dissenting * in part:
I join all but section III.A. of the per curiam opinion. That section equates a request for amendment with a request for a variance, and concludes that plaintiffs’ taking claim for the post-1984 period is not ripe because plaintiffs did not request an amendment to the Plan. This notion is contrary to prior case law and overlooks a fundamental distinction between variances and amendments.
I
In the regulatory area, not all deprivations of property amount to a taking; compensation is required for only those deprivations that go “too far” and have the same effect as an appropriation of the property through eminent domain or physical possession. See Williamson County Regional Planning Comm’n v. Hamilton Bank, 473 U.S. 172, 199, 105 S.Ct. 3108, 3123, 87 L.Ed.2d 126 (1985); Penn Central Transp. Co. v. New York City, 438 U.S. 104, 124, 98 S.Ct. 2646, 2659, 57 L.Ed.2d 631 (1978). It is therefore critical that a reviewing court know precisely what deprivation a regulation has created. “A court cannot determine whether a regulation has gone ‘too far’ unless it knows how far the regulation goes.” MacDonald, Sommer & Frates v. County of Yolo, 477 U.S. 340, 348, 106 S.Ct. 2561, 2565, 91 L.Ed.2d 285 (1986). Knowing how far the regulation goes requires that “the government entity charged with implementing the regulations [have] reached a final decision regarding the application of the regulations to the property at issue.” Williamson, 473 U.S. at 186, 105 S.Ct. at 3116.
Where the regulation or ordinance in question allows for variances, a “final decision” requires that the agency charged with implementing the law have turned down at least one request for a variance. See, e.g., Williamson, 473 U.S. at 188-89, 193-94, 105 S.Ct. at 3117-18, 3120; Kinzli v. City of Santa Cruz, 818 F.2d 1449, 1454, amended 830 F.2d 968 (9th Cir.1987); Herrington v. County of Sonoma, 834 F.2d 1488, amended 857 F.2d 567, 569-70 (9th Cir.1988). This rule makes some sense. A variance is a means by which a land use regulation can be tailored to meet individual needs. Where such flexibility is built into a law, plaintiffs should first demonstrate that the law nevertheless does not accommodate their demands before they attack it as unconstitutional.
II
The per curiam opinion would extend the variance rule beyond its sensible bounds, by requiring plaintiffs to have asked the TRPA to amend the Plan before coming to court. This extension is at odds with both our prior case law and common sense.
*1345A. In Hall v. City of Santa Barbara, 833 F.2d 1270, 1277 (9th Cir.1987), we held that a taking claim is not precluded because a governing body may amend applicable laws at some time in the future. “A governmental taking can always be undone if the government so chooses. That has never defeated a taking claim.” Id (citing several examples).
In Herrington v. County of Sonoma, 834 F.2d 1488, amended at 857 F.2d 567 (9th Cir.1988), we recognized the distinction between variances and amendments. In that case, the Herringtons did not request a variance from the county. Nonetheless, we held that this failure did not render their taking claim unripe because, under California law, the variance they needed couldn’t have been granted anyway. 857 F.2d at 569-70. A variance request was therefore not required, and the Herring-tons met the finality requirement because “the only means of obtaining approval of [their] proposal was through a General Plan amendment.” Id. at 570 (emphasis original).1 We imposed no additional requirement that the Herringtons attempt to have the plan amended. Yet the per cu-riam opinion would do precisely this — and create a square conflict with our prior decisions in Hall and Herrington.
B. The per curiam opinion seeks to distinguish Hall based on the fact that the Compact here allows land owners to ask for amendments and requires action by the TRPA within 180 days. See per curiam op at 1337-1338; Compact Article V(a)(2). The per curiam opinion perceives no difference between the amendment procedure and the variance procedure. However, a difference there is, and it is fundamental.
A decision denying a variance is final, whereas a decision not to amend a law never is. This is because a request for an amendment is directed to a wholly different process than a request for a variance. A variance may be granted or denied according to an administrative process defined by the challenged law itself. Administrative processes are, by their nature, finite; so long as the governing ordinance or regulation remains unchanged, denial of a variance acts as a bar to further requests of the same type. Moreover, the governing law sets guidelines for what variances are permissible. It is therefore possible for the government entity charged with implementing the regulation to give plaintiffs a “final decision” as to whether their desired land use will be allowed under existing law. Indeed, when it is clear that the governing law does not permit the kind of variance plaintiffs need, they needn’t even ask. Herrington, 857 F.2d at 569-70.
An amendment, on the other hand, requires an exercise of political judgment. Political processes are, by their nature, infinite. A change in the makeup of the legislative body, a shift in the political winds, or even a change in attitude based on further experience or additional wisdom, may be a sufficient reason for a political body to change its mind. In fact, a political body needs no ascertainable reason at all for changing course; it is constrained only by the Constitution and the principles of political accountability. There is thus no way for a court to say that a legislative process *1346has come to rest with respect to a challenged law.2
The per curiam opinion suggests that a court could excuse further amendment requests on the theory that such requests would be futile. See per curiam op at 1337 (stating that “barring futility,” a plaintiffs failure to utilize the amendment procedure renders his claim unripe). But how does one determine futility? Is it ever possible to say that a political institution will fail to change its mind about an amendment that it has already rejected? Are courts even competent to render such a judgment? To require plaintiffs to demonstrate that a governing body has made a final decision about what amendments it will allow is to preclude judicial review altogether, as each day brings a new opportunity for the legislature, in its wisdom, to change the law.3
It makes not the least bit of difference that the Compact here establishes a formalized procedure for proposing amendments. Citizens don’t need the government’s permission to petition for a change in the law. That right is already guaranteed by the Constitution. See U.S. Const., Amend. I.
Nor does the “requirement” that the TRPA act within 180 days provide anything of substance. For one thing, it is entirely hortatory; a political body cannot bind itself to take political action within a specific time frame. Failure to act within 180 days is plainly not judicially reviewable. Thus, the characterization of the statute as “an effective means of submitting a specific course of conduct to TRPA and forcing TRPA to act,” per curiam op at 1337, sounds reassuring, but what does it mean? Although the per curiam opinion talks confidently about how landowners are given a means of “forcing the TRPA to act,” it does not explain just how landowners would go about doing such forcing if the TRPA chose not to act.
In any event, I don’t see how the 180-day requirement distinguishes this case from Hall and Herrington. That the TRPA has given some assurance that it will consider and dispose of amendment requests promptly makes it no more likely that it will act on them favorably. Absent a guarantee that amendments proposed via this special procedure are to be given substantively favored treatment, there is no distinction whatever between amendments proposed according to this procedure and those considered under the more traditional rough-and-tumble processes employed by most legislative bodies.
The per curiam opinion thus would allow procedure to triumph over substance: By providing a meaningless mechanism for petitioning for legislative change, the TRPA would be able to delay or even preclude substantive judicial review of plaintiffs’ constitutional claim. We should probably pause to consider what other constitutional claims can thus be shunted away from the courts and into the political arena, and for how long.
The rejoinder that life is more important than property, per curiam op at 1338 n. 5, obscures the fact that both are protected by the Bill of Rights and for that reason alone deserve solicitude — rather than thinly disguised contempt — from members of the *1347judiciary. The fact is, the Constitution protects a variety of rights and liberties and reasonable minds might differ as to the relative importance of each. When we relegate certain of these to collateral status by refusing to give them the full measure of constitutional protection, we undermine the integrity of the constitutional structure and hand a potent weapon to those who may not share our vision as to which rights trump which.
Although, in the time-honored tradition, this is styled a dissent, it is more in the nature of a disagreement. Because section III.A of the per curiam opinion commands only one vote, it is not, of course, a holding of this court. Nevertheless, since two members of the panel find it necessary to reach the issue of whether plaintiffs’ post-1984 claim is ripe, it is appropriate to set forth our disagreement regarding how that issue should be resolved.
. The per curiam’s attempt to torture the holdings in Kitizli and Herrington into supporting its position, per curiam op at 1337, does not survive a careful review of those cases. The error lies in a failure to recognize that when Kinzli (and Herrington by incorporation) require the landowner to have submitted a "development plan,” they are merely saying that the owner must tell the planning board what he would like to do with his property. This has nothing to do with requesting that the governing law be amended.
The per curiam opinion reaches a conclusion by sleight of hand: "Especially in view of the fact that we also excused the Herringtons’ failure to request a variance on the ground that a variance application would also have been futile, there is simply no basis for asserting that Herrington requires us to treat amendments and variances differently for ripeness purposes.” Id. (emphasis omitted). But this is precisely the contrary of what Herrington says. What we held in Herrington is that once it is clear that the proposed development plan does not fit within the governing law, the owner has exhausted and need not go the next step of asking for a change in the law. Far from treating variances and amendments as equivalent, Her-rington treated them as antithetical.
. Granted, at the local government level the distinction between legislative and administrative action is not as clear-cut as at the state or federal level. But, the per curiam’s trivialization of the issue notwithstanding, the distinction has very significant consequences. For example, when the members of a local governing body act legislatively, they are entitled to absolute immunity, Lake Country Estates, Inc. v. TRPA, 440 U.S. 391, 394, 406, 99 S.Ct. 1171, 1173, 1179, 59 L.Ed.2d 401 (1979) (holding that to the extent that members of the TRPA acted legislatively in adopting a land use ordinance, they were entitled to absolute immunity). When they act administratively, however, they are only entitled to qualified immunity. Cinevision Corp. v. City of Burbank, 745 F.2d 560, 577-78, 580 (9th Cir.1984). By suggesting that it’s all the same ball of wax, we would not only deny a distinction we have held relevant in prior cases, but also undermine the absolute immunity heretofore enjoyed by local government officials acting in a legislative capacity.
. Perhaps the per curiam opinion is suggesting that only one request for amendment is required. But what’s the principled basis for stopping there? Having once failed to persuade a legislative body to adopt a desired amendment, a landowner will still be faced with the possibility that the next attempt will be successful. When is enough, enough?