THORNBERRY, Circuit Judge:
Mobil Oil Corporation Marine Transportation Department, Gulf-East Coast Operations [Company] and the Oil, Chemical and Atomic Workers International Union AFL-CIO and Maritime Local No. 8-801 of that Union [Union] entered into a collective bargaining agreement containing an agency shop clause. Shortly thereafter the company brought suit in federal district court for a declaratory judgment under section 301 of the Labor Management Relations Act of 1947, 29 U.S.C. § 185(a), (c), and 28 U.S.C. § 2201, hoping that the court would find the clause violative of the Texas right to work law and thus void and unenforceable.
After a full evidentiary hearing, the district court concluded that the Texas law did apply to this employment relationship and did render the agency shop provision invalid. On appeal, a panel of this court held that a justiciable controversy existed, that jurisdiction properly vested in the district court, and that there was no abuse of discretion by that court in declining to defer to the NLRB. But on the substantive issue the panel concluded that the district court had erred and that the agency shop clause is valid and enforceable with respect to all employees covered by the collective bargaining agreement. The panel held that the Texas right to work law does not apply to this employment relationship because these employees are not employees of Texas or of any other state, but are rather seamen whose job site is on the high seas. Because of the exceptional importance1 and the unique nature of the question presented, we granted rehearing en banc to consider the question of the applicability of the Texas right to work law.2
I.
The parties to this suit, the district court, and the original panel approached the issue presented by analyzing this employment relationship in terms of its contacts with the state of Texas and other jurisdictions. It is conceded by all that employer/employee contacts with the state of Texas far outweigh contacts with any other state, and the evidence supporting that conclusion is neither conflicting nor ambiguous.
The company operates eight oceangoing tankers which move principally between the state of Texas and the state of New York along the East and Gulf Coasts. Since 1962, the headquarters of the company has been located in Beaumont, Texas. All personnel and payroll records are maintained there; all payroll checks are written and mailed from the Beaumont offices; all state and federal taxes are deducted from the seamen’s wages in Beaumont; all monthly check-offs of Union dues for all seamen is performed in Texas; and all grievances filed on the approved grievance forms by the seamen are submitted to [274]*274and considered by the company’s manager at his office in Beaumont, pursuant to the collective bargaining agreement. The state of Texas is the only state administering unemployment compensation for the company’s seamen.
The company employs 289 unlicensed seamen; 123 of the 289 have reported Texas as their state of residence; and 152 of the 289 have requested the company to list Beaumont as their shipping port. All actual hiring of seamen occurs in Texas. While approximately 40% of the seamen first make application in New York and 60% make application in Beaumont, the final hiring decision is made only in Beaumont. As many as 60% of the applicants actually come from the Texas labor force. Furthermore, all terminations of seamen are finalized in Beaumont as a result of decisions made at the Beaumont headquarters. The evidence suggests that from the seaman’s first encounter with the company to his last, the Beaumont headquarters is constantly in the picture.
The above contacts demonstrate that the state of Texas is intimately involved with this employment relationship. And the district court was clearly correct in its finding that a more substantial part of the administration and performance of the collective bargaining agreement occurs in Texas than in any other state.3 We think it • not insignificant to note that this “justiciable controversy” arose in Beaumont, for it was in the context of the hiring process there that the Union and the company began this battle.4
The panel, however, believed that the predominance of Texas contacts with this employment relationship was not decisive. Instead, the panel concluded that the determinative factor in deciding whether the Texas law is applicable is job situs. The record indicates that the seamen spend 80% to 90% of their working time on the high seas. Each seaman works roughly a 120 day cycle consisting of 90 days aboard ship and 30 days ashore “on vacation” during which shore time he continues to be an employee of the company and to receive wages. The panel’s theory is that, with so much of a seaman’s working time reserved for the high seas, the Texas law, and for that matter the law of any state, is rendered inapplicable. The panel does not explain, however, why job situs is the singularly most significant, indeed the decisive, factor.5
Rather than adopt the panel’s limited approach, we choose to weigh all [275]*275of the contacts in the context of our national labor policy. We disagree with the panel’s conclusion that job situs has some talismanic quality. We would agree with the panel opinion that the district court’s finding that a more substantial part of the administration and performance of the collective bárgaining agreement occurs in Texas than in any other state is not decisive. The number of contacts is a highly relevant consideration, but the true importance of these contacts is to be understood not in a quantitative vacuum but in the qualitative background furnished by the Congress.
Whether the agency shop clause in this collective bargaining agreement is valid or not depends entirely upon whether the Texas right to work law applies to this employment relationship.6 In determining whether the Texas law is applicable, we must decide whether it was the intent of Congress in writing section 14(b) ’to permit a state to apply its right to work law to an employment relationship which has contacts with the state as described by the evidence presented to the district court.
Our task is made more difficult by the fact that there is nothing in the legislative history of the Taft-Hartley Act which expressly answers our question one way or the other. Nevertheless, we must determine as best we can what Congress would have intended on this point had it been presented for their explicit consideration. 7 Our job is not to add up all of the contacts and to decide which jurisdiction has “the most.” Our task is to discern whether application of the Texas right to work law would be in furtherance of the Taft-Hartley Act’s expressed purposes.8 See National Woodwork Mfrs. Ass’n v. NLRB, 1967, 386 U.S. 612, 87 S.Ct. 1250, 18 L.Ed.2d 357; Wirtz v. Local 153, Glass Bottle Blowers Ass’n, 1968, 389 U.S. 463, 468, 88 S.Ct. 643, 646, 19 L.Ed.2d 705; NLRB v. Metallic Bldg. Co., 5th Cir. 1953, 204 F.2d 826, 828. And in arriving at a just, proper, and congressionally authorized conclusion, we invoke the eclectic approach of Chief Justice Marshall :
Where the mind labors to discover the design of the legislature, it seizes everything from which aid can be derived .
United States v. Fisher, 1805, 2 Cranch (6 U.S.) 358, 386, 2 L.Ed. 304, 313; see also United States v. Dickerson, 1940, 310 U.S. 554, 562, 60 S.Ct. 1034, 1038, 84 L.Ed. 1356.
Having done so, we conclude that it was not the intent of the Congress to prevent Texas from applying its right to work law in the context of this employment relationship. Indeed, in our judgment, application of the Texas law would be in furtherance of our national labor policies. For the reasons expressed below, we hold that the federal labor legislation, the predominance of Texas contacts over any other jurisdiction, and the significant interest which Texas has in applying its right to work law to this employment relationship warrant application of the Texas law and, consequently, invalidation of the agency shop provision.
[276]*276II.
We begin with “the language of the statute itself.” Jones v. Alfred H. Mayer Co., 1968, 392 U.S. 409, 420, 88 S.Ct. 2186, 2193, 20 L.Ed.2d 1189. Sections 7 and 8(a)(3) of the Labor Management Relations Act, 29 U.S.C. §§ 157, 158(a)(3), specifically authorize employers and labor organizations to enter into union “hop agreements by whose terms membership in a union, after a short waiting period, is required as a condition of employment.9 The Supreme Court has interpreted these sections to authorize an agency shop, a form of union security conditioning employment on the payment of regular union dues and initiation fees in lieu of actual union membership. NLRB v. General Motors Corp., 1963, 373 U.S. 734, 738-739, 83 S.Ct. 1453, 10 L.Ed.2d 670; Retail Clerk’s Inter. Ass’n v. Schermerhorn [.Schermerhorn I], 1963, 373 U.S. 746, 751, 83 S.Ct. 1461, 1464, 10 L.Ed.2d 678.
In the same stroke, section 8(a)(3) was redrafted to prohibit the closed shop which its predecessor section, section 8(3) of the National Labor Relations Act, had permitted. H.R.Conf.Rep.No. 510, 80th Cong., 1st Sess. 60, 1 Leg. Hist.L.M.R.A. 545, U.S.Code Cong.Serv. 1947, p. 1135. The closed shop differs from the union shop in only one regard: the closed shop requires that the job applicant be a member of the union before he is hired. 11 Hou.L.Rev. 709, 710, n. 7 (1974).
Section 14(b) of the Act, 29 U.S.C. § 164(b), provides:
Nothing in this subchapter shall be construed as authorizing the execution or application of agreements requiring membership in a labor organization as a condition of employment in any State or Territory in which such execution or application is prohibited by State or Territorial law.
By enacting this provision, Congress desired to “make certain” that section 8(a)(3) of the Act could not “be said to authorize arrangements of this sort [compulsory unionism] in States where such arrangements were contrary to the State policy.” H.R.Conf.Rep.No. 510, supra, 1 Leg.Hist.L.M.R.A. at 564; Schermerhorn I, supra; U.S.Code Cong. Serv. 80th Cong., 1st Sess.1947, p. 1135. The language of section 14(b) as well as its legislative history suggests an intent on the part of the Congress to save to the states the right to prohibit the practical equivalents of compulsory unionism, including the agency shop. Schermerhorn I, supra; Retail Clerks Inter. Ass’n v. Schermerhorn, 1963, 375 U.S. 96, 98, 84 S.Ct. 219, 220, 11 L.Ed.2d 179 [Schermerhorn II]; NLRB v. Houston Chap., Asso. Gen. Con. of America, Inc., 5th Cir. 1965, 349 F.2d 449, 453.
The national labor policy expressed in these sections of the Taft-Hartley Act is one which directly authorizes union shop arrangements, but only in those states which have not passed the right to work [277]*277laws. In a sense, a conflict between state and federal law may thereby arise
but it is a conflict sanctioned by Congress with directions to give the right of way to state laws barring the execution and enforcement of union-security agreements .
Schermerhorn II, supra, 375 U.S. at 103, 84 S.Ct. at 222, 11 L.Ed.2d at 184. In essence, the congressional judgment is that state right to work laws are in harmony with, and part of, the national labor policy.
The district court focused its attention on the “application” of the agency shop clause requiring the payment of dues and fees “as a condition of employment in” Texas. Believing that the collective bargaining agreement received its most significant application in Texas, the district court concluded that the Texas law applied. The panel, however, began with the premise that state right to work laws are applicable only to employees of those particular states and concluded that the employees in this case are not for the purpose of their collective bargaining agreement employees of Texas or any other state because of their job situs. Fortunately, Congress has given us some indication of the approach it would favor.
When the Congress prohibited the closed shop while permitting the union shop in the Taft-Hartley Act, the Congress was demonstrating its concern with the process by which men and women were hired. As conceived by that legislative body the closed shop enabled union hiring halls to control absolutely the pool of labor available to an employer. H.R.Rep.No. 245, 80th Cong., 1st Sess., 1 Leg.Hist.L.M.R.A. 295, 300. The hiring decision was, in effect, made by the union hiring hall, for the employer never saw a job applicant unless the applicant had been referred by the union, and of course, only union members were referred. The Congress singled out the maritime industry as a serious problem area'.10 The legislative history of the Act exhibits an express intent to give to the employer some measure of freedom in the hiring process. See H. R.Rep.No.245, supra; see also, e. g., 93 Cong.Rec. A1297 (1947) (remarks of Representative Gerald Landis) and 93 Cong.Rec. A2011 (1947) (remarks of Representative Hugh Meade), infra at note 11. The union shop agreement permits an employer to hire a person who has not, at or before the moment of hiring, joined the union. With the proscription of the closed shop, union hiring halls are now required to refer job applicants on a nondiscriminatory basis, so that an employer may consider nonunion applicants on an equal basis with union applicants. NLRB v. Houston Chap., Asso. Gen. Con. of America, Inc., supra. In short, if the congressional discussion on compulsory unionism could be said to have any single reference point, it is not job situs but the hiring process.11 [278]*278And in our judgment, the language of section 14(b) which looks to the “execution or application” of agreements requiring membership as a condition of employment refutes the notion that Congress was suddenly singularly concerned with the job situs of the employee rather than with the effects of such an agreement on men and women seeking employment in a mobile society.12 The language and history of section 14(b) suggest that a state may apply its right to [279]*279work law whenever the state has a significant interest in the process by which a union security agreement is applied. No one disputes the fact that this agency shop agreement receives its application in Beaumont, Texas. Indeed, it is the company’s resolute refusal to apply the agreement in Beaumont, Texas, that has caused this suit.
The preamble to the Texas right to work law records the legislative judgment that because unions exercise great influence over the economic well-being and freedom of the Texas labor force, the state has an important interest in removing union coercion from the hiring process.13 Each of the Texas right to work provisions expresses a concern with that process. Article 5207a, § 2, provides: “No person shall be denied employment on account of membership or nonmembership in a labor union.” Article 5154a, § 8a provides: “It shall be unlawful for any labor union, to collect . . . any fee . whatsoever, as a work permit or as a condition for the privilege to work from any person not a member of the union. . . .” And, lastly, article 5154g, § 1, provides that “ . the right of persons to work shall not be denied or abridged on account of membership or nonmembership in any labor union.” Again, if any single legislative focus can be discerned from the Texas provisions, it is not job situs or the place of job performance but the hiring process. Thus one triggering point for the application of these provisions is hiring within the state. As indicated earlier, this is consistent with the language and intent of section 14(b).
If, as we have noted, the Texas laws were enacted to afford the Texas labor force protection from compulsory unionism, the fact that as many as 60% of the job applicants come from the Texas labor force gives Texas good reason for applying' its right to work provisions. But the state’s interest goes far beyond its predominant association with the hiring process and its numerous other contacts with this employment relationship which have already been listed. During the one-fourth of the year when the employees are “on vacation” in Texas, they presumably have homes and families; they pay taxes; they send their children to school; they eat, sleep, and move about; they act like other citizens of the state. In short, they are precisely the people whom the Texas law was designed to protect.
The high seas definitely have contact with the employees, but what is the interest of the high seas in preventing application of the Texas law? How is some important non-Texas, national, or international policy furthered by refusing to apply the Texas law? The fact that maritime work is interstate, or even international, in character does not imply that the Texas law is inapplicable since federal regulation in these matters is predicated upon interstate commerce. No one who studies the legislative history of section 14(b) can seriously argue that Congress intended to permit the states to prohibit compulsory unionism only in those instances where interstate commerce was not affected.14 Conse[280]*280quently, there is no uniformity principle at stake here since a national or international union may have a union shop in one state and find that arrangement prohibited in another by the state law.15
The panel opinion suggested that allowing Texas to apply its right to work law here would allow an employer to make an “arbitrary choice,” could produce “bizarre consequences,” and could defeat the purposes of sections 7, 8, or 14(b) of the Act. On the contrary, as the legislative history indicates, allowing Texas to apply its laws here is in furtherance of the national labor policy expressed in the Act and in furtherance of the Texas legislative judgment. Furthermore, the panel’s fear that employers might choose headquarter locations in states with right to work laws and thereby defeat national labor policy is, on this record, unfounded. First, the employer’s headquarters is not the decisive factor, any more than job situs is. To reiterate, we have chosen to weigh all of the employer and employee contacts in the context of our national labor policy. Secondly, the Union has never even alleged that Mobil moved to Texas in order to seek the shelter of the Texas right to work law. We will have time enough to consider that factor in the proper case. We should note; however, that employers may set up their operations anywhere they choose, incurring the responsibilities and gaining the advantages of that locale as they do so. The Union has directed us to no language in section 14(b) or its legislative history which suggests that a company cannot freely choose to establish a substantive enterprise in a right to work state for the purpose of obtaining the benefit of the state’s law.
It is the antithesis of the method we use here to permit an employer to make a truly arbitrary choice of the state law to be applied. The applicability of a state right to work law will depend upon whether such application would further the policies and purposes of our national labor laws. For example, where an employer has artifically constructed its operations or its hiring process in such a way that the substance of its activities occurs in and the greatest part of its labor force comes from a non-right to work jurisdiction while a pro forma final hiring decision is made in a right to work state, application of the state’s right to work law would be contrary to the national labor policy. We are not faced with that circumstance here.
Moreover, we see no bizarre consequences flowing from the method used or the result reached today. Yet refusing to allow Texas to apply its law here would create the bizarre consequence of [281]*281exempting the maritime industry from the operation of section 14(b). We have found nothing in the legislative history or the language of any present labor legislation, and the Union has directed us to none, which suggests an intent on the part of the Congress to create a special rule for maritime workers. On the contrary, the legislative history of the Taft-Hartley Act indicates that Congress was aware of the problems in the maritime industry caused by compulsory unionism and sought to deal with them. See S.Rep. No. 105, 80th Cong., 1st Sess., 1 Leg.Hist. 412-413; footnote 9, supra.
In the past, when Congress has decided to supersede section 14(b) and state right to work laws, it has done so expressly. Prior to 1951, the Railway Labor Act, 45 U.S.C. § 151 et seq., prohibited union shop agreements. But in 1951, Congress amended the Act to permit carriers covered by the law and their employees to enter into union shop agreements
[notwithstanding any other provisions of this Act, or of any other statute or law of the United States, or Territory thereof, or of any State.
45 U.S.C. § 152, subd. 11 (emphasis added). Fully aware of the Taft-Hartley Act and its section 14(b), Congress used express language in the 1951 amendment precisely to defeat application of state right to work laws. Railway Employees’ Dept. v. Hanson, 1956, 351 U.S. 225, 232, 76 S.Ct. 714, 718, 100 L.Ed. 1112; see also Sams v. Brotherhood of Railway and Steamship Clerks, 4th Cir. 1956, 233 F.2d 263, 264; Inter. Ass’n of Machinists v. Sandsberry, Tex.Civ.App.1954, 277 S.W.2d 776, 780-781, aff’d, 1956, 156 Tex. 340, 295 S.W.2d 412, cert. denied, 1957, 353 U.S. 918, 77 S.Ct. 669, 1 L.Ed.2d 665. In our judgment, this instance of congressional action demonstrates that state right to work laws are not to be ignored unless the language or legislative history of the federal law permits such a course.16 Railway Employees’ Dept. v. Hanson, supra. State prohibition of compulsory unionism is a “congressional dispensation of grace, not the imperious right of a state,” and thus Congress has always had the power to exempt seamen from the operation of state right to work laws. SeaPAK v. Industrial, Technical & Pro. Emp., Natl. Maritime Union, S.D.Ga.1969, 300 F. Supp. 1197, 1201, aff’d, 5th Cir. 1970, 423 F.2d 1229, aff’d, 1971, 400 U.S. 985, 91 S.Ct. 452, 27 L.Ed.2d 434; Railway Employees’ Dept. v. Hanson, supra. But it has chosen not to do so.
If we were to carve out an exception for seamen here, in the absence of any suggestion to do so from Congress, we would be encroaching on the legislative authority. That course of action, in spite of Justice Cardozo’s frank observation that judge-made law is “one of the existing realities of life,” is beyond our constitutional prerogatives. See Cardozo, The Nature of the Judicial Process (1921 ed.),p. 10.
III.
Alternatively, the Union contends that it is improper to strike out the agency shop clause in its entirety from the collective bargaining contract since some of the seamen have greater contacts with states other than Texas, states without right to work laws. The Union asks that we fashion a rule which renders the Texas right to work law inapplicable to those employees who have greater contracts with non-right to work states than with Texas. We decline the Union’s invitation for two reasons.
The agency shop clause in the collective bargaining contract states:
For the duration of the Agreement all employees hired shall, as a condition of employment, become members of the Union and/or in the alternative pay the regular union dues and initia[282]*282tion fees within 31 days from the employment date.
This clause unambiguously demands that all employees hired join the union or pay the required fees. The clause makes no distinctions between job applicants who are to be protected by the Texas law and those who are not. As the panel majority noted, we cannot rewrite the contract to reflect these allegedly distinguishable classes.17
Secondly, even if we could rewrite the contract, on this record we are unwilling to hold that there is an identifiable group of employees who are not subject to the Texas law. The Union’s brief admits that “the record before the court was insufficient to enable the court” to authorize partial application of the agency shop provision. Since we are not the fact-finders in the judicial process, we have no basis for implementing the Union’s alternative request.
IV.
We have attempted a multi-dimensional analysis of the question presented for our disposition. On one level we have examined this employment relationship in terms of its contacts with the state of Texas and other jurisdictions. On another we have studied Congress’ actions in passing the Taft-Hartley Act and the 1951 amendment to the Railway Labor Act in order to give meaning and context to the undisputed contacts. On still a third level we have scrutinized the interests of different jurisdictions in applying their laws to this employment relationship.
The resulting matrix offers a clear picture. Texas is intimately involved with this employment relationship. Texas has a more significant interest in applying its law than does any other jurisdiction. And the federal labor legislation suggests a congressional intent to allow Texas to apply its law, rather than to exempt the maritime industry from the operation of section 14(b). Having exhausted, as best we can, the reality before us, we are faced with but one result.
The judgment of the district court is affirmed.