Opinion for the Court filed by Chief Judge WALD.
Opinion dissenting in part filed by Circuit Judge RUTH BADER GINSBURG.
WALD, Chief Judge:The Federal Election Commission (FEC or the Commission) appeals from a decision of the district court, see Common Cause v. FEC, 655 F.Supp. 619 (D.D.C.1986), rejecting its interpretation of a provision of the Federal Election Campaign Act (FECA or the Act), which prohibits the use of a candidate’s name in the “name” of any unauthorized “political committee,” 2 U.S.C. § 432(e)(4). The FEC construes the bar as applying only to the name of the committee itself and not to the names of any fundrais-ing projects that the committee sponsors. The Commission also objects to the court’s order that it provide a statement of reasons explaining its 3-3 vote to dismiss a complaint contrary to the recommendation of the FEC General Counsel.
*438Concluding that the scope of § 432(e)(4) is by no means free from doubt, we find that the FEC’s interpretation is an entirely permissible one and reverse the district court’s ruling that the section applies to all projects sponsored by a political committee. Additionally, we hold that our recent ruling in Democratic Congressional Campaign Comm. v. FEC, 831 F.2d 1131 (D.C.Cir.1987), requiring the Commission to provide a statement of reasons whenever it departs from recommendations of the General Counsel in tie-vote dismissals, must be given prospective status only; we decline to apply it retroactively in this case. Accordingly, we vacate the remand order of the district court and instruct it to enter an order dismissing Common Cause’s petition for declaratory and injunctive relief.
I.Background
In July, 1980, the Carter-Mondale Reelection Committee (Carter-Mondale) filed a complaint against five independent committees as well as Ronald Reagan’s officially authorized campaign committee. The complaint alleged that the five committees, Americans for Change (AFC), Americans for an Effective Presidency (AEP), Fund for a Conservative Majority (FCM), North Carolina Congressional Club (NCCC) — now National Congressional Club — and National Conservative Political Action Club (NCPAC) impermissibly coordinated expenditures with Ronald Reagan’s authorized campaign committee and thereby exceeded statutory limits on campaign contributions. See 2 U.S.C. § 431(17) (defining “independent expenditure”); § 441a(a)(7) (B)(i) (providing that coordinated expenditures are treated as contributions); § 441a(a)(l) and (2) (setting out contribution limitations). Carter-Mondale also claimed that the committees had violated § 432 (e)(4)’s ban against the use of a candidate’s name in the name of an unauthorized committee by using the name “Reagan” in the titles of their fund-raising projects on public direct-mail solicitations.
The FEC General Counsel recommended that the Commission find “reason to believe” that a statutory violation had taken place and investigate the coordination charges against all five independent committees. The Commission initially agreed to the recommendation by a 4-2 vote. One Commissioner, Frank P. Reiche, however, took a different position with respect to two committees (NCCC and AEP), voting not to find “reason to believe” that they had engaged in such coordination. His unexplained shift caused the 3-3 deadlock among the Commissioners that resulted in a dismissal of the coordination claims against these two parties.1
The General Counsel suggested further investigation of the allegations against AFC, FCM, NCCC and NCPAC, with respect to the § 432(e)(4) claims. By a 4-2 vote the Commission found “reason to believe” a violation had taken place and ordered further inquiry.2
After an extensive investigation of the Carter-Mondale claims,3 the General Counsel recommended that the Commission take no additional action on the coordination claims, concluding that further inquiry would probably not lead to evidence of *439direct coordination by any of the three unauthorized committees still under investigation. J.A. at 286, 292, 298.
Regarding the § 432(e)(4) claims, the evidence showed that in several campaign communications the unauthorized committees included the name of candidate Reagan in letterheads and return addresses and, in some of the communications, asked for contributions with checks made payable to accounts bearing Reagan’s name. For example, AFC and NCCC used the designation “Americans for Reagan,” FCM, “Citizens for Reagan in 80,” and NCPAC, “Ronald Reagan Victory Fund.”
According to the committees, however, these names referred only to fundraising projects and not to the committees themselves. All of the mailings, with the possible exception of two from FCM, see J.A. at 222, 249, included the disclaimers required by 2 U.S.C. § 441d(a), indicating both the name of the political committee that had paid for the advertisement or solicitation and whether the communication itself had been authorized by any candidate.
Under these circumstances, the General Counsel recommended that the Commission find no probable cause to believe that a violation had occurred. Section 432(e)(4) prohibits the use of a candidate’s name in the name of a political committee, but not in other parts of the material that it publishes. The unauthorized committee had used the name “Reagan” only in reference to fundraising projects, not in the names of the independent political committees themselves. Hence, the General Counsel opined there was no violation of § 432(e)(4). On the General Counsel’s recommendation, a 4-2 Commission majority voted to take no further action on the § 432(e)(4) claims.
On August 1, 1983, Common Cause filed suit in federal district court, arguing that the Commission’s ultimate dismissal of the coordination and the § 432(e)(4) claims that it had investigated and its failure to give reasons for declining to investigate the coordination charges against AEP and NCCC were “contrary to law.” 2 U.S.C. § 437g(a)(8)(C).
The district court reversed the agency on the ground that the FEC’s reading of § 432(e)(4) defied “common sense”:
The political machinery is powered by names and what those names symbolize and identify. Therefore, whatever names the committees presented to the public for identification must also constitute a “name” within the meaning of section 432(e)(4).
Common Cause v. FEC, 655 F.Supp. at 621.
The court also ruled that the agency must explain why it dismissed by a 3-3 vote claims that the General Counsel recommended for further investigation. The court concluded that such a statement of reasons was necessary to enable it to review intelligently the Commission’s determination. Id. at 622-23.
The Commission appeals from both rulings.4
II. The Meaning of “Name” in § 432(e)(4)
Our first task in reviewing an agency’s interpretation of a statute committed to its enforcement is to examine the legislation itself, using traditional tools of statutory construction, to ascertain if its intent is clear. NLRB v. United Food and Commercial Workers Union, Local 23, — U.S. —, 108 S.Ct. 413, 98 L.Ed.2d 429 (1987); INS v. Cardoza-Fonseca, — U.S. —, 107 S.Ct. 1207, 94 L.Ed.2d 434 (1987); see also Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-43 at n. 9, 104 S.Ct. 2778, 2781 n. 9, 81 L.Ed.2d 694 (1984). Clear congressional intent derived from the plain language or legislative history of the statute dictates our result. However, where “the statute and its history are silent or ambiguous with respect to a specific issue,” the *440agency's construction, if reasonable, must ordinarily be honored. Chevron v. NRDC, 467 U.S. at 843, 104 S.Ct. at 2781. Neither the plain language of § 432(e)(4) nor its legislative history unambiguously resolves the dispute before us, though on balance we believe the Commission has the better of the argument.
A. The Language and Structure of the 1979 FECA Amendments
1. The Plain Language of § 432(e)(4)
Our “inquiry focuses first on the language and structure of the statute itself.” American Mining Congress v. EPA, 824 F.2d 1177, 1182 (D.C.Cir.1987). The bare text of § 432(e)(4) could conceivably accommodate either the construction adopted by the FEC or that proposed by Common Cause:
The name of each authorized committee shall include the name of the candidate who authorized such committee.... In the case of any political committee which is not an authorized committee, such political committee shall not include the name of any candidate in its name.
The Commission interprets “name” in § 432(e)(4) to refer only to the official or formal name under which a political committee must register.5 Following this interpretation, a political committee has only one “name,” even though it may rely on various “project” names to help collect money and achieve its other goals. This view obviously comports with the plain language of § 432(e)(4), which refers to the “name” of a political committee in the singular. It is also consistent with the avowed purpose of § 432(e)(4), to eliminate confusion; each committee has only one official name, which identifies it immediately either as an authorized or unauthorized committee,6 and which it must use in disclosing its sponsorship of all paid advertisements.
Thus, the General Counsel reasoned, the language of § 432(e)(4) and the Commission’s implementing regulation7 restrict the scope of the name ban to the officially registered name of a “political committee.”
[I]n order to find that 2 U.S.C. § 432(e)(4) was violated, it would be necessary to determine that “Reagan for President in ‘80’ ” met the definition of a “political committee” at the time this name was used.
J.A. at 284. No candidate’s name appeared in the names of any of the political committees in the case but only in the names of “special projects” of those political committees.8
Common Cause’s alternative construction of § 432(e)(4), however, could also be accommodated within the provision’s literal *441language. Common Cause argues that “name” in § 432(e)(4) does not mean only the officially registered “name” of a political committee but rather any title under which such a committee holds itself out to the public for solicitation or propagandizing purposes. Thus, according to Common Cause, a single political committee may have several “names.” The district court similarly concluded that “whatever names the Committees presented to the public for identification must also constitute a ‘name’ within the meaning of section 432(e)(4).” Common Cause v. FEC, 655 F.Supp. at 621.
This is not a totally implausible interpretation of the statute’s language, but neither is it so obviously the exclusive one as to preclude the Commission’s quite plausible alternative. There is, in short, a genuine ambiguity in § 432(e)(4)’s text.9 See Chevron v. NRDC, 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984).
2. Sections 432(e)(5) and 432(e)(4) Are Not In Pari Materia
Common Cause argues, however, that the traditional practice of construing statutory provisions in pari materia cures any ambiguity. It says that the FEC’s interpretation of “name” in § 432(e)(4) should be the same as its interpretation of the same word in § 432(e)(5), which provides that:
The name of any separate segregated fund established pursuant to section 441b(b) of this title shall include the name of its connected organization.
2 U.S.C. § 432(e)(5).10
Common Cause points out that the Commission has read § 432(e)(5) more broadly than § 432(e)(4). The Commission claims authority11 under § 432(e)(5) to regulate the content of the unofficial as well as official names of PACs; it not only requires PACs to use the full name of their sponsoring organizations in their official titles, but also to use abbreviations that clearly identify the fund’s sponsoring organization in unofficial names. Advisory Opinion 1980-23, 1 Fed.Election Camp.Fin. Guide (CCH) U 5476 (Apr. 14, 1980); Advisory Opinion 1980-86, 1 Fed.Election Camp. Fin.Guide (CCH) ¶ 5534 (Aug. 28, 1980). A PAC may “use a clearly recognized abbreviation or acronym by which the connected organization is commonly known.” 11 C.F. R. § 102.14(c) (1987).12 Common Cause argues that the FEC’s insistence under § 432(e)(5) that all of a PAC’s “names” include identification of its sponsoring organization is at odds with its more lenient stance under § 432(e)(4) that only a political committee’s official name be subject to the candidate-identifying requirement.
For statutory provisions to be construed together under the in pari materia doctrine, however, they must re*442late to the same subject or object. Only in the broadest sense are § 432(e)(5) and § 432(e)(4) analogous to each other; both sections do require political bodies to disclose the identity of persons associated with them. But beyond that generality, the assertion that these two sections must be similarly construed ignores the different purposes served by each section and the dissimilar campaign arrangements to which they apply.
Subsections (e)(4) and (e)(5) serve quite distinct purposes in campaign finance regulation. Subsection (e)(5), by requiring that all PAC names identify their sponsoring organization, is directed at disclosure to the public and to candidates of the true source of PAC contributions — i.e., the corporation, union or other group that is being permitted to avoid the Act’s ordinary prohibitions through the PAC vehicle. See 2 U.S.C. § 441b(a) (corporations and labor unions may not make contributions or expenditures in connection with federal elections). In contrast, subsection (e)(4) is directed solely at disclosure of whether a political committee that solicits funds from the public is part of the authorized campaign machinery of a candidate. (PACs, incidentally, are entirely prohibited from engaging in direct mail solicitations to the general public. 2 U.S.C. § 441b(b)(4)(A).13) Thus, subsection (e)(5) requires full disclosure to the public in all contexts of the sources of PAC support for candidates — i.e., the interest groups to which such a candidate might in the future be beholden. Subsection (e)(4), on the other hand, serves, in conjunction with § 441d(a), a different purpose — i.e., clarifying for readers and potential contributors the candidate authorization status of the political committees who sponsor advertisements and fund solicitations.14
Section 432(e)(4) mainly supplements § 441d(a) by ensuring that once a contributor learns who is paying for the advertisements or who is to be the recipient of his funds, he simultaneously learns by a glance at the title whether that recipient is an authorized or unauthorized vehicle of the candidate. Thus, § 432(e)(4) avoids the kind of confusing disclaimer previously possible, “Paid for by Reagan for President. Not authorized by President Reagan,” and makes § 441d(a)’s disclaimers more effective.15 It is not unreasonable in the absence of more explicit congressional direction to assume, as the Commission has, that now clearer § 441d(a) disclaimers, disclosing the identity of the sponsor of any solicitation and its authorized or unauthorized status, suffice to prevent confusion.16
In essence, we believe the FEC makes a reasonable case that § 432(e)(4) can support a different interpretation than § 432(e)(5) and thus defeats Common Cause’s in pari materia argument that its interpretation of § 432(e)(4) is the only permissible one.
*4433. The Structure of the Statute
The Commission has its own in pari materia argument; the word “name” should be similarly defined in § 432(e)(4) and in § 433(b)(1). Section 433(b)(1) requires unauthorized committees to register one official name with the Commission.17 This requirement of registration by one official name ensures ease of access to the disclosure reports of the committee, since the registered name must be the same one disclosed in § 441d sponsorship disclaimers in advertisements and solicitations. In public direct-mail communications as well, the Commission requires all political committees who are paying for the mailings to identify themselves by their § 432(e)(4)-§ 433(b)(1) name. Specifically, § 441d requires that any such mailing “shall clearly state the name of the person who paid for the communication and state that the communication is not authorized by any candidate or candidate’s committee.” 2 U.S.C. § 441d(a)(3) (emphasis added). Thus, by giving “name” the same meaning in §§ 432(e)(4), 433(b)(1) and 441d(a), the Commission aims to establish a coherent means by which readers and potential contributors can find out the identity and status of those who are soliciting them. If a political committee could register or put in § 441d(a) disclaimers any of its project “names,” that regulatory structure would be shattered. The Commission’s argument, based on the interrelationship between different sections of the Act, makes sense to us and we would be hard-pressed to label it an unreasonable one.
4. The Commission’s Initial Construction of § 432(e)(4)
Both Common Cause and the district court rely, erroneously we think, on Advisory Opinion 1980-84, 1 Fed.Election Camp. Fin.Guide (CCH) 115533 (Aug. 28, 1980), for the proposition that the Commission previously “considered the name of a committee as that which is presented to the public; not simply that which is registered officially in the records out of the public view.” Appellee’s Brief at 18 (quoting Common Cause, 655 F.Supp. at 621).
In Advisory Opinion 1980-84, the Commission told a political committee that it could use stationery bearing a former name, not in compliance with § 432(e)(4), as long as it used its new § 432(e)(4) official name in its § 441d(a) disclaimer.18 The FEC explained that “[o]ther uses of the stationery that are not subject to 2 U.S.C. § 441d could be made without need to include the new official committee name.” Id. at II5534.
The Commission’s view in August, 1980, just after the passage of the Act was thus in conformity with its current view.19 Section 432(e)(4) governs the official names of political committees, but other designations may be used by sub-projects of that committee as long as the public is protected from deception by § 441d(a) disclaimers disclosing who pays for public advertisements or solicitations. We find nothing in Advisory Opinion 1980-84 to support Common Cause’s case.
B. The Statute’s Legislative History
The sparse legislative history on § 432(e)(4) shows nothing definitive to un*444dercut the Commission’s consistent20 interpretation of this provision as applying only to the official name of a political committee. The general purpose of the 1979 amendments to the FECA, which included § 432(e)(4), was to simplify reporting and administrative procedures, H.R.Rep. No. 422, 96th Cong., 1st Sess. 3 (1979), U.S. Code Cong. & Admin.News 1979, p. 2862. Unfortunately there is little or no indication of the precise problem Congress sought to alleviate by enacting § 432(e)(4).
The Commission argues that Congress was merely targeting ambiguous political committee names, which could mislead a reader as to whether the committees were authorized or not. Section 432(e)(4), according to the FEC, simply eliminates confusing names in registration statements and solicitation disclaimers. Under this view, § 432(e)(4) was designed as a modest contribution to further clarification.
Common Cause would give § 432(e)(4) a much broader scope. It argues that the provision was meant to put an end to other common types of confusion as well, e.g., project names that use the name of a candidate and so mislead people into contributing to an unauthorized committee in the mistaken belief that the funds donated will necesary be used for the named candidate’s benefit. Appellee’s Brief at 20-22.
There is, however, no indication, much less any “compelling indication,” in the history of the § 432(e)(4) amendment, that Congress was thinking about the use of project names in solicitations and other campaign communications by unauthorized committees. See Clark-Cowlitz Joint Operating Agency v. FERC, 826 F.2d 1074, 1089 (D.C.Cir.1987) (“what does appear beyond question is that resort to the legisla-five history yields no ‘compelling indications’ of the sort necessary to overturn an agency’s reading that is in harmony with the express language of the legislation”). It seems downright unlikely that Congress would have enacted so broad a reform affecting the projects of unauthorized committees without a single word of explanation or debate. Cf. Finegan v. Leu, 456 U.S. 431, 441 n. 12, 102 S.Ct. 1867, 1873 n. 12, 72 L.Ed.2d 239 (1982) (“virtually inconceivable” that Congress would act to prohibit previously unregulated practice “without any discussion in the legislative history of the Act”).
Congress' silence on the subject seems even more telling against Common Cause’s case when we recall that Congress had only a few years previously addressed at length the problem of potentially misleading campaign communications, culminating in the passage of § 441d(a), which requires all committees paying for advertisements, no matter in whose name, to identify themselves as the source of the solicitation. Nowhere in the legislative history of § 432(e)(4) is there any mention of § 441d(a) or of the problems it sought to alleviate.21
We think the likelihood remote that Congress set out stealthily to alter so dramatically the regulation of campaign advertisements and solicitations in the way Common Cause asserts. This is especially so, since § 441d(a) was itself carefully crafted to accommodate the competing interests of voter information and free political speech. We have recently noted, “A fine balance of interests was deliberately struck by Congress in the name and disclaimer requirements of FECA.” Galliano, No. 836 F.2d at 1370. Section 441d(a) was one of several *445measures designed to facilitate disclosure to the public of the source of political communications “while safeguarding the full enjoyment of the First Amendment rights of individuals and groups to make expenditures for political expression.” H.R.Rep. No. 917, 94th Cong., 2d Sess. 5 (1976). To afford § 432(e)(4) so broad a scope as would the interpretation of Common Cause and the district court, on the basis of a general statement of intent to reduce confusion among the electorate, unduly elevate the rhetoric of purpose above the specifics of text, ignoring in the process the enormous subtleties and complexities inherent in the “FECA’s first-amendment-sensitive regime.” Galliano, 836 F.2d at 1369. Compare Board of Governors v. Dimension Finance Corp., 474 U.S. 361, 373-74, 106 S.Ct. 681, 688-89, 88 L.Ed.2d 691 (1986). In contrast, the Supreme Court has favored narrow interpretations of FECA requirements that implicate first amendment political speech.22 See Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976) (construing FECA provisions narrowly in attempt to avoid impermissible burdens on free expression); see also FEC v. Massachusetts Citizens for Life, 479 U.S. 238, 107 S.Ct. 616, 623, 93 L.Ed.2d 539 (1986) (relying on Buckley to construe independent spending regulation narrowly to apply only to “express advocacy”).
Actually, much of the legislative history of § 432(e)(4) suggests that, rather than the sweeping change proposed by Common Cause, the addition of the requirement that the titles of unauthorized committees not include a candidate’s name was intended only as a comparatively minor clarification in existing law.23 The Commission’s legislative recommendations in its 1978 Annual Report, suggested several FECA “clarifications,” Federal Election Campaign Act Amendments, 1979: Hearing Before the Senate Comm, on Rules and Administration, 96th Cong., 1st Sess. 23 (1979) among them that:
Under the current law, the name of most principal campaign committees identifies the candidate supported. However, in some cases, it is difficult to determine which candidate a principal campaign committee supports. In such cases the committee name does not contain the candidate’s name as, for example, “Good Government Committee” or “Spirit of ’76.”
Id. The Commission concluded that “[i]n order to avoid confusion the Act should require the name of the principal campaign *446committee to include in its name the name of the candidate which designated the committee.” Id.
In the first working draft of the Senate Committee on Rules and Administration, the section entitled “Organization of Political Committees” included this predecessor to § 432(e)(4):24
The name of each principal campaign committee of a candidate shall include the name of such candidate.
Hearing at 65.25 During the hearing process the Committee made the following addition:
The name of each principal campaign committee of a candidate shall include the name of such candidate, but only the principal campaign committee or authorized committee of a candidate may use the candidate’s name in its title.”
Hearing at 110 (emphasis added).26 This counterpoint amendment was understandably advanced as a further step in the clarification process designed to allow anyone reading the name of a political committee to determine immediately whether or not it is an authorized committee by the presence or absence of a candidate’s name.27 It would thus have “regulate[d] the use of a candidate’s name in the name of the candidate’s campaign committee.” S.Rep. No. 319, 96th Cong. 1st Sess. 10 (1979).
Had the new language in the second discussion draft been adopted, however, it would clearly have had other repercussions. Among these, it would have effected the same result as Common Cause’s interpretation of the present § 432(e)(4) would do. “Only” a principal or authorized campaign committee could have used a candidate’s name; the use of that candidate’s name in any unauthorized committee’s title or any of its project titles would have been impermissible. The FEC, however, objected to this very language,28 *447Hearing at 138, and it was never enacted into law.
Although the House version of the bill, which Congress ultimately passed,29 also allows for an immediate identification of a candidate’s authorized committee — (“the name of the candidate must appear in the name of the authorized committee.” H.R. Rep. No. 422, 96th Cong., 1st Sess. 2-3 (1979), U.S.Code Cong. & Admin.News 1979, pp. 2861-2863)30 — it has a less sweeping limitation on the use of a candidate’s name by other entities:
In the case of any political committee which is not an authorized committee, such political committee shall not include the name of any candidate in its name.
2 U.S.C. § 432(e)(4) (emphasis added). Its prohibition on the use of candidate names applies only to unauthorized “political committees.” The Act defines “political committee” as:
[A]ny committee, club, association, or other group of persons which receives contributions aggregating in excess of $1,000 during a calendar year or which makes expenditures aggregating in excess of $1,000 during a calendar year....
2 U.S.C. § 431(4)(A). See also §§ 431(4)(B) & (C) (other groups considered political committees).
The FEC General Counsel explicitly found that the “special projects,” whose names included references to Ronald Reagan, were not “political committees” under this definition and, therefore, § 432(e)(4) did not apply. The special project names referred only to certain activities of the political committees and, the committees themselves, who sponsored the special projects, had all included their officially registered names in the communications at issue, often in their text and always in the § 441d(a) disclaimers. These findings were not overturned in the district court or challenged on appeal.
Thus, little in its legislative history explicitly touches on the precise scope of § 432(e)(4)’s candidate name provision,31 or throws doubt on the reasonableness of the FEC's narrower interpretation. Remembering Sherlock Holmes’ famous clue of the dog that did not bark, we are most impressed by the notion that a major statutory revision in existing campaign disclosure practice would likely have spurred some greater debate or controversy.
C. Chevron’s Second Prong
In sum, then, “[t]he best case to be made for [those challenging the agency], upon analysis, is that when one examines the statute and the specific part of the legislative history upon which they rely, it becomes unclear as to what Congress’ intent actually was.” Eagle-Picher Industries v. EPA, 759 F.2d 922, 930 (D.C.Cir.1985). *448Such an ambiguity directs us to Chevron’s second prong. Where “the statute is silent or ambiguous with respect to the specific issue, the question for the court is whether the agency’s answer is based on a permissible construction of the statute.” NLRB v. United Food and Commercial Workers Union, Local 23, 108 S.Ct. at 421 (quoting Chevron v. NRDC, 467 U.S. at 843, 104 S.Ct. at 2782).
Deference is particularly appropriate in the context of the FECA, which explicitly relies on the bipartisan Commission as its primary enforcer. See FEC v. Democratic Senatorial Campaign Comm., 454 U.S. 27, 37, 102 S.Ct. 38, 44, 70 L.Ed.2d 23 (1981). FECA’s judicial review provision, 2 U.S.C. § 437g(a)(8), is limited; “Congress wisely provided that the Commission’s dismissal of a complaint should be reversed only if ‘contrary to law.’ ” Id. Thus, in resolving questions involving the FEC’s construction of the Act, our task is “not to interpret the statute as [we think] best but rather the narrower inquiry into whether the Commission’s construction [is] ‘sufficiently reasonable’ to be accepted by a reviewing court.” Id. at 39, 102 S.Ct. at 46. Far from being able to conclude that the FEC’s interpretation of § 432(e)(4) is unreasonable, in light of its language and legislative history, we think it is the better interpretation.
It may be, as the district court commented, that by adopting Common Cause’s construction of § 432(e)(4), the FEC might prevent campaign literature employing candidates’ names from misleading some members of the public who, despite proper § 441d disclaimers, will not realize that the candidate’s name in the solicitation “project” does not necessarily mean he will get the money. But Congress neither required nor endorsed this interpretation of the statute, and so the FEC was left to interpret the section in a way that it thought would best fit into the Act’s total administrative scheme, even if this may leave some confusion unabated.
Common Cause admits that even under its broad interpretation of § 432(e)(4) many other potentially misleading uses of candidates’ names would still be allowed. For example, even if the NCCC special project, “Americans for Reagan,” had been called “Americans for a Brighter Day” it still could have discussed, promoted and endorsed Ronald Reagan as a Presidential candidate throughout its campaign communications. The actual content of an advertisement or solicitation surely has as great a potential to mislead as the name of the project. Yet, no one suggests that Congress has authorized the FEC to prohibit or even regulate the use of candidate names in the text of campaign communications.
There are as well countervailing considerations for the FEC to ponder. It must allow the maximum of first amendment freedom of expression in political campaigns commensurate with Congress’ regulatory aims. In this case, it has decided that literal adherence to the language of § 432(e)(4), along with the disclaimer requirement of § 441d(a), strikes the proper balance. We in turn find no authority to create additional restrictions on the use of candidate names where both Congress and the Commission have declined to do so.
III. Deadlock-Vote Dismissals
The second issue in this case falls within the ambit of our recent decision in Democratic Congressional Campaign Comm. v. FEC, 831 F.2d 1131 (D.C.Cir.1987). There we found that a Commission dismissal of a complaint resulting from a 3-3 deadlock vote was reviewable when the General Counsel had made a contrary recommendation based on FEC precedent. In Democratic Congressional Campaign Comm. (DCCC), we upheld the district court’s remand to the Commission for a statement of reasons explaining its deadlock dismissal. The Commission did not seek further review of that decision, and although it tries to distinguish the facts there from those here, in our view the DCCC precedent governs this case.
Nonetheless, we decline to apply the precedent retroactively to this case, which arose before our DCCC decision. See Great Northern Ry. Co. v. Sunburst Co., 287 U.S. 358, 364, 53 S.Ct. 145, 148, 77 *449L.Ed. 360 (1932) (Cardozo, J.) (decision of whether to apply new rule prospectively within discretion of court).
A. The Commission Fails to Distinguish Democratic Congressional Campaign Committee
In DCCC, we found that “[njothing in the text of FECA’s judicial review prescription precludes review of a dismissal due to a deadlock_” 831 F.2d at 1133. Here, as in DCCC, the Commission dismissed the complaint for lack of the requisite four votes in favor of pursuing the investigation. The dismissal, in both cases, was contrary to the General Counsel’s recommendation to proceed.
The Commission argues that here, unlike in DCCC, the General Counsel’s recommendation was not based on clear legal precedent. The General Counsel, however, cited both Commission regulations and advisory opinions in support of his recommendation to investigate the coordination charges against AEP and NCCC. J.A. at 120-22.
We decline the Commission’s invitation to distinguish for remand purposes between deadlock dismissals which run contrary to General Counsel recommendations based on clear legal precedent and recommendations based on the General Counsel’s less definitive assessment of what the law requires in light of the factual allegations in the case. Such fine-tuning would inevitably invite Talmudic discourses about whether General Counsel recommendations were based on precedent in point, not in point, a little in point or a lot in point, or on sound or unsound applications of the law in particular factual situations.
In DCCC, we found that important statutory policies were served by requiring an explanation for deadlock dismissals contrary to the recommendation of the General Counsel. Those policies are equally relevant here, even if there is no Commission precedent directly on point. A statement of reasons, in either situation, is necessary to allow meaningful judicial review of the Commission’s decision not to proceed. Section 437g(a)(8) of the Act requires the court to determine whether FEC decisions are “contrary to law.” Whether the disagreement between the General Counsel and the dissenting Commissioners arises from a dispute about what the governing legal principle is or about how an agreed upon principle should be applied to particular facts, there remains the possibility that similarly situated parties may not be treated evenhandedly. In either case, a Commission decision treating like parties unalike or one based on statutorily impermissible factors should be reversible under the “contrary to law” standard.
Requiring a statement of reasons by the declining-to-go-ahead Commissioners at the time when a deadlock vote results in an order of dismissal also contributes to reasoned decisionmaking by the agency; it ensures reflection and creates an opportunity for self-correction.
Finally, some explanation of the views of the decliners will enhance the predictability of Commission decisions for future litigants. For example, some statement from Commissioner Reiche, while not law,32 would have informed Common Cause of the evidence practically necessary to convince a majority of the Commission to proceed with an investigation of a committee.
Although the panel in DCCC limited its holding to the facts of that case, see 831 F.2d at 1132, we cannot find a principled distinction between the situation in DCCC and the case at bar,33 that would obviate the need for an explanation of the declination to proceed.
*450B. Democratic Central Campaign Committee Should Apply Prospectively Only
Nonetheless, we decline to give retroactive effect to our ruling in DCCC insofar as it would require a statement of reasons to be made in order to permit judicial review. To do so, in this case at least, would be an exercise in futility and a waste of the Commission’s resources.
In Chevron Oil Company v. Huson, 404 U.S. 97, 92 S.Ct. 349, 30 L.Ed.2d 296 (1971), the Supreme Court discussed three factors relevant to the decision whether a court-made rule should be given prospective or retroactive effect. Each of these criteria is met in the instant case. First, if the decision is to be applied prospectively only, it must establish a new principle of law. Id. at 106, 92 S.Ct. at 355. The requirement of a statement of reasons announced in DCCC was new in the sense that it was declared after Commission practice to the contrary that persisted for years prior to the institution of a court challenge.
As to the second factor, the Supreme Court has stated that a court “must ... weigh the merits and demerits [of retroactive application] in each case by looking to the prior history of the rule in question, its purpose and effect, and whether retrospective operation will further or retard its operation.” Id. at 106-07, 92 S.Ct. at 355 (citing Linkletter v. Walker, 381 U.S. 618, 85 S.Ct. 1731, 14 L.Ed.2d 601 (1965)). In this particular case, retrospective application of the rule requiring a statement of reasons in cases of deadlock votes contrary to the recommendation of the General Counsel clearly would not further the purposes of the rule. Today only two of the six Commissioners who participated in the challenged vote still remain at the FEC. Commissioner Reiche, who cast the deciding vote to terminate further inquiry into the allegations against AEP and NCCC, is among those no longer with the Commission. To attempt to reconstruct and review his reasoning at this late date when he is no longer a public official would be to engage in a strange exercise indeed. We are reasonably confident that retroactive application of the rule would produce similar oddities in other cases since FEC Commissioners serve staggered six-year terms and the Commission’s composition, accordingly, changes every two years. See 2 U.S.C. § 437c(a)(2) (setting out term of office for Commissioners).
Furthermore, there is little chance that similarly situated parties have been treated in a materially different fashion in this instance. While the investigations did not proceed against AEP and NCCC, the Commission also eventually dismissed the remaining investigations of NCPAC, AFC and FCM. No conciliation agreement was proposed to these parties and no judicial action taken against them. Therefore, although the various complaints went to different graves, they are all now equally dead.
Nor would requiring a statement of reasons now contribute in any way to reasoned decisionmaking in future cases. Whatever Commissioner Reiche's reasons for distinguishing AEP and NCCC from the other committees, their elucidation would not enhance the predictability of future outcomes before the Commission of which he is no longer a member.
The final relevant factor in retroactivity decisions involves weighing the inequities imposed against the benefits gained by retroactive application of the rule. In this case, the principal inequity would lie in interfering with the Commission’s use of its resources. The district court's remand order did not come until three and one-half years after the Commission had closed the file in this case. There is only a minuscule chance that reopening the proceeding would eventually result in a successful complaint, given the fates of those complaints upon which the Commission voted favorably. More likely, investigative resources would be squandered. See Bethlehem Steel Corporation v. EPA, 782 F.2d 645, 649 (7th Cir.1986) (“[E]ven if wrong there was, [it would] be a poor exercise of our equitable remedial discretion.... The [agency] has limited resources_”). In 1988, an election year, the Commission has far more pressing matters to address than *451to formulate an ex post facto statement of reasons explaining a 1980 dismissal.
While we see no clear way to distinguish a dismissal based on a deadlock vote arising in the circumstances of this case from the situation in DCCC, we decline to implement the statement of reasons requirement of that case retroactively. We will, however, enforce the DCCC rule with respect to all Commission orders of dismissal based on deadlock votes that are contrary to General Counsel recommendations issued subsequent to our decision in that case.34
IV. Conclusion
We reverse the decision of the district court rejecting the FEC’s interpretation of § 432(e)(4) as inapplicable to the use of candidate names in special project titles and vacate its order remanding for a statement of reasons for the FEC’s 1980 tie-vote dismissal of the coordination charges against AEP and NCCC. The district court is, therefore, instructed to dismiss the petition for declaratory and injunctive relief.
. The FECA requires an affirmative vote of four Commissioners to undertake any agency action, 2 U.S.C. 437c(c), including a "reason to believe" finding necessary to initiate an investigation for violation of the statute. 2 U.S.C. § 437g(a)(2).
. On September 29, 1980, while investigation of the undismissed Carter-Mondale claims began, Common Cause filed another administrative complaint with the Commission. This second complaint made substantially the same allegations against the same five independent committees; these claims, like those in the complaint already under investigation, were based on charges of impermissible coordination with the official Reagan campaign. Unlike Carter-Mondale, Common Cause did not specifically make a claim under § 432(e)(4), but noted an apparent violation of this provision in a footnote. Joint Appendix (J.A.) at 158 n. 36. Because Common Cause did not make any new claims or point to any materially different evidence against the political committees, the General Counsel recommended that its complaint be merged with the Carter-Mondale complaint. The Commission agreed to the merger.
.The Commission did not investigate Common Cause's claims against NCCC and AEP because similar charges already had been dropped in the Carter-Mondale case.
. Common Cause did not cross-appeal to challenge the district court’s third ruling — that the investigations of AFC, FCM and NCPAC were properly terminated when it became clear that they were unlikely to reveal evidence of direct coordination. Id. at 623-24. Therefore, that determination is not before us.
. Section 433 governs the registration of all political committees. 2 U.S.C. § 433. It requires that the statement of organization include, inter alia, "the name” of the political committee. See id. at § 433(b)(1). The Commission requires this officially registered name to be used in § 441d(a) identifications:
Whenever any person makes an expenditure for the purpose of financing communications expressly advocating the election or defeat of a clearly identified candidate, or solicits any contribution through any broadcasting station, newspaper, magazine, outdoor advertising facility, direct mailing, or any other type of general public political advertising, such communication—
(3) if not authorized by a candidate, an authorized political committee of a candidate, or its agents, shall clearly state the name of the person who paid for the communication and state that the communication is not authorized by any candidate or candidate’s committee.
2 U.S.C. § 441d(a).
. If a single political committee were allowed to use and even register under multiple names, it would undermine the overriding intent of the 1979 FECA Amendments Act, to simplify the Act’s reporting requirements.
. The regulation states, in relevant part:
The name of each authorized committee shall include the name of the candidate who authorized such committee.... [A]ny political committee which is not an authorized committee shall not include the name of any candidate in its name.
11 C.F.R. § 102.14(a) (1987).
. The money raised by the fundraising projects was deposited in the accounts of the sponsoring political committees and all of the projects’ financial activities were reported in disclosures filed by the committees under their official names. Appellant’s Brief at 23.
. Common Cause’s citation to several state statutes which prohibit the misleading use of multiple names in other situations does not settle the issue here. Many of these statutes contain specific language that expressly covers "name or names.” Presumably, Congress could have used similarly explicit terminology in § 432(e)(4) if it wished to so limit the range of permissible FEC constructions of that provision.
. Separate segregated funds are commonly known as PACs (political action committees). PACs are associated with "connected organizations," a term defined as "any organization which is not a political committee but which directly or indirectly establishes, administers or financially supports a political committee.” 2 U.S.C. § 431(7). For example, when major business concerns and labor unions sponsor and financially support PACs to promote their interests, the business concerns and unions are “connected organizations" under the Act.
. The Commission relies for this authority on the legislative history of the Act. The House Report, in particular, states that § 432(e)(5) "does not prohibit the use of abbreviations or acronyms as long as the official name of the committee is used in all disclosure state-ments_” H.R.Rep. No. 422, 96th Cong., 1st Sess. 13 (1979), U.S.Code Cong. & Admin.News 1979, pp. 2860, 2873.
. While requiring that the full name of a separate segregated fund's sponsoring corporation must appear in the “name” used by the fund on its registration statement, its disclosure report and its public notices, the Commission only requires a "clearly recognized abbreviation or acronym” for the connected organization in other contexts. Thus, even under § 432(e)(5), the Commission has distinguished between official names and other names which may be used more informally.
. Section 441d(a) disclaimers, using names that meet both the requirements of § 432(e)(4) and § 432(e)(5) are, however, required on other campaign materials in which PACs advocate particular candidates.
. Subsection (e)(4) thus allows potential contributors to distinguish between donating directly to a candidate’s official campaign, supporting a candidate indirectly through an unauthorized campaign, or contributing in both ways within the amounts allowed by the statute.
. In this case, no one challenged the adequacy of the statutory disclaimers, but in cases involving a failure to provide adequate disclaimers, the Commission has granted relief under § 441d(a). See Galliano v. United States Postal Workers, 836 F.2d 1362, 1365 (D.C.Cir.1988) (probable cause to find violation of disclaimer provision led Commission to require conciliation agreement). It appears that in two instances the required disclaimers may have been absent from FCM communications. See J.A. at 222, 249. But since no challenge was brought under § 441d(a), we intimate no view regarding the adequacy of the disclaimers in these communications or others.
.Congress agreed that § 441d(a) should give the reader of campaign communications "adequate notice.” 125 Cong.Rec. 37,197 (1979) (Statement of Rep. Thompson). Mr. Thompson, however, also made clear that even the explicit disclaimer requirement of § 441d(a) should not be given an overly restrictive interpretation:
Although the legislation requires a disclaimer for political advertising and solicitation, it is not our intention to require that this disclaimer appear on the front face or page of such material.
Id.
. Section 433(b) sets forth the requirements for the statement of organization to be filed by all political committees, including PACs. See 2 U.S.C. § 433(a) and (b). The requisite information includes of course "the name, address, and type of committee.” Id. at § 433(b)(1).
. Before § 432(e)(4) was enacted in 1979, the Richard C. White Congressional Club of Permian Basin was known as the Congressional Club of Permian Basin. The political committee asked the Commission whether it could still use stationery bearing its old name and a checking account established under that name. The FEC concluded that use of the stationery and the checking account would be permissible as long as the committee used its "full official [§ 432(e)(4) ] name” in § 441d disclaimers. Advisory Opinion 1980-84, 1 Fed.Election Camp. Fin.Guide (CCH) ¶ 5533, 5534 (Aug. 28, 1980).
. Actually, Advisory Opinion 1980-84 answers a different and perhaps even a more difficult question than that raised by this case. The Commission there allowed the political committee to use more than one name for itself. Here, the Commission has only determined that special project names need not meet the requirements set out for the names of “political committee[s].” 2 U.S.C. § 432(e)(4).
. The Commission has maintained its current interpretation of § 432(e)(4) since 1980. Therefore, Common Cause’s citations to decisions in which we refuse to defer to agency positions which have been inconsistent or have been changed without adequate explanation are off-point.
Furthermore, it is without significance that the Commission adopted and reaffirmed this interpretation by a divided vote. Under the statute, a 4-2 vote is sufficient for a valid Commission decision.
. Section 441d(a) is a more powerful antidote to public confusion and has more comprehensive scope than Common Cause’s reading of § 432(e)(4). Section 432(e)(4) only prohibits the use of a candidate’s name in names, titles or designations. The text of an unauthorized committee's campaign communication can still use a candidate's name. Section 441d(a), on the other hand, attributes the entire text of a communication, including names, to an identified sponsor.
. Common Cause’s broad interpretation of § 432(e)(4) would be a mixed blessing. Political communications are enhanced by the ability to use project names like "Americans for Reagan.” Inescapably, some candidates’ names carry a powerful political message, which grabs the eye and commands the loyalty of millions of Americans. The committee’s ability to communicate effectively might be diminished by prohibiting the use of names/slogans like "Americans for Reagan.”
Similarly, a rigid adherence to the name ban would also prohibit a committee opposed to a candidate from fundraising under a project name like "Americans Against Reagan." Our dissenting colleague claims that because "Americans Against Reagan” is not a confusing name for an unauthorized committee, § 432(e)(4)’s flat prohibition on the use of the name of "any candidate” would not apply. See Dissenting Opinion at 3-4 n. 6. This indulgent interpretation would of course ignore the plain import of the name ban. The FEC’s interpretation allowing clearly identified political committees to use candidate names in project solicitations would of course permit slogan-labels like "Americans Against Reagan.”
Allowing candidates' names in fundraising project names may also make particular sense in the context of multi-candidate political committees. It allows the committee to tailor its message to promote candidates or raise funds more effectively when addressing various political or regional audiences.
Authorized committees that do not identify their candidate in their names are also in violation of § 432(e)(4). Common Cause’s interpretation, extended to its logical conclusion, would, therefore, lead to the additional requirement that an authorized committee always use the name of the authorizing candidate in the title of any fundraising project. This, again, is an inhibition on the committee’s political speech.
. According to Robert O. Tierman, the Chairman of the Commission, the amendments sought by the FEC had three primary goals; they were intended to simplify the disclosure process, to encourage grass roots political activity and finally to clarify the existing requirements of FECA. FECA Amendment, 1979: Hearing, 96th Cong., 1st Sess. 26 (1978).
. The language of this provision was drawn directly from an earlier Senate bill, which, although passed by the Senate, was not enacted into law. See S. 926, 95th Cong., 1st Sess. (1977).
. The Act defines the term "principal campaign committee" as "a political committee designated and authorized by a candidate — ” 2 U.S.C. § 431(5). The term "authorized committee” means "the principal campaign committee or any other political committee authorized by a candidate ... to receive contributions or make expenditures on behalf of the candidate.” 2 U.S.C. § 431(6). Section 432(e) elaborates on the relationship between these two types of committees:
(1) Each candidate for Federal office (other than the nominee for the office of Vice President) shall designate in writing a political committee in accordance with paragraph (3) to serve as the principal campaign committee of such candidate. Such designation shall be made no later than 15 days after becoming a candidate. A candidate may designate additional political committees in accordance with paragraph (3) to serve as authorized committees of such candidate. Such designation shall be in writing and filed with the principal campaign committee of such candidate.... (3)(A) No political committee which supports or has supported more than one candidate may be designated as an authorized committee, except that—
(i) the candidate for the office of President nominated by a political party may designate the national committee of such political party as a principal campaign committee, but only if that national committee maintains separate books of account with respect to its function as a principal campaign committee; and (ii) candidates may designate a political committee established solely for the purpose of joint fundraising by such candidates as an authorized committee.
2 U.S.C. § 432(e). The Act’s |1,000 limitation on contributions applies to total amounts given to a "candidate or his authorized political committees,” including, of course, his principal campaign committee. 2 U.S.C. § 441a(l)(A).
. This amendment parallels an earlier House bill, which was not passed, but whose precise language resurfaced later in the House. See H.R. 11315, 95th Cong., 2d Sess. § 302(e)(4) (1978); see also infra note 29.
. The House Report on the predecessor bill, H.R. 11315, offers the clearest explanation of the limited purpose of this provision: "It is the intent of the (House] [C]ommittee that the average contributor or voter be able to determine, by reading the committee's name, on whose behalf the committee is operating." H.R.Rep. No. 982, 95th Cong., 2d Sess. 12 (1978).
. Unfortunately, the record does not disclose the basis of the Commission's objection to the added clause in the Senate's draft bill. Of course, one possible reason is that it objected to the breadth of the limitation on the use of candidates’ names.
. The precise language of § 432(e)(4) in the House bill, H.R. 5010, as introduced by Congressman Thompson on July 30, 1979, survived without amendment.
. Note that, as in the statute itself, the House Report speaks of "the name of a political committee." Id. (emphasis added). Nowhere does it use "names” in the plural.
. The House Report itself, however, did caution the Commission to adopt a narrow reading of § 432(e)(4), at least in one respect:
While the purpose is to require the committee name to identify, clearly and by unambiguous reference, the authorizing candidate, common sense must prevail. In most cases, the use of the last name alone would be sufficient. In rare situations, such as a race with two candidates with the same last name, first names or initials would be necessary. On the other hand, committees which are not authorized may not include the name of the candidate in the name of the committee.
H.R.Rep. No. 422, 96th Cong., 1st Sess. 13 (1979), U.S.Code Cong. & Admin.News 1979, p. 2873.
Nothing in this passage, however, actually reveals whether Congress meant "name" to refer only to the official title of a committee rather than to any designation employed to identify a committee or any of its projects. While it demonstrates Congress’ general intent to prevent confusion arising from misleading committee names, it also indicates Congress’ desire that the FEC not be so overzealous in implementing the provisions as to deter participation in the political process. This view is consistent with another congressional concern in enacting the FECA Amendments of 1979, to encourage greater political participation, particularly at the local level. S.Rep. No. 319, 96th Cong., 1st Sess. 1 (1979).
. Of course, such a statement of reasons would not be binding legal precedent or authority for future cases. The statute clearly requires that for any official Commission decision there must be at least a 4-2 majority vote. To ignore this requirement would be to undermine the carefully balanced bipartisan structure which Congress has erected.
. We emphasize that DCCC sought simply to assure meaningful judicial review under the "contrary to law” standard. 831 F.2d at 1135 n. 5; 2 U.S.C. § 437g(a)(8). The statement of reasons need only be adequate to fulfill that function.
. In cases in which there is an explanation for the declining Commissioners’ action so that meaningful judicial review can be performed, the DCCC holding should apply regardless of the time at which suit was brought. See, e.g., Stark v. FEC, 683 F.Supp. 836 (D.D.C.1988). Here, the likely futility of further proceedings and the corresponding waste of FEC resources arise out of the dissenting Commissioners’ silence about their reasons for declining to go forward.