dissenting,
joined by DeMOSS, STEWART, and DENNIS, Circuit Judges:Despite the utmost esteem in which I hold my colleagues of the en banc majority, I am constrained to dissent. In my view, the majority:
(1) Understates how substantially Herrera’s speech in this case involves matters of public concern;
(2) Overstates the significance of the anti-adornment facet of the employer’s uniform policy in advancing the Hospital’s interest in workplace efficiency;
(3) Ignores some factors pertinent to Herrera’s individual speech interest, and undervalues others, when conducting the Connick/Pickering balancing test;
(4) Inadequately distinguishes the facts and circumstances peculiar to this civilian employer/maintenance-crew employee case from those of the cases proffered as analogs by the majority, involving (a) law enforcement and paramilitary employees, and (b) other sensitive- or confidential-relationship employers;
(5) Fails to parse the Hospital’s total employee pool and focus only on the sub-group of non-healthcare, blue-collar custodial, maintenance, food preparation, and clerical workers for the purpose of weighing the competing interests of free speech and workplace efficiency;
(6) Obverts the effect Texas’s prohibition of a public employer’s recognition of a union and bargaining collectively with it;
(7)For good measure, tosses out a parade of horribles that it speculates would result from a holding in favor of Herrera.
Mindful of the imperative to “keep one’s eyes upon the doughnut and not upon the hole,” I caution all to remain constantly aware that it is not the Hospital’s uniform policy vel non that Herrera violated; he wore the prescribed uniform at all times. Rather, it is the Hospital’s additional prohibition of the wearing of any adornment on those mandatory uniforms — as that restriction is applied to Herrera and his fellow custodial and clerical co-workers— that infringes his First Amendment right to freedom of expression.
1.
Herrera’s Speech and “Matters of Public Concern”
As the majority ultimately acknowledges that Herrera’s expression did indeed address a matter of public concern, I need not comment on each incremental step taken (or not taken) to reach this unavoidable conclusion. I am compelled, however, to flag the majority’s “damning with faint praise” the degree of public concern that Herrera’s workplace speech exhibited.
a. Public Concern or Personal Interest?
First, the majority advances that, because Herrera’s speech occurred in the workplace and was at least implicitly related to his employment, it primarily concerned matters of his own “personal interest.” I must take issue with this bit of overbroad mis-direetion: We have expressly held that speech may warrant protection, even if it occurs only in the workplace.1 And, it is a given that at least a *444modicum of personal interest will inhere in virtually all employee speech, regardless of whether uttered while the speaker is on or off the clock. Herrera concededly had some individual interest in supporting the union organizing drive when he wore the “Union Yes!” button on his uniform. But — at least on this record — he cannot be held to have had any more of a personal stake in the organizing effort than any other similarly situated employee of the Hospital. That he was one of the organizers makes no difference.
In holding that Herrera’s personal interest predominated over the public concern that he addressed, the majority focuses too narrowly on what it perceives to be the motivation for his speech, ignoring his obvious and overarching institutional interest in the hoped-for result of the union organizing effort at this public facility. Such public concern, either pro or con, was shared, I venture, to a greater or lesser degree, by (1) the subset of all of the Hospital’s non-healthcare, custodial and clerical workers, (2) the Hospital’s entire workforce, (3) the healthcare establishment of the entire area, including but not limited to patients and their families, and (4) the community at large. In addition to its potential effect on the compensation, benefits, and working conditions of Hospital employees, a successful union organizing effort would predictably (1) produce fluctuations in the costs of services and changes in the kinds of services offered at the Hospital, and (2) increase political pressure from the public to satisfy hospital workers’ demands. Herrera’s pro-union speech, therefore — irrespective of an inevitable bit of personal motivation — much more directly and substantially addressed a “matter of public concern” than the majority is willing to acknowledge. Yet, courts that have considered the question have uniformly held that speech regarding union activities is almost always speech on a matter of public concern.2 The majority’s subtle trivializing of Herrera’s speech as involving matters of public concern “only insubstantially and in a weak and attenuated sense”3 finds no justification in this record.
b. State Law Proscription of Recognizing Unions
The majority further suggests that any public-concern aspect of Herrera’s pro-union speech is largely negated by the Texas statute that prohibits public agencies from recognizing or collectively bargaining with labor unions.4 As the majority sees it, this Texas law creates a crucial distinction between the instant case, and those like Dep’t of Justice v. FLRA,5 in which the union that prevailed in a federally-supervised election could be and was certified under federal law to act as the non-man*445agement employees’ exclusive bargaining agent. As I interpret this proposition, the majority’s flawed syllogism would go: (1) Unions organize to represent and collectively bargain for workers; (2) Herrera’s union is prohibited by law from doing so; ergo, (3) public-sector employees can have, at most, only a weak and attenuated public concern with union organizing.
Logic dictates a diametrically opposed reasoning and result. Denied the right to be represented officially by a union in collective bargaining and other labor-relation issues, public-sector employees like Herrera would perceive themselves as having a greater need for a strong, collective voice in the arena of public opinion than do employees who can and do have unions as their direct advocates with the general public as well as with their employers. Rather than diminishing the degree of Herrera’s public interest in this organizing effort, this state law prohibition greatly increases the public-interest aspect of Herrera’s expression in support of the organizing campaign here.
Further, the interest of Texans in regulating (or not regulating) labor relations in the public sector is by no means lessened simply because the Legislature has chosen to be restrictive rather than inclusive as regards how government employers may deal with unions. Neither do Texas’s restrictions on official recognition and collective bargaining equate with a lessening of the public’s interest in union activity generally. Texas public employee unions, through collective action in the political arena — such as airing grievances, staging demonstrations, picketing, attending open board or committee meetings, and such — • can and do function influentially as decidedly “public” actors. The larger community inevitably has a substantial interest in the activities of such unions, irrespective of their statutory inability to represent formally, or bargain collectively on behalf of, public employers’ workers.
2.
Connick/Pickering Balancing Process
As noted, the majority ultimately concedes that Herrera’s speech did involve a matter of public concern; and that his speech interest therefore must be balanced against “the interest of the State, as an employer, in promoting the efficiency of the public services it performs through its employees.”6 The Supreme Court has identified a number of factors to be considered in performing this balancing, including whether the employee’s speech (1) impairs discipline by superiors or harmony among co-workers, (2) has a detrimental effect on those close working relationships for which personal loyalty and confidences are essential, (3) impedes the performance of the speaker’s duties, or (4) interferes with the regular operation of the enterprise.7 Except for when a Hospital supervisor fomented confrontations and work interruptions, Herrera’s silent, single-button expression had no deleterious effect on his supervisors’ ability to enforce discipline or on his harmonious interaction with coworkers. Unlike the enlisted ranks in paramilitary agencies or ADAs in a DA’s office, Herrera’s maintenance work was devoid of confidential relationships and requirements for “personal loyalties.” Neither did his work affect the quality or volume of the Hospital’s efforts to accomplish its mission or interfere with the regular functioning of the Hospital, either in *446providing professional services or maintaining its building’s operational condition.
Here, however, the majority eschews (or at least commingles beyond recognition) consideration of the elements of the Court’s prescribed approach or those of any similarly detailed analysis of the real effect of Herrera’s speech. In its place, the majority confects an artificially inflated efficiency interest for the Hospital, then proceeds to balance that overblown interest against the artificially minimized speech interest — not Hererra’s own, or even a hypothetical maintenance worker’s, but that of any hypothetical hospital worker. Yet, even assuming for today’s purpose that the majority reaches the correct result — -and I obviously do not believe that it did — I still would have to question the validity of its methodology. More to the point, in creating and applying this methodology, the majority puts a judicial thumb on the Connick/Pickering scale by using trivializing modifiers to minimize Herrera’s speech interests and aggrandizing modifiers to exaggerate the Hospital’s, efficiency interest.
a. The Hospital’s Efficiency Interest
For example, by describing the public-concern element of Herrera’s speech as being a matter of public concern “only insubstantially and in a weak and attenuated sense,” the majority subtly tips the Connick/Pickering scales against Herrera’s speech interest and in favor of the Hospital’s efficiency interest. This obfuscation should not be allowed to relieve the Hospital of its burden of demonstrating an efficiency interest sufficiently high to justify its denial of Herrera’s First Amendment rights. Yet the majority finds such a predominating interest for the Hospital, first by looking to non-adornment policy cases that it proffers as being similar to the one at issue here, then analogizing efficiency interests of the public employers in those cases to the Hospital’s efficiency interest in requiring that all of its employees’ uniforms be worn free of any items of adornment. In so doing, the majority erroneously equates the efficiency interests of the public employers in those other cases with the efficiency interest of the Hospital here.
Relying on that purportedly analogous case law should be unavailing: Those cases deal almost exclusively with government employers that are either (1) military units, law enforcement agencies, or paramilitary organizations; or (2) quasi-professional civilian agencies such as district attorney’s offices and school boards.8 Just as INS involved uniformed quasi-military Border Patrol agents and Goldman v. Weinberger9 involved a conventional military unit, Connick implicated a district attorney’s penumbral need to enjoy the absolute trust and confidentiality of his lawyer assistants, and Pickering addressed a school board’s need for a heightened professional relationship with career teachers. Even Rankin, itself a law enforcement case, distinguished the special loyalty, confidentially, and discipline needs of a sheriff vis-á-vis his deputies and rejected the applicability of this efficiency-loyalty-discipline interest vis-á-vis a clerical worker — an employee much more analogous to Herrera than to deputy sheriffs, Border Patrol agents, or assistant district attorneys.
*447Cases such as those relied on by the majority are at best minimally comparable to this one, if comparable at all. I readily acknowledge that for law enforcement agencies uniforms serve employer interests in efficiency and esprit de corps. But such purposes are materially distinct from any efficiency interests that unadorned uniforms might conceivably serve for a public hospital in the context of its non-medical maintenance, food preparation, trash removal, and clerical staff. The majority acknowledges (grudgingly) only that “uniforms may be more important in law enforcement than in other fields”10 — once again trivializing a differentiating distinction to support its suggestion that all public agencies, regardless of mission, share a virtually identical interest in requiring unadorned uniforms for every category of employee, regardless of function — just some to a slightly greater or lesser extent than others. Respectfully, this simply is not so.
b. Uniforms, Maybe; Items of Adornment, No
The majority signals its agreement with the Ninth Circuit’s holding in INS,11 by stating that (1) “a uniform requirement [not, I emphasize, a non-adornment requirement, which could apply to uniforms and mufti alike] fosters discipline, promotes uniformity, encourages esprit de corps, and increases readiness,” and (2) having “standardized uniforms encourages the subordination of personal preferences and identities in favor of the overall group mission.”12 Thus, the majority muses, “there is no reason to believe that a uniform requirement will not have somewhat similar efficiency enhancing effects in the non-law enforcement context.”13
I acknowledge the verity of this broad tautology, with its elastic use of “somewhat similar.” To it I must add, however, that, in the much more apropos context of non-law enforcement, public employers like the Hospital, making the additional non-adornment facet of a uniform requirement equally applicable to maintenance and clerical workers as to doctors, nurses, therapists, etc. at most enhances efficiency “only insubstantially and in a weak and attenuated sense” — to quote the majority. The ineluctable fact is that (1) uniforms bearing only the employer-prescribed insignia are central and paramount to the core interests of military and law enforcement agencies; but (2) having Herrera and his subset of workers wear no adornments on their work clothes would contribute minimally, if at all, to such clearly secondary or tertiary interests of civilian institutions like hospitals. And, again, we must remain consciously aware that it is not the uniform requirement per se that Herrera violated; rather, his free speech clashed with the Hospital’s application of the non-adornment appendage of that policy to its custodial and clerical segment of staff.
Finally, the majority treats the Hospital’s avowed interest in requiring maintenance workers to wear unadorned uniforms as though it were a natural extension of, or progression from, this and other courts’ past validations of law enforcement agencies’ efficiency interest in uniforms and thus in non-adornment. The majority might be comfortable with this non sequitur but, as I shall further *448demonstrate, in the discrete context of this case, such a leap does not bridge the gap between unadorned uniforms of military and law enforcement employees, and unadorned uniforms of the very distinguishable subset of blue-collar workers in a civilian public hospital.
c. Efficiency
For openers, it appears to be lost on the majority that “efficiency” as a governmental employer’s interest under Connick /Pickering, necessarily means different things in different settings. As previously observed, “esprit de corps,” “readiness,” and “subordination of personal preferences” are undoubtedly critical to the efficient and effective operations of law enforcement agencies — as they likely are to the efficient functioning of a public hospital’s operating rooms, ICUs, catheter labs, and emergency rooms. But where is there any contribution from esprit de corps or unquestioning responses to orders to the efficient operation of the Hospital’s physical plant, file rooms, and cafeterias? Non-healthcare employees who work those areas, like the sheriffs clerk in Rankin, essentially fly beneath the radar and perform tangential support functions that are subordinate to and separate from the overarching healthcare mission of the Hospital. With respect, I urge that the majority overreaches when it stretches the Ninth Circuit’s INS law-enforcement holding to cover the “somewhat similar” efficiency interest of the Hospital in applying its non-adornment uniform policy to Herrera. This distinction becomes undeniably telling when the majority finally addresses the Hospital’s specific interest in maintaining an unadorned-uniform policy for its custodial and clerical workers — an alleged interest that simply cannot be forced to resemble, even slightly, those of law enforcement or paramilitary agencies. Yet, as the majority flatteringly describes it, the Hospital finds it “highly desirable” for its maintenance workers to “be easily identifiable as such” and “present an appropriate appearance,”14 Even if, arguendo, such self-serving and conclusional protestations were to be credited, how could a single pro-union button lessen the ease of identifying Herrera as a maintenance worker or the propriety of his appearance?
This healthcare facility’s interest, however logical and legitimate it may sound in a vacuum, is a far cry from the interests in esprit de corps, readiness, or unquestioning response to orders that courts have recognized as key to law enforcement agencies’ speech-restricting anti-adornment regulations. But, even granting that a civilian hospital can somehow rationalize an abstract desirability of having its maintenance personnel wear identifying uniforms sans pins or patches, there is still just no way to equate the efficiency value of the non-adornment gloss that the Hospital has engrafted on its basic uniform policy with the indisputably greater value of prohibiting competing adornment on law enforcement and military uniforms. I agree that in cases of that type, addressing as they do law enforcement agencies and military units, uniforms that are required to be adorned with such employers’ own functional insignia — chevrons, bars, leafs, stars, unit patches, and such — serve very real efficiency purposes within and without their ranks. Moreover, it is axiomatic that such function-serving items cannot abide the presence on such uniforms of competing or distracting insignia of the wearer’s personal choice. Not so, however, for the monochromatic denims, dungarees, whites, or khakis required of custodial personnel, cafeteria workers, and file clerks, free, as *449they are, of any functioning insignia of rank, unit, or specialty. There simply can be no confusion or ambiguity when such otherwise insignia-less garb is “adorned” with a single pro-union button.15 It defies logic to conclude that one such item on the otherwise unadorned work uniform of an in-house carpenter, plumber, cafeteria worker, file clerk, electrician, or janitor could negatively affect any efficiency-enhancing function of their work dress — at least not in the real world.
d. Herrera’s Speech Interest
As made clear by the Supreme Court in identifying Rankin’s four non-exclusive factors to be used in balancing the competing interests of a constable and his clerical staff, the weight afforded an employee’s speech interest in the Con-nick/Pickemg balancing process depends to a great extent on the peculiarities of not only the speech itself, but also on the specific work and job description of the individual employee/speaker within the government employer’s operational structure. Stated differently, what’s sauce for a public hospital’s doctor or technician “goose” is not necessarily sauce for its carpenter or file-clerk “gander.” The majority, however, appears disinclined to analyze how Herrera’s particular expression might in reality affect the Hospital’s accomplishment of its principal mission, as contemplated by the Supreme Court in Rankin. Had the majority dutifully conducted such an analysis, it just might have been compelled to conclude that the Hospital’s policy prohibiting items of adornment on uniforms (with notable exceptions, I repeat) was unduly restrictive as applied to Herrera and his “Union Yes” button. Perhaps, then, it was to avoid this troublesome truth that the majority chose to make no personal reference to Herrera (or to the separate subset of maintenance employees) in its evaluation of the competing interests to be balanced. Instead, the majority globally lumps together all “Hospital employees,” as though large segments (maintenance, janitorial, food preparation, clerical) are indistinct from other segments (health care specialists, medical and quasi-medical employees) for the. purpose of assigning relative weights to such interests in the conduct of the Con-nick/Pickering balancing test. I need not labor longer to demonstrate the obvious flaw in the majority’s approach.
e. Blanket Application of Uniform, Adornment Policy
The majority makes much of its belief that the Hospital’s non-adornment policy is content-neutral and therefore may be applied equally to all employees. At first blush, a policy thus configured might appear to be desirable and expedient for public civilian hospitals; but if it is to survive a First Amendment challenge, it must do so on a highly individualized, case-by-case basis. The majority incants this maxim, of course, but goes on to honor it only in the breach. It fails to address the particular circumstances of this case, most notably among which is the fact that Herrera is a maintenance worker — indistinguishable from his counterparts in virtually every functioning edifice, whether public or private — who has some visibility but almost no direct interaction with hospital patients and their families, or, for that *450matter, with M.D.s, RNs, med techs, or executives in management positions. This framework must not be ignored or belittled when balancing Herrera’s speech interest against the Hospital’s interest in efficiency. A hospital policy that is properly found to outweigh the adornment “speech” rights of doctors, nurses, LPNs, and technicians, as well as management executives, can at the same time be constitutionally outweighed by the free-speech rights of rank-and-file non-healthcare workers like Herrera.
The majority nevertheless tests the Hospital’s non-adornment policy without distinguishing between such subsets of its work force and finds the policy appropriate for all employees, across the boards. Such a conclusion may be reached legitimately only after a full and detailed consideration of every factor unique to the case under consideration, i.e., by recognizing that, when it comes to unadorned uniforms, there can be “different strokes for different folks.” I refer in particular to the nature of Herrera’s employment and job description, and the extent to which his kind of work — -and his appearance at work — necessarily affects how much or how little weight should be given to his speech rights in this court’s de novo conduct of the Connick/Pickering balancing process.
3.
Broader Import of This Case
Finally, the majority is quick to warn that, if we allow Herrera’s pro-union speech to trump the anti-adornment aspect of the Hospital’s uniform policy as applied to Herrera, this case could be just the tip of a free-speech iceberg that would threaten the safe passage of many another government employer’s tranquil vessel. But this “horrible” just won’t join the parade: If the Rankin approach is faithfully followed, the nature of an employee’s speech and the context in which it is uttered will always consist of case-specific factors that must be included in the free-speech calculus employed by courts faced with similar public workplace First Amendment challenges. Obviously, the particular Rankin factors of some kinds of speech by some categories of employees will carry more weight on the Connick/Pickering balance beam than will others. To suggest, however, that the approach and result I advocate in this case would produce a precedential pandemic infecting a multitude of civilian public employers with an onerous burden of tolerating any and all manner of on-the-job speech, is nothing more than hype. This warning misapprehends the nature of the analysis that should be undertaken and exaggerates any potential precedential effect of our protecting Herrera’s speech, as a non-paramilitary, non-professional laborer in the civilian sector of public service. An objectively proper exercise of the balancing test will itself contain all the safeguards needed to avoid the scary results predicted by the majority. Its present fears are far less than its horrible imaginings.
I am never completely free of discomfort when dissenting from colleagues in a three-judge panel, much less in the face of a supermajority of my colleagues sitting en banc. This is especially so when, as here, the majority opinion is penned by a jurist of Judge Garwood’s preeminence and reputation. Nevertheless, for the foregoing reasons and those set forth in the panel majority opinion that was vacated to re*451hear this case en banc,161 must respectfully dissent.
. See Branton v. City of Dallas, 272 F.3d 730, 740 (5th Cir.2001) (noting generally that *444"[njeither the [First] Amendment itself nor our decisions indicate that ... freedom [of speech] is lost to the public employee who arranges to communicate privately with his employer rather than to spread his views before the public.”) (citing Givhan v. W. Line Consol. Sch. Dist., 439 U.S. 410, 415-16, 99 S.Ct. 693, 58 L.Ed.2d 619 (1979)).
.See, e.g., Boddie v. City of Columbus, 989 F.2d 745, 750 (5th Cir.1993) ("[S]peech in the context of union activity will seldom be personal; most often it will be political speech.”); see also American Postal Workers Union, AFL-CIO v. United States Postal Serv., 830 F.2d 294, 301 (D.C.Cir.1987) ("The urge to unionize certainly falls within the category of expression that is 'fairly considered as relating to any matter of political, social, or other concern to the community ....'”) (quoting Connick v. Myers, 461 U.S. 138, 146, 103 S.Ct. 1684, 75 L.Ed.2d 708 (1983)).
. Maj. Op. at 439-40.
. See Tex. Gov.Code § 617.002.
. 955 F.2d 998 (5th Cir. 1992).
. Pickering v. Bd. of Educ., 391 U.S. 563, 568, 88 S.Ct. 1731, 20 L.Ed.2d 811 (1968).
. Rankin v. McPherson, 483 U.S. 378, 388, 107 S.Ct. 2891, 97 L.Ed.2d 315 (1987) (a county constable/clerical employee case).
. The majority relies most heavily on the Ninth Circuit’s reasoning in United States Dept, of Justice, Immigration and Naturalization Serv. v. Fed. Labor Relations Auth.,["INS"], 855 F.2d 1454 (9th Cir.1988), even though the Hospital's briefs cite numerous other cases.
. 475 U.S. 503, 106 S.Ct. 1310, 89 L.Ed.2d 478 (1986).
. Maj. Op. at 439 (emphasis added).
. 855 F.2d at 1464.
. Id. (citing Goldman v. Weinberger, 475 U.S. at 508, 106 S.Ct. 1310 ("[T]he traditional outfitting of personnel in standardized uniforms encourages the subordination of personal preferences and identities in favor of the overall group mission.”)).
.Maj. Op. at 439 (emphasis added).
. Maj. Op. at 440 (Emphasis added).
. Indeed, in Lubbock — or Midland or Odessa or and many other communities in this circuit — the partisan football supporters' buttons and stickers that the Hospital permits as one of several exceptions (thereby destroying its claim to the policy’s neutrality), are more likely to cause rancor and confrontation than would even a union label in that right-to-work-state.
. Communication Workers of America v. Ector County Hosp. Dist., 392 F.3d 733 (5th Cir.2004).