RUTAN v. REPUBLICAN PARTY OF ILLINOIS

JUSTICE BRENNAN delivered the opinion of the Court.

To the victor belong only those spoils that may be constitutionally obtained. Elrod v. Burns, 427 U.S. 347 (1976), and Branti v. Finkel, 445 U.S. 507 (1980), decided that the First Amendment forbids government officials to discharge or threaten to discharge public employees solely for not being supporters of the political party in power, unless party affiliation is an appropriate requirement for the position involved. Today we are asked to decide the constitutionality of several related political patronage practices — whether promotion, transfer, recall, and hiring decisions involving low-level public employees may be constitutionally based on party affiliation and support. We hold that they may not.

I

The petition and cross-petition before us arise from a lawsuit protesting certain employment policies and practices instituted by Governor James Thompson of Illinois.1 On November 12, 1980, the Governor issued an executive order proclaiming a hiring freeze for every agency, bureau, board, or commission subject to his control. The order prohibits state officials from hiring any employee, filling any vacancy, creating any new position, or taking any similar action. It affects approximately 60,000 state positions. More than 5,000 of these become available each year as a result of resignations, retirements, deaths, expansions, and reorganizations. The order proclaims that "no exceptions" are permitted without the Governor's "express permission after submission of appropriate requests to [his] office." Governor's Executive Order No. 5 (Nov. 12, 1980), Brief for Petitioners and Cross-Respondents 11 (emphasis added).

Requests for the Governor's "express permission" have allegedly become routine. Permission has been granted or withheld through an agency expressly created for this purpose, the Governor's Office of Personnel (Governor's Office). Agencies have been screening applicants under Illinois' civil service system, making their personnel choices, and submitting them as requests to be approved or disapproved by the Governor's Office. Among the employment decisions for which approvals have been required are new hires, promotions, transfers, and recalls after layoffs.

By means of the freeze, according to petitioners and cross-respondents, the Governor has been using the Governor's Office to operate a political patronage system to limit state employment and beneficial employment-related decisions to those who are supported by the Republican Party. In reviewing an agency's request that a particular applicant be approved for a particular position, the Governor's Office has looked at whether the applicant voted in Republican primaries in past election years, whether the applicant has provided financial or other support to the Republican Party and its candidates, whether the applicant has promised to join and work for the Republican Party in the future, and whether the applicant has the support of Republican Party officials at state or local levels.

Five people (including the three petitioners) brought suit against various Illinois and Republican Party officials in the United States District Court for the Central District of Illinois.2 They alleged that they had suffered discrimination with respect to state employment because they had not been supporters of the State's Republican Party and that this discrimination violates the First Amendment. Cynthia B. Rutan has been working for the State since 1974 as a rehabilitation counselor. She claims that since 1981 she has been repeatedly denied promotions to supervisory positions for which she was qualified because she had not worked for or supported the Republican Party. Franklin Taylor, who operates road equipment for the Illinois Department of Transportation, claims that he was denied a promotion in 1983 because he did not have the support of the local Republican Party. Taylor also maintains that he was denied a transfer to an office nearer to his home because of opposition from the Republican Party chairmen in the counties in which he worked and to which he requested a transfer. James W. Moore claims that he has been repeatedly denied state employment as a prison guard because he did not have the support of Republican Party officials.

The two other plaintiffs, before the Court as cross-respondents, allege that they were not recalled after layoffs because they lacked Republican credentials. Ricky Standefer was a state garage worker who claims that he was not recalled, although his fellow employees were, because he had voted in a Democratic primary and did not have the support of the Republican Party. Dan O'Brien, formerly a dietary manager with the mental health department, contends that he was not recalled after a layoff because of his party affiliation and that he later obtained a lower paying position with the corrections department only after receiving support from the chairman of the local Republican Party.

The District Court dismissed the complaint with prejudice, under Federal Rule of Civil Procedure 12(b)(6), for failure to state a claim upon which relief could be granted. 641 F.Supp. 249 (1986). The United States Court of Appeals for the Seventh Circuit initially issued a panel opinion, 848 F.2d 1396 (1988), but then reheard the appeal en banc. The court affirmed the District Court's decision in part and reversed in part. 868 F.2d 943 (1989). Noting that this Court had previously determined that the patronage practice of discharging public employees on the basis of their political affiliation violates the First Amendment, the Court of Appeals held that other patronage practices violate the First Amendment only when they are the "substantial equivalent of a dismissal." Id., at 955. The court explained that an employment decision is equivalent to a dismissal when it is one that would lead a reasonable person to resign. Ibid. The court affirmed the dismissal of Moore's claim because it found that basing hiring decisions on political affiliation does not violate the First Amendment, but remanded the remaining claims for further proceedings.3

Rutan, Taylor, and Moore petitioned this Court to review the constitutional standard set forth by the Seventh Circuit and the dismissal of Moore's claim. Respondents cross-petitioned this Court, contending that the Seventh Circuit's remand of four of the five claims was improper because the employment decisions alleged here do not, as a matter of law, violate the First Amendment. We granted certiorari, 493 U.S. 807 (1989), to decide the important question whether the First Amendment's proscription of patronage dismissals recognized in Elrod v. Burns, 427 U.S. 347 (1976), and Branti v. Finkel, 445 U.S. 507 (1980), extends to promotion, transfer, recall, or hiring decisions involving public employment positions for which party affiliation is not an appropriate requirement.

II

A

In Elrod, supra, we decided that a newly elected Democratic sheriff could not constitutionally engage in the patronage practice of replacing certain office staff with members of his own party "when the existing employees lack or fail to obtain requisite support from, or fail to affiliate with, that party." Id., at 351, 373 (plurality opinion), and 375 (Stewart, J., joined by BLACKMUN, J., concurring in judgment). The plurality explained that conditioning public employment on the provision of support for the favored political party "unquestionably inhibits protected belief and association." Id., at 359. It reasoned that conditioning employment on political activity pressures employees to pledge political allegiance to a party with which they prefer not to associate, to work for the election of political candidates they do not support, and to contribute money to be used to further policies with which they do not agree. The latter, the plurality noted, had been recognized by this Court as "tantamount to coerced belief." Id., at 355 (citing Buckley v. Valeo, 424 U.S. 1, 19 (1976)). At the same time, employees are constrained from joining, working for, or contributing to the political party and candidates of their own choice. Elrod, supra, at 355-356. "[P]olitical belief and association constitute the core of those activities protected by the First Amendment," the plurality emphasized. 427 U. S., at 356. Both the plurality and the concurrence drew support from Perry v. Sindermann, 408 U.S. 593 (1972), in which this Court held that the State's refusal to renew a teacher's contract because he had been publicly critical of its policies imposed an unconstitutional condition on the receipt of a public benefit. See Elrod, supra, at 359 (plurality opinion) and 375 (Stewart, J., concurring in judgment); see also Branti, supra, at 514-516.

The Court then decided that the government interests generally asserted in support of patronage fail to justify this burden on First Amendment rights because patronage dismissals are not the least restrictive means for fostering those interests. See Elrod, supra, at 372-373 (plurality opinion) and 375 (Stewart, J., concurring in judgment). The plurality acknowledged that a government has a significant interest in ensuring that it has effective and efficient employees. It expressed doubt, however, that "mere difference of political persuasion motivates poor performance" and concluded that, in any case, the government can ensure employee effectiveness and efficiency through the less drastic means of discharging staff members whose work is inadequate. 427 U. S., at 365-366. The plurality also found that a government can meet its need for politically loyal employees to implement its policies by the less intrusive measure of dismissing, on political grounds, only those employees in policymaking positions. Id., at 367. Finally, although the plurality recognized that preservation of the democratic process "may in some instances justify limitations on First Amendment freedoms," it concluded that the "process functions as well without the practice, perhaps even better." Patronage, it explained, "can result in the entrenchment of one or a few parties to the exclusion of others" and "is a very effective impediment to the associational and speech freedoms which are essential to a meaningful system of democratic government." Id., at 368-370.4

Four years later, in Branti, supra, we decided that the First Amendment prohibited a newly appointed public defender, who was a Democrat, from discharging assistant public defenders because they did not have the support of the Democratic Party. The Court rejected an attempt to distinguish the case from Elrod, deciding that it was immaterial whether the public defender had attempted to coerce employees to change political parties or had only dismissed them on the basis of their private political beliefs. We explained that conditioning continued public employment on an employee's having obtained support from a particular political party violates the First Amendment because of "the coercion of belief that necessarily flows from the knowledge that one must have a sponsor in the dominant party in order to retain one's job." 445 U. S., at 516. "In sum," we said, "there is no requirement that dismissed employees prove that they, or other employees, have been coerced into changing, either actually or ostensibly, their political allegiance." Id., at 517. To prevail, we concluded, public employees need show only that they were discharged because they were not affiliated with or sponsored by the Democratic Party. Ibid.5

B

We first address the claims of the four current or former employees. Respondents urge us to view Elrod and Branti as inapplicable because the patronage dismissals at issue in those cases are different in kind from failure to promote, failure to transfer, and failure to recall after layoff. Respondents initially contend that the employee petitioners' and cross-respondents' First Amendment rights have not been infringed because they have no entitlement to promotion, transfer, or rehire. We rejected just such an argument in Elrod, 427 U. S., at 359-360 (plurality opinion) and 375 (Stewart, J., concurring in judgment), and Branti, 445 U. S., at 514-515, as both cases involved state workers who were employees at will with no legal entitlement to continued employment. In Perry, 408 U. S., at 596-598, we held explicitly that the plaintiff teacher's lack of a contractual or tenure right to re-employment was immaterial to his First Amendment claim. We explained the viability of his First Amendment claim as follows:

Likewise, we find the assertion here that the employee petitioners and cross-respondents had no legal entitlement to promotion, transfer, or recall beside the point.

Respondents next argue that the employment decisions at issue here do not violate the First Amendment because the decisions are not punitive, do not in any way adversely affect the terms of employment, and therefore do not chill the exercise of protected belief and association by public employees.6 This is not credible. Employees who find themselves in dead-end positions due to their political backgrounds are adversely affected. They will feel a significant obligation to support political positions held by their superiors, and to refrain from acting on the political views they actually hold, in order to progress up the career ladder. Employees denied transfers to workplaces reasonably close to their homes until they join and work for the Republican Party will feel a daily pressure from their long commutes to do so. And employees who have been laid off may well feel compelled to engage in whatever political activity is necessary to regain regular paychecks and positions corresponding to their skill and experience.7

The same First Amendment concerns that underlay our decisions in Elrod, supra, and Branti, supra, are implicated here. Employees who do not compromise their beliefs stand to lose the considerable increases in pay and job satisfaction attendant to promotions, the hours and maintenance expenses that are consumed by long daily commutes, and even their jobs if they are not rehired after a "temporary" layoff. These are significant penalties and are imposed for the exercise of rights guaranteed by the First Amendment. Unless these patronage practices are narrowly tailored to further vital government interests, we must conclude that they impermissibly encroach on First Amendment freedoms. See Elrod, supra, at 362-363 (plurality opinion) and 375 (Stewart, J., concurring in judgment); Branti, supra, at 515-516.

We find, however, that our conclusions in Elrod, supra, and Branti, supra, are equally applicable to the patronage practices at issue here. A government's interest in securing effective employees can be met by discharging, demoting, or transferring staff members whose work is deficient. A government's interest in securing employees who will loyally implement its policies can be adequately served by choosing or dismissing certain high-level employees on the basis of their political views. See Elrod, supra, at 365-368 (plurality opinion); Branti, supra, at 518, and 520, n. 14. Likewise, the "preservation of the democratic process" is no more furthered by the patronage promotions, transfers, and rehires at issue here than it is by patronage dismissals. First, "political parties are nurtured by other, less intrusive and equally effective methods." Elrod, supra, at 372-373 (plurality opinion). Political parties have already survived the substantial decline in patronage employment practices in this century. See Elrod, supra, at 369, and n. 23 (plurality opinion); see also L. Sabato, Goodbye to Good-time Charlie 67 (2d ed. 1983) ("The number of patronage positions has significantly decreased in virtually every state"); Congressional Quarterly Inc., State Government, CQ's Guide to Current Issues and Activities 134 (T. Beyle ed. 1989-1990) ("Linkage[s] between political parties and government office-holding . . . have died out under the pressures of varying forces [including] the declining influence of election workers when compared to media and money-intensive campaigning, such as the distribution of form letters and advertising"); Sorauf, Patronage and Party, 3 Midwest J. Pol. Sci. 115, 118-120 (1959) (many state and local parties have thrived without a patronage system). Second, patronage decidedly impairs the elective process by discouraging free political expression by public employees. See Elrod, 427 U. S., at 372 (plurality opinion) (explaining that the proper functioning of a democratic system "is indispensably dependent on the unfettered judgment of each citizen on matters of political concern"). Respondents, who include the Governor of Illinois and other state officials, do not suggest any other overriding government interest in favoring Republican Party supporters for promotion, transfer, and rehire.

We therefore determine that promotions, transfers, and recalls after layoffs based on political affiliation or support are an impermissible infringement on the First Amendment rights of public employees. In doing so, we reject the Seventh Circuit's view of the appropriate constitutional standard by which to measure alleged patronage practices in government employment. The Seventh Circuit proposed that only those employment decisions that are the "substantial equivalent of a dismissal" violate a public employee's rights under the First Amendment. 868 F. 2d, at 954-957. We find this test unduly restrictive because it fails to recognize that there are deprivations less harsh than dismissal that nevertheless press state employees and applicants to conform their beliefs and associations to some state-selected orthodoxy. See Elrod, supra, at 356-357 (plurality opinion); West Virginia Bd. of Education v. Barnette, 319 U.S. 624, 642 (1943).8 The First Amendment is not a tenure provision, protecting public employees from actual or constructive discharge. The First Amendment prevents the government, except in the most compelling circumstances, from wielding its power to interfere with its employees' freedom to believe and associate, or to not believe and not associate.

Whether the four employees were in fact denied promotions, transfers, or rehires for failure to affiliate with and support the Republican Party is for the District Court to decide in the first instance. What we decide today is that such denials are irreconcilable with the Constitution and that the allegations of the four employees state claims under 42 U. S. C. § 1983 (1982 ed.) for violations of the First and Fourteenth Amendments. Therefore, although we affirm the Seventh Circuit's judgment to reverse the District Court's dismissal of these claims and remand them for further proceedings, we do not adopt the Seventh Circuit's reasoning.

C

Petitioner James W. Moore presents the closely related question whether patronage hiring violates the First Amendment. Patronage hiring places burdens on free speech and association similar to those imposed by the patronage practices discussed above. A state job is valuable. Like most employment, it provides regular paychecks, health insurance, and other benefits. In addition, there may be openings with the State when business in the private sector is slow. There are also occupations for which the government is a major (or the only) source of employment, such as social workers, elementary school teachers, and prison guards. Thus, denial of a state job is a serious privation.

Nonetheless, respondents contend that the burden imposed is not of constitutional magnitude.9 Decades of decisions by this Court belie such a claim. We premised Torcaso v. Watkins, 367 U.S. 488 (1961), on our understanding that loss of a job opportunity for failure to compromise one's convictions states a constitutional claim. We held that Maryland could not refuse an appointee a commission for the position of notary public on the ground that he refused to declare his belief in God, because the required oath "unconstitutionally invades the appellant's freedom of belief and religion." Id., at 496. In Keyishian v. Board of Regents of Univ. of New York, 385 U.S. 589, 609-610 (1967), we held a law affecting appointment and retention of teachers invalid because it premised employment on an unconstitutional restriction of political belief and association. In Elfbrandt v. Russell, 384 U.S. 11, 19 (1966), we struck down a loyalty oath which was a prerequisite for public employment.

Almost half a century ago, this Court made clear that the government "may not enact a regulation providing that no Republican . . . shall be appointed to federal office." Public Workers v. Mitchell, 330 U.S. 75, 100 (1947). What the First Amendment precludes the government from commanding directly, it also precludes the government from accomplishing indirectly. See Perry, 408 U. S., at 597 (citing Speiser v. Randall, 357 U.S. 513, 526 (1958)); see supra, at 72. Under our sustained precedent, conditioning hiring decisions on political belief and association plainly constitutes an unconstitutional condition, unless the government has a vital interest in doing so. See Elrod, 427 U. S., at 362-363 (plurality opinion) and 375 (Stewart, J., concurring in judgment); Branti, 445 U. S., at 515-516; see also Sherbert v. Verner, 374 U.S. 398 (1963) (unemployment benefits); Speiser v. Randall, supra (tax exemption). We find no such government interest here, for the same reasons that we found that the government lacks justification for patronage promotions, transfers, or recalls. See supra, at 71-76.

The court below, having decided that the appropriate inquiry in patronage cases is whether the employment decision at issue is the substantial equivalent of a dismissal, affirmed the trial court's dismissal of Moore's claim. See 868 F. 2d, at 954. The Court of Appeals reasoned that "rejecting an employment application does not impose a hardship upon an employee comparable to the loss of [a] job." Ibid., citing Wygant v. Jackson Bd. of Education, 476 U.S. 267 (1986) (plurality opinion). Just as we reject the Seventh Circuit's proffered test, see supra, at 75-76, we find the Seventh Circuit's reliance on Wygant to distinguish hiring from dismissal unavailing. The court cited a passage from the plurality opinion in Wygant explaining that school boards attempting to redress past discrimination must choose methods that broadly distribute the disadvantages imposed by affirmative-action plans among innocent parties. The plurality said that race-based layoffs placed too great a burden on individual members of the nonminority race, but suggested that discriminatory hiring was permissible, under certain circumstances, even though it burdened white applicants, because the burden was less intrusive than the loss of an existing job. 476 U. S., at 282-284. See also id., at 294-295 (WHITE, J., concurring in judgment).

Wygant has no application to the question at issue here. The plurality's concern in that case was identifying the least harsh means of remedying past wrongs. It did not question that some remedy was permissible when there was sufficient evidence of past discrimination. In contrast, the Governor of Illinois has not instituted a remedial undertaking. It is unnecessary here to consider whether not being hired is less burdensome than being discharged, because the government is not pressed to do either on the basis of political affiliation. The question in the patronage context is not which penalty is more acute but whether the government, without sufficient justification, is pressuring employees to discontinue the free exercise of their First Amendment rights.

If Moore's employment application was set aside because he chose not to support the Republican Party, as he asserts, then Moore's First Amendment rights have been violated. Therefore, we find that Moore's complaint was improperly dismissed.

III

We hold that the rule of Elrod and Branti extends to promotion, transfer, recall, and hiring decisions based on party affiliation and support and that all of the petitioners and cross-respondents have stated claims upon which relief may be granted. We affirm the Seventh Circuit insofar as it remanded Rutan's, Taylor's, Standefer's, and O'Brien's claims. However, we reverse the Seventh Circuit's decision to uphold the dismissal of Moore's claim. All five claims are remanded for proceedings consistent with this opinion.

It is so ordered.

JUSTICE STEVENS, concurring.

While I join the Court's opinion, these additional comments are prompted by three propositions advanced by JUSTICE SCALIA in his dissent. First, he implies that prohibiting imposition of an unconstitutional condition upon eligibility for government employment amounts to adoption of a civil service system. Second, he makes the startling assertion that a long history of open and widespread use of patronage practices immunizes them from constitutional scrutiny. Third, he assumes that the decisions in Elrod v. Burns, 427 U.S. 347 (1976), and Branti v. Finkel, 445 U.S. 507 (1980), represented dramatic departures from prior precedent.

Several years before either Elrod or Branti was decided, I had occasion as a judge on the Court of Appeals for the Seventh Circuit to evaluate each of these propositions. Illinois State Employees Union, Council 34, Am. Federation of State, Cty., and Municipal Employees, AFL-CIO v. Lewis, 473 F.2d 561 (1972), cert. denied, 410 U.S. 928 (1973). With respect to the first, I wrote:

Denying the Governor of Illinois the power to require every state employee, and every applicant for state employment, to pledge allegiance and service to the political party in power is a far cry from a civil service code. The question in this case is simply whether a Governor may adopt a rule that would be plainly unconstitutional if enacted by the General Assembly of Illinois.1

Second, JUSTICE SCALIA asserts that "when a practice not expressly prohibited by the text of the Bill of Rights bears the endorsement of a long tradition of open, widespread, and unchallenged use that dates back to the beginning of the Republic, we have no proper basis for striking it down." Post, at 95; post, at 102 (a "clear and continuing tradition of our people" deserves "dispositive effect"). The argument that traditional practices are immune from constitutional scrutiny is advanced in two plurality opinions that JUSTICE SCALIA has authored, but not by any opinion joined by a majority of the Members of the Court.2

In the Lewis case, I noted the obvious response to this position: "[I]f the age of a pernicious practice were a sufficient reason for its continued acceptance, the constitutional attack on racial discrimination would, of course, have been doomed to failure." 473 F. 2d, at 568, n. 14. See, e. g., Brown v. Board of Education, 347 U.S. 483 (1954).3 I then added this comment on the specific application of that argument to patronage practices:

With respect to JUSTICE SCALIA'S view that until Elrod v. Burns was decided in 1976, it was unthinkable that patronage could be unconstitutional, see post, at 96-97, it seems appropriate to point out again not only that my views in Lewis antedated Elrod by several years, but, more importantly, that they were firmly grounded in several decades of decisions of this Court. As explained in Lewis:

See also American Federation of State, Cty. and Municipal Employees, AFL-CIO v. Shapp, 443 Pa. 527, 537-545, 280 A.2d 375, 379-383 (1971) (Barbieri, J., dissenting).

To avoid the force of the line of authority described in the foregoing passage, JUSTICE SCALIA would weigh the supposed general state interest in patronage hiring against the aggregated interests of the many employees affected by the practice. This defense of patronage obfuscates the critical distinction between partisan interest and the public interest.4 It assumes that governmental power and public resources — in this case employment opportunities — may appropriately be used to subsidize partisan activities even when the political affiliation of the employee or the job applicant is entirely unrelated to his or her public service.5 The premise on which this position rests would justify the use of public funds to compensate party members for their campaign work, or, conversely, a legislative enactment denying public employment to nonmembers of the majority party. If such legislation is unconstitutional — as it clearly would be — an equally pernicious rule promulgated by the executive must also be invalid.

JUSTICE SCALIA argues that distinguishing "inducement and compulsion" reveals that a patronage system's impairment of the speech and associational rights of employees and would-be employees is insignificant. Post, at 109-110. This analysis contradicts the harsh reality of party discipline that is the linchpin of his theory of patronage. Post, at 105 (emphasizing the "link between patronage and party discipline, and between that and party success").6 More importantly, it rests on the long-rejected fallacy that a privilege may be burdened by unconstitutional conditions. See, e. g., Perry v. Sindermann, 408 U.S. 593, 597 (1972). There are a few jobs for which an individual's race or religion may be relevant, see Wygant v. Jackson Bd. of Education, 476 U.S. 267, 314-315 (1986) (STEVENS, J., dissenting); there are many jobs for which political affiliation is relevant to the employee's ability to function effectively as part of a given administration. In those cases — in other words, cases in which "the efficiency of the public service," Public Workers v. Mitchell, 330 U.S. 75, 101 (1947), would be advanced by hiring workers who are loyal to the Governor's party — such hiring is permissible under the holdings in Elrod and Branti. These cases, however, concern jobs in which race, religion, and political affiliation are all equally and entirely irrelevant to the public service to be performed. When an individual has been denied employment for an impermissible reason, it is unacceptable to balance the constitutional rights of the individual against the political interests of the party in power. It seems to me obvious that the government may not discriminate against particular individuals in hopes of advancing partisan interests through the misuse7 of public funds.

The only systemic consideration permissible in these circumstances is not that of the controlling party, but that of the aggregate of burdened individuals. By impairing individuals' freedoms of belief and association, unfettered patronage practices undermine the "free functioning of the electoral process." Elrod, 427 U. S., at 356. As I wrote in 1972:

The tradition that is relevant in these cases is the American commitment to examine and reexamine past and present practices against the basic principles embodied in the Constitution. The inspirational command by our President in 1961 is entirely consistent with that tradition: "Ask not what your country can do for you—ask what you can do for your country." These cases involve a contrary command: "Ask not what job applicants can do for the State—ask what they can do for our party." Whatever traditional support may remain for a command of that ilk, it is plainly an illegitimate excuse for the practices rejected by the Court today.

JUSTICE SCALIA, with whom THE CHIEF JUSTICE and JUSTICE KENNEDY join, and with whom JUSTICE O'CONNOR joins as to Parts II and III, dissenting.

Today the Court establishes the constitutional principle that party membership is not a permissible factor in the dispensation of government jobs, except those jobs for the performance of which party affiliation is an "appropriate requirement." Ante, at 64. It is hard to say precisely (or even generally) what that exception means, but if there is any category of jobs for whose performance party affiliation is not an appropriate requirement, it is the job of being a judge, where partisanship is not only unneeded but positively undesirable. It is, however, rare that a federal administration of one party will appoint a judge from another party. And it has always been rare. See Marbury v. Madison, 1 Cranch 137 (1803). Thus, the new principle that the Court today announces will be enforced by a corps of judges (the Members of this Court included) who overwhelmingly owe their office to its violation. Something must be wrong here, and I suggest it is the Court.

The merit principle for government employment is probably the most favored in modern America, having been widely adopted by civil service legislation at both the state and federal levels. But there is another point of view, described in characteristically Jacksonian fashion by an eminent practitioner of the patronage system, George Washington Plunkitt of Tammany Hall:

It may well be that the Good Government Leagues of America were right, and that Plunkitt, James Michael Curley, and their ilk were wrong; but that is not entirely certain. As the merit principle has been extended and its effects increasingly felt; as the Boss Tweeds, the Tammany Halls, the Pendergast Machines, the Byrd Machines, and the Daley Machines have faded into history; we find that political leaders at all levels increasingly complain of the helplessness of elected government, unprotected by "party discipline," before the demands of small and cohesive interest groups.

The choice between patronage and the merit principle—or, to be more realistic about it, the choice between the desirable mix of merit and patronage principles in widely varying federal, state, and local political contexts—is not so clear that I would be prepared, as an original matter, to chisel a single, inflexible prescription into the Constitution. Fourteen years ago, in Elrod v. Burns, 427 U.S. 347 (1976), the Court did that. Elrod was limited however, as was the later decision of Branti v. Finkel, 445 U.S. 507 (1980), to patronage firings, leaving it to state and federal legislatures to determine when and where political affiliation could be taken into account in hirings and promotions. Today the Court makes its constitutional civil service reform absolute, extending to all decisions regarding government employment. Because the First Amendment has never been thought to require this disposition, which may well have disastrous consequences for our political system, I dissent.

I

The restrictions that the Constitution places upon the government in its capacity as lawmaker, i. e., as the regulator of private conduct, are not the same as the restrictions that it places upon the government in its capacity as employer. We have recognized this in many contexts, with respect to many different constitutional guarantees. Private citizens perhaps cannot be prevented from wearing long hair, but policemen can. Kelley v. Johnson, 425 U.S. 238, 247 (1976). Private citizens cannot have their property searched without probable cause, but in many circumstances government employees can. O'Connor v. Ortega, 480 U.S. 709, 723 (1987) (plurality opinion); id., at 732 (SCALIA, J., concurring in judgment). Private citizens cannot be punished for refusing to provide the government information that may incriminate them, but government employees can be dismissed when the incriminating information that they refuse to provide relates to the performance of their jobs. Gardner v. Broderick, 392 U.S. 273, 277-278 (1968). With regard to freedom of speech in particular: Private citizens cannot be punished for speech of merely private concern, but government employees can be fired for that reason. Connick v. Myers, 461 U.S. 138, 147 (1983). Private citizens cannot be punished for partisan political activity, but federal and state employees can be dismissed and otherwise punished for that reason. Public Workers v. Mitchell, 330 U.S. 75, 101 (1947); Civil Service Comm'n v. Letter Carriers, 413 U.S. 548, 556 (1973); Broadrick v. Oklahoma, 413 U.S. 601, 616-617 (1973).

Once it is acknowledged that the Constitution's prohibition against laws "abridging the freedom of speech" does not apply to laws enacted in the government's capacity as employer in the same way that it does to laws enacted in the government's capacity as regulator of private conduct, it may sometimes be difficult to assess what employment practices are permissible and what are not. That seems to me not a difficult question, however, in the present context. The provisions of the Bill of Rights were designed to restrain transient majorities from impairing long-recognized personal liberties. They did not create by implication novel individual rights overturning accepted political norms. Thus, when a practice not expressly prohibited by the text of the Bill of Rights bears the endorsement of a long tradition of open, widespread, and unchallenged use that dates back to the beginning of the Republic, we have no proper basis for striking it down.1 Such a venerable and accepted tradition is not to be laid on the examining table and scrutinized for its conformity to some abstract principle of First Amendment adjudication devised by this Court. To the contrary, such traditions are themselves the stuff out of which the Court's principles are to be formed. They are, in these uncertain areas, the very points of reference by which the legitimacy or illegitimacy of other practices is to be figured out. When it appears that the latest "rule," or "three-part test," or "balancing test" devised by the Court has placed us on a collision course with such a landmark practice, it is the former that must be recalculated by us, and not the latter that must be abandoned by our citizens. I know of no other way to formulate a constitutional jurisprudence that reflects, as it should, the principles adhered to, over time, by the American people, rather than those favored by the personal (and necessarily shifting) philosophical dispositions of a majority of this Court.

I will not describe at length the claim of patronage to landmark status as one of our accepted political traditions. Justice Powell discussed it in his dissenting opinions in Elrod and Branti. Elrod, supra, at 378-379; Branti, supra, at 522, n. 1. Suffice it to say that patronage was, without any thought that it could be unconstitutional, a basis for government employment from the earliest days of the Republic until Elrod—and has continued unabated since Elrod, to the extent still permitted by that unfortunate decision. See, e. g., D. Price, Bringing Back the Parties 24, 32 (1984); Gardner, A Theory of the Spoils System, 54 Public Choice 171, 181 (1987); Toinet & Glenn, Clientelism and Corruption in the "Open" Society: The Case of the United States, in Private Patronage and Public Power 193, 202 (C. Clapham ed. 1982). Given that unbroken tradition regarding the application of an ambiguous constitutional text, there was in my view no basis for holding that patronage-based dismissals violated the First Amendment—much less for holding, as the Court does today, that even patronage hiring does so.2

II

Even accepting the Court's own mode of analysis, however, and engaging in "balancing" a tradition that ought to be part of the scales, Elrod, Branti, and today's extension of them seem to me wrong.

A

The Court limits patronage on the ground that the individual's interest in uncoerced belief and expression outweighs the systemic interests invoked to justify the practice. Ante, at 68-72. The opinion indicates that the government may prevail only if it proves that the practice is "narrowly tailored to further vital government interests." Ante, at 74.

That strict-scrutiny standard finds no support in our cases. Although our decisions establish that government employees do not lose all constitutional rights, we have consistently applied a lower level of scrutiny when "the governmental function operating ... [is] not the power to regulate or license, as lawmaker, an entire trade or profession, or to control an entire branch of private business, but, rather, as proprietor, to manage [its] internal operatio[ns] . . . ." Cafeteria & Restaurant Workers v. McElroy, 367 U.S. 886, 896 (1961). When dealing with its own employees, the government may not act in a manner that is "patently arbitrary or discriminatory," id., at 898, but its regulations are valid if they bear a "rational connection" to the governmental end sought to be served, Kelley v. Johnson, 425 U. S., at 247.

In particular, restrictions on speech by public employees are not judged by the test applicable to similar restrictions on speech by nonemployees. We have said that "[a] governmental employer may subject its employees to such special restrictions on free expression as are reasonably necessary to promote effective government." Brown v. Glines, 444 U.S. 348, 356, n. 13 (1980). In Public Workers v. Mitchell, 330 U. S., at 101, upholding provisions of the Hatch Act which prohibit political activities by federal employees, we said that "it is not necessary that the act regulated be anything more than an act reasonably deemed by Congress to interfere with the efficiency of the public service." We reaffirmed Mitchell in Civil Service Comm'n v. Letter Carriers, 413 U. S., at 556, over a dissent by Justice Douglas arguing against application of a special standard to Government employees, except insofar as their "job performance" is concerned, id., at 597. We did not say that the Hatch Act was narrowly tailored to meet the government's interest, but merely deferred to the judgment of Congress, which we were not "in any position to dispute." Id., at 567. Indeed, we recognized that the Act was not indispensably necessary to achieve those ends, since we repeatedly noted that "Congress at some time [may] come to a different view." Ibid.; see also id., at 555, 564. In Broadrick v. Oklahoma, 413 U.S. 601 (1973), we upheld similar restrictions on state employees, though directed "at political expression which if engaged in by private persons would plainly be protected by the First and Fourteenth Amendments," id., at 616.

To the same effect are cases that specifically concern adverse employment action taken against public employees because of their speech. In Pickering v. Board of Education of Township High School Dist., 391 U.S. 563, 568 (1968), we recognized:

Because the restriction on speech is more attenuated when the government conditions employment than when it imposes criminal penalties, and because "government offices could not function if every employment decision became a constitutional matter," Connick v. Myers, 461 U. S., at 143, we have held that government employment decisions taken on the basis of an employee's speech do not "abridg[e] the freedom of speech," U. S. Const., Amdt. 1, merely because they fail the narrow-tailoring and compelling-interest tests applicable to direct regulation of speech. We have not subjected such decisions to strict scrutiny, but have accorded "a wide degree of deference to the employer's judgment" that an employee's speech will interfere with close working relationships. 461 U. S., at 152.

When the government takes adverse action against an employee on the basis of his political affiliation (an interest whose constitutional protection is derived from the interest in speech), the same analysis applies. That is why both the Elrod plurality, 427 U. S., at 359, and the opinion concurring in the judgment, id., at 375, as well as Branti, 445 U. S., at 514-515, and the Court today, ante, at 72, rely on Perry v. Sindermann, 408 U.S. 593 (1972), a case that applied the test announced in Pickering, not the strict-scrutiny test applied to restrictions imposed on the public at large. Since the government may dismiss an employee for political speech "reasonably deemed by Congress to interfere with the efficiency of the public service," Public Workers v. Mitchell, supra, at 101, it follows, a fortiori, that the government may dismiss an employee for political affiliation if "reasonably necessary to promote effective government." Brown v. Glines, supra, at 356, n. 13.

While it is clear from the above cases that the normal "strict scrutiny" that we accord to government regulation of speech is not applicable in this field,3 the precise test that replaces it is not so clear; we have used various formulations. The one that appears in the case dealing with an employment practice closest in its effects to patronage is whether the practice could be "reasonably deemed" by the enacting legislature to further a legitimate goal. Public Workers v. Mitchell, 330 U. S., at 101. For purposes of my ensuing discussion, however, I will apply a less permissive standard that seems more in accord with our general "balancing" test: Can the governmental advantages of this employment practice reasonably be deemed to outweigh its "coercive" effects?

B

Preliminarily, I may observe that the Court today not only declines, in this area replete with constitutional ambiguities, to give the clear and continuing tradition of our people the dispositive effect I think it deserves, but even declines to give it substantial weight in the balancing. That is contrary to what the Court has done in many other contexts. In evaluating so-called "substantive due process" claims we have examined our history and tradition with respect to the asserted right. See, e. g., Michael H. v. Gerald D., 491 U.S. 110 (1989); Bowers v. Hardwick, 478 U.S. 186, 192-194 (1986). In evaluating claims that a particular procedure violates the Due Process Clause we have asked whether the procedure is traditional. See, e. g., Burnham v. Superior Court of California, Marin County, 495 U.S. 604 (1990). And in applying the Fourth Amendment's reasonableness test we have looked to the history of judicial and public acceptance of the type of search in question. See, e. g., Camara v. Municipal Court of San Francisco, 387 U.S. 523, 537 (1967). See also Press-Enterprise Co. v. Superior Court of California, Riverside County, 478 U.S. 1, 8 (1986) (tradition of accessibility to judicial proceedings implies judgment of experience that individual's interest in access outweighs government's interest in closure); Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 589 (1980) (BRENNAN, J., concurring in judgment) ("Such a tradition [of public access] commands respect in part because the Constitution carries the gloss of history"); Walz v. Tax Comm'n of New York City, 397 U.S. 664, 678 (1970) ("unbroken practice of according the [property tax] exemption to churches" demonstrates that it does not violate Establishment Clause).

But even laying tradition entirely aside, it seems to me our balancing test is amply met. I assume, as the Court's opinion assumes, that the balancing is to be done on a generalized basis, and not case by case. The Court holds that the governmental benefits of patronage cannot reasonably be thought to outweigh its "coercive" effects (even the lesser "coercive" effects of patronage hiring as opposed to patronage firing) not merely in 1990 in the State of Illinois, but at any time in any of the numerous political subdivisions of this vast country. It seems to me that that categorical pronouncement reflects a naive vision of politics and an inadequate appreciation of the systemic effects of patronage in promoting political stability and facilitating the social and political integration of previously powerless groups.

The whole point of my dissent is that the desirability of patronage is a policy question to be decided by the people's representatives; I do not mean, therefore, to endorse that system. But in order to demonstrate that a legislature could reasonably determine that its benefits outweigh its "coercive" effects, I must describe those benefits as the proponents of patronage see them: As Justice Powell discussed at length in his Elrod dissent, patronage stabilizes political parties and prevents excessive political fragmentation—both of which are results in which States have a strong governmental interest. Party strength requires the efforts of the rank and file, especially in "the dull periods between elections," to perform such tasks as organizing precincts, registering new voters, and providing constituent services. Elrod, 427 U. S., at 385 (dissenting opinion). Even the most enthusiastic supporter of a party's program will shrink before such drudgery, and it is folly to think that ideological conviction alone will motivate sufficient numbers to keep the party going through the off years. "For the most part, as every politician knows, the hope of some reward generates a major portion of the local political activity supporting parties." Ibid. Here is the judgment of one such politician, Jacob Arvey (best known as the promoter of Adlai Stevenson): Patronage is "`a necessary evil if you want a strong organization, because the patronage system permits of discipline, and without discipline, there's no party organization.'" Quoted in M. Tolchin & S. Tolchin, To the Victor 36 (1971). A major study of the patronage system describes the reality as follows:

See also W. Grimshaw, The Political Economy of Machine Politics, 4 Corruption and Reform 15, 30 (1989); G. Pomper, Voters, Elections, and Parties 255 (1988); Wolfinger, Why Political Machines Have Not Withered Away and Other Revisionist Thoughts, 34 J. Politics 365, 384 (1972).

The Court simply refuses to acknowledge the link between patronage and party discipline, and between that and party success. It relies (as did the plurality in Elrod, supra, at 369, n. 23) on a single study of a rural Pennsylvania county by Professor Sorauf, ante, at 75—a work that has been described as "more persuasive about the ineffectuality of Democratic leaders in Centre County than about the generalizability of [its] findings." Wolfinger, supra, at 384, n. 39. It is unpersuasive to claim, as the Court does, that party workers are obsolete because campaigns are now conducted through media and other money-intensive means. Ante, at 75. Those techniques have supplemented but not supplanted personal contacts. See Price, Bringing Back the Parties, at 25. Certainly they have not made personal contacts unnecessary in campaigns for the lower level offices that are the foundations of party strength, nor have they replaced the myriad functions performed by party regulars not directly related to campaigning. And to the extent such techniques have replaced older methods of campaigning (partly in response to the limitations the Court has placed on patronage), the political system is not clearly better off. See Elrod, supra, at 384 (Powell, J., dissenting); Branti, 445 U. S., at 528 (Powell, J., dissenting). Increased reliance on money-intensive campaign techniques tends to entrench those in power much more effectively than patronage—but without the attendant benefit of strengthening the party system. A challenger can more easily obtain the support of party workers (who can expect to be rewarded even if the candidate loses—if not this year, then the next) than the financial support of political action committees (which will generally support incumbents, who are likely to prevail).

It is self-evident that eliminating patronage will significantly undermine party discipline; and that as party discipline wanes, so will the strength of the two-party system. But, says the Court, "[p]olitical parties have already survived the substantial decline in patronage employment practices in this century." Ante, at 74. This is almost verbatim what was said in Elrod, see 427 U. S., at 369. Fourteen years later it seems much less convincing. Indeed, now that we have witnessed, in 18 of the last 22 years, an Executive Branch of the Federal Government under the control of one party while the Congress is entirely or (for two years) partially within the control of the other party; now that we have undergone the most recent federal election, in which 98% of the incumbents, of whatever party, were returned to office; and now that we have seen elected officials changing their political affiliation with unprecedented readiness, Washington Post, Apr. 10, 1990, p. A1, the statement that "political parties have already survived" has a positively whistling-in-the-graveyard character to it. Parties have assuredly survived—but as what? As the forges upon which many of the essential compromises of American political life are hammered out? Or merely as convenient vehicles for the conducting of national Presidential elections?

The patronage system does not, of course, merely foster political parties in general; it fosters the two-party system in particular. When getting a job, as opposed to effectuating a particular substantive policy, is an available incentive for party workers, those attracted by that incentive are likely to work for the party that has the best chance of displacing the "ins," rather than for some splinter group that has a more attractive political philosophy but little hope of success. Not only is a two-party system more likely to emerge, but the differences between those parties are more likely to be moderated, as each has a relatively greater interest in appealing to a majority of the electorate and a relatively lesser interest in furthering philosophies or programs that are far from the mainstream. The stabilizing effects of such a system are obvious. See Toinet & Glenn, Clientelism and Corruption in the "Open" Society, at 208. In the context of electoral laws we have approved the States' pursuit of such stability, and their avoidance of the "splintered parties and unrestrained factionalism [that] may do significant damage to the fabric of government." Storer v. Brown, 415 U.S. 724, 736 (1974) (upholding law disqualifying persons from running as independents if affiliated with a party in the past year).

Equally apparent is the relatively destabilizing nature of a system in which candidates cannot rely upon patronage-based party loyalty for their campaign support, but must attract workers and raise funds by appealing to various interest groups. See Tolchin & Tolchin, To the Victor, at 127-130. There is little doubt that our decisions in Elrod and Branti, by contributing to the decline of party strength, have also contributed to the growth of interest-group politics in the last decade. See, e. g., Fitts, The Vice of Virtue, 136 U. Pa. L. Rev. 1567, 1603-1607 (1988). Our decision today will greatly accelerate the trend. It is not only campaigns that are affected, of course, but the subsequent behavior of politicians once they are in power. The replacement of a system firmly in party discipline with one in which each officeholder comes to his own accommodation with competing interest groups produces "a dispersion of political influence that may inhibit a political party from enacting its programs into law." Branti, supra, at 531 (Powell, J., dissenting).4

Patronage, moreover, has been a powerful means of achieving the social and political integration of excluded groups. See, e. g., Elrod, 427 U. S., at 379 (Powell, J., dissenting); Cornwell, Bosses, Machines and Ethnic Politics, in Ethnic Group Politics 190, 195-197 (H. Bailey, Jr., & E. Katz eds. 1969). By supporting and ultimately dominating a particular party "machine," racial and ethnic minorities have— on the basis of their politics rather than their race or ethnicity —acquired the patronage awards the machine had power to confer. No one disputes the historical accuracy of this observation, and there is no reason to think that patronage can no longer serve that function. The abolition of patronage, however, prevents groups that have only recently obtained political power, especially blacks, from following this path to economic and social advancement.

While the patronage system has the benefits argued for above, it also has undoubted disadvantages. It facilitates financial corruption, such as salary kickbacks and partisan political activity on government-paid time. It reduces the efficiency of government, because it creates incentives to hire more and less qualified workers and because highly qualified workers are reluctant to accept jobs that may only last until the next election. And, of course, it applies some greater or lesser inducement for individuals to join and work for the party in power.

To hear the Court tell it, this last is the greatest evil. That is not my view, and it has not historically been the view of the American people. Corruption and inefficiency, rather than abridgment of liberty, have been the major criticisms leading to enactment of the civil service laws—for the very good reason that the patronage system does not have as harsh an effect upon conscience, expression, and association as the Court suggests. As described above, it is the nature of the pragmatic, patronage-based, two-party system to build alliances and to suppress rather than foster ideological tests for participation in the division of political "spoils." What the patronage system ordinarily demands of the party worker is loyalty to, and activity on behalf of, the organization itself rather than a set of political beliefs. He is generally free to urge within the organization the adoption of any political position; but if that position is rejected he must vote and work for the party nonetheless. The diversity of political expression (other than expression of party loyalty) is channeled, in other words, to a different stage—to the contests for party endorsement rather than the partisan elections. It is undeniable, of course, that the patronage system entails some constraint upon the expression of views, particularly at the partisan-election stage, and considerable constraint upon the employee's right to associate with the other party. It greatly exaggerates these, however, to describe them as a general "`coercion of belief,'" ante, at 71, quoting Branti, 445 U. S., at 516; see also ante, at 75; Elrod, supra, at 355 (plurality opinion). Indeed, it greatly exaggerates them to call them "coercion" at all, since we generally make a distinction between inducement and compulsion. The public official offered a bribe is not "coerced" to violate the law, and the private citizen offered a patronage job is not "coerced" to work for the party. In sum, I do not deny that the patronage system influences or redirects, perhaps to a substantial degree, individual political expression and political association. But like the many generations of Americans that have preceded us, I do not consider that a significant impairment of free speech or free association.

In emphasizing the advantages and minimizing the disadvantages (or at least minimizing one of the disadvantages) of the patronage system, I do not mean to suggest that that system is best. It may not always be; it may never be. To oppose our Elrod-Branti jurisprudence, one need not believe that the patronage system is necessarily desirable; nor even that it is always and everywhere arguably desirable; but merely that it is a political arrangement that may sometimes be a reasonable choice, and should therefore be left to the judgment of the people's elected representatives. The choice in question, I emphasize, is not just between patronage and a merit-based civil service, but rather among various combinations of the two that may suit different political units and different eras: permitting patronage hiring, for example, but prohibiting patronage dismissal; permitting patronage in most municipal agencies but prohibiting it in the police department; or permitting it in the mayor's office but prohibiting it everywhere else. I find it impossible to say that, always and everywhere, all of these choices fail our "balancing" test.

C

The last point explains why Elrod and Branti should be overruled, rather than merely not extended. Even in the field of constitutional adjudication, where the pull of stare decisis is at its weakest, see Glidden Co. v. Zdanok, 370 U.S. 530, 543 (1962) (opinion of Harlan, J.), one is reluctant to depart from precedent. But when that precedent is not only wrong, not only recent, not only contradicted by a long prior tradition, but also has proved unworkable in practice, then all reluctance ought to disappear. In my view that is the situation here. Though unwilling to leave it to the political process to draw the line between desirable and undesirable patronage, the Court has neither been prepared to rule that no such line exists (i. e., that all patronage is unconstitutional) nor able to design the line itself in a manner that judges, lawyers, and public employees can understand. Elrod allowed patronage dismissals of persons in "policymaking" or "confidential" positions. 427 U. S., at 367 (plurality opinion); id., at 375 (Stewart, J., concurring in judgment). Branti retreated from that formulation, asking instead "whether the hiring authority can demonstrate that party affiliation is an appropriate requirement for the effective performance of the public office involved." 445 U. S., at 518. What that means is anybody's guess. The Courts of Appeals have devised various tests for determining when "affiliation is an appropriate requirement." See generally Martin, A Decade of Branti Decisions: A Government Officials' Guide to Patronage Dismissals, 39 Am. U. L. Rev. 11, 23-42 (1989). These interpretations of Branti are not only significantly at variance with each other; they are still so general that for most positions it is impossible to know whether party affiliation is a permissible requirement until a court renders its decision.

A few examples will illustrate the shambles Branti has produced. A city cannot fire a deputy sheriff because of his political affiliation,5 but then again perhaps it can,6 especially if he is called the "police captain."7 A county cannot fire on that basis its attorney for the department of social services,8 nor its assistant attorney for family court,9 but a city can fire its solicitor and his assistants,10 or its assistant city attorney,11 or its assistant state's attorney,12 or its corporation counsel.13 A city cannot discharge its deputy court clerk for his political affiliation,14 but it can fire its legal assistant to the clerk on that basis.15 Firing a juvenile court bailiff seems impermissible,16 but it may be permissible if he is assigned permanently to a single judge.17 A city cannot fire on partisan grounds its director of roads,18 but it can fire the second in command of the water department.19 A government cannot discharge for political reasons the senior vice president of its development bank,20 but it can discharge the regional director of its rural housing administration.21

The examples could be multiplied, but this summary should make obvious that the "tests" devised to implement Branti have produced inconsistent and unpredictable results. That uncertainty undermines the purpose of both the nonpatronage rule and the exception. The rule achieves its objective of preventing the "coercion" of political affiliation, see supra, at 97, only if the employee is confident that he can engage in (or refrain from) political activities without risking dismissal. Since the current doctrine leaves many employees utterly in the dark about whether their jobs are protected, they are likely to play it safe. On the other side, the exception was designed to permit the government to implement its electoral mandate. Elrod, supra, at 367 (plurality opinion). But unless the government is fairly sure that dismissal is permitted, it will leave the politically uncongenial official in place, since an incorrect decision will expose it to lengthy litigation and a large damages award, perhaps even against the responsible officials personally.

This uncertainty and confusion are not the result of the fact that Elrod, and then Branti, chose the wrong "line." My point is that there is no right line—or at least no right line that can be nationally applied and that is known by judges. Once we reject as the criterion a long political tradition showing that party-based employment is entirely permissible, yet are unwilling (as any reasonable person must be) to replace it with the principle that party-based employment is entirely impermissible, we have left the realm of law and entered the domain of political science, seeking to ascertain when and where the undoubted benefits of political hiring and firing are worth its undoubted costs. The answer to that will vary from State to State, and indeed from city to city, even if one rejects out of hand (as the Branti line does) the benefits associated with party stability. Indeed, the answer will even vary from year to year. During one period, for example, it may be desirable for the manager of a municipally owned public utility to be a career specialist, insulated from the political system. During another, when the efficient operation of that utility or even its very existence has become a burning political issue, it may be desirable that he be hired and fired on a political basis. The appropriate "mix" of party-based employment is a political question if there ever was one, and we should give it back to the voters of the various political units to decide, through civil service legislation crafted to suit the time and place, which mix is best.

III

Even were I not convinced that Elrod and Branti were wrongly decided, I would hold that they should not be extended beyond their facts, viz., actual discharge of employees for their political affiliation. Those cases invalidated patronage firing in order to prevent the "restraint it places on freedoms of belief and association." Elrod, 427 U. S., at 355 (plurality opinion); see also id., at 357 (patronage "compels or restrains" and "inhibits" belief and association). The loss of one's current livelihood is an appreciably greater constraint than such other disappointments as the failure to obtain a promotion or selection for an uncongenial transfer. Even if the "coercive" effect of the former has been held always to outweigh the benefits of party-based employment decisions, the "coercive" effect of the latter should not be. We have drawn a line between firing and other employment decisions in other contexts, see Wygant v. Jackson Bd. of Education, 476 U.S. 267, 282-283 (1986) (plurality opinion), and should do so here as well.

I would reject the alternative that the Seventh Circuit adopted in this case, which allows a cause of action if the employee can demonstrate that he was subjected to the "substantial equivalent of dismissal." 868 F.2d 943, 950, 954 (1989). The trouble with that seemingly reasonable standard is that it is so imprecise that it will multiply yet again the harmful uncertainty and litigation that Branti has already created. If Elrod and Branti are not to be reconsidered in light of their demonstrably unsatisfactory consequences, I would go no further than to allow a cause of action when the employee has lost his position, that is, his formal title and salary. That narrow ground alone is enough to resolve the constitutional claims in the present case. Since none of the plaintiffs has alleged loss of his position because of affiliation,22 I would affirm the Seventh Circuit's judgment insofar as it affirmed the dismissal of petitioner Moore's claim and would reverse the Seventh Circuit's judgment insofar as it reversed the dismissal of the claims of other petitioners and of cross-respondents.

The Court's opinion, of course, not only declines to confine Elrod and Branti to dismissals in the narrow sense I have proposed, but, unlike the Seventh Circuit, even extends those opinions beyond "constructive" dismissals—indeed, even beyond adverse treatment of current employees—to all hiring decisions. In the long run there may be cause to rejoice in that extension. When the courts are flooded with litigation under that most unmanageable of standards (Branti) brought by that most persistent and tenacious of suitors (the disappointed officeseeker) we may be moved to reconsider our intrusion into this entire field.

In the meantime, I dissent.