John Halkias, et al., Plaintiffs,john Halkias and Barry Jackson, Plaintiffs-appellants, v. General Dynamics Corporation, Defendant-appellee.john Anthony Cureington, Plaintiff-appellant, v. General Dynamics Corporation, Defendant-appellee.alvin Staudt, on Behalf of Himself and All Others Similarlysituated, Plaintiff-appellant, v. Glastron, Inc., Defendant-appellee, 31 F.3d 224 (5th Cir. 1994)

Annotate this Case
U.S. Court of Appeals for the Fifth Circuit - 31 F.3d 224 (5th Cir. 1994) Aug. 24, 1994. Rehearing En Banc OrderedSept. 22, 1994

Art Brender, Ft. Worth, TX, for appellants in Nos. 93-1664, 93-1680.

Kary L. Moss, Exec. Dir., Detroit, MI, for amicus--Texas AFL-CIO, et al. in Nos. 93-1664, 93-1680.

Paul D. Inman, Gibson, Dunn & Crutcher, Dallas, TX, for appellee in No. 93-1664.

Paul David Inman, Karl G. Nelson, Gibson, Dunn & Crutcher, Dallas, TX, for appellee in No. 93-1680.

Randal C. Gray, New Braunfels, TX, for appellant in No. 93-8204.

Kary L. Moss, Exec. Dir., Barbara Harvey, NLG, Detroit, MI, for amicus--Texas AFL-CIO & Atomic Workers.

Ellen E. McLaughlin, Lee P. Schafer Seyfarth Shaw Fairweather & Geraldson, Chicago, IL, for appellee in No. 93-8204.

Appeals from the United States District Court for the Northern District of Texas.

Appeal from the United States District Court for the Western District of Texas.

Before WISDOM, BARKSDALE, and EMILIO M. GARZA, Circuit Judges.

RHESA HAWKINS BARKSDALE, Circuit Judge:


These appeals present an issue of first impression in our circuit: the limitations period for an action under the Worker Adjustment and Retraining Notification Act (WARN), 29 U.S.C. §§ 2101-2109. Both district courts applied the six-month period provided by Sec. 10(b) of the National Labor Relations Act (NLRA), 29 U.S.C. § 160(b). We AFFIRM.

Two of the actions (Halkias' and Cureington's) are against General Dynamics Corporation; one (Staudt's), against Glastron, Inc. They concern not receiving timely notice in advance of a layoff, contrary to WARN, referred to by many as a "plant closing" law. E.g., 134 Cong. Rec. S8546 (June 24, 1988) (Senator Grassley); id. at S8665 (June 28, 1988) (Senator Specter).1 

WARN requires a business that employs more than 100 workers to provide at least 60 days' written notice before a "plant closing" or a "mass layoff". 29 U.S.C. §§ 2101-02; see also id. Sec. 2101(a) (3)-(4) (defining "plant closing" and "mass layoff"). Failure to provide such notice results in the business' liability to those who suffered an "employment loss" for back pay and benefits for each day of the violation. Id. Sec. 2104(a) (1); see also id. Sec. 2101(a) (6) (defining "employment loss" as termination, layoff exceeding six months, or reduction of hours of work by more than 50 percent for six months). WARN provides for a federal action to recover these damages, 29 U.S.C. § 2104(a) (5), but does not include a limitations period.

On January 7, 1991, the Department of Defense cancelled a contract with General Dynamics, prompting it, one day later, to institute a "mass layoff" at its facilities in Texas, Oklahoma, and Missouri.2  Halkias was one of the affected employees at the Fort Worth, Texas, facility. Almost two years later, on November 24, 1992, he and other General Dynamics employees at the Fort Worth and Oklahoma (Tulsa) facilities filed a class action in district court, claiming that they were laid off in violation of WARN.3  In its final form, the action was on behalf of approximately 2,000 former salaried, non-union General Dynamics employees at the two facilities.4 

General Dynamics moved for judgment on the pleadings, asserting that the six-month limitations period applicable to unfair labor practice claims under Sec. 10(b) of the NLRA, 29 U.S.C. § 160(b), should be borrowed, and if so, Halkias' action was time-barred. The district court agreed.

Cureington's appeal arises out of the same facts; indeed, the parties to his action agreed to transfer it to the district court adjudicating Halkias', because the claim was identical to, and embraced by, Halkias' class action.5  Although Cureington's action was never formally consolidated with Halkias', the district court dismissed Cureington's complaint sua sponte, because he failed to file within the six-month period.

Staudt's class action against Glastron alleged that it "laid off" over 250 employees at its New Braunfels, Texas, facility between October 31 and December 31, 1990; but Staudt did not file suit until December 17, 1992, approximately two years later.6  Glastron's motion to dismiss, on the basis that the action was barred by the limitations period that should be borrowed from the NLRA, was granted.

As noted, we address an issue of first impression for our court: the WARN limitations period.7  District courts addressing it are divided: like the district courts in these cases, some have applied the NLRA's six-month period;8  others, state limitations periods.9  The Second and Third Circuits, which are the only other circuits to have addressed this issue, rejected the NLRA period and held that a state limitations period was appropriate. United Steelworkers of Am. v. Crown Cork & Seal Co., 32 F.3d 53 (3d Cir. 1994); United Paperworkers Local 340 v. Specialty Paperboard, Inc., 999 F.2d 51, 57 (2d Cir. 1993). Most reluctantly, we part company with our sister circuits, and hold that the NLRA period should be applied.

Congress' failure to provide a limitations period for WARN "is often the case in federal civil law". DelCostello v. International Bhd. of Teamsters, 462 U.S. 151, 158, 103 S. Ct. 2281, 2287, 76 L. Ed. 2d 476 (1983). In such a case,

we do not ordinarily assume that Congress intended that there be no time limit on actions at all; rather, our task is to "borrow" the most suitable statute or other rule of timeliness from some other source. We have generally concluded that Congress intended that the courts apply the most closely analogous statute of limitations under state law.

Id. (footnote omitted). The task of borrowing an appropriate limitations period has been accurately characterized as "a matter of which round peg to stuff in a square hole." Short v. Belleville Shoe Mfg. Co., 908 F.2d 1385, 1393 (7th Cir. 1990) (Posner, J., concurring), cert. denied, 501 U.S. 1250, 111 S. Ct. 2887, 115 L. Ed. 2d 1052 (1991). Due to the many competing reasons for borrowing various periods, this case is no exception. Indeed, it is a classic example.

Complicating our task is the need to consider whether a federal limitations period provides a superior vehicle for WARN's enforcement. See DelCostello, 462 U.S. at 162, 103 S. Ct. at 2289 ("state statutes of limitations can be unsatisfactory vehicles for the enforcement of federal law"). To be sure, the Court continues to caution that "resort to state law remains the norm for borrowing of limitations periods." Id. at 171, 103 S. Ct. at 2294; accord Lampf, Pleva, Lipkind, Prupis & Petigrow v. Gilbertson, 501 U.S. 350, 355, 111 S. Ct. 2773, 2778, 115 L. Ed. 2d 321 (1991) (plurality) ("It is the usual rule that when Congress has failed to provide a statute of limitations for a federal cause of action, a court 'borrows' or 'absorbs' the local time limitation most analogous to the case at hand.") (citations omitted).10  But, under appropriate conditions, we may look to federal law to borrow the period.11 

The Court has stated that:

when a rule from elsewhere in federal law clearly provides a closer analogy than available state statutes, and when the federal policies at stake and the practicalities of litigation make that rule a significantly more appropriate vehicle for interstitial lawmaking, we have not hesitated to turn away from state law.

DelCostello, 462 U.S. at 172, 103 S. Ct. at 2294; accord Lampf, 501 U.S. at 356, 111 S. Ct. at 2778 (plurality).

Consistent with the trend towards utilization of federal limitations periods, discussed in note 11, supra, our court has borrowed the NLRA's limitations period in a number of cases since DelCostello. See, e.g., Landry v. Air Line Pilots Ass'n, 901 F.2d 404, 410-14 (5th Cir.) (applying period to claim for breach of duty of fair representation under Railway Labor Act), cert. denied, 498 U.S. 895, 111 S. Ct. 244, 112 L. Ed. 2d 203 (1990); Trial v. Atchison, T. & S.F. Ry., 896 F.2d 120, 124-26 (5th Cir. 1990) (same); Aluminum, Brick & Glassworkers Int'l Union Local 674 v. A.P. Green Refractories, Inc., 895 F.2d 1053, 1054-55 (5th Cir. 1990) (applying period to "pure" Sec. 301 actions under Labor Management Relations Act); Coyle v. Brotherhood of Ry., Airline, & S.S. Clerks, 838 F.2d 1404, 1405-06 (5th Cir. 1988) (applying period to breach of contract claim against union under Sec. 2 of Railway Labor Act).

The decision to apply a federal, rather than a state, period is a "delicate" one. See Lampf, 501 U.S. at 356, 111 S. Ct. at 2778 (plurality). In DelCostello, the Court selected the NLRA's limitations period after giving due consideration to whether the state period might hinder the federal policy at issue, DelCostello, 462 U.S. at 162-69, 103 S. Ct. at 2289-93, and noting the "family resemblance" between the federal statute at issue and the NLRA. Id. at 170-71, 103 S. Ct. at 2293-94. In Agency Holding Corp. v. Malley-Duff & Assoc., Inc., 483 U.S. 143, 107 S. Ct. 2759, 97 L. Ed. 2d 121 (1987), the Court considered whether a uniform period is desirable, id. at 148-49, 107 S. Ct. at 2763, and whether a federal statute provided a "far closer analogy ... than any state law alternative", paying particular attention to the "similarities in purpose and structure" between the two federal statutes. Id. at 150, 152, 107 S. Ct. at 2764, 2765. It also focused on litigation practicalities, particularly the potential for forum shopping generated by application of diverse state periods. Id. at 153-54, 107 S. Ct. at 2765-66.

In Lampf, a plurality of the Court attempted to promulgate a three-tier "hierarchical inquiry for ascertaining the appropriate limitations period". Lampf, 501 U.S. at 356, 111 S. Ct. at 2778-79 (plurality). Following this approach, one first decides "whether a uniform statute of limitations is to be selected", regardless of whether that uniform period is to be a particular species of state periods or a single, federal one. Id. at 357-58, 111 S. Ct. at 2779 (plurality). If uniformity is desirable, one next chooses either a species of state limitations periods or a federal period, paying particular attention to "the geographic character of the claim" and forum shopping concerns. Id. (plurality). Finally, even if such considerations counsel in favor of the federal period, one must still decide that the "federal source truly affords a 'closer fit' with the cause of action at issue than does any available state-law source." Id. (plurality).

Although the Lampf hierarchy commanded only a plurality,12  its structure is consistent with the earlier approach of eight members of the Court in Agency Holding. On the other hand, the Court has borrowed a federal period without first explicitly ascertaining the need for uniformity. See DelCostello, 462 U.S. at 165-71, 103 S. Ct. at 2291-94. In fact, the discussion in Agency Holding of the desirability of a uniform period for RICO claims may be embraced as a component of DelCostello 's dictate that "litigation practicalities" and policy implications be considered. See Agency Holding, 483 U.S. at 149-50, 107 S. Ct. at 2764 ("a uniform statute of limitations is required to avoid intolerable uncertainty and time-consuming litigation") (internal quotations and citation omitted); DelCostello, 462 U.S. at 172, 103 S. Ct. at 2294 (identifying "litigation practicalities" and the effects on federal policy as factors in choosing a federal limitations period). Accordingly, we follow DelCostello 's general outline, recognizing that subsequent cases have elaborated on that theme.

We first examine whether the NLRA limitations period is more analogous to WARN than available state periods. See DelCostello, 462 U.S. at 172, 103 S. Ct. at 2294 (requiring that the federal period provide a "closer analogy than available state statutes"). To undertake this examination, we first describe the similarities between the NLRA and WARN, and then compare and contrast those similarities with the available state period(s).13 1.

Agency Holding, which held that the Clayton Act's limitations period was applicable to RICO claims, noted that those acts shared "similarities in purpose and structure". Agency Holding, 483 U.S. at 152, 107 S. Ct. at 2765. The same can be said of the NLRA and WARN.

WARN "requires some employers--generally those who are curtailing or closing an operation--to provide sixty days notice to those employees who will be laid off or whose hours will be substantially reduced." Carpenters Dist. Council v. Dillard Dep't Stores, Inc., 15 F.3d 1275, 1278 (5th Cir. 1994).14  It imposes this requirement to provide employees an opportunity to look for other jobs or seek retraining. See 20 C.F.R. Sec. 639.1(a) (1993).

" [T]he NLRA ... [was] enacted to protect the right of workers to join together ... and collectively bargain for the terms and conditions of employment." United Paperworkers, 999 F.2d at 54; see also 29 U.S.C. § 151. To provide such protection, the NLRA, among other things, "conferr [ed] certain affirmative rights on employees and [placed] certain enumerated restrictions on the activities of employers." American Ship Bldg. Co. v. NLRB, 380 U.S. 300, 316, 85 S. Ct. 955, 966, 13 L. Ed. 2d 855 (1965). For example, Sec. 8(a) of the NLRA proscribes "unfair labor practices" by employers. See 29 U.S.C. § 158(a).

Unquestionably, WARN and the NLRA share similar structures; "the family resemblance is undeniable, and indeed there is a substantial overlap" between them. See DelCostello, 462 U.S. at 170, 103 S. Ct. at 2293 (not discussing WARN). In fact, the Department of Labor15  borrowed extensively from the NLRA in promulgating regulations for WARN. See 20 C.F.R. Sec. 639.3(d) (1993) (defining "representative" for WARN purposes by explicit reference to Secs. 9(a) and 8(b) of the NLRA); 20 C.F.R. Sec. 639.3(a) (1) (ii) (1993) (defining "reasonable expectation of recall"); 20 C.F.R. Sec. 639.3(a) (2) (1993) (defining "independent contractors and subsidiaries" by reference to "existing legal rules", i.e., case law interpreting the NLRA); see also 54 Fed.Reg. 16,045 (1989) (explicitly looking to NLRA for guidance in promulgating 20 C.F.R. Sec. 639.3(a) (2) definition of "independent contractors and subsidiaries"); 54 Fed.Reg. 16,044-45 (1989) (discussing promulgation of "reasonable expectation of recall" language in 20 C.F.R. Sec. 639.3(a) (1) (i) by referring to NLRA case law). After WARN's enactment, the NLRB's General Counsel "predicted substantial interplay between the new law [ (WARN) ] ... and the nation's basic labor law administered by the NLRB." NLRB General Counsel Outlines Overlap Between Plant Closing Law and Taft-Hartley, 226 Daily Lab.Rep. (BNA), Nov. 23, 1988, at A-3.16 

Given the linguistic overlap between WARN and the NLRA, it is not surprising that, like the Department of Labor, federal courts have turned to NLRA case law in interpreting WARN. For example, in Damron v. Rob Fork Mining Corp., 739 F. Supp. 341 (E.D. Ky. 1990), aff'd, 945 F.2d 121 (6th Cir. 1991), the district court was required to resolve "the extent to which laid off persons are to be included in the calculation of 'employees' for purposes of the WARN Act." Id. at 342. Its resolution of that issue followed the lead of the Department of Labor and looked to the NLRA:

The specific WARN Act analysis requires a determination of whether an employee would "reasonably experience an employment loss". Rather than await case law development of this phrase, the Secretary [of Labor] adopted a substantially similar analysis formulated under the National Labor Relations Act by the National Labor Relations Board's [NLRB] use of the "reasonable expectation of recall" test. The NLRB case law interpretation of that term, used in determining voter eligibility for representation elections, could then be utilized for the WARN Act. 54 Fed.Reg. 16,044.

The parties appear to agree with this suggestion by the Secretary and it, likewise, appears to this court to be an equally applicable phrase in determining those persons to be counted for WARN Act purposes.

Id. at 344. The Sixth Circuit affirmed, once again employing "the NLRB analogy". See Damron, 945 F.2d at 124-25.

The NLRA is more than just analogous to WARN; in fact, WARN may be thought of as an outgrowth of the NLRA, because Secs. 8(a) (1) and 8(a) (5) of the NLRA, 29 U.S.C. § 158(a) (1), (5), have been interpreted to require an employer to notify a union of its decision to close a plant.

In order to meet its obligation to bargain over the effects on employees of a decision to close, an employer must conduct bargaining in a meaningful manner and at a meaningful time. A concomitant element of meaningful bargaining is timely notice to the union of the decision to close, so that good faith bargaining does not become futile or impossible.

Penntech Papers, Inc. v. NLRB, 706 F.2d 18, 26 (1st Cir.) (citations and internal quotations omitted), cert. denied, 464 U.S. 892, 104 S. Ct. 237, 78 L. Ed. 2d 228 (1983); see also Metropolitan Teletronics Corp., 279 N.L.R.B. 957, 958-59 & n. 14 (1986) (finding NLRA Sec. 8(a) (5) violation because of company's failure to notify union of decision to close and relocate plant), enforced, 819 F.2d 1130 (2d Cir. 1987). In a sense, WARN amends the NLRA by setting a specific time period for notice, in addition to expanding coverage to all employees, regardless of union status.

Furthermore, as noted, collective bargaining agreements frequently require notice of layoffs, shutdowns, and the like. See First Nat'l Maintenance Corp. v. NLRB, 452 U.S. 666, 684, 101 S. Ct. 2573, 2583-84, 69 L. Ed. 2d 318 (1981) (describing that such provisions are "prevalent"); see also Dubuque Packing Co., 303 N.L.R.B. 386, 394 n. 23 (1991) (recognizing collective bargaining agreement that required "6 months' notice be given prior to closing"). Viewed from this perspective, WARN merely codifies a frequent practice facilitated by the NLRA. In fact, the regulations implementing WARN provide:

The provisions of WARN do not supersede any laws or collective bargaining agreements that provide for additional notice or additional rights and remedies. If such law or agreement provides for a longer notice period, WARN notice shall run concurrently with that additional notice period. Collective bargaining agreements may be used to clarify or amplify the terms and conditions of WARN, but may not reduce WARN rights.

20 C.F.R. Sec. 639.1(g) (1993). Thus, the Department of Labor's regulation injects WARN into collective bargaining agreements, both by providing that the WARN notice period automatically increases if provided for by the agreement, and by allowing WARN's "terms and conditions" to be clarified or amplified by the agreement.

These similarities notwithstanding, the Second Circuit held that "the NLRA is not sufficiently analogous to override the traditional assumption that a state limitations period should be applied." United Paperworkers, 999 F.2d at 55. In comparing the acts' purposes, it stated:

The purpose of WARN, unlike that of the NLRA, is not to ensure labor peace but to alleviate the distress associated with job loss for both the workers and the community in which they live. This is demonstrated by the provision in the statute of causes of action for local governments as well as for individual workers and unions. When a union brings an action, it thus serves only as the representative of the class of employees that has been harmed. The NLRA, in contrast, does not protect community interests in avoiding job loss.

Id. at 54; see also United Steelworkers, 32 F.3d at 58-59 (Third Circuit statement that "WARN serves very broad societal goals--to protect workers, their families and their communities in the wake of potentially harmful employment decisions."). Its rejection of the NLRA limitations period hinged on one essential perceived difference between WARN and the NLRA: the NLRA "specifically regulate [s] the collective bargaining relationship", while WARN "remain [s] peripheral to that concern." United Paperworkers, 999 F.2d at 55. It found that "WARN, therefore, neither 'encourages nor discourages' collective bargaining, thus differing in purpose from the NLRA." Id.; see also United Steelworkers, 32 F.3d at 58-59 (Third Circuit stating, at best, WARN has a "tangential" effect on collective bargaining).

As noted, we disagree. A strict identity of purposes is not a requisite for borrowing a limitations period from one federal statute for use in another. To the contrary, a federal statute's limitations period is to be borrowed if, among other things, it "provides a closer analogy than available state statutes". DelCostello, 462 U.S. at 171-72, 103 S. Ct. at 2294 (emphasis added); accord Lampf, 501 U.S. at 356, 111 S. Ct. at 2779 (plurality) ("affords a 'closer fit' with the cause of action at issue than does any available state-law source"); Agency Holding, 483 U.S. at 152, 107 S. Ct. at 2765 (relying, in part, on "similarities in purpose and structure"). If an identity of purposes were required, it would seem that the Court would not have borrowed the Clayton Act's limitations period in establishing a uniform period for RICO. See Agency Holding, 483 U.S. at 156, 107 S. Ct. at 2767.

The purposes of the NLRA are similar to those of WARN. Both regulate labor-management relations: WARN, by requiring notice of impending layoffs and terminations; the NLRA, by facilitating concerted action by employees and governing collective bargaining procedures. Both do so to achieve similar objectives. WARN's broad purpose is to protect "workers, their families and communities" by providing notice so that workers who will be terminated may seek other jobs or retraining. 20 C.F.R. Sec. 639.1(a) (1993). The NLRA's stated purpose, though broader, embraces similar objectives by "encouraging the practice and procedure of collective bargaining ... for the purposes of negotiating the terms and conditions of [workers'] employment or other mutual aid or protection." 29 U.S.C. § 151. As discussed supra, this purpose is achieved, in part, by the proscription of "unfair labor practices" by employers. WARN's purpose may be seen as a subset of the NLRA's: WARN provides a specific protection for workers by requiring that employers render advance notice of terminations or mass layoffs, while the NLRA generally defines and regulates the relationship between employers and employees to promote peaceful labor relations.17 

In sum, WARN and the NLRA share similar language; they undoubtedly overlap. In addition, they share similar, though by no means identical, purposes. The "fit" is reasonably close.18 

The question remains, however, whether a state period provides as close or closer "fit" to WARN than does the NLRA. Several Texas limitations periods19  have been suggested by appellants and amici curiae20  as proper analogies to WARN: Tex.Civ.Prac. & Rem.Code Secs. 16.003 (two-year tort statute), 16.004 (four-year breach of contract statute), and 16.051 (four-year residual statute). Conspicuously absent from these suggestions are persuasive discussions of why any of these statutes contain appropriate limitations periods for WARN. Halkias contends only that " [i]t is clear that the applicable state statute of limitation should be either" from the tort or contract statute. Likewise, Staudt asserts (in his brief) that either the tort or contract limitations period should be applied, although evidencing a preference for the former, on the basis that the failure to give a WARN notice "constitute [s] a taking or conversion of the employee's right to continued employment."21  And, the amici avoid the issue altogether, suggesting that all three periods are suitable candidates, but requesting a remand for the district court to select the closest one.22 

a.

As for the tort period, we disagree with Staudt's assertion that a WARN claim more closely resembles a claim for "conversion of an employee's right to continued employment" than it does an action under the NLRA. Texas is an employment at will State. E.g., Pease v. Pakhoed Corp., 980 F.2d 995, 1000 (5th Cir. 1993). But, even assuming that this is a tort in Texas, it presents a poor analogy to a WARN claim. WARN has nothing to do with a right to continued employment; it concerns only advance notice of an "employment loss". Regardless of whether notice is given, the employer is free to terminate or lay off the employee, or reduce his or her hours of work. Likewise, WARN's resemblance, if any, to any other traditional tort claim is nowhere near as obvious as its resemblance to the NLRA.23 

b.

The request that we apply Texas' residual statute, Sec. 16.051, may be quickly dismissed. As a general matter, the borrowing of a "catchall" period is not favored. See Agency Holding, 483 U.S. at 152-53, 107 S. Ct. at 2765; Wilson v. Garcia, 471 U.S. 261, 278, 105 S. Ct. 1938, 1948, 85 L. Ed. 2d 254 (1985). More specifically, it can hardly be said that Sec. 16.051, which applies to a tremendous diversity of claims, see Tex.Civ.Prac. & Rem.Code Ann. Sec. 16.051, Notes of Decisions (Vernon 1986), is as analogous, or more analogous, to WARN than the NLRA is. See also Haggerty, 952 F.2d at 786 (finding that a state's residual limitations period "offers no analogy [to the federal cause of action at issue], it is simply a fallback position").

c.

Finally, the breach of contract limitations period, Sec. 16.004, likewise fails to provide as close an analogy to WARN as does the NLRA.24  As noted, Texas is an employment at will State. After all, as discussed, no contractual right to continued employment is implicated by WARN.

In sum, canvassing the possible Texas periods reveals none as analogous to WARN as the NLRA.25  None seeks to accommodate the same, or very similar, interests as WARN, much less govern an action that will frequently overlap with WARN.26  And, needless to say, none embraces causes of action that share common language with WARN.27 

Borrowing the NLRA period is supported by litigation practicalities and policy considerations. See DelCostello, 462 U.S. at 172, 103 S. Ct. at 2294 (counseling use of more analogous federal limitations period "when the federal policies at stake and the practicalities of litigation make that rule a significantly more appropriate vehicle for interstitial lawmaking").

The first litigation practicality we consider is forum shopping. See Lampf, 501 U.S. at 357-58, 111 S. Ct. at 2779 (plurality); Agency Holding, 483 U.S. at 154, 107 S. Ct. at 2766 ("the use of state statutes would present the danger of forum shopping"). Raising this concern is the breadth of WARN's venue provision, which permits an action to be brought not only in "any district in which the violation is alleged to have occurred", but also, in any district "in which the employer transacts business ". 29 U.S.C. § 2104(a) (5) (emphasis added). The Second and Third Circuits were not as troubled as we by this generous venue provision, and the concomitant possibility that a party could manipulate it through forum shopping. The Second Circuit began its analysis of this issue by quoting the United Paperworkers district court:

The term "plant closing" as defined by the Act is limited to single sites of employment, and venue is limited to the district where the violation is alleged to have occurred or where the employer does business; unless a single plant site straddles the boundary between two states, it is unlikely prospective plaintiffs will have a broad choice of fora in which to bring their claims or that doubt will arise as to in which state triggering events occurred. Therefore, geographic considerations do not counsel for the application of a uniform federal limitations period for WARN Act claims.

United Paperworkers, 999 F.2d at 56 (citing and quoting district court); see also United Steelworkers, 32 F.3d at 60-61 (Third Circuit relying upon the above United Paperworkers quote to dismiss forum shopping concerns). After endorsing the assertion that plaintiffs would not have "a broad choice of fora"--a conclusion somewhat at odds, it seems, with the judicially-noticeable fact that many businesses with 100 or more employees28  "transact business" in more than one state--the Second Circuit then quoted another district court for the proposition that

choice of law rules would likely point to borrowing the law of the site since that would be the place of the injury, and probably the place of the unlawful action as well. Therefore, it is unlikely that a WARN plaintiff would be able to forum shop for a locale with the most advantageous state of limitations; the law of the site of the layoff would likely be chosen no matter where the suit happened to be filed.

United Paperworkers, 999 F.2d at 56 n. 9 (citation omitted; emphasis added). Choice of law rules appear, however, to pose greater problems than this.

A WARN action is, of course, a federal question case. The choice of laws issue that would be presented to a forum that does not include the site of the WARN violation would be which state's limitations period should be borrowed, and the outcome of that inquiry seems by no means certain. A district court's choice of law principles in a federal question case are derived from federal common law, as the Second and Third Circuits have recognized. Corporacion Venezolana de Fomento v. Vintero Sales Corp., 629 F.2d 786, 795 (2d Cir. 1980) ("This is a federal question case ... and it is appropriate that we apply a federal common law choice of law rule ..."), cert. denied, 449 U.S. 1080, 101 S. Ct. 863, 66 L. Ed. 2d 804 (1981); see Gluck v. Unisys Corp., 960 F.2d 1168, 1179 n. 8 (3d Cir. 1992) ("A state court or legislature does not necessarily seek to further or even consider federal laws when it develops its choice of law provisions. A federal choice of law rule would address those concerns."). If the district court follows the forum state's choice of law principles as a surrogate for federal common law, statutes of limitations are viewed by many jurisdictions as "procedural"; and thus a forum will follow its own period, regardless of the period that would apply in the state in which the cause of action arose. See generally Eugene F. Scoles & Peter Hay, Conflict of Laws Secs. 3.9(b), 3.10, 3.11 at 58-64 (2d ed. 1992).29  This result, of course, would be completely different from that suggested by the Second Circuit.30 

Moreover, examining borrowing cases involving a conflict over which state's limitation period should apply, we find a consistent preference for borrowing the forum state's. See Wang Lab., Inc. v. Kagan, 990 F.2d 1126, 1128 (9th Cir. 1993) ("In an ERISA case, we ordinarily borrow the forum state's statute of limitations....") (emphasis added); Gluck, 960 F.2d at 1179-80 (after reviewing a conflict regarding which state's statute of limitations should be borrowed for an ERISA claim, court decides to "follow the general rule and borrow a limitations period applicable to the forum state claim most analogous to the ERISA claim ...") (emphasis added); Champion Intern. Corp. v. United Paperworkers, 779 F.2d 328, 332-33 (6th Cir. 1985) (refusing to apply forum state's borrowing statute; applying forum state's limitations period to federal cause of action).

Augmenting the dangers that forum shopping may occur, and highlighting the complexity of a federal court's decision as to which limitations period should be borrowed, WARN claims could involve several different states by virtue of the maintenance of a class action. For instance, the district court certified Halkias' class action, allowing him to press WARN claims from facilities in both Texas and Oklahoma. Moreover, as discussed supra, there was another potential forum, Missouri.31  Given the possibility for multistate WARN litigation, the issue of which state's limitations period should control might become an issue of time-consuming litigation. And, it goes without saying, courts do not have time to waste.

Considering, collectively, the likelihood of forum shopping, and the possibility for multistate WARN class actions, the question arises: Should federal courts even address themselves to the vexatious question of which state's period should govern a WARN claim, when the NLRA provides a uniform period? We think not. If nothing else, valuable time and resources should not be consumed litigating such issues. See Short, 908 F.2d at 1389 (discussing pre-Lampf limitations period for securities fraud actions).32  Accordingly, we conclude that a uniform, single limitations period is desirable for WARN claims, a conclusion militating in favor of applying the NLRA's.

This conclusion is buttressed by resort to an examination of WARN's policies, and the likely effect on those policies of applying diverse, and often quite long, limitations periods. See DelCostello, 462 U.S. at 172, 103 S. Ct. at 2294-95 (urging use of more analogous federal limitations period "when the federal policies at stake ... make that rule a significantly more appropriate vehicle for interstitial lawmaking").

First, we recognize that federal labor policy has long favored the rapid settlement of disputes between an employer and an employee. In applying the NLRA's period to Sec. 301 of the Labor Management Relations Act, 29 U.S.C. § 185 (1988), the Seventh Circuit reasoned that " [t]he six-month limitation period will encourage prompt resolution of labor disputes." United Food & Commercial Workers Local 100A v. Hofmeister, 950 F.2d 1340, 1348 (7th Cir. 1991). Appellants urge that this rationale is inapplicable to WARN, contending that although promptness is of unique concern to the collective bargaining process, it is not applicable to WARN's notice requirement.

We disagree. As the Seventh Circuit explained in rejecting a similar argument against the application of the NLRA's period to " 'pure' section 301 actions":33  "The six-month limitations period in section 10(b) was adopted in order to 'bar litigation over past events "after records have been destroyed, witnesses have gone elsewhere, and recollection of the events in question have become dim and confused." ' " Johnson v. Graphic Communications Int'l Union Local 303, 930 F.2d 1178, 1182 (7th Cir.) (citing, inter alia, Local Lodge No. 1424 v. NLRB, 362 U.S. 411, 419, 80 S. Ct. 822, 828, 4 L. Ed. 2d 832 (1960)), cert. denied, --- U.S. ----, 112 S. Ct. 184, 116 L. Ed. 2d 145 (1991).34  This concern is no less present in a WARN claim; indeed, given that WARN is often triggered by plant closings or relocations, the need for prompt litigation is even greater. Allowing a party several years in which to bring a WARN claim could create untold problems concerning availability of evidence.35 

Moreover, as discussed supra, failure to give WARN notice often may create a concurrent claim under the NLRA. When a claim under one federal statute is also an "unfair labor practice under the NLRA, it seems particularly appropriate to borrow the NLRA limitations period." Hofmeister, 950 F.2d at 1348.

Finally, and most important, application of expansive limitations periods would disserve WARN's most specific objective: the provision of a cushion of time for employees to explore other job opportunities and, if necessary, seek retraining. See 20 C.F.R. Sec. 639.1(a) (1993) ("Advance notice provides workers ... some transition time to adjust to the prospective loss of employment, to seek and obtain alternative jobs and, if necessary, to enter skill training or retraining...."). Obviously, providing funds to workers several years after their termination does not serve that objective. Cf. Lloyd v. Department of Labor, 637 F.2d 1267, 1270 (9th Cir. 1980) ("In order to serve the [Trade] Act's purposes of retraining, adjustment, and relocation, it was important that workers claim and receive benefits promptly after discharge. There were also other advantages--such as freshness of records and other evidence--to be gained by promptness.") (citations omitted).36 D.

Halkias and the amici urge that if we apply a federal limitations period, it should be the four-year federal residual period, 28 U.S.C. § 1658. But, by its terms, it applies only to "a civil action arising under an Act of Congress enacted after the date of the enactment of this section". Id. Section 1658 was enacted on December 1, 1990, well after WARN's enactment. Compare Pub. L. No. 101-650, 104 Stat. 5089, 5114-15 (1990) (Sec. 1658 enactment) with Pub. L. No. 100-379, 102 Stat. 890 (1988) (WARN enacted on August 4, 1988).37 

Despite Sec. 1658's plain language, amici urge that we borrow it, if we are to borrow federal law. Specifically, they contend that Sec. 1658's legislative history indicates that Congress refused to give it retroactive effect only because doing so would undermine the settled expectations of parties in those instances where the courts have already settled the limitations period issue. Reasoning that WARN's limitations period has not been so settled, they contend that we should borrow Sec. 1658 despite its plain language to the contrary.

We refuse to reject that plain language. Moreover, we cannot see how the federal residual statute presents a closer analogy to a WARN claim than does the NLRA's six-month period. Also, the application of Sec. 1658 would undermine WARN's central purpose, by discouraging the prompt resolution of WARN claims. For these reasons, traditional borrowing principles do not favor application of Sec. 1658 to WARN claims.

Halkias' last contention is that Congress violated the Fifth Amendment by failing "to prescribe a limitation period for a WARN Act violation", thereby requiring a plaintiff to " 'guess' at which statute the federal courts will 'borrow' ".38  We assume Halkias does not wish to invalidate WARN itself, inasmuch as that statute provides the basis for his requested relief. Rather, we assume that he wants to be excepted from the application of an unforeseen limitations period. We see no need for such an exception.

Generally, Fifth Amendment due process is not affected by the traditional practice of borrowing limitations periods. Cf. Agency Holding, 483 U.S. at 157-65, 107 S. Ct. at 2767-72 (Scalia, J., concurring in judgment) (tracing history of "borrowing" limitations periods). Indeed, the borrowing of limitations periods is a normal component of statutory interpretation, see DelCostello, 462 U.S. at 158-62, 103 S. Ct. at 2287-89; and a borrowed federal limitations period has been applied by the Supreme Court to the litigants before it. See Lampf, 501 U.S. at 363-64, 111 S. Ct. at 2782; see also Lampf, 501 U.S. at 370-71, 111 S. Ct. at 2786 (O'Connor, J., dissenting) (objecting to Court's application of newly-announced limitation period to "the very case in which it announced the new rule").

This is not to say that we do not sympathize with Halkias' underlying concern; we do. Indeed, the rationale of this contention illustrates the necessity for a uniform limitations period for WARN claims, a result we herald by adopting the NLRA's six-month period.

For the foregoing reasons, the judgments of the district courts are

AFFIRMED.

WISDOM, Circuit Judge, dissenting.

This is a difficult case. Reasonable minds can disagree on the proper outcome. To my mind, however, the United States Supreme Court has provided a framework whose strictures dictate the result. Because I find the majority's diligent attempt to escape that outcome unconvincing, I respectfully dissent.

The issue presented, as correctly framed by the majority, is whether the district court erred in applying the six-month statute of limitations from Sec. 10(b) of the National Labor Relations Act (NLRA)1  to the plaintiffs' claims under the Worker Adjustment and Retraining Notification Act (WARN).2  This is not only a question of first impression in this Circuit, we are the only Court of Appeals to consider applying the NLRA's statute of limitations to non-unionized WARN plaintiffs.3  The majority reluctantly parts ways with the Second and Third Circuits on this question and holds that the NLRA six-month limitations applies to WARN.4  I believe that our sister circuits got it right.

I agree with the majority that this case presents a "classic example" of "which round peg to stuff in a square hole". All three consolidated appeals focus on the same task: We must fill in a blank left by Congress, namely, what statute of limitations applies to cases brought under the WARN? Whenever courts are left to a task of this nature, they become part sleuth, part improvisor. The truth is, we do not know what Congress would have preferred and, so, are left with our best guess as to what would most closely approach congressional intent--if it had an intent on the subject.

The starting point in resolving the present issue is the recent Supreme Court decision in Lampf, Pleva, Lipkind, Prupis & Petigrow v. Gilbertson.5  In that case, the Court held that, when Congress creates a federal cause of action but does not include a statute of limitations, the courts are to presume that Congress intended for the analogous state statute of limitations to apply.6  The Supreme Court summarized this general rule as follows:

It is the usual rule that when Congress has failed to provide a statute of limitations for a federal cause of action, a court "borrows" or "absorbs" the local time limitation most analogous to the case at hand....

This rule is founded on the Rules of Decision Act which "has enjoyed sufficient longevity that we may assume that, in enacting remedial legislation, Congress ordinarily intends by its silence that we borrow state law."7 

This general rule is subject to a limited exception. If a state statute of limitations would be "at odds with the purpose or operation of federal substantive law", the courts should ignore the state limitations period and instead borrow the most analogous federal limitations period.8  This exception, however, remains that; it is "a closely circumscribed exception" to the general rule that state, not federal, limitations periods apply when Congress fails to specify a limitations period for a federal cause of action.9 

Moreover, it is important to underscore that this exception applies only when the state statute of limitations would "frustrate" the policies of WARN.10  This approach to the exception makes sense for, as the DelCostello Court explained, it would be "inappropriate to conclude that Congress would choose to adopt state rules at odds with the purpose or operation of federal substantive law".11  Unfortunately, the majority's decision deviates dramatically from this established framework.

The language in Lampf pronouncing the state borrowing rule alive and well did not command "only a plurality", as the majority and General Dynamics would have us believe.12  In fact, a simple head count proves the opposite. Four Justices joined in part II.A of the Court's opinion from which the quoted rule comes. The remaining Justices took positions even more hostile to General Dynamics's position than Justice Blackmun's plurality opinion.13  The Court actually was unanimous in Lampf in expressing the view that state statutes of limitation generally govern if Congress attaches no limitations period to a federal cause of action.

Lampf and DelCostello create a flow chart inquiry. We always look to the state statute of limitations first to see if an analogous state law exists.14  The majority's explanation of the Lampf "hierarchy," however, places the choice between a federal and state statute on par with one another, both subject to the criteria of which provides the best fit in the light of forum shopping concerns and the geographic character of the claim.

That is not Lampf 's teaching. The hierarchy is vertical; Lampf is clear that state statutes of limitations are the rule, federal rules the exception. Courts do not make a simple choice between the federal and state rules. If an analogous state statute of limitations is found, it should be used, provided it is not at odds with the federal substantive law.15  This approach embodies the "presumption of state borrowing".16  In fact, Lampf suggests that the only time a court can go straight to federal law is "where Congress has provided an express limitations period for correlative remedies within the same enactment".17  While that was precisely the case in Lampf, it has nothing to do with the case at hand.

The majority's inverted approach is evidenced by its starting point: "We first examine whether the NLRA limitations period is more analogous to WARN than available state periods".18  That is not the proper first step and dangerously misplaces the proper emphasis. Instead, Lampf requires that we first select the most analogous state period and determine whether it is at odds with the federal substantive law. Hence, the majority's next statement that it will "describe the similarities between the NLRA and WARN, and then compare and contrast those similarities with the available state period(s)" is a task improperly undertaken.19 

Even taken on its own terms, however, the majority's opinion cannot stand. The district courts relied on the mistaken impression that WARN causes of action are analogous to lawsuits brought under the NLRA and, accordingly, within the exception laid down in DelCostello. The majority reiterates that assertion by painting the NLRA and WARN as statutes with similar purposes and structures. I disagree.

This case boils down to the length of time in which a plaintiff may file suit. The main policy reason advanced by the district courts for the restrictive six-month limit was the national interest in the quick resolution of labor disputes.20  Unlike the NLRA, however, nothing in the cause of action created by WARN requires or counsels in favor of an accelerated resolution. Without explanation, the majority adopts the shaky premise that the policy favoring the rapid resolution of labor disputes (presumably so that everyone can get back to business) applies to a situation where the company has closed a plant (and there will be no more business).21 

This is where the majority's statement that the statutes bear a "family resemblance" breaks down. Even a cursory analysis reveals the purposes of these statutes to be markedly different. The reason for preferring rapid resolution of an ordinary labor dispute is that when the dispute is resolved, employees will go back to work, production will resume, the employer's sales will increase, and the positive effects of the business on the overall economy will return. Because strikes disrupt the status quo and hurt the economy, it is important to minimize the economic damage to the company and the community by ending them swiftly. Understandably, the NLRA sets a comparatively short limitations period (six months) as a means of achieving these goals.

Those considerations, unfortunately, have no place in a scenario where WARN comes into play. When a WARN dispute is resolved, the plant stays closed. Employees fired en masse stay fired. Production does not resume. The economy does not return to the status quo ante. Plant closings hurt the economy, but once the closing occurs, the damage is done. Resolution of any resulting WARN Act claims will not remedy that harm. In sum, there is no reason to require that they be rushed to resolution within six months or not at all.

The Second Circuit recognized this fundamental distinction: "The purpose of WARN, unlike that of the NLRA, is not to ensure labor peace but to alleviate the distress associated with job loss for both the workers and the community in which they live".22  No reason to expedite those claims beyond what the state prescribes for the most analogous state limitations period exists. It is important to understand that this alleviation of distress comes from the advance notice of the plant closing, not from the backpay award that comes when a company fails to comply with WARN's provisions.23  Moreover, we must remember that a plaintiff is not paid the day he files suit. If he wins at trial and upon appeal, his backpay award could be years away. The majority's decision will compound--not dissipate--the distress that thousands of workers experience annually.

The most severe aspect of the majority's holding is its decision to subject the non-unionized employees in this case to the strictures of the NLRA, the act governing collective bargaining.24  Returning to the round peg, square hole metaphor, this truly baffles the laws of geometry. The NLRA protects the right to bargain collectively but exacts as a price for that protection that disputed cases be filed within six months. Yet, in this case, the majority holds the plaintiffs to a deal they did not make: They suffer the six-month requirement but benefit from no corresponding protection of their rights.25 

The Supreme Court has drawn an important distinction between statutes that involve the collective bargaining process and those that do not.26  For example, in DelCostello, the Supreme Court stated that it would borrow a federal limitations period because of the national concern for "stable bargaining relationships and finality of private settlements".27  7] The Reed case is equally illustrative. The statute at issue there did not directly touch on collective bargaining concerns and so the Supreme Court applied the usual state limitations period.28  This distinction is well-founded, for a plant closing law "provides protection to individual union and non-union workers alike, and thus neither encourages nor discourages the collective bargaining processes that are the subject of the NLRA".29 

The unfairness of applying the NLRA's limitations period to non-unionized workers is magnified by the practical obstacles that WARN plaintiffs face. Unionized plaintiffs presumably enjoy the benefit of union protection and, probably, union legal representation. The union will be well-versed in the applicable statute of limitations for claims of that sort and the plaintiffs will be advised accordingly. In other words, unionized WARN plaintiffs can rest assured that the union "will handle it".

Non-unionized WARN plaintiffs, on the other hand, are unlikely to have sufficient information to be able to bring suit within six months.30  Whereas unions are readily able to determine whether a sufficient number of their members have lost their jobs to invoke WARN's remedies, non-union employees must learn about the scope of a layoff by talking to co-workers and gradually compiling a list of those affected. This will be a time-consuming laborious undertaking.31 

But workers will hardly be able to direct the energy and attention necessary to that task--they will be out hunting for work. NLRA plaintiffs are preoccupied with going back to their jobs with their company; WARN plaintiffs are preoccupied with finding a new job to replace the one they just lost. The strain they are under will be intense as they will be competing with the perhaps thousands of others similarly situated for any scarce employment opportunities. In spite of this all-too-real scenario, the decision of the Court today requires that they file suit within six months to get the backpay to which they are entitled, knowing that the ultimate award if they win will be far away and ineffective to help with their present needs.

Although the equities seem to me plain, this distinction goes beyond a question of fairness, it is the core of our analysis. For when the purpose of the NLRA is at odds with the purpose of WARN, there is no way that the NLRA can provide a "closer fit" than an analogous state statute of limitations.

Not only is there no strict identity of purposes as the majority reminds us is unnecessary, these are fundamentally different statutes, enacted for very different reasons, creating very different causes of action.32  The purpose of WARN is to give employees advance notice of plant closings and a cause of action when the employer fails to comply. The imposition of the NLRA statute of limitations is at odds with that purpose in that it unduly burdens potential WARN plaintiffs by making them pursue those actions within a very tight time frame.

Comparing these statutes to the securities law at issue in Lampf, where the Court did adopt a uniform federal limitations period, is instructive. In Lampf, the Court was able to borrow the federal limitations period from the remedial provisions passed by the same Congress for the same purpose. Hence, the Court's task was, in many ways, easier. Both the borrowed period and Rule 10(b) were "to protect investors against manipulation of stock prices through regulation of transactions upon securities exchanges...." In Lampf, the Court's borrowed period had a near-identity of purpose with Rule 10(b); we do not.

The majority opinion is no more convincing as to structure. The majority incants the WARN regulations that the Department of Labor borrowed from the NLRA, but fails to explain the significance of that comparison. The borrowing of concepts is understandable: both laws operate within the larger context of employer-employee relations. The Department of Labor is the natural oversight agency for both and the case law applicable to the interpretation of one will likely be useful in construing the other. Still, that they were born into the same extended family does not mean that they bear a family resemblance.33 

In conclusion, I would join the other Courts of Appeals to have considered this issue--the Second Circuit in United Paperworkers and the Third Circuit in Crown Cork & Seal--and hold that WARN lawsuits should be governed by state limitations periods, as suggested by a harmonious reading of Lampf and DelCostello.34  I, like those courts, believe that " [a] WARN cause of action does not fit the limited circumstances under which a federal statute of limitation should be applied".35 

The next step is to decide which state limitations period ought to be borrowed. Although several Texas statutes have been suggested, it is my feeling that the Texas statute of limitations for contract claims provides the best analogy.36  Under Tex.Civ.Prac. & Rem.Code Ann. Sec. 16.004, then, WARN plaintiffs would have four years in which to institute their actions.

As all seem to agree, the fit will never be perfect; that is why it is a question of which round peg to stuff in a square hole. The Texas limitations period for contract claims works in that it is the most analogous state statute.37  I view an action under WARN as

essentially an action for damages caused by an alleged breach of an employer's obligation.... Such an action closely resembles an action for breach of contract cognizable at common law.38 

The Second Circuit in United Paperworkers agreed and applied Vermont's six-year residual limitations period for all contract claims.39  Even acknowledging that Texas is an employment-at-will state, the backpay provisions of WARN resemble damages for a breach of implied contract. Simply put, I perceive no evidence that the Texas contract claim limitations period is at odds with WARN's substantive provisions.40 

Following the proper framework, we look last to whether litigation practicalities and policy considerations make the use of the NLRA limitations period "a significantly more appropriate vehicle for interstitial lawmaking".41  The majority, raising the specter of endless forum shopping among WARN litigants, concludes that they do. I come out the other way.

To start, the forum shopping concerns raised by the majority could be raised in reference to almost any federal law for which we were considering borrowing a state limitations period. If anything, WARN diminishes the threat of forum shopping.42  Any threat of forum shopping that results from the right to file suit wherever the employer does business is stemmed by choice of law rules which dictate that the law of the site of the layoff (and, thus, injury and violation) would be chosen no matter where the suit was filed.43 

We also examine the policy considerations to determine whether WARN calls for a particular degree of uniformity. To me, they do not. The Supreme Court in Auto Workers v. Hoosier Cardinal Corp.44  instructed that the value of uniformity is far greater under the NLRA than under non-collective bargaining statutes. Even if we concede that the subject matter of WARN is "peculiarly one that calls for uniform law," national uniformity is less important because WARN does not involve

those consensual processes that federal labor law is chiefly designed to promote--the formation of the collective agreement and the private settlement of disputes under it.45 

WARN is decidedly not about collective bargaining. It is about what happens when there will be no more employer-employee relationship and, thus, nothing left to bargain for.

The majority's other policy considerations are undercut by experience and common sense. For example, the majority states that the prompt resolution of labor disputes militates in favor of a six-month limitations period. Once a plant has closed, however, whether notice has been given or not, there can be no resolution of a labor dispute. The plant is gone, workers are without jobs, the community is left reeling. Although the majority is appropriately concerned with the availability of witnesses and evidence down the road,46  the state legislatures measure those concerns against other public policy considerations when they enact their statutes of limitations.

The majority also asserts that an expansive limitations period would disserve WARN's objective of providing a time cushion in which to seek other employment options. The majority states that providing funds to workers several years later does not serve that objective.47  The majority has confused time cushions.

The time in which to file suit is not the time cushion Congress sought to give workers; the time cushion is the advance notice that the workers would be losing their jobs. The workers could then prepare for the coming changes while still collecting their paychecks. Congress hoped that notice would be given (the time cushion) and that these causes of action would never accrue. Once they have accrued, however, the restrictive six-month limitations period serves to take away any remedial cushion that workers might get in compensation for their injury.

Worse, the six-month time frame will actually serve to stymie WARN's true objective of giving workers and communities advance notice of impending hard times. By severely limiting the time frame in which potential WARN plaintiffs may file suit, companies can relax a bit and rest assured that they may "make redundant" many legally unsophisticated and unsuspecting workers--particularly non-unionized workers--without suffering the community backlash that will follow an announcement and, better, without the threat of future litigation. An extended limitations period, on the other hand, would function as a deterrent and, thus, an enforcement mechanism.

The majority gives a detailed explication of the trend away from borrowing state statutes of limitations in favor of uniform national rules. Even if that trend does exist--and I do not concede that it does--Lampf explicitly instructs that borrowing state limitations period is still the law. I do not doubt that uniform federal limitations periods might promote certainty, predictability, and minimization, but Congress is well aware of those values and yet frequently (as in this case) chooses to ignore them and leave the question open. Hence, if the state borrowing rule is truly becoming an anachronism, I will look to Congress, not to a law review, for evidence of its decline.48 

The plaintiffs, among countless others, have suffered the indignity of losing their employment without notice--in violation of federal law.49  In seeking the redress to which Congress has made them entitled, this Court has closed the gate on them one last time, on a legal principle so tenuous in foundation it appears as but an academic exercise. Unfortunately for John Halkias, John Cureington, and Alvin Straudt and thousands of other workers similarly situated, it is anything but that.

Because I believe that the Court's decision today promotes expediency and uniformity at the expense of the rights of workers, I dissent.

Before POLITZ, Chief Judge, KING, GARWOOD, JOLLY, HIGGINBOTHAM, DAVIS, JONES, SMITH, DUHE, WIENER, BARKSDALE, EMILIO M. GARZA, DeMOSS, BENAVIDES, STEWART and PARKER, Circuit Judges.

Sept. 22, 1994

BY THE COURT:

A majority of the Judges in active service, on the Court's own motion, having determined to have these cases reheard en banc,

IT IS ORDERED that these causes shall be reheard by the Court en banc with oral argument on a date hereafter to be fixed. The Clerk will specify a briefing schedule for the filing of supplemental briefs.

 1

Although WARN is referred to as a "plant closing" law, it is not addressed solely to the permanent closing of plants. First, it addresses temporary closings. See 29 U.S.C. §§ 2101(a) (2), 2102(a). Second, and more important, it addresses "mass layoffs" that result from far less than plant closure. See 29 U.S.C. §§ 2101(a) (3), 2102(a); see also infra, note 2. Nevertheless, the dissent treats WARN as if it governed only plant closings, and this treatment drives its conclusions. See infra, note 18

 2

WARN generally defines a "mass layoff" as a reduction in force that is not the result of a plant closing and results in either an employment loss at a single site for at least 33 percent of the employees (provided that at least 50 employees suffer an employment loss), or 500 employees. See 29 U.S.C. § 2101(a) (3)

 3

According to General Dynamics, different plaintiffs commenced a WARN action in the Eastern District of Missouri within 10 days of the January 8, 1991, layoff. Halkias attempted to intervene in that action on May 4, 1992; however, that July, his motion to intervene was denied as untimely. Although neither the motion nor the order is part of the record, Halkias does not dispute General Dynamics' statement. Halkias instituted his action five months after the denial of his motion to intervene in the Missouri action (which was filed more than one year after the layoff)

 4

The district court certified the class as:

Each person (i) who has been an employee of General Dynamics Corporation (ii) who, at the time of the termination of his or her employment for General Dynamics Corporation, was not represented by a union, (iii) who, at that time was employed at either the Fort Worth, Texas, plant or the Tulsa, Oklahoma, plant of General Dynamics Corporation, (iv) whose employment was involuntarily terminated between the dates January 7, 1991, and March 1, 1991, and (v) who did not receive written notice of his or her termination of employment at least sixty (60) days prior to such termination.

 5

Cureington filed suit in state court on January 4, 1993, nearly two months after Halkias filed his. General Dynamics removed the case to district court

 6

Staudt states in his brief that the employees were non-union, and advances this as one of the reasons for not adopting the NLRA period. The record, however, is silent on this non-union claim, but Glastron does not dispute it. As discussed infra, we give no weight to union status vel non for fixing the appropriate limitations period

 7

At various places in his briefs, Halkias seems to question the propriety of utilizing Fed. R. Civ. P. 12(c) (allowing entry of a "judgment on the pleadings") as a vehicle for dismissing his complaint; however, he never explicitly contends that a dismissal under Rule 12(c) was inappropriate. Therefore, we do not consider this issue to have been raised

 8

E.g., Newspaper & Mail Deliverers' Union v. United Magazine Co., 809 F. Supp. 185, 188-92 (E.D.N.Y. 1992); Thomas v. North Star Steel Co., 838 F. Supp. 970, 972-75 (M.D. Pa. 1993), rev'd sub nom. United Steelworkers of Am. v. Crown Cork & Seal Co., 32 F.3d 53 (3d Cir. 1994)

 9

E.g., United Steelworkers of Am. v. Crown Cork & Seal Co., 833 F. Supp. 467, 467-70 (E.D. Pa. 1993), aff'd, 32 F.3d 53; Automobile Mechanics' Local No. 701 v. Santa Fe Terminal Serv., Inc., 830 F. Supp. 432, 435-37 (N.D. Ill. 1993); Wholesale & Retail Food Distrib. Local 63 v. Santa Fe Terminal Serv., Inc., 826 F. Supp. 326, 329-31 (C.D. Cal. 1993) (applying California limitations period after determining that defendants waived limitations defense pursuant to NLRA); Frymire v. Ampex Corp., 821 F. Supp. 651, 653-55 (D. Colo. 1993); Wallace v. Detroit Coke Corp., 818 F. Supp. 192, 194-97 (E.D. Mich. 1993)

 10

This usual rule has its genesis in the Rules of Decision Act, 28 U.S.C. § 1652, originally enacted in 1789, which provides:

The laws of the several states, except where the Constitution or treaties of the United States or Acts of Congress otherwise require or provide, shall be regarded as rules of decision in civil actions in the courts of the United States, in cases where they apply.

See Lampf, 501 U.S. at 355-56, 111 S. Ct. at 2778 (plurality) (discussing source of state borrowing principle); see also DelCostello, 462 U.S. at 159 n. 13, 103 S. Ct. at 2288 n. 13:

As we recognized in [Auto Workers v.] Hoosier [Cardinal Corp.], 383 U.S. [696,] 701 [86 S. Ct. 1107, 1110-11, 16 L. Ed. 2d 192 (1966) ], the choice of a limitations period for a federal cause of action is itself a question of federal law. If the answer to that question (based on the policies and requirements of the underlying cause of action) is that a timeliness rule drawn from elsewhere in federal law should be applied, then the Rules of Decision Act is inapplicable by its own terms.

 11

The "norm" of borrowing state limitations periods has been subject to steady erosion. Discussing Lampf, one commentator noted that "the Court furthered a decade-long drift away from the traditional practice of 'borrowing' the statute of limitations of the forum state's most analogous cause of action and moved towards the adoption of uniform national rules." The Supreme Court, 1990 Term--Leading Cases, 105 Harv. L. Rev. 177, 400 (1991). In many respects, this erosion reflects the endorsement of a view once held only in dissent. See Hoosier Cardinal, 383 U.S. at 709-14, 86 S. Ct. at 1115-17 (White, J., dissenting). For obvious reasons, this movement towards uniform, national limitations periods for federal causes of action has much to recommend it. See Leading Cases, supra, at 409 ("The policies advanced by adopting uniform limitations periods--certainty, predictability, and minimization of litigation--would be promoted by uniform periods for all federal statutes. In an age of multistate activity largely regulated by federal rather than state law, the state borrowing rule may be an anachronism.")

The dissent contends that the Leading Cases note "in actuality supports [its] conclusion". Dissent at page 248 n. 48. If its conclusion is that the state borrowing rule is still the usual rule, then we agree. On the other hand, if it believes that the note supports the dissent's conclusion that the existence of a "trend away from the state borrowing rule is pure conjecture", then we disagree.

 12

The dissent charges that " [t]he language in Lampf pronouncing the state borrowing rule alive and well did not command 'only a plurality', as the majority and General Dynamics would have us believe." Dissent at page 242 (footnote omitted). That charge is predicated on a misconstruction; we merely state that "the Lampf hierarchy commanded only a plurality". At no point do we suggest that "only a plurality" of the Court believes that state borrowing rule is anything but the "usual" rule; however, part IIA of the Lampf opinion was, in fact, joined by only four justices. And, only those four justices expressly agreed with the "hierarchical inquiry" distilled in part IIA for determining whether a federal limitations period should be borrowed; no other justice wrote to express a similar view. In any event, as discussed infra, we approach our inquiry through a harmonious reading of DelCostello, Lampf, and other Supreme Court precedent

 13

The dissent takes vigorous exception to the order of our presentation, believing that Lampf mandates that we first address the state limitations period. Dissent at page 242. As to Lampf 's teachings, we are unable to distill the same meaning from the Court's words that the dissent does. Nowhere does Lampf state that federal courts, when confronted with the suggestion that a federal limitations period is more appropriate than a state period, must first address the relevant state period. Instead, Lampf, quoting DelCostello, invites a comparison: federal borrowing is "to be made only when a rule from elsewhere in federal law clearly provides a closer analogy ..., and when the federal policies at stake and the practicalities of litigation make that rule a significantly more appropriate vehicle for interstitial lawmaking." Lampf, 501 U.S. at 356, 111 S. Ct. at 2778 (plurality) (emphasis added; citations and internal quotation marks omitted). Obviously, this comparison involves ascertaining the closeness of both the proposed federal and state periods to the cause of action in issue, and then comparing their fit. Whether one first looks to the state or to the federal period, the resulting inquiry is still the same. Of interest, the Lampf opinion itself did not even consider state law alternatives, because it found "an express limitations period for correlative remedies within the same [federal] enactment." Id. at 362, 111 S. Ct. at 2782 (majority opinion) (footnote omitted). We do not read this to mean that "Lampf suggests that the only time a court can go straight to federal law is 'where Congress has provided an express limitations period for correlative remedies within the same enactment.' " Dissent at page 243 (footnote omitted). In fact, the Supreme Court in Agency Holding addressed the similarities between RICO and the Clayton Act prior to discussing "the lack of any satisfactory state law analogue to RICO" and rejecting the application of a state "catchall" limitations period. See Agency Holding, 483 U.S. at 146-53, 107 S. Ct. at 2762-65

 14

In Carpenters, our court disagreed with the Third Circuit on an aspect of WARN's damages provision. Id. at 1282-86 (disagreeing with Third Circuit's holding that WARN's remedial provision requires "back pay" for each calendar day of violation; calculating damages based only on work days within violation period)

 15

It is significant that the Department of Labor administers WARN. In holding that a claim under the Employee Protection Program of the Airline Deregulation Act should be subject to the NLRA's period, the Third Circuit noted that, "as with the NLRA, the Department of Labor has had a role in administering the EPP". Haggerty v. USAir, Inc., 952 F.2d 781, 787 (3d Cir. 1992)

 16

Although the comments by NLRB General Counsel were far-ranging, the following is illustrative:

I think the two statutes are intended by Congress to operate essentially separately, but I think because of the borrowing of concepts, one from the other, and because of the impact on existing and future collective bargaining relationships that this statute requires ... it is inevitable that there will be a fair amount of interplay between the [two laws].

Id. (ellipses and brackets in original).

 17

The dissent finds this comparison of purposes "weak". Dissent at page 245 n. 32. We think this discussion, along with the other points made in this section (several of which the dissent does not address), to be far "more convincing than the fact that both were passed by Congress." Dissent at page 245 n. 32. This becomes all the more apparent when one considers, as discussed, that the NLRA proscribes conduct almost identical to that proscribed by WARN

 18

The dissent dismisses the relationship between the NLRA and WARN, and does so for essentially one reason: its disagreement with our "shaky premise that the policy favoring the rapid resolution of labor disputes (presumably so that everyone can get back to business) applies to a situation where the company has closed a plant (and there will be no more business)." Dissent at page 243. The dissent returns to this point again and again; a particularly colorful reprise follows: "Once a plant has closed ... there can be no resolution of a labor dispute. The plant is gone, workers are without jobs, the community is left reeling." Dissent at pages 247-48

There are several reasons why the policies of WARN would be better served by application of the NLRA's six-month limitations period, and these reasons are discussed infra, part II.C. For now, four points will suffice. First, as noted, WARN is not merely a plant closing law; it also governs "mass layoffs". Second, to the extent WARN concerns itself with layoffs (in which there is an ongoing business and the possibility that the affected employee may be re-employed), the NLRA's six-month limitations period is consistent with federal law's preference for rapidity in resolving labor disputes--even those having nothing at all to do with the collective bargaining relationship. See infra, note 34 (discussing Age Discrimination in Employment Act and Title VII claims). Third, to the extent that WARN does concern itself with plant closings, the need for promptness in asserting claims becomes even more intense, as is discussed later. Fourth, the six-month NLRA period does apply to any unfair practice claim under the NLRA which might arise from a plant closing without notice; Congress did not create a special limitations period for unfair labor practices connected with a plant closing.

 19

Although Halkias' action includes Oklahoma plaintiffs from General Dynamics' Tulsa facility, no party suggests borrowing an Oklahoma period. This lends credence to our concern, discussed infra, that in a choice of law situation involving borrowing a state limitations period, a district court will look to that of the forum state

 20

A brief was filed on behalf of appellants by the following amici curiae: Texas AFL-CIO, Oil Chemical and Atomic Workers, United Mineworkers of America, Automobile Mechanics' Local No. 701, NLG/Sugar Law Center for Economic and Social Justice, and the UAW

 21

But, at oral argument, when asked which Texas period should be applied, Staudt's counsel replied that he was "afraid you were going to ask that". Although Staudt urged the tort period in his brief, by oral argument, he evidently had changed his mind, stating a preference for the residuary period, and expressly disavowing a preference for the contracts period (the period selected by the Second Circuit). Given the difficulty in identifying a single Texas limitations period that is most analogous, we concur with a sentiment expressed by the Third Circuit in its decision to apply the NLRA's limitations period to actions under the Employee Protection Program of the Airline Deregulation Act: " [I]t is not easy to find a state limitations period that is an appropriate analogy." Haggerty, 952 F.2d at 786

 22

We decline to do so. We review freely, and rule on, legal issues such as this; there is no reason to remand to district courts that have already held that the NLRA presents a closer analogy to WARN than any Texas period

 23

Tort claims exist to compensate an individual for an injury proximately caused by the defendant. But, even if a terminated worker in a WARN situation gets a higher paying job the day after his termination, and therefore arguably has no "injury", WARN still seems to provide for two months pay if the requisite notice was not given. See 29 U.S.C. § 2104(a) (1-2), (7)

 24

The breach of contract period may come closer than any other Texas period. And, if nothing else, utilizing it would create some semblance of uniformity, given the Second Circuit's borrowing a state contract limitations period. But, the rationale utilized by the Second Circuit in choosing this period does not necessarily dictate the same result when Texas law supplies the period. The Second Circuit reasoned that the contract period should govern because it is the period for workers' compensation claims and "wrongful discharge" claims in Vermont. United Paperworkers, 999 F.2d at 57. In Texas, however, a workers' compensation claim must be submitted to the Texas Workers' Compensation Commission within one year. Tex.Lab.Code Ann. Sec. 409.003 (West 1994). Texas' traditional cause of action for wrongful discharge for asserting a workers' compensation claim was subject to the two-year tort limitations period. See Almazan v. United Services Auto. Ass'n, 840 S.W.2d 776, 779-80 (Tex.Ct.App.1992) (discussing wrongful discharge action under Tex.Rev.Civ.Stat.Ann. art. 8307(c) (West 1994) (repealed 1991)), error denied (March 3, 1993). Thus, following the Second Circuit's reasoning may produce a very different result in Texas than it did in Vermont, underscoring the need for uniformity provided by borrowing the limitations period from a federal statute

 25

The Third Circuit recognized that state law provided less than a "perfect analogy". But, because the actions were timely under any of the four potential state periods, it left unanswered which of the four, ranging from two years to six years, is more analogous to WARN than the NLRA. United Steelworkers, 32 F.3d at 61 & n. 5

 26

Recall, as discussed supra, that the NLRA has been interpreted to require WARN-like notice

 27

General Dynamics suggests that if a Texas period is to apply, it should be the six-month period under the Texas Pay Statute. See Tex.Rev.Civ.Code Ann. art. 5155, Sec. 5(a). This provision may be more similar to WARN than those offered by appellants and amici; however, it does not provide as close an analogy as does the NLRA. WARN does not compensate for past services rendered

 28

WARN only applies to employers with "100 or more employees, excluding part-time employees" or "100 or more employees who in the aggregate work at least 4,000 hours per week". 29 U.S.C. § 2101(a) (1) (A-B)

 29

Texas courts generally describe statutes of limitations as procedural and apply their own. See, e.g., Hollander v. Capon, 853 S.W.2d 723, 727 (Tex.Ct.App.1993) ("The statute of limitations is a procedural issue. If the action is barred by the statute of limitations of the forum court in which the lawsuit is pending, no action may be maintained even if the action is not barred in the state where the cause of action arose.")

 30

The dissent does not attempt rigorous analysis of what the effect of choice of law rules would be; it merely states that the "threat of forum shopping ... is stemmed by choice of law rules which dictate that the law of the site of the layoff ... would be chosen no matter where the suit was filed." Dissent at pages 247-48. To support this conclusion, the dissent relies on the district court opinion in Automobile Mechanics'. But, that opinion contains no citation of authority for its sweeping statement that "choice of law rules would likely point to borrowing the law of the site". Automobile Mechanics', 830 F. Supp. at 436

 31

Without deciding the issue, because it is not before us, a class action for several different sites, each of which satisfies WARN's "single site" requirement, appears to be appropriate when the decision to close each site was simultaneous and arose out of the same exigency. See 29 U.S.C. § 2101(a) (2) ("single site" requirement); see also Fed. R. Civ. P. 23(a-b) (class action requirements)

 32

Short 's discussion merits reiteration:

From the perspective of practitioners litigating cases originating in many states (especially class actions), the situation is a nightmare. Lawyers and courts alike devote untold hours to identifying proper state analogies and applying multiple (conflicting or cumulative) tolling doctrines. "This uncertainty and lack of uniformity promote forum shopping by plaintiffs and result in wholly unjustified disparities in the rights of different parties litigating identical claims in different states. Neither plaintiffs nor defendants can determine their rights with any certainty. Vast amounts of judicial time and attorneys' fees are wasted."

Short, 908 F.2d at 1389 (citation omitted).

 33

Referring to Sec. 301 of the Labor-Management Act of 1947, 29 U.S.C. § 185

 34

In fact, federal law's preference for rapidity in labor dispute resolution explicitly extends to areas that have nothing to do with the collective bargaining process. See 29 U.S.C. § 626(d) (1) (180-day filing period before EEOC for claims under Age Discrimination in Employment Act); 42 U.S.C. § 2000e-5(e) (1) (180-day filing period with EEOC for Title VII claims)

 35

Appellants protest that WARN claims are too difficult to prepare within six months. We disagree; moreover, many WARN plaintiffs have brought their claims within six months after the accrual of their cause of action. See, e.g., Local 217, Hotel & Restaurant Employees Union v. MHM, Inc., 976 F.2d 805, 807 (2d Cir. 1992) (within approximately five months of closing of hotel by employer); Local 397, Int'l Union of Electronic, Electrical, Salaried Mach. & Furniture Workers v. Midwest Fasteners, Inc., 763 F. Supp. 78, 80-81 (D.N.J. 1990) (within three months of plant closing). Indeed, General Dynamics states that the Missouri action was filed within 10 days of the layoffs at that facility. See supra, note 3

Likewise, we reject the suggestion that utilization of a six-month period is incompatible with WARN's requirement that a workload reduction or temporary layoff last more than six months before becoming actionable. See 29 U.S.C. § 2101(a) (6) (defining "employment loss" as termination, layoff exceeding six months, or reduction in hours of work exceeding six months). The parties' pleadings are less than clear on exactly what sort of "employment loss" they suffered. If predicated on termination, their claims accrued upon termination, see Automobile Mechanics', 830 F. Supp. at 434; 29 U.S.C. § 2101(a) (6) (A), and a six-month limitations period is not incompatible with such a situation. Because the issue is not before us, we leave for another day when a WARN claim accrues, particularly in the diminished workload or layoff situation. Assuming, without deciding, that a layoff or reduced workload must persist for six months before becoming actionable, the limitations period would not begin to run until after that period expired. See Automobile Mechanics', 830 F. Supp. at 434 (noting differences in time of accrual if claim is predicated on termination or layoff; and, while accrual issue not clear from pleadings, finding action would be timely whether claim arose from layoff or termination). In any event, whether the appellants suffered a layoff or a termination, their claims are untimely, in light of the six-month limitations period we adopt. The question that Halkias presents, "When does the layoff action accrue?", is one that exists regardless of the length of the limitations period.

We also reject the suggestion that a six-month period is unworkable because, unlike the NLRA, WARN does not provide a "complex administrative structure" for pursuit of claims. See United Paperworkers, 999 F.2d at 55. The Supreme Court has applied the NLRA limitations period to actions brought by an employee against his employer for breach of the collective bargaining agreement and against his union (under Sec. 301 of the Labor Management Relations Act) for breach of the duty of fair representation, see DelCostello, 462 U.S. at 154-55, 170-72, 103 S. Ct. at 2285, 2293-94; see also Coyle, 838 F.2d at 1405 (discussing DelCostello; describing action in DelCostello as a "hybrid action ... under Sec. 301"), a cause of action that requires some effort to bring because "the employee will often be unsophisticated in collective-bargaining matters.... He is called upon, within the limitations period, to evaluate the adequacy of the union's representation, to retain counsel, to investigate substantial matters ..., and to frame his suit". DelCostello, 462 U.S. at 166, 103 S. Ct. at 2291. By contrast, a WARN plaintiff need only know that he was terminated or laid off without notice; provided his situation comports with the requisites of WARN, he can sue. Indeed, the absence of a "complex administrative structure" may be emblematic of the simplicity of a WARN claim, rather than a justification for a lengthy limitations period.

The dissent emphasizes that the plaintiffs in these cases are not represented by unions, and thus are deprived of "union legal representation". As discussed supra, an employee bringing hybrid Sec. 301 actions against his union for breach of the duty of fair representation (and against his employer for breach of the collective bargaining agreement) is, after DelCostello, required to bring his claim within the NLRA's six-month limitations period. And, as discussed, he faces obstacles at least as great as those faced by one seeking to assert a simple WARN claim. Yet he, too, will lack union legal representation.

The dissent similarly opines that we "hold [ ] the plaintiffs to a deal they did not make: They suffer the six-month requirement but benefit from no corresponding protection of their rights." Dissent at page 244 (footnote omitted). Of course, WARN confers specific rights on workers, and provides remedies for their violation. It is not as if workers after the enactment of WARN--even with the application of a six-month limitations period--"benefit from no corresponding protection of their rights."

The dissent prompts a question: Can anyone imagine that, with "thousands of workers" in "distress" after a termination or layoff without WARN notice, see dissent at pages 244, 245, a lawyer (or lawyers) will not be far behind? This seems especially probable given WARN's provision of attorneys' fees for prevailing parties. See 29 U.S.C. § 2104(a) (6).

 36

Of course, prompt filing does not necessarily mean prompt resolution. That will depend on numerous factors, including discovery and the district court's caseload. But, obviously, the sooner suit is filed, the sooner the resolution from the date of the employment loss

The dissent asserts that the application of a six-month period would "stymie WARN's true objective". Dissent at page 248. The reason proffered for this assertion is that "companies can relax a bit and rest assured that they may 'make redundant' many legally unsophisticated and unsuspecting workers" with a six-month period, while a longer period would deter violations of WARN. Obviously, the rationale presupposes that an employer chooses not to comply with WARN coldly and rationally. It is quite questionable that a company would intentionally violate WARN in reliance on the possibility that a worker would not sue. In fact, if just one worker approaches a lawyer, then the possibility of a class action becomes very real. In these circumstances, a decision to violate WARN on the basis assumed by the dissent would be illogical. Moreover, if this premise of the dissent has merit, one would suspect that employers would "just relax a bit" regarding the sexual harassment of, or age discrimination against, their "legally unsophisticated and unsuspecting workers" because of the short limitations periods attaching to Title VII or the Age Discrimination in Employment Act. See supra note 34. Current events suggest otherwise.

In addition, the dissent invites a false choice. While the borrowing of state limitations periods may in several states result in long periods (so long that their application to a plant closing or mass layoff seems inappropriate), a state limitations period could be as short as the NLRA's, e.g., the six-month period of the Texas pay statute, see supra note 27, dissent at page 246 n. 36, or perhaps even shorter.

 37

The enactment of Sec. 1658 reflected concern by both Congress and the Federal Courts Study Committee that the process of borrowing limitations periods for federal causes of action was, to say the least, problematic. See H.R.Rep. No. 734, 101st Cong., 2d Sess. 24 (1990), reprinted in 1990 U.S.C.C.A.N. pp. 6802, 6860, 6870

 38

He also contends that Congress likewise violated the Fourteenth Amendment, which, by its terms, applies only to the States

 1

29 U.S.C. § 160(b)

 2

29 U.S.C. §§ 2101-2109

 3

The certified class represented by John Halkias and Barry Jackson consists of about two thousand non-unionized former employees of General Dynamics's Fort Worth and Tulsa operations

 4

See United Paperworkers Local 340 v. Specialty Paperboard, Inc., 999 F.2d 51, 53 (2d Cir. 1993); United Steelworkers of Am. v. Crown Cork & Seal Co., 32 F.3d 53 (3d Cir. 1994)

 5

 501 U.S. 350, 111 S. Ct. 2773, 115 L. Ed. 2d 321, reh'g denied, 501 U.S. ----, 112 S. Ct. 27, 115 L. Ed. 2d 1109 (1991)

 6

Id. at 356, 111 S. Ct. at 2778, 115 L. Ed. 2d at 331 (citations omitted)

 7

Id

 8

DelCostello v. International Bd. of Teamsters, 462 U.S. 151, 161, 103 S. Ct. 2281, 2289, 76 L. Ed. 2d 476 (1983)

 9

Lampf, 501 U.S. at 356, 111 S. Ct. at 2778, 115 L. Ed. 2d at 331-32 (quoting Reed v. United Transportation Union, 488 U.S. 319, 324, 109 S. Ct. 621, 625, 102 L. Ed. 2d 665 (1989)). See also, United Paperworkers, 999 F.2d at 53 (" [The] Supreme Court has set forth limited circumstances under which it might be preferable to borrow a federal limitations period")

 10

Lampf, 501 U.S. at 356, 111 S. Ct. at 2778, 115 L. Ed. 2d at 331

 11

DelCostello, 462 U.S. at 161, 103 S. Ct. at 2289

 12

Maj. op. at 230

 13

Justice Scalia did not join part II.A of Justice Blackmun's plurality opinion. In his concurring opinion, Justice Scalia said: "In my view, absent a congressionally created limitations period state periods govern, or if they are inconsistent with the purposes of the federal act, no limitations period exists ". Lampf, 501 U.S. at 364, 111 S. Ct. at 2783, 115 L. Ed. 2d at 337 (Scalia, J., concurring) (emphasis added). Justices Stevens and Souter dissented. Justice Stevens's dissent explicitly disagrees with " [t]he Court's rejection of the traditional rule of applying a state limitations period when the federal statute is silent ..." Id. at 368, 111 S. Ct. at 2784, 115 L. Ed. 2d at 339 (Stevens, J., dissenting) (emphasis added). Finally, Justices O'Connor and Kennedy dissented from the Court's refusal to make its decision prospective only. Even they agreed, however, with the view that the plaintiffs' "claims were governed by the state statute of limitations for the most analogous state cause of action". Id. at 369, 111 S. Ct. at 2785, 115 L. Ed. 2d at 340 (O'Connor, J., dissenting) (emphasis in original)

 14

At oral argument a member of the panel asked counsel for General Dynamics if it was true that the Court had to look to state law first, before considering federal law. Counsel conceded the point. The following exchange took place

Judge: "When we write this, we have to go through the state statutes first before we get to the federal statutes".

Counsel: "I completely agree with that. That is the proper analysis, there is no doubt."

 15

The Court in Lampf cited both the DelCostello case and Agency Holding Corp. v. Malley-Duff & Associates, Inc., 483 U.S. 143, 107 S. Ct. 2759, 97 L. Ed. 2d 121 (1987)--both of which the majority cites extensively in support of its position--for the proposition that federal law may supply the suitable period "when the operation of a state limitations period would frustrate the policies embraced by the federal enactment" (here, WARN). Id. at 355-56, 111 S. Ct. at 2778, 115 L. Ed. 2d at 331

 16

Id. at 357, 111 S. Ct. at 2779, 115 L. Ed. 2d at 332

 17

See Id. at 362, 111 S. Ct. at 2781-82, 115 L. Ed. 2d at 335-36. See also United Parcel Service v. Mitchell, 451 U.S. 56, 68 n. 4, 101 S. Ct. 1559, 1567 n. 4, 67 L. Ed. 2d 732, 744 n. 4 (1981) (Stewart J., concurring) (state borrowing rule is more appropriate when applied to a congressionally created cause of action as opposed to an implied one)

 18

Maj. op. at 230

 19

Id. at 230. This mistake is repeated throughout the majority's opinion. For example, after detailing its perceived similarities in purpose and structure between WARN and the NLRA, the majority then states: "The question remains, however, whether a state period provides as close or closer 'fit' to WARN than does the NLRA." Id. at 234. Again, the majority has the flow chart running in the wrong direction. We first look to state law and, if it fits and is not at odds with the federal statute, we use it

 20

Judge McBryde's opinion quoted the following passage from Judge Garcia's:

[T]he court concurs with the conclusion that "the federal policy warranting rapid resolution of [labor and employment] disputes favors the NLRA's shorter limitations period; and that a standardized limitations period would promote uniformity in enforcing the WARN act."

Halkias v. General Dynamics Corp., 825 F. Supp. 123, 125 (N.D. Tex. 1993).

 21

The majority criticizes my dissent for failing to appreciate that WARN governs not only plant closings, but mass layoffs and temporary closings. See maj. op. at 227 n. 1, 234 n. 18. The severance of the employment relationship and the repercussions that follow are the same for the victim of a mass layoff as for one who loses his job through a plant closing. That is true even if, in the case of a layoff, the possibility of reuniting the fired employee and the employer exists. Hence, my discussion of the different purposes of WARN and the NLRA applies with equal force to all of the job loss scenarios contemplated by WARN's provisions

 22

See also, Crown Cork & Seal, 32 F.3d at 58 (WARN's "broader purpose" is to "protect workers, their families and their communities in the wake of potentially harmful employment decisions".)

 23

I make this distinction in response to General Dynamics's assertion that the rapid resolution of disputes will better alleviate the distress of job loss by ensuring that the statute's remedies are promptly pursued. The statute presumably was passed with the belief that proper advance notice would be given, rendering the remedial provisions unnecessary in those cases

 24

It bears emphasizing that the plaintiffs in both cases are not represented by labor unions

 25

In this way, the present case is drastically different from the statute at issue in Lampf. In Lampf, the Court held that a federal statute of limitations applied to actions brought pursuant to Rule 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b), and Securities and Exchange Commission Rule 10(b) (5). The Court looked to the contemporaneously enacted remedial provisions that did supply a limitations period. See Lampf, 501 U.S. at 359-60, 111 S. Ct. at 2780, 115 L. Ed. 2d at 333-34. Hence, the Court had the benefit of the balance struck by the Congress in limiting similar provisions of the same act. Unfortunately, we are not presented with those circumstances

 26

The Second Circuit appreciated this distinction. In criticizing the district court's Staudt decision now before us, the court in United Paperworkers stated:

Those courts which have deemed the NLRA period most applicable to WARN actions have failed to grasp this crucial distinction between statutes which specifically regulate the collective bargaining relationship and those which remain peripheral to that concern.

United Paperworkers, 999 F.2d at 55.

 27

DelCostello, 462 U.S. at 171, 103 S. Ct. at 2294 (citation omitted); United Paperworkers, 999 F.2d at 53

 28

Reed, 488 U.S. at 394, 109 S. Ct. at 666; United Paperworkers, 999 F.2d at 53-54

 29

Fort Halifax Packing Co., Inc. v. Coyne, 482 U.S. 1, 20-21, 107 S. Ct. 2211, 2222, 96 L. Ed. 2d 1 (1987) (internal quotations omitted) (evaluating Maine plant closing law)

 30

WARN defines a "mass layoff" sufficient to invoke its terms as a reduction in work force of (1) at least 33 percent of the employees (excluding any part-time employees) and (2) at least 50 employees (excluding any part-time employees); or at least 500 employees (excluding any part-time employees). 29 U.S.C. § 2101(a) (3) (B)

 31

The majority's response to the practical problems that the six-month limitations period will cause is a statement that "many WARN plaintiffs have brought their claims within six months...." Maj. op. at 238 n. 35. It is telling, however, that the cases the majority cites for evidence of that proposition had unions for plaintiffs. The majority never addresses the additional burdens that non-unionized potential WARN plaintiffs face

 32

The majority's weak comparison of the statutes' purposes implicitly concedes this. WARN's purpose is to protect workers and their families by requiring notice, 20 C.F.R. Sec. 639.1(a) (1993), whereas the NLRA's purpose is to facilitate the process of collective bargaining. 29 U.S.C. § 151. The majority's statement that they are similar in that both "regulate labor-management relations," maj. op. at 233, is only slightly more convincing than the fact that both were passed by Congress

 33

Cf. Crown Cork & Seal, 32 F.3d at 57 ("the mere fact that a statute touches upon issues of labor law does not mean that the Court must resort to the statute of limitations contained in Sec. 10(b) of the NLRA")

 34

United Paperworkers, 999 F.2d at 53-54

 35

Id. at 54

 36

The other good candidates are:

(1) The two-year Texas limitations period for personal injury, wrongful discharge, and employment discrimination claims, Tex.Civ.Prac. & Rem.Code Ann. Sec. 16.003;

(2) The four-year Texas residual statute of limitations, Id. Sec. 16.051;

(3) The six-month limitations period of the Texas Pay Statute, Tex.Rev.Civ.Stat.Ann. art. 5155; and

(4) The four-year federal residual statute of limitations Congress recently enacted for all federal causes of action that do not include their own limitations period, 28 U.S.C. § 1658.

At first glance, number (4) appears to cover exactly the instant case. Unfortunately, as the majority correctly indicated, Sec. 1658 by its terms applies only to "civil action [s] arising under an Act of Congress enacted after the date of the enactment of this section." Because WARN was enacted before Sec. 1658, Sec. 1658 does not directly control in WARN cases. Several federal district courts have recognized that Sec. 1658 supersedes the Supreme Court's Lampf analysis for causes of action filed after Sec. 1658 took effect, but because WARN preceded Sec. 1658, the Supreme Court's Lampf analysis governs this case.

 37

See DelCostello, 462 U.S. at 171, 103 S. Ct. at 2294 (" [A]s the courts have often discovered, there is not always an obvious state-law choice for application to a given federal cause of action; yet resort to state law remains the norm for borrowing of limitations periods.")

 38

Auto Workers v. Hoosier Cardinal Corp., 383 U.S. 696, 705 n. 7, 86 S. Ct. 1107, 1113 n. 7, 16 L. Ed. 2d 192. See also, Frymire v. Ampex Corp., 821 F. Supp. 651, 655 (D. Colo. 1993) (applying state contract statute of limitations to WARN claim); Wallace v. Detroit Coke Corp., 818 F. Supp. 192, 196 (E.D. Mich. 1993) (same)

 39

United Paperworkers, 999 F.2d at 53, 57

 40

See Reed, 488 U.S. at 327, 109 S. Ct. at 627; Frymire, 821 F. Supp. at 655. I note, however, that I would have assented to a remand order to the district court to find which state limitations period was most appropriate

 41

DelCostello, 462 U.S. at 172, 103 S. Ct. at 2294. Though I do not wish to beat a dead horse, I reiterate that studying the litigation practicalities and policy considerations is our last task--we reach it only if we have found a state statute that is at odds with the operation of WARN. That is not the case here

 42

See United Paperworkers, 999 F.2d at 56 (forum shopping concerns are not great because "plant closing" is limited to a single site of employment and WARN actions will be filed where the injury took place or the employer does business)

 43

Auto Mechanics Local 701 v. Santa Fe Term. Serv., 830 F. Supp. 432, 436 (N.D. Ill. 1993)

 44

 383 U.S. 696, 701, 86 S. Ct. 1107, 1110-11, 16 L. Ed. 2d 192 (1966)

 45

Id. at 702, 86 S. Ct. at 1111 (internal quotations omitted); DelCostello, 462 U.S. at 162, 103 S. Ct. at 2289

 46

Maj. op. at 238

 47

Maj. op. at 239

 48

At the risk of overkill, I note that the Harvard Law Review note upon which the majority relies as support for this trend in actuality supports my conclusion: "Writing for a plurality, Justice Blackmun acknowledged the continuing validity of the state borrowing rule." The Supreme Court, 1990 Term--Leading Cases, 105 Harv. L. Rev. 177, 400 (1991). The article details that the exception to this accepted rule is just as I have stated: when the federal period "clearly provides a closer analogy" and when federal policies at stake make the federal rule a "significantly more appropriate vehicle" than the state rule. Id. Any reliance, then, on some perceived trend away from the state borrowing rule is pure conjecture, particularly in the face of the clean rule announced in Lampf

 49

WARN has no cruelty provision and, so, Glastron fired three hundred employees on Christmas Eve