Although Milagros Irizarry has lived with the same man for more than two decades and they have two (now adult) children, they have nevermarried. As an employee of the Chicago public school system, she receives health benefits but he does not, even though he is her "domestic partner" (the term for persons who are cohabiting with each other in a relationship similar to marriage), though he would if he were her husband. In July 1999, the Chicago Board of Education extended spousal health benefits to domestic partners--but only if the domestic partner was of the same sex as the employee, which excluded Irizarry's domestic partner, an exclusion that she contends is unconstitutional.

Besides being of the same sex, applicants for domestic-partner status must be unmarried, unrelated, at least 18 years old, and "each other's sole domestic partner, responsible for each other's common welfare." They must satisfy two of the following four additional conditions as well: that they have been living together for a year; that they jointly own their home; that they jointly own other property of specified kinds; that the domestic partner is the primary beneficiary named in the employee's will. Although the board's purpose in entitling domestic partners so defined to spousal benefits was to extend such benefits to homosexual employees, homosexual marriage not being recognized by Illinois, 750 ILCS 5/212, 5/213.1, entitlement to the benefits does not require proof of sexual orientation.

Irizarry's domestic partner satisfies all the conditions for domestic-partner benefits except being of the same sex. She argues that the board's policy denies equal protection and, secondarily, due process. The district court dismissed her suit for failure to state a claim.

The board of education makes two arguments for treating homosexual couples differently from unmarried heterosexual couples. First, since homosexual marriage is not possible in Illinois (or anywhere else in the United States, though it is now possible in the Netherlands), and heterosexual marriage of course is, the recognition of a domestic-partnership surrogate is more important for homosexual than for heterosexual couples, who can obtain the benefits simply by marrying. Second, the board wants to attract homosexual teachers in order to provide support for homosexual students. Cf. Crawford v. City of Chicago, 710 N.E.2d 91, 98-99 (Ill. App. 1999). According to its brief, the board "believes that lesbian and gay male school personnel who have a healthy acceptance of their own sexuality can act as role models and provide emotional support for lesbian and gay students. . . . They can support students who are questioning their sexual identities or who are feeling alienated due to their minority sexual orientation. They can also encourage all students to be tolerant and accepting of lesbians and gay males, and discourage violence directed at these groups."

This line of argument will shock many people even today; it was not that long ago when homosexual teachers were almost universally considered a public menace likely to seduce or recruit their students into homosexuality, then regarded with unmitigated horror. The plaintiff does not argue, however, that the Chicago Board of Education is irrational in having turned the traditional attitude toward homosexual teachers upside down. It is not for a federal court to decide whether a local government agency's policy of tolerating or even endorsing homosexuality is sound. Even if the judges consider such a policy morally repugnant--even dangerous--they may not interfere with it unless convinced that it lacks even minimum rationality, which is a permissive standard. It is a fact that some schoolchildren are homosexual, and the responsibility for dealing with that fact is lodged in the school authorities, and (if they are public schools) ultimately in the taxpaying public, rather than in the federal courts.

The efficacy of the policy may be doubted. Although it had been in effect for a year and a half when the appeal was argued, only nine employees out of some 45,000 had signed up for domestic-partner benefits and none of the nine indicated whether he or she was homosexual; they may not all have been, as we shall see-- perhaps none were. Nor is there any indication that any of the nine are new employees attracted to teach in the Chicago public schools by the availability of health benefits for same- sex domestic partners. Maybe it's too early, though, to assess the efficacy of the policy. No matter; limited efficacy does not make the policy irrational--not even if we think limited efficacy evidence that the policy is more in the nature of a political gesture than a serious effort to improve the lot of homosexual students--if only because with limited efficacy comes limited cost. Because homosexuals are a small fraction of the population, because the continuing stigma of homosexuality discourages many of them from revealing their sexual orientation, and because nowadays a significant number of heterosexuals substitute cohabitation for marriage in response to the diminishing stigma of cohabitation, extending domestic-partner benefits to mixed-sex couples would greatly increase the expense of the program.

Irizarry argues that the child of an unmarried couple ought equally to be entitled to the mentoring and role-model benefits of having teachers who live in the same way as the student's parents do. Cost considerations to one side, the argument collides with a nationwide policy in favor of marriage. True, it is no longer widely popular to try to pressure homosexuals to marry persons of the opposite sex. But so far as heterosexuals are concerned, the evidence that on average married couples live longer, are healthier, earn more, have lower rates of substance abuse and mental illness, are less likely to commit suicide, and report higher levels of happiness--that marriage civilizes young males, confers economies of scale and of joint consumption, minimizes sexually transmitted disease, and provides a stable and nourishing framework for child rearing--see, e.g., Linda J. Waite & Maggie Gallagher, The Case for Marriage: Why Married People Are Happier, Healthier, and Better Off Financially (2000); David Popenoe, Life without Father: Compelling New Evidence That Fatherhood and Marriage Are Indispensable for the Good of Children and Society (1996); George W. Dent, Jr., "The Defense of Traditional Marriage," 15 J.L. & Pol. 581 (1999), refutes any claim that policies designed to promote marriage are irrational. The Chicago Board of Education cannot be faulted, therefore, for not wishing to encourage heterosexual cohabitation; and, though we need not decide the point, the refusal to extend domestic-partner benefits to heterosexual cohabitators could be justified on the basis of the policy favoring marriage for heterosexuals quite apart from the reasons for wanting to extend the spousal fringe benefits to homosexual couples.

Of course, self-selection is important; people are more likely to marry who believe they have characteristics favorable to a long-term relationship. Lee A. Lillard & Constantijn W.A. Panis, "Marital Status and Mortality: The Role of Health," 33 Demography 313 (1996); Lee A. Lillard, Michael J. Brien & Linda J. Waite, "Premarital Cohabitation and Subsequent Dissolution: A Matter of Self- Selection?" 32 Demography 437 (1995). But the Chicago Board of Education would not be irrational (though it might be incorrect) in assigning some causal role to the relationship itself. Linda J. Waite, "Does Marriage Matter?" 32 Demography 483, 498-99 (1995), finds that

cohabitants are much less likely than married couples to pool financial resources, more likely to assume that each partner is responsible for supporting himself or herself financially, more likely to spend free time separately, and less likely to agree on the future of the relationship. This makes both investment in the relationship and specialization with this partner much riskier than in marriage, and so reduces them. Whereas marriage connects individuals to other important social institutions, such as organized religion, cohabitation seems to distance them from these institutions.

Irizarry and her domestic partner may, given the unusual duration of their relationship, be an exception to general izations about the benefits of marriage. We are not aware of an extensive scholarly literature comparing marriage to long-term cohabitation. This may be due to the fact that long-term cohabitation is rare--only ten percent of such relationships last for five years or more, Pamela J. Smock, "Cohabitation in the United States: An Appraisal of Research Themes, Findings, and Implications," 26 Ann. Rev. Sociology 1 (2000). But there is evidence that the widespread substitution of cohabitation for marriage in Sweden has given that country the highest rate of family dissolution and single parenting in the developed world. David Popenoe, Disturbing the Nest: Family Change and Decline in Modern Societies 173-74 (1988). It is well known that divorce is harmful to children (see Jonathan Gruber, "Is Making Divorce Easier Bad for Children? The Long Run Implications of Unilateral Divorce," NBER Working Paper No. 7968 (Oct. 2000), for a survey of the evidence), and presumably the same is true for the dissolution of a cohabitation--and a cohabitation is more likely to dissolve than a marriage. True, Irizarry's cohabitation has not dissolved; but law and policy are based on the general rather than the idiosyncratic, as the Supreme Court noted with reference to other benefits tied to marital status in Califano v. Jobst, 434 U.S. 47, 53-54 (1977). Nor is it entirely clear that this couple ought to be considered an exception to the general concern with heterosexuals who choose to have a family outside of marriage. For when asked at argument why the couple had never married, Irizarry's counsel replied that he had asked his client that question and she had told him that "it just never came up." There may be good reasons why a particular couple would not marry even after producing children, but that the thought of marriage would not even occur to them is disquieting.

The Lambda Legal Defense and Education Fund has filed an amicus curiae brief surprisingly urging reversal-- surprisingly because Lambda is an organization for the promotion of homosexual rights, and if it is the law that domestic-partnership benefits must be extended to heterosexual couples, the benefits are quite likely to be terminated for everyone lest the extension to heterosexual cohabitors impose excessive costs and invite criticism as encouraging heterosexual cohabitation and illegitimate births and discouraging marriage and legitimacy. But Lambda is concerned with the fact that state and national policy encourages (heterosexual) marriage in all sorts of ways that domestic-partner health benefits cannot begin to equalize. Lambda wants to knock marriage off its perch by requiring the board of education to treat unmarried heterosexual couples as well as it treats married ones, so that marriage will lose some of its luster.

This is further evidence of the essentially symbolic or political rather than practical significance of the board's policy. Lambda is not jeopardizing a substantial benefit for homosexuals because very few of them want or will seek the benefit. In any event, it would not be proper for judges to use the vague concept of "equal protection" to undermine marriage just because it is a heterosexual institution. The desire of the board of education to increase the employment of homosexual teachers is admittedly a striking manifestation of the sexual revolution that has character ized, some would say convulsed, the United States in the last forty years. The courts did not try to stop the revolution. On the contrary, they spurred it on, most pertinently to this case by their decisions removing legal disabilities of birth out of wedlock, e.g., Jimenez v. Weinberger, 417 U.S. 628 (1974); Gomez v. Perez, 409 U.S. 535 (1973) (per curiam); Weber v. Aetna Casualty & Surety Co., 406 U.S. 164 (1972); Glona v. American Guarantee & Liability Insurance Co., 391 U.S. 73 (1968); Levy v. Louisiana, 391 U.S. 68 (1968), disabilities that if they still existed might have induced Ms. Irizarry and the father of her children to marry in order to remove those disabilities from their children. Likewise relevant are cases such as Stanley v. Illinois, 405 U.S. 645 (1972), that confer constitutional rights on unwed fathers. But no court has gone so far as to deem marriage a suspect classification because government provides benefits to married persons that it withholds from cohabiting couples. That would be a bizarre extension of case law already criticized as having carried the courts well beyond the point at which the Constitution might be thought to provide guidance to social policy.

To the board's argument that it has extended spousal benefits to the domestic partners of homosexual employees because homosexual marriage is not a status available to its employees, Irizarry replies that the argument depends on the board's groundless decision to provide benefits to spouses, rather than domestic partners, of its employees. She says that all the board has to do to purge the constitutional violation is to condition all nonemployee fringe benefits on satisfaction of its domestic-partnership conditions other than that the domestic partner be of the same sex as the employee; and then the "discrimination" in favor of heterosexuals that the extension of spousal benefits to homosexual domestic partners was intended to erase will be eliminated without discrimination against heterosexual domestic partners. She points to Chicago's Human Rights Ordinance, which forbids discrimination on the basis of marital status. But the purpose, at least the primary purpose, of such a prohibition is surely not to dethronemarriage; it is to prevent discrimination against married women, who employers might think have divided loyalties. Such laws are pro-marriage, not anti- as the plaintiff suggests.

All other considerations to one side, the board reaps cost savings by basing dependent benefits on marital status-- savings distinct from those discussed earlier that depend simply on the much smaller number of homosexuals than heterosexuals likely to seek or qualify for domestic-partner benefits. It is easier to determine whether the claimant is married to an employee than to determine whether the claimant satisfies the multiple criteria for domestic partnership. Earlier we took for granted that cost is an admissible consideration in evaluating the rationality of a classification; here we add that the cases so hold. Bankers Life & Casualty Co. v. Crenshaw, 486 U.S. 71, 83-84 (1988); LaGuerre v. Reno, 164 F.3d 1035, 1041 (7th Cir. 1998); DeSousa v. Reno, 190 F.3d 175, 185 (3d Cir. 1999); Silver v. Baggiano, 804 F.2d 1211, 1218-19 (11th Cir. 1986). And we do not understand the plaintiff to be arguing that the board of education must have anything more than a rational basis for its action in order to defeat the plaintiff's equal protection claim. Only when the plaintiff in an equal protection case is complaining of a form of discrimination that is suspect because historically it was irrational or invidious is there a heavier burden of justifying a difference in treatment than merely showing that it is rational. E.g., Kimel v. Florida Bd. of Regents, 528 U.S. 62, 83-84 (2000); Milner v. Apfel, 148 F.3d 812, 815-16 (7th Cir. 1998); Miller v. United States, 73 F.3d 878, 881-82 (9th Cir. 1995); Disabled American Veterans v. Dept. of Veterans Affairs, 962 F.2d 136, 141-42 (2d Cir. 1992). Heterosexuals cohabiting outside of marriage are not such a class. There is a history of disapproval of (nonmarital) cohabitation, and some states still criminalize it. See, e.g., Ariz. Rev. Stat. Ann. sec. 13-1409; Mich. Comp. Laws Ann. sec. 750.355; N.D. Cent. Code sec. 12.1-20-10--as indeed Illinois did until 1990. United States v. Nichols, 937 F.2d 1257, 1263 (7th Cir. 1991). But the disapproval is not necessarily irrational or invidious, Doe v. Duling, 782 F.2d 1202, 1207 (4th Cir. 1986), given the benefits of marriage discussed earlier. It was rational for the board to refuse to extend domestic-partnership benefits to persons who can if they wish marry and by doing so spare the board from having to make a factual inquiry into the nature of their relationship.

The least rational feature of the board's policy, though not emphasized by the plaintiff, is that although-domestic- partner benefits are confined to persons of the same sex, the partners need not be homosexual. They could be roommates who have lived together for a year and own some property jointly and for want of relatives are each other's "sole domestic partner," and if so they would be entitled to domestic-partner benefits under the board of education's policy. To distinguish between roommates of the same and of different sexes, as the policy implicitly does, cannot be justified on the ground that the latter but not the former could marry each other!

So the policy does not make a very close fit between end and means. But it doesn't have to, provided there is a rational basis for the loose fit. See, e.g., Kimel v. Florida Bd. of Regents, supra, 528 U.S. 62 at 83-84; Vance v. Bradley, 440 U.S. 93, 108-109 (1979); Zehner v. Trigg, 133 F.3d 459, 463 (7th Cir. 1997); Wedderburn v. INS, 215 F.3d 795, 800 (7th Cir. 2000). This follows from our earlier point that cost is a rational basis for treating people differently. Economy is one of the principal reasons for using rules rather than standards to govern conduct. Rules single out one or a few facts from the welter of possibly relevant considerations and make that one or those few facts legally determinative, thus dispensing with inquiry into the other considerations. A standard that takes account of all relevant considerations will produce fewer arbitrary differences in outcome, but at a cost in uncertainty, administrative burden, and sometimes even--as here--in invading people's privacy. It is easy to see why the board of education does not want to put applicants to the proof of their sexual preference. That would be resented. The price of avoiding an inquiry that would be costly because it would be obnoxious is that a few roommates may end up with windfall benefits. We cannot say that the board is being irrational in deciding to pay that price rather than snoop into people's sex lives. Cf. Califano v. Jobst, supra, 434 U.S. at 53-54.

If the result is, as it may be, that none of the nine employees who have opted for domestic-partner benefits is homosexual (or at least that none is willing to acknowledge his homosexuality publicly, for that is not required by the board's policy though it would seem implicit in the board's desire to attract homosexuals who have "a healthy acceptance of their own sexuality"), this would lend a note of irony to the board's policy and would reinforce our earlier conjecture that the purpose is to make a statement rather than to confer actual monetary benefits. But "making a statement" is a common purpose of legislation and does not condemn it as irrational.

The plaintiff has a second ground of appeal. Interpreting the city ordinance to which we referred earlier as forbidding discrimination on the basis of marital status, she argues that by depriving her of the right created by those laws the board has deprived her of "property" without due process of law. It is true that a legal claim can be "property" within the meaning of the due process clause Logan v. Zimmerman Brush Co., 455 U.S. 422, 428-31 (1982); Mullane v. Central Hanover Bank & Trust Co., 339 U.S. 306, 311, 13 (1950); Shvartsman v. Apfel, 138 F.3d 1196, 1199 (7th Cir. 1998), but it is not true that a violation of state or local law (or for that matter of federal law) as such is a deprivation of property. Daniels v. Williams, 474 U.S. 327, 331 (1986); Paul v. Davis, 424 U.S. 693, 700-01 (1976). How could it be? It is the violation that gives rise to the legal claim! Had the board not discriminated against the plaintiff on the basis of her (non)marital status, she would not have a claim that she could press (that she is pressing) before the Chicago Commission on Human Relations. We suspect that she has misinterpreted the ordinance as forbidding any preference for marital status. But that is an issue for the Commission to resolve in the first instance, and it is irrelevant to whether she has alleged a deprivation of constitutional property.

Affirmed.