George M. WALKER; Jere Beckman; Joel Sigrist; Susan M. Newell; and Lee Slocum, Plaintiffs, Appellants, v. EXETER REGION COOPERATIVE SCHOOL DISTRICT; Oyster River Cooperative School District; John Stark Regional School District; Winchester School District; and Amherst-Souhegan Cooperative School District, Defendants, Appellees, 284 F.3d 42

BOUDIN, Chief Judge. This case concerns a challenge on equal protection grounds to the vote requirements for issuance of bonds in certain New Hampshire school districts. For present purposes, school districts in New Hampshire are treated as municipalities. N.H.Rev.Stat. Ann. § 33:8 (2000). While some communities use a “city form” of government where bond issues along with legislation are approved by elected officials {e.g., a city council), id. § 33:9, there are two other types of municipal government recognized by state law where legislation including bond issues requires voter approval.

One such form of organization is the traditional town meeting in which voters assemble on a regular basis to vote upon legislation, including bond issues. N.H.Rev.Stat. Ann. § 39:1. In a variation, other municipalities have retained direct voter control but adopted a so-called “official ballot” regime in which residents vote directly on legislation including bond issues not at a town meeting but by paper ballot at the polls on a designated day. Id. § 40:13 (2000 & Supp.2001). Voters in school districts may alter their form of governance by a three-fifths vote. Id. § 40:14.

Prior to 1999, state law required a super-majority of two-thirds of those voting, either in town meetings or by official ballot, in order to approve local bond issues. N.H.Rev.Stat. Ann. § 33:8. Then, in 1999, the state statute was amended so that the two-thirds vote requirement was continued for town meeting districts, but the super-majority required for approval was reduced to three-fifths in municipalities using the official ballot system. In other words, a different super-majority is required depending on whether the voters vote in a town meeting or official ballot school district.

In June 2001, plaintiffs filed a declaratory judgment action in federal district court in New Hampshire. Each plaintiff claimed to be a resident in one of the defendant school districts using the official ballot method and further claimed that each of the districts after the 1999 amendment had approved bond issues by a vote greater than three-fifths but less than two-thirds.

The complaint sought a declaration that the lesser percentage required in official ballot districts, by contrast to the greater percentage required in town meeting districts, violated plaintiffs’ rights under the Equal Protection Clause of the Fourteenth Amendment by according different weights to the votes of similarly situated voters. The defendant school districts moved to dismiss the complaint for failure to state a claim. Fed.R.Civ.P. 12(b)(6).

They contended that voters in the two different types of district are not similarly situated and, even if they were, the legislature’s vote-percentage distinction is rational and does not violate equal protection standards. In August 2001, the district court granted the motion to dismiss, ruling that voters residing in the two different types of school district are not similarly situated, that “uniformity among a state’s local subdivisions is not a constitutional requisite,” and that no equal protection claim could be based on the different super-majority requirements. Plaintiffs now appeal to this court.

On appeal, plaintiffs’ central argument is that voters in official ballot and town meeting districts are similarly situated because even though each school district votes separately on its own bond issues, school financing in New Hampshire is inherently a “state” matter that requires uniformity across the state. See Claremont Sch. Dist. v. Governor, 142 N.H. 462, 703 A.2d 1353 (N.H.1997).

Defendants say that the local districts are not similarly situated and distinguish Claremont; but they argue in any event that the different super-majority requirements are rationally based and therefore valid even if comparison is warranted.

Supreme Court cases have often described as the basic equal protection principle that “all persons similarly situated should be treated alike.” This may lead one to suppose that an equal protection challenge automatically fails whenever there are substantial differences between the two groups and without further inquiry into whether those differences justify the classification; the New Hampshire Supreme Court’s decision rejecting plaintiffs’ challenge to the 1999 amendment under the state equal protection clause may seem to support this view. McGraw v. Exeter Region Coop. Sch.

Dist., 145 N.H. 709, 765 A.2d 710, 712 (N.H.2001). But so far as federal equal protection law is concerned, this is an over-simplification. Rather, putting aside the special problem of suspect classifications, the underlying equal protection inquiry — at least under federal law — is whether different treatment of two separately classified groups is at least marginally reasonable.

Thus, Supreme Court cases refer to the “similarly situated” rubric as the cardinal principle but then implement it through the “general rule ... that legislation is presumed to be valid and will be sustained if the classification drawn by the statute is rationally related to a legitimate state interest.” Decisions may sometimes use the similarly situated language to conflate the two inquiries — by pointing to a differentiating characteristic so self-evidently a basis for a reasonable classification as to show both dissimilarity and reasonableness at the same time.

Imagine state-enacted water conservation measures directed to any county with a drought but not to any with a rich supply of water. But in our case, merely to point to a difference-in voting mechanisms (between official ballots and town meetings) is to identify a differentiating characteristic without automatically explaining why it justifies the difference in treatment (different super-majorities). This does not make it irrelevant to ask whether two groups are similarly situated. Our own cases do ask this question.

If there are no differences at all, that is likely to be fatal to the classification; and if there are differences between the two groups, this will focus attention on distinctions that may rationally justify the difference in treatment. But unless the distinctions are self-evidently a rational justification for the discrepant treatment — which is not true in our case — the justification question remains to be addressed.

The district court appears to have assumed that a per se rule exists that differential treatment of local units within a state is always and everywhere lawful because they are not similarly situated. There may be a hint of this view in the oldest of the cases invoked, but the more recent case law appears to go no further than to recognize that differential treatment of local units may be easily justified. Holt Civic Club v. City of Tuscaloosa, 439 U.S. 60, 68-70, 99 S.Ct. 383, 58 L.Ed.2d 292 (1978); Salsburg v.

Maryland, 346 U.S. 545, 553 & n. 9, 74 S.Ct. 280, 98 L.Ed. 281 (1954). “Normally” is not the same as “always”; and it does not take too much imagination to conjure up differential treatment by the state of different communities that might well raise equal protection concerns. We note in passing that even if there were an independent similarly situated requirement, plaintiffs would not be helped by Claremont.

There, the state supreme court held that using widely varying property tax rates for schools in different communities violated a state constitutional requirement that state taxes be “proportional and reasonable.” 703 A.2d at 1355 (quoting N.H. Const. pt. 2 art. 5). Although the court applied this requirement to locally prescribed and collected taxes, its concern was based on the perceived impact of local discrepancies on public education within the state. Id. at 1356-57.

In short, Claremont addressed a different constitutional provision (the state’s proportional and reasonable requirement), a different subject (tax rates themselves as opposed to voting regimes for bond issuance), and a different concern (the impact of differential tax rates on state education). That it had nothing to do with judging different voting regimes was made clear by the same court’s later decision in McGraw, which brushed aside plaintiffs’ state equal protection challenge to the 1999 amendment. 765 A.2d at 712.

This brings us to the question of what test should be used in judging the different super-majorities in this ease. Super-majorities are not inherently unlawful for referenda, Gordon v. Lance, 403 U.S. 1, 6-7, 91 S.Ct. 1889, 29 L.Ed.2d 273 (1971), but the issue here is unequal treatment. There is some indication that heightened scrutiny exists where a state distinguishes among voters within the relevant jurisdiction. See, e.g., Harper v. Va. Bd. of Elections, 383 U.S. 663, 665, 86 S.Ct. 1079, 16 L.Ed.2d 169 (1966).

Certainly, the Supreme Court has regularly invalidated schemes of this kind, although more recent cases have sometimes allowed voting to be limited to or weighted unequally toward specially affected residents. But here, New Hampshire has prescribed different voting regimes for decisions made separately in differently structured local communities, though admittedly for the same kind of decision.

The few Supreme Court cases in point indicate that in this situation, equal protection claims are properly tested by asking whether there is a rational basis for the difference in treatment based on some conceivable legitimate legislative objective. This seems fair enough: the votes are on local bond issues and each vote within the district is counted equally. There is no issue here of a state-wide vote tabulated differently in constituent districts. Cf. Bush v. Gore, 531 U.S. 98, 121 S.Ct. 525, 148 L.Ed.2d 388 (2000).

The defendants’ brief refers to legislative history indicating that the 1999 amendment was prompted by several related concerns: first, the need in town meeting districts for a larger super-majority rested in part on the risk of smaller turnouts at a town meeting, whereas in ballot communities voting is much easier, lessening the need for a large super-majority; and, second, such easy voting in the latter districts had led to a substantial decline in bond approvals in such communities because of a passive minority who might not trouble to attend town meetings but would vote no almost automatically on school financing under the official ballot mechanism.

Although legislative history is by no means necessary to sustain a classification as rational, both distinctions articulated by the legislature seem plausible enough to satisfy the lenient rational basis test. See FCC v. Beach Communications, Inc., 508 U.S. 307, 313-15, 113 S.Ct. 2096, 124 L.Ed.2d 211 (1993); McGowan v. Maryland, 366 U.S. 420, 425-26, 81 S.Ct. 1101, 6 L.Ed.2d 393 (1961). If the state legislature had allocated different super-majority requirements at random, this would be a quite different case.

But here, different requirements apply to different methods of voting, and the explanation as to why the different methods might warrant the small adjustment in the super-majority requirement (amounting to an additional one-fifteenth of the vote in town meeting districts) can hardly be described as arbitrary on its face in view of legislative history. We need not pursue the issue much farther because the plaintiffs have failed to make any serious effort to counter this claim of rational basis.

Defendants moved in the district court for a dismissal of the complaint on two separate grounds — the similarly situated rubric and the claim of rational basis. The district court did not reach the second ground, and therefore plaintiffs had no obligation to address the issue in their opening brief. But it is well-settled that on appeal the defendants may defend the judgment below on any ground that would serve. See Roeder v. Alpha Indus., Inc., 814 F.2d 22, 28, 30-31 (1st Cir.1987).

Accordingly, if the plaintiffs did not anticipate this alternative ground and address it in their opening brief, prudence dictated that they counter with a reply brief showing that the defendants were wrong or that the issue was not subject to disposition on the pleadings. Plaintiffs have made no serious effort to anticipate the issue, save for a claim, not seriously developed and so disregarded, Mass. Sch. of Law v. Am. Bar Ass’n, 142 F.3d 26, 43 (1st Cir.1998), that the legislature was im-permissibly trying to dilute the votes of bond issue opponents. And plaintiffs filed no reply brief at all.

Absent any sustained challenge to defendants’ rational basis claims, we have no cause to doubt the defendants’ proffered rationales. In any event, the claim of “impermissible motive” is based on a misunderstanding. All super-majority requirements “dilute” the votes of those who would win if a bare majority sufficed. This is just another way of saying that the legislature (or Constitution drafter for that matter, e.g., U.S. Const. art. II, § 2, cl. 2; id. art.

V) believed that the type of action should be attended by special caution and should be taken only with more support than a simple majority. Examples include the ratification of treaties, amendments to the Constitution, and votes to cut off Senate debate. So, when New Hampshire adopted and then modified a super-majority requirement for local bond issues, we assume with plaintiffs that its purpose was to weight voting and affect outcomes. But, contrary to plaintiffs’ assumption, this does not, standing alone, condemn the measure or invite some form of special scrutiny. See Gordon v.

Lance, 403 U.S. at 6-7, 91 S.Ct. 1889, 29 L.Ed.2d 273. Still less does the legislative action taken here — a reduction in the super-majority requirement (from two-thirds to three-fifths) for official ballot districts — require special justification. Affirmed. . The state itself was not joined as a defendant. Because we find that the challenged statute does not violate equal protection, we bypass the question whether the state Attorney General is a necessary party. See Fed. R.Civ.P. 19(a); cf. Norwood v. Harrison, 581 F.2d 518, 519 (5th Cir.1978); Liquifin Aktiengesellschaft v.

Brennan, 383 F.Supp. 978, 983-84 (S.D.N.Y.1974). . City of Cleburne v. Cleburne Living Ctr., 473 U.S. 432, 439, 105 S.Ct. 3249, 87 L.Ed.2d 313 (1985); see also Plyler v. Doe, 457 U.S. 202, 216, 102 S.Ct. 2382, 72 L.Ed.2d 786 (1982) ("The Equal Protection Clause directs that ‘all persons similarly circumstanced shall be treated alike.’ ” (quoting F.S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415, 40 S.Ct. 560, 64 L.Ed. 989 (1920))); Atchison, Topeka & Santa Fe R.R. Co. v.

Matthews, 174 U.S. 96, 104, 19 S.Ct. 609, 43 L.Ed. 909 (1899) ("[T]he equal protection guaranteed by the Constitution forbids the legislature to select a person, natural or artificial, and impose upon him or it burdens and liabilities which are not cast upon others similarly situated.”). .

Cleburne, 473 U.S. at 440, 105 S.Ct. 3249; accord Plyler, 457 U.S. at 216, 102 S.Ct. 2382 (usual requirement is "that the classification at issue bears some fair relationship to a legitimate public purpose”); Tribe, American Constitutional Law §§ 16-1 to 16-2 (2d ed.1988); Tussman & tenBroek, The Equal Protection of the Laws, 37 Cal. L.Rev. 341, 346 (1949). . See Barrington Cove Ltd. P’ship v. R.I. Hous. & Mortgage Fin. Corp., 246 F.3d 1, 8 (1st Cir.2001); Rubinovitz v. Rogato, 60 F.3d 906, 909-10 (1st Cir.1995). The similarly situated rubric was also used in The Dartmouth Review v.

Dartmouth College, 889 F.2d 13, 19 (1st Cir.1989), for a quite different purpose, namely, to refute a factual inference of race discrimination. Once again, the same words may play different roles in different contexts. . Missouri v. Lewis, 101 U.S. 22, 30, 25 L.Ed. 989 (1879) (stating that equal protection "has not respect to local and municipal regulations that do not injuriously affect or discriminate between persons or classes of persons within the places or municipalities for which such regulations are made ” (emphasis added)).

Even in Lewis, however, the Court went on to address justification, saying that the state's establishment of a special court of appeals with exclusive jurisdiction over appeals from certain counties was justified by "[c]onve-nience, if not necessity.” Id. at 30-31. . Compare, e.g., Hill v. Stone, 421 U.S. 289, 95 S.Ct 1637, 44 L.Ed.2d 172 (1975) (invalidating "dual box” municipal bond election procedure that required majority of property taxpayers and majority of all voters), and Gray v.

Sanders, 372 U.S. 368, 83 S.Ct. 801, 9 L.Ed.2d 821 (1963) (invalidating Georgia’s county-unit method of voting, which gave disproportionate weight to rural voters), with Town of Lockport v. Citizens for Cmty. Action, 430 U.S. 259, 97 S.Ct. 1047, 51 L.Ed.2d 313 (1977) (holding that divergent interests of cities and towns justified requirement of majority of voters in both to approve new county charter), and Salyer Land Co. v.

Tulare Lake Basin Water Dist., 410 U.S. 719, 93 S.Ct. 1224, 35 L.Ed.2d 659 (1973) (allowing election for water storage board, which oversaw distribution of water to farms, to be limited to landowners). See generally Gelfand, Federal Constitutional Law and American Local Government § 1-3 (1984); Briffaull, Who Rules at Home?: One Person/One Vote and Local Governments, 60 U. Chi. L.Rev. 339, 345-369 (1993). . Holt, 439 U.S. at 68-70, 99 S.Ct. 383; Salsburg, 346 U.S. at 553 & n. 9, 74 S.Ct. 280; see Neuman, Territorial Discrimination, Equal Protection, and Self-Determination, 135 U. Pa. L.Rev. 261, 273-75 (1987). .

See, e.g., Hearing on H.B. 487 Before the Senate Comm. on Education, 1999 Gen. Ct., 156th Sess. (N.H. June 9, 1999) (statements of Rep. Lynde and Dean Michener, Dir. of Gov't Relations, N.H. School Boards Ass’n); Public Hearing on H.B. 487 Before the House Comm, on Education, 1999 Gen. Ct., 156th Sess. (N.H. Mar. 24, 1999) (statement of Rep. Estabrook).