concurring.
My colleagues treat a simple case governed by settled doctrine as if it were complex and novel. Much could be said for their discussion as an original matter, but it is not an original matter. The Supreme Court has held that, without violating the first amendment, a public body may forbid its employees to run for elective office. Clements v. Fashing, 457 U.S. 957, 971-73, 102 S.Ct. 2836, 2847-49, 73 L.Ed.2d 508 (1982); Broadrick v. Oklahoma, 413 U.S. 601, 616-17, 93 S.Ct. 2908, 2918, 37 L.Ed.2d 830 (1973); CSC v. Letter Carriers, 413 U.S. 548, 556, 93 S.Ct. 2880, 2886, 37 L.Ed.2d 796 (1973). Cf. United Public Workers v. Mitchell, 330 U.S. 75, 67 S.Ct. 556, 91 L.Ed. 754 (1947). These cases cover not only civil service employees (Broadrick and Letter Carriers) but also poli-cymaking officials (Clements). Sheriff Ma-han took the lesser step of putting the candidate on unpaid leave until the election was over. Whatever deep problems there may be in the resign-to-run statutes that the Court has upheld, the problems in today’s case are no deeper.
Nonetheless, the majority treats this as a difficult and unsettled issue, hinting broadly that the plaintiff loses only because deputy sheriffs in Illinois are “policymakers” (implying that a civil service employee may not be put on unpaid leave while he runs for elective office) and adding that an elected official may not insist that even policymaking subor*220dinates toe the party line. Much of this discussion is unnecessary given Broadrick and Clements and implicitly contradicts those opinions. Indeed, a reader might infer from portions of the majority’s opinion that Broad-rick came out the other way and that the court therefore needs to distinguish policy-making appointees (who may be dismissed for having the wrong politics) from civil service employees (who may run for political office). Broadrick and Clements show, however, that there is no such dichotomy. I therefore join the judgment but not the opinion.
The majority mentions Broadrick in passing yet proceeds as if the Supreme Court had been mum about the application of the first amendment to resign-to-run rules. It does not cite Clements or Letter Carriers. Mahan ignores them too, concentrating on the defense of official immunity, on which he prevailed in the district court. The majority bypasses immunity; having elected to discuss at length a subject to which the defendant gives only cursory attention, my colleagues ought not proceed as if the whole law must be teased out of the few first amendment eases Mahan cites. That he did not find apt cases in the Supreme Court’s jurisprudence does not require us to make up a new body of doctrine. See United States National Bank of Oregon v. Independent Insurance Agents of America, Inc., — U.S. -, -, 113 S.Ct. 2173, 2177-79, 124 L.Ed.2d 402 (1993); Kamen v. Kemper Financial Services, Inc., — U.S. -, -, 111 S.Ct. 1711, 1718, 114 L.Ed.2d 152 (1991). But cf. Elder v. Holloway, 975 F.2d 1388 (9th Cir.1992), rehearing in bane denied, 984 F.2d 991 (1993), cert. granted, — U.S. -, 113 S.Ct. 3033, 125 L.Ed.2d 721 (1993). That the sheriffs rule cannot be called a “baby Hatch Act” also is irrelevant. Nothing in Broadrick or Letter Carriers depends on the fact that those statutes contained many rules on top of the prohibition against running for public office. So much is clear from Clements, where the prohibition stood alone. The Court treated the absence of other restraints on political activity as an extra reason to find the resign-to-run rule consistent with the Constitution. 457 U.S. at 972, 102 S.Ct. at 2848.
There remains a possibility that this case is special because Mahan’s rule gives the sheriff discretion. According to the majority, “it would be rather a stretch of the Hatch Act cases to interpret them as authorizing a public official selectively to apply a ‘resign to run’ provision to his political enemies”. Why so? Anyone running for his seat is an incumbent’s “enemy,” and Mahan no doubt would have suspended a Republican who sought his job. Treating in a political fashion those who stand for political office is not problematic. Selectivity means that the sheriff might let a deputy run for some other office — say, recorder of deeds or dog catcher. An incumbent planning to retire might even let a deputy run for sheriff. A rule providing that employees of Agency X may run for a job outside the agency, or within it if the incumbent is leaving, increases rather than reduces freedom of speech. The (lawful) alternative, recall, is a blanket prohibition on political activity. To say that the rule is less restrictive than it might be is to make it easier, not harder, to sustain.
There is of course the possibility that a sheriff “might invoke the regulation only against a deputy of whose views he disapproved” (opinion at 216). So, too, an absolute resign-to-run rule could be enforced sporadically, with higher-ups looking the other way when employees whose positions they approve run for office. The possibility of viewpoint discrimination (proper only when the employee holds a policymaking or confidential job) does not undermine the enforcement of a rule against someone who cannot plausibly make such a claim. Wilbur may not invoke the rights of third parties — especially not hypothetical third parties. Although the first amendment’s overbreadth doctrine sometimes permits contentions that are deceptively similar to third-party standing, see Henry P. Monaghan, Overbreadth, 1981 Sup.Ct.Rev. 1, the Court’s conclusion that the first amendment tolerates sweeping resign-to-run rules removes any basis for entertaining a “what-if ... ?” approach to Sheriff Mahan’s regulation.
In addition to doubting a government’s ability to keep civil servants out of politics, *221the majority doubts a politician’s ability to require his subordinates to promote his agenda. Although the majority (correctly) disparages the plaintiffs effort to subject all political discharges to an “open ended inquiry”, it includes a discussion that will have a similar effect, implying that a political official may not condition the employment of even a confidential or policymaking official on adherence to debatable positions unrelated to the job. It is hard for judges to tell what is related to the job in politics; the people are free to vote for reasons we may think beside the point. They may elect a sheriff because of his position on hunting, even though judges can’t see why. And it is easy for discharged (or suspended) employees to say that the particular grounds of the adverse action were not job-related, embroiling public officials in litigation that the judicial branch will handle with a high error rate. The majority’s own hypothetical illustrates these points nicely. How to treat animals is a much-debated question in politics, and sheriffs may have good reasons for wanting deputies to be political ciphers. They may be called on to evict animal-rights protesters surrounding a clinic, and a sheriff would not be comfortable putting an animal-rights activist in charge.
Consider some parallels. Suppose President Clinton decides that because his Administration supports higher taxes, funding for abortions, and other hotly debated positions, all of his personal appointees must support these positions in public. (That is a standard instruction to political subordinates.) An assistant deputy undersecretary in the Department of Energy, whose job has no conceivable relation to abortion or taxes, gives a speech recommending a flat tax or the extension of the Hyde Amendment and is sacked. Must the President appear on the stand in some district court to prove to a jury’s satisfaction that a show of agreement within the executive branch of government is important? Suppose the President instructs all of his top appointees to keep their mouths shut about any issue within the portfolio of some other department — so that only employees of the FDA and few other agencies may discuss experimentation on animals? Presidents, governors, and other chief executives believe that their subordinates work for “the Administration” rather than a particular subunit, and they insist that everyone pitch in for “the Administration’s” program; members of Congress take the same view about their staffs. I assume that any political appointee at the Department of State who puts forward his own health care proposal will discover that his “resignation” has been accepted. Having scorned plaintiffs invitation to hold an inquest into the discharge of a Secretary of HHS, we should not invite equally open-ended litigation about the extent to which holders of high office may consider their subordinates’ statements about the whaling policies of friendly governments, homosexuals in the military, and other political issues of the day.