Citations

Full opinion text

TRAGER, District Judge: In this case we revisit the question of reconciling the proper role of the courts and of labor arbitrators where a claim is made that a labor arbitration award should not be enforced because it contravenes public policy. The fáetual context, nuclear safety, is one that this Court" has' not confronted before, and one which makes the issue more difficult than it otherwise might be. Plaintiff-appellant,, International Brotherhood of Electrical Workers, Local 97, appeals from a judgment of, the United States District Court for the Northern District of New York vacating an arbitration award which ordered the conditional reinstatement of Patrick Rando, a technician at. a nuclear power plant owned by defendant-appellee Niagara Mohawk Power ' Corporation. Rando was discharged after he adulterated a drug test and subsequently tested positive for cocaine. On appeal, plaintiff argues that enforcement of the arbitration award reinstating Rando would not violate public policy and that the district court exceeded its reviewing power in setting aside the arbitration award. We conclude that public policy here is defined by the regulations of the Nuclear Regulatory Commission and that the arbitration panel’s award of conditional reinstatement, following rehabilitation, does not violate public policy as defined by those regulations. We also conclude that the district court impermissibly employed a de novo standard when reviewing the arbitration panel’s factual findings. Accordingly, the district court erred in vacating the arbitration award and we reverse. Background Defendant Niagara Mohawk Power Corporation (“Company”) is engaged in the. business of generating and supplying electrical power and supplying natural gas to residential and commercial users throughout upstate New York. Among the facilities-operated by the Company are two nuclear power plants located at Nine Mile Point in Lycoming, New York. Plaintiff International Brotherhood of Electrical Workers, Local 97 (“Union”) is the collective bargaining representative for a bargaining unit consisting of hourly paid production, maintenance, non-technieal, office and clerical employees of Niagara Mohawk. The Company and the Union (and the Union’s predecessor local unions affiliated with the International Brotherhood of Electrical Workers) have been parties to a series of collective bargaining agreements concerning the terms and conditions of employment for employees in the collective bargaining unit which included Rando. The events which preceded Rando’s discharge are essentially undisputed. At the time of his discharge, Rando was employed as a chemistry technician at the Company’s Nine Mile Point, Unit 2 plant, a position he held for approximately four and a half years. As a chemistry technician, Rando was responsible for ensuring that the plant chemistry was maintained within the technical specifications required by the federal Nuclear Regulatory Commission (“NRC”), including testing “air effluence, influence, water effluence, pump oils” and other related activities. Opinion and Decision of the Arbitration Panel (“0 & D”) at 4. Prior to the incident that ultimately would give rise to this appeal, Rando had a clean disciplinary record. As a nuclear power plant operator, Niagara Mohawk is required to comply with extensive regulations promulgated by the NRC. The regulations are designed to ensure the safety of nuclear power plant workforces and the public at large. One of the NRC regulations mandates that nuclear power plants maintain “Fitness For Duty Programs” which provide for random drug screening of employees. See 10 C.F.R. §§ 26.1, 26.27. On February 4, 1993, pursuant to the Company’s Fitness for Duty Program, Rando was selected on a random basis to provide a urine sample for testing. The Company’s policy also required that, prior to submitting the sample, Rando sign a consent form in which he certified that the specimen he was about to provide was not adulterated or altered in any way. The signed consent form acknowledged Rando’s understanding that the results of the test or his lack of cooperation in providing the specimen “[would] be used to make decisions regarding [his] future access to nuclear facilities.” Consent Form, Joint Appendix at 34. The sample that Rando provided on February 4 was detected by technicians to smell of chlorine, and subsequent tests and laboratory analysis confirmed the presence of sodium hydrochloride (a chloride compound) in the specimen. See 0 & D at 4. Rando was contacted at home and advised that he would be required to provide a second sample as soon as possible. On February 5, Rando submitted a second urine sample, this time under observation as required by the consent form which accompanied his first sample. The second sample submitted-by Rando tested positive for cocaine. See id. On the same day, a meeting was convened; attending were Robert LaDue, Niagara Mohawk’s then General Supervisor of Labor Relations for the Nuclear Division; Mike Kenifie, Rando’s union representative; and Rando. At this meeting, Rando admitted that he had altered his initial drug test with chlorine but stated that he did not use drugs or illegal substances. When asked why he had adulterated the test, Rando responded that he did it to “test the system.” 0 & D at 5. A subsequent search of Mr. Rando’s locker by security personnel revealed the existence of a ten ounce bottle of chlorine solution, a small vial of urine, and another small vial of a chlorine solution. The Company claims that the results of the search were significant because they demonstrated that “[the] act on [Rando’s] part to alter his drug test was planned well in advance, premeditated.” 0 & D at 5. According to Rando, he obtained the chlorine on the morning of the test. A disciplinary meeting was held on February 8, 1993 pursuant to Article XVI of the collective bargaining agreement. At this meeting, Rando admitted that he had a substance abuse problem and that he had altered the first sample in an attempt to conceal this problem. He and the Union requested leniency from the Company regarding the imposition of a disciplinary penalty. Rando’s union representative noted that the Company had found no fault with Rando’s job performance, despite conducting a full audit of his work records. The union representative further noted that Rando had no prior disciplinary infractions and had once been selected as safety employee of the month. The Company, however, represented by LaDue; decided to terminate Rando that day “on the basis of his intentional behavior to defraud the company as far as his drug test was concerned and his falsification of the document when he certified that it. was not altered in any way.” 0 & D at 5. The Union appealed the termination through the Step Three grievance hearing procedure specified in the collective bargaining agreement. Rando’s union representative argued that Rando had voluntarily begun treatment for chemical abuse following his discharge and that he deserved credit for taking that step. More significantly, Rando’s union representative cited Section 3.2.3 of the Nuclear Division Directive, a Company-issued document governing nuclear employees, which provides that “[alteration or interference with the proper collection of samples or conduct of tests shall be treated as a positive test and shall result in the appropriate management sanctions.’* Rando’s union repre-' sentative noted that pursuant to the Company’s own Fitness for Duty Program, an initial positive test subjected an employee to a two week suspension. See 0 & D at 6. Representing the Company at the Step Three grievance hearing, LaDue responded that the termination had nothing to do with a positive drug test, but rather was appropriate because Rando had proved himself untrustworthy by falsifying a document and altering his sample. See id. Citing NRC regulations, LaDue stated: “[They] require, if someone is untrustworthy and unreliable, he can’t have unescorted access to the site.” Id. LaDue admitted that a .first positive test had never resulted in more than a two-week suspension, combined with referral to the Employee Assistance Program (“EAP”). See id. The matter was ultimately submitted to arbitration as required by the collective bargaining agreement. See Article XXII, Joint Appendix at 72. The parties stipulated that the issue to be decided by the arbitration panel (“Panel”) was: “Did the Company have just cause for the discharge of the Grievant, Patrick Rando? If not, what shall the remedy be?” 0 & D at 1. On July 23, 1995, the Panel issued its “Opinion and Decision.” The Panel found that Rando had been discharged without just cause and ordered that he be reinstated to his former position “contingent on his producing a negative drug test and satisfying any requirements that may be imposed by the Employee Assistance Program after an evaluation.” 0 & D at 14. The award also provided that, pursuant to the Fitness for Duty Program, Rando would be subject to follow-dp testing at the Company’s discretion for eighteen months, and that the Company could require that Rando provide urine samples under observation. See id. The Panel further ordered that if Rando met the terms'for reinstatement, he was to receive back pay for the period from the date of his discharge to the date of his reinstatement, less eighteen months. See id. At this point, almost thirty months had passed since Rando’s discharge. The Company refused to reinstate Rando, and the Union brought this- action in the United States District Court for the Northern District of New York seeking confirmation of the award pursuant to § 301 of the Labor Management Relations Act, 29 U.S.C. § 185. The Company counterclaimed to vacate the award on the ground that its enforcement would violate public policy. The Union moved for summary judgment to confirm the award, dismiss the Company’s counterclaim, and further sought an award of attorney’s fees and costs. The Company cross-moved for summary judgment to vacate the award, dismiss the complaint in its entirety, and also sought an award of costs and fees. The district court granted the Company’s cross-motion for summary judgment and vacated the award, holding that Rando’s reinstatement would contravene public policy. See International Bhd. of Elec. Workers, Local 97 v. Niagara Mohawk Power Corp., 950 F.Supp. 1227, 1236 (N.D.N.Y.1996). The district court further denied both parties’ motions for attorney’s fees. See id. The Union appeals from that portion of the district court’s December 30,1996 judgment denying the Union’s motion for summary judgment and granting the Company’s cross-motion for summary judgment vacating the award. The “Opinion and Decision” of the Arbitration Panel The opinion of the arbitration panel begins with a review of the relevant provisions of the collective bargaining agreement, the applicable NRC regulations, and the Company’s Fitness for Duty Program. Next follow the details of Rando’s discharge, a summary of the parties’ respective contentions, and finally, a discussion of the majority’s rationale for its determination that Patrick Rando was terminated without cause. In analyzing the propriety of the Company’s disciplinary response to Rando’s behavior, the Panel examined at length the Company’s response within the context of both the NRC regulations and the programs enacted by the Company in compliance with these regulations. Under the NRC’s Fitness for Duty regulations, the sanction to be applied when an employee first tésts positive for drug use (a “first positive”) is left to the discretion of the employer, but must involve at minimum a two-week denial of unescorted access to protected areas and referral to an employee assistance program. See 10 C.F.R. § 26.27. No sanction is specified for adulteration in the NRC regulations. In 1987, in accordance with NRC regulations, the Company unilaterally implemented an Alcohol/Drug Screening/Testing-Fitness for Duty Program that applied to both represented and non-represented employees. In November, 1988, the Company and the Union reached agreement on an “Alcohol/Drug Testing-Fitness for Duty Program” (“Joint Program”) which superceded the Company’s unilaterally-imposed program. The Joint Program applies to all represented employees, including those in nuclear operations. In the Joint Program, the parties agreed upon a testing protocol, penalties for offenders, and opportunities for employees to seek help for substance abuse. See O & D at 3. The Joint Program specifies that the penalty for a first positive test result is a suspension of up to two weeks, and that following such suspension, an employee must take a repeat drug test, yield a negative result, and must sign an authorization permitting the Company to conduct unscheduled alcohol/drug tests for a period of eighteen months after his return to work. Although the Joint Program does not address the penalty for adulteration, an additional, Company-issued document, referred to as the Nuclear Division Directive, specifies that adulteration of a drug test is to be treated as a positive test. In determining whether the just cause standard specified in Article XXII of the collective bargaining agreement was met when the Company discharged Rando for adulterating his drug test, the Panel stated that just cause requires notice of the potential consequences of misconduct prior to the imposition of discipline. See 0 & D at 9. The Panel found that Rando was discharged without cause principally because he had not received sufficient notice of discharge as a potential disciplinary penalty for the offense of adulteration. See id. at 9-14. The majority found ‘-‘[t]he notion that an employee could be subject to a disciplinary penalty that was never announced [to be] alien to the concept of due process.” Id. at 9. The Panel described “[p]roper notice [as] the first of the canonical ‘seven tests’ of just cause.” Id. (citing Grief Brothers, 42 LA 555 (Daugherty, 1964)). Furthermore, the Panel observed, in order to satisfy the just cause standard, such notice must be “full and unmistakable.” 0 & D at 9. The Panel concluded that Patrick Rando had not received adequate notice that the disciplinary consequences attendant to adulterating a urine sample included termination. In so finding, the Panel rejected the Company’s contention that the act of adulterating the sample was tantamount to falsification of documents, an offense which the Company claimed “intrinsically warrants discharge,” and the punishment for which the Company asserted was within management’s sole authority and of which employees were aware. See id. at 9. In support of this argument, the Company noted that annual training sessions for nuclear employees emphasized that falsification of documents, such as work records, would lead to discharge. However, the Panel found no evidence that the training explicitly equated adulteration of drug tests with falsification of documents, stating that “[t]he link is by no means self-evident.” ■ Id. The Panel concluded that Rando’s act of adulterating the urine sample constituted an insubordinate refusal to submit a specimen as required by the Joint Fitness Program rather than falsification of a work record. See 0 & D at 9-14. Furthermore, the Panel found that nothing in the language of the consent form that accompanies a submitted urine sample places an employee on notice that discharge is a potential consequence of adulterating a sample. Moreover, according to the Panel, the language of the consent form is inconsistent with the Company’s argument that this offense constitutes falsification of work records meriting dismissal. To the contrary, the Panel noted that the consent form signed by Rando merely put an employee on notice that tampering with the sample may lead to the employee being required to produce a second sample, perhaps conducted under observation. The Panel observed: “It is paradoxical to call for a retest in cases of adulteration and at the same time deem that conduct a summarily dischargeable offense (falsification).” 0 & D at 9. Also, according to the Panel, the consent form could not be compared to a falsified work record because when Rando signed the form, he had not yet submitted the urine sample, and thus, no work record was yet in existence. Id. at 10. Rather, the Panel observed, the offense actually occurred later when, in submitting an adulterated sample, Rando “failed to comply with the undertaking he gave on the form — a failing akin to insubordination.” Id. The Panel also found evidence in the Company’s history which suggested that the Company had at one time considered adulteration to be a matter of insubordination, rather than one necessarily related to integrity. While the 1988 Joint Fitness Program did not address the issue of adulteration at all, under the Company’s prior, unilaterally-imposed Fitness for Duty Program, a “Company Drug Testing Information and Instruction Sheet” had been issued to all employees which cautioned, among other things, that “[a]ny adulteration or switching of urine by an employee will be considered insubordination and will result in disciplinary action up to and including discharge.” 0 & D at 15. The Panel considered the omission of a similar provision in the Joint Program to be significant. See 0 & D at 9. Furthermore, the Panel took note of the Company’s unilaterally-enacted Nuclear Division Directive, effective October 15, 1992, which states with respect to adulteration only that “[ajlteration or interference with the proper collection of samples or conduct of tests shall be treated as a positive test and shall result in appropriate management sanctions.” Nuclear Division Directive, § 3.2.3, Joint Appendix at 31. Given the Company’s prior view of adulteration as synonymous with insubordination, and in light of the lack of direct notice to employees of an intention to treat this conduct as a dischargeable offense, the Panel rejected the Company’s efforts to re-eharacterize the offense as falsification warranting summary dismissal. See 0 & D at 10. In characterizing Rando’s adulteration of the sample as insubordination as opposed to falsification, the Panel also found the past practice of the Company in addressing similar occurrences to be highly instructive. During the Step Three grievance hearing, LaDue admitted: “I’m not aware of someone being terminated for a first time drug, or alcohol, when they tested positive.” O & D at 6. The Panel found that the record contained no example of an employee discharged for adulteration since the Joint Program was adopted in 1988. The parties’ history did reveal one other case in which an employee was found to have adulterated a urine sample in connection with a drug test, but the employee had received only a three-week suspension. In that incident, which occurred in 1992, Doe, a gas division employee, tested positive for cocaine. Doe’s discipline consisted of a two week suspension for testing positive, a requirement that he continue his rehabilitation with the Company’s Employee Assistance Program and follow up on all prescribed courses of treatment, the possibility of spot drug screening for an eighteen month period, and the further penalty of an additional week of suspension for tampering with his urine specimen. Doe’s decision to tamper with his urine sample was viewed by the Company as a refusal to- be drug tested, an act which the Company claimed subjected him to further discipline under the Company’s Fitness for Duty Program. See 0 & D at 12. The Panel found the Doe case to be relevant to the issue of Rando’s discharge: In the Doe case adulteration was treated as a refusal to be tested, that is, as a form of insubordination, rather than a form of falsification. Doe was not a nuclear employee, ... but the difference between gas and nuclear cannot mutate the essential nature of the offense, transforming insub- - ordination into falsification. 0 & D at 13. The Company argued that one ease did not establish a binding practice and that Doe was given a “break” by management. See id. The Panel rejected this argument, stating that “the company’s characterization of [Doe’s] offense cannot be ignored.” Id. In short, there was no indication that the Company had ever treated Rando’s conduct, or any conduct close to it, as a cause for immediate termination. The Panel also found significant the fact that the Company’s Nuclear Division Directive, which conformed Company policy to the NRC regulations, did not treat adulteration as a disqualifying character offense. Rather, the Directive “brackets adulteration with ‘interference’ and orders that the adul-terator be treated 'as if he tested positive. Access is denied to those who are positive but, for a first positive, only briefly.” 0 & D at 10. The Panel observed that if adulteration were considered to be a form of fraud, the privilege of access certainly could not be re-conferred after such a brief hiatus. The Panel also noted that other offenses resulted in withdrawal of access for considerably longer periods of time, such as possession or use of .drugs or alcohol within protected areas, which carries with it a denial of access privileges for five years. See Nuclear Division Directive. § 3.3.1 . (citing 10 C.F.R. 26.27(b)(3)). “In the scheme propounded by the directive, then, adulteration does not rank among the most serious offenses.” 0 & D at 10. The Panel determined that because the right of unescorted access significantly impacts safety in the nuclear industry, the response established by the Directive could be taken -as an index of the gravity of the offense of adulteration. See id. The Panel noted: A person who was considered deficient in character would not be granted access to a regulated facility. The company cannot plausibly claim that an offense renders an employee inherently lacking in trust when the watchdog agency has no qualms about his returning to the facility. It is thus manifestly reasonable to take into account the extent of access denial in determining the justice of the discipline imposed on the grievant. It would be incongruous if adulteration merited only brief loss of access and at the same time automatic discharge. 0 & D at 10-11. In concluding, the Panel acknowledged: There can be no doubt that adulteration is contrary to the Joint Program, which afforded an employee an authorized way to avoid taking the test: he would have to accept two weeks off for refusal. The grievant decided to avoid taking the test in an unauthorized way — by rendering the urine useless. His conduct clearly invites a proportionate penalty of some kind, even though it does not justify discarding him as an unemployable moral reprobate. 0 & D at 13. The Panel thus ordered that Rando be reinstated to his former position, contingent on his producing a negative drug test and satisfying any requirements that might be imposed under the Employee Assistance Program after an evaluation. The Panel further ordered that if Rando met the requirements for reinstatement, he was to receive back pay for the period from the date of his discharge to the date of his reinstatement, less eighteen months, with any earn: ings during the eighteen month period to be deducted from the back pay. Rando was also deemed subject to follow-up testing at the Company’s discretion for eighteen months following his reinstatement, and the Company was permitted to require that samples be produced under observation. See 0 & D at 14. The District Court Opinion After examining the facts surrounding Patrick Rando’s discharge, the district court analyzed the proper role for a court when it is asked to review an arbitrator’s decision. The district court articulated the well-accepted proposition that courts are not authorized to reconsider the merits of an arbitration award, recognizing that the strong “federal policy of settling labor disputes by arbitration” would be undermined by a contrary position. International Bhd. of Electrical Workers, Local 97 v. Niagara Mohawk Power Corp., 950 F.Supp. 1227, 1231 (N.D.N.Y.1996)(citing Newsday, Inc. v. Long Island Typographical Union, 915 F.2d 840, 844 (2d Cir.1990)). Citing precedent of this Court expressing the deference to be accorded the decision of an arbitrator, the district court observed that an award will not be vacated “ ‘even if the arbitrator’s interpretation of the contract is clearly erroneous, so long as such Award is explained in terms that offer even a barely colorable justification for the outcome reached.’ ” 950 F.Supp. at 1231 (quoting Hygrade Operators, Inc. v. Local 333, ILA, 945 F.2d 18, 22 (2d Cir.1991)). The district court went on to note thát the Second Circuit’s position is entirely consistent with the standard expressed in United Paperworkers Int’l Union v. Misco, Inc., 484 U.S. 29, 38, 108 S.Ct. 364, 370-71, 98 L.Ed.2d 286 (1987), where the Supreme Court stated that “[a]s long as the arbitrator is even arguably construing or applying the contract and acting within the scope of his authority, that a court is convinced he committed serious error does not suffice to overturn his decision.” The district court then considered an exception to this general rule: “a court may, however, refuse to enforce an [arbitration] award pursuant to a collective bargaining agreement if the award is contrary to public policy.” 950 F.Supp. at 1231 (citing Misco, 484 U.S. at 42-43, 108 S.Ct. at 372-74; Newsday, 915 F.2d at 844). After citing a number of eases in which courts have vacated arbitrators’ awards that were held to violate public policy, the district court proceeded to articulate the public policy implicated by the present case. See 950 F.Supp. at 1232. The court stated that “federal statutes, regulations, and case law support a finding of a well-defined and dominant public policy against employment of individuals who deliberately violate nuclear safety rules and cannot meet NRC standards of trustworthiness and reliability.” Id. at 1234. The district court also declared that because of the strong public policy issues involved in this case, it intended to “review the findings of the arbitration panel de novo.” Id. In describing the level of deference that a reviewing court must give the factual findings underlying an arbitrator’s award, the district court stated that “[although reviewing courts must ordinarily accept an arbitrator’s findings of fact, in addressing public policy concerns, the courts are not to defer to the arbitrator, but rather consider these issues de novo.” Id. at 1232. The district court sought support for this proposition from Misco, in which the Supreme Court stated that “ ‘the question of public policy is ultimately one for resolution by the courts.’ ” Id. (quoting Misco, 484 U.S. at 43, 108 S.Ct. at 373). Having determined that it had the authority to conduct a de novo review, the district court rejected a number of findings critical to the Panel’s determination that Patrick Rando had not been afforded due process because he was not on notice that his conduct could result in termination. Instead, the court concluded that “[a] reasonable person in the grievant’s employment position would be on notice that violating the federally mandated drug policy set forth by Niagara Mohawk, adulterating a urine sample, falsifying a certification that he would not adulterate his urine sample, and falsely denying illegal drug use could result in termination.” 950 F.Supp. at 1236. Thus, after reviewing the same facts that had been considered by the Panel, the district court concluded first, that Rando’s conduct “supports a finding that he is neither trustworthy or reliable as required by the NRC regulatory scheme implemented by Niagara Mohawk[,]” and second, that his actions are “substantial evidence that [he] can no longer be considered trustworthy or reliable, at least to the degree required to hold the position of chemistry technician at a nuclear power plant.” Id. The district court thus concluded that the reinstatement of Patrick Rando “would contravene public policy requiring strict adherence to nuclear safety rules.” Id. Discussion We first address the law concerning judicial review of arbitration awards generally and under the public policy.exception, before discussing whether the district court’s vaca-tur was appropriate under the circumstances of this case. (1) It is well-settled that courts play a limited, role when asked to review the decision of an arbitrator made pursuant to a grant of authority to interpret a collective bargaining agreement. Courts are not empowered to reexamine the merits of an arbitration award, even though th.e parties to the agreement may argue that the award arises out of á misinterpretation of the contract or a factual error. See United Paperworkers Int’l Union v. Misco, Inc., 484 U.S. 29, 36, 108 S.Ct. 364, 369-70, 98 L.Ed.2d 286 (1987). As the Supreme Court stated almost forty years ago, “[t]he refusal of courts to review the merits of an arbitration award is the proper approach to arbitration' under collective bargaining agreements. The federal policy of settling labor disputes by arbitration would be undermined if courts had the final say on the merits of the awards.” United Steelworkers v. Enterprise Wheel & Car Corp., 363 U.S. 593, 596, 80 S.Ct. 1358, 1360, 4 L.Ed.2d 1424 (1960). Thus, an arbitrator’s award is legitimate and enforceable as long as it “draws its essence from the collective bargaining agreement” and is not merely an exercise of the arbitrator’s “own brand of industrial justice.” Id. at 597, 80 S.Ct. at 1361. The reasons for according such deference to arbitral decisions by insulating them from judicial review can be traced to the federal statutes which regulate labor-management relations. See Misco, 484 U.S. at 37, 108 S.Ct. at 370. The Labor-Management Relations Act of 1947 reflects a clear preference for the private resolution of labor disputes without government, intervention: “ ‘[F]inal adjustment by a method agreed upon by the parties is hereby declared to be the desirable method for settlement of grievance disputes arising over the application or interpretation of an existing collective-bargaining agreement.’” Id. (quoting 61 Stat. 154, 29 U.S.C. § 173(d)). In accordance with this deferential approach, the Supreme Court has unambiguously stated that a reviewing court is bound by both the arbitrator’s factual findings and his judgment regarding remedies. See Misco, 484 U.S. at 37-38, 108 S.Ct. at 370-71. It is abundantly clear that courts must tread lightly when reviewing arbitral decisions. That a court believes an arbitrator to have committed serious legal or factual error will not justify overturning his decision, provided that the arbitrator is “even arguably construing or applying the contract and acting within the scope of his authority.” Id. at 38, 108 S.Ct. at 371; see also Wackenhut Corp. v. Amalgamated Local 515 and Int’l Union, 126 F.3d 29, 32 (2d Cir.1997)(“The contractual theory of arbitration ... requires a reviewing court to affirm an award it views as incorrect — even very incorrect — so long as the decision is plausibly grounded in the parties’ agreement.”); Saint Mary Home, Inc. v. Service Employees Int’l Union, Dist. 1199, 116 F.3d 41 (2d Cir.1997); Denver & Rio Grande Western R.R. Co. v. Union Pac. R.R. Co., 119 F.3d 847, 849 (10th Cir.1997)(“Errors in either the arbitrator’s factual findings or his interpretation of the law (unless that interpretation shows a manifest disregard of controlling law) do not justify review or reversal on the merits of the controversy.”)(citing Misco, 484 U.S. at 36-38, 108 S.Ct. at 369-71). Against this background, we now examine a narrow exception to the deferential approach that generally characterizes judicial review of arbitration awards. In W.R. Grace & Co. v. Local Union 759, Int’l Union of United Rubber, Cork, Linoleum & Plastic Workers, 461 U.S. 757, 766, 103 S.Ct. 2177, 2183-84, 76 L.Ed.2d 298 (1983), the Supreme Court held that “a court may not enforce a collective-bargaining agreement that is contrary to public policy” and that “the question of public policy is ultimately one for resolution by the courts.” The Court stated that “[i]f the contract as interpreted by [the arbitrator] violates some explicit public policy, we are obliged to refrain from enforcing it.” Id. However, in W.R. Grace, the Court concluded that enforcement of the arbitration award at issue there did not compromise the public policy requiring obedience to court orders. In W.R. Grace, a damages award granted to employees for breach of seniority provisions under a collective bargaining agreement was held not to violate public policy, even though a court-ordered conciliation agreement signed by the employer conflicted with the seniority provisions of the collective bargaining agreement. See id. at 768-70, 103 S.Ct. at 2184-86. Emphasizing the limited nature of the public policy exception, the Supreme Court stated that a court’s refusal to enforce an arbitrator’s interpretation of a collective bargaining agreement is limited to situations where the contract as interpreted would violate “some explicit public policy” that is “well defined and dominant.” Id. at 766, 103 S.Ct. at 2183-84 (citing Muschany v. United States, 324 U.S. 49, 66, 65 S.Ct. 442, 451, 89 L.Ed. 744 (1945)). Four years later, in factual circumstances closer to those raised by this case, the Supreme Court again discussed the public policy exception, stating once more that “[a] eourt[] [may refuse] to enforce an arbitrator’s award under a colieetive-bargaining agreement because it is contrary to public policy.” United Paperworkers Int’l Union v. Misco, Inc., 484 U.S. 29, 42, 108 S.Ct. 364, 373, 98 L.Ed.2d 286 (1987). Nonetheless, the Court cautioned that its decision in W.R. Grace “does not ... sanction a broad judicial power to set aside arbitration awards as against public policy.” Id. at 43, 108 S.Ct. at 373-74. Rather, the Court stressed the limited parameters within which a court must operate when reviewing an arbitral award for possible violation of public policy: Because the parties have contracted to have disputes settled by an arbitrator chosen by them rather than by a judge, it is the arbitrator’s view of the facts and of the meaning of the contract that they have agreed to accept. Courts thus do not sit to hear claims of factual or legal error by an arbitrator as an appellate court does in reviewing decisions of lower courts. To resolve disputes about the application of a collective-bargaining agreement, an arbitrator must find facts and a court may not reject those findings simply because it disagrees with them. The same is true of the arbitrator’s interpretation of the con-tract____ So, too, where it is contemplated that the arbitrator will determine remedies for contract violations that he finds, courts have no authority to disagree with his honest judgment in that respect. If the courts were free to intervene on these grounds, the speedy resolution of grievances by private mechanisms would be greatly undermined. Misco, 484 U.S. at 37-38, 108 S.Ct. at 370-71 (emphasis added). Thus, in Misco, in a unanimous ruling that reversed a Fifth Circuit decision setting aside an arbitration award which ordered the reinstatement of an employee who had been suspended for suspected possession and use of marijuana on company property, the Misco Court emphasized that the decision in W.R. Grace “turned on our examination of whether the award created any explicit conflict with other laws and legal precedents.” Id. at 43, 108 S.Ct. at 373 (internal quotations omitted). It is evident that in referring to the “award,” the Misco Court was focusing on the final result of the arbitrator’s remedy, namely, whether the reinstatement itself “would actually violate” the public policy ‘•‘against the operation of dangerous machinery by persons under the influence of drugs or alcohol.” Id. at 44, 108 S.Ct. at 374 (internal quotations omitted). At no point did the Court evaluate or question the reasoning which led the arbitrator to fashion the remedy of reinstatement. Rather, the Court criticized the Fifth Circuit both for its reasoning on the merits, and for substituting its own judgment for that of the arbitrator. The Court expressed a clear intent to insulate the arbitrator’s fact-finding and reasoning from judicial scrutiny, stating that [I]t was inappropriate for the Court of Appeals itself .to draw the necessary inference ... [that the employee] had ever been or would be under the influence of marijuana while he was on the job____ [T]he fact that it is inquiring into a possible violation of public policy [does not] excuse a court for doing the arbitrator’s task. Id. at 44-45, 108 S.Ct. at 374-75. The Supreme Court’s emphasis on result, as opposed to the arbitrator’s reasoning, is consistent with the deference that must be accorded the arbitrator’s role in resolving disputes arising under collective bargaining agreements, while at the same time providing the courts with a necessary check on the compatibility of the chosen remedy with public policy. See id. at 44, 108 S.Ct. at 374 (“A refusal to enforce an- award must rest on more than speculation or assumption.”); see also Saint Mary, 116 F.3d at 46 (“[C]ourts may refuse to enforce arbitral awards only in those rare cases when enforcement of the award would be directly at odds with a well defined and dominant public policy resting on clear law and legal precedent.”); Stead Motors v. Automotive Machinists Lodge No. 1173, 886 F.2d 1200, 1212-13 (9th Cir.1989)(en banc)(“[A] faithful reading of Misco requires ... [that a] court must both delineate an overriding public policy rooted in something more than ‘general considerations of supposed public interests,’ and, of equal significance, it must demonstrate that the policy is one that specifically militates against the relief ordered by the arbitrator. ’’Xquoting Misco, 484 U.S. at 44, 108 S.Ct. at 374)(emphasis added). We, therefore, read the Supreme Court’s decision in Misco to mean that a court’s task in reviewing an arbitral award for .possible violations of public policy is limited to determining whether the award itself, as contrasted with the reasoning that underlies the award, “create[s][an] explicit conflict with other laws and legal precedents” and thus clearly violates an identifiable public policy. Misco, 484 U.S. at 43, 108 S.Ct. at 373-74 (internal quotations omitted). A court is not authorized to revisit or question the fact-finding or the reasoning which produced the award. Moreover, this approach is compatible with the well-recognized principle that arbitrators are not required to provide reasons for their awards. See Enterprise Wheel, 363 U.S. at 598, 80 S.Ct. at 1361-62 (“A mere ambiguity in the opinion accompanying an award ... is not a reason for refusing to enforce the award. Arbitrators have no obligation to the court to give then-reasons for an award. To require opinions free of ambiguity may lead arbitrators to play it safe by writing no supporting opinions.”)(footnote omitted). The Second Circuit’s recent articulation of the public policy exception fully comports with the result-oriented approach which in our view should govern a federal court’s review of an arbitration award on public policy grounds. In Saint Mary Home, Inc. v. Service Int’l Union, Dist. 1199, 116 F.3d 41 (2d Cir.1997), this Court upheld an arbitration award which ordered reinstatement of a nursing facility employee who had been discharged for possession of marijuana with intent to sell while on the job. This Court’s opinion began by reiterating the basic premise that a court “may not substitute its view for that of the arbitrator nor may it even review the award for clear error.” Id. at 44 (citing Leed Architectural Prods., Inc. v. United Steelworkers, Local 6674, 916 F.2d 63, 65 (2d Cir.1990)). It further stated that where an arbitrator explains his conclusions “in terms that offer even a barely colorable justification for the outcome reached, confirmation of the award cannot be prevented by litigants who merely argue, however persuasively, for a different result.” Saint Mary Home, 116 F.3d at 44 (quoting Andros Compania Maritima, S.A. and Marc Rich & Co., AG., 579 F.2d 691, 704 (2d Cir.1978)). Thus, Saint Mary Home rejected the argument that public policy categorically prohibits reinstatement in all cases where drug related conduct occurs in the workplace, finding instead that “the public policy relating to the response for drug related conduct in the workplace is flexible and remedial.” 116 F.3d at 46. The opinion also noted the'absence of an established and dominant public policy, as demonstrated through ease law or regulations, “prohibiting reinstatement after suspension of those convicted of drug offenses in the health care industry, one of the most heavily regulated industries in the country.” Id. at 47. And while the opinion in dicta stated that the case before the court was to be contrasted with .cases from other courts in which arbitral awards calling for reinstatement of employees to various safety-sensitive positions were vacated, (citing Union Pac. R.R. v. United Transp. Union, 3 F.3d 255 (8th Cir.1993)(railroad); Exxon Shipping Co. v. Exxon Seamen’s Union, 993 F.2d 357 (3d Cir.1993)(shipping); Gulf Coast Indus. Workers Union v. Exxon Co., 991 F.2d 244 (5th Cir.1993)(petro-chemical refinery); Georgia Power Co. v. International Bhd. of Elec. Workers, Local 34, 707 F.Supp. 531 (N.D.Ga.1989), aff'd 896 F.2d 507 (11th Cir.1990)(utility plant); Delta Air Lines, Inc. v. Air Line Pilots Ass’n, Int’l, 861 F.2d 665 (11th Cir.1988) (airline); Iowa Elec. Light & Power Co. v. Local Union 204 of the Int’l Bhd. of Elec. Workers, 834 F.2d 1424 (8th Cir.1987)(nuclear power plant)) the opinion, nonetheless, confirms that an appropriate analysis of the public policy question requires that a court focus on the outcome of the arbitrator’s interpretation of the contract— namely, whether the remedy imposed can be fairly and unequivocally shown to violate a well-established public policy. See Saint Mary Home, 116 F.3d at 46 — 47. Our reading of Supreme Court jurisprudence, as well as the most recent words of this Court, indicates that the approach does not change when the case involves the reinstatement of an employee to a safety-sensitive position in the nuclear power industry, as does the case immediately before us. In determining whether an arbitral award which orders reinstatement of an employee previously discharged from a nuclear power plant violates public policy, the issue should be whether the reinstatement itself specifically contravenes a “well defined and dominant public policy.” Misco, 484 U.S. at 44, 108 S.Ct. at 374. As we have already demonstrated, such an analytical framework is the only one which can adequately reconcile the competing interests at stake here; namely, the deference that must be accorded an arbitrator’s judgment, on the one hand, and the substantial public interest in nuclear safety, on the other. (2) Having identified the proper framework which a district court should employ when reviewing an arbitral award for possible violation of public policy, we must now determine whether the conditional reinstatement of an employee to a highly safety-sensitive position in a nuclear power plant necessarily violates public policy, where that employee adulterated a drug test, subsequently tested positive for drugs, and denied having a substance abuse problem when initially confronted with the test results. There can be no doubt, nor does the Union dispute, that there exists a strong public policy in favor of promoting a safe, drug-free working environment in the nuclear power industry. We agree with the district court that the existence of extensive federal legislation governing safety in the nuclear power context, as well as Congressional authorization for a specific agency, the Nuclear Regulatory Commission, to oversee and ensure nuclear safety, reflects “the level of public concern over the safety of the nuclear power industry.” International Bhd. Of Elec. Workers, Local 97 v. Niagara Mohawk Power Corp., 950 F.Supp. 1227, 1232 (N.D.N.Y.1996). Observing that the NRC regulations represent “a strict regulatory scheme devised by Congress for the protection of the public from the hazards of nuclear radiation[,]” the district court correctly concluded that “[t]here is a dominant and well-defined policy ‘requiring strict adherence to nuclear safety rules.’” Id. (quoting Iowa Elec. Light & Power Co. v. Local Union 204, IBEW, 834 F.2d 1424, 1427-28 (8th Cir.1987)). See also Daniel Constr. Co. v. Local 257, IBEW, 856 F.2d 1174, 1182 (8th Cir.1988). The question before us is thus whether the award ordering the conditional reinstatement of Patrick Rando necessarily violates public policy concerning nuclear safety. We find that it does not. A close analysis of the Nuclear Regulatory Commission regulations, arguably the most persuasive evidence of the public policy at issue in this case, indicates that these regulations do not proscribe the reinstatement of employees previously found to have adulterated a drug test or who have tested positive for drugs or alcohol. First, given the comprehensiveness of the regulations, it is significant that they do not discuss adulteration, let alone specify any penalty or sanction for adulterating a drug test. Second, the NRC regulations require only a relatively minimal response when nuclear power plant operators learn that an employee has failed a drug test for the first time. Specifically, 10 C.F.R. § 26.27 requires licensees (i.e., operators of nuclear power plants subject to NRC regulations) to “take the following actions ... [A first positive] must, as a minimum, result in immediate removal from activities [within the facility] for at least 14 days and referral to the EAP for assessment and counseling during any suspension period.” 10 C.F.R. § 26.27(b)(2). The lack of any mention of adulteration and the fact that the NRC regulations specify only a two-week minimum sanction for a first positive test result is powerful evidence that the regulations themselves do not contemplate dismissal as the ordinary penalty for an initial drug offense. To the contrary, § 26.27 clearly contemplates the notion of both rehabilitation and reinstatement: Plans for treatment, follow-up, and future employment must be developed, and any rehabilitation program deemed appropriate must be initiated during such suspension period. Satisfactory management and medical assurance of the individual’s fitness to adequately perform activities within the scope of this part must be obtained before permitting the individual to be returned to these activities. 10 C.F.R. § 26.27(b)(2). Indeed, the favorable attitude of the regulations toward rehabilitation and reinstatement is illustrated by this same provision, which begins by stating that “[Hacking any other evidence to indicate the use, sale, or possession of illegal drugs onsite, a confirmed positive test result must be presumed to be an indication of offsite drug use.” Id. This is significant because of the sharp distinction which the regulations draw between onsite and offsite use of drugs. The former leads to a minimum five year denial of access to the plant, 10 C.F.R. § 26.27(b)(3), whereas, off-site use of drugs requires only a minimum two-week suspension. 10 C.F.R. § 26.27(b)(2). Surely, if rehabilitation was not a significant consideration, the presumption would work the other way, placing the burden on the employee to demonstrate that he or she had not used drugs onsite before being allowed to return to the premises. It is thus evident that nothing within the NRC regulations prohibits the re-employment of an employee who yields a first positive on a drug test, provided that adequate assurance of the employee’s rehabilitation is obtained. Furthermore, as the NRC regulations do not speak to the issues of either adulteration or false- denials, there 'is no evidence that the NRC intended to deny the possibility of rehabilitation and re-employment to an employee who initially falsely denies that he has a drug problem, or to bar the eventual reinstatement of an employee found to have adulterated a drug test or to have falsely certified on a consent form that he would not adulterate the test. Moreover, the Company’s own Nuclear Division Directive specifies that an incident of adulteration is to be treated as a “first' positive,” an offense meriting only a two week suspension under the Company’s Fitness for Duty Program. See Nuclear Division Directive, § 3.2.3. Given the NRC’s express contemplation of eventual reinstatement as a viable response to evidence’ of offsite'drug use by an employee and its silence on the issue of adulteration, we do not see how the district court could conclude that the conditional' reinstatement of Patrick Rando constitutes a clear violation of the public policy favoring a safe nuclear work environment. See DiRussa v. Dean Witter Reynolds, Inc., 121 F.3d 818, 825 (2d Cir.1997), cert. denied, — U.S. -, 118 S.Ct. 695, 139 L.Ed.2d 639 (1998)(“The party seeking to prevent enforcement of the award must ‘clearly fehów[ ]’ a violation of public poliey.”)(quoting United Paperworkers Int’l Union v. Misco, Inc., 484 U.S. 29, 43, 108 S.Ct. 364, 373-74, 98 L.Ed.2d 286 (1987)); Saint Mary Home, Inc. v. Service Employees Int’l' Union, Dist. 1199, 116 F.3d 41, 45 (1997). The Panel here ordered that Rando be reinstated “contingent on his producing a negative drug test and satisfying any requirements that may be imposed by the Employee Assistance Program after an evaluation.” 0 & D at 14. The award also provided for follow-up testing of Rando at the Company’s discretion for an eighteen month period. Id. Thus, the Panel’s award contains án implicit recognition that if an employee is to be reinstated to his former position, it must be in a manner consistent with NRC regulations and protective of public safety. The award fashioned by the Panel meets that test. By conditioning Rando’s reinstatement on satisfactory proof of his drug-free, rehabilitated status, the Panél specifically sought to prevent a situation where nuclear safety might be compromised in the future. As such,- it cannot be said that enforcement of the Panel’s award providing for the conditional reinstatement of Rando “would actually violate the public policy” favoring a safe nuclear work environment. Misco, 484 U.S. at 44, 108 S.Ct. at 374. (3) The Company makes two principal arguments on appeal; first, that the public policy relating to nuclear safety includes a mandate that employees granted unescorted access to nuclear fácilities be trustworthy and reliable, and second, that because Rando’s actions render him untrustworthy within the meaning of these regulations, the arbitration panel’s award reinstating Rando necessarily violates public policy. To support its argument, the Company points to two provisions of the NRC regulations. The first of these provisions provides that individuals granted unescorted access to protected areas within the nuclear power plant must be “trustworthy and reliable” and that they “not constitute an unreasonable risk to the health and safety of the public.” 10 C.F.R. § 73.56(b)(1). The Company also relies on a provision in the regulations dealing with Fitness for Duty Programs, which articulates as one of three objectives for these programs: to provide reasonable assurance that nuclear power plant personnel “will perform their tasks in a reliable and trustworthy manner.” 10 C.F.R. § 26.10(a). The Company argues that Ran-do proved himself to be dishonest and untrustworthy by “[lying] on his consent form, adulterating his urine sample in a premeditated fashion and [lying] about his actions and drug use when [initially] confronted.” 0 & D at 7. The district court, in reviewing the provisions of the NRC regulations which require that employees of nuclear power plants be trustworthy and reliable, determined that these requirements should be read into the public policy at issue in this case. This was an appropriate exercise of the court’s reviewing authority, as “the question of public policy is ultimately one for resolution by the courts” and is to be ascertained “by reference to ... laws and legal precedents.” W.R. Grace & Co. v. Local Union 759, Int’l Union of Rubber, Cork, Linoleum, & Plastic Workers, 461 U.S. 757, 766, 103 S.Ct. 2177, 2183-84, 76 L.Ed.2d 298 (1983). After identifying the relevant public policy, the district court then essentially concluded that any employee who acts as Rando did (i.e., who adulterates a drug test, falsely certifies that he would not adulterate his urine sample, and falsely denies illegal drug use) is not trustworthy and reliable. We would agree that Rando’s actions were dishonest, but in our view, this determination does not end the discussion. Assuming that the public policy relating to nuclear safety requires in part that employees of nuclear power plants be trustworthy and reliable, the real question is whether Rando’s dishonest acts render him an inherently untrustworthy person within the meaning of the NRC regulations when read as a whole. The only basis on which we or the district court could overturn the Panel’s award would be an unequivocal showing that Rando’s actions make him untrustworthy and unreliable within the meaning of the regulations. The central question in this inquiry is, therefore, the meaning of the regulations. In attempting to discern the meaning and scope of the NRC provisions dealing with the trustworthiness and reliability requirements, we find it significant that nowhere do they (or any other NRC provisions, for that matter) mention the offenses of adulteration or false. denials of drug use. Furthermore, §§ 26.10 and 73.56 neither define the terms “trustworthy” and “reliable” nor do they offer examples of conduct which would place an employee in violation of these provisions. This, of course, is not determinative. The regulations must be cqnsidered in their entirety in order to determine whether Rando’s behavior constitutes a violation of the NRC requirement that employees be trustworthy and reliable. ,. . . We do not believe that we can fairly read these regulations to say that Rando’s conduct in connection with his drug test necessarily violates the provisions which require that employees be trustworthy and reliable. Put another way, the NRC regulations, both those dealing with the trustworthy and reliable requirement, as well as the regulatory scheme as a whole, cannot accurately be interpreted to preclude a finding that someone who acts as Rando acted could, nonetheless, be trustworthy and reliable as those terms are used by the Nuclear Regulatory Commission. Indeed, surprising as it máy seem, it is fair to say that the NRC regulations actually contemplate the rehabilitation of employees who act dishonestly, as Rando did. For the reasons we have earlier discussed, it is clear that the NRC regulations contemplate rehabilitation as a significant goal; they do so even at the risk of compromising public safety to some degree. It is quite apparent that the regulations do not limit the availability of rehabilitation only to those employees who voluntarily come forward and confess to their-employer- that they have a drug problem. Rather, a close examination of the regulations indicates that the NRC’s rehabilitative approach to substance abuse is also applicable to employees who seek to conceal their drug problems, either by remaining quiet and hoping that a spot drug test will, not expose their secret, or by adulterating a drug test and lying about having a drug problem. This is apparent because of the presumption that absent'evidence to the contrary, a positive test result indicates off-site drug use. In terms of public safety, it does not matter whether a nuclear employee takes drugs on plant premises or just before he comes to work. Had the NRC truly intended the elimination of drug use among nuclear employees as its chief goal, the presumption would operate the other way, placing the burden on the employee to show, at the very least, that a positive drug test was the result of óff-site drug use and that his drug use presents no danger to the public. Thus, much as we might desire to agree with the district court’s analysis, we cannot fairly read the two provisions in the NRC regulations requiring that nuclear employees be trustworthy and reliable to mean that an employee who adulterates a drug -test and falsely denies having a drug problem when initially confronted is categorically incapable of being regarded as trustworthy and reliable. This is particularly true where such an employee’s actions are an inherent and inseparable part of his underlying drug problem, as is also the case with an employee who does not affirmatively step forward to admit his problem. Despite the dishonest nature of Rando’s acts, those acts can be viewed as symptomatic of his drug problem, a problem for which the NRC regulations clearly contemplate rehabilitation as a viable management response. His behavior, while reprehensible, does not brand him an untrustworthy person as a matter of law. Nor can -we fairly read the two provisions on which the Company relies as overriding the NRC’s relatively, sympathetic attitude toward the problem of drug .abuse and its overall favorable approach to rehabilitation, even at the risk of having employees with undisclosed drug problems working at nuclear power plants. Accordingly, the district court erred in concluding that Rando’s conditional reinstatement was in conflict with public policy as defined by the NRC regulations. Our conclusion is not changed by the Company’s argument that there need not be an explicit law or governmental regulation prohibiting the remedy granted in an arbitration award for the award to be vacated on public policy grounds. In construing the public policy at issue in this case, the district court observed in a footnote that “[although public policy is to be determined by examining statutes and regulations, an arbitration award need not violate positive law or regulations before vacatur is appropriate.” International Bhd. of Elec. Workers, Local 97 v. Niagara Mohawk Power Corp., 950 F.Supp. 1227, 1234, n. 13 (N.D.N.Y.1996). In response, the Union argues that the award can be set aside “only if the arbitration panel’s interpretation of the Agreement ... violate[s] some explicit public policy embodied in [the NRC] regulations.” Appellant’s Brief, p. 31. The Company directs the Court’s attention to the fact that only two circuits, the District of Columbia and the Ninth Circuit, have adopted this more restrictive view. See, e.g., Local 588 v. Foster Poultry Farms, 74 F.3d 169, 174 (9th Cir.1995); Stead Motors v. Automotive Machinists Lodge No. 1173, 886 F.2d 1200 (9th Cir.1989)(en banc); Northwest Airlines, Inc. v. Air Line Pilots Ass’n, Int’l, 808 F.2d 76, 83-84 (D.C.Cir.1987); Postal Workers v. Postal Serv., 789 F.2d 1, 8 (D.C.Cir.1986). We need not adopt this narrow view because even the courts that have adopted a broader view have, nonetheless, drawn upon federal regulations where they exist as the principal source for the public policy to be applied. See Exxon Shipping Co. v. Exxon Seamen’s Union, 993 F.2d 357, 363-64 (3d Cir.1993); Iowa Elec. Light & Power Co. v. Local Union 204 IBEW, 834 F.2d 1424, 1429 (8th Cir.1987); Union Pacific R.R. v. United Transp. Union, 3 F.3d 255, 261-62 (8th Cir.1993). Indeed, the Company concedes that “[the] extensive regulation and involvement by the federal government in the area of nuclear safety obviously provides the source of a