RESEARCH NOTES Minority Group Interests and Political Representation: Gay Elected Officials in the Policy Process* Donald P. Haider-Markel, Mark R. Joslyn, and Chad J. Kniss University of Kansas Of key importance to groups in a democracy is the political representation of their interests in the policy process. The most obvious strategy of groups to achieve representation is to elect officials that identify with group interests. Our research examines the political representation of lesbian and gay interests, exploring the influence of openly gay elected officials on domestic partner policies. Based on the literature, we select and operationalize variables that may influence policy adoption. Analysis on a dataset of 270 localities suggests that elected gay officials are an important determinant for achieving substantive political representation. Our findings also suggest that supportive non-gay elected officials can effectively represent gays in the policy process. Political representation is essential to any democratic system of government. Through electoral institutions individuals and groups are able to select candidates who represent their political preferences. Women, ethnic, and racial minorities are especially conscious of electoral representation, each believing their interests may be brought to the fore and kept there through public officials sharing group identification (Mladenka 1989; Matland 1993; Thomas 1994). Although shared identification and experience cannot guarantee representation, research suggests the presence of elected black and Hispanic officials increases the likelihood that black and Hispanic interests are represented in policy processes (Eisinger 1982; Mladenka 1989; Saltzstein 1989). Lesbians and gay men share similar concerns over political representation. However, it has yet to be determined whether elected gay officials effectively represent the interests of the gay community in the policy process. The exceedingly negative attitudes toward homosexuals as a group and the relatively *A previous version of this paper was presented at the 1998 annual meeting of the Midwest Political Science Association. The authors would like to thank David Lowery and three anonymous reviewers for their suggestions. THE JOURNAL OF POLITICS, Vol. 62, No. 2, May 2000, Pp. 568–577 © 2000 Blackwell Publishers, 350 Main St., Malden, MA 02148, USA, and 108 Cowley Road, Oxford OX4 1JF, UK. Minority Group Interests and Political Representation 569 few gays in the general population suggest lesbian and gay concerns about representation may be particularly acute (Sherrill 1996). As a stigmatized numerical minority, therefore, gays and lesbians may try to achieve political representation through election of openly gay candidates to public office, or by influencing the behavior of elected sympathetic heterosexuals and closeted homosexuals. In this study, we seek to add to the existing literature on political representation in three ways. First, we consider a previously neglected group—gays and lesbians—and attempt to ascertain whether the election of gay candidates produces substantive as opposed to symbolic representation (Eisinger 1982; Saltzstein 1989). In light of the fact that approximately 124 gays and lesbians held public office in 1997, and 64% of these were at the local level (Ness 1999), this question merits some attention. Second, as distinct from previous research that simply identified correlations between group presence and policies related to group interests (see Eisinger 1982; Saltzstein 1989), our study addresses temporal causation by controlling for when elected gays take office and when the related policy is adopted. Finally, we examine the issue of domestic partner policies in American localities to determine if policy is significantly influenced by the presence of gay or lesbian elected officials. Unlike some minority policies previously examined, domestic partner policies clearly—and nearly exclusively— benefit gays and lesbians. As such, this study offers an explicit test of substantive political representation of minorities. Political Representation of Group Interests The representation of citizen interests in most democratic systems occurs through the election of officials who represent those interests. Government officials are elected, in large measure, based on party and group affiliation and issue positions (Campbell et al. 1960). As such, constituents can expect representation based on the extent to which their affiliations and positions coincide with those of their elected officials (Kingdon 1989). When an official clearly belongs to a particular ethnic, racial, or religious group, the group is said to achieve symbolic or descriptive representation (see Fox 1997, Saltzstein 1989, and Swain 1993 for a summary of the literature). The normative inference then follows: Elected officials will spend part of their time representing the interests of their community, thereby translating symbolic representation into substantive representation. Although groups may of course achieve substantive representation without symbolic by electing sympathetic elites (Browning, Marshall, and Tabb 1984), symbolic representation is viewed as a more efficient and reliable means for achieving representation in the policy process (Gerber, Morton, and Rietz 1998). Initial research on black political representation suggests that the election of blacks results in policy benefits to the black community (Cole 1976). In the 1980s 570 Donald P. Haider-Markel, Mark R. Joslyn, Chad J. Kniss and 1990s, researchers continued to find significant links between the election of minorities and women and policies adopted by local and state governments (Browning, Marshall, and Tabb 1984; Eisinger 1982; Mladenka 1989; Saltzstein 1989; Thomas 1994). Although it appears that minority groups are able to employ symbolic representation to advance substantive policy goals, we should not conclude that symbolic representation has no other value. Consequences evidently flow from the mere presence of specific groups, affecting decision processes among both elites and voters (Pitkin 1967; Browning, Marshall, and Tabb 1984). In fact, Wahlke (1971, 288) asserted that “symbolic satisfaction with the process of government is probably more important than specific, instrumental satisfaction with the policy output of the process.” Thus, it appears policy-making profits from the symbolism of minority group inclusion, advancing the legitimacy of governmental outputs. Indeed, the presence of gay officials may challenge many of the arguments against gay-related policies; gay officials can function to undermine negative conventions by effectively articulating and personalizing the wishes of the gay community. In short, gay officials can thus be prominent in city hall politics by sensitizing others to the demands of lesbians and gays.1 Domestic Partner Policy and Politics In the aforementioned literature, political representation links two components: the election of group members and subsequent government action that benefits the represented groups (Saltzstein 1989, 526). However, identifying group interests is difficult. Here we chose a policy that is consistently demanded by gay interests, generally exclusive to gays, and has both symbolic and tangible benefits—domestic partner policies (Murphy 1993). To solve some of the problems associated with the inability to marry, gay advocates have proposed policies for same-sex partners. One such policy is domestic partner registration. Similar to a marriage license, same-sex couples can register their relationship and receive a certificate (Badgett 1994; Murphy 1993). A related policy assists the domestic partners of employees by supplying tangible benefits. In an increasing number of local governments, employees and their 1 On gay officials and domestic partner policies, we can also find anecdotal evidence of symbolic and substantive representation. In Santa Barbara, California, gay council member Tom Roberts was the main force behind the enactment of a 1997 domestic partner benefits policy to city employees (Price 1997). Christine Kehoe, San Diego’s first lesbian city council member, successfully sponsored legislation providing insurance to domestic partners of city employees in 1994. Similarly, Santa Cruz mayor John Laird ensured the passage of a benefits policy as well as other gayfriendly policies (Button, Rienzo, and Wald 1997). Gay and lesbian officials also spearheaded passage of domestic partner policies in Ann Arbor, Chapel Hill, San Francisco, Seattle, and West Hollywood (Haider-Markel 1997a; 1998). Minority Group Interests and Political Representation 571 domestic partners are granted the financial benefits of married partners, including insurance benefits (Badgett 1994).2 Methods and Data To examine the relationship between elected officials and the adoption of domestic partner policies, we constructed a dataset of 270 American cities and counties for our empirical analysis. To do this, we first identified the 105 localities that adopted domestic partner registries, domestic partner benefits, or both as of June 1998 (see Haider-Markel 1998). Second, as a control group, we added the 125 randomly selected localities used by Wald, Button, and Rienzo (1996). Third, to increase the number of localities in our sample that have elected gay officials, we also included the 126 non-randomly selected localities with antidiscrimination laws for gays and lesbians used by Wald, Button, and Rienzo (1996). Due to overlap between the samples, adding the final cases increases our N by 40 cases (those localities with antidiscrimination policies but no domestic partner policy), for a total of 270 localities.3 Dependent Variables As of June 1998, 38 localities had adopted some form of domestic partner registration policy, while 67 localities offered some benefits to non-married government employees in committed relationships.4 For example, Boston began offering limited benefits in 1993, while Atlanta offers both registration and benefits. The fact that significantly more localities have adopted benefits rather than registration suggests the political dynamics of the two policies may differ, with registries perhaps more controversial. As such, we examine registration and benefits as distinct policies. For each domestic partner policy, each locality is assigned a one if the policy was adopted and a zero if the policy was not adopted. This formulation thus requires logistic regression. Political Representation We capture the possible effects of substantive representation with a variable that is a simple count of the number of openly lesbian or gay council members (57) and mayors (5) elected prior to the adoption of a domestic partner policy, 2 Although domestic partner benefit policies can cost local governments a significant amount of money, these policies may actually be less controversial than registries. Registries may provoke significant resistance as opponents fear that this government recognition of homosexual relationships too closely resembles that of heterosexual marriage (i.e., the issuance of a certificate by the government). 3 Dropping the 40 non-random cases does not significantly change our findings, but including these cases provides a more conservative test because it ensures that most localities with elected gay officials are included in our sample. 4 A complete list is available from the lead author at prex@lark.cc.ukans.edu. 572 Donald P. Haider-Markel, Mark R. Joslyn, Chad J. Kniss and were also in office when the policy was adopted.5 For example, Ann Arbor, Michigan, had two gay officials elected before the adoption of its domestic partner registry, and those same two officials were in office when the policy was adopted. As such, Ann Arbor is coded as a two on political representation. New Orleans, Louisiana, meanwhile, adopted a domestic partner registry in 1993, but had no elected gay officials at the time. Thus, New Orleans is coded as zero on political representation. A key contribution here is that unlike previous studies of political representation, our measure of political representation accounts for temporal causality. Past aggregate-level research has not attempted to determine if officials were elected before policies were adopted or when either event occurred—researchers simply tested for potential correlations between the presence of officials and policies (Eisinger 1982; Mladenka 1989; Saltzstein 1989).6 Control Variables Because domestic partner policies have become a significant issue in lesbian and gay politics, the factors likely to influence these policies should be similar to those that influence other gay-related policies. This section outlines the basic logic and operationalization of other determinants of domestic partner policy adoption. Lesbians and gays, the main beneficiaries of domestic partner policies, are more likely to obtain such policies if they can effectively mobilize (HaiderMarkel 1997b). In some American localities, gay groups have considerable strength in the policy-making process (Button, Rienzo, and Wald 1997). We employ two measures of potential gay interest group resources—the potential financial resources of the gay community and a surrogate measure of potential gay interest group members. We measure potential financial resources with a count of the number of gay-owned bars, newspapers, and services per 100,000 persons in a locality.7 Our measure of potential gay and lesbian interest group members is the number of same-sex unmarried partner households per 1,000 local population.8 Similar to previous efforts (Haider-Markel and Meier 1996; Mooney and Lee 1995; Wald, Button, and Rienzo 1996), we capture the effect of conservative religious forces by including a measure of the percentage of a locality’s popu5 Most of the data on gay elected officials can be found in (Haider-Markel 1997a), but are available from the authors on request. The data were collected in a thorough search of gay politics literature, gay newspapers (especially The Washington Blade), the Gay and Lesbian Victory Fund, and the National Gay and Lesbian Task Force. 6 It is also important to recognize that our measure includes many forms of influence, whether it is largely symbolic or purely substantive as aggressive sponsors of domestic partner legislation. In either event, the presence of gays may or may not afford a substantive policy outcome. 7 For a description see Wald, Button, and Rienzo’s (1996) use of the Damron Road Atlas and Haider-Markel and Meier’s (1996) similar measure. 8 Unless otherwise noted, all data are from the U.S. Bureau of the Census (various years). See Haider-Markel (1997b) and Wald, Button, and Rienzo (1996) for a full description of this variable. Minority Group Interests and Political Representation 573 lation that belong to conservative (Protestant or evangelical) denominations.9 Similarly, the Republican party’s focus on traditional family values has led many to suggest Republican opposition to non-traditional gender roles and homosexuality (Layman and Carmines 1997). As such, the percent of a locality’s vote for President Bush in 1992 is included as a measure of local partisanship and, indirectly, support for traditional family values. During policy formulation, the values of both political elites and the masses are often crucial to adoption (Kingdon 1995). Elite support on gay issues is measured as the average percent a locality’s House representative voted for a progay position in the 103rd Congress.10 Finally, because studies find increased levels of education and urbanism are positively associated with greater levels of support for gay civil rights (Haeberle 1996; Donovan and Bowler 1996), we include the percentage of the local population with a college degree and the population of the locality. Results The results from a logistic regression analysis of domestic partner registries in 270 localities are shown in Table 1. The first column displays the results for the model predicting the likelihood a locality will adopt domestic partner registration absent the political representation variable. The second column displays the coefficients of the fully specified model that includes the political representation variable. Moreover, influence statistics appear in the third column in order to clarify the relative impact of the independent variables. Overall, the model performs quite well, predicting approximately 88% of the cases correctly. Moreover, the direction of estimated relationships generally conforms to expectations. For example, support for family values is negatively related to the adoption of registries, while education, elite sentiment, gay businesses, and gay households are positively related to adoption of domestic partner registries. Conservative Protestants, population, and gay households do not, however, achieve statistical significance. After incorporation of the political representation variable, the model’s fit statistics improve. More important, results indicate the presence of lesbian or gay officials significantly increases the probability of domestic partner registration. Inspection of influence statistics indicates that gay officials affect the adoption of registration policies more so than any other factor. The evidence of substantive representation thus appears strong and comparable with findings concerning black and Hispanic political representation in local government. 9 See Wald, Button, and Rienzo (1996) for the denominations classified as religious conservative. The data are from Bradley et al. (1992) and are county-level. 10 For example, a representative who voted for the pro-gay position on three out of four votes would receive a score of 75%. The measure is also weighted by a locality’s share of the population for a House district. See Wald, Button, and Rienzo (1996) for data source and measurement. TABLE 1 Determinants of Domestic Partner Registration and Benefits Policies in Localities Independent Variables Family Values Support/Partisanship Education Religious Conservatives Political Elite Support Gay Businesses Gay Households Population Gay Political Representation Constant Pseudo R-square -2 Log Likelihood Goodness of Fit Chi-Square Percent Correctly Predicted Number of Cases Registration w/o Representation Variable 2.093** (.042) .037** (.014) 2.036 (.023) .018** (.009) 1.154** (.368) 1.168 (.764) .030 (.010) — 21.487 (1.730) .46 140.44 176.38 78.97 88.52 270 Registration with Representation Variable 2.079* (.042) .028* (.016) 2.018 (.022) .021** (.009) 1.045** (.360) .707 (.920) .030 (.020) 1.263** (.370) 22.450 (1.824) .53 124.15 137.30 95.26 88.89 270 Influence 2.037 .021 2.019 .044 .033 .007 .011 .050 Benefits w/o Representation Variable 2.032 (.031) .025** (.012) 2.018 (.015) .029** (.007) .994** (.390) 3.065** (1.437) .090** (.040) — 23.348** (1.428) .53 185.18 193.99 119.58 84.07 270 Benefits with Representation Variable 2.023 (.032) .018 (.014) 2.009 (.015) .032** (.008) .932** (.382) 2.84* (1.45) .090** (.040) 1.193** (.388) 24.00** (1.50) .57 172.80 199.00 131.95 86.30 270 Influence 2.025 .035 2.023 .201 .076 .092 .113 .122 Notes: Coefficients are Logistic regression coefficients. Standard errors are in parentheses. Two-tailed significance test 5 ** Sig. , .05; * Sig. , .10; — indicates omitted variable; holding all other independent variables at their means, “influence” is the change in predicted probability of Y 5 1 of changing this independent variable one standard deviation from its mean. The population coefficients were multiplied by 100,000 for ease of viewing. Minority Group Interests and Political Representation 575 Estimates for the adoption of domestic partner benefits also appear in Table 1. The fourth column displays the estimates for controls only, and the fifth column includes the political representation variable. As expected, relationships differ from the previous model, reflecting a somewhat different pattern of politics. Political elites, gay businesses and households, and mass tolerance—as indicated by population size—have relatively more influence on the adoption of benefits policies. Family values/partisanship, education, and religious conservatives have relatively less influence, however. This finding ostensibly supports the notion that benefits policies are less controversial. Religious forces and moral concerns exert less influence on benefits policy, while potential resources of gay interest groups have more influence on benefits policies than on registration policies. Most important, gay political representation significantly influences the adoption of domestic partner benefits. However, unlike the registration model, it is not the most influential factor. Instead, elite support has the greatest influence. If benefits policies are actually less controversial than registration policies, this might explain why the support of sympathetic elites has more influence on policy than do gay representatives. Heterosexual officials may find it less risky politically to support benefits rather than registration. Conclusions This study examined the question of substantive political representation of gay and lesbian interests in the policy process. We suggested that gay activists are more likely to be successful in the policy-making process if they elect openly gay officials. To test this hypothesis, we examined the role of gay elected officials in local adoption of domestic partner policies using a dataset of 270 American localities. Overall, our empirical results suggest a number of key points. First, the patterns of politics surrounding both types of domestic partner policies differ, but the adoption of these policies is driven by many of the same forces that influence the adoption of gay civil rights policies (Haeberle 1996; Haider-Markel and Meier 1996; Wald, Button, and Rienzo 1996). Second, the presence of supportive factors, such as an educated populace, sympathetic elites, and gay interest group resources, matters more than opposition forces, including support for traditional values and the presence of conservative religious denominations. Finally, while most of these factors significantly influence the adoption of domestic partner policies, the single most important factor influencing registration policy is political representation of lesbians and gays. Elected gays also exert a strong influence on the adoption of domestic partner benefit policies, but the influence of supportive elites is greater. We suggest this distinction arises because heterosexual officials may find it easier to support benefits policies over registration policies. This combination of results suggests that election of gay officials is not simply a symbolic action, as they evidently represent the interests of their commu- 576 Donald P. Haider-Markel, Mark R. Joslyn, Chad J. Kniss nity. In some policy arenas, this substantive political representation may make the difference between policy adoption or failure, but in others, gay officials are one political influence among many. Whether as a symbolic force or legislative entrepreneur, elected gays are important for achieving the policy goals of the gay community. However, our findings also suggest that gays should not ignore alliances with heterosexual officials interested in gay concerns. This finding is especially relevant to gays as a group since they have achieved a minimal level of success in electing gay officials (Ness 1999). More broadly, the results presented in this article underscore the importance of individual legislators. Indeed, the structural designs, behavioral conventions, and normative processes within governing institutions are evidently pliant, accommodating the various personalities and perspectives that electoral democracy provides. Further research would in fact do well to examine how gay, minority, and female officials promote policy ideas, interact with policy experts, frame ideas, and form political coalitions in local, state, and national legislatures. Manuscript submitted 23 February, 1999 Final manuscript received 30 September, 1999 References Badgett, M.V. Lee. 1994. “Equal Pay for Equal Families,” Academe, May–June, 26–30. Bradley, Martin B., Norman M. Green, Dale E. Johnson, Mac Lynn, and Lou McNeil. 1992. Churches and Church Membership in the United States, 1990. Atlanta: Glenmary Research Center. Browning, Rufus P., Dale R. Marshall, and David H. Tabb. 1984. Protest Is not Enough: The Struggle or Blacks and Hispanics for Equality in Urban Politics. Berkeley, CA: University of California Press. Button, James W., Barbara A. Rienzo, and Kenneth D. Wald. 1997. Private Lives, Public Conflicts: Battles Over Gay Rights in American Communities. Washington, DC: CQ Press. Campbell, Angus, Philip E. Converse, Warren E. Miller, and Donald E. Stokes. 1960. The American Voter. Chicago: The University of Chicago Press. Cole, Leonard. 1976. Blacks in Power: A Comparative Study of Black and White Elected Officials. Princeton, NJ: Princeton University Press. Donovan, Todd, and Shaun Bowler. 1997. “Direct Democracy and Minority Rights: Opinions on Anti-Gay and Lesbian Ballot Initiatives.” In Anti-Gay Rights: Assessing Voter Initiatives, ed. S. Witt and S. McCorkle. Westport, CT: Praeger. Eisinger, Peter K. 1982. “Black Employment in Municipal Jobs: The Impact of Black Political Power.” The American Political Science Review 76(2): 380–92. Fox, Richard Logan. 1997. Gender Dynamics in Congressional Elections. Thousand Oaks, CA: Sage Publications. Gerber, Elisabeth R., Rebecca B. Morton, and Thomas A. Rietz. 1998. “Minority Representation in Multimember Districts.” American Political Science Review 92(1): 127–44. Haeberle, Steven H. 1996. “Gay Men and Lesbians at City Hall.” Social Science Quarterly 77(1): 190–97. Haider-Markel, Donald P. 1998. “Policy Entrepreneurs and the Political Representation of Minority Group Interests: The Case of Domestic Partner Policies.” Presented at the annual meetings of the Midwest Political Science Association, Chicago. Haider-Markel, Donald P. 1997a. “From Bullhorns to PACs: Lesbian and Gay Politics, Interest Groups, and Policy.” Ph.D. diss. University of Wisconsin–Milwaukee. Minority Group Interests and Political Representation 577 Haider-Markel, Donald P. 1997b. “Interest Group Survival: Shared Interests Versus Competition for Resources.” Journal of Politics 59(3): 903–12. Haider-Markel, Donald P., and Kenneth J. Meier. 1996. “The Politics of Gay and Lesbian Rights: Expanding the Scope of the Conflict.” Journal of Politics 58(2): 332–49. Kingdon, John. 1989. Congressmen’s Voting Decisions, 3rd ed. New York: Harper and Row. Kingdon, John. 1995. Agendas, Alternative, and Public Policies. 2nd ed. New York: HarperCollins. Layman, Geoffrey C., and Edward G. Carmines. 1997. “Cultural Conflict in American Politics: Religious Traditionalism, Postmaterialism, and U.S. Political Behavior.” Journal of Politics 59(3): 751–77. Matland, Richard E. 1993. “Institutional Variables Affecting Female Representation in National Legislatures: The Case of Norway.” The Journal of Politics 55(3): 737–55. Mladenka, Kenneth R. 1989. “Blacks and Hispanics in Urban Politics.” American Political Science Review 83(1): 165–91. Mooney, Christopher Z., and Mei-Hsien Lee. 1995. “Legislating Morality in the American States: The Case of Pre-Roe Abortion Regulation Reform.” American Journal of Political Science 39(3): 599–627. Murphy, Elizabeth. 1993. Understanding the Domestic Partner Dilemma: Perspectives of Employer and Insurer. 2nd ed. West Hollywood, CA: City of West Hollywood. Ness, Carol. “Openly Gay Officialdom,” The San Francisco Examiner, 25 January 1999. Pitkin, Hannah F. 1967. The Concept of Representation. Berkeley, CA: University of California Press. Price, Deb. “New Gay Rights Projects Need Some Support From New Jersey’s Re-Elected Gov. Whitman,” The Detroit News, 7 November 1997. Saltzstein, Grace Hall. 1989. “Black Mayors and Police Policies.” The Journal of Politics 51(3): 525–44. Sherrill, Kenneth S. 1996. “The Political Power of Lesbians, Gays, and Bisexuals.” PS: Political Science and Politics 29: 469–73. Swain, Carol. 1993. Black Faces, Black Interests. Cambridge, MA: Harvard University Press. Thomas, Sue. 1994. How Women Legislate. New York: Oxford University Press. U.S. Bureau of the Census. Various Years. Statistical Abstract of the United States. Washington, DC: U.S. Bureau of the Census. Wahlke, John C. 1971. “Policy Demands and System Support: The Role of the Represented.” British Journal of Political Science 1: 271–90. Wald, Kenneth D., James W. Button, and Barbara A. Rienzo. 1996. “The Politics of Gay Rights in American Communities: Explaining Antidiscrimination Ordinances and Policies.” American Journal of Political Science 40(4): 1152–78. Donald P. Haider-Markel is assistant professor of political science, University of Kansas, Lawrence, KS 66045. Mark R. Joslyn is assistant professor of political science, University of Kansas, Lawrence, KS 66045. Chad J. Kniss is a Ph.D. candidate in political science, University of Kansas, Lawrence, KS 66045.