concurring in part and dissenting in part.
After examining and dismissing Ameri-tech’s proposed justifications for admitting its collective bargaining agreement (“CBA”) evidence on the issue of liability, the majority sets forth that, from this day forward, CBAs can never be admitted in any Title VII dispute on the issue of liability. I believe that the adoption of such a sweeping dictate represents a marked departure from our evidentiary jurisprudence, and may in the future drastically alter the manner in which Title VII cases are adjudicated. In an area of developing law, where we should proceed with due caution, in my view the majority has vaulted forward, cementing a categorical rule that is without warrant and perhaps counterproductive. Therefore, while I join Parts I and III of the majority’s decision, I must respectfully dissent from Part II.
I.
That Amos’s actions were reprehensible, no rational person could dispute. Like*532wise, it is not debatable that Amos deserved to be discharged well before he was. However, as Judge Posner correctly points out, our obligation in an en banc setting is greater than any single dispute. While meting out justice on an individual case basis, it is critical that we remain vigilant of the manner in which any rule we articulate will be employed in the future. Thus, we must to a degree detach ourselves from the disturbing facts of this case, admittedly a task not easily accomplished, and focus on the overriding legal question. Once we put aside our disgust over totally inappropriate conduct, this case devolves to an evidentiary issue of considerable significance. As we articulated in anticipation of this en banc review, that issue is “whether, as a matter of law, evidence regarding arbitration and a company’s collective bargaining agreement is inadmissible in a Title VII suit to show ... that an employer’s response to sexual harassment was reasonable for the purpose of determining employer liability.”
In response, the majority flatly states that such “evidence is not relevant to liability and therefore is inadmissible under Fed.R.Evid. 402.” In making that determination, the majority opinion emphasizes that none of the proffered reasons for which Ameritech sought to introduce the CBA could have been “sufficient to defeat liability under Title VII”; and I do not dispute that assertion. It is uncontested that employers cannot use collective bargaining agreements to contract around anti-discrimination laws like Title VIL See Trans World Airlines, Inc. v. Hardison, 432 U.S. 63, 79, 97 S.Ct. 2264, 53 L.Ed.2d 113 (1977). However, Ameritech is not seeking to use its obligations under the CBA as an affirmative defense precluding liability under Title VII. Rather, Ameri-tech seeks only to introduce evidence as to its obligations under its CBA to show that its response to Amos’s conduct was reasonable. As we have stated in other areas, “[rjeasonableness depends upon the circumstances presented in a given situation and upon balancing the public, governmental, and private interests at stake in that situation.” Shields v. Burge, 874 F.2d 1201, 1204 (7th Cir.1989). The adoption of a categorical rule of exclusion in an area of case-specific analysis seems incongruous. Indeed, the majority’s explication that Am-eritech’s CBA evidence was insufficient to defeat Title VII liability would, at most, justify only a finding that the exclusion of that evidence during trial was of an inconsequential nature. However, the majority rushes onward, stating that because Amer-itech’s proffered reasons for introducing the CBA evidence would not have been sufficient to defeat liability under Title VTI, the evidence was irrelevant as a matter of law, and thus subject to exclusion under Federal Rule of Evidence 402.1
Yet we, nor any other court, have ever supposed that because a piece of evidence, standing alone, could not operate to negate liability, that such evidence has no part in a trial. Rule 401 of the Federal Rules of Evidence sets a much more liberal stan*533dard. Under that Rule, evidence is relevant if it has “any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence.” Fed.R.Evid. 401 (emphasis added). This definition of relevant evidence is an expansive one. To be relevant, evidence need not conclusively decide the ultimate issue in a case, nor make the proposition appear more probable, “but it must in some degree advance the inquiry.” 1 J. Weinstein & M. Berger, Weinstein’s Federal Evidence § 401.04[2][b]; see also Davis v. Lane, 814 F.2d 397, 399 (7th Cir.1987). As we have noted, this “is not a difficult standard to meet.” United States v. Brisk, 171 F.3d 514, 525 (7th Cir.1999). For that reason, relevance under Rule 402 must be construed liberally, Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 587, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), and the “[e]xclusion of relevant evidence should be used sparingly.” Morgan v. United Air Lines, Inc., 750 F.Supp. 1046, 1055 (D.Colo.1990); see also Gentile v. County of Suffolk, 926 F.2d 142, 151 (2d Cir.1991). Here therefore, the question should not be whether Ameritech’s obligations under the CBA made its actions reasonable, but rather whether Ameri-tech’s obligations under the CBA assist us in that determination.
In fact, our definition of what constitutes relevant evidence has been expanded even further beyond the lax standard articulated in the Federal Rules. For example, even when a piece of evidence did not technically make a fact of consequence more or less probable, we have allowed the entry of that evidence when it fills the gap that its exclusion might create. “Evidence is relevant if its exclusion would leave a chronological and conceptual void in the story.” Wilson v. Groaning, 25 F.3d 581, 584 (7th Cir.1994) (“Leaving out this evidence would have left the jury with an unduly sanitized and incomplete version of the facts.”); see also United States v. Vretta, 790 F.2d 651, 655 (7th Cir.1986). Similarly, evidence is often considered relevant background information when it permits a jury to better understand the context of an employer’s actions. See Fed.R.Evid. 401 advisory committee’s note; cf. Old Chief v. United States, 519 U.S. 172, 186-89, 117 S.Ct. 644, 136 L.Ed.2d 574 (1997); Blue Cross and Blue Shield of N.J., Inc. v. Philip Morris, Inc., 138 F.Supp.2d 357, 365-72 (E.D.N.Y.2001) (Weinstein, J.). In this instance, faced with voluminous evidence of Amos’s lamentable conduct, the jury no doubt wondered why Ameritech did not terminate Amos sooner. Indeed, its determination on liability may well have hinged on that fact. Ameritech’s CBA evidence, while by no means exculpating the company from Title VII liability, would have provided the jury with the motivation behind Ameritech’s delay. Any particular disciplinary decision is much more likely to appear reasonable if an employer is able to explain the considerations that led it to that choice, including the disciplinary system that it has formulated to address employee infractions. Our sister circuits have recognized this and have considered evidence of an employer’s own internal investigatory and disciplinary procedures in determining whether its response to harassment was reasonable. See Curry v. District of Columbia, 195 F.3d 654, 660 n. 13 (D.C.Cir.1999); Dhyne v. Meiners Thriftway, Inc., 184 F.3d 983, 988 (8th Cir.1999); Waymire v. Harris County, Tex., 86 F.3d 424, 429 (5th Cir.1996). Of course, the jury in this case could still have found Ameritech’s actions to be blameworthy, and even determined that Ameritech’s proffered motivation made it more probable that the company acted unreasonably rather than less. However, that decision *534would have been made with the benefit of all relevant evidence.
II.
In support of its decision, the majority asserts that a CBA is a consensual agreement, akin to any private contract, and that employers cannot by their consent avoid duties under federal law. I do not quarrel with the latter proposition. As for the former, while the supposition that a CBA is tantamount to a private contract for Title VII does have some superficial appeal, it ignores the undeniably unique federal character of CBAs.
Unlike private employment contracts, the creation and interpretation of CBAs are governed by federal law. A large body of decisions from both the federal courts and the National Labor Relations Board (“NLRB”) place significant restraints on the kinds of negotiating tactics that both employers and unions can utilize. While negotiating a CBA, an employer must be concerned that its actions will form the basis of a charge causing it to be disciplined by the NLRB for committing an unfair labor practice, a concern notably absent in most contract negotiations.
Federal law clearly shapes and restricts the menu of terms a unionized firm will be able to obtain in a CBA. In negotiating most contracts, a party has the very important ability to walk away from the bargaining table, or threaten to do so. If the opposite party is taking an unreasonable position, one can seek more agreeable partners and better terms elsewhere. Not so for the unionized employer. It is required by federal law to negotiate with the union elected by its workers, forbidden by those laws from bargaining with another union or directly bargaining with its actual or potential employees. See United Steelworkers of Am. v. Warrior & Gulf Navigation Co., 363 U.S. 574, 580, 80 S.Ct. 1347, 4 L.Ed.2d 1409 (1960) (recognizing that, unlike regular contract negotiations where parties have “no real compulsion to deal with one another, as opposed to dealing with other parties,” unionized employers are locked into a relationship with a particular union). Moreover, the incentives of a recognized union, which wishes to show its value to the workers it represents, will lead it to bargain hard for increased job security, see Duffy Tool & Stamping, L.L.C. v. NLRB, 233 F.3d 995, 998-99 (7th Cir.2000), which will normally involve demanding (and receiving) a just cause provision, see Estelle D. Franklin, Maneuvering Through the Labyrinth: The Employers’ Paradox in Responding to Hostile Environment Sexual Harassment — A Proposed Way Out, 67 Fordham L. Rev. 1517, 1561 n. 203 (“Ninety-four percent of all collective bargaining agreements contain just cause or equivalent clauses.”), backed by arbitration, see Lingle v. Norge Div. of Magic Chef, Inc., 486 U.S. 399, 411 n. 11, 108 S.Ct. 1877, 100 L.Ed.2d 410 (1988) (“Arbitrators are delegated by nearly all collective-bargaining agreements as the adjudicators of contract disputes.”).
One of the primary purposes of the federal labor laws imposing collective bargaining was to achieve provisions such as the just cause and arbitration clauses to which Ameritech is subject. See United Steelworkers of Am. v. Enterprise Wheel & Car Corp., 363 U.S. 593, 596, 80 S.Ct. 1358, 4 L.Ed.2d 1424 (1960) (noting “federal policy of settling labor disputes by arbitration”); Warrior & Gulf, 363 U.S. at 578, 80 S.Ct. 1347 (“The present federal policy is to promote industrial stabilization through the collective bargaining agreement. A major factor in achieving industrial peace is the inclusion of a provision for arbitration of grievances in the collective bargaining agreement.”) (citation and footnote omitted); id. at 582, 80 S.Ct. 1347 (discuss*535ing, in the context of labor relations governed by CBAs, “congressional policy in favor of settlement of disputes by the parties through the machinery of arbitration”).
Given the federally-influenced constraints that a CBA imposes on a company in disciplining employees, it would be inexact to liken a CBA to a voluntarily-adopted private employment contract. As part of its explanation for why it did not react with greater asperity or severity in response to Amos, Ameritech should have been permitted to expound on how its obligations stemming from federal labor law impacted the specific actions that were reasonable for it take in attempting to comply with Title VII. The jury would have been free to decide that even with the constraints of the CBA, Amos’s harassment was severe enough that Ameritech failed to exercise reasonable care by not firing Amos sooner and thus violated Title VII. However, the jury should have been permitted to consider how Ameritech’s obligations resulting from the federal law impact the content of its duty to exercise reasonable care to prevent co-worker harassment under Title VII.2
III.
Relying on the Court’s decisions in Los Angeles Department of Water & Power v. Manhart, 435 U.S. 702, 98 S.Ct. 1370, 55 L.Ed.2d 657 (1978) and Automobile Workers v. Johnson Controls, Inc., 499 U.S. 187, 111 S.Ct. 1196, 113 L.Ed.2d 158 (1991), the majority states that a cost-based justification is not available in Title VII actions. I do not dispute the accuracy of that statement in the context of the cases relied upon, and the hypotheticals articulated by the majority. Undoubtably, a corporate defendant is strictly liable for acts that can fairly be said to be acts of the company itself, such as torts committed by employees acting in the scope of employment. See Restatement (Second) of Agency § 219(1). Likewise, when the company’s management, to whom the formulation of corporate policy is entrusted, enacts a company policy that discriminates against women, this is considered an act by the company itself. Thus, under traditional agency principles, the corporation is strictly liable for the effects of such a decision and will not be heard to complain that costs or even beneficence motivated the discriminatory policy.
But, those agency principles for imposing vicarious strict liability on the company do not apply in the cases of co-worker harassment. Sexual harassment is not within the scope of the harasser’s employment. See Faragher v. City of Boca Raton, 524 U.S. 775, 798-800, 118 S.Ct. 2275, 141 L.Ed.2d 662 (1998). Thus, a co-worker does not possess the kind of company authority to make a company strictly liable for that individual’s harassment. It is for that reason that men or women victimized by co-worker sexual harassment can recover from an employer only by showing that the employer was negligent. See id. at 799, 118 S.Ct. 2275 (noting that circuit courts “uniformly judg[e] employer liabili*536ty for co-worker harassment under a negligence standard”).
Costs have long been a consideration in the question of whether a person or company was negligent, see United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir.1947) (L. Hand, J.), and to say to the contrary, as the majority does today, is to depart from what has been implicit in our decisions up until this point. In Zimmerman v. Cook County Sheriff’s Department, 96 F.3d 1017, 1019 (7th Cir.1996), we noted that an employer need not adopt an “Orwellian program of continuous surveillance,” even though it is undeniable that employees would engage in less harassment if they were constantly watched. Likewise, for hostile environments that are not too severe, remedies such as warnings, probation, or transfers are reasonable, see Savino v. C.P. Hall Co., 199 F.3d 925, 936 (7th Cir.1999); McKenzie v. Illinois Dep’t of Transp., 92 F.3d 473, 480 (7th Cir.1996); Baskerville v. Culligan Int’l Co., 50 F.3d 428, 432 (7th Cir.1995); accord Adler v. Wal-Mart Stores, Inc., 144 F.3d 664, 676 (10th Cir.1998); Harris v. L & L Wings, Inc., 132 F.3d 978, 984 (4th Cir.1997); Knabe v. Boury Corp., 114 F.3d 407, 414 (3d Cir.1997), even though less harassment would occur if the company took the costly step of summarily firing every verified harasser, or even every employee who was accused of harassment. The underlying theme in these decisions is that an employer will not be held liable for failing to take unreason ably burdensome measures to prevent co-worker harassment, even when the adoption of those measures would guarantee that harassment would not take place.
If one concludes that a company could drastically reduce (if not eradicate) harassment in the workplace if it spent an unlimited amount of resources on the problem (e.g., pervasive screening, daily training, constant monitoring), and that the costs of such measures are irrelevant for Title VII purposes, then I fail to see how a company could ever defend against liability if it did not fully implement all conceivable measures. Thus, despite paying lip service to the concept of negligence, I am concerned that the majority may have insured that companies in future Title VII suits will be judged by a standard more akin to strict liability. One can only conjecture as to whether the imposition of such a liability regime will result in increased or decreased employer expenditures on preventing (and responding to) sexual harassment in the workplace.
Despite its assertion that cost-based justifications met their Waterloo in Manhart, the majority nonetheless acknowledges that “costs matter to reasonableness.” However, the majority seeks to cabin the import of that acknowledgment by stating that while costs may factor into a reasonableness determination, those “additional” costs which an employer voluntarily elects to bear are inadmissible in Title VII liability disputes. While this apparent demarcation may appear attractive, the waters become muddied the instant one attempts to discern what constitutes an additional, voluntarily-elected cost as opposed to a cost that all employers must face. For instance, in an effort to prevent sexual harassment, an employer will adopt a -set of administrative procedures, tailored to that company, for detecting, and responding to, employee co-worker harassment. If an employee were to sue the company, alleging that it did not discharge a harasser with proper rapidity, it is unclear whether the company would be able to explain how the procedures it chose to adopt to deal with harassment influenced the timeliness of its response. Surely, the specific measures a company has chosen to implement constitute additional, voluntarily-elected costs. But, as the Fifth Circuit *537has noted, “in analyzing the promptness of response it is important to keep in mind the entity’s lines of command [and] organization format.” Waymire, 86 F.3d at 429 (emphasis added and citation omitted). Of course, it is possible to argue that every company must establish guidelines for dealing with harassment, and thus, an individual employer’s procedures are not an additional, voluntarily-elected cost, and are therefore admissible. What becomes apparent is that, depending on the level of abstraction employed, most any cost incurred in the prevention of sexual harassment may be designated as an additional, voluntarily-elected cost, or a cost faced by all employers. While it is impossible to foresee how the majority’s approach will impact future Title VII disputes, I fear the result will be random and inconsistent trial-level admissibility determinations, based largely on litigants’ abilities to characterize costs.
IV.
Judge Posner’s concurrence provides what some might consider an attractive middle ground. While it criticizes the expansive rule articulated by the majority, it finds any error in the district court’s decision to be harmless. Ultimately, I believe that the district court did in fact err, and I cannot state unequivocally that such error was harmless. Here, the decision of the district court left Ameritech with two alternatives: one unappealing, the other disingenuous. Ameritech was forced to assert either that Amos’s conduct did not violate Title VII or that the company was not on notice of the conduct. It is certainly possible, perhaps even probable, that had the jury heard the evidence of Ameri-tech’s obligations under the CBA, it still would have found the company guilty. However, in the end, that is a judgment call on which I respectfully disagree with Judge Posner. The district court’s decision to exclude the CBA evidence effectively tied the hands of Ameritech, leaving the company with no defense to raise. Under those circumstances, I cannot conclude that the district court’s decision was harmless.
I firmly believe that the majority opinion’s broad stroke today will prove to be a harmful impediment to litigants in future Title VII cases. The majority has proposed numerous hypotheticals, in all of which, CBA evidence would be inadmissible. However, no matter how many vignettes are raised in which the evidence would be deemed inadmissible, that does not justify the adoption of a categorical rule. Perhaps a case will seldom arise where such evidence is relevant in a liability setting. Yet that is no justification to rush towards an absolute precept. Given Amos’s egregious conduct and Ameritech’s all too deliberate pace in disciplining him, few may quibble with the application of the majority’s approach in this case. However, I remain concerned with the case that we will face in the future and the way this decision will be applied. Whether the rule articulated by the majority will prove to be beneficial or deleterious in future cases is difficult to foretell. I simply believe that neither our case law nor prudence justifies such a step. Therefore, I must respectfully dissent from Part II of the majority’s opinion.
. According to Rule 402, "[a]ll relevant evidence is admissible, except as otherwise provided by the Constitution of the United States, by Act of Congress, by these rules, or by other rules prescribed by the Supreme Court pursuant to statutory authority. Evidence which is not relevant is not admissible.” This rule establishes two types of evidence which are inadmissible: relevant evidence that is ex-cludable pursuant to requirements of the Constitution, Congress, etc., and evidence that is irrelevant, as defined in Rule 401. The majority decision has deemed the CBA evidence here to be "irrelevant on the issue of liability.” Thus, the majority must believe that this evidence is excluded because the CBA does not meet the definition of relevant evidence. That being so, I fail to see how an absolute rule of exclusion is appropriate in this area.
. It is important to emphasize once again that Ameritech does not argue that an arbitration provision in a CBA waives an employer’s obligations under Title VII to prevent sexual harassment in contravention of Alexander v. Gardner-Denver Co., 415 U.S. 36, 51, 94 S.Ct. 1011, 39 L.Ed.2d 147 (1974). Rather, it only wishes to introduce the CBA to assist in convincing the jury that its actions constituted reasonable care. As stated above, other circuits have considered evidence of an employer's own internal investigatory and disciplinary procedures in determining whether a response to harassment was reasonable. The fact that this defendant's apparatus for imposing discipline is embodied in a CBA is no reason for denying it the opportunity to explain its action that other defendants receive.