Getting Us Where We Live

Published in: September-October 2004 issue.


WHILE THE WORDS “we are everywhere” can be heard frequently at gay and lesbian political events, the 2000 United States Census provided the first empirical confirmation of this rallying cry. The finding that same-sex unmarried partners were present in 99.3 percent of all counties in the United States (Smith and Gates 2001) was one of the most commonly reported statistics from the release of Census 2000. News that gay and lesbian couples live in nearly every community may not seem all that startling. Yet the census data solidify what was only conjecture among some people. In fact, these data can be used to open people’s minds. When informed that 55 same-sex couples were counted in his hometown in Mississippi, Republican State Senator Kean Kirby told The Clarion-Leader (Jackson, Miss.), “Surely you jest. Wow! I have never met any of these people.”

The Census data can help dispel stereotypes and present a more accurate picture of gay and lesbian families. Lobbyists from the Human Rights Campaign (HRC), the largest national gay and lesbian political organization, and from other gay/lesbian civil rights groups now regularly use this information to convince congressional representatives that gay and lesbian people live, and most likely vote, in their districts. Elizabeth Birch, formerly HRC Executive Director, cited the Census data in her 2003 testimony before a Senate Judiciary Subcommittee during hearings on a possible constitutional amendment to ban civil marriage for same-sex couples.

Of course, the importance of understanding the location patterns of gay and lesbian couples goes beyond simply acknowledging that they exist. It goes beyond recognition of their political clout. Gay and lesbian service providers, activist organizations, and an increasing number of companies seeking to market to the gay and lesbian population can all benefit from a more precise understanding of the location patterns and demographic characteristics of this population.

Issues such as civil marriage for same-sex couples, gay and lesbian adoption rights, domestic partner benefits, and hate crime and antidiscrimination statutes (that include sexual orientation) constitute a broad public policy agenda that could be influenced by these new Census data. The same-sex marriage issue in particular looms large in the 2004 presidential campaign, as some conservative groups consider banning such contractual unions a higher immediate priority than abortion restrictions. These high-profile debates are marked by an astonishing lack of empirical data. Until now, it was difficult to assess the potential impact of these policies because so little was known about the gay/lesbian population. That’s about to change.

In addition to enlightening policy debate, hard facts on the gay and lesbian population illustrate that they are a sizable voting bloc and an increasingly visible constituency in many American communities. In the California gubernatorial recall election of 2003, for example, four percent of voters identified as gay, lesbian, or bisexual, a figure similar to national polls in the 1996 and 2000 presidential elections. It is probably no coincidence that eight of the ten states with the highest concentration of gay and lesbian couples voted for Al Gore in the 2000 presidential election, and that fourteen of the twenty states (and the District of Columbia) with higher-than-average gay and lesbian concentrations also supported the Democratic candidate.

Census 2000 makes it possible to document the extent to which states with high proportions of same-sex couples generally have more favorable laws regarding gay men and lesbians. The data can confirm that nine of the thirteen states that have not passed a “Defense of Marriage Act” or “DOMA” restricting marriage to heterosexual couples are among the twenty states with a gay and lesbian couple concentration above the U.S. average. Similarly, nine of the thirteen states with laws banning discrimination based on sexual orientation are among these same twenty states. Statistically, states with more gay/lesbian-supportive laws have higher concentrations of gay and lesbian couples.

DURING the Census collection period in spring 2000, each housing unit in the fifty states, the District of Columbia, and Puerto Rico received a Census questionnaire with six basic questions about each person in the household: name, sex, age, relationship to the householder, Hispanic origin, and race. The householder (the person filling out the Census form) was also asked whether the housing unit was rented or owned. These seven questions make up what is commonly called the Census “short form.”

The Census does not ask any questions about sexual orientation, sexual behavior, or sexual attraction, three common ways used to identify gay men and lesbians in surveys. Rather, Census forms include a number of relationship categories to define how individuals in a household are related to the householder. These fall into two broad categories: related persons (including husband/wife, son/daughter, brother/sister, and so on), and unrelated persons. Since 1990, the Census Bureau has included an “unmarried partner” category to describe an unrelated household member’s relationship to the householder. If the householder designates another adult of the same sex as his or her unmarried partner, the household counts as a same-sex unmarried-partner household.

For this analysis we developed the “Gay/Lesbian Index,” which serves as a measure of the concentration of gay and lesbian couples among households in a particular geographic region. The Index is a ratio of the proportion of same-sex couples living in a region to the proportion of households that are located in a region. This ratio, then, measures the over- or under-representation of same-sex couples in a geographic area relative to the population. A Gay/Lesbian Index value of 1.0 indicates that a same-sex couple is just as likely as a randomly picked household to locate in a particular region. A value of 2.0 indicates that gay- and lesbian-couple households are twice as likely as the “average” U.S. household to locate in that region, while values less than 1.0 indicate that gay and lesbian couples are less likely than the average household to locate there.

Where, then, do gay and lesbian couples live? Census 2000 counts same-sex couples in 99 percent of U.S. counties and 97 percent of Census tracts (geographic units with approximately 2,000 occupied households, regardless of the physical size of the area). However, the distribution of gay and lesbian families is far from uniform across the nation. This analysis explores the location patterns of gay and lesbian couples by comparing these patterns with those of all households in a given county or tract.

County-level location patterns show high concentration levels of gay and lesbian couples along nearly the entire California coast, in southern Florida, and throughout New England. County level location patterns show only a few counties in the Midwest with high concentrations of gay and lesbian couples. The top ten counties ranked by a county-level Gay/Lesbian Index are San Francisco County, Monroe County (Key West, Florida), Hampshire County (Amherst Massachusetts), Washington, D.C., DeKalb County (Atlanta), New York County (Manhattan), Arlington County (Virginia), Suffolk County (Boston), Sonoma County (California), and Denver County.

Gay and lesbian couples are more than three times as likely as the average American household to live in the San Francisco metropolitan area, which ranks first among all metropolitan statistical areas (MSA’s) in the Gay/Lesbian Index. Rankings of MSA’s by population size demonstrate that metropolitan areas in nearly every region of the country have relatively large concentrations of lesbian and gay couples. Note that “university” towns like Madison (Wisconsin), Boulder (Colorado), and Bloomington (Indiana), are common among small and medium-sized MSA’s.

Even though San Francisco ranks highest in the gay- and lesbian-couple concentration at both the county and MSA levels, it is not actually the “gayest” town in America. That distinction goes to Provincetown, Massachusetts, a Cape Cod mecca for gay and lesbian tourists. Same-sex couples are 22 times more likely than the average household to live in Provincetown, where over one in eight households is a gay or lesbian couple.

The proportion of lesbian and gay couples in some communities far exceeds the national average of nearly one in a hundred households. This is also true of many neighborhoods. More than 100 neighborhoods (defined roughly as zip codes) in the U.S. have gay- and lesbian-couple concentrations that exceed one in fifty households. These neighborhoods can be found in 52 different cities located in eighteen states and the District of Columbia, and include ten neighborhoods in San Francisco, nine in Atlanta, six in Seattle, and five each in New York and Portland, Oregon.

A RECENT Massachusetts Supreme Judicial Court decision (Goodridge et al. v. Department of Public Health) that compels the state to grant civil marriage licenses to same-sex couples further heightens the most divisive debate about the lives of gay and lesbian Americans. Proponents of civil marriage for same-sex couples share such traits as commitment, stability, and child rearing with their married counterparts. Denying them the option of marriage amounts to disparate treatment for essentially equal households. Opponents counter that adding same-sex couples to the ranks of the married further erodes a revered institution and potentially undermines its preeminence as the core of family stability and child well-being. So what can Census data tell us about this debate?

Many same-sex couples share characteristics with their married counterparts. One in four same-sex couples is raising children, and, like their married counterparts, gay and lesbian couples with their children have on average two children per household. In the absence of marriage, same-sex couples raising children face a variety of legal complications in terms of parental rights. When same-sex couples cannot establish dual parental rights through a process known as second-parent adoption, the surviving parent and children are left vulnerable if a partner dies. A child without a legal parent may be denied Social Security benefits and health insurance, and parents may not be able to authorize medical treatment for the child in the case of an emergency. Census data show that two-thirds of children being raised by same-sex couples live in states where second-partner adoptions are not a guaranteed option for parents. The lack of second-parent adoption leaves these children at both financial and medical risk.

While the Census does not ask specific questions about the stability of commitment of relationships, respondents are asked about home ownership and whether both members of a partnership lived in the same house five years ago. Both measures provide evidence of stability. More than 67 percent of gay and lesbian couples own their home, a figure above the sixty percent rate of U.S. home ownership and below that of the 78 percent rate among heterosexual couples. Among same-sex couples who did not move in the last five years, 85 percent have lived in their homes with their partner for at least five years. While that number falls below the rate of 95 percent for heterosexual couples, it still offers evidence that a large portion of same-sex couples live in stable homes and maintain long-term relationships.

Ongoing analysis offers the promise of constructing a rich and compelling portrait of same-sex couples and their families. Such efforts honor and respect the important decision of more and half a million gay and lesbian couples simply to be counted.


Gary J. Gates and Jason Ost both serve as research associates in the Center on Labor, Human Services, and Population at the Urban Institute in Washington, D.C. They are the authors of The Gay & Lesbian Atlas, from which this article has been excerpted and adapted (Urban Institute Press, 2004).


Read More from Gary J. Gates