ORDER

The memorandum disposition filed January 19, 1995 is withdrawn. With the memorandum disposition withdrawn, the Petition for Rehearing and the Motion to Certify Questions of State Law to the Oregon Supreme Court are moot.

OPINION

We must decide whether the City of Portland’s ordinance requiring non-residents to consent to annexation as a condition of receiving a subsidy, or reduction in hook-up costs, for mandated sewer connections, violates their federal constitutional rights to free speech or equal protection.

Michelle and James Hussey along with several of their neighbors are homeowners residing in an area of East Multnomah County outside the City of Portland called “Mid-County” who are required to connect to the City sewer system. When Portland decided to offer connection at a reduced charge to those who signed irrevocable consents to annexation, Hussey sued for declaratory and injunctive relief under 42 U.S.C. §§ 1983 and 1988. On cross-motions for summary judgment, the City prevailed. Hussey appeals, arguing that since consent of a majority of electors (registered voters), as well as homeowners, is required for annexation under Oregon law, the City has infringed the electors’ rights to political speech, and that by imposing financial distress upon those who refuse to consent, the City has violated their rights to equal protection. We hold that Portland’s offer of a subsidy to electors who consent to annexation impermissibly burdens Hussey’s right to vote. We have jurisdiction under 28 U.S.C. § 1291 , and we reverse.

I

In 1986, the State of Oregon’s Environmental Quality Commission ordered the City of Portland to provide sewer services to residents of Mid-County, who must connect to the sewers. The Commission forbade the City from requiring annexation as a condition of using its sewer systems.

Some of the Mid-County area is within the City; some is not. Historically, Portland has charged the actual costs of installing sewers against the properties benefitted. In this case, however, it adopted a program to minimize financial distress faced by City residents, funded by sewer customers within the City, and decided to extend the same program to owners of single family residential property outside the City limits but inside the Portland Urban Services Boundary if they consented to annexation. Hussey is in this group.

City Ordinance 165188, implementing the program, was enacted March 11, 1992. It provides a sewer connection subsidy to homeowners who irrevocably consent to annexation as both landowners and electors. The subsidy, which depends on lot size and the date on which the property owner consents, can be thousands of dollars.

Under Oregon law, annexation may be accomplished in several ways, two of which are relevant here: in a conventional election by a majority of the ballots cast, ORS § 222.120(4)(a); or by the written consent of a majority of all voters registered in the territory to be annexed, and the written consent of owners of a majority of the land in that territory — the so-called “double majority” method. ORS § 199.490(2)(a)(B). Under both approaches, either all the territory will be annexed or none of it will be.

II

Hussey argues that imposing financial distress only on electors who oppose annexation violates his personal right to equal protection under the Fourteenth Amendment because once a governmental entity extends the right to vote on a particular subject to its citizens, that right to vote has the same constitutional significance as the right to vote on any other issue, and that the government therefore should be bound by the same constitutional standards as in a conventional election. Since Oregon law requires the assent of electors for annexation, Hussey submits that Portland may not restrict that form of voting without a compelling state interest. There can be none, in his view, as the integrity of the political process is at stake.

Portland, on the other hand, denies that there is any financial penalty involved and argues that, in any event, as a matter of law there is no inherent state or federal constitutional right to vote on annexation. Further, the City contends that because ORS § 199.490(2)(a)(B), the statute applicable here, does not provide for a right to vote, the constitutional standards that govern a conventional election are not implicated in Hus-sey’s case.

A

As the position of both parties turns on whether consents by electors are the constitutional equivalent of “voting,” we address this issue first. We conclude that they are. Both must be returned by registered voters; both are official expressions of an elector’s will; both are required to resolve political issues; and both require a majority for success. Without the consent of a double majority of registered voters and landowners, Portland would have had to conduct an election to annex Mid-County. And, the consent forms themselves state that if Portland attempts to annex Mid-County by election, “this agreement constitutes a waiver of the right to vote.... [and] such persons shall count as yes votes.”

Portland correctly points out that there is no federal or Oregon state constitutional right to vote on annexation. “There is no federal constitutional right to vote on municipal annexations. We are also unable to find any such right in the Oregon Constitution.” Mid-County Future Alternatives Comm. v. City of Portland, 310 Or. 152 , 795 P.2d 541, 549 , cert. denied, 498 U.S. 999 , 111 S.Ct. 558 , 112 L.Ed.2d 564 (1990) (citation omitted); Hunter v. City of Pittsburgh, 207 U.S. 161, 178-79 , 28 S.Ct. 40, 46 , 52 L.Ed. 151 (1907) (“The State ... at its pleasure, may ... expand or contract the territorial area, unite the whole or a part of it with another municipality.... with or without the consent of the citizens.”). Nevertheless, once citizens are granted the right to vote on a matter, the exercise of that vote becomes protected by the Constitution even though the state was not obliged to allow any vote at all. The Supreme Court has subjected to strict scrutiny statutes governing the eligibility of voters to participate in school district elections, for example. “The need for exacting judicial scrutiny of statutes distributing the franchise is undiminished simply because, under a different statutory scheme, the offices subject to election might have been filled through appointment.” Kramer v. Union Free School Dist., 395 U.S. 621, 628-29 , 89 S.Ct. 1886, 1890 , 23 L.Ed.2d 583 (1969).

Although we acknowledge a state’s considerable latitude in matters of suffrage, we are not persuaded by Portland’s argument that “consents” are not votes but are just “consents.” Labelling cannot be dispositive; otherwise a state could escape the laws protecting voters in a gubernatorial election, for example, merely by declaring that the governorship would be determined by which candidate has the most “consents.”

The best support for Portland’s position that this annexation does not involve voting rights is found in two decisions by other circuits. In Carlyn v. City of Akron, 726 F.2d 287 (6th Cir.1984), the Sixth Circuit held that voting rights were not implicated by an annexation procedure initiated by the petition of a landowner. Although without the landowner’s petition, annexation could only be initiated by a majority vote of the electorate, the ultimate authority for any annexation rested with the county commission ers. It was this fact that the court thought dispositive.

Id. at 289 .

The Fourth Circuit has taken the same view, explaining that although “[i]t is true that three-fourths of the [landowners] in the area to be annexed must request annexation before the ... annexing city may” do so, this procedure does not implicate voting rights because the ultimate annexing authority is vested in the governing board of the annexing city. Berry v. Bourne, 588 F.2d 422, 424 (4th Cir.1978).

As in Berry and Carlyn , ultimate annexation authority in this case rests not with the voters, but with the Oregon Boundary Commission. We decline, however, to follow the reasoning of those cases.

Contrary to the statements in Berry and Carlyn , traditional voting often has no direct, dispositive effect, but rather takes effect only when acted upon by others. For example, voters do not choose the president, the electoral college does. But that does not show that citizens do not vote in presidential elections. Similarly, voter approval of a bond referendum does not compel a municipality to issue the bonds. If other financing is available, or interest rates move adversely, the government may decline to issue the bonds notwithstanding the voters’ authorization. Yet the Supreme Court has explicitly held that municipal bond referendums do involve voting. Cipriano v. City of Houma, 395 U.S. 701, 706 , 89 S.Ct. 1897, 1900 , 23 L.Ed.2d 647 (1969). We decline to hold, therefore, that the annexation proceeding here did not involve voting merely because the Boundary Commission — as well as electors and landowners — would have to approve any boundary changes before they took effect.

Although we do not agree with the reasoning of Berry and Carlyn , we have no quarrel with their results. Neither of the annexation methods at issue in those cases granted electors any say in the proceedings; the consent of landowners alone was required. Without the participation of voters, there can be no voting.

We find more persuasive two state supreme court decisions which held that annexation proceedings similar to the one at issue here were sufficiently similar to voting to be treated as such. In Town of Fond du Lac v. City of Fond du Lac, 22 Wis.2d 533 , 126 N.W.2d 201 (1964), the Wisconsin Supreme Court invalidated an annexation ordinance because some of the signatures on the annexation petition had been coerced: two residents were offered a year’s free rent if they would sign the petition, while two others were told they would be evicted if they failed to sign. The court declared that:

Id., 126 N.W.2d at 203-04 .

The California Supreme Court reached a similar conclusion in Curtis v. Board of Supervisors of Los Angeles County, 7 Cal.3d 942 , 104 Cal.Rptr. 297 , 501 P.2d 537 (1972). California permitted persons owning a majority of land within a proposed annexation area to prevent an annexation election by filing a written protest. Residents who wished to vote in the election challenged the law. Defendants argued that no voting rights were involved since the landowners’ written protest was not an election. The court, however, determined that since the protest prevented an election, it was still subject to strict scrutiny:

Id., 104 Cal.Rptr. at 304 , 501 P.2d at 544 (footnote omitted).

Id. at 306, 501 P.2d at 546 .

Portland may pursue annexation either by calling for an election, which no one disputes would involve voting, or by the consent of a double majority of landowners and registered voters. The conclusion is inescapable that the common thread between these procedures is that both require the consent of the governed. That is what voting is. Because the consent forms are analytically like votes, and are a substitute for them, legally they must be treated as votes.

B

Hussey contends that the Portland ordinance impinges on fundamental “personal rights protected by the Constitution,” and is therefore subject to strict or close scrutiny. City of Cleburne v. Cleburne Living Center, 473 U.S. 432, 440 , 105 S.Ct. 3249, 3254 , 87 L.Ed.2d 313 (1985). The right allegedly impinged on is the right to vote.

While “[i]t is beyond cavil that voting is of the most fundamental significance under our constitutional structure,” Burdick v. Takushi, 504 U.S. 428, 433 , 112 S.Ct. 2059, 2063 , 119 L.Ed.2d 245 (1992) (quotations omitted), courts do not “subject every voting regulation to strict scrutiny.” Id.

“[W]hen a state election law provision imposes only reasonable, nondiscriminatory restrictions upon the First and Fourteenth Amendment rights of voters, the State’s important regulatory interests are generally sufficient to justify the restrictions.” Id. at 434 , 112 S.Ct. at 2063-64 (quotations omitted). However, “when those rights are subjected to severe restrictions, the regulation must be narrowly drawn to advance a state interest of compelling importance.” Id. (quotations omitted).

The Court has “upheld generally applicable and even-handed restrictions that protect the integrity and reliability of the electoral process itself.” Anderson v. Celebrezze, 460 U.S. 780 , 788 n. 9, 103 S.Ct. 1564, 1570 , 75 L.Ed.2d 547 (1983). For example, the Court found reasonable a prohibition on write-in votes where it was relatively easy for candidates to appear on the ballot. Burdick, 504 U.S. at 441 , 112 S.Ct. at 2067 . Similarly permissible are brief residency requirements for voters. Marston v. Lewis, 410 U.S. 679, 680-81 , 93 S.Ct. 1211, 1212 , 35 L.Ed.2d 627 (1973) (fifty-day residency requirement permits county recorder time to certify correctness and completeness of registrations).

Nevertheless, the Court has subjected to strict scrutiny—and struck down as unconstitutional—other statutes that more seriously interfere with the right to vote. See, e.g., Harper v. Virginia Board of Elections, 383 U.S. 663, 670 , 86 S.Ct. 1079, 1083 , 16 L.Ed.2d 169 (1966) (“[W]here fundamental rights and liberties are asserted under the Equal Protection Clause, classifications which might invade or restrain them must be closely scruti-nized_ [T]he right to vote is too precious, too fundamental to be so burdened or conditioned” by a poll tax.); Kramer, 395 U.S. at 633 , 89 S.Ct. at 1892 (granting franchise in school district elections only to property owners and parents of school children fails strict scrutiny).

We have no difficulty concluding that the subsidy here is unlike the reasonable, minor restrictions present in Burdick and Marston , and is similar to, but more distorting than, the poll tax at issue in Harper . In Burdick and Marston , the statutes promoted traditional goals of elections: accurate and complete voter registration, and channeling votes to legitimate candidates. While “there must be a substantial regulation of elections if they are to be fair and honest,” Storer v. Brown, 415 U.S. 724, 730 , 94 S.Ct. 1274, 1279 , 39 L.Ed.2d 714 (1974), the Portland ordinance does not fall into this category of legislation which is subject only to limited scrutiny.

Instead, the Portland ordinance severely and unreasonably interferes with the right to vote. Like the poll tax in Harper which was “closely scrutinized,” 383 U.S. at 670 , 86 S.Ct. at 1083 , the subsidy here disproportionately affects the poor. But unlike a poll tax, which applies regardless of how a voter casts his ballot, this subsidy is conditioned on how an elector votes. In this way, the Portland ordinance is even more distorting than a poll tax.

Portland argues that its ordinance should only be subjected to the rational basis test, relying on Blackwell v. City of St. Charles, 726 F.Supp. 256 (E.D.Mo.1989), aff'd per curiam, 917 F.2d 1150 (8th Cir.1990). Although Blackwell does apply rational basis scrutiny to an annexation scheme similar to Portland’s, the annexation there differed in one critical respect: landowners, and not voters, were asked to consent to annexation. As we scrutinize the Portland ordinance more closely only because it involves the consent of voters, which we equate with voting under the facts of this case, and not because it involves the consents of landowners, Blackwell does not speak to the annexation proceeding at issue here.

Like other “severe restrictions” on the right to vote, therefore, the Portland ordinance can only stand if it is “narrowly drawn to advance a state interest of compelling importance.” Norman v. Reed, 502 U.S. 279, 289 , 112 S.Ct. 698, 705 , 116 L.Ed.2d 711 (1992).

C

Portland does not argue that its ordinance can survive strict scrutiny, and it cannot. While Portland’s stated goals of promoting stability of neighborhoods and aligning service and tax boundaries are certainly legitimate, they are not compelling. Portland is free to charge residents of unincorporated areas its estimated actual cost of servicing them. Moreover, it is not necessary that Portland link its subsidy to a person’s vote. It could offer the subsidy to all Mid-County homeowners as an inducement to vote for annexation, but has chosen not to. The Portland ordinance has, in effect, created a classic Prisoners’ Dilemma, thereby subverting the process through which citizens consent to be governed.

The Portland ordinance fails the close scrutiny test both because Portland’s goals are not “compelling” and because the ordinance is not “narrowly drawn” to achieve that goal. Norman, 502 U.S. at 289 , 112 S.Ct. at 705 . As such, the Portland ordinance is an unconstitutional infringement on the fundamental right to vote. It is designed to distort the political process by granting substantial subsidies based solely on whether a voter consents to annexation, and it cannot stand.

REVERSED.

. Mary Fran Mathis, John Rutherford and Teresa Rutherford. Together they are referred to as "Hussey.” . We do not reach Hussey's other constitutional argument, that the ordinance violates his First Amendment rights to speech.