American Bankers v. National Credit Union Admin.

Full title: AMERICAN BANKERS ASSOCIATION, Plaintiff, Independent Bankers Association…

Court: United States District Court, D. Columbia

Date published: Mar 10, 1999

38 F. Supp. 2d 114 (D.D.C. 1999)

Opinion

No. Civ.A. 99-00042(CKK).

March 10, 1999.

Michael S. Helfer, Wilmer, Cutler Pickering, Washington, D.C., for plaintiff.

Leonard J. Rubin, Albert B. Krachman, Bracewell Patterson, L.L.P., Washington, D.C., for Plaintiff-Intervenor.

Arthur R. Goldberg, Lois Bonsal Osler, United States Department of Justice, Federal Programs Branch, Civil Division, Washington, D.C., for Defendant.

David M. Malone, Ronald R. Glancz, Venable, Baetjer, Howard Civiletti, L.L.P., Washington, D.C., Paul J. Lambert, Bingham Dana, L.L.P., Washington, D.C., for Defendant-Intervenor.

MEMORANDUM OPINION

KOLLAR-KOTELLY, District Judge.

For years, the National Credit Union Administration (NCUA) interpreted Section 109 of the Federal Credit Union Act (FCUA), 12 U.S.C. § 1759, to permit various employment groups, each one united by its own peculiar occupational bond but otherwise unrelated to another group, to coalesce and form a "multiple common-bond" credit union. See Interpretative Ruling and Policy Statement (IRPS) 82-1, 47 Fed.Reg. 16775 (1982). Early last year, after the issue had been volleyed twice between this Circuit's trial and appellate courts, the Supreme Court held that the NCUA's interpretation of Section 109 violated the unambiguous provisions of the FCUA. See National Credit Union Admin. v. First Nat'l Bank Trust Co., 522 U.S. 479, 118 S.Ct. 927, 939-40, 140 L.Ed.2d 1 (1998). Six months later, by overwhelming votes in both houses, Congress enacted the Credit Union Membership Access Act (CUMAA) "to amend existing law and specifically authorize multiple common bond federal credit unions." S.REP. No. 105-193, at 6 (1998). After a notice-and-comment period, the NCUA promulgated IRPS 99-1, which, since January 1, 1999, has set forth the criteria by which the agency now evaluates applications to add new groups to multiple common-bond credit unions. See 63 Fed.Reg. 71,998 (1998).

Eight days after the effective date of IRPS 99-1, Plaintiff American Bankers Association (ABA) submitted an application for a preliminary injunction, seeking to enjoin seven aspects of the rule on substantive grounds and the rule in its entirety due to an alleged procedural error.¹ Defendant NCUA and two Intervenors, the National Association of Federal Credit Unions (NAFCU) and the Credit Union National Association (CUNA), maintain that IRPS 99-1 comports with the plain language of the CUMAA, or where the statute is ambiguous, represents a reasonable interpretation that warrants deference from the judicial branch under Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843-44, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). The parties also disagree over the extent to which IRPS 99-1 threatens ABA member-banks with irreparable harm. Pending before the Court are the Plaintiffs' two applications for a preliminary injunction, the Defendants' oppositions, the reply memoranda thereto, and the Defendants' sur-replies. After reviewing the briefs and the arguments that counsel presented at a hearing held on March 1, 1999, the Court denies the applications.

1 Plaintiff-Intervenor Independent Bankers Association of America (IBAA) submitted a memorandum in support of its application for a preliminary injunction that echoes, almost verbatim, the memorandum filed by the ABA. Although the Court hereinafter refers only to the ABA, the holdings and rationales advanced in this Memorandum Opinion apply equally to both the ABA's and the IBAA's applications.

I. BACKGROUND

A. 1934-1998: Administrative Interpretation and Judicial Review of the FCUA

Enacted during the Great Depression, the FCUA authorizes the chartering of federal credit unions, which by statutory directive, are "cooperative association[s] organized . . . for the purpose of promoting thrift among [their] members and creating a source of credit for provident or productive purposes." 12 U.S.C. § 1752(1). Over the years, the FCUA has spawned the growth of almost 7000 federal credit unions. One variable that explains their ubiquity is that they are, unlike the banks and thrifts against which they compete for consumers, exempt from federal and state taxes. Congress, however, has bestowed this putative competitive advantage on credit unions because, unlike banks and thrifts, they "are member-owned, democratically operated, not-for-profit organizations generally managed by volunteer boards of directors . . . [that] have the specific mission of meeting the credit and savings needs of consumers, especially persons of modest means." CUMAA, Pub.L. No. 105-219, § 2(4), 112 Stat. 913, 914 (1998). By statutory directive, credit unions traditionally have been managed according to an ethos of volunteerism: But for one director, no member of a credit union's board of directors, supervisory committee, or any other committee may receive compensation for the services that he or she performs. See 12 U.S.C. § 1761(c), 1761a.

As distinct as a credit union's management structure is compared to that of its private counterpart, so too is its customer base. From its inception, the FCUA narrowly circumscribed "Federal credit union membership . . . to groups having a common bond of occupation or association, or to groups within a well-defined neighborhood, community, or rural district." FCUA, ch. 750, § 9, 48 Stat. 1216, 1219 (1934). Rooted in the cooperative spirit that animated the credit-union movement, the "common bond" requirement was intended to "ensure both that those making lending decisions would know more about applicants and that borrowers would be more reluctant to default. . . . The common bond was seen as the cement that united credit union members in a cooperative venture." First Nat'l Bank and Trust Co. v. NCUA, 988 F.2d 1272, 1274 (D.C. Cir. 1993).

For forty-eight years, the NCUA and its predecessors interpreted the FCUA's common-bond provision to require that every member of a credit union share the same common bond of occupation. Then, in 1982, the NCUA modified its enduring interpretation to permit multiple occupational groups, each one united by its own particular common bond, to coalesce into a single credit union regardless of whether any common denominator linked the disparate constituent groups. See IRPS 82-1, 47 Fed.Reg. 16775 (1982). In the years that followed, the NCUA reiterated and clarified its novel understanding of the FCUA, culminating in IRPS 89-1, which explained that "[a] select group of persons seeking credit union service from an occupational, associational or multiple group Federal credit union must have its own common bond," but "[t]he group's common bond need not be similar to the common bond(s) of the existing Federal credit union." 54 Fed.Reg. 31165, 31176 (1989).

IRPS 89-1 ushered in an era of unprecedented growth in the credit-union industry. AT T Family Federal Credit Union (ATTF), for example, though initially chartered with rather modest field-of-membership and geographic restrictions, swelled under IRPS 89-1 to provide services for 110,000 members in more than 150 distinct occupational groups throughout the United States. Operating for the benefit of not only AT T employees, ATTF swept within its ranks the employees of other corporate giants such as the Lee Apparel Company, the Coca-Cola Bottling Company, the Diba-Geigy Corporation, the Duke Power Company, and the American Tobacco Company. See First Nat'l, 118 S.Ct. at 931.

As more credit unions evolved under IRPS 89-1 into large multiple common-bond institutions, the ABA and similar trade groups feared that an increasing number of consumers would abandon traditional private banks and thrifts to pursue the lower interest rates on loans and higher yields on savings that credit unions have typically offered. In 1990, the banking industry fired the opening salvo, in what has proven to be a lengthy battle, when it challenged the NCUA's decision to expand ATTF's field of membership pursuant to IRPS 89-1. After the litigation traveled back and forth between the trial and appellate courts of this Circuit to resolve the banks' prudential standing under the Administrative Procedure Act, see First Nat'l, 988 F.2d 1272, 1275-79 (D.C. Cir. 1993), the D.C. Circuit subsequently held that IRPS 89-1 violated the unambiguously expressed intent of the FCUA. See First Nat'l Bank Trust Co. v. NCUA, 90 F.3d 525, 528-30 (D.C. Cir. 1996). In early 1998, the Supreme Court affirmed. See First Nat'l, 118 S.Ct. at 938-40. Like the court below it, the Supreme Court concluded that because "Congress has made it clear that the same common bond of occupation must unite each member of an occupationally defined federal credit union, . . . the NCUA's contrary interpretation is impermissible under the first step of Chevron." Id. at 938-39 (emphasis in original).

B. The CUMAA

Although the Supreme Court's First National decision quelled the judicial dispute over NCUA's multiple common-bond chartering policy, the debate continued, moving from an Article III case or controversy to an Article I lobbying initiative. What emerged from a summer-long legislative effort was the CUMAA. Enacted to "amend existing law and specifically authorize multiple common bond federal credit unions," S.REP. No. 105-193, at 6 (1998); accord H.R. REP. No. 105-472, at 18 (1998), U.S. Code Cong. Admin.News 1998, the CUMAA garnered broad support from both houses of Congress, passing by a vote of 411-8 in the House of Representatives and by a count of 92-6 in the Senate. Displacing the FCUA's traditional field-of-membership restrictions, the CUMAA established three distinct types of credit unions: single common-bond credit unions, multiple common-bond credit unions, and community credit unions. See 12 U.S.C. § 1759(b)(1)-(3).² A single common-bond credit union, defined as "[o]ne group that has a common bond of occupation or association," simply reflects the traditional common-bond requirement that the FCUA had always mandated. See 12 U.S.C. § 1759(b)(1).

2 In this background section, the Court sets aside the community credit union because it plays no role in the ABA's application for a preliminary injunction.

Of far greater importance to the parties in this case, however, is how Congress defined a multiple common-bond credit union. While, to be sure, the CUMAA unambiguously authorized the chartering of multiple common-bond credit unions, it did not simply ratify the NCUA's preexisting policy. Rather, Congress crafted "certain additional group size and geographic expansion limits," H.R. REP. No. 105-472, at 18, to ensure the "sense of cohesion or identity [that] is essential to the fulfillment of the public mission of credit unions." CUMAA, Pub.L. No. 105-219, § 2(3), 112 Stat. at 913. Under the CUMAA, a multiple common-bond credit union consists of more than one group

(A) each of which has (within the group) a common bond of occupation or association; and

(B) the number of members, each of which (at the time the group is first included within the field of membership of a credit union described in this paragraph) does not exceed any numerical limitation applicable under subsection (d).

12 U.S.C. § 1759(b)(2)(A)-(B). By enacting subparagraph (B), Congress significantly modified the NCUA's former multiple common-bond chartering policy by restricting the size of groups that may be added to a multiple common-bond credit union. Subsection (d), which contains these quantitative limitations, permits "only a group with fewer than 3,000 members [to] be eligible to be included in the field of membership" of a multiple common-bond credit union. § 1759(d)(1). Even groups with fewer than 3000 members, however, may not automatically enlist within the ranks of a multiple common-bond credit union; the NCUA is expressly directed to charter separate credit unions "instead of approving an application to include an additional group within the field of membership of an existing credit union whenever practicable and consistent with reasonable standards for the safe and sound operation of the credit union." § 1759(f)(1)(A).

Although groups with more than 3000 members presumptively may not consociate with a multiple common-bond credit union, the CUMAA recognizes certain exceptions to this general rule. The 3000-member limit does not apply with respect to

(A) any group that the Board determines, in writing and in accordance with the guidelines and regulations [promulgated by the NCUA], could not feasibly or reasonably establish a new single common-bond credit union . . . because —

(i) the group lacks sufficient volunteer and other resources to support the efficient and effective operation of a credit union;

(ii) the group does not meet the criteria that the Board has determined to be important for the likelihood of success in establishing and managing a new credit union . . .; or

(iii) the group would be unlikely to operate a safe and sound credit union.

§ 1759(d)(2)(A)(i)-(iii).³

3 By adding a new subsection (c) to § 1759, Congress also saw fit to provide, in the words of the House Report, "a broad grandfather [clause] for all persons and organizations who could be forced out of credit unions by the Supreme Court's decision in the AT T case." H.R. REP. No. 105-472, at 19 (1998). While the parties dispute the scope of this grandfather provision, no one questions that § 1759(c) authorizes multiple common-bond credit unions chartered under the former IRPS 89-1 regime to continue to provide services to individuals who were members on the date of the CUMAA's enactment regardless of whether the constituent groups satisfy the CUMAA's numerical restrictions.

Beyond size restrictions, the CUMAA also tempers the growth of multiple common-bond credit unions by imposing an important proximity requirement. When the NCUA determines that a group — whether it contains fewer or more than 3000 members — cannot independently charter a single common-bond credit union consistent with reasonable safety and soundness concerns, the agency must, whenever practicable in light of those same concerns, add the group to an existing credit union "that is within reasonable proximity to the location of the group." § 1759(f)(1)(B). How to gauge "reasonable proximity," however, is a task that Congress left to the NCUA to determine.

C. IRPS 99-1

On August 31, 1998, the NCUA issued a proposed rule to revise and update the agency's chartering and field-of-membership policies, many of which had been rendered obsolete after the President signed the CUMAA into law. See 63 Fed.Reg. 49164 (1998). After the sixty-day notice-and-comment period elapsed, the NCUA examined 369 comments submitted by various federal and state credit unions, national credit-union trade associations, banks and their trade associations, and other interested public and private entities. The agency's final rule, IRPS 99-1, was published in the Federal Register on December 30, 1998. Invoking the good-cause exception set forth at 5 U.S.C. § 553(d)(3), the NCUA bypassed the normal thirty-day waiting period between a final rule's publication and effective date; IRPS 99-1 took effect on January 1, 1999. See 63 Fed.Reg. at 72017.

IRPS 99-1 heralded a number of important and fundamental policy changes, many of which precipitated this latest round of litigation. Because each of the provisions that the ABA challenges is addressed later in this Memorandum Opinion, the Court need not explore the finer contours of IRPS 99-1 here. It is enough simply to note that the ABA seeks preliminary injunctive relief against seven discrete aspects of the final rule: the criteria for adding groups in excess of 3000 members to multiple common-bond credit unions, the standard for adding groups with fewer than 3000 members to multiple common-bond credit unions, the calculus by which the NCUA measures a group's membership, the definition of "reasonable proximity," the scope of "grandfathered" membership, the definition of a single occupational common bond, and rules governing voluntary mergers of financially sound multiple common-bond credit unions.

II. THE STANDARD FOR EVALUATING PRELIMINARY INJUNCTIONS

When considering an application for a preliminary injunction, federal courts in this Circuit examine whether: (1) there is a substantial likelihood that the plaintiff will succeed on the merits of its claims; (2) the plaintiff will suffer irreparable injury if an injunction does not issue; (3) an injunction will substantially injure other parties; and (4) the public interest will be furthered by interim injunctive relief. See Serono Lab. v. Shalala, 158 F.3d 1313, 1317-18 (D.C. Cir. 1998); Sea Containers Ltd. v. Stena AB, 890 F.2d 1205, 1208 (D.C. Cir. 1989); Washington Metro. Area Transit Comm'n v. Holiday Tours, Inc., 559 F.2d 841, 842 (D.C. Cir. 1977). "This calculus reflects a sliding-scale approach in which an injunction may issue if the arguments for one factor are particularly strong `even if the arguments in other areas are rather weak.'" Kelso v. United States Dep't of State, 13 F. Supp.2d 1, 3 (D.D.C. 1998) (quoting CityFed Fin. v. Office of Thrift Supervision, 58 F.3d 738, 746 (D.C. Cir. 1995)); accord Davenport v. International Bhd. of Teamsters, AFL — CIO, 166 F.3d 356, 360 (D.C. Cir. 1999) ("These factors interrelate on a sliding scale and must be balanced against each other.").

III. LIKELIHOOD OF SUCCESS ON THE MERITS

Although the ABA seeks to enjoin seven different rules set forth in IRPS 99-1, the issue presented in each instance is the same: whether the NCUA validly interpreted an act of Congress — in this case, the CUMAA. To resolve this inquiry, the Court looks to the Supreme Court's seminal decision in Chevron U.S.A, Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). The first task under Chevron is to ask "whether Congress has directly spoken to the precise question at issue." To determine whether Congress has, in fact, spoken directly to the question presented — an analytic process known as Chevron step one — the Court "must first exhaust the traditional tools of statutory construction." Natural Resources Defense Council, Inc. v. Browner, 57 F.3d 1122, 1125 (D.C. Cir. 1995). As always, this examination proceeds from "the fundamental canon that statutory interpretation begins with the language of the statute itself." Pennsylvania Dep't of Pub. Welfare v. Davenport, 495 U.S. 552, 557-58, 110 S.Ct. 2126, 109 L.Ed.2d 588 (1990). If, after employing the traditional tools of statutory construction, it appears that Congress has already spoken to the issue, "that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress." Chevron, 467 U.S. at 842-43, 104 S.Ct. 2778.

If, on the other hand, Congress has been silent or ambiguous with respect to the particular issue, the Court must defer to the agency's interpretation so long as it is "based on a permissible construction of the statute." Id. at 843, 104 S.Ct. 2778. Denominated as Chevron step two, this deferential inquiry looks only to whether the agency's interpretation is "reasonable and consistent with the statutory scheme and legislative history." Cleveland v. United States Nuclear Regulatory Comm'n, 68 F.3d 1361, 1367 (D.C. Cir. 1995); see also Chevron, 467 U.S. at 845, 104 S.Ct. 2778 (holding that if an agency's interpretation "represents a reasonable accommodation . . ., we should not disturb it unless it appears from the statute or its legislative history that the accommodation is not one that Congress would have sanctioned"). Moreover, as is the case here, the "view of the agency charged with administering the statute is entitled to considerable deference; and to sustain it, we need not find that it is the only permissible construction that [the NCUA] might have adopted." Chemical Mfrs. Ass'n v. Natural Resources Defense Council, Inc., 470 U.S. 116, 125, 105 S.Ct. 1102, 84 L.Ed.2d 90 (1985).

A. Whether IRPS 99-1 impermissibly liberalizes exceptions to the CUMAA's 3000-member limit on groups that can be added to multiple common-bond credit unions

Generally, the CUMAA forecloses groups with more than 3000 members from joining the field of membership of an existing credit union. See 12 U.S.C. § 1759(d)(1)-(2). Where a group of that same size, however, could not feasibly or reasonably charter its own credit union, Congress has authorized the NCUA to add the group to a multiple common-bond credit union. See § 1759(d)(2)(A). Under IRPS 99-1, among the factors that the NCUA will examine in considering a group's "economic advisability" are the "desire and intent of the group and the sponsor support." 63 Fed.Reg. at 72002. Seizing on this single sentence, the ABA contends that IRPS 99-1 violates § 1759(d)(2) because, when the NCUA considers whether to recognize an exception to the 3000-member restriction, its "determination turns essentially on whether the new group wants to form a separately chartered entity." ABA's App. for Prelim. Inj. at 15 (emphasis in original).

Recognizing that "statutory interpretation begins with the language of the statute itself," Butler v. West, 164 F.3d 634, 639 (D.C. Cir. 1999) (internal quotations omitted), the Court concludes that Congress has not spoken directly to the issue at hand. Deference to agency interpretation is warranted "when Congress has left a gap for the agency to fill pursuant to an express or implied `delegation of authority to the agency.'" Chevron, 467 U.S. at 843-44, 104 S.Ct. 2778. As this Circuit has explained the dichotomy of express and implied delegations, Congress generates interstices "explicitly by authorizing the agency to adopt implementing regulations, or implicitly by enacting an ambiguously worded provision that the agency must interpret." National Fuel Gas Supply Corp. v. FERC, 811 F.2d 1563, 1569 (D.C. Cir. 1987). Section 1759(d)(2)(A) is replete with both express and implied delegations. Perhaps most telling of the discretion that the NCUA enjoys is the second exception to the 3000-member limit that Congress adopted: "the group does not meet the criteria that the Board has determined to be important for the likelihood of success in establishing and managing a new credit union." § 1759(d)(2)(A)(ii) (emphasis added). Beyond this express authorization, Congress implicitly delegated to the NCUA the power to define such ambiguous terms as "sufficient volunteer and other resources," "efficient and effective operation," "other factors that may affect the financial viability and stability of a credit union," and "unlikely to operate a safe and sound credit union." § 1759(d)(2)(A)(i)-(iii). Statutory terms such as these, bound by no fixed meaning, compel judicial deference to reasonable agency interpretations.

The field-of-membership criteria set forth in IRPS 99-1 is eminently reasonable. As an initial matter, it would seem that the Plaintiffs' concerns about the final rule stem more from their own myopic reading of IRPS 99-1 than from an inherent flaw in the regulation itself. All that animates the ABA's objection is one spare sentence: "Important factors in making this determination [whether a group could charter an independent, financially sound credit union], however, are the desire and intent of the group and the sponsor support." 63 Fed.Reg. at 72002. To read the ABA's briefs, one would believe that desire and intent are all that the NCUA will ever examine. A more contextual and faithful reading of the final rule, however, belies the Plaintiffs' claims. In the sentence that immediately precedes the one just quoted, IRPS 99-1 provides: "As the legislation [CUMAA] directs, the Board will encourage the formation of separately chartered credit unions if it is prudent and economically advisable." Id. (emphasis added). It is only to resolve this fundamental question — whether it would be prudent and economically advisable for a group to charter its own credit union — that the group's desire, intent, and sponsor support become "[i]mportant factors in making this determination." Id. Notably, IRPS 99-1 does not state, as the ABA somewhat deceptively paraphrases, that group desire and intent are the "most important factors"; it only states that they are "important factors." Compare ABA's App. for Prelim. Inj. at 15 with 63 Fed.Reg. at 72002. Indeed, two sentences later, the regulation concludes: "While the intent of the group and sponsor support cannot be ignored and will carry great weight, they are not the sole factors. The final decision must be based on an independent regulatory analysis in consideration of the remaining factors specified in the regulation." 63 Fed.Reg. at 72002; see also id. at 72019-20 (enumerating other factors to be considered in determining economic advisability, including the proposed management's character and fitness, the group's proposed business plan, and present and future market conditions).

Although it might very well amount to arbitrary and capricious decision-making were the NCUA to add a group of more than 3000 members to a multiple common-bond credit union based on nothing more than desire and intent, IRPS 99-1 does not purport to establish such a regime. Notwithstanding the ABA's fears that the NCUA will elevate group desire and intent to dispositive importance, "agency action comes to us with a presumption of regularity." Advanced Micro Devices v. CAB, 742 F.2d 1520, 1546 (D.C. Cir. 1984). As the final rule plainly states, "[t]hose groups that can or should be able to meet [financial marketplace challenges], regardless of size, will be required to form a separate credit union." 63 Fed.Reg. at 72002. (emphasis added). The Court's conclusion that IRPS 99-1 is a reasonable interpretation of the CUMAA is predicated upon "certain assumptions about how the policy [will] be implemented," assumptions that are "based on a straightforward reading of the policy and on the Board's representations in its Brief to this court." Advanced Micro Devices, 742 F.2d at 1546. "Thus the parade of horribles that the [Plaintiffs] predict will result from implementation of [IRPS 99-1], if in fact it does result, will have to be dealt with in future litigation." Id. Having won the right to contest NCUA chartering decisions in First National, the ABA may proceed on a case-by-case basis if it suspects the NCUA of arbitrary and capricious agency action.⁴

4 Empirical data collected during the first six weeks that IRPS 99-1 governed NCUA's chartering decisions suggest that the ABA's fears may be more imaginary than real. Of the 1786 separate groups that have come before the NCUA for chartering decisions, only eight have contained more than 3000 members. Most significantly, not one of these eight groups has been permitted to join a multiple common-bond credit union; three have been denied and five have been deferred for insufficient documentation. See Congress Improved Access by Small Groups (chart presented to the Court at the March 1, 1999 hearing, summarizing data gathered from NCUA's Regional Offices for the period January 1, 1999 through February 12, 1999); Tr. of Hr'g, Mar. 1, 1999, at 63:16-:21 (statement of counsel for the NCUA).

Having placed IRPS 99 1 in a proper context, the question remains whether group desire and intent and sponsor support are factors that are reasonably consistent with statutory structure and legislative history. See City of Cleveland v. NRC, 68 F.3d 1361, 1367 (D.C. Cir. 1995). There can be little doubt that they are. Federal credit unions, since their inception in 1934, have been and remain cooperative enterprises. See 12 U.S.C. § 1752(1). In the CUMAA itself, Congress expressly observed in its findings that credit unions "are member-owned, democratically operated, not-for-profit organizations generally managed by volunteer boards of directors." CUMAA, Pub.L. No. 105-219, § 2(4), 112 Stat. 913, 914 (1998). Obviously cognizant of the cooperative and voluntary nature inherent in credit-union management, Congress provided that a group with more than 3000 members could nonetheless join an existing credit union if it "lack[ed] sufficient volunteer and other resources to support the efficient and effective operation of [a separate] credit union." § 1759(d)(2)(A)(i) (emphasis added). Where Congress has explicitly recognized that insufficient volunteer resources may justify an exception to the CUMAA's 3000-member limit, it unquestionably is reasonable for the NCUA to inquire into a group's desire, intent, and sponsor support — three proxies well suited to evaluating a group's volunteer resources.

Moreover, the CUMAA authorizes the NCUA to waive the 3000-member restriction when a "group does not meet the criteria that the Board has determined to be important for the likelihood of success in establishing and managing a new credit union, including . . . factors that may affect the financial viability and stability of a credit union." § 1759(d)(2)(A)(ii). IRPS 99-1 is entirely consistent with this rather broad grant of discretion. In an institution where all but one of its directors and executive committee members must forgo compensation, see §§ 1761(c), 1761a, the desire and intent of its organizers and members are peculiarly important in assessing that entity's stability and viability. Moreover, at the same time that they manage without remuneration, a credit union's board of directors may nevertheless be held personally liable under applicable state law for actions taken in their official capacity.⁵ Whether a group will be able to attract and maintain qualified directors and committee members who are willing to accept these conditions of service is an important area into which the agency may delve. For the ABA to suggest that the NCUA flaunts the will of Congress when it examines group desire, intent, and sponsor support ignores the nature of credit unions and the structure of the FCUA and the CUMAA.

5 Under the FCUA, a credit union's board of directors are also prone to civil monetary penalties for, inter alia, violating any law or regulation, recklessly engaging in unsafe or unsound practices, and knowingly or recklessly causing a substantial loss to the credit union. See 12 U.S.C. § 1786(k)(2).

Finally, nothing in the various committee reports impeaches the validity of IRPS 99-1. Although not particularly enlightening, the Senate Report indicates that the CUMAA permits the NCUA to add a group with more than 3000 members to an existing credit union if it lacks "volunteers." S.REP. No. 105-193, at 7. To be sure, both houses did "not intend for these exceptions to provide broad discretion to the Board to permit larger groups to be incorporated within or merged with other credit unions." H.R. REP. No. 105-472, at 19; accord S.REP. No. 105-193, at 7. Yet this only means that Congress did not intend the NCUA to add large groups to multiple common-bond credit unions based on considerations divorced from safety and soundness. See H.R. REP. No. 105-472, at 19 ("The exceptions are intended to apply where the Board has sufficient evidence to support a finding that creation of a separately chartered credit union . . . present[s] safety and soundness concerns."). Certainly, where safety and soundness are at issue, the CUMAA delegates considerable discretion to the NCUA to determine whether a group of more than 3000 members would be able to "feasibly or reasonably" charter its own single common-bond credit union. See 12 U.S.C. § 1759(d)(2)(A)(i)-(iii). Assessing a group's desire and intent in chartering its own credit union is an inquiry that is reasonably calculated to resolve whether the group could feasibly or reasonably operate an independent credit union. For the foregoing reasons, the ABA has failed to establish a substantial likelihood of prevailing on the merits of this claim.

B. Whether IRPS 99-1 violates the CUMAA by taking a "hard look" at the ability of a group with fewer than 3000 members to charter its own single common-bond credit union

Depending on whether a group's membership exceeds or falls short of 3000 members, IRPS 99-1 establishes different standards for assessing whether to add the group to a multiple common-bond credit union. Groups with more than 3000 members "must be able to demonstrate why they cannot satisfactorily form a separate credit union if they want to be added to another credit union." 63 Fed.Reg. at 72001. On the other hand, groups with fewer than 3000 members "must be able to demonstrate why they can successfully operate a credit union." Id.; see also id. at 72000 ("[A] charter applicant with a proposed field of membership of fewer than 3,000 primary potential members may have to provide more support than a proposed credit union with a larger field of membership in order to demonstrate that it is economically advisable and that it will have a reasonable chance to succeed."). The ABA suggests that this differential review violates the CUMAA by creating an impermissible presumption that a group with fewer than 3000 members cannot charter a separate credit union.

Congress's intent, whatever it may be, is by no means "express" or "clear" as the ABA maintains. Far from having "directly spoken to the precise question at issue," Chevron, 467 U.S. at 842, 104 S.Ct. 2778, Congress said very little in the CUMAA about groups with fewer than 3000 members. Unlike their larger counterparts, groups below the 3000-member threshold need not demonstrate that they fit within one of the exceptions set forth at § 1759(d)(2)(A)(i)-(iii) before being added to an existing credit union. See § 1759(d)(1). Rather, the only check that Congress imposed was a general command to the NCUA to

encourage the formation of separately chartered credit unions instead of approving an application to include an additional group within the field of membership of an existing credit union whenever practicable and consistent with reasonable standards for the safe and sound operation of the credit union.

§ 1759(f)(1)(A). Thus at an abstract level, Congress's intent is manifestly clear: where practicable with safety and soundness concerns, the NCUA should a charter a separate credit union before adding a group to a multiple common-bond credit union. Yet precisely how the NCUA implements this aspiration is a matter Congress never addressed. The necessary policy choices that the NCUA must make, therefore, warrant judicial deference to a reasonable agency interpretation under Chevron step two. See National Ass'n of Mfrs. v. DOI, 134 F.3d 1095, 1106 (D.C. Cir. 1998); Kennecott Utah Copper Corp. v. DOI, 88 F.3d 1191, 1206 (D.C. Cir. 1996)

The parties disagree over how this portion of the rule should be characterized; the ABA describes it as a "presumption" while the NCUA suggests that it is merely a benign "threshold." Compare ABA's App. for Prelim. Inj. at 16-17 with 63 Fed.Reg. at 72001. This debate over nomenclature, however, is hardly illuminating; for whether IRPS 99-1 is classified as a presumption or simply a threshold (whatever the difference between the two may be), the question remains: Will the rule, when implemented, transgress the CUMAA? To answer this, perhaps the obvious needs to be stated first: Congress clearly perceived some difference between groups with more than 3000 members and those with fewer than 3000 members when it enacted the CUMAA. Were group size irrelevant, Congress either would have dispensed entirely with the detailed exceptions to the 3000-member limit in § 1759(d)(2) or required all groups, irrespective of size, to satisfy § 1759(d)(2)'s criteria. From this distinction, concededly, it cannot be inferred that all groups under 3000 were presumed by Congress to be incapable of chartering their own credit unions. The legislative history makes that clear. See H.R. REP. No. 105-472, at 19 ("The Committee does not intend for this numerical limitation to be interpreted as permitting all groups with 3,000 or fewer members to be included within the field of membership of an existing credit union."). Yet Congress's two-tiered approach in the CUMAA reflects an understanding that, in contrast to larger groups, those with fewer than 3000 members are generally less likely to be able to meet the demands of chartering their own credit union. For the NCUA to acknowledge this reality by taking a harder look at the economic advisability of smaller groups is hardly unreasonable. Treating groups differently based on whether their membership rolls eclipse 3000, therefore, finds support in the structure of the CUMAA.

Moreover, Congress's command to "encourage the formation of separately chartered credit unions" is not absolute. The duty arises only when to do so would be "practicable and consistent with reasonable standards for the safe and sound operation of the credit union." § 1759(f)(1)(A). As it did throughout the CUMAA, here again Congress designated safety and soundness concerns as the lodestars that should ultimately guide the NCUA's chartering decisions. The proper question, therefore, is not whether IRPS 99-1 has the effect of discouraging the formation of separate credit unions, but whether it discourages the formation of financially safe and sound separate credit unions.

This it does not do. In the NCUA's considerable experience, it has found that the modern financial-services market renders small credit unions peculiarly susceptible to insolvency. During 1998, for example, all eighteen credit unions that the NCUA either involuntarily liquidated or merged with other institutions had 3000 or fewer members. See NAFCU Opp'n to Mot. for Prelim. Inj. at Ex. 4 (Decl. of Dr. Tun A. Wai ¶ 6). As of June 1998, over 65% of the federally insured credit unions that were suffering from severe financial difficulties had memberships under 3000. See id. (Decl. of Wai ¶ 4). Moreover, every federal credit union that had been classified with the lowest financial-performance rating had 3000 or fewer members. See id. Nonetheless, as IRPS 99-1 candidly acknowledges, today there are approximately 3100 federal credit unions with fewer than 3000 members. See 63 Fed.Reg. at 72001. The vast majority of them, however, were chartered long before the advent of sophisticated electronic banking services, when both the economic conditions and the financial-services expectations of credit-union members were dramatically different. These differences, in turn, "provided the credit unions an opportunity to become established and develop a loyalty base under marketplace expectations that significantly differ from those of today." Id. Presently, to be competitive and financially viable, a new credit union must be able to provide the full panoply of services — ATMs, debit cards, and electronic banking just to name a few — that modern banks, thrifts, and established credit unions offer as standard fare. Reflecting how difficult modernity has made it for a group with a small membership base to charter its own credit union, of the twenty-nine credit unions chartered since 1996, only one had fewer than 3000 members. See 63 Fed.Reg. at 7200.⁶

6 Although "[t]he remarks of a single legislator, even the sponsor, are not controlling in analyzing legislative history," Chrysler Corp. v. Brown, 441 U.S. 281, 311, 99 S.Ct. 1705, 60 L.Ed.2d 208 (1979), the comments of Rep. Vento underscore the "need to modernize the credit union field of membership definitions which do not fit the socio-economic reality of the 1990's. . . . Credit union law needs to be modernized, addressing the membership base of credit unions because they would not be able to sustain a membership base and reasonable services under the strict interpretation of a 1934 federal credit union law." 144 CONG.REC. H1868-02, H1874 (1998) (statement of Rep. Vento).

By emphasizing fiscal security throughout the CUMAA, Congress obviously did not intend that the NCUA would turn a blind eye to market conditions, consumer expectations, and systemic barriers to entry when making its chartering decisions. In fact, because federal credit unions are insured by the National Credit Union Share Insurance Fund (NCUSIF), Congress and the NCUA share a unique concern for the financial integrity of any new charter. Obligated to preserve the corpus of the NCUSIF, the NCUA may draw upon its expertise and experience in adopting regulations that attempt to measure a prospective credit union's economic advisability. While the NCUA's regulations subject groups containing fewer than 3000 members to greater scrutiny, greater circumspection is reasonable and by no means inconsistent with Congress's intent to charter separate, financially sound credit unions. To be sure, such a rule may very well discourage the formation of some single common-bond credit unions. But that is the purpose of the final rule: to encourage not only "new charters, but also ensure to the fullest extent possible that those groups receiving a separate charter will have a reasonable basis for success and thereby avoid unnecessary risks to the NCUSIF." 63 Fed.Reg. at 72001.

Nor does this "hard look" mean that the NCUA will automatically approve an application to add a group with fewer than 3000 members to an existing credit union. For though the NCUA "recognizes that newly chartered credit unions in today's financial marketplace have unique challenges," IRPS 99-1 nonetheless directs that "[t]hose groups that can or should be able to meet those challenges, regardless of size, will be required to form a separate credit union." Id. at 72002 (emphasis added). Since January 1, 1999, when IRPS 99-1 became effective, the NCUA has permitted 2610 groups to join multiple common-bond credit unions. Though that number may seem large, half of these groups contained only 28 or fewer members, see NCUA Submission of Additional Info. Requested by the Court ¶ 2 (indicating that as of February 26, 1999, the median number of members among groups added to multiple common-bond credit unions is 28), and on average, each group boasted approximately⁷ 76 members. See Congress Improved Access by Small Groups (Data from NCUA's Regional Offices for the period Jan. 1, 1999 through Feb. 12, 1999 presented to the Court at the March 1, 1999 hearing). Where the median and mean composite of these groups reflects such low membership, it should cause no great alarm to learn that the NCUA has determined that most would be unable to charter an independent credit union. Nonetheless, the NCUA has denied 69 applications based on, inter alia, safety and soundness concerns and the ability of the group to form a separately-chartered single common-bond credit union. See NCUA Submission of Additional Info. Requested by the Court ¶ 4. For the NCUA to scrutinize the financial viability and resources of a group under 3000 members more rigorously than it does a group of more than 3000 members is quite consistent with congressional intent. The ABA, therefore, is unlikely to succeed on the merits in proving that IRPS 99-1 is an unreasonable interpretation of the CUMAA.

7 On this point, the record is not perfectly clear. As of February 12, 1999, the mean value of the 1786 groups added to multiple common-bond credit unions was 76. When this data were presented at the March 1, 1999 hearing, the Court asked the NCUA to calculate the median value as well in order to guard against statistical outliers that might have skewed the mean. In its response, the NCUA reported the median size of these groups as of February 26, 1999, but did not calculate the updated mean value. Although comparing data culled from two sets of records would rightly offend statisticians, for the purposes of adjudicating this application for a preliminary injunction, the information offers a rough appreciation for how IRPS 99-1 has operated.

C. Whether IRPS 99-1 violates the CUMAA by excluding family and household members when calculating the membership of a group

Because under both the CUMAA and IRPS 99-1, a group consisting of more than 3000 members presumptively may not join a multiple common-bond credit union, how group size is measured assumes an important role. Under IRPS 99-1, the only relevant variables that factor into this calculus are individuals designated as "primary potential members" — the actual employees in an occupational common bond or the actual members of an associational common bond. See 63 Fed.Reg. at 72000. The ABA assails the primary potential member classification as contrary to the unambiguous intent of the CUMAA. For the reasons that follow, the Court concludes that the ABA is highly unlikely to succeed on the merits of this claim.

Undergirding the ABA's attack is a fundamentally flawed understanding of § 1759(e)(1) and how it relates to the remainder of the CUMAA. Paragraph (1) provides:

No individual shall be eligible for membership in a credit union on the basis of the relationship of the individual to another person who is eligible for membership in the credit union, unless the individual is a member of the immediate family or household (as those terms are defined by the Board, by regulation) of the other person.

12 U.S.C. § 1759(e)(1). As the ABA reads this provision, family and household members must be counted when calculating group size because "[t]hey have full membership rights identical to those of so-called primary members, by both statute and regulation." ABA's App. for Prelim. Inj. at 20. Section 1759(e)(1), however, does not automatically bestow credit-union membership on family and household members. Had Congress wished in all cases to draw family and household members into the credit union's field of membership, it would have provided that such individuals "shall be members." Instead, Congress used less definitive language, providing only that family and household members "shall be eligible for membership." § 1759(e)(1) (emphasis added). Recognizing this distinction, IRPS 99-1 provides that, at "the charter applicant's option," immediate family and household members, among others, may be added to the field of membership. 63 Fed.Reg. at 72027 (emphasis added).⁸

8 By fusing together stray clauses of the final rule, the ABA strives mightily to manufacture an internal tension within IRPS 99-1. According to the ABA's creative editing, IRPS 99-1 "list[s] among the `other persons sharing [the] common bond' of a given group `member[s] of the immediate family or household.'" ABA App. for Prelim. Inj. at 20 (quoting selectively 63 Fed.Reg. at 72027). What this provision actually states is: "A number of persons, by virtue of their close relationship to a common bond group, may be included, at the charter applicant's option, in the field of membership." 63 Fed.Reg. at 72027 (emphasis added). Contrary to the ABA's conception of it, this portion of the rule permits individuals such as family and household members to be added as credit-union members not because they constitute part of a common-bond group but because they maintain a close relationship to a common-bond group. Although this provision falls under a caption titled "Other Persons Sharing Common Bond," the clear language of the provision itself resists the ABA's fanciful misquoting.

Yet even were family and household members entitled, as a matter of law, to benefit always from credit-union membership, the ABA would be no more likely to succeed on the merits. Section 1759(d)(1), which contains the numerical limitations that make membership calculations so important, provides that "only a group with fewer than 3,000 members shall be eligible to be included in the field of membership" of a multiple common-bond credit union. § 1759(d)(1) (emphasis added). By employing the word "group," Congress refers back to § 1759(b)(1)-(2), which restricts "membership" in a single or multiple common-bond credit union to one "group" or more "groups" that share "a common bond of occupation or association." § 1759(b)(1)(2). Family and household members, by virtue of nothing more than their close relationship with a primary member, do not share a "common bond of occupation or association."

Section 1759(e)(1), though it renders family and household members "eligible for membership in a credit union," does not declare that they are members of a group. § 1759(e)(1) (emphasis added). Indicative of this distinction, § 1759(e) is captioned: "Additional Membership Eligibility Provisions." By "additional," Congress strongly suggests that it did not understand family and household members to compose part of a common bond, but simply wished to permit them to enjoy the advantages of credit-union membership, notwithstanding that they lacked any meaningful connection to the rest of the group. What little legislative history there is on this issue buttresses the Court's conclusion. See H.R. REP. No. 105-472, at 12 ("Credit union members in the occupation category are employed by the same enterprise, or in the same trade. As associational common bond is available to groups of individuals who participate in activities that develop common loyalties, mutual benefits, and mutual interests."). The ABA has not demonstrated a substantial likelihood of succeeding on the merits.

D. Whether IRPS 99-1 interprets the CUMAA's "reasonable proximity" requirement in an impermissibly expansive manner

Where the NCUA determines that a group with fewer than 3000 members cannot charter its own financially safe credit union or that a group with more than 3000 members satisfies an exception in § 1759(d)(2)(A), the agency may incorporate that group into the field of membership of a multiple common-bond credit union. In determining which existing credit union should absorb the new group, the agency must honor the CUMAA's "reasonable proximity" requirement. See 12 U.S.C. § 1759(f)(1)(B). Believing that "credit union members who live, work and interact in the same geographic area are likely to have more of a meaningful and common bond than those who do not," H.R. REP. No. 105-472, at 20, Congress directed the NCUA to include a group "in the field of membership of a credit union that is within reasonable proximity to the location of the group." § 1759(f)(1)(B).

In its attempt to define more concretely the parameters of "reasonable proximity," the NCUA operated from the premise that the requirement imposed a geographic limitation — that the group "must be within reasonable proximity geographically to the credit union." 63 Fed.Reg. at 72002. A credit union's main office, however, is not the sole point from which a group's geographic proximity is measured under the final rule. Purporting to exploit the "advantages acquired from advancing technologies," IRPS 99-1 provides that "the group to be added must be within the service area of a service facility of the credit union." Id. A service facility, in turn, encompasses a credit-union-owned branch, a shared branch, a mobile branch that provides services at the same location on a weekly basis, and a "credit union owned electronic facility." Id. Although the Plaintiffs' briefs suggest that the agency's definition extends as far as an ATM, "the final rule excludes an ATM as a service facility." Id.

As the ABA concedes, the "CUMAA does not expressly define the phrase `reasonable proximity.'" ABA App. for Prelim. Inj. at 23. This Court's review, accordingly, is limited to the deferential contours of Chevron step two. "Deference, however, does not mean abdication of careful and thorough judicial review." Amerada Hess Pipeline Corp. v. FERC, 117 F.3d 596, 604 (D.C. Cir. 1997) (quoting Baltimore Gas Elec. Co. v. FERC, 26 F.3d 1129, 1135 (D.C. Cir. 1994)). The Court will examine the statutory scheme and legislative history to determine whether the agency's interpretation is reasonable. See Military Toxics Project v. EPA, 146 F.3d 948, 954-55 (D.C. Cir. 1998). Nonetheless, it is important to recognize that where, as here, Congress has "fail[ed] to legislate in sufficient detail as to resolve a particular question of interpretation," the "overriding principle is that as long as Congress has no clearly discernable intent on the point in question, it is the agency which is vested with primary responsibility for interpreting the statute." Investment Co. Inst. v. Conover, 790 F.2d 925, 935 (D.C. Cir. 1986).

The ABA's claim of unreasonableness, predicated on isolated comments from a House legislator and equivocal observations in the House report, fail to demonstrate that the NCUA impermissibly interpreted the CUMAA's reasonable proximity requirement. Turning first to the legislative history, the ABA stakes great reliance on the following passage: "The term `facility' in the Act is meant to be defined in the same way that the [NCUA] has defined `service facility,' that is, an automatic teller machine or similar device would not qualify." H.R. REP. No. 105-472, at 19. This single sentence hardly upsets the validity of the NCUA's interpretation. Notably, this portion of the Report pertains not to the reasonable proximity requirement set forth in § 1759(f)(1)(B), but to an entirely different section that addresses exceptions to the common-bond requirements for underserved areas. See § 1759(c)(2)(A)(ii). Moreover, in defining the term "facility," the House Report limits its discussion to how "[t]he term `facility' in the Act is meant to be defined." H.R. REP. No. 105-472, at 19 (emphasis added). The only section of the CUMAA in which Congress used the word "facility" was § 1759(c)(2)(A)(ii)'s exception for underserved areas; never does the word "facility" appear in the reasonable proximity provision of the CUMAA.

Nonetheless, even if the House Report accurately summarizes how Congress wanted the NCUA to define service facility in IRPS 99-1, the final rule retains its validity. By extending the definition of service facility to reach credit-union-owned electronic facilities, IRPS 99-1, according to the ABA, violates Congress's understanding that "an automatic teller machine or similar device would not qualify" as a service facility. H.R. REP. No. 105-472, at 19. As already discussed, the final rule explicitly excludes ATMs from the definition. See 63 Fed.Reg. at 72002. Assuming, as the ABA does, that the NCUA's definition must conform precisely to the letter of the House Report, the question becomes whether a credit-union-owned electronic facility is "similar" to an ATM. Unfortunately, the record with respect to this matter is not as complete as it could be. It appears that, unlike an ATM, a credit-union-owned electronic facility can accept a member's loan application and disburse the proceeds from approved loans.⁹ Beyond these differences, in neither their briefs nor in their answers to this Court's questions at oral argument, have the Defendants identified any other function that distinguishes an ATM from a credit-union-owned electronic facility.

9 By the terms of IRPS 99-1, a credit-union-owned electronic facility must be able to offer these services. See 63 Fed.Reg. at 72003 ("At a minimum, to qualify as a service facility, the member must be able to deposit funds, apply for a loan, and obtain funds on approved loans.").

Whatever the precise difference between these two machines may be, the reasonableness of IRPS 99-1 does not rise or fall based on how distinct an ATM is from a credit-union-owned electronic facility. In the past, the NCUA has defined a service facility as a place where: "(1) Shares are accepted for members' accounts; (2) loan applications are accepted or loans are disbursed; (3) a member can deal directly with a credit union representative; and (4) the service provided is clearly associated with that particular credit union." IRPS 94-1, 59 Fed.Reg. 29066, 29078 (1994). All that IRPS 99-1 has arguably changed is that, now, a member need not have to deal directly with a credit-union representative. See 63 Fed.Reg. at 72003 (requiring a service facility to be able to accept deposits, accept loan applications, and disburse loan proceeds). Congress, however, did not indicate that personal service was the glue that binds a credit union. Rather, the House Report "believe[d] credit union members who live, work and interact in the same geographical area are more likely to have more of a meaningful affinity and common bond than those who do not." H.R. REP. No. 105-472, at 20 (emphasis added). True, the proponent of the provision intended "that NCUA give a conservative interpretation to the term `reasonable proximity.'" 144 CONG.REC. H7037, H7050 (statement of Rep. LaFalce). But, "[t]he remarks of a single legislator, even the sponsor, are not controlling in analyzing legislative history." Independent Bankers Ass'n v. Farm Credit Admin., 164 F.3d 661, 668 (D.C. Cir. 1999) (quoting Chrysler Corp. v. Brown, 441 U.S. 281, 311, 99 S.Ct. 1705, 60 L.Ed.2d 208 (1979)). On the present record, the ABA is unlikely to be able to demonstrate that an electronic facility impermissibly attenuates the CUMAA's reasonable proximity requirement.

E. Whether IRPS 99-1 defines a single common bond too expansively

Congress restricted membership in a single common-bond credit union to "[o]ne group that has a common bond of occupation or association." 12 U.S.C. § 1759(b)(1). IRPS 99-1 recognizes a number of circumstances under which persons may share a single occupational common bond. See 63 Fed.Reg. at 72022. Among these, the ABA takes issue with two. The first, which the Court will denominate as the 10-percent-ownership rule, provides that: "Employment in a corporation or other legal entity with a controlling ownership interest (which shall not be less than 10 percent) in or by another legal entity makes that person part of a single occupational common bond." Id. The second, which the Court shall refer to as the dependency-relationship rule, reads: "Employment in a corporation or other legal entity which is related to another legal entity (such as a company under contract and possessing a strong dependency relationship with another company) makes that person part of a single occupational common bond." Id. The ABA maintains that these definitions unreasonably interpret the CUMAA by including within a single occupational common bond, what it perceives to be, "multiple employer groups having little or no meaningful affinity." ABA App. for Prelim. Inj. at 25.

Of the many challenges to IRPS 99-1 that the ABA has mounted in its Application for a Preliminary Injunction, this is the easiest to dispose of. Although this Circuit invalidated the NCUA's former multiple common-bond chartering policy, it held that, with respect to single occupational common bonds, "[t]he NCUA may identify and approve other types of common bonds, subject only to the rule of reason embedded in Chevron step two." First National, 90 F.3d at 529. To resolve the first issue issued presented by the ABA, however, the Court need not even resort to the Chevron framework; for as it turns out, the ABA contests a phantom regulation that nowhere appears in IRPS 99-1. In its initial Application and again in its Reply brief, the ABA reads the 10-percent-ownership rule to mean that so long as one company owns 10 percent of another company, both companies' employees would share a single occupational common bond. See ABA App. for Prelim. Inj. at 27; ABA Reply at 12 (indicating that, so far as the ABA understands the 10-percent-ownership rule, a single occupational common bond would exist "regardless of any other facts, including whether some other entity owns the other 90 percent of Company B"). What eludes the ABA is the rule's fundamental limiting provision: for a single occupational common bond to exist among employees of two companies, one company must not only possess at least a 10 percent interest in the other, but its interest must be controlling. See 63 Fed.Reg. at 72022 (permitting single occupational common bond where one corporation has "a controlling ownership interest (which shall not be less than 10 percent)") (emphasis added). Thus, if Company A owns 15 percent of Company B, but Company C owns 30 percent of Company B, IRPS 99-1 would not recognize a single occupational common bond between the employees of Company A and Company B. Because the ABA contests a rule that the NCUA has never promulgated, it enjoys no likelihood of succeeding on the merits of its claim.

With respect to the dependency-relationship rule, the ABA dedicates only two sentences in all of its briefs to this issue. See ABA App. for Prelim. Inj. at 27-28 ("Similarly, the `dependency relationship' exception would allow large suppliers to be bonded with almost a limitless number of customers and service providers. For example, Microsoft could presumably be in the same `single' common bond as nearly all makers of computer software and hardware, including those that have an adversarial relationship with Microsoft, and its outside law firms."); ABA Reply at 12 (offering no argument with respect the dependency-relationship rule). Assessed simply for reasonableness, the dependency-relationship rule withstands the ABA's faint attack.

Unlike many of the other rules that IRPS 99-1 sets forth, the dependency-relationship rule is not of recent vintage. As far back as 1989, the NCUA permitted employees of entities bound by a strong dependent relationship to join in a single occupational common bond. See IRPS 89-1, 54 Fed.Reg. 31165, 31169 (1989). That this interpretation has endured for a decade is significant for many reasons, not the least of which is that the NCUA's "consistent and longstanding interpretation" as "the agency charged with administration of the [FCUA], while not controlling, is entitled to considerable weight." United States v. National Ass'n of Sec. Dealers, Inc., 422 U.S. 694, 719, 95 S.Ct. 2427, 45 L.Ed.2d 486 (1975); see also Anderson Shipping Co. v. EPA, 852 F.2d 1387, 1391 (D.C. Cir. 1988) ("We are inclined to give considerable weight to this longstanding Agency interpretation of the statute entrusted to its administration."). But beyond doctrinal significance, it is noteworthy that the parade of horribles forecasted by the ABA has yet to manifest. Surely, were the dependency-relationship rule so expansive that a corporate behemoth could form a single occupational common-bond credit union with each of its contractors, the ABA would be able to identify at least one. Looking at how the NCUA has traditionally applied the dependent-relationship rule, though, it is clear why the ABA's fears have yet to materialize: The rule has been used "primarily to determine when military base credit unions should be permitted to serve the employees of contractors operating on the base." NAFCU Opp'n at 18. Accordingly, recognizing that the "NCUA may identify and approve other types of common bonds, subject only to the rule of reason embedded in Chevron step two," First National, 90 F.3d at 529, the Court cannot say that the NCUA's longstanding dependency-relationship rule is an unreasonable interpretation of the FCUA or the CUMAA. The ABA, therefore, is unlikely to succeed on the merits.¹⁰

F. Whether IRPS 99-1 violates the CUMAA by interpreting too permissively the Act's grandfather clause

10 By only challenging the dependent-relationship rule now for the first time, the ABA can hardly claim that, without an injunction, it will suffer irreparable injury. Thus quite apart from whether its challenge to the rule is likely to succeed on the merits, the ABA would not be entitled to a preliminary injunction against this aspect of IRPS 99-1. See Beacon Theatres, Inc. v. Westover, 359 U.S. 500, 506-07, 79 S.Ct. 948, 3 L.Ed.2d 988 (1959) ("The basis of injunctive relief in the federal courts has always been irreparable harm and inadequacy of legal remedies.").

When Congress enacted the CUMAA, its most pressing and obvious purpose was to abrogate the Supreme Court's ruling in First National by expressly authorizing the NCUA to charter multiple common-bond credit unions. See H.R. REP. No. 105-472, at 18 ("[R]egarding the Supreme Court's decision on the common bond issue in the AT T case[,] [t]he Committee has determined that it is appropriate to change existing law and specifically authorize multiple common bond federal credit unions."); accord S.REP. No. 105-193, at 6. At the same time, Congress also sought to preserve the credit unions that had been chartered under the old, albeit unlawful, NCUA regulations. This grandfather clause provides:

[A] member of any group whose members constituted a portion of the membership of any Federal credit union as of that date of enactment shall continue to be eligible to become a member of that credit union, by virtue of membership in that group, after that date of enactment.

12 U.S.C. § 1759(c)(1)(A)(ii). There is no disagreement between the parties that the grandfather clause permits persons to become credit-union members if, as of the CUMAA's enactment, they were members of a group that formed part of a multiple common-bond credit union.¹¹ IRPS 99-1, however, extends the grandfather provision not only to individuals who were group members as of the CUMAA's enactment but also to all persons who subsequently become members of the group. Specifically, IRPS 99-1 "permits a member, or subsequent new member, of any group whose members constituted a portion of the membership of any federal credit union at the date of enactment, to continue to be eligible for membership in the credit union." 63 Fed.Reg. at 72015 (emphasis added).

11 A separate clause of the CUMAA grandfathers persons who already were members of a credit union. See 12 U.S.C. § 1759(c)(1)(A)(i).

The ABA argues that IRPS 99-1 violates the unambiguous and plain language of the CUMAA's grandfather clause. Because only someone who was a "member of any group" as of the CUMAA's enactment can " continue to be eligible to become a member of that credit union," § 1759(c)(1)(A)(ii), the ABA contends that the NCUA may not stretch the grandfather clause so wide as to reach persons who later become members of a group. Under this plausible interpretation, a person cannot possibly "continue" to be eligible for membership in an entity if he or she has never previously been eligible.

Without explaining why, the NCUA conclusorily asserts that the grandfather provision is "somewhat ambiguous" because it fails to distinguish between "members on August 7, 1998, and members joining the group at a later date." NCUA Opp'n at 39. Hoping that, because it thinks the statute to be ambiguous so too might the Court, the NCUA wastes no time retreating to legislative history to vindicate its interpretation under Chevron step two. The NAFCU, on the other hand, articulates a slightly more sophisticated interpretation of the grandfather clause. "A member of a group," according to the NAFCU, "is a term of art" that defies the literal meaning ascribed to it in the ABA's briefs. See NAFCU Opp'n at 19. Although the NAFCU cites no interpretative regulations or decisions to support its claim, it maintains that in the context of field-of-membership decisions, the NCUA has construed the term "a member of a group" to encompass a sponsoring entity's current and future employees, without distinction. Under this conception of the grandfather clause, "a member of a group" refers not to an individual within the group but to the general characteristics of the group's membership, i.e., employees of Company A. Read in this manner, Congress intended to grandfather all persons who currently share or will come to share the bond that unites that particular group. Although not as natural a reading as the one that the ABA advances, the NAFCU's interpretation of the grandfather provision is not implausible. Indeed, in the CUMAA's findings, Congress, after concluding that credit unions fulfill a valuable public purpose, observed that "current members and membership groups should not face divestiture from the financial services institution of their choice as a result of recent court action." CUMAA, Pub.L. No. 105-219, § 2(2), 112 Stat. 913, 913 (1998) (emphasis added). Congress's concern for "membership groups," as opposed to current, individual group members, suggests that it intended the grandfather clause to reach present as well as future persons who come to share the group characteristic.

Nonetheless, whether the grandfather clause is unambiguous, as the ABA contends, this Circuit has consistently observed that "even where the language of a statute is `superficially clear, legislative history may call such apparent clarity into question.'" American Scholastic TV Programming Found. v. FCC, 46 F.3d 1173, 1180 (D.C. Cir. 1995) (quoting Tataranowicz v. Sullivan, 959 F.2d 268, 277-78 (D.C. Cir. 1992)). Thus while it may be that the "immediate statutory text is the best evidence of congressional intent, it is not the only such evidence." Natural Resources Defense Council, Inc. v. Browner, 57 F.3d 1122, 1127 (D.C. Cir. 1995) (internal quotations omitted); see also McCarthy v. Bronson, 500 U.S. 136, 139, 111 S.Ct. 1737, 114 L.Ed.2d 194 (1991); Crandon v. United States, 494 U.S. 152, 158, 110 S.Ct. 997, 108 L.Ed.2d 132 (1990); Tataranowicz, 959 F.2d at 276.¹² Indeed, the Court "may consider a provision's legislative history in the first step of Chevron analysis to determine whether Congress' intent is clear from the plain language of the statute." City of Cleveland v. NRC, 68 F.3d 1361, 1366 n. 4 (D.C. Cir. 1995); accord Halverson v. Slater, 129 F.3d 180, 187 n. 10 (D.C. Cir. 1997). Cognizant that "[r]eference to statutory design and pertinent legislative history may often shed new light on congressional intent, notwithstanding statutory language that appears superficially clear," NRDC, 57 F.3d at 1127 (internal quotation marks omitted), the Court proceeds to examine the relevant House and Senate Reports to determine whether "there is evidence in the legislative history that Congress intended something else." Building Const. Trades Dep't v. United States Dep't of Labor Wage Appeals Bd., 932 F.2d 985, 990 (D.C. Cir. 1991).

12 The ABA's solipsistic interpretation of the grandfather clause calls to mind Learned Hand's sardonic observation that "[t]here is no surer way to misread any document than to read it literally." Guiseppi v. Walling, 144 F.2d 608, 624 (2d Cir. 1944) (Hand, J., concurring), aff'd sub nom. Gemsco, Inc. v. Walling, 324 U.S. 244, 65 S.Ct. 605, 89 L.Ed. 921 (1945). With reasoning as relevant to this case as it was in 1944, Hand wrote: "As nearly as we can, we must put ourselves in the place of those who uttered the words . . . and, although their words are by far the most decisive evidence of what they would have done, they are by no means final." Id.

If, as this Circuit has suggested before, only a "show stopper" in the legislative history will suffice before a Court may ignore what otherwise appears to be clear statutory text, First National, 90 F.3d at 530, this case is it. Described as a "broad grandfather" in the House Report, § 1759(c)(1)(A) is intended to

cover all persons or organizations or successors who were members of a federal credit union on the date of enactment of this Act, as well as anyone who is or becomes a member of a group representing a portion of the credit union's membership.

H.R. REP. No. 105-472, at 19 (emphasis added). Lest this be an unintended mistake, the House Report prominently emphasizes in its "Purpose and Summary" section that Title I of the CUMAA "also grandfathers all current members as well as current groups contained within the membership of a credit union as of the date of enactment of this legislation. The grandfather will permit such groups to continue accepting new members." Id. at 11 (emphasis added). Nor does this reflect merely a unicameral understanding of the grandfather clause. The Senate Report similarly provides:

This section also grandfathers all persons or organizations who are members of Federal credit unions on the date of enactment of the Act. Furthermore, any individual member of a group that is part of a credit union shall continue to be eligible to become a member of that credit union and any new member of such group is also eligible.

S.REP. No. 105-193, at 7 (emphasis added). Like the House, the Senate also reinforced its understanding of the grandfather clause in the Report's summary. See id. at 3-4.

Whatever merit there may be in the ABA's construction of the grandfather clause, pellucid legislative history, coupled with statutory design and effect, insulate IRPS 99-1 from attack. Unlike the majority of cases in which the legislative history offers equivocal or contradictory evidence of congressional intent, see, e.g., Halverson, 129 F.3d at 187-88; NRDC, 57 F.3d at 1127-28; Building Const. Trades, 932 F.2d at 990-91, both the House and Senate Reports in this case evince an intent demonstrably at odds with the literal interpretation of the CUMAA that the ABA advocates. Moreover, a literal interpretation would yield results that undermine the very purpose of the grandfather clause and the CUMAA. For although Congress enacted the grandfather clause to protect "all persons and organization who could be forced out of credit unions as a result of the February 25, 1998 Supreme Court decision," S.REP. No. 105-193, at 3, a literal construction of this provision would gradually emaciate grandfathered credit unions; as employees retired or otherwise left their present jobs, a credit union's membership base would slowly erode. Given that Congress elevated safety and soundness concerns to a level of primacy throughout the CUMAA, it strains credulity to believe that it would have enacted a grandfather clause that jeopardizes the long-term stability of credit unions that it specifically intended to protect.

Therefore, after examining the grandfather provision and its legislative history, the Court concludes that § 1759(c)(1)(A)(ii) is, at the least, ambiguous with respect to whether it covers persons who join credit-union groups after August 7, 1998. While it is the infrequent case in which legislative history will trump a putatively unambiguous statute, those cases do exist. See, e.g., Associated Gen'l Contractors v. California State Council of Carpenters, 459 U.S. 519, 529-31, 103 S.Ct. 897, 74 L.Ed.2d 723 (1983) (rejecting literal reading of § 4 of the Clayton Act, 15 U.S.C. § 15, based on legislative history); American Scholastic TV Programming Found., 46 F.3d at 1180-81 (holding that § 553(b)(1) of the Cable Act did not reach noncable video programming even though there was "nothing to indicate that it is anything but a blanket prohibition on all video programming") (emphasis in original). Having found that the grandfather clause is ambiguous, and thus must be reviewed under Chevron step two, the Court easily concludes that IRPS 99-1 is a reasonable interpretation. By extending the benefits of the grandfather clause to both present and future group members, IRPS 99-1 reasonably implements congressional intent. Accordingly, the ABA is unlikely to succeed on the merits of this claim.

G. Whether the voluntary-merger rule in IRPS 99-1 violates the unambiguous intent of the CUMAA

Purporting to "[r]ecogniz[e] the importance of mergers to a stable healthy credit union system," the NCUA's final rule

permits the voluntary merger of healthy multiple common bond credit unions containing select employee groups of less than 3,000 primary potential members without regard to the statutory analysis that is required when non-affiliated groups of less than 3,000 members seek to join an existing credit union.

63 Fed.Reg. at 72003 (emphasis added).¹³ The ABA contends that the NCUA's voluntary-merger rule violates the CUMAA's unambiguous mandate to "encourage the formation of separately chartered credit unions . . . whenever practicable and consistent with reasonable standards for the safe and sound operation of the credit union." 12 U.S.C. § 1759(f)(1)(A). In an effort to escape the plain meaning of the CUMAA and seek refuge in the deferential contours of Chevron step two, the NCUA maintains that because Congress did not expressly prohibit voluntary mergers, an ambiguity exists in the CUMAA that, in turn, invites judicial deference to the NCUA's putatively reasonable interpretation. Because IRPS 99-1's voluntary merger rule, as it pertains to groups under 3000 primary potential members, violates the unambiguous requirements of the CUMAA, the ABA has demonstrated a strong likelihood of success on the merits.

13 The final rule requires that groups exceeding 3000 primary potential members satisfy one of the exceptions in § 1759(d)(2) in order to merge into an existing credit union. Apparently satisfied that the NCUA will scrutinize mergers of groups over 3000 members according to the CUMAA's standards, the ABA limits its challenge to IRPS 99-1's voluntary merger rule to groups with fewer than 3000 primary potential members. See ABA Reply at 9-10 n. 9.

"As always, our inquiry starts from `the fundamental canon that statutory interpretation begins with the language of the statute itself.'" Butler v. West, 164 F.3d 634, 639 (D.C. Cir. 1999). The CUMAA adds a new subsection to 12 U.S.C. § 1759 that is captioned "Criteria for Approval of Expansion of Multiple Common-Bond Credit Unions." 12 U.S.C. § 1759(f). Measured by the plain language of the caption alone, mergers certainly would fall within the express terms of § 1759(f) insofar as they vastly expand the field of membership of a multiple common-bond credit union. More than just § 1759(f)'s pithy caption, however, underscores its application to mergers. It goes on to require that:

The Board shall . . . encourage the formation of separately chartered credit unions instead of approving an application to include an additional group within the field of membership of an existing credit union whenever practicable and consistent with reasonable standards for the safe and sound operation of the credit union.

§ 1759(f)(1)(A) (emphasis added). "Shall" is "language of an unmistakably mandatory character," Hewitt v. Helms, 459 U.S. 460, 471, 103 S.Ct. 864, 74 L.Ed.2d 675 (1983), and, when employed by Congress, indicates mandatory action and an absence of discretion. See Her Majesty the Queen v. EPA, 912 F.2d 1525, 1533, (D.C. Cir. 1990); MCI Telecommunications Corp. v. FCC, 765 F.2d 1186, 1191 (D.C. Cir. 1985). The specific mandatory duty that Congress has assigned to the NCUA is to "encourage the formation of separately chartered credit unions instead of approving an application to include an additional group within the field of membership of an existing credit union, whenever practicable and consistent with reasonable standards for the safe and sound operation of the credit union." 12 U.S.C. § 1759(f)(1)(A) (emphasis added). By the NCUA's own account, the mechanics of a "merger involve combining . . . two separate credit unions with only one surviving as a continuing entity." Def.'s Opp'n at 15. During a voluntary merger, therefore, groups from one credit union will invariably be added to the field of membership of an existing credit union. According to the manifest terms of § 1759(f)(1)(A), the NCUA has a mandatory duty, whenever practicable and consistent with safety and soundness concerns, to "encourage the formation of separately chartered credit unions instead of approving an application to include an additional group within the field of membership of an existing credit union." 12 U.S.C. § 1759(f)(1)(A). For the NCUA to approve applications to merge groups consisting of fewer than 3000 primary potential members into existing credit unions without evaluating whether any of these groups might possess the ability to charter their own credit unions violates the unambiguous intent of Congress.

Nothing in the legislative history of the CUMAA mandates a different conclusion. For the most part, the relevant committee reports simply parrot the statute's text. Yet to the extent that they illuminate Congress's intent, they buttress rather than undermine the Court's conclusion. In discussing § 1759(f)(1)(A), the Senate Report observed: "This section provides for the NCUA to encourage the formation of separately chartered credit unions wherever possible, consistent with safety and soundness, instead of including an additional group within an existing credit union's field of membership." S.REP. No. 105-193, at 7 (1998) (emphasis added). Elaborating on the provision's meaning, if only slightly more, the House Report indicated:

It is the Committee's position that the NCUA should charter new credit unions wherever possible and such formation would be consistent with safety and soundness. . . . The NCUA shall encourage groups, regardless of size, to form their own credit unions where such formation would be consistent with safety and soundness and not pose a significant risk to the share insurance fund.

H.R. REP. No. 105-472, at 20 (1998) (emphasis added). Both committee reports slightly reformulate the language of § 1759(f)(1)(A) to emphasize that the NCUA must charter separate credit unions "wherever possible." When a merger is proposed, § 1759(f)(1)(A) does not accord the NCUA any discretion to rubber-stamp an application to merge a group with fewer than 3000 primary potential members into an existing credit union without first conducting an inquiry into the group's resources and abilities. If a group could effectively charter its own credit union, consistent with safety and soundness concerns, then Congress's mandate to charter separate credit unions "instead of approving an application to include an additional group within the field of membership of an existing credit union," § 1759(f)(1)(A), must be given effect "wherever possible." S.REP. No. 105-193, at 7; H.R. REP. No. 105-472, at 20.

As already discussed, when an unaffiliated group with fewer than 3000 primary potential members, i.e., a group that is not already associated with an existing credit union, applies to join an existing multiple common-bond credit union, the NCUA has properly interpreted § 1759(f)(1)(A) to require the agency to examine whether the group, consistent with safety and soundness concerns, would be capable of independently chartering its own credit union. See 63 Fed.Reg. at 72001-02. While there is a superficial distinction between a previously unaffiliated group joining a credit union and a group that already participates in a credit union, it is a distinction without a substantive difference under the CUMAA. In both cases, there is an application to add a group with fewer than 3000 primary potential members to the field of membership of an existing credit union. Given Congress's clear intent that the NCUA encourage the formation of separate credit unions " wherever possible," H.R. REP. No. 105-472, at 20; S.REP. No. 105-193, at 7, the agency may not turn a blind eye to a group's ability to charter its own credit union when considering whether to add that group to a multiple common-bond credit union pursuant to a voluntary merger.

Despite the seemingly unambiguous and unqualified command of § 1759(f)(1)(A), the NCUA suggests that when a group with fewer than 3000 primary potential members is added to an existing credit union pursuant to a voluntary merger, the agency need not inquire into whether the group would be capable of chartering its own credit union. How the NCUA arrives at this conclusion is at times unclear, and at other times, incorrect as a matter of law. Banished to an after-thought footnote, all that the NCUA says about the language of § 1759(f)(1)(A) is that it "does not apply to mergers." NCUA Opp'n at 32 n. 18. Why this is so, the NCUA never says.

By brushing aside the literal language of § 1759(f)(1)(A) without comment, the NCUA strives to manufacture an "ambiguity" in the CUMAA that would compel this Court to defer to its reasonable interpretation under Chevron step two. In the NCUA's own words, "most significantly, however, Congress did not bar mergers in the Act; surely if it were intent on forbidding such mergers it would have said so." Id. at 30. Thus, because Congress did not affirmatively proscribe the NCUA from merging groups with fewer than 3000 primary potential members into existing credit unions, the agency maintains that the CUMAA is "ambiguous," and that IRPS 99-1's voluntary-merger rule commands deference.

By no means is the NCUA the first agency to propose that a statute boasts an ambiguity simply because Congress has not expressly negated the power that the agency claims. Time and again, this Circuit has confronted similar entreaties from other agencies, and each time, the court has reiterated that

[t]o suggest, that Chevron step two is implicated any time a statute does not expressly negate the existence of a claimed administrative power ( i.e., when the statute is not written in "thou shalt not" terms), is both flatly unfaithful to the principles of administrative law and refuted by precedent. Were courts to presume a delegation of power absent an express withholding of such power, agencies would enjoy virtually limitless hegemony, a result plainly out of keeping with Chevron and quite likely with the Constitution as well.

Halverson v. Slater, 129 F.3d 180, 187 (D.C. Cir. 1997) (quoting Railway Labor Executives' Ass'n v. National Mediation Bd., 29 F.3d 655, 671 (D.C. Cir. 1994)) (emphasis in original) (ellipses omitted); accord Backcountry Against Dumps v. EPA, 100 F.3d 147, 150-51 (D.C. Cir. 1996); Ethyl Corp. v. EPA, 51 F.3d 1053, 1060 (D.C. Cir. 1995); Oil, Chem. and Atomic Workers Int'l Union, AFL — CIO v. NLRB, 46 F.3d 82, 90 (D.C. Cir. 1995). Reviewing Chevron's first principles, it is clear that "deference is warranted only when Congress has left a gap for the agency to fill pursuant to an express or implied `delegation of authority to the agency.'" Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843-44, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984) (emphasis added); see also National Fuel Gas Supply Corp. v. FERC, 811 F.2d 1563, 1569 (D.C. Cir. 1987) ("When Congress leaves gaps . . ., either explicitly by authorizing the agency to adopt implementing regulations, or implicitly by enacting an ambiguously worded provision that the agency must interpret, it has explicitly or implicitly delegated to the agency the power to fill those gaps."). Here, as the NCUA concedes, Congress never explicitly authorized the agency to promulgate regulations to govern voluntary-mergers. Neither, however, did Congress leave an implicit interstice for the NCUA to fill; there is no "ambiguously worded provision" in the CUMAA that might reasonably be construed to permit the NCUA to disregard the clear, unambiguous language of § 1759(f)(1)(A). Therefore, this Court "refuse[s], once again, to presume a delegation of power merely because Congress has not expressly withheld such power." Ethyl, 51 F.3d at 1060.

Nor, of course, does the NCUA's general authority to prescribe rules and regulations governing mergers permit the agency to promulgate a voluntary-merger rule that otherwise violates a specific statutory prohibition. See 12 U.S.C. § 1766(a) ("The Board may prescribe rules and regulations for the administration of this chapter (including, but not by way of limitation, the merger, consolidation, and dissolution of corporations organized under this chapter)."). Where Congress has sought to liberate the NCUA's merger authority from generally applicable statutory restrictions, it has expressly amended the FCUA. In 1982, in order to imbue the NCUA with greater flexibility in structuring emergency mergers of financially unsound credit unions, Congress enacted the Garn-St Germain Depository Institutions Act of 1982, Pub.L. No. 97-320, § 131, 96 Stat. 1469, 1486 (codified at 12 U.S.C. § 1785(h)).¹⁴ Recognizing that the NCUA's merger authority did not empower it to flaunt general prohibitions in the FCUA, Congress expressly

14 Section 1785(h) provides, in pertinent part: "Notwithstanding any other provision of law, the Board may authorize a merger or consolidation of an insured credit union which is insolvent or is in danger of insolvency with any other insured credit union. . . ." 12 U.S.C. § 1785(h).

authorize[d] the NCUA Board to approve a merger or consolidation of an insured credit union with another insured credit union without regard to the traditional common bond requirements of Section 109 of the Federal Credit Union Act and without regard to requirements of state law, provided the Board determines that no other reasonable alternatives are available and the public interest would best be served if approval is granted.

S.REP. No. 97-536, at 50-51 (1982), reprinted in 1982 U.S.C.C.A.N. 3054, 3104-05. No similar provision in the CUMAA delegates to the NCUA the discretion to ignore the generally applicable provisions of § 1759(f)(1)(A). For the foregoing reasons, the ABA has demonstrated a substantial likelihood of success in establishing that the voluntary-merger rule in IRPS 99-1, as applied to groups with fewer than 3000 primary potential members, is contrary to law under Chevron step one.

H. Whether the NCUA validly promulgated IRPS 99-1

The NCUA published IRPS 99-1 in the Federal Register on December 30, 1998, just forty-eight hours before its New Year's Day effective date. See 63 Fed. Reg. at 71998. The ABA contends that this brief lacuna between the final rule's publication and effectiveness violates the Administrative Procedure Act's notice requirement. Section 10(e) of the APA provides that the

required publication or service of a substantive rule shall be made not less than 30 days before its effective date, except —

(1) a substantive rule which grants or recognizes an exemption or relieves a restriction;

(2) interpretative rules and statements or policy; or

(3) as otherwise provided by the agency for good cause found and published with the rule.

5 U.S.C. § 553(d)(1)-(3). Among the three arguments proffered by the NCUA in defense of its actions is an attack on the ABA's standing to contest this alleged procedural violation.¹⁵ While the arguments that the NCUA advances with respect to standing are well taken, they are better conceptualized under the APA's "prejudicial error" rubric codified at § 706 rather than the law of standing. When refracted through this analytic prism, the ABA's procedural attack on IRPS 99-1 appears highly unlikely to succeed on the merits.

15 In addition to the standing argument, the NCUA claims that IRPS 99-1 is a substantive rule that relieves a restriction pursuant to 5 U.S.C. § 553(d)(1) and that the agency had "good cause" under § 553(d)(3) to exempt IRPS 99-1 from the APA's notice requirement.

In reviewing agency action, the APA provides that "due account shall be taken of the rule of prejudicial error." 5 U.S.C. § 706. As this Circuit has recently observed, "[t]his sentence `sums up in succinct fashion the "harmless error" rule applied by courts in the review of lower court decisions as well as of administrative bodies.'" Doolin Sec. Savings Bank, F.S.B. v. Office of Thrift Supervision, 139 F.3d 203, 212 (D.C. Cir. 1998) (quoting UNITED STATES DEP'T OF JUSTICE, ATTORNEY GENERAL'S MANUAL ON THE ADMINISTRATIVE PROCEDURE ACT 110 (1947)). Similar to the manner in which the harmless error rule governs other instances of appellate review, § 706 "tempers judicial consideration of challenges to `preliminary, procedural, or intermediate agency action.'" Id. (quoting 5 U.S.C. § 704).

The legislative history of the APA reveals that the purpose for deferring the effectiveness of a final rule under § 553(d) was to "afford persons affected a reasonable time to prepare for the effective date of a rule or rules or to take other action which the issuance may prompt." S.REP. No. 79-752, at 15 (1946). As this Circuit has explained, "the purpose of the thirty-day waiting period is to give affected parties a reasonable time to adjust their behavior before the final rule takes effect." Omnipoint Corp. v. FCC, 78 F.3d 620, 630 (D.C. Cir. 1996) (emphasis added); see also American Federation of Gov't Employees v. Block, 655 F.2d 1153, 1156 (D.C. Cir. 1981). Although the ABA accurately cites the Supreme Court's First National decision to establish that its members have standing to contest NCUA rules, see First National, 118 S.Ct. at 936-38, it never articulates how any of its members have been prejudiced by the expedited effectiveness of the final rule. Nor would it appear that the ABA could identify any prejudice that it has suffered as a result of IRPS 99-1 taking effect on January 1, 1999 as opposed to January 29, 1999. The NCUA's final rule imposed no new regulatory burdens or requirements to which ABA members had to adjust. In analogous contexts, where an agency has improperly shortened the APA's notice-and-comment period, this Circuit has found harmless error where a petitioner did "not explain what it would have said had it been given" an opportunity to respond. See Air Transp. Ass'n of Am. v. Civil Aeronautics Bd., 732 F.2d 219, 224 n. 11 (D.C. Cir. 1984). The only prejudice that the ABA could conceivably identify would be simply having to compete with multiple common-bond credit unions twenty-eight days sooner. But such an injury is not only negligible, it is not of the kind that § 553(d) was principally adopted to protect against. See Omnipoint, 78 F.3d at 630. Accordingly, because the ABA cannot articulate any prejudicial error that resulted from IRPS 99-1's premature date of effectiveness, it has failed to demonstrate a substantial likelihood of success on the merits. See 5 U.S.C. § 706.

IV. OTHER FACTORS CONSIDERED IN DETERMINING WHETHER TO GRANT PRELIMINARY INJUNCTIVE RELIEF

With respect to seven of the eight bases on which the ABA seeks a preliminary injunction — the voluntary-merger rule being the lone exception — this Court's conclusion that the Plaintiffs are "not likely to succeed on the merits effectively decides the preliminary injunction issue." Serono Lab., Inc. v. Shalala, 158 F.3d 1313, 1326 (D.C. Cir. 1998). Although the preliminary-injunction calculus reflects a sliding-scale approach in which a strong argument in favor of one factor may excuse a relatively weaker showing on another factor, see CityFed Fin. v. Office of Thrift Supervision, 58 F.3d 738, 747 (D.C. Cir. 1995), absent a "substantial indication of probable success [on the merits], there would be no justification for the court's intrusion into the ordinary processes of administration and judicial review." Washington Metro. Area Transit Comm'n v. Holiday Tours, Inc., 559 F.2d 841, 843 (D.C. Cir. 1977); see also Demjanjuk v. Meese, 784 F.2d 1114, 1117-18 (D.C. Cir. 1986) (per Bork, J.) (denying equitable relief where, despite threat of irreparable harm, petitioner failed to demonstrate likelihood of success on the merits). Without any probability of prevailing on the merits, the Plaintiffs' purported injuries, no matter how compelling, do not justify preliminary injunctive relief.

Even if IRPS 99-1 were assumed to pose irreparable harm to ABA institutions, that injury must be weighed against the remaining factors: harm to others and the public interest. Certainly, lost business and revenue for which the law affords no legal remedy may well rise to the level of irreparable harm. Cf. Sampson v. Murray, 415 U.S. 61, 90, 94 S.Ct. 937, 39 L.Ed.2d 166 (1974).¹⁶ Yet to grant an injunction in order to protect the ABA from potential harm is to impose a reciprocal injury on credit unions and their current and prospective members. Congress passed the CUMAA, in part, "to ensure the continued safety and soundness of credit unions by permitting multiple common bond formations." H.R. REP. No. 105-472, at 10. Were the Court to enjoin the NCUA from implementing IRPS 99-1, credit unions confronted with safety and soundness problems would be unable to assuage their concerns by adding new groups to their field of membership.

16 Based on findings made by Judge Thomas Penfield Jackson in an Order that imposed a nationwide injunction against the NCUA's former chartering policies, the ABA maintains that the Defendants are collaterally estopped from denying that banks face irreparable harm when the NCUA charters a multiple common-bond credit union. Like the larger question in which it is subsumed, the issue of whether collateral estoppel operates in this proceeding need not be decided by the Court.

Moreover, the final preliminary-injunction factor, the public interest, militates against the ABA's prayer for relief because it is "inextricably linked with the merits of the case." Serono Lab, 158 F.3d at 1326. If, as the Court has concluded, the multiple common-bond chartering policies embedded in IRPS 99-1 are valid, an injunction would gravely frustrate the will of Congress. Galvanized by a determination that "it is appropriate to change existing law and specifically authorize multiple common bond credit unions," H.R. REP. No. 105-472, at 18; accord S.REP. No. 105-193, at 6, super-majorities in both houses of Congress voted to enact the CUMAA. See also Statement by William J. Clinton upon Signing H.R. 1151, 34 WEEKLY COMP. PRES. DOC. 1591 (Aug. 7, 1998) ("This bill will restore membership flexibility to credit unions, allowing, for example, employees of a number of smaller companies or members of a number of churches to join together to form a credit union."). An injunction would paralyze the NCUA from administering IRPS 99-1, a final rule that, at least based on the record before the Court, validly interprets and implements the goals of the CUMAA. For the foregoing reasons, the ABA's Application for a Preliminary Injunction is denied with respect to all issues except for the voluntary-merger rule.

Although the four preliminary-injunction factors must be weighed differently when considering the voluntary-merger rule, they nonetheless produce the same result. Unlike with the remainder of their claims, the ABA is likely to prevail in its challenge to the voluntary-merger rule. Probable success, however, is not itself sufficient to warrant preliminary equitable relief. Rather, "`[t]he basis of injunctive relief in the federal courts has always been irreparable harm.'" Sampson, 415 U.S. at 88, 94 S.Ct. 937 (quoting Beacon Theatres, Inc. v. Westover, 359 U.S. 500, 506, 79 S.Ct. 948, 3 L.Ed.2d 988 (1959)). As this Circuit held over forty years ago, "[t]he key word in this consideration is irreparable." Virginia Petroleum Jobbers Ass'n v. FPC, 259 F.2d 921, 923 (D.C. Cir. 1958) (emphasis in original). Whatever harm ABA institutions may suffer as a result of IRPS 99-1's other provisions, their claims of injury flowing from the voluntary-merger rule are substantially attenuated. In its briefs, the ABA argues that IRPS 99-1 will principally injure its members by permitting the NCUA to add various select employment groups to multiple common-bond credit unions: the more new groups that are added, the more customers potentially will abandon banks to pursue credit unions.

When looking exclusively at the voluntary-merger rule, however, this concern is substantially mitigated, if not wholly absent. A merger necessarily involves groups that already are incorporated in an existing credit union. Members of these groups, in turn, already enjoy the option of joining a credit union instead of patronizing a bank. Because a merger adds no groups that did not already exist in a credit union, the ABA's principal concern — that multiple common-bond credit unions will proliferate in size as new groups are added — is simply not present. In fact, nowhere in its briefs does the ABA address how the voluntary-merger rule threatens its members with irreparable injury. At oral argument, counsel for the ABA conceded that "it's a fair point . . . to say that it is not as compelling because they're not taking away customers from banks." Tr. of Hr'g, Mar. 1, 1999, at 21:11-:13. All that counsel could say about the voluntary-merger rule is that "the increased size, the increased marketing, the increased advertising capability of these large credit unions, make them more fierce competitors." Id. at 21:14-:16. That this rule permits credit unions eventually to achieve economies of scale is hardly the type of immediate and irreparable injury that mandates a preliminary injunction. Certainly, on this record, the ABA has failed to meet its burden of demonstrating how the voluntary-merger rule will exact an irreparable toll on its members' businesses.

On the flip side of this inquiry, the harm to others, it is also highly unlikely that credit unions would suffer greatly if the voluntary-merger rule were enjoined. After all, by its own terms, the rule extends only to financially sound credit unions. To foreclose a stable, financially sound credit union from merging with another stable, financially sound credit union would impose as insignificant a burden on those credit unions as the voluntary-merger rule poses to banks. With respect to the public interest, the ABA, quoting Judge Harold Greene of this Court, avers that "the public interest is best served by having federal agencies comply with the requirements of federal law, particularly the requirements of the APA." ABA App. for Prelim. Inj. at 36 (quoting Patriot Inc. v. Department of Hous. Urban Dev., 963 F. Supp. 1, 6 (D.D.C. 1997)). While this undoubtedly represents an important interest, the touchstone of this Court's equity powers remains a plaintiff's ability to demonstrate irreparable harm. See Sampson, 415 U.S. at 88, 94 S.Ct. 937; CityFed, 58 F.3d at 747. Absent a showing of irreparable harm, this Court, after careful consideration, concludes that equity does not militate in favor of granting the ABA's Application for a Preliminary Injunction against the voluntary-merger rule in IRPS 99-1. See CityFed, 58 F.3d at 747 ("Because CityFed has made no showing of irreparable injury here, that alone is sufficient for us to conclude that the district court did not abuse its discretion.").

V. CONCLUSION

For the foregoing reasons, the Court denies the ABA's Application for a Preliminary Injunction against IRPS 99-1. A separate Order accompanies this Memorandum Opinion.

ORDER

For the reasons expressed in the accompanying Memorandum Opinion, it is this 10 day of March 1999, hereby

ORDERED that the Application for Preliminary Injunction filed by the American Bankers Association shall be, and hereby is, DENIED; and it is

FURTHER ORDERED that the Application for Preliminary Injunction filed by Independent Bankers Association of America shall be, and hereby is, DENIED.

SO ORDERED.