with whom Circuit Judges BRUNETTI and KOZINSKI concur, dissenting:
I respectfully dissent.
The principal issue in this case is whether a prehire agreement entered into under the authority of section 8(f) of the National Labor Relations Act (“NLRA”) can be repudiated by the employer until the union establishes majority status. The Supreme Court, in a unanimous opinion and in language that could not be clearer, has held that it can be. The Court stated:
*1138A § 8(f) prehire agreement is subject to repudiation until the union establishes majority status.
Jim McNeff, Inc. v. Todd, 461 U.S. 260, 271, 103 S.Ct. 1753, 1759, 75 L.Ed.2d 830 (1983).
Despite this clear statement in McNeff, the majority upholds a contrary view of the NLRB. It dismisses the language in McNeff on the ground that the Court there merely followed its prior decision, NLRB v. Iron Workers, 434 U.S. 335, 98 S.Ct. 651, 54 L.Ed.2d 586 (1978) (hereinafter Hig-don ). The majority feels, in turn, that the Court in Higdon did not actually construe section 8(f), but instead simply held that the Board’s construction of section 8(f) was within a range of possible reasonable constructions. Such a limited holding would leave room for other, equally reasonable, constructions of the Act. Thus the majority contends that a new construction by the Board, even one which is directly contrary to the pronouncement of the Supreme Court in McNeff, is permissible if it can be justified as being reasonable. See Opinion at 1129-1131.
I part company from the majority because I believe the Supreme Court did definitively construe section 8(f) in both Hig-don and McNeff, although the Court may have given deference to the Board’s interpretation in doing so. The Court did not, as the majority suggests, decide only that the Board’s interpretation was a reasonable construction of the Act subject to change at the agency’s whim. Instead — after giving heightened consideration to the Board’s arguments — the Court passed judgment upon the meaning of section 8(f) and such judicial interpretation is binding under our principle of stare decisis.
I.
This case presents a vital question concerning the relative roles of the judiciary and administrative agencies in construing statutes, the importance of which, in my view, transcends the particular point of labor law at issue. We are here faced with a question of pure statutory interpretation. We are concerned with determining Congressional intent. Specifically, our inquiry is how Congress intended a prehire agreement to operate. The Supreme Court in Higdon and McNeff interpreted the statute after giving some deference to the interpretation of the NLRB, the agency charged with administering the Act. The NLRB, at that time, gave persuasive arguments that Congress intended to allow either party to repudiate the contract until the Union achieved majority status.
The NLRB’s position, however, was by no means binding on the Court. Congress did not delegate to the NLRB the discretion to determine how a prehire agreement was to operate, allowing the Board to treat the agreement in any manner that the Board considered would best implement sound labor policy. Congress simply enacted section 8(f), which was to be interpreted by the courts. That section has been interpreted by the highest court authority — the United States Supreme Court — after careful consideration of the statutory language, legislative history, and the interpretation placed upon it by the agency charged with its enforcement. That interpretation must stand until it is overruled by the Supreme Court or until Congress amends the statute.
I therefore believe we are bound by Hig-don and McNeff to reject the Board’s new Deklewa rule. The majority apparently believes that the stare decisis value of Hig-don and McNeff is vitiated by virtue of the deference accorded in those cases to the Board’s interpretation of the Act. I submit that the majority misconceives the true nature of the deference granted in those cases. An examination of the role of deference in the two cases reveals that the Court did, in fact, render its own interpretation of section 8(f) and, in doing so, created precedent to which we must adhere.
Confusion in this area is quite understandable. A vast number of cases discuss the proper deference accorded to an agency’s interpretation of a statute, but, as one distinguished author notes, such discussions are often laced with “verbalisms [that] have rarely been helpful and have usually been harmful; [and] are often un*1139certain, conflicting and even confusing.” 5 K. Davis, Administrative Law Treatise, § 29.1, at 334 (2d ed. 1984). Behind the confusing verbiage lies, I believe, two very distinct principles of judicial deference to agencies. Moreover, the two breeds of deference have vastly different effects on the principle of stare decisis. It is evident to me that the majority has mistaken one type of deference for the other, and thereby misjudged the stare decisis effect of both Higdon and McNeff.
II.
The type of deference underlying the Supreme Court’s ruling in Higdon and McNeff is appropriate in cases when the Court confronts a pure question of statutory construction. When the Court seeks to ascertain the meaning of an ambiguous statutory provision, it gives special consideration to the viewpoint of the agency charged with implementing the statute. Such deference is due because the agency “constitute^] a body of experience and informed judgment to which courts ... may properly resort for guidance.” Federal Maritime Bd. v. Isbrandtsen Co., 356 U.S. 481, 499, 78 S.Ct. 851, 862, 2 L.Ed.2d 926 (1958). As Davis notes, the deference given by judges for interpretations of administrators “is not at all surprising”: it is simply a matter of “[djeference of generalists for the views of specialists [which] could be deemed a part of the law of nature....” K. Davis, supra, at 400.
However, such deference does not intrude on the Court’s role as the final authority on questions of statutory construction. As the Supreme Court stated in Securities Indus. Ass’n. v. Board of Governors, 468 U.S. 137, 143, 104 S.Ct. 2979, 2982, 82 L.Ed.2d 107 (1984), “Judicial deference to an agency’s interpretation of a statute ‘only sets “the framework for judicial analysis; it does not displace it.” ’ ” See also Barlow v. Collins, 397 U.S. 159, 166, 90 S.Ct. 832, 837, 25 L.Ed.2d 192 (1970) (“[When] the only or principal dispute relates to the meaning of the statutory term, the controversy must ultimately be resolved, not on the basis of matters within the special competence of the [agency], but by judicial application of canons of statutory construction.”); Batterton v. Francis, 432 U.S. 416, 424, 97 S.Ct. 2399, 2405, 53 L.Ed.2d 448 (1977) (“[A]dministrative interpretations of statutory terms are given important but not controlling significance.”).
Over the years, the Supreme Court has repeatedly affirmed its role as the final authority on issues of statutory construction. See, e.g., SEC v. Sloan, 436 U.S. 103, 118, 98 S.Ct. 1702, 1711, 56 L.Ed.2d 148 (1978) (“[T]he courts are the final authorities on issues of statutory construction [and] ‘are not obliged to stand aside and rubber-stamp their affirmance of administrative decisions....’” (Citations omitted.)); INS v. Cardoza-Fonseca, 480 U.S. 421, 107 S.Ct. 1207, 1221, 94 L.Ed.2d 434 (1987). (“The judiciary is the final authority on issues of statutory construction....”) (quoting Chevron U.S.A., Inc. v. Natural Resources Defense Council, 467 U.S. 837, 843 n. 9, 104 S.Ct. 2778, 2781 n. 9, 81 L.Ed.2d 694 (1984)); Federal Election Comm’n v. Democratic Senatorial Campaign Committee, 454 U.S. 27, 32, 102 S.Ct. 38, 42, 70 L.Ed.2d 23 (1981); Bureau of Alcohol, Tobacco and Firearms v. F.L.R.A., 464 U.S. 89, 97, 104 S.Ct. 439, 444, 78 L.Ed.2d 195 (1983).
It is equally well-settled that deference is not due to an agency’s construction of a statute where the statutory language is clear. United States v. Missouri Pacific R. Co., 278 U.S. 269, 280, 49 S.Ct. 133, 137, 73 L.Ed. 322 (1929); Swift Co. v. United States, 105 U.S. 691, 695, 26 L.Ed. 1108 (1881); United States v. Tanner, 147 U.S. 661, 663, 13 S.Ct. 436, 437, 37 L.Ed. 321 (1893). When the statutory language is ambiguous, the amount of weight given to the agency’s interpretation depends on a variety of factors, such as the thoroughness of the agency’s consideration, whether the agency’s construction has been consistent over the years, and whether the timing of the agency’s construction was contemporaneous with the passage of the statute. United States v. Sweet, 189 U.S. 471, 473, 23 S.Ct. 638, 638, 47 L.Ed. 907 (1902); Federal Maritime Bd. 356 U.S. at 499-500, 78 S.Ct. at 862-63; United States v. John*1140ston, 124 U.S. 236, 253, 8 S.Ct. 446, 455, 31 L.Ed. 389 (1888); United States v. American Trucking Ass’ns., Inc., 310 U.S. 534, 549, 60 S.Ct. 1059, 1067, 84 L.Ed. 1345 (1940); United States v. Leslie Salt Co., 350 U.S. 383, 396-97, 76 S.Ct. 416, 423-24, 100 L.Ed. 441 (1956). See also Cardoza-Fonseca, 107 S.Ct. at 1221 n. 30 (less deference accorded to agency when agency has changed its mind).
With these principles in mind, it is evident that a court faced with an issue of pure statutory interpretation must make its own independent judgment as to the meaning of the statute. Because it is the final authority on such matters, it cannot delegate its function to the agency. Though a court may give a certain amount of deference to the agency’s interpretation of the statute, its analysis must not end there. Rather, the court must either adopt the agency’s position as its own interpretation of the Act, or adopt a construction other than that promoted by the agency. In either case, the court makes its own final judgment as to the meaning of the statute. The court’s determination as to the meaning of the statute is no less its own simply because it accorded deference to the agency and accepted the agency’s construction.
Because courts cannot escape their function as final authorities on issues of statutory interpretation, their decisions on such matters are binding regardless of the degree of deference given to the agency in reaching the decision. As the Supreme Court noted in Estate of Sanford v. Commissioner, 308 U.S. 39, 60 S.Ct. 51, 84 L.Ed. 20 (1939), “[W]e should be ... free to reject [an administrative] practice when it conflicts with our own decisions. A change of practice ... will be accepted as controlling when consistent with our decisions.” Id. at 53, 60 S.Ct. at 60 (emphasis added). In a similar vein, Justice Stevens noted the binding effect of the Supreme Court’s decisions of statutory interpretation in Shearson American Express, Inc. v. McMahon, 482 U.S. 220, 107 S.Ct. 2332, 96 L.Ed.2d 185 (1987), where he stated:
Gaps in the law must, of course, be filled by judicial construction. But after a statute has been construed, either by this Court or by a consistent course of decision by other federal judges and agencies, it acquires a meaning that should be as clear as if the judicial gloss had been drafted by the Congress itself.
Id. 107 S.Ct. at 2359 (J. Stevens, concurring and dissenting). To summarize, in cases where the Supreme Court confronts an issue of pure statutory interpretation, the deference accorded to the agency’s construction of the statute does not impinge in the least upon the stare decisis effect of the decision. The decision is binding upon the agency and upon lower courts.1
III.
The other type of deference is markedly different and readily distinguishable from the type discussed above. This is the deference given to an administrative agency when Congress has delegated to the agency the authority to further define the specifics of a general proposition of law. Unlike the deference discussed above, when this type of deference underlies a court’s decision, stare decisis is affected. Exploring the roots of such deference demonstrates why.
In some instances, Congress purposefully does not resolve competing economic or social interests in particular areas and instead leaves this to the agency. In other situations, Congress enacts quite general provisions, with the specifics to be filled in by the agency. As noted in Chevron,
If Congress has explicitly left a gap for the agency to fill, there is an express *1141delegation of authority to the agency to elucidate a specific provision of the statute by regulation. Such legislative regulations are given controlling weight unless they are arbitrary, capricious, or manifestly contrary to the statute. Sometimes the legislative delegation to an agency on a particular question is implicit rather than explicit. In such a case, a court may not substitute its own construction of a statutory provision for a reasonable interpretation made by the administrator of an agency.
We have long recognized that considerable weight should be accorded to an executive department’s construction of a statutory scheme it is entrusted to administer. ...
Chevron, 467 U.S. at 843-44, 104 S.Ct. at 2782 (footnotes omitted). Thus there are two categories of delegation: (a) explicit delegation; and (b) implicit delegation.
The Chevron case itself provides excellent examples of both explicit and implicit delegation. The case concerned the Clean Air Act, the administration of which is entrusted to the Environmental Protection Agency (“EPA”). Congress directed the EPA to promulgate National Ambient Air Quality Standards, to publish a list of categories of pollution sources, and to establish new source performance standards for each. See Chevron, 467 U.S. at 846, 104 S.Ct. at 2783. This was an express delegation of authority by Congress to develop the specifics of general legislation. At issue in Chevron was the meaning of the statutory term “stationary source.” There had been no specific delegation of authority to define that term. However, the Court found that Congress intended that there not be a static judicial definition of the term but, instead, a flexible definition to be applied by the EPA in order to carry out the general policy of the Act. Thus the Court found an implicit delegation of authority to the EPA to determine the meaning of “stationary source.” See id. at 862-66, 104 S.Ct. at 2791-93.
When there exists an explicit delegation of authority to fill gaps in the legislation, this function is generally fulfilled by the adoption of regulations. When the agency enacts regulations or procedures designed to apply the statute to various factual situations its rule-making assumes a quasi-legislative character. See Batterton, 432 U.S. at 424 n. 8, 97 S.Ct. at 2405 n. 8 (“ ‘Administration, when it interprets a statute so as to make it apply to particular circumstances, acts as a delegate to the legislative power.’ ”) (Citations omitted.) See also NLRB v. Hearst Publications, Inc., 322 U.S. 111, 130-31, 64 S.Ct. 851, 860-61, 88 L.Ed. 1170 (1944); State of Montana v. Clark, 749 F.2d 740, 745 (D.C.Cir.1984), cert. denied, 474 U.S. 919, 106 S.Ct. 246, 88 L.Ed.2d 255 (1985). A court, in reviewing such regulations, is not actually confronted with an issue of statutory interpretation, for there is nothing in the statute per se to interpret: Congress left a gap, entrusting the agency to fill it. A reviewing court is not free to set aside those regulations simply because it would have refined the statute in a different manner. Batterton, 432 U.S. at 425, 97 S.Ct. at 2405. Thus, the judicial task when reviewing an agency’s regulations designed to fill a gap left by Congress differs fundamentally from the judicial task in determining the meaning of the statute. In the former case, the court does not confront an issue of statutory interpretation and, thus, its authority is limited to reviewing the regulations to ensure that Congress gave the agency the power to make such regulations, and to ensure that the regulations are consistent with the statutory scheme. In the latter case, the court is the final authority as to the statute’s meaning and may defer to the agency’s interpretation as it deems appropriate, but its holding reflects a judicial construction of the statute. In both types of cases, the principle of deference is employed by courts, but the principle operates very differently in each context.
When the delegation from Congress is implicit rather than explicit, the principle is the same. In Chevron, the Court determined that the term “stationary source” in the Clean Air Act had been intended by Congress as a flexible definition to be refined by the EPA, to which it had entrusted wide discretion in administering the legisla*1142tion. Thus, the Court's inquiry was only whether the EPA’s construction of that term was reasonable. As in the case of an express delegation, where Congress contemplates future regulations, the Court deferred to the EPA’s interpretation of the term as being a reasonable exercise of the delegated authority. Chevron, 467 U.S. at 865-66, 104 S.Ct. at 2793. This differs markedly from the deference given to an agency in resolving a pure question of statutory interpretation.
As an example of how the principle of deference operates in each context, it is useful to contrast Chevron with Cardoza-Fonseca. In Chevron, the Court determined that Congress left a gap in the statute for the agency to fill; consequently, the Court exercised very limited review of the agency’s regulations. Chevron, 467 U.S. at 845, 104 S.Ct. at 2783. In Cardo-za-Fonseca, however, the Court found that the question before it was a “pure question of statutory construction for the courts to decide” and rejected the agency’s interpretation of the statute. Cardoza-Fonseca, 107 S.Ct. at 1220-21.
As previously noted, stare decisis operates with full force whenever a court decides a question of statutory interpretation, regardless of the amount of deference given to the agency in reaching its decision.
IV.
With this framework of statutory analysis in mind, the next step is to examine what the Supreme Court actually did in the Higdon and McNeff cases, when it construed section 8(f) of the NLRA. Did the Court, as I maintain, make a determination of Congressional intent as to the type of contract that a prehire agreement was meant to be? Or did the Court merely find that Congress delegated to the NLRB the task of determining the contractual nature of a prehire agreement, and that the NLRB’s determination was reasonable, thus leaving the agency free to adopt other positions at a later time?
I begin with Higdon. In that case, the Court held that the Board properly applied section 8(b)(7)(C) of the NLRA to a section 8(f) prehire agreement. The Board had held that picketing to enforce a section 8(f) prehire agreement was tantamount to ree-ognitional picketing, and that section 8(b)(7)(C) was violated when the union failed to request an election within 30 days.
In analyzing the Court’s discussion of section 8(f) within the framework of statutory analysis I have set forth above, it is apparent to me that the interpretation of section 8(f) is one of pure statutory construction. The deference given to the Board was that of considering the Board’s position in arriving at that construction.
There is no indication that there was either an express or implicit delegation to the Board to refine standards or to fill in gaps in section 8(f). Nor is it a situation in which the statute is merely being applied to the facts of a particular case. It is purely a construction of what Congress intended in enacting section 8(f). There are extensive references in Higdon to the legislative history and statutory policy of Congress.2 *1143The Court concluded that although the Act made prehire agreements in any other industry an unfair labor practice, Congress intended to create an exception for the construction industry. The Court found that Congress authorized an employer and a union to enter into a voluntary prehire agreement in the construction industry without the union having achieved majority status, and that Congress did not give pre-hire agreements any other status within the NLRA. It was a voluntary agreement that could be repudiated until the union achieved majority status.
The majority cites the following language in the Higdon opinion.
We have concluded that the Board’s construction of the Act, although perhaps not the only tenable one, is an acceptable reading of the statutory language and a reasonable implementation of the purposes of the relevant statutory sections.
Id. 434 U.S. at 341, 98 S.Ct. at 656 (footnote omitted). The majority interprets this language to mean that the Court did not “independently construe the reach and scope of section 8(f).” Maj. op. at 1129-30. I do not read it that way. In my mind, the Court simply articulated the deference it gave to the Board’s views in reaching its own interpretation of the statute.
Certainly, if Higdon leaves any doubt, there is abundant language in McNeff to demonstrate that, in both cases, the Court arrived at its own interpretation of Congressional intent in enacting section 8(f). The issue in McNeff was whether the monetary obligations of a section 8(f) prehire agreement could be enforced prior to repudiation. The opinion repeatedly acknowledged that a section 8(f) agreement was voidable. The references throughout McNeff are to the intent of Congress in enacting section 8(f). For example, the Court stated:
In upholding the Board’s view that a union commits an unfair labor practice by picketing to enforce a prehire agreement before it has attained majority status, we noted in Higdon that this view protects two interests that Congress intended to uphold when it enacted § 8(f).
[O]ur decision in Higdon promotes Congress’ “intention ... that prehire agreements were to be arrived at voluntarily_” Higdon, 434 U.S., at 348, n. 10 [98 S.Ct., at 659, n. 10], In accord with this intention, we approved the Board’s conclusion that a “prehire agreement is voidable” “until and unless [the union] attains majority support in the relevant unit.” Id., at 341 [98 S.Ct., at 655]. Allowing the union to picket to enforce a prehire agreement before it attains majority status is plainly inconsistent with the voidable nature of a pre-hire agreement.
The concerns with the § 7 rights of employees to select their own bargaining representative and our fidelity to Congress’ intent that prehire agreements be voluntary — and voidable — that led to our decision in Higdon are not present in this case.
In a § 301 suit, the District Court merely enforces a contract entered into by the employer — a contract that Congress has legitimated to meet a special situation even though employees themselves have no part in its negotiation or execution. Such enforcement does not grant the plaintiff union a right otherwise enjoyed only by a majority union except in the very narrow sense, expressly intended by Congress, that employers and minority unions in the construction industry do not violate the Act by entering into prehire agreements. There is no sense in which respondents’ contract action has a recognitional purpose like that forbidden in Higdon.
Neither does respondents’ § 301 action trench on the voluntary and voidable *1144characteristics of a § 8(f) prehire agreement. It is clear in this case that petitioner entered into the prehire agreement voluntarily. Moreover, although the voidable nature of prehire agreements clearly gave petitioner the right to repudiate the contract, it is equally clear that petitioner never manifested an intention to void or repudiate the contract.... Whatever may be required of a party wishing to exercise its undoubted right to repudiate a prehire agreement before the union attains majority support in the relevant unit, no appropriate action was taken by petitioner to do so in this case. Consequently, respondents’ suit does not enervate the voluntary and voidable characteristics of the prehire agreement.
Apart from not offending the concerns noted in Higdon, allowing a minority union to enforce overdue obligations accrued under a prehire agreement prior to its repudiation vindicates the policies Congress intended to implement in § 8(f). Congress clearly determined that prehire contracts should be lawful to meet problems unique to the construction industry.
Id. 461 U.S. at 267-71, 103 S.Ct. at 1757-59 (footnotes omitted) (emphasis added).
Throughout the opinion, the Court repeatedly referred to the prehire agreements as being voluntary and voidable with the undoubted right of a party to repudiate the agreement prior to the union achieving majority status. The constant reference to the “voidable nature” of such agreements serves as a strong reminder that the Supreme Court has decided the issue before us in this case and, in doing so, has discerned the Congressional intent behind section 8(f) and rendered its own independent interpretation of that section. As previously discussed, the Court had no choice but to render its independent construction of section 8(f) for it was faced with a pure question of statutory construction. The Supreme Court is the final authority on such matters and the deference it gives to the Board’s view cannot impinge on the stare decisis effect of its ruling.
V.
The majority points to a line of cases in which this circuit adopted a changed Board position, even in face of prior circuit precedent upholding the former Board position. See Opinion at 1135. The cases cited are ones in which we reviewed the Board’s test for determining whether an employer’s interrogation of an employee regarding union activities violates the employee’s rights under section 8(a)(1) of the NLRA. The majority is quite correct in pointing out that the Board over the years changed its test from an “all the circumstances” test to a “per se” rule and back again. Despite the vacillation, and even though we had previously upheld the per se rule, (see, e.g., NLRB v. Fort Vancouver Plywood Co., 604 F.2d 596, 599 n. 1 (9th Cir.1979)), we affirmed the Board’s return to the “all the circumstances” test in Hotel Employees & Restaurant Employees Union, Local 11 v. NLRB, 760 F.2d 1006, 1009 (9th Cir.1985).
The majority, however, has again failed to examine the type of deference operating in those cases. Those cases did not involve questions of pure statutory interpretation, but rather questions relating to the Board’s application of a statutory provision to the facts in particular cases. Specifically, the cases dealt with the standard developed by the Board to apply section 8(a)(1) of the NLRA to circumstances in which an employer interrogates an employee about the employee’s union sympathies. Section 8(a)(1) makes it an unfair labor practice for an employer “to interfere with, restrain, or coerce employees in the exercise of the [employee organizational] rights.... ” 29 U.S.C. § 158(a)(1). In applying the statutory proscription on a case-by-case basis, the Board, at different times, developed two different evidentiary requirements. Under the “per se” standard developed by the Board, any questions concerning union sympathies were deemed inherently coercive in violation of section 8(a)(1). See Hotel, 760 F.2d at 1007. Under the “all the circumstances” test, the interrogation was found unlawful only where “ ‘under all the circumstances the interrogation reasonably tend[ed] to restrain or interfere with *1145the employees in the exercise of rights guaranteed by the Act.’ ” Id. at 1008 (citation omitted).
These differing standards represent a classic example of the Board applying the statute to everyday situations. The review in those cases is directed towards the Board’s manner of enforcing the Act, not to the meaning of the statutory language itself. As in all cases which apply the law to everyday circumstances, the proper scope of review of the Board’s standard in Hotel was “for rationality and consistency with the Act.” Id. The panel concluded, “A standard which considers the totality of the circumstances surrounding an employee interrogation is a realistic approach to the enforcement of section 8(a)(1). It is a standard that is consistent with the Act because the Board ... can determine, on a case-by-case basis, whether all the facts demonstrate coercive behavior.” Id. at 1009 (emphasis added).
In sum, the line of cases cited by the majority do not provide authority for this court’s departure from either Supreme Court precedent or this circuit’s precedent in which questions of statutory meaning are decided.3
VI.
I agree with the majority that we must treat our own circuit precedent in a manner consistent with our treatment of Supreme Court precedent. The majority achieves its consistency by overruling Royal Development v. NLRB, 703 F.2d 363 (9th Cir.1983). My belief is simply that, when a panel of our court decides an issue of pure statutory construction, its ruling is binding on future panels and may be departed from only by an en banc panel. This is so despite the deference accorded to the agency in arriving at that interpretation of the statute.
CONCLUSION
I believe the ramifications of the majority’s ruling extend far beyond the area of labor law into the entire realm of administrative law. Underlying the majority’s holding is, in my mind, a misconception of the nature of deference given to an agency on a purely legal question of statutory interpretation. In the name of administrative deference, the majority would deprive this court of its role of divining Congressional intent behind a statutory provision, and assign that role to the agency charged with administering the statute. The potentially grave consequences resulting from this misapplication of the deference principle need not be enumerated here.
I reiterate my position that McNeff and Higdon constitute binding precedent which *1146we are obliged to follow. The Supreme Court’s holdings, in my view, cannot be dismissed as simply affirming the Board’s then-position while leaving room for the Board to adopt the opposite approach at a later point in time. Rather, the Supreme Court in those cases construed section 8(f) and found, after giving a certain amount of deference to the Board’s views, that Congress intended prehire agreements to be voidable. We are bound by that determination unless and until such time the Court overrules its prior opinions or Congress amends the statute.
. Of course, after operating under a decision of the Court, the agency may persuade the Court that the Court's initial construction of the statute was erroneous, and may urge the Court to adopt an alternative construction. The Court may be persuaded to do so, for the agency's experience in implementing the statute under the former construction may demonstrate that such a construction could not have been intended by Congress in light of the difficulties it poses for implementation, or in light of other policy reasons unveiled since the Court’s decision. But for the Court to adopt the later agency position, it would have to overrule its prior decision adopting the former construction.
. See e.g.,
The Board’s position is rooted in the generally prevailing statutory policy that a union should not purport to act as the collective bargaining agent for all unit employees, and may not be recognized as such, unless it is the voice of the majority of the employees in the unit.
As for § 8(b)(7), which, along with § 8(f), was added in 1959, its major purpose was to implement one of the Act’s principal goals — to ensure that employees were free to make an uncoerced choice of bargaining agent. As we recognized in Connell Construction Co. v. Plumbers & Steamfitters, 421 U.S. 616 [95 S.Ct. 1830, 44 L.Ed.2d 418] (1975), ”[o]ne of the major aims of the 1959 Act was to limit 'top down’ organizing campaigns, in which unions used economic weapons to force recognition from an employer regardless of the wishes of his employees.” Id., at 632 [95 S.Ct., at 1840], and references cited therein.
Congressional concern about coerced designations of bargaining agents did not evaporate as the focus turned to the construction industry, (n. 10)
(n. 10) Congress was careful to make its intention clear that prehire agreements were to be arrived at voluntarily, and no element of coercion was to be admitted into the narrow exception being established to the majority principle. Representative Barden, an impor*1143tant House floor leader on the bill and a conferee, introduced as an expression of legislative intent Senator Kennedy’s explanation the year before of the voluntary nature of the prehire provision[.]
Higdon, 434 U.S. at 344-48, 98 S.Ct. at 657-59 (emphasis added).
. Another example of the agency changing its position with this court’s approval is the Board’s treatment of representation elections. Over a period of 20 years, the NLRB changed its position four times on the issue of when a representation election should be set aside on the grounds of misrepresentation. NLRB v. Best Products Co., Inc., 765 F.2d 903, 910 (9th Cir.1985). The NLRB vacillated between the Hollywood Ceramics standard, where an election would be set aside when there was a "substantial departure from the truth,” and the Shopping Kart Food Market standard, under which an election would not be set aside solely because there had been misrepresentations of fact. See Best, 765 F.2d at 910-11 (discussing NLRB’s changes of policy). In Best, this court upheld the Board’s adherence to the Shopping Kart standard even though the Board had vacillated between the two standards, and even though this court had previously upheld the Hollywood Ceramics standard. See, e.g., NLRB v. Sauk Valley Mfg. Co., Inc., 486 F.2d 1127, 1131 (9th Cir.1973). We recognized in Best, "The Board may alter its standards provided that its new rules are both rational and consistent with the Act.” 765 F.2d at 912. We noted that the "NLRB has wide discretion to determine representation matters and questions arising during election proceedings,” id. at 908, and confined our standard of review to two inquiries: (1) "whether the Board acted within an area of regulation committed to it by Congress,” and (2) “whether the Board properly applied the correct legal standard.” Id. at 907. We concluded that there was a “‘reasonable basis in law’ for the Board’s change of policy" and deferred to the Board’s decision. Id. at 913. Again, this is the type of deference which is properly employed in reviewing a Board’s application of the statute to the facts of particular cases. It must not be confused with the deference accorded to an agency’s view of the meaning of the statute. As with the line of cases mentioned in the majority opinion, this example provides no support for departing from prior circuit precedent on issues of statutory construction.