763 N.W.2d 862 (2009)

Katherine VARNUM, Patricia Hyde, Dawn Barbouroske, Jennifer Barbouroske, Jason Morgan, Charles Swaggerty, David Twombley, Lawrence Hoch, William M. Musser, Otter Dreaming, Ingrid Olson, and Reva Evans, Appellees,
v.
Timothy J. BRIEN, In His Official Capacities as the Polk County Recorder and Polk County Registrar, Appellant.

No. 07-1499.Supreme Court of Iowa.

April 3, 2009.869*869 Roger J. Kuhle and Michael B. O’Meara, Assistant County Attorneys, for appellant.

Dennis W. Johnson of Dorsey & Whitney, LLP, Des Moines, and Camilla B. Taylor and Kenneth D. Upton, Jr. of Lambda Legal Defense & Education Fund, Inc., Chicago, IL, for appellees.

John M. Murray of Murray & Murray, PLC, Storm Lake; Paul Benjamin Linton, Special Counsel, Thomas More Society, Northbrook, IL; Paul R. Devin, Supreme Advocate, Knights of Columbus, New Haven, 870*870 CT; and Thomas Brejcha, President and Chief Counsel, Thomas More Society, Chicago, IL, for amicus curiae Knights of Columbus.

Norman L. Springer, Jr. of McGinn McGinn Jennings & Springer, Council Bluffs, Mathew D. Staver, Stephen M. Crampton, Mary E. McAlister, and David M. Corry, Liberty Counsel, Lynchburg, VA, for amicus curiae Liberty Counsel.

Michael J. Manno, West Des Moines, for amici curiae Jews Offering New Alternatives to Homosexuality, Parents and Friends of Ex-Gays & Gays, and Evergreen International.

Jason M. Steffens of Simmons Perrine, PLC, Cedar Rapids, and Roger T. Severino of The Becket Fund for Religious Liberty, Washington, DC, for amicus curiae The Becket Fund for Religious Liberty.

Andrew J. Boettger of Hastings & Gartin, LLP, Ames, and Steven W. Fitschen, Barry C. Hodge, and Nathan A. Driscoll of The National Legal Foundation, Virginia Beach, VA, for amicus curiae The National Legal Foundation.

David James Hanson of Hofmeyer & Hanson, P.C., Fayette, and Joshua K. Baker of Institute for Marriage and Public Policy, Manassas, VA, for amici curiae James Q. Wilson, et al., Legal and Family Scholars.

Timm W. Reid of Galligan, Reid & Galligan, P.C. and Iowa Liberty and Justice Center, Des Moines, and Benjamin W. Bull, Brian W. Raum, and James A. Campbell of Alliance Defense Fund, Scottsdale, AZ, for amicus curiae Iowa Legislators.

Robert R. Anderson, Vinton, Stuart J. Roth of American Center for Law & Justice, Washington, DC, Laura B. Hernandez of The American Center for Law & Justice, Virginia Beach, VA, and Vincent P. McCarthy of American Center for Law & Justice NE, Litchfield, CT, for amicus curiae The American Center for Law & Justice.

Daniel DenBeste of Elderkin & Pirnie, PLC, Cedar Rapids, and Monte Neil Stewart of Marriage Law Foundation, Orem, UT, for amici curiae United Families International, Family Watch International, and Family Leader Foundation.

Michael A. Giudicessi, Christian S. Walker, Karin A. Johnson, and Nicole Nayima of Faegre & Benson, LLP, Des Moines; Michael A. Ponto of Faegre & Benson, LLP, Minneapolis, MN; and Evan Wolfson, Executive Director of Freedom to Marry, New York City, for amicus curiae Freedom to Marry.

Randall C. Wilson of Iowa Civil Liberties Union, Des Moines, and John A. Knight of American Civil Liberties Union Foundation, Chicago, IL, for amici curiae The Bazelon Center for Mental Health Law, Iowa Protection and Advocacy Services, Inc., National Council on Independent Living, and the National Senior Citizens Law Center.

David H. Goldman, Brent A. Cashatt, and Kodi A. Petersen of Babich, Goldman, Cashatt & Renzo, P.C., Des Moines, and Suzanne B. Goldberg, Clinical Professor of Law and Director of Sexuality and Gender Law Clinic Columbia Law School, New York City, for amici curiae Robert C. Hunter, Jean C. Love, and Maura Strassberg, Iowa Constitutional Law Scholars.

Gordon R. Fischer of Bradshaw, Fowler, Proctor & Fairgrave, P.C., Des Moines, and Hugh S. Balsam and Kara J. Bruce of Locke Lord Bissell & Liddell, LLP, Chicago, IL, for amici curiae Certain Current and Former Iowa Elected Officials.

Ivan T. Webber of Ahlers & Cooney, P.C., Des Moines, and Tobias Barrington Wolff of University of Pennsylvania Law School, Philadelphia, PA, for amici curiae 871*871 OneIowa; The Iowa Chapters of Parents, Families and Friends of Lesbians and Gays; The Equal Justice Society; The National Black Justice Coalition; The Mexican American Legal Defense and Educational Fund; The Zuna Institute; The Asian American Justice Center; The Southern Poverty Law Center; People for the American Way Foundation; The Asian American Legal Defense and Education Fund; and The Immigrants’ Rights Program of the American Friends Service Committee.

Harvey L. Harrison of Harrison & Dietz-Kilen, P.L.C., Des Moines, and Mary Bonauto, Bennett H. Klein, and Janson Wu of Gay & Lesbian Advocates & Defenders, Boston, MA, for amici curiae Massequality, Massachusetts Gay and Lesbian Political Caucus, Gay & Lesbian Advocates & Defenders, various Massachusetts Senators and Representatives, Equality Federation, Garden State Equality, Love Makes a Family, Vermont Freedom to Marry Task Force, Human Rights Campaign, Human Rights Campaign Foundation, and National Gay and Lesbian Task Force.

Catherine C. Dietz-Kilen of Harrison & Dietz-Kilen, P.L.C., Des Moines, for amicus curiae Iowa and National Faith Leaders, Communities and Scholars.

Charles E. Gribble and Matthew T. Oetker of Parrish, Kruidenier, Dunn, Boles, Gribble, Cook, Parrish, Gentry & Fisher, L.L.P., Des Moines, and Aderson B. Francois, Caroline Boucher, Amy E. Hatcher, and Damien G. Scott of Civil Rights Clinic, Howard University School of Law, for amicus curiae Howard University School of Law Civil Rights Clinic.

Mark E. Schantz of College of Law University of Iowa, Iowa City, and Stephen Sanders and Jeffrey W. Sarles of Mayer Brown, LLP, Chicago, IL, for amici curiae Iowa Professors of Law and History.

Jim Quilty of Crawford & Quilty, Des Moines, and Vivian L. Polak, Jonathan A. Damon, Spencer R. Wood, Suman Chakraborty and Emily A. Gianquinto of Dewey & LeBoeuf, LLP, New York City, for amici curiae The National Association of Social Workers, The National Association of Social Workers-Iowa Chapter, Youth and Shelter Services, and Middleton Center for Children’s Rights at Drake University Law School.

Roxanne Barton Conlin of Roxanne Conlin & Associates, P.C., Des Moines, Jennifer K. Brown and Julie F. Kay of Legal Momentum, New York City, and Laura W. Brill, Elizabeth L. Rosenblatt, and Richard M. Simon of Irell & Manella, Los Angeles, CA, for amici curiae Women’s Rights Organizations.

Joseph M. Barron, Des Moines; Richard T. Greenberg of McGuirewoods, LLP, Chicago, IL; and Shannon Price Minter of National Center for Lesbian Rights, Washington, DC, for amici curiae Professors of Family Law and Jurisprudence.

Amy L. Reasner of Lynch Dallas, P.C., Cedar Rapids, and Carmine D. Boccuzzi of Cleary Gottlieb Steen & Hamilton, LLP, New York City, for amicus curiae Social Science Academics and Associations.

Thomas A. Newkirk and Paige E. Fiedler of Fiedler & Newkirk, PLC, Urbandale; Paul M. Smith, William M. Hohengarten, and Anjan Choudhury of Jenner & Block, LLP, Washington, DC, and Nathalie F.P. Gilfoyle of American Psychological Association, Washington, DC, for amicus curiae The American Psychological Association.

Daniel L. Bray of Bray & Klockau, P.L.C., Iowa City, and Diana E. Richmond of Sideman & Bancroft, LLP, San Francisco, CA, for amici curiae The American Academy of Matrimonial Lawyers, Joan 872*872 and Lyle Middleton Center for Children’s Rights, Drake Legal Clinic, and Pediatricians and Family Physicians.

CADY, Justice.

In this case, we must decide if our state statute limiting civil marriage to a union between a man and a woman violates the Iowa Constitution, as the district court ruled. On our review, we hold the Iowa marriage statute violates the equal protection clause of the Iowa Constitution. Therefore, we affirm the decision of the district court.

I. Background Facts and Proceedings.

This lawsuit is a civil rights action by twelve individuals who reside in six communities across Iowa. Like most Iowans, they are responsible, caring, and productive individuals. They maintain important jobs, or are retired, and are contributing, benevolent members of their communities. They include a nurse, business manager, insurance analyst, bank agent, stay-at-home parent, church organist and piano teacher, museum director, federal employee, social worker, teacher, and two retired teachers. Like many Iowans, some have children and others hope to have children. Some are foster parents. Like all Iowans, they prize their liberties and live within the borders of this state with the expectation that their rights will be maintained and protected — a belief embraced by our state motto.[1]

Despite the commonality shared with other Iowans, the twelve plaintiffs are different from most in one way. They are sexually and romantically attracted to members of their own sex. The twelve plaintiffs comprise six same-sex couples who live in committed relationships. Each maintains a hope of getting married one day, an aspiration shared by many throughout Iowa.

Unlike opposite-sex couples in Iowa, same-sex couples are not permitted to marry in Iowa. The Iowa legislature amended the marriage statute in 1998 to define marriage as a union between only a man and a woman.[2] Despite this law, the six same-sex couples in this litigation asked the Polk County Recorder to issue marriage licenses to them. The recorder, following the law, refused to issue the licenses, and the six couples have been unable to be married in this state. Except for the statutory restriction that defines marriage as a union between a man and a woman, the twelve plaintiffs met the legal requirements to marry in Iowa.

As other Iowans have done in the past when faced with the enforcement of a law that prohibits them from engaging in an activity or achieving a status enjoyed by other Iowans, the twelve plaintiffs turned to the courts to challenge the statute. They seek to declare the marriage statute unconstitutional so they can obtain the array of benefits of marriage enjoyed by heterosexual couples, protect themselves and their children, and demonstrate to one another and to society their mutual commitment.

In turning to the courts, the twelve plaintiffs filed this lawsuit in the Polk County District Court. They claimed the 873*873 statutory same-sex marriage ban violates certain liberty and equality rights under the Iowa Constitution. The individual rights claimed by plaintiffs to be adversely affected (by the action of the legislative branch in enacting the same-sex marriage ban and the action of the government officials of the executive branch in enforcing the ban) included the fundamental right to marry, as well as rights to privacy and familial association. Additionally, plaintiffs claimed the legislative and the executive actions unconstitutionally discriminated against them on several bases, including sexual orientation.

The case was presented to the district court by means of a summary judgment motion. The record was developed through witness affidavits and depositions. This record included an explanation by some of the plaintiffs of the disadvantages and fears they face each day due to the inability to obtain a civil marriage in Iowa. These disadvantages and problems include the legal inability to make many life and death decisions affecting their partner, including decisions related to health care, burial arrangements, autopsy, and disposition of remains following death. Various plaintiffs told of the inability to share in their partners’ state-provided health insurance, public-employee pension benefits, and many private-employer-provided benefits and protections. They also explained how several tax benefits are denied. Adoption proceedings are also more cumbersome and expensive for unmarried partners. Other obstacles presented by the inability to enter into a civil marriage include numerous nongovernmental benefits of marriage that are so common in daily life they often go unnoticed, such as something so simple as spousal health club memberships. Yet, perhaps the ultimate disadvantage expressed in the testimony of the plaintiffs is the inability to obtain for themselves and for their children the personal and public affirmation that accompanies marriage.

The parties also explored the reasons for defining marriage in a way that denies these benefits to same-sex couples. The County offered five primary interests of society in support of the legislature’s exclusive definition of marriage. The first three interests are broadly related to the advancement of child rearing. Specifically, the objectives centered on promoting procreation, promoting child rearing by a mother and a father within a marriage, and promoting stability in an opposite-sex relationship to raise and nurture children. The fourth interest raised by the County addressed the conservation of state resources, while the final reason concerned the governmental interest in promoting the concept and integrity of the traditional notion of marriage.

Much of the testimony presented by the County was in the form of opinions by various individuals that same-sex marriage would harm the institution of marriage and also harm children raised in same-sex marriages. Two college professors testified that a heterosexual marriage is, overall, the optimal forum in which to raise children. A retired pediatrician challenged the accuracy of some of the medical research that concludes there is no significant difference between children raised by same-sex couples and opposite-sex couples. A clinical psychologist testified sexual orientation is not as defined and stable as race and gender and can change over time. He acknowledged, however, it is difficult to change a person’s sexual orientation, and efforts to do so can be harmful to the person.

The plaintiffs produced evidence to demonstrate sexual orientation and gender have no effect on children raised by same-sex couples, and same-sex couples 874*874 can raise children as well as opposite-sex couples. They also submitted evidence to show that most scientific research has repudiated the commonly assumed notion that children need opposite-sex parents or biological parents to grow into well-adjusted adults. Many leading organizations, including the American Academy of Pediatrics, the American Psychiatric Association, the American Psychological Association, the National Association of Social Workers, and the Child Welfare League of America, weighed the available research and supported the conclusion that gay and lesbian parents are as effective as heterosexual parents in raising children. For example, the official policy of the American Psychological Association declares, “There is no scientific evidence that parenting effectiveness is related to parental sexual orientation: Lesbian and gay parents are as likely as heterosexual parents to provide supportive and healthy environments for children.”[3] Almost every professional group that has studied the issue indicates children are not harmed when raised by same-sex couples, but to the contrary, benefit from them. In Iowa, agencies that license foster parents have found same-sex couples to be good and acceptable parents. It is estimated that more than 5800 same-sex couples live throughout Iowa, and over one-third of these couples are raising children.

The district court concluded the statute was unconstitutional under the due process and equal protection clauses of the Iowa Constitution and granted summary judgment to the plaintiffs. It initially ordered the county recorder to begin processing marriage licenses for same-sex couples, but stayed the order during the pendency of an appeal.

II. Standard of Review.

Summary judgment is appropriate only when “there is no genuine issue as to any material fact and … the moving party is entitled to a judgment as a matter of law.” Iowa R. Civ. P. 1.981(3). “An issue of fact is `material’ only when the dispute is over facts that might affect the outcome of the suit, given the applicable governing law.” Junkins v. Branstad, 421 N.W.2d 130, 132 (Iowa 1988). The party requesting summary judgment shoulders the burden to demonstrate no genuine issue of material fact exists. Hunter v. City of Des Moines Mun. Hous. Auth., 742 N.W.2d 578, 584 (Iowa 2007). We review the legal issues necessary for resolution of the constitutional claims presented within the context of the summary judgment proceeding de novo. Kistler v. City of Perry, 719 N.W.2d 804, 805 (Iowa 2006).

III. Constitutional Separation of Powers.

We approach the resolution of this case with a keen and respectful understanding of our Iowa Constitution and the vital roles of the three branches of government, as well as the role of the people. It is important for these roles to be identified and expressed from time to time when individuals seek recognition of rights, if only to serve as a reminder of the process of governing that has served us so well as a state for over 150 years.

875*875 The Iowa Constitution is the cornerstone of governing in Iowa. Like the United States Constitution, the Iowa Constitution creates a remarkable blueprint for government. It establishes three separate, but equal, branches of government and delineates the limited roles and powers of each branch. See Iowa Const. art. III, § 1 (“The powers of the government of Iowa shall be divided into three separate departments — the legislative, the executive, and the judicial: and no person charged with the exercise of powers properly belonging to one of these departments shall exercise any function appertaining to either of the others, except in cases hereinafter expressly directed or permitted.”). Among other basic principles essential to our form of government, the constitution defines certain individual rights upon which the government may not infringe. See Iowa Const. art. I (“Bill of Rights”). Equal protection of the law is one of the guaranteed rights. See Iowa Const. art. I, § 6. All these rights and principles are declared and undeniably accepted as the supreme law of this state, against which no contrary law can stand. See Iowa Const. art. XII, § 1 (“This constitution shall be the supreme law of the state, and any law inconsistent therewith, shall be void.”).

This case, as with most other civil rights actions before it, implicates these broad constitutional principles of governing. The legislature, in carrying out its constitutional role to make public policy decisions, enacted a law that effectively excludes gay and lesbian people from the institution of civil marriage. The executive branch of government, in carrying out its role to execute the law, enforced this statute through a county official who refused to issue marriage licenses to six same-sex couples. These Iowans, believing that the law is inconsistent with certain constitutional mandates, exercised their constitutional right to petition the courts for redress of their grievance. This court, consistent with its role to interpret the law and resolve disputes, now has the responsibility to determine if the law enacted by the legislative branch and enforced by the executive branch violates the Iowa Constitution.

A statute inconsistent with the Iowa Constitution must be declared void, even though it may be supported by strong and deep-seated traditional beliefs and popular opinion. Iowa Const. art. XII, § 1 (providing any law inconsistent with the constitution is void). As Chief Justice John Marshall wrote over two centuries ago, “It is a proposition too plain to be contested, that the constitution controls any legislative act repugnant to it….” Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177, 2 L.Ed. 60, 73 (1803).

It is also well established that courts must, under all circumstances, protect the supremacy of the constitution as a means of protecting our republican form of government and our freedoms. As was observed by Justice Robert H. Jackson decades ago in reference to the United States Constitution, the very purpose of limiting the power of the elected branches of government by constitutional provisions like the Equal Protection Clause is “to withdraw certain subjects from the vicissitudes of political controversy, to place them beyond the reach of majorities and officials and to establish them as legal principles to be applied by the courts.” W. Va. State Bd. of Educ. v. Barnette, 319 U.S. 624, 638, 63 S.Ct. 1178, 1185, 87 L.Ed. 1628, 1638 (1943).

The same principle applies to the provisions of the Iowa Constitution that limit government power. The idea that courts, free from the political influences in the other two branches of government, are better suited to protect individual rights 876*876 was recognized at the time our Iowa Constitution was formed. See Koehler v. Hill, 60 Iowa 543, 667, 15 N.W. 609, 640-41 (1883) (Beck, J., dissenting) (“Judges ought not to be partisans, and be influenced by partisan control. Their duty is to interpret and apply the law, to the end that the liberty, and the rights and property, of the people may be secured.”); 1 The Debates of the Constitutional Convention; of the State of Iowa 453 (W. Blair Lord rep.) (Davenport, Luse, Lane & Co. 1857) (containing expression of one delegate’s desire “to have one department of our State government in regard to which we can say, there is no political taint or bias, there is no partisan complexion to it; it is of such a character that when we go before it to have our dearest rights decided, we may rest assured that they will be decided upon principles of law and equity, and not upon political or party principles”).

In fulfilling this mandate under the Iowa Constitution, we look to the past and to precedent. We look backwards, not because citizens’ rights are constrained to those previously recognized, but because historical constitutional principles provide the framework to define our future as we confront the challenges of today.

Our responsibility, however, is to protect constitutional rights of individuals from legislative enactments that have denied those rights, even when the rights have not yet been broadly accepted, were at one time unimagined, or challenge a deeply ingrained practice or law viewed to be impervious to the passage of time. The framers of the Iowa Constitution knew, as did the drafters of the United States Constitution, that “times can blind us to certain truths and later generations can see that laws once thought necessary and proper in fact serve only to oppress,” and as our constitution “endures, persons in every generation can invoke its principles in their own search for greater freedom” and equality. See Lawrence v. Texas, 539 U.S. 558, 578-79, 123 S.Ct. 2472, 2484, 156 L.Ed.2d 508, 526 (2003) (acknowledging intent of framers of Federal Constitution that Constitution endure and be interpreted by future generations); Callender v. Skiles, 591 N.W.2d 182, 190 (Iowa 1999) (“Our constitution is not merely tied to tradition, but recognizes the changing nature of society.”).

When individuals invoke the Iowa Constitution’s guarantees of freedom and equality, courts are bound to interpret those guarantees. In carrying out this fundamental and vital role, “we must never forget that it is a constitution we are expounding.” M’Culloch v. Maryland, 17 U.S. (4 Wheat.) 316, 407, 4 L.Ed. 579, 602 (1819). It speaks with principle, as we, in turn, must also. See State v. Wheeler, 145 Wash.2d 116, 34 P.3d 799, 807 (2001) (Sanders, J., dissenting).

Finally, it should be recognized that the constitution belongs to the people, not the government or even the judicial branch of government. See Iowa Const. art. I, § 2 (“All political power is inherent in the people. Government is instituted for the protection, security, and benefit of the people, and they have the right, at all times, to alter or reform the same, whenever the public good may require it.”). While the constitution is the supreme law and cannot be altered by the enactment of an ordinary statute, the power of the constitution flows from the people, and the people of Iowa retain the ultimate power to shape it over time. See Iowa Const. art. X (“Amendments to the Constitution”).

IV. Equal Protection.

A. Background Principles. The primary constitutional principle at the heart of this case is the doctrine of equal protection. The concept of equal protection is 877*877 deeply rooted in our national and state history, but that history reveals this concept is often expressed far more easily than it is practiced. For sure, our nation has struggled to achieve a broad national consensus on equal protection of the laws when it has been forced to apply that principle to some of the institutions, traditions, and norms woven into the fabric of our society. This observation is important today because it reveals equal protection can only be defined by the standards of each generation. See Cass R. Sunstein, Sexual Orientation and the Constitution: A Note on the Relationship Between Due Process and Equal Protection, 55 U. Chi. L.Rev. 1161, 1163 (1988) (“[T]he Equal Protection Clause looks forward, serving to invalidate practices that were widespread at the time of its ratification and that were expected to endure.”).

The process of defining equal protection, as shown by our history as captured and told in court decisions, begins by classifying people into groups. A classification persists until a new understanding of equal protection is achieved. The point in time when the standard of equal protection finally takes a new form is a product of the conviction of one, or many, individuals that a particular grouping results in inequality and the ability of the judicial system to perform its constitutional role free from the influences that tend to make society’s understanding of equal protection resistant to change. As Justice Oliver Wendell Holmes poignantly said, “It is revolting to have no better reason for a rule of law than that so it was laid down in the time of Henry IV. It is still more revolting if the grounds upon which it was laid down have vanished long since, and the rule simply persists from blind imitation of the past.” Oliver Wendell Holmes, Justice, Supreme Judicial Court of Massachusetts, The Path of the Law, address dedicating new hall at Boston University School of Law (January 8, 1897), in 10 Harv. L.Rev. 457, 469 (1897). This concept is evident in our past cases.

In the first reported case of the Supreme Court of the Territory of Iowa, In re Ralph, 1 Morris 1 (Iowa 1839), we refused to treat a human being as property to enforce a contract for slavery and held our laws must extend equal protection to persons of all races and conditions. 1 Morris at 9. This decision was seventeen years before the United States Supreme Court infamously decided Dred Scott v. Sandford, 60 U.S. (19 How.) 393, 15 L.Ed. 691 (1856), which upheld the rights of a slave owner to treat a person as property. Similarly, in Clark v. Board of Directors, 24 Iowa 266 (1868), and Coger v. North West. Union Packet Co., 37 Iowa 145 (1873), we struck blows to the concept of segregation long before the United States Supreme Court’s decision in Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954). Iowa was also the first state in the nation to admit a woman to the practice of law, doing so in 1869. Admission of Women to the Bar, 1 Chicago Law Times 76, 76 (1887). Her admission occurred three years before the United States Supreme Court affirmed the State of Illinois’ decision to deny women admission to the practice of law, see Bradwell v. Illinois, 83 U.S. (16 Wall.) 130, 139, 21 L.Ed. 442, 445 (1873), and twenty-five years before the United States Supreme Court affirmed the refusal of the Commonwealth of Virginia to admit women into the practice of law, see Ex parte Lockwood, 154 U.S. 116, 118, 14 S.Ct. 1082, 1083, 38 L.Ed. 929, 930 (1894). In each of those instances, our state approached a fork in the road toward fulfillment of our constitution’s ideals and reaffirmed the “absolute equality of all” persons before the law as “the very foundation principle of our government.”[4] See Coger, 37 Iowa at 153.

878*878 So, today, this court again faces an important issue that hinges on our definition of equal protection. This issue comes to us with the same importance as our landmark cases of the past. The same-sex-marriage debate waged in this case is part of a strong national dialogue[5] centered on a fundamental, deep-seated, traditional institution that has excluded, by state action, a particular class of Iowans. This class of people asks a simple and direct question: How can a state premised on the constitutional principle of equal protection justify exclusion of a class of Iowans from civil marriage?

B. Legal Tests to Gauge Equal Protection. The foundational principle of equal protection is expressed in article I, section 6 of the Iowa Constitution, which provides: “All laws of a general nature shall have a uniform operation; the general assembly shall not grant to any citizen or class of citizens, privileges or immunities, which, upon the same terms shall not equally belong to all citizens.” See also Iowa Const. art. I, § 1 (“All men and women are, by nature, free and equal….”); id. art. I, § 2 (recognizing “[a]ll political power is inherent in the people” and “[g]overnment is instituted for the protection, security, and benefit of the people”). Like the Federal Equal Protection Clause found in the Fourteenth Amendment to the United States Constitution, Iowa’s constitutional promise of equal protection “`is essentially a direction that all persons similarly situated should be treated alike.'”[6] Racing Ass’n of Cent. Iowa v. Fitzgerald, 675 N.W.2d 1, 7 (Iowa 2004) 879*879 [hereinafter RACI II] (quoting City of Cleburne v. Cleburne Living Ctr., 473 U.S. 432, 439, 105 S.Ct. 3249, 3254, 87 L.Ed.2d 313, 320 (1985)).

Even in the zealous protection of the constitution’s mandate of equal protection, courts must give respect to the legislative process and presume its enactments are constitutional. We understand that Iowa’s tripartite system of government requires the legislature to make difficult policy choices, including distributing benefits and burdens amongst the citizens of Iowa. In this process, some classifications and barriers are inevitable. As a result, courts pay deference to legislative decisions when called upon to determine whether the Iowa Constitution’s mandate of equality has been violated by legislative action. More specifically, when evaluating challenges based on the equal protection clause, our deference to legislative policy-making is primarily manifested in the level of scrutiny we apply to review legislative action.

In most cases, we apply a very deferential standard known as the “rational basis test.” Id. Under the rational basis test, “[t]he plaintiff has the heavy burden of showing the statute unconstitutional and must negate every reasonable basis upon which the classification may be sustained.” Bierkamp v. Rogers, 293 N.W.2d 577, 579-80 (Iowa 1980). In deference to the legislature, a statute will satisfy the requirements of the equal protection clause

“so long as there is a plausible policy reason for the classification, the legislative facts on which the classification is apparently based rationally may have been considered to be true by the governmental decisionmaker, and the relationship of the classification to its goal is not so attenuated as to render the distinction arbitrary or irrational.”

RACI II, 675 N.W.2d at 7 (quoting Fitzgerald v. Racing Ass’n of Cent. Iowa, 539 U.S. 103, 107, 123 S.Ct. 2156, 2159, 156 L.Ed.2d 97, 103 (2003)). Although the rational basis test is “deferential to legislative judgment, `it is not a toothless one’ in Iowa.” Id. at 9 (quoting Mathews v. de Castro, 429 U.S. 181, 185, 97 S.Ct. 431, 434, 50 L.Ed.2d 389, 394 (1976)). The rational basis test defers to the legislature’s prerogative to make policy decisions by requiring only a plausible policy justification, mere rationality of the facts underlying the decision and, again, a merely rational relationship between the classification and the policy justification. Nonetheless, the deference built into the rational basis test is not dispositive because this court engages in a meaningful review of all legislation challenged on equal protection grounds by applying the rational basis test to the facts of each case. Id. (citing Bierkamp, 293 N.W.2d at 581).[7]

880*880 The constitutional guarantee of equal protection, however, demands certain types of statutory classifications must be subjected to closer scrutiny by courts. See, e.g., Plyler v. Doe, 457 U.S. 202, 216, 102 S.Ct. 2382, 2394, 72 L.Ed.2d 786, 799 (1982) (“[W]e would not be faithful to our obligations under the Fourteenth Amendment if we applied so deferential a standard to every classification.”). Thus, courts apply a heightened level of scrutiny under equal protection analysis when reasons exist to suspect “prejudice against discrete and insular minorities … which tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities.” United States v. Carolene Prods. Co., 304 U.S. 144, 152 n. 4, 58 S.Ct. 778, 783 n. 4, 82 L.Ed. 1234, 1242 n. 4 (1938).

Under this approach, classifications based on race, alienage, or national origin and those affecting fundamental rights are evaluated according to a standard known as “strict scrutiny.” Sherman v. Pella Corp., 576 N.W.2d 312, 317 (Iowa 1998). Classifications subject to strict scrutiny are presumptively invalid and must be narrowly tailored to serve a compelling governmental interest. In re S.A.J.B., 679 N.W.2d 645, 649 (Iowa 2004).

A middle tier of analysis exists between rational basis and strict scrutiny. This intermediate tier has been applied to statutes classifying on the basis of gender or illegitimacy and requires the party seeking to uphold the statute to demonstrate the challenged classification is substantially related to the achievement of an important governmental objective. Sherman, 576 N.W.2d at 317. It is known as “intermediate scrutiny” or “heightened scrutiny,”[8] and groups entitled to this tier of review are often called “quasi-suspect” groups. See Cleburne Living Ctr., 473 U.S. at 445, 105 S.Ct. at 3258, 87 L.Ed.2d at 324. To survive intermediate scrutiny, the law must not only further an important governmental interest and be substantially related to that interest, but the justification for the classification must be genuine and must not depend on broad generalizations. United States v. Virginia, 518 U.S. 515, 533, 116 S.Ct. 2264, 2275, 135 L.Ed.2d 735, 751 (1996).

C. Determination of Constitutional Facts. The parties expended considerable effort developing a summary judgment record to assist the district court in deciding the legal issues presented in this case, including which level of scrutiny to apply. Before proceeding to determine these legal issues, we consider the role of the evidence offered by the parties to support their legal arguments. The district court excluded some of the offered testimony, 881*881 which the County has raised as an issue on appeal.

Our law recognizes a distinction between “adjudicative” and “legislative” facts. Greenwood Manor v. Iowa Dep’t of Pub. Health, 641 N.W.2d 823, 836 (Iowa 2002). Most often, judicial decision-making is predicated solely on a finding of facts relating to the parties and their particular circumstances. Id. These facts are referred to as “adjudicative” facts, see id., and the resolution of a dispute over these facts is done within the framework of a set of rules to determine the admissibility of evidence tending to prove such facts. See generally Iowa Rs. Evid. At times, however, judicial decision-making involves crafting rules of law based on social, economic, political, or scientific facts. See 2 John W. Strong, McCormick on Evidence § 328, at 369 (5th ed. 1999) [hereinafter McCormick on Evidence]. These facts have been denominated as “legislative” facts and become relevant to judicial decision-making when courts are required to decide the constitutionality of a statute, among other occasions. Id. As a result, judicial decision-making in the context of constitutional issues can involve the “process of adapting law to a volatile social-political environment.” Id. at 370. Legislative facts are relevant in deciding these constitutional issues because courts must normally analyze “whether there exist circumstances which constitutionally either legitimate the exercise of legislative power or substantiate the rationality of the legislative product.” Id. In fact, the common role of legislative facts in constitutional cases has led to an alternative designation of legislative facts called “constitutional facts” to better describe those facts “which assist a court in forming a judgment on a question of constitutional law.” Kenneth C. Davis, An Approach to Problems of Evidence in the Administrative Process, 55 Harv. L.Rev. 364, 403 (1942).

Unlike adjudicative facts, legislative or constitutional facts “may be presented either formally or informally.” Welsh v. Branstad, 470 N.W.2d 644, 648 (Iowa 1991). There is no formalized set of rules governing a court’s ability to consider legislative or constitutional facts. See Iowa R. Evid. 5.201 (applying rule governing judicial notice only to adjudicative facts); Fed.R.Evid. 201 advisory committee’s note (“No rule deals with judicial notice of `legislative facts.'”). See generally City of Council Bluffs v. Cain, 342 N.W.2d 810, 816-17 (Iowa 1983) (McCormick, J., dissenting). Thus, constitutional facts are introduced into judicial decisions through independent research by judges and written briefs of the parties, as well as testimony of witnesses. See McCormick on Evidence at 381-84. Importantly, constitutional facts are not subject to the rules of evidence when presented by a party in the form of witness testimony. Conceptually, testimony relating to constitutional facts is only presented as authority for the legal decision the court is required to make, and it would be inconsistent to apply formal rules of evidence to facts in the form of testimony that a court can independently obtain and consider in deciding the case.

Nonetheless, courts consider the “actual truth-content” of constitutional facts. See id. at 382-83. Such facts are generally disputable, and courts must rely on the most compelling data in order to give needed intellectual legitimacy to the law or rule crafted by the court. Id. at 383.

Consequently, we review all of the material tendered by the parties in this case to assist us in our review of the constitutionality of the civil marriage statute. The error committed by the trial court in failing to do so is of no consequence under our de novo reviewing standard.

882*882 D. Similarly Situated People. The County seeks to undercut the plaintiffs’ equal protection claim by asserting the plaintiffs are not similarly situated to heterosexuals. We consider this threshold argument before proceeding to the application of our equal protection test.

We begin by recognizing the constitutional pledge of equal protection does not prohibit laws that impose classifications. Chicago & Nw. Ry. v. Fachman, 255 Iowa 989, 996, 125 N.W.2d 210, 214 (1963) (recognizing “it is often necessary in accomplishing efficient and beneficial legislation to divide the subjects upon which it operates into classes”). Many statutes impose classifications by granting special benefits or declaring special burdens, and the equal protection clause does not require all laws to apply uniformly to all people. Nordlinger v. Hahn, 505 U.S. 1, 10, 112 S.Ct. 2326, 2331, 120 L.Ed.2d 1, 12 (1992). Instead, equal protection demands that laws treat alike all people who are “`similarly situated with respect to the legitimate purposes of the law.'” RACI II, 675 N.W.2d at 7 (quoting Coll. Area Renters & Landlord Ass’n v. City of San Diego, 43 Cal.App.4th 677, 50 Cal.Rptr.2d 515, 520 (1996)) (emphasis omitted).

This requirement of equal protection — that the law must treat all similarly situated people the same — has generated a narrow threshold test. Under this threshold test, if plaintiffs cannot show as a preliminary matter that they are similarly situated, courts do not further consider whether their different treatment under a statute is permitted under the equal protection clause. See, e.g., Timberland Partners XXI, LLP v. Iowa Dep’t of Revenue, 757 N.W.2d 172, 176-77 (Iowa 2008) (applying threshold analysis); In re Det. of Hennings, 744 N.W.2d 333, 338-40 (Iowa 2008) (same); Grovijohn v. Virjon, Inc., 643 N.W.2d 200, 204 (Iowa 2002) (same). Not only have we utilized this test in the past, but courts from other jurisdictions have confronted it in cases involving equal protection challenges to statutes that restrict marriage to opposite-sex couples. See In re Marriage Cases, 43 Cal.4th 757, 76 Cal.Rptr.3d 683, 183 P.3d 384, 435 n. 54 (2008) (analyzing and rejecting the government’s threshold argument that same-sex couples are not similarly situated to opposite-sex couples); Kerrigan v. Comm’r of Pub. Health, 289 Conn. 135, 957 A.2d 407, 423-24 & n. 19 (2008) (same).

The County references this threshold test in this case and asserts the plaintiffs are not similarly situated to opposite-sex couples so as to necessitate further equal protection analysis because the plaintiffs cannot “procreate naturally.” In other words, the County argues the statute does not treat similarly situated persons differently, but merely treats dissimilar persons differently.

In considering whether two classes are similarly situated, a court cannot simply look at the trait used by the legislature to define a classification under a statute and conclude a person without that trait is not similarly situated to persons with the trait. See Racing Ass’n of Cent. Iowa v. Fitzgerald, 648 N.W.2d 555, 559 (Iowa 2002) (RACI I); Joseph Tussman & Jacobus ten-Broek, The Equal Protection of the Laws, 37 Cal. L.Rev. 341, 344-47 (1949) [hereinafter Tussman & tenBroek]. The equal protection clause does not merely ensure the challenged statute applies equally to all people in the legislative classification. “`[S]imilarly situated’ cannot mean simply `similar in the possession of the classifying trait.’ All members of any class are similarly situated in this respect and consequently, any classification whatsoever would be reasonable by this test.” Tussman & tenBroek, 37 Cal. L.Rev. at 345. In the same way, the similarly situated 883*883 requirement cannot possibly be interpreted to require plaintiffs to be identical in every way to people treated more favorably by the law. No two people or groups of people are the same in every way, and nearly every equal protection claim could be run aground onto the shoals of a threshold analysis if the two groups needed to be a mirror image of one another. Such a threshold analysis would hollow out the constitution’s promise of equal protection.

Thus, equal protection before the law demands more than the equal application of the classifications made by the law. The law itself must be equal. See Fachman, 255 Iowa at 998, 125 N.W.2d at 215 (“`The equal protection of the laws is a pledge of the protection of equal laws….'” (quoting Yick Wo v. Hopkins, 118 U.S. 356, 369, 6 S.Ct. 1064, 1070, 30 L.Ed. 220, 226 (1886))). In other words, to truly ensure equality before the law, the equal protection guarantee requires that laws treat all those who are similarly situated with respect to the purposes of the law alike. RACI II, 675 N.W.2d at 7. This requirement makes it “impossible to pass judgment on the reasonableness of a [legislative] classification without taking into consideration, or identifying, the purpose of the law.” Tussman & tenBroek, 37 Cal. L.Rev. at 347. The purposes of the law must be referenced in order to meaningfully evaluate whether the law equally protects all people similarly situated with respect to those purposes. For these reasons, the trait asserted by the County is insufficient to support its threshold argument.

Nevertheless, we have said our marriage laws “are rooted in the necessity of providing an institutional basis for defining the fundamental relational rights and responsibilities of persons in organized society.” Laws v. Griep, 332 N.W.2d 339, 341 (Iowa 1983); see also Baehr v. Lewin, 74 Haw. 530, 852 P.2d 44, 58 (1993) (stating civil marriage is “`a partnership to which both partners bring their financial resources as well as their individual energies and efforts'” (quoting Gussin v. Gussin, 73 Haw. 470, 836 P.2d 484, 491 (1992))). These laws also serve to recognize the status of the parties’ committed relationship. See Madison v. Colby, 348 N.W.2d 202, 206 (Iowa 1984) (stating “`the marriage state is not one entered into for the purpose of labor and support alone,'” but also includes “`the comfort and happiness of the parties to the marriage contract'” (quoting Price v. Price, 91 Iowa 693, 697-98, 60 N.W. 202, 203 (Iowa 1894)) (emphasis added)); Hamilton v. McNeill, 150 Iowa 470, 478, 129 N.W. 480, 482 (1911) (“The marriage to be dissolved is not a mere contract, but is a status.”); Turner v. Hitchcock, 20 Iowa 310, 325 (1866) (Lowe, C.J., concurring) (observing that marriage changes the parties’ “legal and social status”).

Therefore, with respect to the subject and purposes of Iowa’s marriage laws, we find that the plaintiffs are similarly situated compared to heterosexual persons. Plaintiffs are in committed and loving relationships, many raising families, just like heterosexual couples. Moreover, official recognition of their status provides an institutional basis for defining their fundamental relational rights and responsibilities, just as it does for heterosexual couples. Society benefits, for example, from providing same-sex couples a stable framework within which to raise their children and the power to make health care and end-of-life decisions for loved ones, just as it does when that framework is provided for opposite-sex couples.

In short, for purposes of Iowa’s marriage laws, which are designed to bring a sense of order to the legal relationships of 884*884 committed couples and their families in myriad ways, plaintiffs are similarly situated in every important respect, but for their sexual orientation. As indicated above, this distinction cannot defeat the application of equal protection analysis through the application of the similarly situated concept because, under this circular approach, all distinctions would evade equal protection review. Therefore, with respect to the government’s purpose of “providing an institutional basis for defining the fundamental relational rights and responsibilities of persons,” same-sex couples are similarly situated to opposite-sex couples.[9]

E. Classification Undertaken in Iowa Code Section 595.2. Plaintiffs believe Iowa Code section 595.2 classifies on the bases of gender and sexual orientation. The County argues the same-sex marriage ban does not discriminate on either basis. The district court held section 595.2 classifies according to gender. As we will explain, we believe the ban on civil marriages between two people of the same sex classifies on the basis of sexual orientation.

The County initially points out that section 595.2 does not explicitly refer to “sexual orientation” and does not inquire into whether either member of a proposed civil marriage is sexually attracted to the other. Consequently, it seizes on these observations to support its claim that the statute does not establish a classification on the basis of sexual orientation because the same-sex civil marriage ban does not grant or withhold the benefits flowing from the statute based on sexual preference. Instead, the County argues, section 595.2 only incidentally impacts disparately upon gay and lesbian people.

The County’s position reveals the importance of accurately and precisely defining the classification in analyzing all equal protection challenges. The manner in which a classification is defined impacts the utility of an equal protection analysis as a means of revealing discrimination. Therefore, it is critical that a court reviewing the statute 885*885 identify the true nature of the classification.

It is true the marriage statute does not expressly prohibit gay and lesbian persons from marrying; it does, however, require that if they marry, it must be to someone of the opposite sex. Viewed in the complete context of marriage, including intimacy, civil marriage with a person of the opposite sex is as unappealing to a gay or lesbian person as civil marriage with a person of the same sex is to a heterosexual. Thus, the right of a gay or lesbian person under the marriage statute to enter into a civil marriage only with a person of the opposite sex is no right at all. Under such a law, gay or lesbian individuals cannot simultaneously fulfill their deeply felt need for a committed personal relationship, as influenced by their sexual orientation, and gain the civil status and attendant benefits granted by the statute. Instead, a gay or lesbian person can only gain the same rights under the statute as a heterosexual person by negating the very trait that defines gay and lesbian people as a class — their sexual orientation. In re Marriage Cases, 183 P.3d at 441. The benefit denied by the marriage statute — the status of civil marriage for same-sex couples — is so “closely correlated with being homosexual” as to make it apparent the law is targeted at gay and lesbian people as a class. See Lawrence, 539 U.S. at 583, 123 S.Ct. at 2486, 156 L.Ed.2d at 529 (O’Connor, J., concurring) (reviewing criminalization of homosexual sodomy and concluding that “[w]hile it is true that the law applies only to conduct, the conduct targeted by this law is conduct that is closely correlated with being homosexual. Under such circumstances, [the] sodomy law is targeted at more than conduct. It is instead directed toward gay persons as a class.”). The Court’s decision in Romer v. Evans, 517 U.S. 620, 116 S.Ct. 1620, 134 L.Ed.2d 855 (1996), supports this conclusion. Romer can be read to imply that sexual orientation is a trait that defines an individual and is not merely a means to associate a group with a type of behavior. See Romer, 517 U.S. at 632, 116 S.Ct. at 1627, 134 L.Ed.2d at 865-66 (holding an amendment to a state constitution pertaining to “homosexual … orientation” expresses “animus toward the class that it affects”).

By purposefully placing civil marriage outside the realistic reach of gay and lesbian individuals, the ban on same-sex civil marriages differentiates implicitly on the basis of sexual orientation. See Kerrigan, 957 A.2d at 431 n. 24; Conaway v. Deane, 401 Md. 219, 932 A.2d 571, 605 (2007). Thus, we proceed to analyze the constitutionality of the statute based on sexual orientation discrimination.

F. Framework for Determining Appropriate Level of Judicial Scrutiny. Our determination that the marriage statute employs a sexual-orientation-based classification does not, of course, control the outcome of our equal protection inquiry. Most statutes, one way or the other, create classifications. Clements v. Fashing, 457 U.S. 957, 967, 102 S.Ct. 2836, 2845, 73 L.Ed.2d 508, 518 (1982) (“Classification is the essence of all legislation, and only those classifications which are invidious, arbitrary, or irrational offend the Equal Protection Clause of the Constitution.”). To determine if this particular classification violates constitutional principles of equal protection, we must next ask what level of scrutiny applies to classifications of this type. The County argues the more deferential rational basis test should apply, while plaintiffs argue closer scrutiny is appropriate.

Although neither we nor the United States Supreme Court has decided which level of scrutiny applies to legislative classifications 886*886 based on sexual orientation, numerous Supreme Court equal protection cases provide a general framework to guide our analysis under the Iowa Constitution.[10] To say a general framework exists is not to say the Supreme Court has provided a precise formula for determining when legislative action is subject to a heightened form of scrutiny. See Cleburne Living Ctr., 473 U.S. at 472 n. 24, 105 S.Ct. at 3272 n. 24, 87 L.Ed.2d at 341 n. 24 (Marshall, J., concurring in part and dissenting in part) (“No single talisman can define those groups likely to be the target of classifications offensive to the Fourteenth Amendment and therefore warranting heightened or strict scrutiny; experience, not abstract logic, must be the primary guide.”); Conaway, 932 A.2d at 606 (“There is no brightline diagnostic, annunciated by … the U.S. Supreme Court, by which a suspect or quasi-suspect class may be readily recognized.”). Instead, the Supreme Court has expressed a number of general principles to assist in identifying the appropriate level of scrutiny.

Classifications based on factors like race, alienage, national origin, sex, or illegitimacy are “so seldom relevant to achievement of any legitimate state interest that laws grounded in such considerations are deemed to reflect prejudice and antipathy.” Cleburne Living Ctr., 473 U.S. at 440, 105 S.Ct. at 3254, 87 L.Ed.2d at 320. Rather than bearing some relationship to the burdened class’s ability to contribute to society, such classifications often reflect irrelevant stereotypes. Id. at 440-41, 105 S.Ct. at 3254-55, 87 L.Ed.2d at 320-21. “For these reasons and because such discrimination is unlikely to be soon rectified by legislative means,” laws based on these types of classifications must withstand more intense judicial scrutiny than other types of classifications. Id.

Instead of adopting a rigid formula to determine whether certain legislative classifications warrant more demanding constitutional analysis, the Supreme Court has looked to four factors.[11] Conaway, 932 887*887 A.2d at 606 (discussing factors examined by Supreme Court in considering use of heightened scrutiny). The Supreme Court has considered: (1) the history of invidious discrimination against the class burdened by the legislation;[12] (2) whether the characteristics that distinguish the class indicate a typical class member’s ability to contribute to society;[13] (3) whether the distinguishing characteristic is “immutable” or beyond the class members’ control;[14] and (4) the political power of the 888*888 subject class.[15] In considering whether sexual orientation is a suspect class, a number of our sister jurisdictions have referenced similar factors. See In re Marriage Cases, 183 P.3d at 442-43; Kerrigan, 957 A.2d at 426; Conaway, 932 A.2d at 606-07; Andersen v. King County, 158 Wash.2d 1, 138 P.3d 963, 974 (2006).

Both parties recognize the relevance of these factors. They disagree, however, over how the factors should be applied to decide whether sexual orientation is a suspect or quasi-suspect class. The County essentially views the factors as elements, asserting each must be fulfilled before we may abandon our deferential level of scrutiny. To this end, the County argues the immutability and political powerlessness “elements” are not satisfied in this case.

In its effort to treat the factors as essential elements, the County overlooks the flexible manner in which the Supreme Court has applied the four factors in the past.[16] For purposes of state constitutional 889*889 analysis, we likewise refuse to view all the factors as elements or as individually demanding a certain weight in every case. Instead, we analyze each of the four factors and assess how each bears on the question of whether the Iowa Constitution requires a more searching scrutiny be applied to the specific classification at issue. We note the first two factors — history of intentional discrimination and relationship of classifying characteristic to a person’s ability to contribute — have always been present when heightened scrutiny has been applied. They have been critical to the analysis and could be considered as prerequisites to concluding a group is a suspect or quasi-suspect class. However, we consider the last two factors — immutability of the characteristic and political powerlessness of the group — to supplement the analysis as a means to discern whether a need for heightened scrutiny exists.

G. Determination of Appropriate Level of Scrutiny. Guided by the established framework, we next consider each of the four traditional factors and assess how each bears on the question of whether the constitution demands a more searching scrutiny be applied to the sexual-orientation-based classification in Iowa’s marriage statute.

1. History of discrimination against gay and lesbian people. The first consideration is whether gay and lesbian people have suffered a history of purposeful unequal treatment because of their sexual orientation. The County does not, and could not in good faith, dispute the historical reality that gay and lesbian people as a group have long been the victim of purposeful and invidious discrimination because of their sexual orientation. The long and painful history of discrimination against gay and lesbian persons is epitomized by the criminalization of homosexual conduct in many parts of this country until very recently. See Lawrence, 539 U.S. at 578-79, 123 S.Ct. at 2483-84, 156 L.Ed.2d at 520 (invalidating criminalization of homosexual sodomy in 2003). Additionally, only a few years ago persons identified as homosexual were dismissed from military service regardless of past dedication and demonstrated valor. Public employees identified as gay or lesbian have been thought to pose security risks due to a perceived risk of extortion resulting from a threat of public exposure. School-yard bullies have psychologically ground children with apparently gay or lesbian sexual orientation in the cruel mortar and pestle of school-yard prejudice. At the same time, lesbian and gay people continue to be frequent victims of hate crimes. See Criminal Justice Information Servs. Div., FBI, Hate Crime Statistics 2007, http:// www.fbi.gov/ucr/hc2007/victims.htm (according to FBI-collected data, the only hate crimes occurring more frequently than sexual-orientation-motivated hate crimes are crimes based on race or religious bias).

The Iowa General Assembly has recognized the need to address sexual-orientation-based discrimination by including sexual orientation as a characteristic protected in the Iowa Civil Rights Act, by defining hate crimes to include certain offenses committed because of the victim’s sexual orientation, and by prohibiting “harassing or bullying” behavior in schools based on sexual orientation. See Iowa Code §§ 216.2-.18A (Iowa Civil Rights Act) (sexual-orientation-based discrimination); id. § 280.28 (school harassment and 890*890 bullying); id. § 729A.2 (hate crimes committed because of the victim’s sexual orientation). These statutory enactments demonstrate a legislative recognition of the need to remedy historical sexual-orientation-based discrimination.[17]

In sum, this history of discrimination suggests any legislative burdens placed on lesbian and gay people as a class “are more likely than others to reflect deep-seated prejudice rather than legislative rationality in pursuit of some legitimate objective.” Plyler, 457 U.S. at 216 n. 14, 102 S.Ct. at 2394 n. 14, 72 L.Ed.2d at 799 n. 14. This observation favors an elevated scrutiny to uncover any such prejudice.

2. Sexual orientation and the ability to contribute to society. A second relevant consideration is whether the characteristic at issue — sexual orientation — is related to the person’s ability to contribute to society. Heightened scrutiny is applied when the classification bears no relationship to a person’s ability to contribute to society. The existence of this factor indicates the classification is likely based on irrelevant stereotypes and prejudice. Kerrigan, 957 A.2d at 453. A classification unrelated to a person’s ability to perform or contribute to society typically reflects “prejudice and antipathy — a view that those in the burdened class are not as worthy or deserving as others” or “reflect[s] outmoded notions of the relative capabilities of persons with the characteristic.” Cleburne Living Ctr., 473 U.S. at 440-41, 105 S.Ct. at 3254-56, 87 L.Ed.2d at 320-21.

Not surprisingly, none of the same-sex marriage decisions from other state courts around the nation have found a person’s sexual orientation to be indicative of the person’s general ability to contribute to society.[18] See, e.g., In re Marriage Cases, 183 P.3d at 442; Kerrigan, 957 A.2d at 434-36; Dean v. District of Columbia, 653 A.2d 307, 345 (D.C.1995); Conaway, 932 A.2d at 609. More importantly, the Iowa legislature has recently declared as the public policy of this state that sexual orientation 891*891 is not relevant to a person’s ability to contribute to a number of societal institutions other than civil marriage. See Iowa Code § 216.6 (employment); id. § 216.7 (public accommodations); id. § 216.8 (housing); id. § 216.9 (education); id. § 216.10 (credit practices).[19] Significantly, we do not construe Iowa Code chapter 216 to allow marriage between persons of the same sex, a construction expressly forbidden in the Iowa Code. See id. § 216.18A (“[Chapter 216] shall not be construed to allow marriage between persons of the same sex, in accordance with chapter 595.”). Rather, we merely highlight the reality that chapter 216 and numerous other statutes and regulations demonstrate sexual orientation is broadly recognized in Iowa to be irrelevant to a person’s ability to contribute to society.[20] Those statutes and regulations reflect at least some measure of legislative and executive awareness that discrimination based on sexual orientation is often predicated on prejudice and stereotype and further express a desire to remove sexual orientation as an obstacle to the ability of gay and lesbian people to achieve their full potential. Therefore, we must scrutinize more 892*892 closely those classifications that suggest a law may be based on prejudice and stereotype because laws of that nature are “incompatible with the constitutional understanding that each person is to be judged individually and is entitled to equal justice under the law.” Plyler, 457 U.S. at 217 n. 14, 102 S.Ct. at 2394 n. 14, 72 L.Ed.2d at 799 n. 14. Thus, although we do not interpret chapter 216 to allow same-sex marriage, we rely on the legislative judgment underlying chapter 216 to determine the appropriate level of scrutiny when sexual orientation is the basis for a statutory classification. Based on Iowa statutes and regulations, it is clear sexual orientation is no longer viewed in Iowa as an impediment to the ability of a person to contribute to society.

3. Immutability of sexual orientation. The parties, consistent with the same-sex-marriage scholarship, opinions, and jurisprudence, contest whether sexual orientation is immutable or unresponsive to attempted change. The County seizes on this debate to argue the summary judgment granted by the district court in this case was improper because plaintiffs could not prove, as a matter of fact, that sexuality is immutable. This argument, however, essentially limits the constitutional relevance of mutability to those instances in which the trait defining the burdened class is absolutely impervious to change. To evaluate this argument, we must first consider the rationale for using immutability as a factor.

A human trait that defines a group is “immutable” when the trait exists “solely by the accident of birth,” Frontiero v. Richardson, 411 U.S. 677, 686, 93 S.Ct. 1764, 1770, 36 L.Ed.2d 583, 591 (1973) (Brennan, J., plurality opinion), or when the person with the trait has no ability to change it, Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 360, 98 S.Ct. 2733, 2784, 57 L.Ed.2d 750, 815 (1978). Immutability is a factor in determining the appropriate level of scrutiny because the inability of a person to change a characteristic that is used to justify different treatment makes the discrimination violative of the rather “`basic concept of our system that legal burdens should bear some relationship to individual responsibility.'” Frontiero, 411 U.S. at 686, 93 S.Ct. at 1770, 36 L.Ed.2d at 591 (Brennan, J., plurality opinion) (quoting Weber v. Aetna Cas. & Sur. Co., 406 U.S. 164, 175, 92 S.Ct. 1400, 1407, 31 L.Ed.2d 768, 779 (1972)); accord Plyler, 457 U.S. at 217 n. 14, 102 S.Ct. at 2394 n. 14, 72 L.Ed.2d at 799 n. 14 (“Legislation imposing special disabilities upon groups disfavored by virtue of circumstances beyond their control suggests the kind of `class or caste’ treatment that the Fourteenth Amendment was designed to abolish.”). Put another way, when a characteristic is immutable, different treatment based on this characteristic seems “all the more invidious and unfair.” Nan D. Hunter, The Sex Discrimination Argument in Gay Rights Cases, 9 J.L. & Pol’y 397, 403 (2001). Additionally, immutability can relate to the scope and permanency of the barrier imposed on the group. Temporary barriers tend to be less burdensome on a group and more likely to actually advance a legitimate governmental interest. Consequently, such barriers normally do not warrant heightened scrutiny. See Sosna v. Iowa, 419 U.S. 393, 406, 95 S.Ct. 553, 561, 42 L.Ed.2d 532, 544-45 (1975) (one-year residency requirement for divorce permitted in part because the constraint was only temporary); Vlandis v. Kline, 412 U.S. 441, 453, 93 S.Ct. 2230, 2236, 37 L.Ed.2d 63, 72 (1973) (bona fide state resident requirement for college tuition permissible when students are provided an opportunity to prove they have become residents). The permanency of the barrier also depends on the ability of the individual 893*893 to change the characteristic responsible for the discrimination. This aspect of immutability may separate truly victimized individuals from those who have invited discrimination by changing themselves so as to be identified with the group. As implied by Justice Ferren, in dissent, in Dean:

The degree to which an individual controls, or cannot avoid, the acquisition of the defining trait, and the relative ease or difficulty with which a trait can be changed, are relevant to whether a classification is “suspect” or “quasi-suspect” because this inquiry is one way of asking whether someone, rather than being victimized, has voluntarily joined a persecuted group and thereby invited the discrimination.

653 A.2d at 346 (Ferren, J., dissenting).

Importantly, this background reveals courts need not definitively resolve the nature-versus-nurture debate currently raging over the origin of sexual orientation in order to decide plaintiffs’ equal protection claims. The constitutional relevance of the immutability factor is not reserved to those instances in which the trait defining the burdened class is absolutely impossible to change. Compare Sherman, 576 N.W.2d at 317 (suggesting heightened scrutiny is applicable to gender classifications), with Iowa Code § 144.23 (providing legal procedure to obtain new birth certificate indicating change in gender). That is, we agree with those courts that have held the immutability “prong of the suspectness inquiry surely is satisfied when … the identifying trait is `so central to a person’s identity that it would be abhorrent for government to penalize a person for refusing to change [it].'” Kerrigan, 957 A.2d at 438 (quoting Watkins v. U.S. Army, 875 F.2d 699, 726 (9th Cir. 1989) (Norris, J., concurring in the judgment)); see also In re Marriage Cases, 183 P.3d at 442 (“Because a person’s sexual orientation is so integral an aspect of one’s identity, it is not appropriate to require a person to repudiate or change his or her sexual orientation in order to avoid discriminatory treatment.”).

In this case, the County acknowledges sexual orientation is highly resistant to change. Additionally, “sexual orientation `forms a significant part of a person’s identity.'” Kerrigan, 957 A.2d at 438 (quoting Able v. United States, 968 F.Supp. 850, 863 (E.D.N.Y.1997), rev’d on other grounds, 155 F.3d 628 (2d Cir.1998)). Sexual orientation influences the formation of personal relationships between all people — heterosexual, gay, or lesbian — to fulfill each person’s fundamental needs for love and attachment. Accordingly, because sexual orientation is central to personal identity and “`may be altered [if at all] only at the expense of significant damage to the individual’s sense of self,'” classifications based on sexual orientation “are no less entitled to consideration as a suspect or quasi-suspect class than any other group that has been deemed to exhibit an immutable characteristic.” Id. at 438-39 (quoting Jantz v. Muci, 759 F.Supp. 1543, 1548 (D.Kan.1991), rev’d on other grounds, 976 F.2d 623 (10th Cir.1992)). Sexual orientation is not the type of human trait that allows courts to relax their standard of review because the barrier is temporary or susceptible to self-help.

4. Political powerlessness of lesbian and gay people. As observed, the political power of the burdened class has been referenced repeatedly in Supreme Court cases determining the level of scrutiny to be applied to a given piece of legislation. Unfortunately, the Court has never defined what it means to be politically powerless for purposes of this analysis, nor has it quantified a maximum amount of political power a group may enjoy while still 894*894 receiving the protection from unfair discrimination accompanying heightened scrutiny. The County points to the numerous legal protections gay and lesbian people have secured against discrimination, and the County argues those protections demonstrate gay and lesbian people are not a politically powerless class. The County’s argument implies gay and lesbian people must be characterized by a complete, or nearly complete, lack of political power before courts should subject sexual-orientation-based legislative burdens to a heightened scrutiny.

Notwithstanding the lack of a mathematical equation to guide the analysis of this factor, a number of helpful general principles related to the political power of suspect classes can be culled from the Supreme Court’s cases. First, these cases show absolute political powerlessness is not necessary to subject legislative burdens on a certain class to heightened scrutiny. For example, females enjoyed at least some measure of political power when the Supreme Court first heightened its scrutiny of gender classifications. See Frontiero, 411 U.S. at 685-88 & n. 17, 93 S.Ct. at 1769-72 & n. 17, 36 L.Ed.2d at 591-92 & n. 17 (Brennan, J., plurality opinion) (subjecting gender classifications to heightened scrutiny after observing Civil Rights Act of 1964 and Equal Pay Act of 1963 prohibited sex discrimination and noting “the position of women in America [had] improved markedly” by 1973 such that “women [did] not constitute a small and powerless minority”).

Second, Supreme Court jurisprudence establishes that a group’s current political powerlessness is not a prerequisite to enhanced judicial protection. “[I]f a group’s current political powerlessness [was] a prerequisite to a characteristic’s being considered a constitutionally suspect basis for differential treatment, it would be impossible to justify the numerous decisions that continue to treat sex, race, and religion as suspect classifications.” In re Marriage Cases, 183 P.3d at 443. Race continues to be a suspect classification, Grutter v. Bollinger, 539 U.S. 306, 326, 123 S.Ct. 2325, 2337, 156 L.Ed.2d 304, 331 (2003), even though racial minorities enjoy growing political power.[21] Likewise, gender classifications receive various forms of heightened scrutiny, even though women continue to gain political power. See, e.g., Virginia, 518 U.S. at 532-33, 116 S.Ct. at 2275, 135 L.Ed.2d at 751-52 (applying intermediate scrutiny).

While a more in-depth discussion of the history of the political-power factor is possible, see Kerrigan, 957 A.2d at 439-44, we are satisfied, for the purpose of analyzing the Iowa Constitution, the political powerlessness factor of the level-of-scrutiny inquiry does not require a showing of absolute political powerlessness. Rather, the touchstone of the analysis should be “whether the group lacks sufficient political strength to bring a prompt end to the prejudice and discrimination through traditional political means.” Id. at 444.

It is also important to observe that the political power of gays and lesbians, while responsible for greater acceptance and decreased discrimination, has done little to remove barriers to civil marriage. Although a small number of state legislatures have approved civil unions for gay and lesbian people without judicial intervention, no legislature has secured the right to civil marriage for gay and lesbian 895*895 people without court order.[22] The myriad statutes and regulatory protections against discrimination based on sexual orientation in such areas as employment, housing, public accommodations, and education have not only been absent in the area of marriage, but legislative bodies have taken affirmative steps to shore up the concept of traditional marriage by specifically excluding gays and lesbians. Like Iowa, over forty other states have passed statutes or constitutional amendments to ban same-sex marriages. See Ben Schuman, Note, Gods & Gays: Analyzing the Same-Sex Marriage Debate From a Religious Perspective, 96 Geo. L.J. 2103, 2107-08 (2008) (recognizing forty-two states with laws prohibiting same-sex marriage as of November 8, 2007) [hereinafter Schuman]; Human Rights Campaign, Statewide Marriage Prohibitions (2008), available at htt p://www.hrc.org/documents/marriage prohibitions.pdf (mapping states’ enactment of marriage prohibitions). Thus, although equal rights for gays and lesbians have been increasingly recognized in the political arena, the right to civil marriage is a notable exception to this trend. Consequently, the specific right sought in this case has largely lacked any extensive political support and has actually experienced an affirmative backlash.

We are convinced gay and lesbian people are not so politically powerful as to overcome the unfair and severe prejudice that history suggests produces discrimination based on sexual orientation. Gays and lesbians certainly possess no more political power than women enjoyed four decades ago when the Supreme Court began subjecting gender-based legislation to closer scrutiny. Additionally, gay and lesbian people are, as a class, currently no more powerful than women or members of some racial minorities. These facts demonstrate, at the least, the political-power factor does not weigh against heightened judicial scrutiny of sexual-orientation-based legislation.

5. Classifications based on sexual orientation demand closer scrutiny. In summarizing the rationale supporting heightened scrutiny of legislation classifying on the basis of sexual orientation, it would be difficult to improve upon the words of the Supreme Court of Connecticut:

Gay persons have been subjected to and stigmatized by a long history of purposeful and invidious discrimination that continues to manifest itself in society. The characteristic that defines the members of this group — attraction to persons of the same sex — bears no logical relationship to their ability to perform in society, either in familial relations or otherwise as productive citizens. Because sexual orientation is such an essential component of personhood, even if there is some possibility that a person’s sexual preference can be altered, it would be wholly unacceptable for the state to require anyone to do so. Gay persons also represent a distinct minority of the population. It is true, of course, that gay persons recently have made significant 896*896 advances in obtaining equal treatment under the law. Nonetheless, we conclude that, as a minority group that continues to suffer the enduring effects of centuries of legally sanctioned discrimination, laws singling them out for disparate treatment are subject to heightened judicial scrutiny to ensure that those laws are not the product of such historical prejudice and stereotyping.

Kerrigan, 957 A.2d at 432. But see Conaway, 932 A.2d at 609-14 (holding sexual-orientation-based legislation is not entitled to heightened scrutiny because lesbian and gay people are not politically powerless); Andersen, 138 P.3d at 974 (determining plaintiffs failed to satisfy burden to prove homosexuality is not an immutable trait and consequently holding sexual-orientation-based distinctions do not demand closer judicial scrutiny). We agree with the observations of the Connecticut Supreme Court. The factors established to guide our determination of the level of scrutiny to utilize in our examination of the equal protection claim in this case all point to an elevated level of scrutiny. Accordingly, we hold that legislative classifications based on sexual orientation must be examined under a heightened level of scrutiny under the Iowa Constitution.

H. Application of Heightened Scrutiny. Plaintiffs argue sexual-orientation-based statutes should be subject to the most searching scrutiny. The County asserts Iowa’s marriage statute, section 595.2, may be reviewed, at most, according to an intermediate level of scrutiny. Because we conclude Iowa’s same-sex marriage statute cannot withstand intermediate scrutiny, we need not decide whether classifications based on sexual orientation are subject to a higher level of scrutiny. Thus, we turn to a discussion of the intermediate scrutiny standard.[23]

1. Intermediate scrutiny standard. “To withstand intermediate scrutiny, a statutory classification must be substantially related to an important governmental objective.” Clark v. Jeter, 486 U.S. 456, 461, 108 S.Ct. 1910, 1914, 100 897*897 L.Ed.2d 465, 472 (1988). In applying an intermediate standard to review gender-based classifications, the Supreme Court has stated: “Focusing on the differential treatment or denial of opportunity for which relief is sought, the reviewing court must determine whether the proffered justification is `exceedingly persuasive.'” Virginia, 518 U.S. at 532-33, 116 S.Ct. at 2275, 135 L.Ed.2d at 751. To this end, courts evaluate whether the proffered governmental objectives are important and whether the statutory classification is “`substantially related to the achievement of those objectives.'” Id. at 533, 116 S.Ct. at 2275, 135 L.Ed.2d at 751 (quoting Miss. Univ. for Women v. Hogan, 458 U.S. 718, 724, 102 S.Ct. 3331, 3336, 73 L.Ed.2d 1090, 1098 (1982)).

2. Statutory classification: exclusion of gay and lesbian people from civil marriage. To identify the statutory classification, we focus on the “differential treatment or denial of opportunity for which relief is sought.” Id. at 532-33, 116 S.Ct. at 2275, 135 L.Ed.2d at 751 (considering “categorical exclusion” of women from institution of higher education). Plaintiffs bring this lawsuit complaining of their exclusion from the institution of civil marriage. In response, the County offers support for the legislature’s decision to statutorily establish heterosexual civil marriage. Because the relevant focal point is the opportunity sought by the plaintiffs, the issue presented by this lawsuit is whether the state has “exceedingly persuasive” reasons for denying civil marriage to same-sex couples, not whether state-sanctioned, heterosexual marriage is constitutional. See id. at 531, 116 S.Ct. at 2274, 135 L.Ed.2d at 751. Thus, the question we must answer is whether excluding gay and lesbian people from civil marriage is substantially related to any important governmental objective.

3. Governmental objectives. The County has proffered a number of objectives supporting the marriage statute. These objectives include support for the “traditional” institution of marriage, the optimal procreation and rearing of children, and financial considerations.[24]

The first step in scrutinizing a statutory classification can be to determine whether the objectives purportedly advanced by the classification are important. “The burden of justification is demanding and it rests entirely on the State.” Id. at 533, 116 S.Ct. at 2275, 135 L.Ed.2d at 751. Where we find, or can assume, the proffered governmental interests are sufficiently weighty to be called “important,” the critical inquiry is whether these governmental objectives can fairly be said to be advanced by the legislative classification. See, e.g., Fed. Land Bank v. Arnold, 426 N.W.2d 153, 156 (Iowa 1988) (“First we must examine the legitimacy of the end to be achieved; we then scrutinize the means used to achieve that end.”). In this analysis, we drill down to analyze the “link 898*898 between classification and objective.” Romer, 517 U.S. at 632, 116 S.Ct. at 1627, 134 L.Ed.2d at 866.

a. Maintaining traditional marriage. First, the County argues the same-sex marriage ban promotes the “integrity of traditional marriage” by “maintaining the historical and traditional marriage norm ([as] one between a man and a woman).” This argument is straightforward and has superficial appeal. A specific tradition sought to be maintained cannot be an important governmental objective for equal protection purposes, however, when the tradition is nothing more than the historical classification currently expressed in the statute being challenged. When a certain tradition is used as both the governmental objective and the classification to further that objective, the equal protection analysis is transformed into the circular question of whether the classification accomplishes the governmental objective, which objective is to maintain the classification. In other words, the equal protection clause is converted into a “`barren form of words'” when “`discrimination… is made an end in itself.'” Tussman & tenBroek, 37 Cal. L.Rev. at 357 (quoting Truax v. Raich, 239 U.S. 33, 41, 36 S.Ct. 7, 10, 60 L.Ed. 131, 135 (1915)).

This precise situation is presented by the County’s claim that the statute in this case exists to preserve the traditional understanding of marriage. The governmental objective identified by the County — to maintain the traditional understanding of marriage — is simply another way of saying the governmental objective is to limit civil marriage to opposite-sex couples. Opposite-sex marriage, however, is the classification made under the statute, and this classification must comply with our principles of equal protection. Thus, the use of traditional marriage as both the governmental objective and the classification of the statute transforms the equal protection analysis into the question of whether restricting marriage to opposite-sex couples accomplishes the governmental objective of maintaining opposite-sex marriage.

This approach is, of course, an empty analysis. It permits a classification to be maintained “`for its own sake.'” Kerrigan, 957 A.2d at 478 (quoting Romer, 517 U.S. at 635, 116 S.Ct. at 1629, 134 L.Ed.2d at 868). Moreover, it can allow discrimination to become acceptable as tradition and helps to explain how discrimination can exist for such a long time. If a simple showing that discrimination is traditional satisfies equal protection, previous successful equal protection challenges of invidious racial and gender classifications would have failed. Consequently, equal protection demands that “`the classification ([that is], the exclusion of gay [persons] from civil marriage) must advance a state interest that is separate from the classification itself.'” Id. (quoting Hernandez v. Robles, 7 N.Y.3d 338, 821 N.Y.S.2d 770, 855 N.E.2d 1, 33 (2006) (Kaye, C.J., dissenting)); see also Romer, 517 U.S. at 635, 116 S.Ct. at 1629, 134 L.Ed.2d at 868 (rejecting “classification of persons undertaken for its own sake”).

“[W]hen tradition is offered to justify preserving a statutory scheme that has been challenged on equal protection grounds, we must determine whether the reasons underlying that tradition are sufficient to satisfy constitutional requirements.” Kerrigan, 957 A.2d at 478-79 (emphasis added). Thus, we must analyze the legislature’s objective in maintaining the traditional classification being challenged.

The reasons underlying traditional marriage may include the other objectives asserted by the County, objectives we will separately address in this decision. However, some underlying reason other than 899*899 the preservation of tradition must be identified.[25] Because the County offers no particular governmental reason underlying the tradition of limiting civil marriage to heterosexual couples, we press forward to consider other plausible reasons for the legislative classification.

b. Promotion of optimal environment to raise children. Another governmental objective proffered by the County is the promotion of “child rearing by a father and a mother in a marital relationship which social scientists say with confidence is the optimal milieu for child rearing.” This objective implicates the broader governmental interest to promote the best interests of children. The “best interests of children” is, undeniably, an important governmental objective. Yet, we first examine the underlying premise proffered by the County that the optimal environment for children is to be raised within a marriage of both a mother and a father.

Plaintiffs presented an abundance of evidence and research, confirmed by our independent research, supporting the proposition that the interests of children are served equally by same-sex parents and opposite-sex parents. On the other hand, we acknowledge the existence of reasoned opinions that dual-gender parenting is the optimal environment for children. These opinions, while thoughtful and sincere, were largely unsupported by reliable scientific studies.[26]

Even assuming there may be a rational basis at this time to believe the legislative classification advances a legitimate government interest, this assumed fact would not be sufficient to survive the equal protection analysis applicable in this case. In order to ensure this classification based on sexual orientation is not borne of prejudice and stereotype, intermediate scrutiny demands a closer relationship between the legislative classification and the purpose of the classification than mere rationality. Under intermediate scrutiny, the relationship between the government’s goal and the classification employed to further that goal must be “substantial.” Clark, 486 U.S. at 461, 108 S.Ct. at 1914, 100 L.Ed.2d at 472. In order to evaluate that relationship, it is helpful to consider whether the legislation is over-inclusive or under-inclusive. See RACI II, 675 N.W.2d at 10 (considering under-inclusion and over-inclusion even in the rational basis context).

A statute is under-inclusive when the classification made in the statute “does not include all who are similarly situated with respect to the purpose of the law.” Tussman & tenBroek, 37 Cal. L.Rev. at 348. An under-inclusive statute means all people 900*900 included in the statutory classification have the trait that is relevant to the aim of the statute, but other people with the trait are not included in the classification. See id. A statute is over-inclusive when the classification made in the statute includes more persons than those who are similarly situated with respect to the purpose of the law. See id. at 351. An over-inclusive statute “imposes a burden upon a wider range of individuals than are included in the class of those” with the trait relevant to the aim of the law. Id. As the degree to which a statutory classification is shown to be over-inclusive or under-inclusive increases, so does the difficulty in demonstrating the classification substantially furthers the legislative goal.

We begin with the County’s argument that the goal of the same-sex marriage ban is to ensure children will be raised only in the optimal milieu. In pursuit of this objective, the statutory exclusion of gay and lesbian people is both under-inclusive and over-inclusive. The civil marriage statute is under-inclusive because it does not exclude from marriage other groups of parents — such as child abusers, sexual predators, parents neglecting to provide child support, and violent felons — that are undeniably less than optimal parents. Such under-inclusion tends to demonstrate that the sexual-orientation-based classification is grounded in prejudice or “overbroad generalizations about the different talents, capacities, or preferences” of gay and lesbian people, rather than having a substantial relationship to some important objective. See Virginia, 518 U.S. at 533, 116 S.Ct. at 2275, 135 L.Ed.2d at 751 (rejecting use of overbroad generalizations to classify). If the marriage statute was truly focused on optimal parenting, many classifications of people would be excluded, not merely gay and lesbian people.

Of course, “[r]eform may take one step at a time, addressing itself to the phase of the problem which seems most acute to the legislative mind.” Knepper v. Monticello State Bank, 450 N.W.2d 833, 837 (Iowa 1990) (citing Williamson v. Lee Optical of Okla., 348 U.S. 483, 489, 75 S.Ct. 461, 465, 99 L.Ed. 563, 573 (1955)). Thus, “[t]he legislature may select one phase of one field and apply a remedy there, neglecting the others.” Williamson, 348 U.S. at 489, 75 S.Ct. at 465, 99 L.Ed. at 573. While a statute does not automatically violate equal protection merely by being under-inclusive, the degree of under-inclusion nonetheless indicates the substantiality of the relationship between the legislative means and end.

As applied to this case, it could be argued the same-sex marriage ban is just one legislative step toward ensuring the optimal environment for raising children. Under this argument, the governmental objective is slightly more modest. It seeks to reduce the number of same-sex parent households, nudging our state a step closer to providing the asserted optimal milieu for children. Even evaluated in light of this narrower objective, however, the ban on same-sex marriage is flawed.

The ban on same-sex marriage is substantially over-inclusive because not all same-sex couples choose to raise children. Yet, the marriage statute denies civil marriage to all gay and lesbian people in order to discourage the limited number of same-sex couples who desire to raise children. In doing so, the legislature includes a consequential number of “individuals within the statute’s purview who are not afflicted with the evil the statute seeks to remedy.” Conaway, 932 A.2d at 649 (Raker, J., concurring in part and dissenting).

At the same time, the exclusion of gay and lesbian people from marriage is under-inclusive, even in relation to the narrower 901*901 goal of improving child rearing by limiting same-sex parenting. Quite obviously, the statute does not prohibit same-sex couples from raising children. Same-sex couples currently raise children in Iowa, even while being excluded from civil marriage, and such couples will undoubtedly continue to do so. Recognition of this under-inclusion puts in perspective just how minimally the same-sex marriage ban actually advances the purported legislative goal. A law so simultaneously over-inclusive and under-inclusive is not substantially related to the government’s objective. In the end, a careful analysis of the over- and under-inclusiveness of the statute reveals it is less about using marriage to achieve an optimal environment for children and more about merely precluding gay and lesbian people from civil marriage.

If the statute was truly about the best interest of children, some benefit to children derived from the ban on same-sex civil marriages would be observable. Yet, the germane analysis does not show how the best interests of children of gay and lesbian parents, who are denied an environment supported by the benefits of marriage under the statute, are served by the ban. Likewise, the exclusion of gays and lesbians from marriage does not benefit the interests of those children of heterosexual parents, who are able to enjoy the environment supported by marriage with or without the inclusion of same-sex couples.

The ban on same-sex civil marriage can only logically be justified as a means to ensure the asserted optimal environment for raising children if fewer children will be raised within same-sex relationships[27] or more children will be raised in dual-gender marriages. Yet, the same-sex-marriage ban will accomplish these outcomes only when people in same-sex relationships choose not to raise children without the benefit of marriage or when children are adopted by dual-gender couples who would have been adopted by same-sex couples but for the same-sex civil marriage ban. We discern no substantial support for this proposition. These outcomes, at best, are minimally advanced by the classification. Consequently, a classification that limits civil marriage to opposite-sex couples is simply not substantially related to the objective of promoting the optimal environment to raise children. This conclusion suggests stereotype and prejudice, or some other unarticulated reason, could be present to explain the real objectives of the statute.

c. Promotion of procreation. The County also proposes that government endorsement of traditional civil marriage will result in more procreation. It points out that procreation is important to the continuation of the human race, and opposite-sex couples accomplish this objective because procreation occurs naturally within this group. In contrast, the County points out, same-sex couples can procreate only through assisted reproductive techniques, and some same-sex couples may choose not to procreate. While heterosexual marriage does lead to procreation, the argument by the County fails to address the real issue in our required analysis of the objective: whether exclusion of gay and lesbian individuals from the institution of civil marriage will result in more procreation? If procreation is the true objective, 902*902 then the proffered classification must work to achieve that objective.

Conceptually, the promotion of procreation as an objective of marriage is compatible with the inclusion of gays and lesbians within the definition of marriage. Gay and lesbian persons are capable of procreation. Thus, the sole conceivable avenue by which exclusion of gay and lesbian people from civil marriage could promote more procreation is if the unavailability of civil marriage for same-sex partners caused homosexual individuals to “become” heterosexual in order to procreate within the present traditional institution of civil marriage. The briefs, the record, our research, and common sense do not suggest such an outcome. Even if possibly true, the link between exclusion of gay and lesbian people from marriage and increased procreation is far too tenuous to withstand heightened scrutiny. Specifically, the statute is significantly under-inclusive with respect to the objective of increasing procreation because it does not include a variety of groups that do not procreate for reasons such as age, physical disability, or choice. In other words, the classification is not substantially related to the asserted legislative purpose.

d. Promoting stability in opposite-sex relationships. A fourth suggested rationale supporting the marriage statute is “promoting stability in opposite sex relationships.” While the institution of civil marriage likely encourages stability in opposite-sex relationships, we must evaluate whether excluding gay and lesbian people from civil marriage encourages stability in opposite-sex relationships. The County offers no reasons that it does, and we can find none. The stability of opposite-sex relationships is an important governmental interest, but the exclusion of same-sex couples from marriage is not substantially related to that objective.

e. Conservation of resources. The conservation of state resources is another objective arguably furthered by excluding gay and lesbian persons from civil marriage. The argument is based on a simple premise: couples who are married enjoy numerous governmental benefits, so the state’s fiscal burden associated with civil marriage is reduced if less people are allowed to marry. In the common sense of the word, then, it is “rational” for the legislature to seek to conserve state resources by limiting the number of couples allowed to form civil marriages. By way of example, the County hypothesizes that, due to our laws granting tax benefits to married couples, the State of Iowa would reap less tax revenue if individual taxpaying gay and lesbian people were allowed to obtain a civil marriage. Certainly, Iowa’s marriage statute causes numerous government benefits, including tax benefits, to be withheld from plaintiffs.[28] Thus, the ban 903*903 on same-sex marriages may conserve some state resources. Excluding any group from civil marriage — African-Americans, illegitimates, aliens, even red-haired individuals — would conserve state resources in an equally “rational” way. Yet, such classifications so obviously offend our society’s collective sense of equality that courts have not hesitated to provide added protections against such inequalities.

One primary requirement of the equal protection clause is a more substantial relationship between the legislative goal and the means used to attain the goal. When heightened scrutiny is applicable, the means must substantially further the legislative end. Consequently, in this case, the sexual-orientation-based classification must substantially further the conservation-of-resources objective.

As observed in our analysis of the other reasons offered in support of the marriage statute, significant degrees of over-inclusion and under-inclusion shed light on the true relationship between exclusion of gay and lesbian people from civil marriage and the goal of conserving governmental resources. Exclusion of all same-sex couples is an extremely blunt instrument for conserving state resources through limiting access to civil marriage. In other words, the exclusion of same-sex couples is over-inclusive because many same-sex couples, if allowed to marry, would not use more state resources than they currently consume as unmarried couples. To reference the County’s example, while many heterosexual couples who have obtained a civil marriage do not file joint tax returns — or experience any other tax benefit from marital status — many same-sex couples may not file a joint tax return either. The two classes created by the statute — opposite-sex couples and same-sex couples — may use the same amount of state resources. Thus, the two classes are similarly situated for the purpose of conserving state resources, yet the classes are treated differently by the law. In this way, sexual orientation is a flawed indicator of resource usage.

Just as exclusion of same-sex couples from marriage is a blunt instrument, however, it is also significantly undersized if the true goal is to conserve state resources. That is to say, the classification is under-inclusive. The goal of conservation of state resources would be equally served by excluding any similar-sized group from civil marriage. Indeed, under the County’s logic, more state resources would be conserved by excluding groups more numerous than Iowa’s estimated 5800 same-sex couples (for example, persons marrying for a second or subsequent time). Importantly, there is also no suggestion same-sex couples would use more state resources if allowed to obtain a civil marriage than heterosexual couples who obtain a civil marriage.

Such over-inclusion and under-inclusion demonstrates the trait of sexual orientation is a poor proxy for regulating aspiring spouses’ usage of state resources. This tenuous relationship between the classification and its purpose demonstrates many people who are similarly situated with respect to the purpose of the law are 904*904 treated differently. As a result, the sexual-orientation-based classification does not substantially further the suggested governmental interest, as required by intermediate scrutiny.

4. Conclusion. Having examined each proffered governmental objective through the appropriate lens of intermediate scrutiny, we conclude the sexual-orientation-based classification under the marriage statute does not substantially further any of the objectives. While the objectives asserted may be important (and many undoubtedly are important), none are furthered in a substantial way by the exclusion of same-sex couples from civil marriage. Our equal protection clause requires more than has been offered to justify the continued existence of the same-sex marriage ban under the statute.

I. Religious Opposition to Same-Sex Marriage. Now that we have addressed and rejected each specific interest advanced by the County to justify the classification drawn under the statute, we consider the reason for the exclusion of gay and lesbian couples from civil marriage left unspoken by the County: religious opposition to same-sex marriage. The County’s silence reflects, we believe, its understanding this reason cannot, under our Iowa Constitution, be used to justify a ban on same-sex marriage.

While unexpressed, religious sentiment most likely motivates many, if not most, opponents of same-sex civil marriage and perhaps even shapes the views of those people who may accept gay and lesbian unions but find the notion of same-sex marriage unsettling.[29] Consequently, we address the religious undercurrent propelling the same-sex marriage debate as a means to fully explain our rationale for rejecting the dual-gender requirement of the marriage statute.

It is quite understandable that religiously motivated opposition to same-sex civil marriage shapes the basis for legal opposition to same-sex marriage, even if only indirectly. Religious objections to same-sex marriage are supported by thousands of years of tradition and biblical interpretation.[30] The belief that the “sanctity of marriage” would be undermined by the inclusion of gay and lesbian couples bears a striking conceptual resemblance to the expressed secular rationale for maintaining the tradition of marriage as a union between dual-gender couples, but better identifies the source of the opposition. Whether expressly or impliedly, much of society rejects same-sex marriage due to sincere, deeply ingrained — even fundamental — religious belief.

Yet, such views are not the only religious views of marriage. As demonstrated 905*905 by amicus groups, other equally sincere groups and people in Iowa and around the nation have strong religious views that yield the opposite conclusion.[31]

This contrast of opinions in our society largely explains the absence of any religion-based rationale to test the constitutionality of Iowa’s same-sex marriage ban. Our constitution does not permit any branch of government to resolve these types of religious debates and entrusts to courts the task of ensuring government avoids them. See Iowa Const. art. I, § 3 (“The general assembly shall make no law respecting an establishment of religion….”). The statute at issue in this case does not prescribe a definition of marriage for religious institutions. Instead, the statute declares, “Marriage is a civil contract” and then regulates that civil contract. Iowa Code § 595A.1. Thus, in pursuing our task in this case, we proceed as civil judges, far removed from the theological debate of religious clerics, and focus only on the concept of civil marriage and the state licensing system that identifies a limited class of persons entitled to secular rights and benefits associated with civil marriage.

We, of course, have a constitutional mandate to protect the free exercise of religion in Iowa, which includes the freedom of a religious organization to define marriages it solemnizes as unions between a man and a woman. See Iowa Const. art. I, § 3 (“The general assembly shall make no law … prohibiting the free exercise [of religion]….”). This mission to protect religious freedom is consistent with our task to prevent government from endorsing any religious view. State government can have no religious views, either directly or indirectly, expressed through its legislation. Knowlton v. Baumhover, 182 Iowa 691, 710, 166 N.W. 202, 208 (1918). This proposition is the essence of the separation of church and state.

As a result, civil marriage must be judged under our constitutional standards of equal protection and not under religious doctrines or the religious views of individuals. This approach does not disrespect or denigrate the religious views of many Iowans who may strongly believe in marriage as a dual-gender union, but considers, as we must, only the constitutional rights of all people, as expressed by the promise of equal protection for all. We are not permitted to do less and would damage our constitution immeasurably by trying to do more.

The only legitimate inquiry we can make is whether [the statute] is constitutional. If it is not, its virtues … cannot save it; if it is, its faults cannot be invoked to accomplish its destruction. If the provisions of the Constitution be not upheld when they pinch as well as when they comfort, they may as well be abandoned.

Home Bldg. & Loan Ass’n v. Blaisdell, 290 U.S. 398, 483, 54 S.Ct. 231, 256, 78 L.Ed. 413, 452 (1934) (Sutherland, J., dissenting).

In the final analysis, we give respect to the views of all Iowans on the issue of same-sex marriage — religious or otherwise — by giving respect to our constitutional principles. These principles require 906*906 that the state recognize both opposite-sex and same-sex civil marriage. Religious doctrine and views contrary to this principle of law are unaffected, and people can continue to associate with the religion that best reflects their views. A religious denomination can still define marriage as a union between a man and a woman, and a marriage ceremony performed by a minister, priest, rabbi, or other person ordained or designated as a leader of the person’s religious faith does not lose its meaning as a sacrament or other religious institution. The sanctity of all religious marriages celebrated in the future will have the same meaning as those celebrated in the past. The only difference is civil marriage will now take on a new meaning that reflects a more complete understanding of equal protection of the law. This result is what our constitution requires.

J. Constitutional Infirmity. We are firmly convinced the exclusion of gay and lesbian people from the institution of civil marriage does not substantially further any important governmental objective. The legislature has excluded a historically disfavored class of persons from a supremely important civil institution without a constitutionally sufficient justification. There is no material fact, genuinely in dispute, that can affect this determination.

We have a constitutional duty to ensure equal protection of the law. Faithfulness to that duty requires us to hold Iowa’s marriage statute, Iowa Code section 595.2, violates the Iowa Constitution. To decide otherwise would be an abdication of our constitutional duty. If gay and lesbian people must submit to different treatment without an exceedingly persuasive justification, they are deprived of the benefits of the principle of equal protection upon which the rule of law is founded. Iowa Code section 595.2 denies gay and lesbian people the equal protection of the law promised by the Iowa Constitution.[32]

V. Remedy.

Because our civil marriage statute fails to provide equal protection of the law under the Iowa Constitution, we must decide how to best remedy the constitutional violation. The sole remedy requested by plaintiffs is admission into the institution of civil marriage. The County does not suggest an alternative remedy. The high courts of other jurisdictions have remedied constitutionally invalid bans on same-sex marriage in two ways. Some courts have ordered gay and lesbian people to be allowed to access the institution of civil marriage. See In re Marriage Cases, 183 P.3d at 453; Kerrigan, 957 A.2d at 480 (“definition of marriage necessarily must be expanded” to include same-sex couples); Opinions of the Justices to the Senate, 440 Mass. 1201, 802 N.E.2d 565, 571 (2004). Other courts have allowed their state legislatures to create parallel civil institutions for same-sex couples. See Lewis v. Harris, 188 N.J. 415, 908 A.2d 196, 221 (2006); Baker v. State, 170 Vt. 194, 744 A.2d 864, 887 (1999).

Iowa Code section 595.2 is unconstitutional because the County has been unable to identify a constitutionally adequate justification for excluding plaintiffs from the institution of civil marriage. A new distinction based on sexual orientation would be equally suspect and difficult to square with the fundamental principles of equal protection embodied in our constitution. This record, our independent research, and the appropriate equal protection analysis 907*907 do not suggest the existence of a justification for such a legislative classification that substantially furthers any governmental objective. Consequently, the language in Iowa Code section 595.2 limiting civil marriage to a man and a woman must be stricken from the statute, and the remaining statutory language must be interpreted and applied in a manner allowing gay and lesbian people full access to the institution of civil marriage.

VI. Conclusion.

The district court properly granted summary judgment to plaintiffs. Iowa Code section 595.2 violates the equal protection provision of the Iowa Constitution. Our decision becomes effective upon issuance of procedendo.[33]

AFFIRMED.

All justices concur.

[1] The state motto of Iowa is: “Our liberties we prize and our rights we will maintain.” It is inscribed on the Great Seal of Iowa and on our state flag. See Iowa Code §§ 1A.1, 1B.1 (2009).

[2] Iowa Code section 595.2(1) provides “[o]nly a marriage between a male and a female is valid.” All statutory references are to the 2009 Code of Iowa. While some statutes referenced here have been amended since this lawsuit originated, none of the amendments dictate the outcome of this case.

[3] This statement is the official policy of the American Psychological Association regarding sexual orientation, parents, and children. See Am. Psychological Ass’n Council of Representatives, Am. Psychological Ass’n, Resolution on Sexual Orientation, Parents, and Children (2004), in Ruth Ullmann Paige, Proceedings of the American Psychological Association for the Legislative Year 2004: Minutes of the Annual Meeting of the Council of Representatives July 28 & 30, 2004, Honolulu, HI, 60 Am. Psychologist 436-511 (July-August 2005), available at http://www.apa.org/pi/lgbc/policy/parents. html (reporting adoption of resolution).

[4] The cases we have cited are not meant to imply this court has been at the forefront in recognizing civil rights in all areas and at all times. See, e.g., In re Carragher, 149 Iowa 225, 229-30, 128 N.W. 352, 354 (1910) (upholding a law that effectively denied women pharmacists the right to sell alcohol, stating “discrimination between the sexes is neither arbitrary nor capricious, and the fact that in many instances individuals of one sex are in general better fitted than those of the other sex for a given occupation or business is one of such common knowledge and observation that the Legislature may properly recognize it in enacting regulations therefor”). These cases do, however, reflect this court has, for the most part, been at the forefront in recognizing individuals’ civil rights. The path we have taken as a state has not been by accident, but has been navigated with the compass of equality firmly in hand, constructed with a pointer balanced carefully on the pivot of equal protection.

[5] One commentator has found that, since the same-sex-marriage debate started, twenty-seven states have passed constitutional amendments prohibiting same-sex marriage, and seventeen of those state amendments also ban other official forms of same-sex relationships, such as civil unions. Ben Schuman, Note, Gods & Gays: Analyzing the Same-Sex Marriage Debate from a Religious Perspective, 96 Geo. L.J. 2103, 2106-08 (2008). Only one state has recognized same-sex marriage, while several other states recognize civil unions or another form of same-sex relationship. Id. Elizabeth F. Emens, Intimate Discrimination: The State’s Role in the Accidents of Sex and Love, 122 Harv. L. Rev. 1307, 1315 & n.18 (2009) (noting only Massachusetts and Connecticut allow same-sex marriage, while the legality of same-sex marriage in California is currently “in flux”).

[6] Plaintiffs’ challenge to Iowa Code section 595.2 is based on the equal protection guarantee in the Iowa Constitution and does not implicate federal constitutional protections. Generally, we view the federal and state equal protection clauses as “identical in scope, import, and purpose.” Callender, 591 N.W.2d at 187. At the same time, we have jealously guarded our right to “employ a different analytical framework” under the state equal protection clause as well as to independently apply the federally formulated principles. Racing Ass’n of Cent. Iowa v. Fitzgerald, 675 N.W.2d 1, 4-7 (Iowa 2004). Here again, we find federal precedent instructive in interpreting the Iowa Constitution, but we refuse to follow it blindly.

The United States Supreme Court has not resolved the broad question of whether an absolute ban of marriages between persons of the same sex violates the Federal Equal Protection Clause. See Lawrence, 539 U.S. at 578, 123 S.Ct. at 2484, 156 L.Ed.2d at 525 (noting that case does not decide “whether the government must give formal recognition to any relationship that homosexual persons seek to enter”). Nor has the Court resolved many of the narrower legal questions presented by this lawsuit. Nonetheless, the federal framework traditionally employed for resolution of equal protection cases provides a useful starting point for evaluation of Iowa’s constitutional equal protection provision.

[7] Under a traditional rational basis review, courts are required to accept generalized reasons to support the legislation, even if the fit between the means and end is far from perfect. See McGowan v. Maryland, 366 U.S. 420, 425-26, 81 S.Ct. 1101, 1105, 6 L.Ed.2d 393, 399 (1961) (“[L]egislatures are presumed to have acted within their constitutional powers despite the fact that, in practice, their laws result in some inequality.” (Citations omitted.)). Moreover, the challengers bear the burden of negating every conceivable rational basis that might support the classification drawn in the statute. FCC v. Beach Commc’ns, Inc., 508 U.S. 307, 315, 113 S.Ct. 2096, 2102, 124 L.Ed.2d 211, 221 (1993). Some legal commentators have suggested that the occasional practice of courts to examine the purpose of the law more closely under the rational basis test has actually created an additional category of equal protection analysis called “rational basis with bite.” See Gayle Lynn Pettinga, Rational Basis with Bite: Intermediate Scrutiny By Any Other Name, 62 Ind. L.J. 779, 780 (1987) (“[R]ational basis with bite is simply intermediate scrutiny without an articulation of the factors that triggered it….”); Jeremy B. Smith, Note, The Flaws of Rational Basis with Bite: Why the Supreme Court Should Acknowledge Its Application of Heightened Scrutiny to Classifications Based on Sexual Orientation, 73 Fordham L.Rev. 2769, 2774 (2005); Steven P. Wieland, Note, Gambling, Greyhounds, and Gay Marriage: How the Supreme Court Can Use the Rational-Basis Test to Address Varnum v. Brien, 94 Iowa L.Rev. 413, 438-42 (2008) (suggesting the Iowa Supreme Court apply “rational basis with bite”).

[8] References to “heightened” scrutiny in this opinion are meant to be general; heightened scrutiny includes any judicial inquiry more searching than the rational basis test. References to “intermediate” scrutiny discuss a specific level of scrutiny between the rational basis test and strict scrutiny.

[9] While we have applied the threshold analysis in previous cases, we have, at times, directly or indirectly infused that analysis with principles traditionally applied in the complete equal protection analysis. See, e.g., Hennings, 744 N.W.2d at 338-39 (disposing of an equal protection claim with a threshold similarly situated analysis, but within that threshold analysis evaluating the relationship of the state’s interest and the classification made by the statute); see also Timberland Partners XXI, LLP, 757 N.W.2d at 176-77 (implicitly considering relationship between classifications and taxing interests of the state by focusing on distinctions in the use of commercial and residential property); Grovijohn, 643 N.W.2d at 204 (using threshold test to find the notice provisions of the dramshop statute treat all dramshop plaintiffs the same, but determining plaintiff failed to articulate how the adoption of the comparative fault statute altered prior case law finding the notice requirement did not violate equal protection). This approach is almost inevitable for the test to have any real value as an analytical tool to resolve equal protection claims. Consequently, we question the usefulness of the threshold test and express caution in the future use of the threshold analysis. See, e.g., Angie Baker, Note, Leapfrogging Over Equal Protection Analysis: The Eighth Circuit Sanctions Separate and Unequal Prison Facilities for Males and Females in Klinger v. Department of Corrections, 76 Neb. L.Rev. 371, 385 (1999) (noting United States Supreme Court has not applied “similarly situated” analysis as a threshold test); Donna Laddy, Can Women Prisoners Be Carpenters? A Proposed Analysis for Equal Protection Claims of Gender Discrimination in Educational and Vocational Programming at Women’s Prisons, 5 Temple Pol. & Civ. Rts. L.Rev. 1, 22-23 (1995) (same). Because the plaintiffs here satisfy the threshold test we have followed in the past, the outcome in this case would not be affected by abandoning that test now. Therefore, we leave to future parties the task of arguing the applicability of the threshold similarly situated analysis in future cases.

[10] In the past, when a dispute has required this court to choose which level of scrutiny to apply to a given classification, the United States Supreme Court has already determined the appropriate level of scrutiny under the Federal Equal Protection Clause. See, e.g., In re Det. of Williams, 628 N.W.2d 447, 452-53 (Iowa 2001) (relying on Supreme Court’s decision that the mentally ill do not comprise a suspect class); Bennett v. City of Redfield, 446 N.W.2d 467, 473 (Iowa 1989) (deferring to Supreme Court precedent to determine employees of city governments do not comprise a suspect class). We have routinely followed the Supreme Court’s lead, and as a result we have not previously developed an independent analysis under the Iowa Constitution. See, e.g., Williams, 628 N.W.2d at 452-53 (noting Supreme Court holding and containing no independent analysis); Gilleland v. Armstrong Rubber Co., 524 N.W.2d 404, 407 (Iowa 1994) (stating “strict scrutiny analysis only required in few cases involving suspect classes and fundamental rights as defined by Supreme Court”); Bennett, 446 N.W.2d at 473 (noting Supreme Court holding and containing no independent analysis); Stracke v. City of Council Bluffs, 341 N.W.2d 731, 734 (Iowa 1983) (“Because no fundamental right or suspect class, as defined by the United States Supreme Court, is involved in this case, our task is to determine whether a rational basis existed….”). While we again note our authority to develop independent analyses under the Iowa Constitution, we nonetheless view the Supreme Court’s general framework for determining the constitutional “suspectness” of a class as a useful analytical starting point.

[11] Justice Brennan’s opinion for the Supreme Court in Plyler suggests the significance of each of the four factors. Although Plyler does not formulate a specific test, nor even discuss “factors” per se, the case touches upon each of the four traditional factors:

Several formulations might explain our treatment of certain classifications as “suspect.” Some classifications are more likely than others to reflect deep-seated prejudice rather than legislative rationality in pursuit of some legitimate objective. Legislation predicated on such prejudice is easily recognized as incompatible with the constitutional understanding that each person is to be judged individually and is entitled to equal justice under the law. Classifications treated as suspect tend to be irrelevant to any proper legislative goal. Finally, certain groups, indeed largely the same groups, have historically been “relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process.” The experience of our Nation has shown that prejudice may manifest itself in the treatment of some groups. Our response to that experience is reflected in the Equal Protection Clause of the Fourteenth Amendment. Legislation imposing special disabilities upon groups disfavored by virtue of circumstances beyond their control suggests the kind of “class or caste” treatment that the Fourteenth Amendment was designed to abolish.

Plyler, 457 U.S. at 217 n. 14, 102 S.Ct. at 2395 n. 14, 72 L.Ed.2d at 799 n. 14 (citations omitted); see also Kerrigan, 957 A.2d at 426 (identifying two “required factors” and two additional considerations); Dean v. District of Columbia, 653 A.2d 307, 339-40 (D.C.1995) (Ferren, J., dissenting) (identifying four factors and explaining the Supreme Court has not applied all four factors in every case).

[12] See Virginia, 518 U.S. at 531-32, 116 S.Ct. at 2274-75, 135 L.Ed.2d at 750 (observing “`long and unfortunate history of sex discrimination’ ” (quoting Frontiero v. Richardson, 411 U.S. 677, 684, 93 S.Ct. 1764, 1769, 36 L.Ed.2d 583, 590 (1973) (Brennan, J., plurality opinion))); Lyng v. Castillo, 477 U.S. 635, 638, 106 S.Ct. 2727, 2729, 91 L.Ed.2d 527, 533 (1986) (noting subject class had “not been subjected to discrimination”); Cleburne Living Ctr., 473 U.S. at 443, 105 S.Ct. at 3256, 87 L.Ed.2d at 332 (mentally retarded not victims of “continuing antipathy or prejudice”); Mass. Bd. of Ret. v. Murgia, 427 U.S. 307, 313, 96 S.Ct. 2562, 2567, 49 L.Ed.2d 520, 525 (1976) (considering “`history of purposeful unequal treatment'” (quoting San Antonio Indep. Sch. Dist. v. Rodriguez, 411 U.S. 1, 28, 93 S.Ct. 1278, 1294, 36 L.Ed.2d 16, 40 (1973))).

[13] See Cleburne Living Ctr., 473 U.S. at 440, 105 S.Ct. at 3254, 87 L.Ed.2d at 320 (certain classifications merely “reflect prejudice and antipathy”); Miss. Univ. for Women v. Hogan, 458 U.S. 718, 725, 102 S.Ct. 3331, 3336, 73 L.Ed.2d 1090, 1098 (1982) (“Care must be taken in ascertaining whether the statutory objective itself reflects archaic and stereotypic notions.”); Murgia, 427 U.S. at 313, 96 S.Ct. at 2566, 49 L.Ed.2d at 525 (considering whether aged have “been subjected to unique disabilities on the basis of stereotyped characteristics not truly indicative of their abilities”); Frontiero, 411 U.S. at 686, 93 S.Ct. at 1770, 36 L.Ed.2d at 591 (Brennan, J., plurality opinion) (“[T]he sex characteristic frequently bears no relation to ability to perform or contribute to society.”).

[14] Lyng, 477 U.S. at 638, 106 S.Ct. at 2729, 91 L.Ed.2d at 533 (close relatives “do not exhibit obvious, immutable, or distinguishing characteristics that define them as a discrete group”); Cleburne Living Ctr., 473 U.S. at 442, 105 S.Ct. at 3255-56, 87 L.Ed.2d at 322 (mentally retarded people are different from other classes of people, “immutably so, in relevant respects”); Plyler, 457 U.S. at 220, 102 S.Ct. at 2396, 72 L.Ed.2d at 801 (children of illegal aliens, unlike their parents, have “legal characteristic[s] over which children can have little control”); Mathews v. Lucas, 427 U.S. 495, 505, 96 S.Ct. 2755, 2762, 49 L.Ed.2d 651, 660 (1976) (status of illegitimacy “is, like race or national origin, a characteristic determined by causes not within the control of the illegitimate individual”); Frontiero, 411 U.S. at 686, 93 S.Ct. at 1770, 36 L.Ed.2d at 591 (Brennan, J., plurality opinion) (“[S]ex, like race and national origin, is an immutable characteristic determined solely by the accident of birth….”).

[15] Lyng, 477 U.S. at 638, 106 S.Ct. at 2729, 91 L.Ed.2d at 533 (close relatives of primary household are “not a minority or politically powerless”); Cleburne Living Ctr., 473 U.S. at 445, 105 S.Ct. at 3257, 87 L.Ed.2d at 324 (refusing to find “that the mentally retarded are politically powerless”); San Antonio Indep. Sch. Dist., 411 U.S. at 28, 93 S.Ct. at 1294, 36 L.Ed.2d at 40 (considering whether minority and poor school children were “relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process”).

[16] The Supreme Court has not required, nor even discussed, every factor in every case. See, e.g., Palmore v. Sidoti, 466 U.S. 429, 433-34, 104 S.Ct. 1879, 1882-83, 80 L.Ed.2d 421, 426 (1984) (foregoing analysis of political power); Nyquist v. Mauclet, 432 U.S. 1, 9 n. 11, 97 S.Ct. 2120, 2125 n. 11, 53 L.Ed.2d 63, 71 n. 11 (1977) (jettisoning immutability requirement and scrutinizing classification of resident aliens closely despite aliens’ voluntary status as residents); Mathews, 427 U.S. at 505-06, 96 S.Ct. at 2762-63, 49 L.Ed.2d at 660-61 (according heightened scrutiny to classifications based on illegitimacy despite mutability and political power of illegitimates); Murgia, 427 U.S. at 313-14, 96 S.Ct. at 2567, 49 L.Ed.2d at 525 (omitting any reference to immutability); San Antonio Indep. Sch. Dist., 411 U.S. at 25, 93 S.Ct. at 1292, 36 L.Ed.2d at 38 (omitting any reference to immutability); Frontiero, 411 U.S. at 685-88, 93 S.Ct. at 1770-71, 36 L.Ed.2d at 591-92 (Brennan, J., plurality opinion) (scrutinizing classification based on gender closely despite political power of women); Graham v. Richardson, 403 U.S. 365, 371-72, 91 S.Ct. 1848, 1852, 29 L.Ed.2d 534, 541-42 (1971) (foregoing analysis of immutability); see also Lyng, 477 U.S. at 638, 106 S.Ct. at 2729, 91 L.Ed.2d at 533 (referring to whether members of the class “exhibit obvious, immutable, or distinguishing characteristics that define them as a discrete group” (emphasis added)); Dean, 653 A.2d at 346 (Ferren, J., dissenting) (observing Supreme Court has “`never held that only classes with immutable traits can be deemed suspect'” (quoting Watkins v. U.S. Army, 875 F.2d 699, 725 (9th Cir.1989) (Norris, J., concurring))). Some courts view the Supreme Court’s precedent as according greater weight to the first two factors, see, e.g., Kerrigan, 957 A.2d at 426 (“It bears emphasis, however, that the United States Supreme Court has placed far greater weight — indeed, it invariably has placed dispositive weight — on the first two factors, that is, whether the group has been the subject of long-standing and invidious discrimination and whether the group’s distinguishing characteristic bears no relation to the ability of the group members to perform or function in society.”), or as suggesting the factors are alternative means to heightening scrutiny of a legislative classification, id. at 440 (noting “`a suspect class is one saddled with such disabilities, or subjected to such a history of purposeful unequal treatment, or relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process'” (quoting Murgia, 427 U.S. at 313, 96 S.Ct. at 2567, 49 L.Ed.2d at 525) (emphasis added)). But see Conaway, 932 A.2d at 609-14 (holding sexual-orientation-based legislation is not entitled to heightened scrutiny because gay and lesbian people are not politically powerless, even though other factors are satisfied); Andersen, 138 P.3d at 974 (determining plaintiffs failed to show homosexuality is an immutable trait and consequently holding sexual-orientation-based distinctions do not demand closer judicial scrutiny).

[17] Additional legal authority abounds demonstrating the history of discrimination. See, e.g., In re Marriage Cases, 183 P.3d at 442 (finding “sexual orientation is a characteristic… that is associated with a stigma of inferiority and second-class citizenship, manifested by the group’s history of legal and social disabilities”); Kerrigan, 957 A.2d at 432-33 (concluding gay and lesbian people have “been subjected to and stigmatized by a long history of purposeful and invidious discrimination,” and recounting numerous legal and scientific authorities); Dean, 653 A.2d at 344-45 (“Discrimination against homosexuals has been pervasive in both the public and the private sectors.”); Conaway, 932 A.2d at 609 (“Homosexual persons have been the object of societal prejudice by private actors as well as by the judicial and legislative branches of federal and state governments.”).

[18] The County references plaintiffs’ inability to procreate “naturally,” presumably pointing out each couple’s inability to procreate without assistance. Plaintiffs’ inability to contribute children to society by procreation through sexual intercourse with each other does not dictate the outcome of our consideration under this factor. The inquiry into gay and lesbian people’s ability to contribute to society is a general one, designed to signal whether such classifications routinely risk elevating stereotype over ability. A person’s ability to procreate is merely one of many ways in which the person can contribute to society. While the narrower consideration of plaintiffs’ procreative abilities may be relevant to whether section 595.2 ultimately passes judicial scrutiny, consideration of those abilities is less helpful in determining which level of scrutiny to apply. That is, the inability of gay and lesbian partners to contribute by procreation through sexual intercourse with each other does not indicate whether legislative classifications based on sexual preference — which can conceivably occur in any legislative subject matter area — will generally be based on “stereotyped characteristics not truly indicative of their abilities.” Murgia, 427 U.S. at 313, 96 S.Ct. at 2566, 49 L.Ed.2d at 525.

[19] The legislature has further indicated the irrelevancy of sexual orientation by mandating sex education in the state’s public schools be free of biases relating to sexual orientation, Iowa Code § 279.50, and by securing personal freedom from violence and intimidation due to sexual orientation, id. § 729A.1. Likewise, numerous state administrative regulations indicate sexual orientation is not relevant to a person’s ability to contribute to society. See Iowa Admin. Code r. 191-48.9 (prohibiting discrimination in making or solicitation of viatical settlement contracts on basis of sexual orientation); id. r. 281-12 (preamble) (ensuring access to education meeting child’s needs and abilities regardless of sexual orientation); id. r. 281-12.1 (ordering equal opportunity in educational programs regardless of sexual orientation); id. r. 281-12.3 (ordering school boards to consider the potential disparate impact of student responsibility and discipline policies on students because of students’ sexual orientation); id. r. 281-68.4 (prohibiting discrimination in admission process to public charter schools based on sexual orientation); id. r. 282-25.3 (labeling denial of participation in benefits of educational program based on sexual orientation an “unethical practice”); id. r. 282-26.3 (prohibiting licensed educators from discriminating based on sexual orientation); id. r. 641-131.7 (allowing public health department to take numerous adverse actions against emergency medical care personnel who “practice, condone, or facilitate” discrimination against a patient on the basis of sexual orientation); id. r. 641-131.8 (allowing public health department to take numerous adverse actions against training program or continuing education providers who “practice, condone, or facilitate” discrimination against a patient on the basis of sexual orientation); id. r. 641-132.10 (allowing denial, probation, revocation, and suspension of authorized emergency medical service programs that discriminate on the basis of sexual orientation); id. r. 645-282.2 (prohibiting licensed social workers from discriminating on the basis of sexual orientation); id. r. 645 — 363.2 (providing that sexual-orientation-based discrimination by sign language interpreters or transliterators is unethical); id. r. 657 — 3.28 (providing that sexual-orientation-based discrimination by pharmacy technicians is unethical); id. r. 657-8.11 (same for licensed pharmacies, licensed pharmacists, and registered pharmacist-interns); id. r. 661-81.2 (prohibiting entrance of information regarding sexual orientation into Iowa law enforcement intelligence network information system in most circumstances).

[20] Other federal and state authority supports such a conclusion. See Kerrigan, 957 A.2d at 435 (relying on Connecticut statutes banning discrimination based on sexual orientation “in every important economic and social institution and activity that the government regulates”); cf. Frontiero, 411 U.S. at 687, 93 S.Ct. at 1771, 36 L.Ed.2d at 592 (Brennan, J., plurality opinion) (interpreting congressional protections against gender discrimination as suggesting legislative determination such classifications are “inherently invidious” and implying significance of “conclusion of coequal branch of Government” in deciding whether to apply heightened scrutiny).

[21] By one measure — occupation of public office — the political power of racial minorities is unbounded in this country today. This fact was on display January 20, 2009, when Barack H. Obama, the African-American son of a native Kenyan, was inaugurated as the forty-fourth President of the United States of America.

[22] See Misha Isaak, Comment, “What’s in a Name?”: Civil Unions & the Constitutional Significance of “Marriage,” 10 U. Pa. J. Const. L. 607, 608 & n. 7 (2008) [hereinafter Isaak] (noting limited success in obtaining legal status for same-sex marriage, including passage of same-sex civil unions in only two states and same-sex civil marriage in only one); Justin Reinheimer, What Lawrence Should Have Said: Reconstructing an Equality Approach, 96 Cal. L.Rev. 505, 516-17 & n. 46 (2008) (reviewing status of state laws relating to legal recognition of same-sex relationships and noting only the California legislature has (twice) passed a law allowing same-sex civil marriage (which was twice vetoed by that state’s governor)).

[23] Plaintiffs and some amici urge us to abandon the traditional tiered analysis. This view is supported by commentators, at least one of which points out no new suspect or quasi-suspect class has been identified since the 1970s and argues the notion of suspect classes is a dead letter in federal constitutional law. See Evan Gerstman, Same Sex Marriage & the Constitution 61-63 (2004). The “rigidified” tiered approach has long had its detractors, see, e.g., San Antonio Indep. Sch. Dist., 411 U.S. at 98, 93 S.Ct. at 1330, 36 L.Ed.2d at 81 (Marshall, J., dissenting), and other states have exercised corresponding sovereign authority to adopt balancing tests in equal protection cases, see, e.g., State v. Ostrosky, 667 P.2d 1184, 1192-93 (Alaska 1983); Greenberg v. Kimmelman, 99 N.J. 552, 494 A.2d 294, 302 (1985); Baker v. State, 170 Vt. 194, 744 A.2d 864, 878 (1999). Although a more flexible analysis is arguably more reflective of Iowa’s constitutional equality mandates and related jurisprudence, we recognize it is sometimes prudent to delay consideration of a new analysis to subsequent cases when the change can be more fully explored. See RACI II, 675 N.W.2d at 6 (“[I]t is prudent to delay any consideration of whether a different analysis is appropriate to a case in which this issue was thoroughly briefed and explored.” (citing In re Det. of Garren, 620 N.W.2d 275, 280 n. 1 (Iowa 2000))). And while plaintiffs have urged us to adopt a balancing standard, the parties’ application of equal protection principles to the facts of this case has focused on the tiered approach. We also conclude we are able to adequately fulfill our “constitutional obligation as the highest court of this sovereign state to determine whether the challenged classification violates Iowa’s constitutional equality provision” by using the traditional approach in this case. Id. at 4. While we once again reaffirm our sovereign authority “to employ a different analytical framework under state constitutional provisions,” RACI II, 675 N.W.2d at 5, we decline to adopt a new approach in this case.

[24] Other jurisdictions considering the validity of legislative exclusion of gay and lesbian people from civil marriage have considered alternative justifications. See, e.g., Kerrigan, 957 A.2d at 476 (uniformity with laws of other jurisdictions); id. at 518 (Zarella, J., dissenting) (regulation of heterosexual procreation); Hernandez v. Robles, 7 N.Y.3d 338, 821 N.Y.S.2d 770, 855 N.E.2d 1, 32-34 (2006) (Kaye, C.J., dissenting) (moral disapproval, uniformity with other jurisdictions); Andersen, 138 P.3d at 982 (avoid “the need to resolve the sometimes conflicting rights and obligations of the same-sex couple and the necessary third party in relation to a child”). We need not independently analyze these alternative justifications as they are not offered to support the Iowa statute. See Virginia, 518 U.S. at 533, 116 S.Ct. at 2275, 135 L.Ed.2d at 751 (“The justification must be genuine, not hypothesized or invented post hoc in response to litigation.”).

[25] The preservation of traditional marriage could only be a legitimate reason for the classification if expanding marriage to include others in its definition would undermine the traditional institution. The County has simply failed to explain how the traditional institution of civil marriage would suffer if same-sex civil marriage were allowed. There is no legitimate notion that a more inclusive definition of marriage will transform civil marriage into something less than it presently is for heterosexuals. Benjamin G. Ledsham, Note, Means to Legitimate Ends: Same-Sex Marriage through the Lens of Illegitimacy-Based Discrimination, 28 Cardozo L.Rev. 2373, 2388 (2007).

[26] The research appears to strongly support the conclusion that same-sex couples foster the same wholesome environment as opposite-sex couples and suggests that the traditional notion that children need a mother and a father to be raised into healthy, well-adjusted adults is based more on stereotype than anything else. In any event, we do not address whether there is a rational basis for the marriage statute, as the sexual-orientation classification made by the statute is subject to a heightened standard of scrutiny.

[27] The County does not specifically contend the goal of Iowa’s marriage statute is to deter gay and lesbian couples from having children. Such a claim would raise serious due process concerns. See Eisenstadt v. Baird, 405 U.S. 438, 453, 92 S.Ct. 1029, 1038, 31 L.Ed.2d 349, 362 (1972) (noting due process concern with governmental interference with decision to conceive children).

[28] Plaintiffs identify over two hundred Iowa statutes affected by civil-marriage status. See, e.g., Iowa Code § 85.31 (dependent surviving spouse receives benefits when spouse death caused by work injury); id. § 135J.1(4) (hospice patient’s family includes spouse); id. § 142C.4 (spouse has power to make decision concerning anatomical gifts); id. § 144A.7 (patient’s spouse determines application of life-sustaining procedures in absence of declaration); id. § 144C.5 (surviving spouse controls disposition of decedent’s remains in absence of declaration); id. § 252A.3(1) (spouse liable for support of other spouse); id. § 252A.3(4) (children of married parents legitimate); id. § 422.7 (spouses may file joint tax return); id. § 422.9(1) (optional standard deduction for married taxpayers); id. § 422.12(1)(b) (spouses eligible for personal exemption credit); id. § 450.3 (inheritance rights of surviving spouses); id. § 450.9 (surviving spouse exempt from inheritance tax on property passed from decedent spouse); id. § 450.10(6) (spousal allowance for surviving spouse); id. § 523I.309 (surviving spouse must consent to decedent spouse’s interment); id. § 613.15 (spouse may recover value of services and support of decedent spouse for wrongful death or negligent injury); id. § 622.9 (restriction of testimony of communication between husband and wife); id. § 633.211(1) (surviving spouse receives decedent spouse’s entire estate in intestacy); id. § 633.236 (surviving spouse has right to elective share); id. § 633.272 (surviving spouse takes under partial intestacy if elective share not exercised); id. § 633.336 (damages for wrongful death). The Government Accounting Office, as of 2005, had identified more than 1000 federal legal rights and responsibilities derived from marriage. Isaak, 10 U. Pa. J. Const. L. at 607 n. 6.

[29] A survey in the Des Moines Register in 2008 found 28.1% of individuals surveyed supported same-sex marriage, 30.2% opposed same-sex marriage but supported civil unions, and 32% of respondents opposed both same-sex marriage and civil unions. Erin Jordan, About 6 in 10 Iowans back same-sex unions, poll finds, Des Moines Register, Nov. 26, 2008, at 4B. The Des Moines Register survey is consistent with a national survey by the PEW Research Center in 2003. This PEW survey found that 59% of Americans oppose same-sex marriage, and 32% favor same-sex marriage. Schuman, 96 Geo. L.J. at 2108. However, opposition to same-sex marriage jumped to 80% for people “with a high level of religious commitment,” with only 12% of such people in favor of same-sex marriage. Id.

[30] Schuman, 96 Geo. L.J. at 2109-12 (discussing the religious arguments against same-sex marriage found in both the Old and New Testaments of the Bible, supporting a conclusion that homosexuality is considered to be a sin and same-sex marriage to be an extension of that sin).

[31] Many religions recognize same-sex marriage, such as Buddhists, Quakers, Unitarians, and Reform and Reconstructionist Jews. Schuman, 96 Geo. L.J. at 2108. Amicus curiae Iowa and National Faith Leaders, Communities, and Scholars point out the United Church of Christ encourages, but does not require, its local congregations to adopt wedding policies that do not discriminate between heterosexual, gay, and lesbian couples, while the Episcopal Church permits priests to perform liturgies and blessings at same-sex weddings as a matter of pastoral care. Additionally, many groups and clergy within various religions are working to achieve inclusion of same-sex marriage. Id. at 2108-09.

[32] Our decision that the statute violates the equal protection clause of the Iowa Constitution makes it unnecessary to address the other bases for plaintiffs’ challenge to Iowa’s marriage statute.

[33] See Iowa R.App. P. 6.1208 (stating procedendo shall issue twenty-one days after the opinion is filed unless a petition for rehearing is filed).