Opinion for the Court filed by Circuit Judge WALD.

The National Black Police Association and other organizations and individuals (“plaintiffs”) brought suit to challenge the campaign contribution limits enacted in the District of Columbia (“District” or “D.C.”) by the D.C. Campaign Contributions Limitation Initiative of 1992 (“Initiative 41”). The district court enjoined the contribution limits as unconstitutional and the District filed notice of appeal. During pendency of the District’s appeal, however, legislation passed by the District that significantly increased Initiative 41’s contribution limits became effective. We agree with the District that passage of this legislation has mooted this case and that the district court’s opinion should be vacated.

[*348] I. Background

Initiative 41, which was enacted by voters in the District on November 3, 1992, and took effect on March 17,1993, imposed strict contribution limits on District races. Contributions to candidates for Mayor, D.C. Council Chairman or at-large Council member were limited to a maximum of $100 per candidate, and contributions to ward Council member candidates or Board of Education member candidates were limited to a maximum of $50 per candidate. Similar restrictions were imposed on contributions seeking to recall members from these offices. In addition, Initiative 41 prohibited any contributor from giving more than $600 to all candidates in any one election. This overall cap did not apply to contributions made in regard to initiative, referendum or recall measures. D.C. Law 9-204, D.C.Code § 1-1441.1 (Supp. 1996), amended by D.C. Law 11-144, 43 D.C.Reg. 2174 (1996).

Initiative 41 took effect on March 17,1993. Less than one year later, in February 1994, a bill was introduced in the D.C. Council that proposed raising the campaign contribution limits to $1,000 for the position of Mayor, D.C. Council Chairman and at-large Council member, and $400 for Council ward member and Board of Education positions, but the bill was not enacted. See Jonetta Rose Barras, Campaign Limits Face Repeal, Wash. Times, Mar. 11, 1994, at C6; Yolanda Woodlee, A Quest for Campaign Money, Wash. Post, Feb. 26, 1994, at B4. However, a similar measure was eventually passed by the D.C. Council on April 2, 1996, and signed into law by the Mayor on April 18, 1996. This new campaign legislation increased Initiative 41’s contribution ceilings so that the limits became $2,000 for mayoral candidates, $1,500 for Council Chairman candidates, $1,000 for at-large Council member candidates, $500 for ward Council member candidates and at-large Board of Education member candidates, and $200 for ward Board of Education member candidates, and changed the maximum cap on contributions per election to $8,500. ^[1] D.C. Law 11-144, 43 D.C.Reg. 2174, 2174-75 (1996). These new limits are nearly identical to thosein effect before passage of Initiative 41. ^[2] See D.C.Code§§ l-1441(a), 1-1441(b) (1992 repl.), repealed by D.C. Law 11-144, 43 D.C.Reg. 2174 (1996). The legislation became effective on June 13, 1996, after the required thirty-day period for congressional review. See D.C.Code § 1-233(c)(1) (1992 repl.).

Meanwhile, plaintiffs filed suit on July 6, 1994, seeking to have Initiative 41’s contribution limits enjoined on the grounds that the limits violated the First and Fifth Amendments of the U.S. Constitution and the District of Columbia Self Government and Governmental Reorganization Act of 1973. The suit was initially brought against only the D.C. Board of Elections and Ethics (“Board”), but the district court granted a motion by the District to intervene as a defendant. After the district court denied plaintiffs’ motion for a preliminary injunction and an injunction pending appeal, a five-day bench trial was held in February 1996. On April 19,1996, when the new campaign legislation had been passed by the D.C. Council and signed by the Mayor and was pending congressional review, the district court held that the contribution limits enacted by Initiative 41 were unconstitutional and enjoined their enforcement.

The Board chose not to appeal the district court’s decision, but the District filed a notice of appeal on May 16, 1996. On appeal, however, the District does not challenge the merits of the district court’s decision. Instead, it seeks only to have this court declare the case moot because of passage of the new legislation and vacate the decision below. Plaintiffs, for their part, contend that the case is[*349] either not moot or comes under the exception for cases that are capable of repetition yet evading review. They farther argue that, even if the ease is moot, vacatur is not appropriate here. We turn first to a determination of whether this ease has become moot and, concluding that it has, to the question of whether vacatur should be granted.

II. Mootness

Article III of the Constitution restricts the federal courts to deciding only “actual, ongoing controversies,” Honig v. Doe, 484 U.S. 305, 317, 108 S.Ct. 592, 601, 98 L.Ed.2d 686 (1988), and a federal court has no “power to render advisory opinions [or] ... ‘decide questions that cannot affect the rights of litigants in the ease before them.’ ” Preiser v. Newkirk, 422 U.S. 395, 401, 95 S.Ct. 2330, 2334, 45 L.Ed.2d 272 (1975) (quoting North Carolina v. Rice, 404 U.S. 244, 246, 92 S.Ct. 402, 404, 30 L.Ed.2d 413 (1971)). Moreover, a live controversy must exist at all stages of review. Lewis v. Continental Bank Corp., 494 U.S. 472, 477, 110 S.Ct. 1249, 1253, 108 L.Ed.2d 400 (1990); Preiser, 422 U.S. at 401-02, 95 S.Ct. at 2334-35. Hence, “[e]ven where litigation poses a live controversy when filed, ... [this] court [must] refrain from deciding it if ‘events have so transpired that the decision will neither presently affect the parties’ rights nor have a more-than-speculative chance of affecting them in the future.’ ” Clarke v. United States, 915 F.2d 699, 701 (D.C.Cir.1990) (in banc) (quoting Transwestern Pipeline Co. v. FERC, 897 F.2d 570, 575 (D.C.Cir.1990)). The intervening event arguably ending any live controversy between plaintiffs and the District was the District’s enactment of the new campaign contribution legislation that amended Initiative 41’s contribution limits. Thus, voluntary cessation analysis governs our mootness inquiry. Although generally voluntary cessation of challenged activity does not moot a case, a court may conclude that voluntary cessation has rendered a ease moot if the party urging mootness demonstrates that (1) “there is no reasonable expectation that the alleged violation will recur,” and (2) “interim relief or events have completely or irrevocably eradicated the effects of the alleged violation.” County of Los Angeles v. Davis, 440 U.S. 625, 631, 99 S.Ct. 1379, 1383, 59 L.Ed.2d 642 (1979) (internal quotations omitted); Reeve Aleutian Airways, Inc. v. United States, 889 F.2d 1139, 1142-43 (D.C.Cir.1989); see also Clarke, 915 F.2d at 705 (noting that voluntary cessation analysis has been extended to apply to legislative bodies other than Congress).

Both of these requirements for mootness are satisfied here. To begin with, there is no reasonable expectation that the alleged violation will recur. Although voluntary cessation analysis applies where a challenge to government action is mooted by passage of legislation, the mere power to reenact a challenged law is not a sufficient basis on which a court can conclude that a reasonable expectation of recurrence exists. Rather, there must be evidence indicating that the challenged law likely will be reenacted. See Jones v. Temmer, 57 F.3d 921, 923 (10th Cir.1995) (“[Defendants assért that the claims are not moot because the Colorado legislature remains free to reinstate the old law at a later date. We view this possibility as too conjectural and speculative to avoid a finding of mootness”); Native Village of Noatak v. Blatchford, 38 F.3d 1505, 1510 (9th Cir.1994) (“a statutory change ... is usually enough to render a ease moot, even if the legislature possesses the power to reenact the statute after the lawsuit is dismissed” unless “it is virtually certain that the repealed law will be reenacted”); see also City of Mesquite v. Aladdin’s Castle, Inc., 455 U.S. 283, 296 n. *, 102 S.Ct. 1070, 1078 n. *, 71 L.Ed.2d 152 (1982) (White, J., concurring) (challenge to city ordinance was not mooted by city’s amendment of the ordinance because city announced an intention to reenact ordinance if lower opinion vacated, whereas change in university’s regulations in Princeton Univ. v. Schmid, 455 U.S. 100, 102 S.Ct. 867, 70 L.Ed.2d 855 (1982) (per curiam), had mooted case because “Princeton gave no indication that it desired to return to the original regulatory scheme”). There is no evidence in the record to suggest that the D.C. Council might repeal the new legislation and reenact strict contribution limits. The Council has not “announced ... such an intention,” City of Mesquite, 455 U.S. at 289 n. 11, 102[*350] S.Ct. at 1075 n. 11, and indeed its enactment of the new legislation evinces an intent to the contrary. It is also clear that enactment of the new legislation does not represent a sudden shift in position on the part of the Council; the challenged contribution limits were originally enacted not by any action by the D.C. Council but rather by a voter initiative and legislative efforts to amend Initiative 41 began within one year of the Initiative’s becoming effective.

There is also no reason to conclude that D.C. voters are likely to enact a new initiative that imposes contribution limits akin to those found in Initiative 41. Although Initiative 41’s backers proposed such a new initiative on March 14, 1996, they withdrew this proposed initiative a little over a month later, on April 25, 1996, and no initiative to reimpose strict contribution limits is currently pending. See Joint Appendix at 159; Brian Blomquist, Campaign Cash Cap Contested, Wash. Times, Mar. 17, 1996, at A12. Plaintiffs also point to the high passage rate of Initiative 41 and of popular initiatives generally in the District as demonstrating the likelihood that voters will enact a new initiative incorporating Initiative 41’s limits. But we are not willing to presume that the District’s electorate remains committed to Initiative 41 simply because it was initially passed by a wide margin. Voters may well have changed their minds on the merits of strict contribution limits in light of the public debate that occurred during consideration of the new contribution limits legislation and after seeing how Initiative 41’s limits worked during the 1994 elections. And, needless to say, the fact that voter initiatives are usually very successful is a wholly inadequate basis for us to find that there is a reasonable expectation that a voter initiative incorporating strict contribution limits will be proposed and passed.

Second, passage of the new legislation has completely and irrevocably eradicated the effects of the alleged violation. Plaintiffs sought only to have Initiative 41’s limits declared unconstitutional and enjoined. . But as a result of the new legislation Initiative 41’s limits are no longer in force, and plaintiffs do not contend that these limits continue to have any residual effect. Hence, declaratory and injunctive relief would no longer be appropriate. See Diffenderfer v. Central Baptist Church of Miami, 404 U.S. 412, 414-15, 92 S.Ct. 574, 575-76, 30 L.Ed.2d 567 (1972); see also In re Bunker Ltd. Partnership, 820 F.2d 308, 311 (9th Cir.1987) (“[wjhere intervening legislation has settled a controversy involving only injunctive or declaratory relief, the controversy has become moot”). Nor do plaintiffs suggest that the contribution limits enacted by the new legislation, which are substantially higher than those imposed by Initiative 41 and nearly identical to the limits in place prior to the initiative, are also unconstitutional. Indeed, neither plaintiffs nor the District have presented any arguments to this court regarding the constitutionality of campaign contribution limits in general or the limits imposed by the new legislation or Initiative 41 in particular.

Finally, it is also apparent that this case does not trigger the exception in the mootness doctrine for controversies that are capable of repetition yet evading review. The capable of repetition exception only applies if “ ‘(1) the challenged action was in its duration too short to be fully litigated prior to its cessation or expiration, and (2) there is a reasonable expectation that the same complaining party [will] be subjected to the same action again.’” Doe v. Sullivan, 938 F.2d 1370, 1376 (D.C.Cir.1991) (quoting Weinstein v. Bradford, 423 U.S. 147, 149, 96 S.Ct. 347, 349, 46 L.Ed.2d 350 (1975)) (alterations in original); Clarke, 915 F.2d at 704. There is nothing in the nature of Initiative 41’s contribution limits that inherently prevents review. See Conyers v. Reagan, 765 F.2d 1124, 1128 (D.C.Cir.1985) (proper question to determine whether future challenge would provide an opportunity to fully litigate is “whether the challenged activity is by its very nature short in duration, so that it could not, or probably would not, be able to be adjudicated while fully live”) (internal quotations and additions omitted). Plaintiffs argue that the capable of repetition exception nonetheless applies here because if strict contribution limits akin to Initiative 41’s were enacted by another initiative, the D.C. Council would pass legislation[*351] amending the limits before plaintiffs could obtain judicial review. But there is no reason to presume either that the D.C. Council would pass such legislation again — thereby twice thwarting the popular will of the D.C. electorate — or that it would do so too quickly to allow for judicial review. The latter claim is particularly dubious in light of the fact that it took the Council four years to amend Initiative 41, nearly double the two-year period that we have held is sufficient to sustain an evading review claim. See, e.g., Burlington N. R.R. Co. v. Surface Transp. Bd., 75 F.3d 685, 690 (D.C.Cir.1996) (noting that “both Supreme Court and circuit precedent hold that orders of less than two years’ duration ordinarily evade review”). In any event, as discussed above, there is no reasonable probability that Initiative 41’s limits will be reenacted, so the second requirement of the capable of repetition exception is not met here. See Native Village, 38 F.3d at 1511 (noting that reasonable probability inquiry under voluntary cessation analysis and the capable of repetition exception is the same).

Thus, we conclude that passage of the new campaign contribution legislation amending Initiative 41’s contribution limits has mooted this appeal. We now turn to the question of whether our conclusion that the ease is moot should lead us to vacate the district court’s decision holding Initiative 41’s limits to be unconstitutional.

III. Vacatur

Relying on the Supreme Court’s statement that vacatur “clears the path for future relitigation of the issues between the parties and eliminates a judgment, review of which was prevented by happenstance,” United States v. Munsingwear, Inc., 340 U.S. 36, 39, 71 S.Ct. 104, 106, 95 L.Ed. 36 (1950), for many years lower courts followed a practice of vacating the decision below and dismissing the ease in most instances when a matter was rendered moot pending appeal. However, two years ago in U.S. Bancorp Mortgage Co. v. Bonner Mall Partnership the Supreme Court clarified that vacatur is an equitable remedy, not an automatic right, and held that vacatur is usually inappropriate when “the party seeking relief from the judgment below caused the mootness by voluntary action.” U.S. Bancorp Mortg. Co. v. Bonner Mall Partnership, 513 U.S. 18, 23-25, 115 S.Ct. 386, 391, 130 L.Ed.2d 233 (1994). According to the Court, to grant vacatur automatically in such instances would be to treat judicial precedents as “the property of private litigants” and “would — quite apart from any considerations of fairness to the parties — disturb the orderly operation of the federal judicial system.” Id. at 27, 115 S.Ct. at 392. Instead, where mootness results from voluntary action, vacatur should not be granted unless to do so would serve the public interest. Id. at 25-29, 115 S.Ct. at 392-93; see also Dilley v. Gunn, 64 F.3d 1365, 1370 (9th Cir.1995) (describing Bancorp); National Football League Players Ass’n v. Pro-Football, Inc.; 79 F.3d 1215, 1216-17 (D.C.Cir.1996), vacating in part on rehearing, 56 F.3d 1525, 1530 (D.C.Cir.1995). Bancorp accords with the longstanding practice of this court, under which we have denied vacatur in some instances so as “to avoid unfairness to parties who prevailed in the lower court,” and in particular refused to order vacatur where mootness resulted from “‘the deliberate action of the losing party before the district court.’ ” United States v. Garde, 848 F.2d 1307, 1310 (D.C.Cir.1988) (quoting Center for Science in the Public Interest v. Regan, 727 F.2d 1161, 1165-66 (D.C.Cir.1984)).

The issue we face here is whether Bancorp’s presumption against vacatur should apply where the party seeking relief from the judgment below is the government and the case has been mooted by passage of new legislation. Clearly, the passage of new legislation represents voluntary action, and thus on its face the Bancorp presumption might seem to govern. We believe, however, that application of the Bancorp presumption in this context is not required by the Ban-corp opinion’s rationale and would be inappropriate, at least if there is no evidence indicating that the legislation was enacted in order to overturn an unfavorable precedent. The rationale underlying the Bancorp presumption is that litigants should not be able to manipulate the judicial system by “roll[ing] the dice ... in the district court”[*352] and then “wash[ing] away5’ any “unfavorable outcome55 through use of settlement and va-catur. Bancorp, 513 U.S. at 27-29, 115 S.Ct. at 393; see also Garde, 848 F.2d at 1311 (“We do not wish to encourage litigants who are dissatisfied with the decision[s] of ... trial eourt[s] to have them wiped from the books by merely filing an appeal, then complying with the order or judgment below and petitioning for vacatur.”) (internal quotations omitted). The mere fact that a legislature has enacted legislation that moots an appeal, without more, provides no grounds for assuming that the legislature was motivated by such a manipulative purpose. The legislature may act out of reasons totally independent of the pending lawsuit, or because the lawsuit has convinced it that the existing law is flawed. In American Library Association v. Barr, 956 F.2d 1178 (D.C.Cir.1992), we rejected the argument that vacatur was not appropriate where a case had become moot on appeal due to Congress’ passage of new legislation, arguing that Congress’ action “to repair what may have been a constitutionally defective statute ... represents responsible lawmaking, not manipulation of the judicial process.” Id. at 1187.

Moreover, the respect that courts owe other organs of government should make us wary of impugning the motivations that underlie a legislature’s actions. As we stated in Clarke v. United States, where an appeal had become moot because Congress had allowed the challenged law to sunset and this court in banc granted vacatur, “it would seem inappropriate for the courts either to impute such manipulative conduct to a coordinate branch of government or to apply against that branch a doctrine that appears to rest on the likelihood of a manipulative purpose.” 915 F.2d at 705. This belief that legislative actions are presumptively legitimate animates other areas of the law, such as equal protection jurisprudence and statutory construction doctrines. See, e.g., Washington v. Davis, 426 U.S. 229, 242, 96 S.Ct. 2040, 2048-49, 48 L.Ed.2d 597 (1976) (refusing to hold “that a law, neutral on its face and serving ends otherwise within the power of government to pursue, is invalid under the Equal Protection Clause simply because it may affect a greater proportion of one race than of another”); Edward J. DeBartolo Corp. v. Florida Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 575, 108 S.Ct. 1392, 1397-98, 99 L.Ed.2d 645 (1988) (describing rule of statutory construction providing that courts will construe statutes so as “to avoid [serious constitutional] problems unless such construction is plainly contrary to the intent of Congress,” as based in part on the fact that “courts will ... not lightly assume that Congress intended to infringe constitutional liberties or usurp power constitutionally forbidden it”).

Significantly, the record here demonstrates that the District did not adopt the new legislation in order to erase an unfavorable decision from the books. All of the acts required to be performed by the District in order for a proposed bill to become law— namely passage by the D.C. Council and signing by the Mayor — had occurred before the district court held Initiative 41’s limits to be unconstitutional. The D.C. Council passed the measure on April 2nd and the Mayor signed it on April 18th, the day before the district court issued its opinion. Moreover, legislative efforts to repeal Initiative 41’s strict limits had been ongoing for over two years, and the first unsuccessful bill to amend Initiative 41’s limits was introduced even before this litigation was filed. The only evidence that calls the District’s motivations at all into question is that in this appeal the District has offered no defense of Initiative 41 on the merits; it only argues that this case has become moot and that the decision down below should be vacated. However, a thorough defense of the constitutionality of Initiative 41’s limits would have been a time-consuming endeavor, and we accept the District’s claim that it should not be required to expend resources to defend laws that are no longer in force. ^[3]

[*353] Separation of powers concerns provide further reason to exempt from Ban-corp’s presumption against vacatur the situation of a case which has become moot on appeal due to passage of legislation. As is true of the District, in most multi-branch governments defense of existing laws falls to the executive whereas initiation of legislation is the responsibility of the legislature. Although the executive has the option of refusing to sign legislation, so as to avoid mooting litigation, this option is a hollow one if the executive believes both that the new legislation would be beneficial and that the pending challenge has no merit. To a degree, therefore, the executive branch is in a position akin to a party who finds its case mooted on appeal by “happenstance,” rather than events within its control. This argument suggests that the Bancorp presumption against vacatur might apply if the case has been rendered moot on appeal by enactment or repeal of a regulation, even though the courts accord the executive branch the same presumption of legitimate motive as is given the legislative branch. See, e.g., Cammermeyer v. Perry, 97 F.3d 1235, 1239 (9th Cir.1996) (refusing to vacate lower court decision that became moot on appeal when government reinstated officer and replaced challenged regulation). We need not reach this question, however, and therefore express no views on the matter other than to note the limits of our holding here that the Bancorp presumption is usually inapplicable when legislative action moots a case and the government seeks vacatur.

Plaintiffs use the Tenth Circuit’s decision in 19 Solid Waste Department Mechanics v. City of Albuquerque, 76 F.3d 1142 (1996) (“19 [Solid Waste Department] Mechanics”), to argue against vacating the district court’s decision here. The court in IS Solid Waste Department Mechanics held that a challenge to Albuquerque’s drug testing policy had become moot on appeal because the city adopted a new testing policy, but refused vacatur, citing the Bancorp presumption against vacating where mootness results from voluntary action. Id. at 1144. However, 19 Solid Waste Department Mechanics is clearly distinguishable from this case, since Albuquerque admitted that it had adopted the new policy in response to the district court’s decision enjoining the existing policy as unconstitutional whereas the District’s enactment of new legislation in this case was not motivated by the district court’s decision below. It is also not apparent whether the change in Albuquerque’s drug testing policy represented legislative or regulatory action. In any event, the Tenth Circuit proceeded to grant vacatur in a later case where a challenge to prison conditions had become moot because Albuquerque had complied with a settlement agreement; the court there described the city’s actions as “responsible governmental conduct to be commended.” McClendon v. City of Albuquerque, 100 F.3d 863, 868 (1996). In McClendon, the Tenth Circuit emphasized that the decision whether or not to grant vacatur should be determined on the basis of “particular circumstances,” id. (internal quotations omitted), thus suggesting that 19 Solid Waste Department Mechanics should be viewed as simply a specific instance where refusing vacatur served the public interest and not as establishing a general rule against vacatur where mootness results from voluntary governmental action. See 19 Solid Waste Department Mechanics, 76 F.3d at 1145 (arguing that vacatur “does not appear to serve any interest other than the City’s own”).

Finally, we also note that granting vaca-tur serves the public interest here. The issue in this case was the constitutionality of contribution limits, and it is a well-established principle that courts should avoid unnecessarily deciding constitutional questions. See Ashwander v. TVA 297 U.S. 288, 345-[*354] 47, 56 S.Ct. 466, 482-83, 80 L.Ed. 688 (1936) (Brandéis, J., concurring). Indeed, we have previously “emphasized the avoidance of constitutional questions as one of the bases for vacatur.” Clarke, 915 F.2d at 708; see also Kremens v. Bartley, 431 U.S. 119, 128, 133-34 & n. 15, 97 S.Ct. 1709, 1714-15, 1717-18 & n. 15, 52 L.Ed.2d 184 (1977) (underscoring policy of avoiding unnecessary constitutional decisions in holding that passage of new legislation mooted case and ordering that lower decisions be vacated); In re City of El Paso, 887 F.2d 1103, 1106 (D.C.Cir.1989) (affirming decision to quash subpoenas on grounds of mootness but vacating part of decision that reached constitutional question because of principle of avoiding constitutional questions). Moreover, since the district court’s opinion will remain “on the books” even if vacated, albeit without any preclusive effect, future courts will be able to consult its reasoning.

The presumption of integrity that attaches to legislative action and the difficulties that separation of powers creates for attributing one branch’s actions to another support not applying the Bancorp rule to situations where the party seeking vacatur is the government and mootness results on appeal because of legislative action. In this context, absent additional evidence of an illegitimate motive, we believe the general rule in favor of vacatur still applies. Needless to say, this does not mean that vacatur should be granted in all cases of this kind. As the Court underscored in Bancorp, vacatur is an equitable remedy and the record in particular cases may militate in favor of denying vacatur. Here, however, we find that the equitable balance tips clearly in favor of granting vacatur; the timing of the District’s actions precludes any inference of manipulative intent, and there is no suggestion of such an intent in the record. Vacatur also ensures that a decision on a constitutional matter of great significance and current public interest does not remain in force unreviewed.

We hold that this appeal is moot and that equity would best be served by granting vacatur. We therefore dismiss this appeal as moot, vacate the district court’s decision and remand for this case to be dismissed.

So ordered.