To what extent may a city regulate the political activities of its employees when city elections are conducted on a "nonpartisan" basis? Broadly stated, this is the principal issue we face in the present case.

Dallas fireman Richard Wachsman and the Dallas Police & Fire Action Committee ("Committee")1 requested injunctive relief against the City of Dallas ("City") and the chiefs of the police and fire departments.2 At issue were sections 16(b) and 16(c) of Chapter XVI of the City Charter of the City of Dallas, Texas, which provide:

"(b) To avoid undue influence of city employees on the outcome of city council elections and to avoid undue influence of city councilmen or candidates for city council on city employees, the following restrictions are imposed:

"(1) No employee of the city or association of such employees may publicly endorse or actively support candidates for the city council or any political organization or association organized to support candidates for the city council;

"(2) No employee of the city may circulate petitions for city council candidates, although he may sign such a petition;

"(3) No employee of the city may contribute, directly or indirectly or through an organization or association to such a campaign nor solicit or receive contributions for a city council candidate;

"(4) No employee of the city may wear city council campaign buttons nor distribute campaign literature at work or in a city uniform or in the offices or buildings of the City of Dallas.

"(c) In elections other than for city council of the City of Dallas, an employee of the city may not:

"(1) Use the prestige of his position with the city for any partisan candidate;

"(2) Manage a partisan political campaign;

"(3) Solicit or receive contributions for such a campaign;

"(4) Actively support a candidate except on his own time while not in a city uniform nor in an office or building of the City of Dallas."

Appellants Wachsman and the Committee did not and do not challenge subsections (b)(4), (c)(1), and (c)(4).

The district court held section 16(b)(1) unconstitutional insofar as it prohibited "an employee of the Dallas Police Department or Dallas Fire Department from endorsing a city council candidate before any group of more than 15 people, unless such group is a convention, caucus, rally or similar gathering."3 The City has not appealed from this holding. The district court upheld the constitutionality of the remainder of the sections as challenged by the plaintiffs. On appeal Wachsman and the Committee present the following issues:

1. Whether the City can prohibit financial contributions by these city employees or their organization to city council candidates.

2. Whether the City can prohibit public endorsements of city council candidates by these city employees or their organization, even as narrowed by the district court's order.

3. Whether the City can prohibit these city employees from circulating (though not from signing) endorsement petitions for city council candidates.

4. Whether the City can restrict these city employees or their organization from soliciting and/or receiving campaign funds on behalf of city council candidates, or for partisan political campaigns.

5. Whether the City can prohibit these city employees from managing partisan political campaigns.

Entwined in these issues are two novel questions for this Circuit. In the "Hatch Act" cases, United States Civil Service Commission v. National Association of Letter Carriers, 413 U.S. 548, 93 S. Ct. 2880, 37 L. Ed. 2d 796 (1973), and Broadrick v. Oklahoma, 413 U.S. 601, 93 S. Ct. 2908, 37 L. Ed. 2d 830 (1973), the Supreme Court approved certain restrictions on the political activities of federal and state employees in a partisan setting. The Hatch Act,4 however, does not impose a ban upon campaign contributions. Therefore, we must decide whether the City can impose restrictions in a nonpartisan setting similar to those approved in Letter Carriers and Broadrick, and, if so, whether the rationale of these cases also allows the City to prohibit city employee contributions to city council candidates.

The parties stipulated that Dallas municipal elections and offices are "nonpartisan," and that "there is no substantial party involvement in the city council elections."5 Relying on this fact and the emphasis in Letter Carriers and Broadrick on partisan political activity by government employees, 413 U.S. at 557-67, 93 S.Ct. at 2886-2887 and 413 U.S. at 606, 93 S.Ct. at 2912-2913, Wachsman and the Committee argue that the Hatch Act cases do not control this case. Appellants also heavily rely on Morial v. Judiciary Commission, 565 F.2d 295 (5th Cir.1977) (en banc ), cert. denied, 435 U.S. 1013, 98 S. Ct. 1887, 56 L. Ed. 2d 395 (1978), and Magill v. Lynch, 560 F.2d 22 (1st Cir.1977), cert. denied, 434 U.S. 1063, 98 S. Ct. 1236, 55 L. Ed. 2d 763 (1978). In the latter case, without citation or explanation, the Court stated: "Political oppression of public employees will be rare in an entirely nonpartisan system." 560 F.2d at 29. Magill concerned firemen's political participation in Pawtucket, Rhode Island city elections which were "nominally" nonpartisan. The Court held that "the government may constitutionally restrict its employees' participation in nominally nonpartisan elections if political parties play a large role in the campaigns." Id. (footnote omitted). In Pawtucket parties did play such a role.6 The Court made no firm pronouncement on whether the employees' first amendment rights would outweigh the city's interests in controlling their political activities in elections in which party involvement was absent, but it did state that in such a situation "the [governmental] interests identified by the Letter Carriers Court [would] lose much of their force." Id.

In Morial, which concerned Louisiana's statutory requirement that judges resign before running for elective nonjudicial office, the Court also found significant party involvement in a "nonpartisan" election.7 The Court stated, "In any given case, the relevant inquiry must be whether the threat to the state's interests in the impartiality of its public servants stems from party involvement or from political involvement." 565 F.2d at 303-04 n. 8. The opinion also observes that "[i]n the context of the Hatch Act which regulates the activities of federal employees, it was natural for the Supreme Court [in Letter Carriers ] to discuss the dangers of partisanship as if they were substantially identical to the dangers of party domination," and that "[a] faction may form around a man as much as around a party label; the judicial office may be abused by using it to promote the interest of a faction as well as a formal party." Id. Wachsman and the Committee insist that simple political involvement does not raise concerns of sufficient significance to warrant infringement of employees' rights, and that the problems raised by party involvement are what make such infringement constitutionally permissible. This, according to appellants, is why Morial states that courts must distinguish between party and political involvement. The City counters by arguing that Morial means courts should not rely on labels in deciding these cases.

While each side in the case at bar can derive a measure of comfort from various expressions in the Morial and Magill opinions, we regard neither holding as decisive in the present context. Neither decision invalidated the restrictions there in issue. However, though the New Orleans and Pawtucket municipal elections were legally structured as "nonpartisan," political parties nevertheless played a more significant and overt role in the actual results in these cities than is the case in Dallas. Neither decision purports to definitively address the issue we face here.8

We are unwilling to strike down Dallas' Charter provisions merely because of the noted nonpartisan nature of the Dallas city council elections.

Not only did Letter Carriers uphold the Hatch Act, but the Supreme Court's opinion contains no direct statement that the result would have been otherwise had the restrictions not been limited to partisan activities. Admittedly, the opinion stresses the partisan factor, but, consistent with Morial, we believe that this is because political activity at the federal level is partisan, in both an inter-party and an intra-party sense.9 This is not simply a coincidence, but is rather a factor establishing the requisite closeness of the relationship between the restrictions in question and the legitimate governmental ends to be served thereby. One aspect of this was accurately capsulized by the statement of the City's expert witness at trial that "[w]hat the Hatch Act tries to do is keep the employee from being involved in the politics that elects his boss." Accordingly, even if Hatch Act type restrictions imposed on its employees by the federal government, where the relevant politics are essentially partisan, would be constitutionally invalid if generally extended to nonpartisan political activity, it would not necessarily follow that municipalities whose relevant politics are essentially nonpartisan would not be able to constitutionally impose restrictions applicable to their nonpartisan elections on municipal employees.

While Letter Carriers indisputably contains frequent references to partisan politics and political parties, it also contains several crucial passages, identifying the harms to important societal interests attendant on the politicalization of governmental service. The discussion in these passages is in terms of politics generally rather than simply party or partisan politics. Quoting with apparent approval from United Public Workers v. Mitchell, 330 U.S. 75, 98, 67 S. Ct. 556, 569, 91 L. Ed. 754 (1947), Justice White, in Letter Carriers, 413 U.S. at 555, 93 S.Ct. at 2885, observed that in Mitchell the Court stated that in enacting the Hatch Act, Congress had "recognized the 'danger to the [government] service in that political rather than official effort may earn advancement and to the public in that governmental favor may be channeled through political connections.' " Justice White's opinion in Letter Carriers continues by stating that a decision to uphold the Hatch Act

"... would no more than confirm the judgment of history, a judgment made by this country over the last century that it is in the best interest of the country, indeed essential, that federal service should depend upon meritorious performance rather than political service, and that the political influence of federal employees on others and on the electoral process should be limited." Id. at 557, 93 S.Ct. at 2886.10

These governmental interests subserved by restrictions on the political activities of federal employees are restated and elaborated on in these subsequent passages from Letter Carriers:

"There is another consideration in this judgment: it is not only important that the Government and its employees in fact avoid practicing political justice, but it is also critical that they appear to the public to be avoiding it, if confidence in the system of representative Government is not to be eroded to a disastrous extent. [Id. at 565, 93 S. Ct. at 2890.]

"....

"A related concern, and this remains as important as any other, was to further serve the goal that employment and advancement in the Government service not depend on political performance, and at the same time to make sure that Government employees would be free from pressure and from express or tacit invitation to vote in a certain way or perform political chores in order to curry favor with their superiors rather than to act out their own beliefs." [Id. at 566, 93 S. Ct. at 2890-2891.]

And, Letter Carriers identifies still another "major concern" as being "the conviction that the rapidly expanding Government work force should not be employed to build a powerful, invincible, and perhaps corrupt political machine." Id. at 565, 93 S. Ct. at 2890.

All these concerns led the Court in Letter Carriers to conclude that "plainly identifiable acts of political management and political campaigning on the part of federal employees may constitutionally be prohibited." Id. at 567, 93 S.Ct. at 2891.

We read Letter Carriers as recognizing several important societal interests which would be sufficiently adversely affected by certain conditions potentially attendant on unrestrained political activity of governmental employees as to justify substantial restrictions on those activities. At least four such societal interests may be identified: the interest in an efficient government; that in a government which enjoys public confidence; that in the right of individual citizens to be free of governmental discrimination based on their political activities or connections; and that in the right of governmental employees to be free of employer pressure in their personal political decisions. The potential conditions which would be harmful or injurious to these important societal interests include the following three. First, the condition could exist in which "employment and advancement in Government service" is made to "depend on political performance" rather than on "official effort" or "meritorious performance." This not only impairs the efficiency of government directly, by depriving it of the services of the more capable for the positions so affected, but also indirectly, by its adverse effects on employee morale. This condition is also injurious to the governmental employees' rights to be free of employer pressure to affiliate with a candidate they may personally abhor or "to perform political chores in order to curry favor with their superiors rather than to act out their own beliefs." Moreover, this condition obviously tends to undermine public confidence in government. A second harmful condition is that of governmental employees "practicing political justice," or exercising "political influence ... on others," or channeling "governmental favor" "through political connections." Such a condition is seriously injurious to the rights of individual citizens to be free of governmental discrimination based on their political activities or connections. It also impairs public confidence in government. Governmental efficiency is also adversely affected in that governmental administration is deflected from the direct and purposeful implementation of the laws and policy choices made by those charged with that responsibility, generally the elected representatives, and is instead focused on furtherance of the political objectives of the civil servant. Third, the condition may occur under which "the political influence of federal employees" is brought to bear without restraint "on the electoral process," or the governmental work force is employed to build a powerful "political machine." This condition undermines public confidence in government both because it tends to cast government in the light of master rather than servant of the people, and because it involves use of governmental power for purposes other than those for which the government was instituted. The latter factor also impairs governmental efficiency in the exercise of its proper purposes.

We believe that the foregoing conditions are no less harmful merely because they may be brought about by political pressures generated in a nonpartisan, rather than a partisan, political context. If Police Officer Smith is best suited for a promotion vacancy, but the promotion goes instead to Officer Jones, who is not qualified for the position, because Jones helped in the Mayor's reelection campaign by soliciting campaign contributions from the businesses and residents on his beat while Smith, whose personal preference was the Mayor's opponent, refused to do so, does it make any difference whether the campaign was partisan or nonpartisan? Will not the injuries to the efficiency of the police force and its morale, and to the rights of Officer Smith, be fully as grievous in the one case as in the other? Will not the confidence of the public be just as undermined? Have not the citizens on Officer Jones' beat been just as imposed on by being solicited by the very governmental employee to whom, day in and day out, they must look for both effective police protection and for evenhanded, reasonable enforcement of the laws regulating their own conduct?

Nor are we able to say with any confidence, on the basis of history or common knowledge, that if mayoralty campaigns were nonpartisan rather than partisan, the result would be a reduction in the likelihood of occurrences, of the kind above hypothesized, of such magnitude and reasonable consistency among different localities as to mandate one federal constitutional requirement for the cities with "partisan" elections and another for all those whose elections were "nonpartisan."11

An across-the-board constitutional distinction for these purposes between "partisan" and "nonpartisan" elections to local legislative and executive type offices seems questionable on other grounds as well. We doubt that such human traits as personal ambition, greed, fear, and the like, on the one hand, and commitment to principle, unselfishness, honor, and similar characteristics, on the other hand, are distributed among such candidates and their supporters in significantly different proportions according to whether the election by which they are to be chosen is "partisan" or "nonpartisan." Experience in years past in most areas of the former Confederacy also cautions against making such a distinction. There, generally the only relevant elections at the state and local levels were the Democratic primaries, and the only relevant contending forces and organizations were those which coalesced around the various candidates and interest and ideological groupings, rather than around any formal party structure or machinery. Yet this state of affairs did not substantially eliminate the adverse effects of active governmental employee electioneering.12 We would be reluctant to hold that in such situations, some of which persist even to this day, "little Hatch Acts" would be unconstitutional. Nor in that setting would it seem reasonable to uphold a "little Hatch Act" merely because of the formality of a general election legally structured as partisan, even though normally such an election, typically involving only one candidate, would have no true relevance whatever to the actual workings of the political process.

We are taught by Magill, and at least inferentially by Morial, that Hatch Act type restrictions imposed by a governmental entity on its employees are not invalid merely by reason of the fact "so far as the laws of ... [that entity] can make them, ... [its] elections are nonpartisan." Magill, 560 F.2d at 26. But, rejecting as we must the contention that a municipality's Hatch Act type restrictions are invalid merely because the formal legal structure of its elections is as nonpartisan as the law can make elections, should we nevertheless hold that the determinative consideration is whether the elections are in fact nonpartisan? Were we to do so, we would embark on a course with few meaningful guideposts for judicial decision. What is a political party for such purposes? Were the "New Party" and the "Old Party," which contended for political power in certain areas of South Texas, political parties? What of the "Good Government" and "Citizens Leagues" and similar organizations that have been significant forces for long periods of time in the legally "nonpartisan" political affairs of certain Texas municipalities? What of organizations such as the "West Austin Democrats," which frequently endorse candidates in the legally "nonpartisan" municipal elections of the capital city of Texas? Moreover, it is common knowledge that when other state and local elections within the geographical area of a municipality are legally and factually conducted on a partisan basis, partisanship will almost inevitably have at least some bearing on the municipal elections even though they are in legal form nonpartisan and are held in different years from the legally partisan elections. See note 5, supra. How much "partisan" activity or influence is necessary to cause an election, the legal structure of which is "nonpartisan," to nevertheless be sufficiently "partisan" in fact for purposes of the rule for which appellants contend? To what extent, if any, should this depend on how close the elections are, given that the closer the election, the less partisan activity or influence is required to render partisanship potentially outcome-determinative? And, what kind of activity is "partisan" for these purposes? Is it necessary that the political parties act through their formal structures, or will action of informal party "clubs" suffice? What of endorsement advertisements by citizens, perhaps including local officials elected in legally partisan elections, identifying themselves by party affiliation and the supported city council candidate as a long-time member of that party? Moreover, a determination of whether elections are partisan in fact, apart from their legal structure, involves primarily the consideration of matters which by their nature will tend to vary over time. Some variations will come about abruptly, others incrementally. If in the last two municipal elections the campaigns have become progressively more partisan on each occasion, may the city adopt a "little Hatch Act" in anticipation that the trend will continue? Will we hold the city's "little Hatch Act" invalid as to one election but valid as to the next, or vice versa, though there has been no intervening change in the legal structure of the elections?13

We decline to adopt as the touchstone for decision in this area an across-the-board distinction based purely on whether the elections are partisan or nonpartisan. For purposes of judging the validity of restrictions on election-related activity in the light of first amendment considerations, we believe it more meaningful to distinguish between elections on the basis of whether they are candidate elections or noncandidate elections, such as referenda and constitutional amendment elections. See Citizens Against Rent Control v. Berkeley, 454 U.S. 290, 298-99, 102 S. Ct. 434, 438-439, 70 L. Ed. 2d 492, 501 (1981)14; First National Bank of Boston v. Bellotti, 435 U.S. 765, 790, 98 S. Ct. 1407, 1423, 55 L. Ed. 2d 707, 726 (1978) ("Referenda are held on issues, not candidates for public office."); Let's Help Florida v. McCrary, 621 F.2d 195, 199-200 (5th Cir.1980), aff'd mem. sub nom., Firestone v. Let's Help Florida, 454 U.S. 1130, 102 S. Ct. 985, 71 L. Ed. 2d 284 (1982). That distinction is preserved in the Dallas City Charter.

Since we have identified governmental interests served by the restrictions on city employee activities in city council elections that are as significant as those found in Letter Carriers, and since, with the exception of contributions, the Supreme Court upheld very similar restrictions in that case,15 we see no reason to distinguish these provisions on a partisan/nonpartisan basis. They are constitutional.16

Appellants argue that whatever the City's interests may be, they could not apply to the restrictions on campaigns for offices other than city council, as the fear of employer overreaching would not be present. In other words, the City has no interest in preventing Wachsman from managing, or soliciting funds for, the partisan campaign of a candidate for a local, statewide, or national elective office, as long as the political activity does not interfere with Wachsman's work.17

Virtually all the numerous restrictions on federal employee political activity upheld in Letter Carriers (see note 15, supra ) apply as much to strictly state and local elections and political affairs as to elections for federal office and political activities attendant thereto. Thus, for example, the Hatch Act's prohibitions of federal employees publicly endorsing candidates or driving voters to the polls apply as fully to "off-year" elections for village constable as they do to elections for the United States Senate. Accordingly, the mere fact that restrictions on governmental employee political activity extend to elections for positions in governmental entities other than the entity imposing the restrictions on its employees does not of itself invalidate the restrictions. Nevertheless, the Hatch Act restrictions on federal employees are limited to partisan politics, and the political parties which operate at the federal level are generally the same ones which are significant at the state and local levels. This factor thus furnishes a nexus between the federal and the state and local politics as to which the Hatch Act restricts federal employee political activity. It is argued that in the present case no such comparable nexus exists, because the city elections are nonpartisan, while the other elections to which the Charter restrictions apply are partisan and thus unrelated to the City. This contention is not without force.

However, we conclude that while the case at bar does not present as close a nexus between the nonemployer elections and the affairs of the employer governmental entity imposing the restrictions on its employees as does the Hatch Act, nevertheless such a nexus is not entirely lacking here. We believe it unrealistic to assume that politics within the geographical boundaries of a city are divided into completely unrelated watertight compartments of city and noncity politics. On the candidate and officeholder level, it is certainly not unheard of for a person prominent in local partisan politics, as a former officeholder or otherwise, to become a city councilperson in a nonpartisan election, or for a member of the nonpartisan city council to thereafter become a local, state, or even federal elective officeholder through the partisan political process. Moreover, significant operating relationships frequently exist within the geographical area of a city, between the city government, whether partisan or not, and the county, state, and federal governments. City politics, then, whether or not "partisan," cannot be viewed as wholly divorced from the politics, within the area of the city, of the local, state, and federal governments.

Further, we note that the full range of Hatch Act prohibitions applies to state and local political activities, while the kinds of activities in which the Dallas City Charter prohibits its employees from engaging respecting noncity elections are much more limited than the Charter's prohibitions respecting city elections. Aside from the prohibitions against activities in uniform or in city buildings or on city time and against use of the prestige of city office, which are not challenged, the only prohibitions respecting noncity elections are those against soliciting funds and serving as campaign manager. Here the rights of the city employees are less fundamental.18

Finally, appellants' challenge to these portions of the Charter fails to recognize the considerable power wielded by public employees by virtue of their positions. Of necessity, individual governmental employees, as a practical matter, often have considerable discretion in the actual administration and enforcement of laws and the provision of governmental services. And, the timing, manner, and nature of the performance of such governmental functions in actual practice may frequently be subject to variation according to the attitudes of the governmental employees concerned. A Dallas private citizen might well be hesitant to refuse a political contribution or favor sought by a campaign manager or solicitor in a county, legislative or congressional campaign if the private citizen were dependent on that campaign manager or solicitor for fire or police protection or were subject to discretionary law or regulatory enforcement actions by that campaign manager or solicitor. Certainly the City has a legitimate interest in minimizing the exposure of its citizens to such pressures from city employees. And the Dallas private citizen is no less pressured simply because the election campaign for which the contribution or favor is sought by the city employee is for an office such as sheriff of Dallas County, or a local legislative or congressional seat, rather than for a place on the Dallas city council.19

Accordingly, the City has a significant interest in limiting the activities of its employees in local "partisan" campaigns. While these interests are somewhat less extensive than those present in city council elections, the restrictions imposed are less onerous, and less fundamental employee rights are involved. Therefore, we uphold these provisions.

One further problem deserves our attention before we address the restrictions on contributions. The district court's invalidation of section 16(b)(1) regarding endorsements presented at nonpolitical gatherings applied only to individual employees, and not to their organizations. The Committee asserts that this ruling ignores its first amendment rights. These rights are derived from the individual members' associational and speech rights, see Citizens Against Rent Control, and from the public's right to free and uninhibited comment on political issues. See Bellotti. Although the district court failed to state why it reached different conclusions regarding Wachsman and the Committee, we perceive an appropriate distinction that justifies its order.

The Committee must act through a spokesman. Regarding individual employees, the court was obviously concerned with limiting an employee's right to endorse a candidate at a private and/or nonpolitical gathering (e.g., a Kiwanis Club meeting or a neighborhood barbecue). Such a setting suggests a public employee acting as a private citizen. This suggestion, however, does not fit a situation in which an official spokesman for an organization of city employees announces in that capacity the organization's endorsement of a particular city council candidate. The latter carries with it all the pernicious possibilities inherent in allowing individual employees to so address political gatherings. The arrival of an official spokesman bearing such an endorsement largely imbues a gathering with a political flavor. Indeed, the appellant Committee here is the avowedly political arm of the city employee police and firefighters organizations. Thus, the City's interests in prohibiting endorsements will be present in any situation in which an organization of city employees desires to make an endorsement. Therefore, the trial court's order in this regard is affirmed.20

We address separately the trial court's decision regarding the City Charter's prohibition on contributions to city council candidates because the Hatch Act does not contain such a ban. To resolve this question, we invoke the test developed in Morial by Judge Goldberg, which he distilled from Elrod v. Burns, 427 U.S. 347, 96 S. Ct. 2673, 49 L. Ed. 2d 547 (1976), Buckley v. Valeo, 424 U.S. 1, 96 S. Ct. 612, 46 L. Ed. 2d 659 (1976), and Letter Carriers, supra. See also Clements v. Fashing, --- U.S. ----, 102 S. Ct. 2836, 73 L. Ed. 2d 508 (1982); Citizens Against Rent Control, supra; Branti v. Finkel, 445 U.S. 507, 100 S. Ct. 1287, 63 L. Ed. 2d 574 (1980); and Bellotti, supra.

"The standard to be applied ... is a function of the severity of impairment of first amendment interests. As the burden comes closer to impairing core first amendment values, ... or impairs some given first amendment value more substantially, ... the requisite closeness of fit of means and end increases accordingly.... [R]estrictions on the partisan political activity of public employees and officers, where such activity contains substantial nonspeech elements, ... are constitutionally permissible if justified by a reasonable necessity ... to burden those activities to achieve a compelling public objective." Morial, 565 F.2d at 300 (citations and footnote omitted).

Although the contribution ban affects a substantial first amendment right, we find it is reasonably necessary to achieve a compelling public objective.

Limitations on campaign contributions unquestionably involve substantial first amendment rights. These rights are primarily associational, Buckley, 424 U.S. at 21-25, 96 S.Ct. at 635-636, 637-638, although they also affect freedom of expression, id. at 21, 96 S.Ct. at 635-636; see also Citizens Against Rent Control, 454 U.S. at 299, 102 S.Ct. at 439, 70 L. Ed. 2d at 501, especially when the ban is absolute. In fact, since a total prohibition prevents the "symbolic expression of support evidenced by a contribution," Buckley, 424 U.S. at 21, 96 S.Ct. at 636, the City Charter limits employees' freedom of expression in the constitutionally sensitive area of political speech.

However, Buckley does hold that private citizen candidate contributions may be limited. And, we have noted the distinction for these purposes made between candidate elections and noncandidate, referenda type elections. Citizens Against Rent Control, supra; Bellotti, supra; Let's Help Florida v. McCrary, supra. We are dealing with contributions to candidate campaigns here, and while here there is a prohibition, rather than merely a limitation as in Buckley, in Buckley the limitation was across the board, both as to contributors and types of candidate elections, and here the only contribution restriction is that city employees are prohibited from contributing to city council elections.

In our view the City's interest in enacting this limited prohibition, preventing city council candidate election contributions by city employees, is undoubtedly compelling.

A century ago our Supreme Court, over but a single dissent, upheld the constitutionality of the act of Congress prohibiting federal employees from, among other things, "giving to ... any other officer or employe of the government any money or property or other thing of value for political purposes." Ex parte Curtis, 106 U.S. 371, 1 S. Ct. 381, 27 L. Ed. 232 (1882). We believe that for present purposes city employee contributions to the campaign of a city council candidate can be regarded much the same as contributions to the candidate. Further, in light of the considerations militating against having different rules respecting incumbents and challengers and the ever present possibility (in some instances, certainty) that the councilperson elected will not be an incumbent, we also conclude that the City Charter's prohibition in this respect is in no substantial sense subject to a stronger challenge than a prohibition against city employee contributions to city council members for their reelection efforts would be. Moreover, we note that the statute at issue in Ex parte Curtis extended to contributions for any political purpose, while here the prohibition is limited to contributions for candidate elections in which the winner will be, in a sense, the contributor's boss. Accordingly, we view Ex parte Curtis as being at the very least extremely persuasive if not actually controlling on the contribution issue here. Chief Justice Waite's forceful reasoning in that case bears repeating:

"A feeling of independence under the law conduces to faithful public service, and nothing tends more to take away this feeling than a dread of dismissal. If contributions from those in public employment may be solicited by others in official authority, it is easy to see that what begins as a request may end as a demand, and that a failure to meet the demand may be treated by those having the power of removal as a breach of some supposed duty, growing out of the political relations of the parties. Contributions secured under such circumstances will quite as likely be made to avoid the consequences of the personal displeasure of a superior, as to promote the political views of the contributor--to avoid a discharge from service, not to exercise a political privilege." Id. at 373-74, 1 S.Ct. at 384-385.

Additionally, the City's prohibition on campaign contributions is designed to prevent "quid pro quo corruption between a contributor and a candidate," Let's Help Florida v. McCrary, supra, 621 F.2d at 199, as well as the appearance of undue influence. It guards both against less qualified employees being given favored assignments based on either their political contributions or their roles in an employee organization making such contributions, and against discrimination directed at qualified employees because they do not make or bring about such contributions. More generally, it prevents city council members from pressuring city employees, and prevents employees and their organizations from excessively or improperly influencing city council members or candidates. The latter also helps ensure that it does not appear to the public that political "insiders" have undue effect on council elections.

Having identified compelling governmental interests, we must consider whether the absolute ban is a reasonable necessity for achieving those ends. Appellants rely on Bruno v. Garsaud, 594 F.2d 1062, 1064 (5th Cir.1979), in which the Court, in dictum, expressed doubt that the state could constitutionally prohibit "employees, at least when acting as private citizens without any fanfare or publicity, from making contributions to a political candidate or party." These remarks, however, were directed to a statute prohibiting governmental employee contributions "for any political organization or purpose." Id. at 1063 n. 2 (emphasis added).21 We share this doubt as applied to across-the-board bans. But here, while it is absolute, the ban is on contributions in a particular type of election, not all elections. It is directly tied to the City's interests. Simply limiting the allowable amount of a contribution would not prevent employees from being discriminated against or from attempting to achieve favor with council candidates. Moreover, a provision making it unlawful for council members to exploit public employees is not sufficient to achieve the City's interests. Not only does it fail to address the question of undue employee influence, it also does not provide full protection to employees. See note 16, supra.

One might argue that the City is protecting employees who do not wish to be protected.22 This simply overlooks the general public's interest in an independent and efficient civil service. Appellants argue that upholding the lower court will make "political eunuchs" of city employees. However, this is simply not so, as the lower court's findings correctly demonstrate that the City leaves unregulated a considerable scope of city employee political activity. See note 3, supra. Rather than being emasculated, city employees are limited only to an extent that furthers their ability to perform optimally.

AFFIRMED.