A cross-national analysis of the guarantees of rights

Abstract

What are the predictors of rights guarantees at the level of individual countries? We examine this question within the context of what factors lead certain countries, but not others, to have legislation prohibiting sexual orientation discrimination in the workplace between 1972 and 2002. In the theoretical framework, a combination of domestic forces (past inclusion of minorities, culture, and democratic conditions) and global trends (regulation by supranational bodies and globalization) predict guarantees of rights. To test the theory, generalized estimating equation time-series cross-sectional analyses are performed on data from 161 countries. The results, which are robust to changes in model specification and alternative measurement schemes, confirm our key hypotheses. We conclude by discussing the implications of this research for the study of rights generally.

Keywords

employment equality gay rights generalized estimating equation models

Rights are codified in a variety of ways in different countries. Yet only some have banned discrimination on the basis of sexual orientation in court rulings or by statutory or constitutional provisions (Greenberg et al., 1993; Moustafa, 2003; Sartori, 1994; Sathe, 2004; Songer and Johnson, 2007; Stone Sweet, 2000; Widner, 2001). Legislation passed in Costa Rica, for instance, prohibits discrimination against gay individuals in the workplace and elsewhere. Likewise, certain constitutions, such as the South African, ban discrimination on the basis of sexual orientation (Gibson, 2004; Gloppen, 2005; Haynie, 2003 Schmid, 2000). Whereas Costa Rica, South Africa, and other nations have those rights on the books, the general picture for gay individuals worldwide is rather bleak. The majority of countries, including some established democracies, have yet to include in their legal code a prohibition against discrimination on the basis of sexual orientation in the workplace.

The issue of rights generally, and the rights afforded to sexual minorities specifically, has been studied with a focus on case studies (Adam et al., 1999; Graupner and Tahmindjis, 2005; Wintemute, 1995) or on certain regions of the world, such as Europe (Beger, 2004; Waaldijk, 2000). There have been attempts at tracking systematic influences, for instance, in the context of liberalization of policy related to same-sex sexual activity (Frank and McEneaney, 1999; Frank et al., 2010) or in the case of ratification of international rights covenants (Cole, 2005). Yet there is insufficient research on the question of this article, which examines what variables systematically explain the provision of the guarantees of rights at the level of the state. Previous studies that have used case studies or focused on specific regions do not allow us to dig deeper and examine cross-national, cross-regional, and temporal factors that may have systematic effects.

This project makes several theoretical and empirical contributions to existing work on rights and equality. At the level of theory, equality under law is crucially important for the political inclusion, recognition, and empowerment of a minority group. In fact, some argue that basic rights, such as the right not to be discriminated against in employment, are a stepping stone for full equality under law in all aspects of life (for instance, in marriage (Merin, 2002) and otherwise (Waaldijk, 2000)). Sanders (1996) identifies three stages in state practices towards lesbian, gay, bisexual, and transgender (LGBT) individuals. The first stage is decriminalization, which entails the repeal of legal prohibitions against same-sex sex. ‘[R]epeal of criminal prohibitions,’ according to Sanders, allows then for the ‘enactment of antidiscrimination laws’ (Sanders, 1996: 18). The third stage concerns legal recognition of gay relationships. This project, thus, focuses on the second stage of the theory in Sanders (1996), namely, legal guarantees against discrimination, and identifies two groups of variables predicting equality: domestic and global predictors. The key factor in the group of domestic predictors is political inclusion spillover from one minority group to another. The impact of supranational institutions (such as the International Labor Organization) and globalization belong in the second group. Accordingly, our work extends the ‘norm cascade’ argument (Finnemore and Sikkink, 1998) both globally, as it relates to the rights of the LGBT community, and to the domestic playing field of rights, where it has been explored much less. Our study highlights the importance of how inclusion of one group can influence the inclusion of another as well as the impact a commitment to international treaties can have on the legal protections discriminated groups can enjoy. Our findings suggest that this combination of domestic and international factors influences the likelihood that a government would protect LGBT individuals from sexual orientation discrimination in the workplace.

At the empirical level, controlling for additional alternative accounts, this theory is tested with data for discrimination in employment against LGBT individuals in 161 countries over a period of three decades starting in 1972. The study of rights and liberties offered to members of sexual minority groups in a cross-national framework naturally controls for idiosyncrasies, which make it harder to compare other minorities. For instance, comparing the rights in different countries of other types of minorities is likely to involve groups that are dissimilar on variables ranging from their proportion in the population to the reason why they are treated as a minority. Accordingly, the study of discrimination against this group in a cross-national perspective is conducive to a systematic explanation of the guarantees of equality. Furthermore, we propose time-series cross-sectional analyses, whose explanatory power is superior to cross-sectional snapshots taken at different points in time (e.g. Frank and McEneaney, 1999). Finally, to the best of our knowledge, the number of states included in our study is larger than any quantitative study of this or a related topic in the past, which allows us to draw broader conclusions.¹

Explaining rights

The right not to be discriminated against in employment is fundamental to the ability to sustain oneself and to support a family as much as it is critical for the realization of one’s dreams and aspirations. While not always analyzed in a quantitative, systematic fashion, discrimination has been studied in a variety of contexts (e.g., for marriage, see Merin, 2002; Rayside, 2008; Waaldijk, 2000) and we focus on discrimination in employment (Wintemute, 1995: 13). It is important to recognize at the outset that legal recognition may not always translate into equality on the ground (Epp, 1996).² Still, we argue, the legal codification of rights is often meaningful and worth focusing on (Graupner and Tahmindjis, 2005; Wintemute, 1995). While a court decision or a legal measure prohibiting discrimination may fail to translate into full equality for the minority they aim to protect, having the right etched into law has a declaratory value, is educational, and provides members of the minority group with venues to claim redress (McCann, 1994; Sanders, 1996; Scheingold, 2004).

What might be intuitively considered the predictors of guarantees of rights may in fact be only somewhat accurate. Indeed, assumptions about which countries would be on the cutting edge of employment rights for sexual minorities are often incorrect. It might be expected that the Netherlands, a country that has a history of political inclusion, would be one of the first countries, in 1992, to prohibit employment discrimination based on sexual orientation, but so did Israel in the same year – a country with less stellar human rights records.³ On the other hand, a country like the US that has been at the forefront of laws against discrimination since the 1960s is yet to enforce laws protecting sexual minorities in employment.⁴ To further examine the systematic influences on the guarantees of rights, we now turn to discuss the two groups of sources of rights – domestic and global – and their causal mechanisms.

Domestic forces

The first group of predictors of rights relates to the level of domestic politics. In the theoretical framework developed here, the key predictor in this group is a spillover effect due to past political inclusion of minority groups. In addition, we examine the effects of religion and culture as systematic predictors.

Political inclusion and domestic norm spillover

We argue that the inclusion of other previously excluded groups has a spillover effect, wherein existing legal prohibitions on discrimination are extended to additional groups (gay individuals in our case). In this study, we focus on the spillover effect created by the inclusion of women.

As more women are elected members of parliament, the passage of laws taking women’s concerns into account increases (Meyer, 2003; Swers, 2001, 2002; Vega and Firestone, 1995). This is particularly true when the presence of women legislators is substantial (Berkman and O’Connor, 1993; Grey, 2002; Thomas, 1991) and has implications for other branches of government as well (Deitch, 1993; Manfredi, 2004).

Key to our argument is the notion that when women are included as active participants in the political process, they can serve as a reference group to other types of minorities. Political inclusion and legal equality for women would spur sexual minorities to become more active, mobilize, and eventually form a support structure of organized political activity and advocacy. Such a support structure is a key condition for the legal protection of the rights of those minority groups (Baird, 2004, 2007; Epp, 1996, 1998). Below we contend that a ‘norm cascade’ (Finnemore and Sikkink, 1998) should apply to the protection of the LGBT community. Yet we believe that part of this argument – that as norms spread, the norm will diffuse more quickly – is applicable domestically as well. As more groups are included, the likelihood of inclusion generally increases.

Elements of the very same support structure that had been conducive to women’s rights may in fact be instrumental in the context of LGBT rights. If we take the US case as an example, certain groups such as the American Civil Liberties Union (ACLU), which had been pivotal in the fight for women’s equality, also became heavily involved in efforts to guarantee LGBT rights (see, for instance, the ACLU involvement in the decision of the Supreme Court in Romer v. Evans⁵ as a friend of the court). In addition to mobilization of other minority groups, the inclusion of women is likely to either be a part of a general trend in the direction of inclusion of underrepresented minorities or facilitate such a trend. Indeed, women’s inclusion is an exceptionally good proxy for the political inclusion of the other in the context of gender and sex (Robinson and Spivey, 2007; Seidman, 2009). In sum, the support structure for the inclusion of women and the trend towards inclusion more generally are likely to serve as the infrastructure for the political inclusion of sexual minorities. Hence, the inclusion of women would have a spillover effect and would, thus, be a predictor of guarantees of rights for LGBT individuals, including equality in employment:

H1: Higher levels of political inclusion of women increase the likelihood of a prohibition on employment discrimination against gay individuals being on the books.

The effects of culture and religion

Beyond the effect of political inclusion spillover from women to sexual minorities, we identify culture and religion as predictors as well. The legal guarantee of equality may be rooted in certain cultural traditions; one example is European, or Western, cultures. Religiosity may be another influence on equality generally; this may be true in particular in the context of the rights afforded to sexual minorities.

A great deal of research has focused on the impact of culture on democracy and rights. The culture most often referred to as the birthplace of democracy and human rights is the West (see, among others, Huntington, 1996). Human rights are related to ‘a particular historical phase in the evolution in the West of the modern state system’ (Galtung, 2000: 11). Indeed, Huntington (1996) argues that it is only the strength of the West that spreads human rights. If we accept the argument that Europe was the birthplace of the modern conception of human rights, then it should make sense that the protection of previously excluded and discriminated groups would be more likely in Western countries (Jacobsen and Bruun, 2000). Donnelly (1982: 303) goes as far as to argue that ‘most non-Western cultural and political traditions lack not only the practice of human rights but the very concept. As a matter of historical fact, the concept of human rights is an artifact of modern Western civilization.’

This, nevertheless, is only one side of a hotly contested debate (Sen, 1998, 1999). As Lauren (1998: 5) points out, world religions from around the globe teach the worth of people. From an empirical perspective, however, Ramirez et al. (1997) find support for the conclusion that membership in the West is important for political rights. In sum, it is not clear from the extant literature whether European or Western cultures should be predisposed to guarantee the rights of sexual minorities. However, there is a good deal of literature suggesting that this may be the case. To account for this possible effect, we hypothesize that:

H2: Being a European (or Western) country increases the likelihood of having a prohibition on employment discrimination against gays.

Since the sample includes all the countries in the world, the analyses proposed here offer a stringent test of H2; the US is one example of a Western country that is a laggard when it comes to the rights of sexual minorities (Barclay et al., 2009). As all countries are included in the analyses, it becomes more difficult to reject the null hypothesis that being a Western country does not increase the likelihood of legal guarantees against discrimination.

While Western culture may increase the likelihood of equality and rights for minority groups, certain variables work in the opposite direction. This is true in particular for the effects of religiosity on policies related to minority groups. Frank and McEneaney (1999) do not find effects of the nation’s religiosity on policies pertaining to sexual minorities. However, the research question in Frank and McEneaney (1999) concerns decriminalization, rather than equality under law. According to Wald et al. (1996), due to ‘communal protest’ against social practices that threaten tradition, groups will react to potential changes in the political status quo. In the context of rights for LGBT individuals, this is true in particular for religious groups (Wald et al., 1996: 1161). Indeed, Wald et al. (1996) examine the effect of Protestant communities and find a significant effect. Their study, however, focuses on communities in the US, while ours is done within a comparative framework. Thus, examination of the effects of additional religious groups is in place.

While its origins are Judeo-Christian, the proscription on sexual relations between people of the same sex is found in other religions as well. Islam condemns same-sex intercourse. Not all nations with a Muslim majority treat same-sex-related activity as a crime; for example, no prohibitions against same-sex intercourse exist in certain nations with a large Muslim constituency because they are relatively secular in nature (Indonesia), are multi-religious, or because tolerance of same-sex-related activity has been entrenched in the system for years (Turkey). Yet we do expect that there is considerable pressure on public officials not to advance gay rights in nations with a strong Muslim constituency. Akin to Islam, Catholic dogma also condemns sexual relations between members of the same sex as sinful. Therefore, we predict that the higher the percentage of Catholics in the population, the greater the likelihood that the country will fail to prohibit discrimination in employment against gay individuals. We examine the effect of the dominance of religions with Judeo-Christian origins in countries worldwide, and expect the likelihood of discrimination to increase with the size of the religious group:

H3: States with a greater share of Catholics, Muslims, or Protestants are less likely to prohibit employment discrimination against gay individuals.

Global influences

In the theoretical framework developed here, we recognize that while politically consequential, domestic forces could not fully account for the guarantees of rights. As global influences permeate numerous aspects of political life (Tsutsui and Wotipka, 2004), the realm of rights and equality is no exception. To complement the domestic influences in the theoretical framework developed here, therefore, we also include global forces; supranational bodies and broader processes of globalization are likely to influence rights at the domestic level. This is true in particular of agencies or covenants of the United Nations pertaining to rights in general (human rights covenants) and more specifically to employment (International Labor Organization). Ceteris paribus, we expect regulation by international bodies and the broader processes of globalization to influence minority rights.

Human rights covenants

Adopted by the United Nations General Assembly on 16 December 1966, the International Covenant on Civil and Political Rights (ICCPR) came into force on 23 March 1976. Along with the International Covenant on Economic, Social and Cultural Rights (ICESCR), which in almost all cases was ratified jointly with the ICCPR by the different countries, the ICCPR sets standards for a range of rights and liberties, including for labor rights and for issues of discrimination.

The relations, in the realm of rights, between the ratification of the covenants and national politics are well studied. In fact, some understandings of gay rights are as human rights (Beger, 2004: ch. 4; Graupner and Tahmindjis, 2005). For instance, according to Waaldijk (2000), processes of decriminalization in Europe were first a function of gay rights becoming a civil rights issue. Later on, this turned into a human rights issue. The implications of the ICCPR for rights at the domestic level was also demonstrated in the decision handed down by the Human Rights Committee of the United Nations in Toonen v. Australia.⁶ The ICCPR was used to find Tasmanian laws prohibiting same-sex sex between men in violation of the privacy provision in Article 17.

We code for date of ratification by the country of the international human rights covenants, which include the ICCPR and the ICESCR (Cole, 2005). As international law has been shown to influence national politics in this sphere, the theoretical framework developed here suggests that ratification would increase the likelihood of inclusion of rights in the country’s legal code; ratification of the ICCPR and the ICESCR will make it more likely that the country prohibits discriminatory practices against gay individuals in employment:

H4: Ratifying the international human rights covenants increases the likelihood of legal prohibitions on employment discrimination against gay individuals.

International Labor Organization

While we control for the covenants pertaining to human rights, which may affect rights generally, some agencies of the United Nations, and specifically the International Labor Organization (ILO), are concerned with issues of labor and employment directly. Covering a wide range of issues, from discrimination in employment to child labor, and from work safety to forced labor, the ILO seeks to set an international labor code. With the adoption of conventions and recommendations by the International Labor Conference, and the ratification by the national governments of the member states, the ILO sets international labor standards; ratifying a convention amounts to an obligation by the state to apply it in law. According to its website (ilo.org), almost all states are members of the ILO.

The ILO has adopted 188 conventions to date.⁷ Of particular interest to our project for both its timing and its content is Discrimination (Employment and Occupation) Convention, 1958 (C111). The Convention defines discrimination, states the obligation of each of the member states to prevent discrimination using their legal codes, and sets a comprehensive and high standard for the application of the principles put forth. We expect the ratification of the Convention to increase the likelihood of antidiscriminatory legal measures in employment regulation:

H5: Ratifying the Discrimination (Employment and Occupation) Convention, 1958 (C111) increases the likelihood of legal prohibitions on employment discrimination against gay individuals.

The European Union

Using regulations, directives and decisions of its governing bodies, the European Union (EU) is a supranational entity with the power to influence legal and economic conditions within its Member States. From political rights to inflation and from national deficits to unemployment rates, the EU dictates a considerable amount of the internal affairs of its members. EU legislation provides for minimum requirements (with Member States having the ability to provide higher levels of protection in their laws) for protection against discrimination in vocational training, membership in employer and employee organizations, and any other type of discrimination in employment and occupation. The EU has put a general emphasis on the human rights of sexual minorities, for example, in pushing for the rights of such minorities by strong-arming Romania to lift its sodomy prohibition (Somner et al., 2013).⁸ Furthermore, the institutions of the EU have been historically committed to human rights and equality (Beger, 2004: 130).⁹

While the EU is not considered by all to be a good address to claim rights (Elman, 2001), membership in the EU itself is likely to increase the likelihood that a country would guarantee the rights of members of sexual minority groups not to be discriminated against in employment. Anticipation of future regulation of that sort by the EU may lead a country to prohibit discrimination against LGBT individuals. Alternatively, related EU regulation in the past (e.g. a requirement to repeal sodomy laws) put in place the necessary political, legal, and institutional infrastructure, which is likely also to lead Member States to pass legislation prohibiting discrimination. As the correlation table in the Appendix indicates, the correlation between the EU predictor and the Western culture predictor is .51, which is unlikely to entail problems in model estimation:

H6: Membership in the EU (or the EC) increases the likelihood of legal prohibitions on employment discrimination against gays being a part of the legal code.

Cross-country norm cascade – political globalization

Beyond the effects of international organizations and laws, more general processes of globalization may also influence the likelihood of the legal entrenchment of rights. Tsutsui and Wotipka (2004) discuss a ‘norm cascade’ between countries. Finnemore and Sikkink (1998) posit that norm entrepreneurs push norms domestically and then internationally until the number of countries that have accepted these norms reaches a tipping point. Once this tipping point has been reached, the new norm becomes expected behavior. Indeed, in their empirical analysis of women’s suffrage, Ramirez et al. (1997) see such a pattern. Along the same lines, we argue that in the case of employment discrimination, there is a norm diffusion that happens not only between minority groups(i.e. domestically from women to sexual minorities), but also between countries.

In the post-Second World War period, the embrace of human rights becomes the right, proper, and legitimate thing for the state to do. In a globalized nation, political entrepreneurs, public opinion (Lax and Phillips, 2009), political organizations, and social movements (Barclay et al., 2009; Bernstein, 2002; Patternote and Kollman, 2010) are able to recognize alternative legal arrangements within which to settle gay rights. This in turn would increase the likelihood of a legal prohibition on discrimination in employment based on sexual orientation. A more globalized political sphere is more likely to set such processes in motion. With more globalization, public opinion, political entrepreneurs, and social movements are more likely to push in that direction. And thus, with higher levels of globalization, the likelihood is greater for a norm cascade influencing domestic politics to incorporate rights in the legal code. The logic of global norm spillover, as presented here, draws simultaneously from constructivist and rationalist understandings of international learning, which are generally thought to be different. Yet, in line with recent scholarship which argues that there is room for synthesis across these different approaches to international learning, we combine the two. In sum, we expect a greater likelihood of prohibitions on discrimination in states that are more politically globalized and are thus more likely to be part of the cross-country norm cascade:

H7: More politically globalized states are also more likely to prohibit employment discrimination against gay individuals.

We include as control variables in our model the democratic conditions in the nation and modernization. ‘The distribution of political and economic rights depends largely on characteristics of the political regime. A democratic regime will normally accord greater political and economic rights to the people’ (Olzak and Kiyoteru, 1998: 697). Political inclusion is more likely to take place in democracies, and thus democratic conditions facilitate legal equality (Wilensky, 2002). We argue that the same kind of encouragement of inclusion that has been observed for ethnic and class issues in democracies should be true as well for sexual minorities. Hence, laws that prevent discrimination in employment based on sexual orientation are more likely when the polity is more democratic. Studying gay rights at the local community level within the US, Wald et al. (1996) control for political opportunity structure. Wald et al. (1996) measure political opportunity structure, such as civil rights for minority groups,¹⁰ and identify it as an important predictor. The national-level equivalent to political opportunity structure at the local level is democratic conditions; as our unit of analysis is the country, the measure of democratic conditions accounts for the extent to which the political climate allows for new rights claims. The presence of democratic political institutions, for instance, is crucial for the success of campaigns in favor of passing antidiscrimination measures (Tarrow, 1988, 1991). We thus control for democratic conditions in the country.

Lastly, an additional control variable is modernization. In a variety of ways, such as through literacy, education, and cultural change, modernization can change the view of who should be accepted in society and who should be protected by the state. Modernization has been shown to act as a causal variable in increasing levels of democracy generally (Doorenspleet, 2004; Huntington, 1991; Lipset, 1960; Przeworski and Limongi, 1997; Ramirez et al., 1997) and specifically enhancing the rights of women (Inglehart and Norris, 2003; Steel et al., 1992) and ethnic minorities (Gurr, 2000). More closely related to the effects of modernization on issues of sexuality in the society, Lienemann (1998: 8) argues that with modernization comes ‘removal of sexuality from the realm of taboo and a new respect for the private sphere.’ Inglehart (1997) explicitly argues that a shift to post-modern values puts the rights of women and sexual minorities in the center of political conflict. We therefore expect modernization to increase the likelihood of legal measures against discrimination.

Data and methods

The dependent variable is coded based on the May 2009 report of the International Lesbian, Gay, Bisexual and Intersex Association (Ottoson, 2009; see also Teorell et al., 2009). The data are time-series cross-sectional and the dependent variable, Legal Prohibition, is equal to 1 if the state prohibited discrimination against gay individuals in the workplace that year, and 0 otherwise. The coding of the dependent variable is based on the standards set by Teorell et al. (2009). The coding scheme considers cases where prohibitions consist of specific employment non-discrimination rules and cases that consist of broad lesbigay non-discrimination rules. Furthermore, Teorellet al. (2009) also deal with cases where some highly decentralized countries (e.g. the US and Australia) prohibit employment discrimination in some jurisdictions (e.g. states and provinces, respectively) and not others, in which case the coding scheme considers changes that apply to the jurisdiction as a whole. This coding also considers court resolutions and constitutional prescriptions and proscriptions. Since the prohibition on discrimination can be overruled, every year in which it is on the books (due to legislation, judicial ruling or constitutional proscription) is coded 1 for this country.

To measure domestic norm spillover, Political Inclusion is operationalized as the inclusion of women in the political process; as a measure of political inclusion, we examine the percentage of women representatives in the national legislative body. To measure this, we use the Inter-Parliamentary Union ‘Women in national parliaments’ data (Inter-Parliamentary Union, 2005).¹¹ The culture and religion predictors include the religion variables and the Western culture variable. Based on the Central Intelligence Agency (CIA) factbook, we code for Percent Catholic, Percent Protestant, and Percent Muslim. Western Political Culture is a dummy coded 1 for Western countries.¹² We use three measures to examine the effects supranational institutions have on our dependent variable – the European Union, ratification of the ILO’s C111, and ratification of the ICCPR. Membership in the EU was determined according to official records of the EU and its predecessors the European Economic Community and the European Community. Date of ratification of C111 of the International Labor Organization was coded according to data on the ILO’s official website.¹³ Date of ratification of the International Covenant on Civil and Political Rights was coded according to the website for the United Nations Human Rights Treaties.¹⁴ To measure the effect of the international norm cascade, we use a measure of globalization – the KOF Index of Globalization (Dreher, 2006; Dreher et al., 2008). The index for the globalization variables is measured in line with the standard in comparative public law (Frank and McEneaney, 1999; Tsutsui and Wotipka, 2004) and ranges from 0 to 100. Higher values indicate higher levels of globalization. The index of Political Globalization is measured by the number of embassies and high commissions in a country, the number of memberships the country has in international organizations, participation in United Nations peacekeeping missions, and the number of international treaties signed since 1945 (Dreher, 2006; Dreher et al., 2008). As for our control variables, to measure Democratic Conditions, we utilize the POLITY score, which was imputed using Freedom House data where it was missing. The scale ranges from 0 (least democratic) to 10 (most democratic) (Hadenius and Teorell, 2005). Since, in a comparative framework, Modernization is commonly measured as Gross Domestic Product (GDP) per capita (Gleditsch, 2002), we measure this variable the same way here. GDP per capita is measured in constant US dollars at base year 2000 (Gleditsch, 2002).

Our data indicate that the type of prohibitions of interest appeared relatively recently. As mentioned earlier, the first two countries to prohibit employment discrimination on the basis of sexual orientation were the Netherlands and Israel, both in 1992. Since these two countries outlawed this type of discrimination, until the year 2002, 17 additional states entered such a prohibition into their legal codes.¹⁵

For the purposes of the multivariate analysis, we use time-series cross-sectional data, listing all states in the abovementioned data set for which data are available for the years 1972–2002. A total of 161 states with data on average for 23 years per nation make for a total of 3782 observations. To test the hypotheses, we employ a generalized estimating equation (GEE) model (Zorn, 2001). A marginal approach, such as the GEE, is appropriate in this case since we are interested generally in what variables influence whether a prohibition on discrimination in employment is on the books in a certain country in a specific year (Zorn, 2001: 475). Since the dependent variable is dichotomous, and due to the data structure described earlier, we employ a GEE model with first-order autoregressive component and logit as the link function. For the period studied, the coding was 1 in 103 of the 3782 observations, with a mean of .03 and a standard deviation of .16. Given the percentage of events compared with non-events (2.7%) and the size of the sample (3782), an alternative estimation approach, a rare-events logit, is possible. However, a rare-events logit would not make a considerable improvement on the accuracy of the coefficients (King and Zeng, 1999: 15–17) and would not allow us to account for auto-regressive effects.

Results

What factors influence the likelihood of having legal guarantees of rights on the books? And more specifically, what variables affect the likelihood that a country has a prohibition on discrimination in employment based on sexual orientation in its legal code? The overall goodness-of-fit measures of the models in Table 1 is notable. Furthermore, the results lend strong support to most of our key hypotheses.

Table 1.

Generalized estimating equation model: Cross-sectional time-series analyses of predictors of legal prohibitions on employment discrimination against gay individuals (1972–2002).

Variables	Model I (Robust Standard Errors)	Model II (Robust Standard Errors)
Domestic forces
Political inclusion and domestic norm spillover	.047* (.023)	.074** (.03)
Religion – % Catholics	n/a	−.017 (.012)
Religion – % Protestants	n/a	−.03** (.01)
Religion – % Muslims	n/a	−.06 (.05)
Western political culture	n/a	−1.7 (1.3)
Global effects
The European Union	.62 (.57)	.74 (.68)
International Labor Organization	2.2* (1.3)	4.1** (1.8)
International Covenant on Civil and Political Rights	2.4*** (.84)	2.7** (1.1)
Political globalization	.004 (.02)	.02 (.02)
Control variables
Democratic conditions	.25* (.12)	.33* (.18)
Modernization	.00009* (.00004)	.00014* (.00007)
Constant	−12.3*** (1.8)	−15.07*** (2.5)
	N = 3782Number of groups = 163Observations per group – average 23.2Wald χ² (7) = 119.17Prob > χ² = .0	N = 3763Number of groups = 161Observations per group – average 23.4Wald χ² (11) = 70.6Prob > χ² = .00

Notes: Due to data availability, the following countries were not included in the analyses: Andorra, Antigua and Barbuda, Brunei Darussalam, Dominica, Grenada, Kiribati, Libya, Liechtenstein, Marshall Islands, Micronesia, Monaco, Nauru, Oman, Palau, Qatar, Samoa, San Marino, Sao Tome and Principe, Seychelles, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Taiwan, Tibet, Timor-Leste, Tonga, Tuvalu, Vanuatu, Vietnam, Yemen, Yugoslavia, and Zanzibar.

***

p < .001; **p < .01; *p < .05; #p < .1; one-tailed tests where directionality hypothesized.

Both domestic and global components of the theoretical framework proposed here are substantiated in the analyses. Model 1 in Table 1 examines the effects of domestic spillover, supranational institutions, global norm cascade, democracy, and modernization on equality under law. The specification of Model 2 in that table adds the cultural and religious variables as well. As expected, the effect of domestic norm spillover is positive and significant. As the percentage of women representatives in the legislature increases (i.e. the more politically included women are), the more likely is a legal prohibition on discrimination. Two of the three supranational institutions studied here have a positive and statistically significant effect on rights at the state level. Ratification of the ILO’s C111 and of the ICCPR independently increase the likelihood of equality of the type studied here. These effects are either significant or highly significant and with coefficients of substantial magnitude. Conversely, the hypothesis concerning a broader effect for globalization in the form of a global norm cascade fails to win support in the data. It is via international institutions and international law (the ILO and the human rights covenants) that global influences lead to guarantees of rights in national legislation.

As for our control variables, the effect of democratic conditions is positive and within standard levels of statistical significance. Unsurprisingly, as the polity becomes more democratic, the likelihood of rights for sexual minorities increases. Furthermore, as GDP per capita grows, the likelihood of legal measures against discrimination increases. When adding the culture and religion variables in Model 2, the support for the hypotheses discussed so far remains unchanged. Of the variables in this last category, however, only the percentage of Protestants in the polity is significant and in the anticipated direction (which is along the lines of Wald et al., 1996).

In sum, we find support for the effects of a subset of the variables in each of the two groups of predictors. In the group of domestic forces, norm spillover is significant and in the anticipated direction, while only one of the culture and religion variables has a significant effect. In the group of global forces, two of the three supranational bodies studied systematically affect the likelihood of equality. As far as our research question is concerned, degree of globalization, however, is inconsequential.

Supporting analyses

As supporting analysis and since the first prohibitions do not appear until 1992, we present in Table 2 an identical analysis but for a shorter period of time. The results in Table 2 present the equivalent analyses to those in Table 1 for the period of two decades following 1982. While the sizes of the coefficients are different (largely due to the limited period of time analyzed – two rather than three decades), the findings are substantively indistinguishable. In sum, in addition to the duration of time studied, the findings are also robust to model specification and alternative measurements.

Table 2.

Generalized estimating equation model: Cross-sectional time-series analysis of predictors of legal prohibitions on employment discrimination against gay individuals (1982–2002).

Variables	Model (Robust Standard Errors)
Domestic forces
Political inclusion and domestic norm spillover	.063* (.03)
Religion – % Catholics	−.017 (.012)
Religion – % Protestants	−.027* (.014)
Religion – % Muslims	−.06 (.05)
Western political culture	−1.5 (1.2)
Global effects
The European Union	.7 (.69)
International Labor Organization	4.0** (1.7)
International Covenant on Civil and Political Rights	2.5* (1.2)
Political globalization	.023 (.017)
Control variables
Democratic conditions	.33# (.2)
Modernization	.00013* (.00007)
Constant	−14.58*** (2.54)
	N = 2739Number of groups = 161Observations per group – average 17Wald χ² (11) = 62.97Prob > χ² = .0

***

p < .001; **p < .01; *p < .05; #p < .1; one-tailed tests where directionality hypothesized.

Discussion and conclusions

Two stories are often told in the context of rights guarantees: a transnational-international account and a domestic one. Told separately, however, neither is complete. In the case we explore here, it is clear that a focus on one or the other would be myopic. As this project indicates, the best way to understand why some countries outlaw discrimination on the basis of sexual orientation is by taking into consideration both the international and the domestic perspectives. At the level of domestic politics, we present evidence that earlier widening of the circle of power to a previously excluded group may start a cascade leading to protection of other groups. Indeed, a crucially important domestic force is the spillover of norms from the inclusion of one minority group to another. When women are politically included, they serve as a reference group to other minorities that in turn mobilize and eventually form a support structure, which is a key condition to win legal protections (Epp, 1996, 1998). Elements in the support structure for the equality of certain minorities in the past (e.g. interest groups involved in the campaign for women’s equality) may be instrumental in the fight for equality of additional minorities (e.g. sexual minorities). This analysis suggests a need for a closer examination of how the norm cascade works, and not just at the transnational level.

On the global level, our findings underscore the consequentiality of supranational institutions. The ILO and international law have a robust and substantively meaningful effect as predictors of rights guarantees. Ratification of the ILO’s C111 or the ICCPR increases the likelihood of having guarantees of rights on the books.

While some accounts in the literature claim that, in Europe for instance, the mere decriminalization of same-sex sex inexorably leads to the introduction of anti-discrimination measures (e.g. Waaldijk, 2000), the theory proposed here offers an alternative analysis that is richer and more complex. Considering domestic influences side by side with global effects provides for a useful theoretical framework for the study of the second stage of legal guarantees for gay individuals in Sanders (1996), namely, guarantees against discrimination. In addition, the results move forward some of the key debates in the literature on rights and equality. For instance, Frank and McEneaney (1999) do not find the effect of religion on policy outcome found in Wald et al. (1996). The reason for this discrepancy may lie with the difference in the research question. Frank and McEneaney are interested in the liberalization of policies limiting the ability to engage in same-sex sexual activities. Wald et al.’s question pertains to anti-discrimination measures. The latter find an effect for religiosity. In sum, the effect of religious constituencies is closely related to the policy question studied (Sanders, 1996).

The theoretical framework proposed here may have wider validity and may teach us about rights more broadly; in a world where international law and international organizations become more complex and where the effects of such institutions permeate almost every aspect of political life, those effects may not be unique to the rights of sexual minorities or to the issue of discrimination in employment. On the other hand, in support of the many case studies on this topic, this study also points to the importance of religion in sanctioning LGBT rights. What is more, while the inclusion of women was theoretically linked in this study to equality of sexual minorities, studies of equality for other groups may need to focus on other types of norm spillover. It may well be that the inclusion of an ethnic group or a racial minority (rather than women) leads to different types of spillover effects.

Lastly, the research presented here not only contributes to our understanding of rights, but may also have policy implications; as gay individuals in the majority of countries are yet to win legal equality, the findings in this study may have implications for the policies countries decide to endorse and for the strategic choices made by social movements planning their future moves. In our future efforts, we plan to extend our investigation of rights spillover to other types of rights as well as to other types of minority groups.

Footnotes

Appendix

Table A2.

Descriptive statistics.

	Observations	Mean	SD	Min.	Max.
Women in parliament	3943	9.9	8.8	0	42.7
Democracy	6173	5.5	3.4	0	10
GDP	5625	7711.3	8330.7	170.55	84408.23
Political globalization	6245	49.2	24.5	1.55	98.78
Membership EU	7380	.05	.23	0	1
European nation	7380	.22	.41	0	1
% Catholic	6732	31.13	35.58	0	99.1
% Protestant	6624	13.19	21.2	0	97.8
% Muslim	6732	23.2	35.68	0	99.9
ILO	7380	.54	.49	0	1
ICCPR	7380	.47	.49	0	1

Funding

Udi Sommer’s Marie Curie grant was used to support this research.

Notes

Author biographies

Udi Sommer is Lecturer in the Department of Political Science in Tel Aviv University, Israel (PhD, Stony Brook University, 2007; MA, Tel Aviv University, 2002; BA, The Hebrew University, 1999). Sommer’s broad research interests are in comparative public law and constitutionalism, American politics, and political methodology. He specializes in the analysis of political institutions. He is a recipient of several grants and fellowships including the Fulbright Doctoral Fellowship, Marie Curie and a grant from the American National Science Foundation.

Victor Asal is Director of the Center for Policy Research and an Associate Professor in the Department of Political Science at the University at Albany, State University of New York, USA. He is the director of the Homeland Security Certificate and MPA Concentration. He received his PhD in Government and Politics from the University of Maryland, College Park. He is also the co-director of the Project on Violent Conflict. Dr Asal is affiliated with the National Consortium for the Study of Terrorism and Responses to Terrorism (START), a Department of Homeland Security Center of Excellence.

References

Adam

Duyvendak

Krouwel

(1999) The Global Emergence of Gay and Lesbian Politics: National Imprints of a Worldwide Movement. Philadelphia, PA: Temple University Press.

Baird

(2004) The effect of politically salient decisions on the U.S. Supreme Court’s agenda. The Journal of Politics 66: 755–772.

Baird

(2007) Answering the Call of the Court: How Justices and Litigants Set the Supreme Court’s Agenda. Charlottesville, VA: University of Virginia Press.

Barclay

Bernstein

Marshall

A-M

(2009) Queer Mobilizations: LGBT Activists Confront the Law. New York, NY: New York University Press.

Beger

(2004) Tensions in the Struggle for Sexual Minority Rights in Europe: Que(e)rying Political Practices. Manchester: Manchester University Press.

Berkman

O’Connor

(1993) Do women legislators matter? American Politics Research 21(1): 102–124.

Bernstein

(2002) Identities and politics: Toward a historical understanding of the lesbian and gay movement. Social Science History 26(3): 531.

Cole

(2005) Sovereignty relinquished? Explaining commitment to the international human rights covenants, 1966–1999. American Sociological Review 70: 472–495.

Deitch

(1993) Gender, race, and class politics and the inclusion of women in Title VII of the 1964 Civil Rights Act. Gender and Society 7(2): 183–203.

10.

Donnelly

(1982) Human rights and human dignity: An analytic critique of non-Western conceptions of human rights. American Political Science Review 76(2): 303–316.

11.

Doorenspleet

(2004) The structural context of recent transitions to democracy. European Journal of Political Research 43(May): 309–335.

12.

Dreher

(2006) Does globalization affect growth? Evidence from a new index of globalization. Applied Economics 38(10): 1091–1110.

13.

Dreher

Gaston

Martens

(2008) Measuring Globalization: Gauging its Consequences. New York, NY: Springer.

14.

Elman

(2001) Testing the Limits of European Citizenship: Ethnic Hatred and Male Violence. National Women’s Studies Association Journal 13(3): 49–69.

15.

Epp

(1996) Do bills of rights matter? The Canadian Charter of Rights and Freedoms. American Political Science Review 90(4): 765–779.

16.

Epp

(1998) The Rights Revolution: Lawyers, Activists and Supreme Courts in Comparative Perspective. Chicago, IL: The University of Chicago Press.

17.

Finnemore

Sikkink

(1998) International norm dynamics and political change. International Organization 52: 887–917.

18.

Frank

McEneaney

(1999) The individualization of society and the liberalization of state policies on same-sex sexual relations, 1984–1995. Social Forces 77(3): 911–944.

19.

Frank

Camp

Boutcher

(2010) Worldwide trends in the criminal regulation of sex, 1945 to 2005. American Sociological Review 75(6): 867–893.

20.

Galtung

(2000) 9 alternative models for global democracy. In: Holden

(ed.) Global Democracies: Key Debates. London: Routledge.

21.

Gibson

(2004) Truth, reconciliation, and the creation of a human rights culture in South Africa. Law and Society Review 38: 5–40.

22.

Gleditsch

(2002) Expanded trade and GDP data. Journal of Conflict Resolution 46: 712–724.

23.

Gloppen

(2005) Social rights litigation as transformation: South African perspectives. In: Jones

Stokke

(eds) Democratising Development: The Politics of Socio-economic Rights in South Africa. Leiden: Martinus Nijhoff Publishers.

24.

Graupner

Tahmindjis

(2005) Sexuality and Human Rights. New York: Harrington Park Press.

25.

Greenberg

Katz

Oliviero

. (eds) (1993) Constitutionalism and Democracy: Transitions in the Contemporary World. New York, NY: Oxford University Press.

26.

Grey

(2002) Does size matter? Critical mass and New Zealand’s women MPs. Parliamentary Affairs 55(1): 19–29.

27.

Gurr

(2000) People vs. States. Washington, DC: United States Institute of Peace.

28.

Hadenius

Teorell

(2005) Assessing Alternative Indices of Democracy, C&M Working Papers 6, IPSA, August 2005. Available at: http://www.concepts-methods.org/Files/WorkingPaper/PC%206%20Hadenius%20Teorell.pdf (accessed 23 July 2013).

29.

Haynie

(2003) Judging in Black and White: Decision Making in the South African Appellate Division, 1950–1990. New York, NY: Peter-Lang Publishing.

30.

Huntington

(1991) The Third Wave: Democratization in the Late Twentieth Century. Norman, OK: University of Oklahoma Press.

31.

Huntington

(1996) The Clash of Civilizations: Remaking the World Order. New York, NY: Simon and Schuster.

32.

Inglehart

(1997) Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 Societies. Princeton, NJ: Princeton University Press.

33.

Inglehart

Norris

(2003) Rising Tide: Gender Equality and Cultural Change around the World. Cambridge: Cambridge University Press.

34.

Inter-Parliamentary Union (2005) Women in national parliaments. Available at: http://www.ipu.org/

35.

Jacobsen

Bruun

(2000) Human Rights and Asian Values: Contesting National Identities and Cultural Representations in Asia. London: Curzon Press.

36.

King

Zeng

(1999) Logistic regression in rare events data. Department of Government, Harvard University. Available at: http://GKing.Harvard.Edu

37.

Lauren

(1998) The Evolution of International Human Rights. Oxford: Clarendon Press.

38.

Lax

Phillips

(2009) Gay rights in the States: Public opinion and policy responsiveness. American Political Science Review 103(3): 367–386.

39.

Lienemann

(1998) Churches and Homosexuality: An Overview of Recent Official Church Statements on Sexual Orientation. The Ecumenical Review 50(1): 7–22.

40.

Lipset

(1960) Political Man. New York, NY: Doubleday.

41.

Manfredi

(2004) Feminist Activism in the Supreme Court. Vancouver: University of British Columbia Press.

42.

McCann

(1994) Rights at Work: Pay Equity Reform and the Politics of Legal Mobilization. Chicago, IL: University of Chicago Press.

43.

Merin

(2002) Equality for Same-Sex Couples: The Legal Recognition of Gay Partnerships in Europe and the United States. Chicago, IL: University of Chicago Press.

44.

Meyer

(2003) Much ado about nothing? Political representation policies and the influence of women parliamentarians in Germany. Review of Policy Research 20(3): 401–421.

45.

Moustafa

(2003) Law versus the state: The judicialization of politics in Egypt. Law & Social Inquiry 28(4): 883–930.

46.

Olzak

Kiyoteru

(1998) Status in the world system and ethnic mobilization. Journal of Conflict Resolution 42(6): 691–720.

47.

Ottoson

(2009) State-sponsored homophobia: A world survey on laws prohibiting same sex activity between consenting adults. International Lesbian, Gay, Bisexual and Intersex Association. Available at: http://www.ilga.org/statehomophobia/ILGA_State_Sponsored_Homophobia_2008.pdf (accessed 6 August 2008).

48.

Patternote

Kollman

(2010) Explaining convergence: Same-sex unions activism and policy. Unpublished paper.

49.

Przeworski

Limongi

(1997) Modernization: Theories and facts. World Politics 49: 155–183.

50.

Ramirez

Soysal

Shanahan

(1997) The changing logic of political citizenship: Cross-national acquisition of women’s suffrage rights. American Sociological Review 62(5): 735–745.

51.

Rayside

(2008) Queer Inclusions, Continental Divisions: Public Recognition of Sexual Diversity in Canada and the United States. Toronto: University of Toronto Press.

52.

Robinson

Spivey

(2007) The politics of masculinity and the ex-gay movement. Gender & Society 21(5): 650–675.

53.

Sanders

(1996) Getting lesbian and gay issues on the international human rights agenda. Human Rights Quarterly 18: 67–106.

54.

Sartori

(1994) Comparative Constitutional Engineering: An Inquiry into Structures, Incentives and Outcomes. New York, NY: New York University Press.

55.

Sathe

(2004) Judicial Activism in India: Transgressing Borders and Enforcing Limits. New York, NY: Oxford University Press.

56.

Scheingold

(2004) The Politics of Rights: Lawyers, Public Policy, and Political Change. Ann Arbor, MI: University of Michigan Press.

57.

Schmid

(2000) Decriminalization of Sodomy under South Africa’s 1996 Constitution: Implications for South African and US law. Cardozo Journal of International and Comparative Law 8: 163.

58.

Seidman

(2009) Critique of compulsory heterosexuality. Sexuality Research and Social Policy: Journal of NSRC 6(1): 18–28.

59.

Sen

(1998) Universal truths: Human rights and the Westernizing illusion. Harvard International Review 20: 40–43.

60.

Sen

(1999) Democracy as a universal value. Journal of Democracy 10: 3–17.

61.

Sommer

Asal

Zuber

Parent

(2013) Institutional Paths to Policy Change: Judicial versus Nonjudicial Repeal of Sodomy Laws. Law and Society Review 47(1): 409–439.

62.

Songer

Johnson

(2007) Judicial decision making in the Supreme Court of Canada: Updating the personal attribute model. Canadian Journal of Political Science 40: 911–934.

63.

Steel

Warner

Stieber

. (1992) Postmaterialist values and support for feminism among Canadian and American women and men. The Western Political Quarterly 45(2): 339–353.

64.

Stone Sweet

(2000) Governing with Judges: Constitutional Politics in Europe. New York, NY: Oxford University Press.

65.

Swers

(2001) Understanding the policy impact of electing women: Evidence from research on Congress and state legislatures. PS: Political Science and Politics 34(2): 217–220.

66.

Swers

(2002) The Difference Women Make: The Policy Impact of Women in Congress. Chicago, IL: University of Chicago Press.

67.

Tarrow

(1988) National politics and collective action: Recent theory and research in Western Europe and the United States. In: Scott

Blake

(eds) Annual Review of Sociology (vol. 14). Palo Alto, CA: Annual Reviews.

68.

Tarrow

(1991) Comparing social movement participation in Western Europe and the United States: Problems, uses, and a proposal for synthesis. In: Rucht

(ed.) Research on Social Movements. Frankfurt am Main: Campus Verlag.

69.

Teorell

Charron

Samanni

. (2009) The Quality of Government Dataset. Version 17 June 09. University of Gothenburg: The Quality of Government Institute. Available at: http://www.qog.pol.gu.se

70.

Thomas

(1991) The impact of women on state legislative policies. The Journal of Politics 53(4): 958–976.

71.

Tsutsui

Wotipka

(2004) Global civil society and the international human rights movement: Citizen participation in human rights international nongovernmental organizations. Social Forces 83(2): 587–620.

72.

Vega

Firestone

(1995) The effects of gender on congressional behavior and the substantive representation of women. Legislative Studies Quarterly 20(2): 213–222.

73.

Waaldijk

(2000) Civil developments: Patterns of reform in the legal position of same-sex partners in Europe. Canadian Journal of Family Law 17(1): 62–88.

74.

Wald

Button

Rienzo

(1996) The politics of gay rights in American communities: Explaining antidiscrimination ordinances and policies. American Journal of Political Science 40(4): 1152–1178.

75.

Widner

(2001) Building the Rule of Law: Francis Nyalali and the Road to Judicial Independence in Africa. New York: W.W. Norton.

76.

Wilensky

(2002) Rich Democracies: Political Economy, Public Policy, and Performance. Berkeley, CA: University of California Press.

77.

Wintemute

(1995) Sexual Orientation and Human Rights: The United States Constitution, the European Convention, and the Canadian Charter. New York, NY: Oxford University Press.

78.

Zorn

CJW

(2001) Generalized estimating equation models for correlated data: A review with applications. American Journal of Political Science 45(2): 470–490.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB

0.08 MB