Services (Wachovia) is a subsidiary of Wachovia Bank of Georgia and is an issuer of Mastercard and Visa credit cards, including bank cards. (Pendleton 351/24-352/6.) Wachovia has 3 million active bank card holders and $ 5.3 billion outstanding. (Pendleton 410/1-2.). 56 F 243.

57 F 240, 242.

58 F 245.

59 F 246.

60 F 247, 249-250, 252.

61 F 248, 251.

62 F 253-254.

63 FCC National Bank (First Card) is a division of First Chicago NBD and is an issuer of Visa and Mastercard credit cards. (Koppin 476/5-15.) First Card has 14 million credit card holders, 11 million active credit card holders, and about $ 17.8 billion outstanding. (Koppin 482/7-11.) 64 F. 258.

65 F. 268-269.

TRANS UNION CORPORATION 633 Initial Decision There is no single criterion in credit granting that qualifies a consumer for credit.73 In prescreening, each element from a consumer's credit report used by a credit grantor to make a firm offer of credit is a factor in eligibility for credit. The credit grantor does not have a permissible purpose to obtain a consumer report unless every consumer who meets the prescreen criteria is given a firm offer of credit. The 1996 amendments to the FCRA recognize prescreening as a permissible purpose. Congress intended prescreen lists to be treated as Acredit reports.@74 The prescreen criteria are factors in the eligibility for credit. 66 Chase Manhattan Bank issues a variety of credit cards and has 20 million credit card holders. (Zancola 662/6-12; 665/9-11.) 67 F 278. For purposes of this criterion, there is no requirement that the trade be open -- only the number of trades. CX 280-Z-46; F 281. 68 F 276, 278-281. [redacted] F 280.

69 F 279, 281.

70 F 281.

71 F 283 [redacted] F 284 [redacted] F 285 [redacted] F 286 [redacted]. 72 F 282.

73 F 241, 256, 272.

74 H. Rep. 103-486, 103rd Cong., 2nd Sess., 32 (1994)(AA prescreened list is a series of credit reports because the consumers whose names appear on the list have been determined, through information in their credit files, to meet credit criteria identified by a credit grantor.@) VOLUME 129 Initial Decision 3. Invitations to Apply Some banks solicit consumers to apply for credit in Ainvitations to apply.@ The bank mails an application for credit to consumers. This is not a guaranteed offer of credit. The credit grantor decides whether to grant credit on the completed application.75 When a consumer responds, the application is scored by the credit grantor;76 the decision to grant credit depends on the score.77 [redacted]78 [redacted]79 [redacted]80 [redacted]81 [redacted]82 [redacted]83 [redacted]84 [redacted]85 [redacted]86 [redacted]87 [redacted]88 [redacted]89.

75 F 213.

76 F 214, 221.

77 F 219, 230.

78 F 216. [redacted] F 217.

79 F 38, 44, 218.

80 F 220.

81 F 221-222.

82 F 223.

83 F 226.

84 F 227.

85 F 228.

86 F 38 (high credit), 45, 56 (open date). 87 F 229.

88 F 231.

89 F 230.

TRANS UNION CORPORATION 635 Initial Decision In deciding credit eligibility on invitations to apply, credit grantors use the presence of a tradeline, a bank card, and a mortgage. Credit grantors also use as factors: FICO scores, the number of retail cards, and the open dates and credit limits for tradelines.90 Credit grantors have used presence of a student loan in credit eligibility.

Each of these elements, by their use in credit decisions, is a factor in credit eligibility.

C. Trans Union's Target Marketing Lists 1. Target Marketing Databases The Trans Union target marketing lists come primarily from CRONUS, the company's credit report database. a. CRONUS CRONUS is the Trans Union database for individual consumer credit data used to supply consumer credit reports to Trans Union's customers. It contains 600 million records, including 250 million records of previous addresses.91 A Arecord@ on CRONUS means name and address. Additional consumer information -- account number, open date, account type, payment history -- is known as a tradeline. A record can be created on CRONUS by an inquiry or public record entry alone; having a record on CRONUS does not imply the existence of a tradeline.92 CRONUS contains credit data supplied by Trans Union 90 F 232.

91 F 16.

92 F 22-23.

VOLUME 129 Initial Decision subscribers, public record data (which Trans Union purchases from vendors who gather the information at courthouses and similar public records sources), and demographic information, most of which also comes from Trans Union's credit grantor subscribers.93 About 85,000 subscribers supply data to Trans Union in a form known as the Metro format.94 Trans Union updates tradelines on CRONUS nightly, completing a full update of the entire database every week.95 b. The Master File Database Trans Union's database for its target marketing activities is its Master File. The source of individual consumer information for the Master File is CRONUS, and the TransMark database, an intermediate file which is a Asnapshot@ of CRONUS. The TransMark database is created each month of records from CRONUS; it combines all the CRONUS records on the same name and address. The Master File database is built using the records for each person in the TransMark database.96 93 F 17-19.

94 F 17-18, 22.

95 F 19. There is a second database -- ANCI -- that accumulates information daily -- showing inquiries when a Trans Union customer obtains a credit report on a consumer, and manual changes to a consumer file done in response to a consumer dispute. The information on ANCI is checked whenever a consumer report is delivered, and is used in the update of CRONUS. F 21.

96 F 25.

TRANS UNION CORPORATION 637 Initial Decision The Master File uses individual information from CRONUS.97 This process rejects CRONUS records that do not have the minimum number of qualifying trades and applies other criteria.98 The Master File rejects tradelines in CRONUS reported by collection agencies indicating derogatory information.99 Prior to the December 1997 Master file, the minimum requirement for a consumer to be included was two open tradelines, one verified within twelve months.100 After December 1997, it is either two trades reported in the last six months, or one trade reported in the last six months matched to an outside vendor file.101 If a person meets the minimum requirement in the Master File, the consumer's record is further examined for individual information from CRONUS.102 The Master File is rebuilt every four months.103 97 Id.

98 F 27-30.

99 F 29. If a consumer's only tradeline is a collection account, that person will not appear on Trans Union's Master File. The impact is to shield from the list collection accounts denoting less creditworthy individuals. F 200-201. 100 F 31.

101 F 33.

102 F 25, 34, 39, 45, 52, 56-58, 60-64, 70-73, 75. 103 F 26. List orders are handled by ACXIOM Corporation, a data processing firm in Conway, Arkansas. Trans Union owns an interest in ACXIOM. Trans Union also contracts with ACXIOM to perform computer functions for Trans Union, in Chicago and in Conway. F 35. VOLUME 129 Initial Decision c. Characteristics and Attribute File Trans Union uses over 300 credit attributes from CRONUS. This Attribute File, also known as Standard Characteristics, contains credit information about creditworthiness, i.e., the number of trades with payments over 180 days or the maximum balance owed on all mortgage trades. Standard Characteristics are used by Trans Union in target marketing products such as PIC, P$ YCLE, SOLO, TIE, and E-Val.104 Trans Union released characteristics, with names and addresses attached, to its target marketing clients. Trans Union discontinued the disclosure of these characteristics on October 1, 1997.105 Trans Union's Standard Characteristics are no longer available for use in Trans Union's target marketing business; they are available as part of the PerformanceBase/ RelationBase File only for use in firm offers of credit.106 2. Promotion and Sale of Trans Union's Target Marketing Products Trans Union salespeople around the country sell its target marketing lists.107 They identify prospects by mail solicitation or telemarketing. Trans Union also sells lists to brokers, list mangers, and wholesalers (Metromail, R.L. Polk, ACXIOM and 104 F 108. The source for the Attribute File is CRONUS. In order to be in the Attribute File, a person must have at least two open tradelines, one of which is verified within 12 months. F 109.

105 F 110. Among the characteristics are individual level information such as: months since oldest trade opened, and number of personal finance inquiries, finance installment trades opened in 24 months, and bank card trades. Id. 106 F 111.

107 F 118. PerformanceData employs 46 people including 10 salespersons. PerformanceData had 440 target marketing customers (many of whom are third parties representing other customers) and $ 34 million in revenue in 1997. F 122.

TRANS UNION CORPORATION 639 Initial Decision Donnelley Marketing), who sell Trans Union lists or information from Trans Union's lists and their own database, to direct marketers. Trans Union provides target marketing lists for onetime use by its customers by rental or license.108 Trans Union promotes lists by emphasizing the unique source of Trans Union's creditbased consumer data.109 Trans Union equates target marketing products with credit information by stating, for example, that Trans Union's Standard Characteristics correlate highly with Alending activity,@ and Trans Union data are Ahighly predictive . . . especially with financial offers.@110 Trans Union salespeople refer to the credit-based character of the target marketing lists. Trans Union's finance list provides persons who have Agenerally had trouble with their credit in the past and are highly responsive to credit offers@ (CX 68-A); for persons on the student loan list, Asince credit has been established, one could argue that this list would have higher pass rates through the credit bureaus@ (CX 136); and Trans Union's premium bank card list indicates persons Awho have been approved for this high credit amount in the past@ (CX 64-A).111 Trans Union promotes its target marketing lists as providing Adeepest access to credit-based information@ (CX 61-A) and Athis data should not be looked upon as merely credit-based data, but as an individual-level data source unmatched by any other offering@ (CX 61-B).112 108 F 118-119.

109 F 123.

110 F 124; CX 263-A and CX 265-A.

111 F 124.

112 F 133.

VOLUME 129 Initial Decision The requirements for Trans Union's target marketing lists are disclosed to potential customers; Trans Union promotes the fact that each individual in the Master File must have at least two active lines of credit.113 Trans Union also publicizes that a person who shows no activity within a 12 month period, such as payments or credit use, is dropped from the Master File.114 Trans Union publicizes criteria for lists, such as the premium bank card list which is described as identifying persons who hold a bank card with a credit limit above $ 9,999.115 The Master File lists chosen by Trans Union's target marketing clients show the specificity which Trans Union provides in individual-level credit information.116 The diversity and source of this credit-based information permits target promotions by credit grantor marketers.117 Trans Union emphasizes that the Master File is Athe freshest and most comprehensive@ source for data due to its Arobust and extensive source of original credit based information,@ and that Trans Union's database is the largest database of consumer credit information in the United States.118 Trans Union also promotes the Master File as a unique source for individual-level, observed behavior data. According to a PerformanceData brochure, the 113 F 126.

114 F 126.

115 F 44, 126.

116 F 127.

117 F 128-130.

118 F 133. Trans Union describes the Master File as the Arichest source of individual-level financial data available@ (CX 321-F), and that its database is Akept fresh and current by nearly two billion updates supplied by credit grantors every month, and is maintained for accuracy and quality@ (CX 72-B). TRANS UNION CORPORATION 641 Initial Decision Master File is Awithout equal@ and its information is Ahighly accurate@ and is Abased on actual behavior -- not self-reported or neighborhood values.@119 Trans Union sellers testified that Trans Union's lists are superior because Master File data is reported by third parties, and is not the self-reported information relied upon by other target marketers, such as Polk, Donnelley, and Metromail, whose target marketing databases contain survey information directly from consumers, who may provide inaccurate information. Trans Union promotes the fact that its data is more objective because it is reported by third-party credit grantors.120 Target marketing customers can obtain Trans Union lists with credit information similar to prescreen criteria, such as whether the individual has at least three open tradelines and the types of tradelines, or whether a person has two open tradelines, one open for more than 36 months.121 Clients may also have certain information from the Master File printed out, permitting target mail and telephone solicitations.122 119 F 135. The Master File is characterized as Aliving and breathing data,@ Athe most comprehensive available in terms of observed behavior (not selfreported)@ (CX 264-A) and as the Aonly source of individual-level financial data@ that is Abehavioral@ (CX 78-Z-34). 120 F 136.

121 F 130. Trans Union's Master File has also been used for mailing a Apre-approved@ credit offer. F 132.

122 F 131. For the September 1997 Master File, Aprintable@ information includes: open dates for the first and second most current auto loans/leases and mortgages, home value ranges, and length of residence. (CX 1-C, J, N, P). Other information is printable upon request, including high credit amounts and loan type for auto loans and mortgages and the aggregate high credit for all student loans (CX 1-B -- C, O -- P, W).

VOLUME 129 Initial Decision 3. Master File Lists Show Detailed Credit Information Only Trans Union offers target marketing lists based on individual-level credit data. These lists are unique in their source, in the extent of individual information they reveal, and in their use in target mail and telephone promotions. a. Automobile Trans Union offers lists indicating the presence and number of open automobile loans, loan type (lease, refinanced loan, equity transfer loan), the open and expiration dates for the lease or loan, and the high credit amount of the lease or loan. Trans Union also offers a Adriver@ list.123 Other target marketing firms offer automobile lists limited to ownership information and not the range of credit-based data that Trans Union supplies. Other lists obtain data from self-reported survey responses or from the state departments of motor vehicles registration data, not from consumer reporting information.124 [redacted].125 [redacted]126 123 F 38. The source for the Master File Aauto expiration date@ (CX 1-A), Aauto high credit@ (CX 1-B), Aauto loan type@ (CX 1-B), and Aauto open date@ (CX 1-C), is individual records in CRONUS. A person is a Adriver@ (CX 1-E) if the CRONUS record for the person shows an auto loan or a tradeline with a business that issues gas cards. When the Master File program examines CRONUS records to determine whether a person has an auto loan, it will not consider the record if the open date is more than five years old. F 39. 124 F 40, 134.

125 F 41. [redacted] F 43.

126 F 42.

TRANS UNION CORPORATION 643 Initial Decision b. Bank Card Trans Union sells lists that permit target marketing of persons who have an open bank card (including the open date of the most recent bank card) and holders of an open premium bank card (including the open date of the most recent premium bank card). Trans Union defines its premium bank card list as disclosing a bank card with a credit limit of more than $ 9,999.127 Other firms offer bank card lists, with the information at the household level, from self-reported survey responses or from suppliers of credit card information not from consumer credit reporting databases.128 This material is not comparable to the range, depth, accuracy or timelines of Trans Union's CRONUSderived, individual-level credit card information.129 [redacted].130 Nothing in the self-reported credit card information nor in the information provided by other suppliers shows the open date of the bank cards or premium bank cards.131 Target marketing lists from other list providers show credit card use; they are not comparable to Trans Union's bank card list because they do not show whether a consumer has a bank card tradeline, nor any of the information about the tradeline that can be gleaned only from a consumer credit reporting database -- open date, open date of 127 F 44, 126. To generate the Apresence of an open bank card,@ the Master File looks to the date open of the individual CRONUS record for bank card tradelines. For a person to qualify as a premium bank card holder, the individual CRONUS record must indicate a tradeline with a kind of business, account type, and a credit limit greater than $ 9,999. F 45. 128 F 46.

129 F 46-48, 131-146, 148-154, 157-160.

130 F 47-48.

131 F 48.

VOLUME 129 Initial Decision most recent, and high credit limit; these lists are available only from Trans Union.132 [redacted]133 c. Department Store Card Trans Union offers a list for presence of an open department store trade, including open date of the most recent department store trade.134 Other list providers offer department store card information from self-reported survey responses or from suppliers of credit card information whose source is not obtained from consumer credit reporting databases.135 [redacted]. [redacted].136 d. Finance Trade Trans Union offers its customers finance loan lists including presence of an open finance trade (along with the open date of the most recent finance loan), presence of a A30/60/90 day@ finance trade, a financial loan identifier (identifying mortgage or auto loans from a finance company), and presence of a closed finance loan trade.137 The finance loan lists are promoted by Trans Union, 132 F 44, 48-51.

133 F 49.

134 F 52. To determine the Apresence of an open department store card@ (CX 1-E), the Master File looks to the individual CRONUS record. F 52. 135 F 53.

136 F 54.

137 F 55. To determine the Apresence of an open finance trade@ (CX 1-F), the Master File examines the individual CRONUS record showing whether the subscriber's business is Afinance,@ the account type is installment, and if the finance trade has an open date. F 56. Trans Union defines a A30/60/90 day@ trade where payment is due in 30/60/90 days. F 55. To determine the Apresence TRANS UNION CORPORATION 645 Initial Decision and used by Trans Union list clients, to market to riskier segments of the population, because finance companies historically lend to more marginal credit risks.138 e. Head of Household Trans Union offers a list for head of household, derived from individual CRONUS records, naming the person with the greatest number of trades in a household as the head of household.139 No other list provider offers a list showing the person in a household with the greatest number of tradelines.140 f. Mortgage Trans Union offers mortgage related lists including: the presence of an open mortgage, presence of a second open mortgage, and the open and closed dates and high credit amounts (stated in range values) of both mortgage trades. Trans Union's list customers can also obtain information about the type of mortgage loan (VA and FHA loans, refinanced and secured mortgages, and secured home improvement loans).141 The lists for presence of mortgages, mortgage high credit range, mortgage closed dates, and mortgage type are derived from individual CRONUS records.142 The Master File list AHome Value Range@ of a 30/60/90 day finance trade,@ the Master File examines an individual CRONUS record for the subscriber's kind of business and the account type. A Afinance loan closed@ also uses individual CRONUS records. F 57. 138 F 124, 128, 203.

139 F 58.

140 F 58.

141 F 59.

142 F 60.

VOLUME 129 Initial Decision uses Trans Union's Standard Characteristics, which is CRONUSderived individual credit information, to calculate a consumer's home value.143 List providers other than Trans Union offer mortgage lists. Their information is from self-reported sources such as surveys, and the public record -- including county registrar and tax assessor files.144 [redacted].145 [redacted]146 [redacted].147 Except when a house is sold, any calculation of home value or equity is an estimate. Trans Union's up-to-date mortgage balance information makes calculations more accurate than extrapolations from deed information, which may be old. [redacted] Polk's internal analysis determined that the Trans Union data were more accurate than the public record data.148 Trans Union's home equity list uses credit attributes from CRONUS, including presence of a mortgage, the original open date, and the opening mortgage balance, to calculate the current home market value for a house, and then uses the current 143 F 61-63. This is the same model used by Trans Union to create E-Val. Among the standard characteristics used by Trans Union for the home value calculation are: months since oldest revolving trade opened, total mortgage high credit/credit limit, maximum balance owed on all mortgage trades, and total balance of all mortgage trades. F 62. 144 F 65, 66. [redacted] F 66.

145 F 66.

146 F 67.

147 F 69. [redacted] 148 F 68.

TRANS UNION CORPORATION 647 Initial Decision mortgage balance to calculate the estimated equity available.149 Knowing the equity in a residence is valuable information to sellers of home equity loans; it permits them to solicit those likely to be qualified to purchase. Using Trans Union's mortgage data, a telemarketer of home equity loans can obtain names of consumers who have multiple mortgages, are homeowners, age 24-55, with home market value $ 50,000 - $ 299,000. (CX-24.) Lists can use open dates of loans (e.g., month that a mortgage was taken out);150 loan types or sources (e.g., mortgages issued by finance companies (CX-40); and types of credit cards (e.g., upscale retail cards (CX-35).151 g. Length of Residence Trans Union offers a length of residence list showing the number of years a person has resided at a current address. The Master File length of residence indicator uses individual data from CRONUS, including mortgage open dates.152 149 F 63. The Master File list Ahome equity actual,@ the actual dollar amount of estimated equity in an individual's home, and Ahome equity range,@ similar information stated as a range value, are calculated by using mortgage high credit and mortgage balance information from CRONUS and subtracting them from the modeled home value. F 64.

150 Knowing that a homeowner took out a mortgage during a period of high interest rates facilitates telemarketing and other promotions that specifically mention refinancing the prospect's high-interest loan. F 129; CX 19-F.

151 F 127-129.

152 F 70.

VOLUME 129 Initial Decision h. Mail Order Trans Union offers mail order buyer lists; many are based on information purchased from outside vendors. One top seller, however -- Trans Union's AMOB1" -- is based on individual credit information in CRONUS.153 i. Singles Trans Union offers a Asingles@ list of unmarried consumers, based on tradelines for the person in CRONUS.154 j. Student Loan Trans Union offers several student loan lists including presence and open date of a student loan, the aggregate high credit amount of all of a person's student loan, and a closed student loan, all derived from CRONUS individual credit information.155 k. Upscale Card Trans Union offers a list for the presence and open date of an open upscale retail card (Neiman Marcus, Saks, Tiffany). To find the upscale retail card and open date, the Master File uses individual CRONUS records, including the specific subscriber number on the individual consumer's tradeline. If the subscriber code matches the code for one of the companies on the National Retail Federation list of upscale stores, the person is named as having an open upscale retail card on the Master File.156 153 F 71. Trans Union sells its lists to mail-order sellers (Spiegel's, L.L. Bean, Eddie Bauer); it examines individual consumer records on CRONUS and identifies as a mail-order buyer on the Master File those consumers who have open tradelines with mail order companies. Id. 154 F 72.

155 F 73.

156 F 74-75.

TRANS UNION CORPORATION 649 Initial Decision List providers other than Trans Union do not differentiate their retail card lists to provide an Aupscale@ list. [redacted].157 4. September 1997 Master File a. E-Val Trans Union offers E-VAL, a scoring system that predicts the amount of equity available in a consumer's home. E-Val is also available on Trans Union's Master File in the form of its AHome Value Ranges,@ AHome Equity (Actual),@ and AHome Equity Range@ lists. E-Val lists were available for sale to target marketing customers prior to October 1, 1997; they are now available only for firm offers.158 b. PIC PIC, created by Trans Union and an outside modeling firm, predicts the likelihood that a person owns a financial service. Trans Union used surveys and characteristics from the consumers' CRONUS files in PIC. A PIC score is calculated using individual credit data to predict the likelihood of a person owning an IRA account or certificate of deposit. Until October 1, 1997, Trans Union offered PIC as a Master File select; PIC is now available only for firm offers of credit.159 157 F 76.

158 F 61, 77.

159 F 78.

VOLUME 129 Initial Decision c. P$ YCLE P$ YCLE, created by between Trans Union and Claritas, assigns consumers to one of 42 segments (known as Aclusters@ or Abuckets@). P$ YCLE uses individual-level Standard Characteristics to calculate an estimate of a person's income producing assets, and assign the person to a P$ YCLE's Abucket.@ Trans Union offered P$ YCLE for non-firm offers prior to October 1, 1997; P$ YCLE is now available only for firm offers.160 [redacted].161 d. SOLO SOLO puts consumers into groups (Aclusters@) based on individual level CRONUS data, including 35 credit characteristics. SOLO is a Asister product@ to SILHOUETTE, a Trans Union product identifying people likely to respond to a firm offer of credit and is available only to those with a permissible purpose under the FCRA.162 SILHOUETTE's characteristics include Anumber and types of trades . . . age of trades, credit limits, credit utilization, payment history . . . .@ According to the SILHOUETTE User's Guide, Aassignment to a cluster is based on a credit view of consumers . . . Cluster 8, because of its generally high level of credit activity, might contain consumers who are good candidates for credit offers.@ SILHOUETTE distinguishes clusters based on some of the same factors used by Trans Union in its Master File target 160 F 79.

161 F 80.

162 F 81.

TRANS UNION CORPORATION 651 Initial Decision marketing list products -- e.g., finance accounts, bank revolving, department store, retail.163 SOLO and SILHOUETTE use the same 35 Trans Union Standard Characteristics; SOLO's 41 clusters were developed from SILHOUETTE's 25 clusters. SOLO and SILHOUETTE evaluate individual credit behavior. SOLO is described by Trans Union as Aa disguised version of SILHOUETTE for list applications.@164 Trans Union's internal Seller's Guide describes SOLO as Aobjectively-reported, behavior-based information, rather than self-reported (and therefore inevitably biased) information.@ It notes that ASOLO is most often used by credit grantors for nonpreapproved offers, such as home equity offers or secured card offers.@165 Until October 1, 1997, Trans Union offered SOLO to its target marketing customers; now SOLO is available only for firm offers of credit. Trans Union withdrew SOLO from its target marketing customers because of the FCRA Amendments and because customers had learned that SOLO clusters correlated with how consumers pay their bills.166 Prior to October 1, 1997, Trans Union attached SOLO codes to a customer's own client lists. Trans Union's internal seller's guide for SOLO states that SOLO 163 F 82-83.

164 F 82, 85; CX 110. ASILHOUETTE provides . . . actual credit data to catgorize [sic] individuals. Therefore the FTC requires permissible purpose . . . to use SILHOUETTE. SOLO is also based on credit information; however, when SOLO clusters are reported back, the credit characteristics are masked . . .@ F 85; CX 115-Z-49.

165 F 84-89.

166 F 86.

VOLUME 129 Initial Decision clusters can be used to Aprofile@ one's Abest customers.@167 Trans Union intends that SOLO is used for credit eligibility. The internal Trans Union SOLO seller's guide discusses its use for preapproved offers of credit, and tracking Aactivation, utilization, retention, and/or profitability@ of credit accounts, and use with Trans Union's income estimator, ATIE,@ for prescreened offers.168 Competitors' models similar to Trans Union's SOLO and P$ YCLE are not comparable because they do not use individual credit information. ACXIOM received SOLO and P$ YCLE codes from Trans Union until October 1, 1997. The models that ACXIOM uses in place of Trans Union's P$ YCLE and SOLO (Claritas' Affluence and WealthWise and Experian's P$ YCLE), do not predict as well as the Trans Union data.169 e. TIE TIE, Trans Union's income estimator, calculates an individual consumer's estimated income based on credit data from CRONUS. Prior to October 1, 1997, TIE was offered for target marketing purposes; it is now available only for firm offers.170 To receive a TIE score, a consumer must have at least two tradelines, one of which must have a credit line greater than zero, and one must have been open for at least twelve months. To calculate a consumer's TIE score, Trans Union uses 23 Standard Characteristics, such as number of tradelines.171 167 F 87. Until September 30, 1997, Trans Union appended SOLO clusters to TransLink lists. F 88, 105.

168 F 89.

169 F 90.

170 F 91.

171 F 92.

TRANS UNION CORPORATION 653 Initial Decision Trans Union intends TIE to be used in credit granting. A product brochure for TIE states that TIE can be used to Afine tune credit limits and loan conditions on credit applications,@ to A'Red Flag' applicants whose low income estimate may indicate the need for additional verification,@ and to Aflag accounts to increase/decrease lines of credit.@ Trans Union's internal seller's guide for TIE recommends uses of TIE: credit risk scoring for new or existing accounts; use in prescreen criteria; as a supplement to credit applicant data; and to set initial credit limits.172 Competitive list providers offer estimated income developed from public record and self-reported data, subjective information that has not been verified, and household income rather than individual level income. TIE is updated every seven days, whereas the income estimates of other list providers rely on census data updated every ten years.173 [redacted] offer an income product, based on self-reported information and census figures. None of that information is individual credit data nor is it from credit reporting agencies.174 5. Trans Union's Other Target Marketing Lists a. TransLink TransLink is Trans Union's reverse append product that associates a name and address with a bank card number. It works as follows: consumers charge purchases at a merchant using credit cards other than the merchant's own credit card; the merchant sends to Trans Union a list of the credit card account numbers; 172 F 93-94.

173 F 95.

174 F 95-98.

VOLUME 129 Initial Decision Trans Union, using individual account information from CRONUS, returns to the merchant the names and addresses of the consumers. The merchant can then send those consumers promotions.175 The merchant Aprints a mailing label and sends [the consumer] a piece of junk mail.@ Until September 30, 1997, Trans Union appended SOLO, TIE and age data to TransLink lists. Trans Union promotes TransLink as a file that is created from consumers who frequently use their Mastercard, Visa, American Express, Discover, Optima and other third party cards. This file contains a consumer's active tradeline number and address . . . Successful address matches can also be linked to Trans Union's Master File for other demographic appends.176 TransLink is among Trans Union's most profitable list products.177 Trans Union is the only consumer reporting agency that offers this service.

b. New Issues The New Issues file is a list of consumers receiving credit in the last 90 days. A Trans Union customer knows the date (30/60/90 days) and type of credit (retail, finance, mortgage or auto loan trades). The New Issues file gets this data from CRONUS. On October 1, 1997, Trans Union discontinued the sale of the New Issues file for target marketing customers and now the file is only available for firm offers of credit.178 175 F 99, 101.

176 CX 58-D.

177 F 104.

178 F 106.

TRANS UNION CORPORATION 655 Initial Decision c. Emerging Consumers The Emerging Consumers file gets from CRONUS names of consumers with one tradeline with a date verified within the last twelve months. Trans Union discontinued the Emerging Consumer file due to concern that it was Acommunicating information that we shouldn't be communicating.@179 D. Burden of Proof Respondent argues that180 complaint counsel has the burden of proof by Aclear and convincing evidence.@ The two Commission cases cited by respondent do not support this proposition.181 Respondent also cites Alioto v. Cowles Communications, Inc., 519 F.2d 777, 779 (9th Cir. 1975), which used the clear and convincing standard in a case where the burden of proof requires a showing that a libelous statement was made in reckless disregard of the truth. In general, the clear and convincing burden of proof is used where the Court considers a particular type of claim should be disfavored on policy grounds.182 No such precedent is cited here, and the preponderance of the evidence standard applies in this case.

179 F 107.

180 Brief pp. 17-18.

181 Id. p. 18 n.9. McGraw-Hill Pub. Co., 57 F.T.C. 1152 (1960), cited by respondent as requiring Aclear and convincing evidence,@ specifically used the preponderance of the evidence burden of proof, id. at 1171. 182 TREATISE ON CONSTITUTIONAL LAW, ROTUNDA, Vol. 3 at p. 168 (1986).

VOLUME 129 Initial Decision E. Trans Union Target Marketing Products As Consumer Reports 1. FCRA The FCRA defines Aconsumer report@ to mean any written, oral, or other communication of any information by a consumer reporting agency bearing on a consumer's credit worthiness, credit standing, credit capacity, character, general reputation, personal characteristics, or mode of living which is used or expected to be used or collected in whole or in part for the purpose of serving as a factor in establishing the consumer's eligibility for: (A) credit or insurance to be used primarily for personal, family, or household purposes;

(B) employment purposes; or (C) any other purpose authorized under section 604. 15 U.S.C. ' 1681a(d).

In the context of this litigation, as framed by the Court of Appeals, to qualify as the sort of Ainformation@ that can constitute a consumer report, then, an entry on a Trans Union mailing list must (A) Abear[ ] on@ at least one of the seven factors and (B) be used, expected to be used, or collected for one of three types of purposes.

The first element does not seem very demanding, and we do not understand Trans Union to even contest the proposition that a person's having two tradelines Abear[s]@ on one or more of the seven enumerated factors...

In addressing the next factor, whether information in the lists is Aused or expected to be used or collected in whole or in part for the purpose of serving as a factor in establishing the TRANS UNION CORPORATION 657 Initial Decision consumer's eligibility@ for various benefits, the Commission considered only credit eligibility, and we follow suit. Trans Union, 81 F.3d at 231.

The issue here is whether information that Trans Union conveys in its target marketing lists is also information that is used by credit grantors in eligibility determinations. The Court of Appeals indicated that to be Aa factor,@ information should play a role in the credit decision but that it need not be the decisive factor. Id. at 233. Thus, the Court of Appeals noted that on remand the FTC could show that Trans Union used or expected existence-of-tradeline information to be used as a factor in creditgranting decisions by demonstrating Athat credit decisions could be made, even in part, on such 'existence' information.@ Id. Under the language of the statute, existence-of-tradeline information could constitute a consumer report because it serves as Aa factor@ in credit grantors' decisions to grant credit. Trans Union's credit scoring and the criteria they use for prescreen and invitation to apply offers, show that the presence of a tradeline may be a factor in determining whether a consumer is eligible for credit. Other factors used by credit grantors, including presence, type and date of tradelines, are used in Trans Union's target marketing lists. Trans Union's lists may be consumer reports because the information is Aused ... or collected in whole or in part for the purpose of serving as a factor in establishing the consumer's eligibility for ... credit.@ 15 U.S.C. 1681a(d)(1). 2. Facts The factors that Trans Union uses in target marketing products are also used by credit grantors deciding credit.183 That makes 183 Respondent called witnesses who testified that data from the Master List File could not be used in part to determine credit eligibility. Respondent's witnesses were not credible on this issue. Dr. Coffman, who works with Trans VOLUME 129 Initial Decision them Aconsumer reports.@ 15 U.S.C. 1681a(d)(1). Trans Union, 81 F.3d at 233.

Every name on Trans Union's target marketing lists must have at least one qualifying tradeline.184 This is also required in Trans Union's credit scoring, and in credit grantors' decisions to make firm offers of credit in prescreen promotions.185 When Trans Union sells a target marketing list, it is selling a list of people who have at least one tradeline; a list of people who have one tradeline is a series of consumer reports, and the recipient of the list must have a Apermissible purpose@ in order for Trans Union lawfully to furnish the list.186 The Court of Appeals has already held that target marketing is not a Alegitimate business purpose under the Act.@ Trans Union, 81 F.3d at 233-34. Therefore, Trans Union's sales of target marketing lists constitute furnishing consumer reports to those who do not have a permissible purpose to receive them, in violation of Sections 604 and 607(a) of the FCRA. 15 U.S.C. '' 1681b, 1681e(a).

Union on predictive models, showed bias when he testified, inconsistently, that a post office box may be part of eligibility because some creditors will not mail to a post office box. (Tr. 3840.) Furthermore, Dr. Coffman testified that based on his experience with credit risk scoring models he has seen the number of mortgages and their amounts, the number of auto loans, the number of open bank cards, and the age of the oldest trade and credit limits used as predictive attributes in scoring models. (F 195, 204-07; Coffman 3862/5-17; 3868/16 -- 3869/10; 3869/6 -- 3871/4; 3876/14 -- 3877/20; 3882/7 -- 3884/4.) Another respondent witness, Kenneth Scott, president of a direct marketing advertising agency, has experience in marketing credit cards, not in eligibility determinations. (Scott 2608/23 -- 2609/18; 2614/17 -- 2616/16; 2622/23 -- 2624/1.) 184 F 28-31, 33, 92, 106-107, 109, 126.

185 F 185-188, 243-245, 258, 278-282, 288, 290. 186 Age of newest and oldest tradeline is also an eligibility factor. F 117, 196, 198-199, 228; 18 F.T.C. at 841-42.

TRANS UNION CORPORATION 659 Initial Decision Trans Union's target marketing customers can get a list of consumers who have an open bank card tradeline.187 The same criterion is used by credit grantors in their eligibility decisions,188 based on consumer reports, and, Trans Union's sale of such a list to a target marketer who has no permissible purpose to receive the consumer reports is a violation of the FCRA. Similarly, Trans Union's SOLO and TIE use credit eligibility factors.189 They are marketed by Trans Union for both target marketing and credit eligibility.190 F. Trans Union's Constitutional Rights Trans Union asserts First Amendment and Equal Protection arguments in its defense. The Commission, in its September 28, 1994 opinion, rejected Trans Union's constitutional arguments. 118 F.T.C. at 879-890. The Court of Appeals did not reach Trans Union's First Amendment claim, but signaled its views. The Court stated:

Trans Union raises a serious First Amendment claim with respect to the Act's application. Its target marketing list competitors who aren't consumer reporting agencies under the Act can use any information they gather -- including credit information -- in the preparation of their lists. In fact, because of its interpretation of Acollected ... for the purpose ...@ in the Act, the Commission would evidently permit Trans Union to sell its target marketing lists if the data were Aseparately obtained for target marketing purposes.@ 187 F 44.

188 E.g., F 216, 224, 248-252.

189 F 82, 85, 92, 108.

190 F 87-89, 93-94.

VOLUME 129 Initial Decision Trans Union, 81 F.3d at 235.

Trans Union argues that it has been denied equal protection rights by the Commission's enforcement action,191 but any difference in treatment between Trans Union and its list competitors who aren't consumer reporting agencies is attributable to the FCRA, not to the Commission's enforcement policy. The FCRA was designed to stop unfair information practices in the credit reporting industry that were harming consumers and undermining confidence in the banking system.192 191 The Commission has not been unfairly discriminatory in applying the FCRA to the sale of target marketing information by competing credit reporting firms. Trans Union's practices go beyond the activity allowed by the TRW consent order. (F 357.) Respondent argues that credit grantors disseminate information about their customers for target marketing though it is just as invasive of privacy as anything done by PerformanceData. E.g., to a lesser degree, TRW used credit attributes from its consumer reporting database, aggregated to a nine digit zip code. (F 360.) Administrative agencies, however, have the prosecutorial discretion to go after one law breaker at a time. Moog Industries, Inc. v. FTC, 355 U.S. 411, 413 (1958); FTC v. Universal-Rundle Corp., 387 U.S. 244 (1967). This implies the right to shape remedies, to some extent, to meet the eccentricities of negotiated settlement of litigation. (F 360.) Respondent also decry that its competitors in the sale of target marketing lists obtain their data from public records such as telephone books (Aseparately obtained@ information). The record in this case shows that the target marketing lists from that information is not as effective as the lists sold by respondent which are from the credit report database. 192 The Congressional purpose in the FCRA is that Athe banking system is dependent upon fair and accurate credit reporting. Inaccurate credit reports directly impair the efficiency of the banking system, and unfair credit reporting methods undermine the public confidence which is essential to the continued functioning of the banking system.@ 15 U.S.C. ' 1681(a)(1). TRANS UNION CORPORATION 661 Initial Decision 1. PerformanceData Lists Are Commercial Speech Complaint counsel argue that PerformanceData lists are a product, not speech, and are not protected by the First Amendment. But the lists are part of commercial transactions, providing Ainformation of import to significant issues@ serving as a vital part of the process to Ainform the public of the availability, nature, and prices of products and services,@ and performing Aan indispensable role in the allocation of resources in a free enterprise system.@ Bates v. State Bar of Arizona, 433 U.S. 350, 364 (1977).

Target marketing lists are the means by which direct and telemarketers tell consumers about their business. Information on home sales and mortgages are found in these lists (Tr. 2657-59), as is information on automobile ownership, F 38, bank cards, F 44, department store cards, F 52, finance loans, F 55, mail order buyers, F 71, student loans, F 10, and upscale retail credit cards, F 74. Target marketing lists are as much a part of the process of speech as loudspeakers, Saia v. New York, 334 U.S. 558 (1948); public sidewalks, United States v. Grace, 461 U.S. 171 (1983); or free standing newsracks used to distribute free magazines, City of Cincinnati v. Discovery Network, Inc., 507 U.S. 410 (1993). Target marketing lists involve information taken from the database gathered and used in the sale of credit reports. They constitute commercial speech.193 193 PerformanceData has provided a list to a wholesaler whose client was the National Republican Committee, which used the list to solicit consumers to make campaign contributions (CX 22; Tr. 308-09); the predominant use of the lists, however, is for commercial purposes. TU 171-73, TU 175; TU. 307-08, TU-315-19, TU-1317-21.

VOLUME 129 Initial Decision 2. Commercial Speech Law The sale of consumer reports receives limited First Amendment protection as Acommercial speech.@ Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U.S. 749, 762, n.8 (1985). A'Commercial speech [enjoys] a limited measure of protection, commensurate with its subordinate position in the scale of First Amendment values,' and is subject to modes of regulation that might be impermissible in the realm of noncommercial expression.@ Board of Trustees of State Univ. of N.Y. v. Fox, 492 U.S. 469, 477 (1989). Restrictions on truthful commercial speech will be upheld if the government asserts a substantial interest, the regulation directly advances that interest, and the regulation is narrowly drawn. Florida Bar v. Went For It, Inc., 515 U.S. 618, 623 (1995); Central Hudson Gas & Electric Corp., v. Public Serv. Commu of N.Y., 447 U.S. 557, 564 (1980).194 a. Privacy Interest in the FCRA The FCRA protects consumers' privacy by prohibiting consumer reporting agencies from communicating information covered by the Act to marketers for impermissible purposes. Protecting the privacy of consumers may be a Asubstantial@ governmental interest. Florida Bar v. Went For It Inc., 515 U.S. at 194 Though not yet explicitly overruled, the Central Hudson test for commercial speech must be used with some apprehension. 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484 (1996) held a state ban on price advertising of alcoholic beverages to be impermissible under the First Amendment, and the varying opinions of the Court show minority support for the Central Hudson test and adumbrate change. Sean P. Costello, Strange Brew: The State of Commercial Speech Jurisprudence Before and After 44 Liquormart, Inc. v. Rhode Island, 47 CASE W. RES. L. REV. 681, 685 (1997). Nevertheless, Central Hudson continues in use. International Dairy Food Assn v. Amestoy, 92 F.3d 67, 72 (2d Cir. 1996); Anheuser-Busch, Inc. v. Schmoke, 101 F.3d 325, 327, 330 (4th Cir. 1996); Miller v. Stuart, 117 F.3d 1376, 1382 (11th Cir. 1997), cert. denied, 139 L.E. 2d 753, 118 S.Ct. 852. TRANS UNION CORPORATION 663 Initial Decision 621 (substantial government interest in protecting the privacy of personal injury victims against invasive contact by lawyers). Trans Union sells information from its consumer credit database to target marketers who engage in the direct promotion. Trans Union invades consumers' privacy when it sells consumers' credit histories to third-party marketers without consumers' knowledge or consent; that privacy interest is substantial.195 The FCRA is based on Aa need to insure that consumer reporting agencies exercise their grave responsibilities with fairness, impartiality and a respect for the consumer's right to privacy.@ Section 602 of the FCRA, 15 U.S.C. ' 1681. Congress aimed at the risks to privacy from unregulated use of personal information;196 the dissemination of credit reports for purposes other than granting credit was the problem.197 195 F 316-19, 325, 353-55. In Ohralik v. Ohio State Bar Assn, 436 U.S. 447, 465 n.25 (1978), the Court held that an in-person solicitation by an attorney visiting a potential client in her hospital bed recovering from an accident was intimidating and invaded her privacy. In Shapero v. Kentucky Bar Assn, 486 U.S. 466, 476 (1988), the Court held that there was no invasion of the potential client's privacy by a lawyer's solicitation letter, but the Court stated that the invasion of privacy, if any, occurred when the lawyer discovered the recipient's legal affairs, not when he confronts the recipient with the discovery. In this case, the invasion of privacy occurs when respondent takes information from the credit report database, puts it in the target marketing lists, and sells the lists to direct and telemarketers. 196 Alan Westin, PRIVACY AND FREEDOM 163-165 (Athenaeum 1967). 197 F 325-26. Fair Credit Reporting Hearings on H.R. 16340 Before the Subcomm. on Consumer Affairs of the House Comm. On Banking and Currency, 91st Cong., 2nd Sess. 31 (1970) at 109; Hearings on S. 823 before the Subcomm. On Financial Institutions Of The Senate Banking and Currency Comm., 91st Cong., 1st Sess. 92 (1969) at 359, 433. VOLUME 129 Initial Decision b. Public Confidence in Credit Reporting Congress also intended to assure the integrity of the credit reporting system and the public's confidence in the credit reporting system.198 Credit reporting is vital to the United States economy and to consumers, and Aunfair credit reporting methods undermine the public confidence which is essential to the continued functioning of the banking system.@ Section 602 of the FCRA, 15 U.S.C. ' 1681.

c. Citizens' Concerns Citizens worry about Asecondary@ uses of their personal information. (F 316-18, 325-26.) In 1996, Congress narrowed disclosure of credit report information for Alegitimate business need@ by adding that a Abusiness transaction@ must be consumerinitiated. Section 604(a)(3)(F)(I). A permissible purpose to obtain a consumer report without the consent of the consumer exists only for firm offers of credit or insurance and only with notice and an Aopt-out@ privacy safeguard.199 A firm offer of credit is needed to make a prescreen offer permissible.200 Congress deleted the proposed Senate language authorizing credit report information to be used in target marketing.201 198 F 325.

199 F 331.

200 F 329. A prescreen (Afirm offer@) is only permissible if the consumer reporting agency offers the consumer an election to be excluded from marketing lists of potential borrowers and publicizes this option to consumers. Section 604(c) and Section 604(e)(1) and (5). The statute now limits the type of information that may be disclosed to the credit or insurance grantor (Section 604(c)(2)) and requires that the solicitation disclose that information contained in the consumer's consumer report was used and inform the consumer of the right and procedure to opt-out. Section 615(d). 201 F 367.

TRANS UNION CORPORATION 665 Initial Decision d. Government Interest Directly Advanced The Order here must advance an interest Ain a direct and material way.@ Florida Bar, 515 U.S. at 625-6. There, the Court relied on a survey of the effects of lawyer advertising on public opinion, to show that Florida's 30-day ban on solicitation directly advanced the interests of consumer privacy. Id. at 625. This record contains uncontradicted evidence of consumer privacy interests supporting limitation on uses of consumer credit information beyond those Apermissible@ uses specified in the Act.202 Consumers have a privacy interest in the use of information from CRONUS for compiling target marketing lists.203 The FCRA and the Order entered here directly advance the governmental interest in protecting consumers' right not to have covered information communicated by consumer reporting agencies to target marketers without a permissible purpose. Trans Union, 118 F.T.C. at 884-85; 81 F.3d at 230.204 Consumer reporting agencies warrant special restrictions because of their unique position as comprehensive repositories of consumer information and their critical role for the nation's economy.205 202 F 316-319.

203 F 318.

204 AWe find no resemblance between target marketing and [the Act's permissible] purposes: extension of credit, employment, underwriting of insurance, and license eligibility ... If Trans Union's provision of lists derived from its 'base list' for target marketing is to be lawful under the Act, it must be because the information is not so sensitive as to rise to the level of a consumer report.@ Trans Union, 81 F.3d at 234.

205 F 6-12, 16-19, 21, 325-26.

VOLUME 129 Initial Decision The Order must be narrowly tailored to achieve its goal. Florida Bar, 515 U.S. at 632. Since the Order follows the statute, the issue is whether the statute is tailored to achieve its goal. Further, the opt-out procedure required by the FCRA does not cure the problem. While the right to opt-out theoretically allows the consumers to request their names to be removed from target marketing lists (F 334), most consumers are unaware of this procedure (F 336). Although Trans Union complies with the notice requirement for opt-out under the FCRA (F 342), there is no credible, direct evidence of the success rate of opt-out actually stopping direct mail or telemarketing calls (F 343, 348). In enacting the FCRA Congress addressed consumers' concern that the information in their credit histories be protected from misuse. The FCRA is reasonably tailored to achieve the governmental interest in consumer privacy. The FCRA emphasizes controls on the use of credit report information rather than controls on collection of data. Recipients must have a Apermissible purpose@ to obtain the information. The FCRA does not outlaw the secondary use of credit information for target marketing; it merely requires credit reporting agencies to include consumers in the decision to use their information.206 The limitation on secondary use of credit information imposed by the Act is no more extensive than necessary to serve the substantial government interest in protecting consumers from impermissible uses of their credit information.

F. Summary Trans Union's target marketing lists are Aconsumer reports@ as defined in the Fair Credit Reporting Act. A tradeline in a consumer's credit file used in Trans Union's target marketing lists is collected by Trans Union Ato serve as a factor in credit-granting 206 F 328-29, 331.

TRANS UNION CORPORATION 667 Initial Decision decisions.@ Credit grantors Aused or expected it to be used for the purpose.@ Trans Union, 81 F.3d 228 at 233. Other elements from individual consumer credit files, used by Trans Union in its target marketing lists, also are used in credit grantors' decisions to grant credit, including tradelines by type (bank card, finance company, mortgage, automobile loan), open date of tradelines, high credit limit, current mortgage balance, and estimated individual income.207 The same elements from credit files at Trans Union are used as factors in credit grantors' decisions to grant credit to consumers. Target marketing lists assembled from these elements are Aconsumer reports.@ Target marketing is not a permissible purpose for furnishing a consumer report; sale by Trans Union of its target marketing lists thus violates the FCRA.208 Trans Union, 81 F.3d at 234.

The Fair Credit Reporting Act complies with First Amendment commercial free speech and equal protection standards: the FCRA protects a substantial government interest in the privacy of individual consumer information in credit bureau files; the law directly advances that governmental interest in protecting consumer privacy and provides a reasonable fit to serve the governmental interest. The record shows substantial citizen concern over the privacy of the individual information collected, stored, and sold by credit bureaus. The FCRA affords privacy protection tailored to the sensitive nature of the data collected by 207 Certain data from the consumer files are used for identification rather than as an element in credit determination (name, telephone number, mother's maiden name, address, zip code, year of birth, age, any generational designation, social security number). (F 357; but see F 360.) 208 On October 1, 1997 -- the first day that it would face potential civil penalty liability for violations of the FCRA -- Trans Union reduced the information from its consumer credit database that it makes available through its target marketing lists.

VOLUME 129 Initial Decision consumer reporting agencies and their unique status as repositories of the data.

III. CONCLUSIONS OF LAW 1. The Federal Trade Commission has jurisdiction over the subject matter of this proceeding and over the respondent, Trans Union Corporation.

2. Trans Union is a corporation doing business under the laws of the state of Delaware, with its office located at 555 West Adams Street, Chicago, Illinois 60661.

3. Trans Union assembles information on consumers to furnish consumer reports to subscribers and consumers. Trans Union furnishes these consumer reports in interstate commerce. 4. Trans Union is a consumer reporting agency. Section 603(f) of the Fair Credit Reporting Act (FCRA), 15 U.S.C. ' 1681a(f). 5. Trans Union's target marketing lists are consumer reports. Section 603(d)(1) of the FCRA, 15 U.S.C. ' 1681a(d)(1). Trans Union Corp. v. FTC, 81 F.3d 228 (D.C. Cir. 1996). 6. Trans Union furnishes consumer report information in target marketing lists to persons who do not have a permissible purpose under Section 604 of the FCRA, 15 U.S.C. ' 1681b; Trans Union Corp. v. FTC, 81 F.3d 228, 233-34 (D.C. Cir. 1996). 7. By this conduct, Trans Union violates Section 604 and Section 607(a) of the FCRA. 15 U.S.C. '' 1681b, 1681e(a). 8. An appropriate Order follows.

ORDER IT IS HEREBY ORDERED that respondent, Trans Union Corporation:

TRANS UNION CORPORATION 669 Initial Decision a) Cease and desist from distributing or selling consumer reports in the form of target marketing lists to any person unless respondent has reason to believe that such person either intends to make a firm offer of credit to all consumers on the lists or to use such lists for purposes authorized under Section 604 of the FCRA. b) Maintain for at least five (5) years from the date of service of this Order and upon request make available to the Federal Trade Commission for inspection and copying, all records and documents necessary to demonstrate fully its compliance with this Order.

c) Deliver a copy of this Order to all present and future management officials having administrative, sales, advertising, or policy responsibilities with respect to the subject matter of this Order.

d) For the five (5) year period following the entry of this Order, notify the Commission at least thirty (30) days prior to any proposed change in respondent such as dissolution, assignment, or sale resulting in the emergence of a successor corporation, the creation or dissolution of subsidiaries, or any other change in the corporation that might affect compliance obligations arising out of this Order.

e) Within one hundred and eighty (180) days of service of this Order, deliver to the Commission a report, in writing, setting forth the manner and form in which it has complied with this Order as of the date.

VOLUME 129 Dissenting Statement DISSENTING STATEMENT OF COMMISSIONER SHEILA F. ANTHONY I oppose the issuance of a stay in this matter. Administrative tribunals such as the Commission Amay properly stay their own orders when they have ruled on an admittedly difficult legal question and when the equities of the case suggest that the status quo should be maintained.@ Washington Metropolitan Area Transit Comm=n v. Holiday Tours, Inc., 559 F.2d 841, 844-45 (D.C. Cir. 1977). In the present case, however, Trans Union has failed to make an adequate showing on any of the relevant factors. To begin with, Trans Union=s arguments on the merits, which simply repeat arguments made previously, do not convince me that the issues in this case are close or difficult, much less that Trans Union has a substantial chance of success on appeal. Nor are the equities in Trans Union=s favor. Quite to the contrary, Trans Union has failed to make any plausible showing that it will suffer irreparable injury. Trans Union is first and foremost a credit reporting agency, and it makes no claim that compliance with our order will interfere with its principal activity, that of selling credit reports. Even with respect to its target marketing activities, moreover, our order does nothing more than require Trans Union to comply with the law, in the same manner that all credit reporting agencies must. Furthermore, Trans Union=s attempt to equate its economic and speech interests with those that have justified stays in wholly dissimilar cases is grossly unpersuasive. Merely invoking constitutional arguments in support of its business activities does not entitle Trans Union to an automatic stay pending appeal.

Finally, even if I were to assume that Trans Union stands to suffer some level of injury, I would still conclude that the equities strongly weigh against a stay, in light of the vital interests of consumers and the general public at stake here. Trans Union has a legal obligation to comply with the FCRA by furnishing consumer credit reports only to those with a legally permissible purpose to TRANS UNION CORPORATION 671 Dissenting Statement receive them. Conforming its target marketing business to comply with the FCRA and our order will protect consumers= substantial privacy interests in their financial transactions. Moreover, once Trans Union exploits consumers= confidential financial information, it is extremely difficult to compensate for such an invasion of privacy. Therefore, the public interest weighs in favor of halting this violative behavior and effectuating the Commission=s order at the earliest possible date. Accordingly, in my view, Trans Union has not met its burden in demonstrating the necessity of a stay.

VOLUME 129 Complaint