Persistent evidence of employment discrimination against lesbian, gay, bisexual, and transgender (LGBT) people has prompted policymakers in some states and local jurisdictions to outlaw such discrimination. Twenty-one states and the District of Columbia currently ban sexual orientation and gender identity (SOGI) discrimination in employment, and one additional state bans only sexual orientation discrimination. Half of the LGBT population lives in states that do not have any laws prohibiting discrimination in private employment.¹ In those remaining states, 24 have one or more cities and counties that offer some form of protection.

Progress at the federal level has been slower, however, and there is no explicit ban on SOGI discrimination in federal statutes. President Obama signed an Executive Order in 2014 that requires companies that do business with the federal government to have policies that pledge nondiscrimination on the basis of sexual orientation and gender identity.

More importantly, in 2013 the Equal Employment Opportunity Commission (EEOC) began allowing anyone who believed that they experienced SOGI-based discrimination to file charges of sex discrimination under Title VII of the Civil Rights Act. The EEOC argues that discrimination against LGBT employees is sex discrimination because it is rooted in gender stereotypes and other sex-based considerations.²  Federal court rulings have been split on whether Title VII prohibits SOGI discrimination, however, and two SOGI cases are currently being appealed to the U.S. Supreme Court.³  For now, the EEOC’s inclusion of SOGI discrimination as sex discrimination under Title VII remains, effectively extending federal protection for SOGI to employees in every state. 

In this study, we analyze data on the 9,121 SOGI discrimination charges filed with the EEOC or with a state or local fair employment practices agency (FEPA) between 2013 and 2016. These charges provide an unprecedented view into the situations where employees believe that they have faced discrimination because of their sexual orientation and/or gender identity. We report on trends in charge filing over time and on patterns by industry, issues raised and outcomes of the charges. 

 

Key Findings

Overall, the EEOC rulings and subsequent acceptance of charges appear to have dramatically increased the number of people able to file SOGI-related charges. Almost half of the charges were filed in states that do not have an explicit state SOGI nondiscrimination law, giving thousands of people an opportunity to seek redress that they did not have before 2013. The importance of the EEOC’s acceptance is clear: in the non-SOGI law states, the EEOC received and processed 82% of those charges (and FEPAs received the rest); in states with state SOGI protections, the EEOC received one third of the cases, with the FEPAs receiving two thirds.

The number of SOGI discrimination charges filed increased steadily from 2013 to 2015 and then leveled off in 2016. The increase in the number of charges filed was most rapid for states without SOGI protections, suggesting that people learned over time about their right to file a charge. Charges filed in states with SOGI protections also increased over time, which might mean that the EEOC action also increased awareness in those states.

Some states might have more charges filed than others because the population of LGBT people is larger, meaning more people are at risk of SOGI-based discrimination. Therefore, we divided the number of charges per state by the estimated LGBT population size in that state. These adjusted rates show that states with SOGI protections have 102 charges filed per 100,000 LGBT adults, while states without SOGI laws have only 83 charges filed per 100,000 LGBT adults. The higher charge rate in states with SOGI protection suggests that we might expect even more charges of SOGI discrimination to be filed in the non-SOGI protection states. That is, passage of an explicit state or federal SOGI nondiscrimination law would likely spread the knowledge about the right to challenge discrimination more broadly and result in additional charges being filed in those states lacking state-level protections. In addition to more knowledge, stronger state or federal laws could also reduce fear of retaliation and make it more likely that employees who have experienced discrimination will feel safe enough to file charges. 

Looking at the characteristics of charging parties and their employers reveals other findings:  

• Demographics: For sexual orientation charges, a disproportionate number of charges were filed by men and by Black individuals. In contrast, for gender identity charges, a disproportionate number of charges were filed by women and White individuals.
• Intersectionality: Some SOGI discrimination charges include allegations of another kind of discrimination, such as race and disability.
• Industry: More than 40% of charges come from a small number of industries, which vary somewhat across types of states or charge. Several of these common industries pay low wages relative to the economy as a whole: Retail, Accommodations and Food Services, and Administration Waste Management & Remediation Services.
• Issue: The most common “issue” alleged (the employment practice that is being challenged) was discharge (54%), followed by harassment (47%), terms and conditions (29%), discipline (15%), and sexual harassment (14%).
  • Charges of harassment, sexual harassment, and discharge were more common for sexual orientation charges than for gender identity charges. Issues of hiring and benefits were more common for gender identity charges than for sexual orientation charges.
  • States without SOGI protections saw higher rates of serious charges like harassment, discharge, and retaliation than states with protections.
• Outcomes: Approximately 15% of SOGI charges result in some kind of benefit or positive outcome for the charging party. States without SOGI protections had a higher rate of benefits (17%) compared with states with SOGI protections (13%). Benefits include back pay, fringe benefits, compensatory or punitive damages, attorney’s fees, policy changes, training, or accommodations. 

 

The Data

This study is based on confidential data provided by the Equal Employment Opportunity Commission to the authors after approving their application to work with the data under the Intergovernmental Personnel Act. The data include all 9,262 charges that included an allegation of discrimination based on “sex-sexual orientation” or “sex-gender identity/transgender” in the EEOC’s charge processing database. These charges were filed with the EEOC or with a state or local Fair Employment Practices Agency (FEPA) from 2012-2016. The small number of charges filed with the EEOC in 2012 likely had a SOGI charge added later. We limit this study to the 9,121 charges filed by people living in the 50 states or District of Columbia, excluding those charges filed in U.S. territories. When calculating whether a charge results in a merit outcome, we limit the analysis to the 86% of charges that have been fully processed and closed.

Any charge related to Title VII of the Civil Rights Act may be filed with the EEOC directly or with one of the state or local nondiscrimination agencies that have worksharing agreements with the EEOC. Charges filed with either the EEOC or a FEPA are “dual filed” with the other. Therefore, the EEOC’s charge database typically includes SOGI cases filed with a state or local agency. However, the EEOC and FEPAs do not necessarily collect and/or enter into the electronic record the same demographic information about the charging party. Many of the charges filed with FEPAs do not include data on the race, sex, or other demographic characteristics of the charging party. Also, the EEOC does not collect data on the sexual orientation or gender identity of the charging party from the EEOC data; the charging parties are likely to be LGBT people or heterosexual people who are perceived as others to be LGBT.

To calculate the rate of charges filed per 100,000 LGBT adults per state, we drew on estimates of the LGBT population by state that were provided by the Williams Institute from the Gallup Daily Tracking Poll. The Gallup survey includes a question asking respondents whether they identify as LGBT. The percentage by state ranges from 2.4% in Idaho to 5.2% in Vermont and 9.0% in the District of Columbia. We then multiplied a state’s average percentage of adult Gallup respondents identifying as LGBT (averaged over the 2012-2016 surveys) by the adult population of that state estimated from the 2012-2016 American Community Surveys. 

2.1 Trends in Charge Filing Over Time

Between 2012 and 2016, a total of 9,121 charges were filed against respondents located in one of the 50 states or Washington, D.C., claiming some form of SOGI based discrimination. The EEOC received 5,352 SOGI charges, and FEPAs received 3,769 SOGI charges. 

Graph 1 shows the overall growth in charges filed during this period. As reflected in the graph, there was a sharp increase in the number of charges filed in 2013, which was the first year the EEOC began to formally accept SOGI charges. Some charges were filed earlier and may have had a SOGI charge added after the fact. Between 2012 and 2015, the number of charges filed annually increased dramatically, from 634 to 2,448, with a leveling off in 2016 when 2,275 charges were filed.⁴

Between 2012 and 2016, there was a notable increase in charges filed for both sexual orientation and gender identity discrimination, though the growth was much steeper for sexual orientation charges (Graph 2). In 2012, the EEOC and FEPAs received 396 charges based on sexual orientation and the number increased fivefold to 2,107 by 2015 and 1,992 in 2016. For gender identity, there was an 82% increase from 245 in 2012 to 445 in 2015.

2.2 TRENDS BY STATE

Currently, 21 states and D.C. have laws prohibiting SOGI discrimination, and Wisconsin prohibits discrimination based on sexual orientation only.⁵ Combined, these 22 states and the District of Columbia are home to half of the LGBT population in the United States. In our findings here we called them “states with SOGI protections” even though SOGI charges can now be filed with either the EEOC or through a state or local Fair Employment Practice Agency (FEPA) office in all states.

Although we see growth in charge filing between 2012 and 2016, most of the growth came in states without legal protections (shown in the dashed lines in Graph 3). Charges in states without protections grew from 126 charges filed in 2012 to 1,213 in 2016, a 10-fold increase in filing. In states with SOGI nondiscrimination policies (shown in the solid lines), charge filings increased more modestly, doubling from 508 filed in 2012 to 1,063 in 2016. These trends indicate that the EEOC policy change has created a new layer of protection against discrimination for LGBT people. 

In addition, the charge data reveal a division of labor between states with SOGI nondiscrimination laws and those without such laws (Graph 3). In states with SOGI or SO nondiscrimination laws, the state FEPAs (shown by the solid orange line) handled the majority of SOGI charges, or 63% of all charges filed. In states without SOGI protections, in contrast, the EEOC offices (shown by the dashed blue line) saw the majority of SOGI charges and handled 82% of all charges filed during our period. In the non-SOGI states, the remaining charges were filed with state or local FEPAs (dashed orange line). 

2.3 Population-adjusted rates of charge filings:

States with more LGBT people are likely to have more charges of discrimination, but the charge figures reported above do not take into account the LGBT population size. Therefore, we calculate a rate of SOGI charges per 100,000 LGBT adults by dividing the number of charges in each state by the size of the LGBT population in each state. We then calculated the charge rate for states with SOGI antidiscrimination laws and states without. States that have laws protecting either sexual orientation or both sexual orientation and gender identity have a 19% higher filing rate compared to states without any law—102 per 100,000 in the states with protection versus 83 per 100,000 in the states without protections.

A 2015 study of SOGI charges reported by FEPAs in states with SOGI protections used a similar methodology but found a much lower charge rate of 46 per 100,000.⁶ The number of charges filed with FEPAs in our data is very similar to the numbers reported by state agencies to the authors of the 2015 study, but our figures also take into account the charges filed with the EEOC, making the total charge rate much higher in the SOGI protection states. The higher rate in our study suggests that many more charges were also being filed in states with SOGI protections as a result of the EEOC change in policy. 

2.4 Discussion of trends:

The trends in filing SOGI charges suggest that the EEOC’s policy change dramatically increased the number of people who could report and seek redress for workplace experiences that they believe were discriminatory. These data indicate a rapid increase in LGBT individuals, or individuals perceived as being LGBT, using their rights to file discrimination charges. While it is possible that some heterosexual people might be filing charges of discrimination based on heterosexuality, it is unlikely these account for many (if any) of the charges. 

One obvious reason for the increase over time is the expansion of that right in states without explicit protections against SOGI discrimination. A second important reason is likely the spread of knowledge about the ability to file SOGI charges. In the states without SOGI protections, where the opportunity to file is new, the increasing number of charges over time shows the sharpest increase. The jump in the SOGI-protection states’ population-adjusted charge rates from an earlier study of state FEPA charges (2008-2014) suggests that knowledge of the EEOC policy increased filings everywhere as people learned of that change. Also, the number of charges in states with SOGI protections might have increased because the extension of federal recognition adds legitimacy to SOGI charges and encourages some charges that might not have been filed otherwise. It is also important to note that President Obama signed the executive order prohibiting federal contractors from discrimination based on SOGI status in 2014, an event that may have also increased knowledge of SOGI protections for some workers.

Another implication of the adjusted charge rates is that there is still underreporting of experiences of discrimination, particularly in the non-SOGI-protection states. The rate of charges is approximately 20% higher in states with SOGI protections than in the states without them. However, we might expect the SOGI-protection states to have lower rates of discrimination: laws are more likely to pass in more tolerant states, less discrimination is likely to occur in more tolerant states, and enforcement of the law is likely to reduce discrimination. Even if the level of actual discrimination is the same in both sets of states, the gap in adjusted rates of charge filings implies that more people in non-SOGI law states would file charges if federal law or state law made the ban on SOGI discrimination more explicit and, therefore, more visible. Those actions might also reduce fears of retaliation for filing a charge. 

 

Patterns of Charges Filed by Demographics, Industry, Basis, and Issue

3.1 Demographics

Demographic data on charging parties can reveal two important perspectives on discrimination: who is experiencing potentially discriminatory actions in the workplace and who is taking advantage of the right to file a charge of discrimination. While the EEOC database includes information on the race, sex, and some other characteristics of charging parties, the data are incomplete, mainly because the FEPAs are less likely to enter that data. For those charges with available data on sex, more are filed by men (55%) than by women (45%). For those with data on race/ethnicity, approximately 52% are White, 43% are Black, and 10% are Hispanic. (Approximately 16% of all charges do not include data on the sex of the charging party, and 41% are missing data on race.) To put these numbers in context, approximately 78% of the labor force is White, 12% is Black, and 17% is Hispanic.⁷ Black individuals, therefore, are notably overrepresented among SOGI charging parties.

• Demographics by charge type: When we compare the demographic differences between sexual orientation and gender identity charges, we see different patterns by sex and race/ethnicity. Within gender identity charges, women filed 63% of those charges—a much greater proportion than men’s 37% of gender identity charges. Men, in contrast, dominate charges of discrimination based on sexual orientation, with 58% of these charges being filed by men compared to only 42% filed by women. Interestingly, 25% of gender identity charges have missing data on sex, compared with 14% of sexual orientation charges with missing data. There are only two options on the intake questionnaire (male and female), and some people might not identify with those binary options. Also, that difference might reflect a lack of understanding about how to classify the sex of transgender people, even though EEOC intake personnel are supposed to report the gender identity of the charging party for the sex question. Thus, we do not know for certain whether the people marked as “female” on the intake form for gender identity charges are transgender women or people whose sex was assigned female at birth. Race comparisons show somewhat divergent patterns in filing between gender identity and sexual orientation claims. For gender identity charges, 55% of charging parties are White, 39% Black, and 7% Hispanic. For sexual orientation charges, 50% of claimants are White, 44% are Black, and 10% are Hispanic. Approximately 39% of sexual orientation charges and 45% of the gender identity charges did not include race data. 
• Demographics across states: We also compared demographic differences between states with and without SOGI protections. We see very few differences in filing by sex, with approximately 56% of charges in SOGI states filed by men compared to 55% in non-SOGI states. As noted above, data on sex is more likely to be missing in SOGI states because most of those charges are filed with FEPAs, which are less likely to report race and sex (approximately 27% of charges are missing sex information for SOGI states vs. 4% for non-SOGI states). We find more differences when we compare race between SOGI and non-SOGI protection states. Charging parties in non-SOGI protection states are slightly more White than those in SOGI states (50% vs. 54%). Notably, however, a greater proportion of charging parties in non-SOGI states are Black, at 47%, compared to 38% in SOGI states. This pattern is likely at least partially attributable to the fact that approximately 16% of the population is Black in states without SOGI laws, as compared to 10% in states with SOGI laws.⁸ Also, 11% of charging parties are Hispanic in non-SOGI states and 9% in SOGI states. 

3.2 Industry

The product or service that an employer produces provides some information about the types of jobs that charging parties might hold. We analyzed our charge dataset using the industry classification used in the EEOC database (2-digit NAICS industry codes) to identify the employment sectors most likely to see charges filed. Unfortunately, 58% of charges did not list an industry at all. Of the charges with valid industry information, the industries with the largest number of charges filed include the Health Care and Social Assistance industry (15%), Retail (14%), Accommodation and Food Services (12%), and Manufacturing (10%). Notably, the average hourly earnings of workers in the Retail and in Accommodations and Food Services are significantly below the average hourly earnings for all workers.⁹

• Industry across charge type: When we compare industry level data between sexual orientation and gender identity, we find some slight differences in where charges emerge. Sexual orientation charges mirror the overall trend with Health Care and Social Assistance (16%), Retail (14%), and Accommodation and Food Services (12%) being the three largest represented industries in the charge dataset. For gender identity charges, Retail (17%) was the largest industry represented, with Health Care (13%) and Manufacturing (13%) following closely behind. 
• Industry across states: We also compared industry level differences between states with and without SOGI nondiscrimination policies. We found minor differences between these two sets of states in terms of where charges are focused. In states with SOGI nondiscrimination policies, the top three industries represented are Healthcare (16%), Administration Waste Management and Remediation Services (13%), and Retail (13%). Note that Administration Waste Management and Remediation Services is also a low wage industry.¹⁰ In states without legal protections, the top three industries are Retail (16%), Health Care (15%), and Accommodation and Food Services (15%).

3.3 Intersectionality

Although our focus is on charges that contain an allegation of discrimination based on sexual orientation and/or gender identity, employees often allege discrimination on multiple bases. Looking at whether other bases of discrimination are alleged provides an angle on “intersectionality”, or whether combining people’s marginalized identities affects their vulnerability to discrimination. For example, an individual might be harassed because of their sexual orientation and deprived of accommodations because of their disability. Or being LGBT and having a disability might make someone more vulnerable to a discriminatory discharge than LGBT people without a disability. For SOGI discrimination, the most common other bases of alleged discrimination include race (17%) and disability (14%). The overrepresentation of Black charging parties, coupled with these allegations of race discrimination, suggests that the intersection of these stigmatized identities could shape experiences of employment discrimination for this group.

• Intersectionality across charge type: We found differences in the other bases alleged for charges of sexual orientation discrimination in comparison to gender identity. Race discrimination was alleged in a greater proportion of sexual orientation charges (19%) than gender identity (10%), as was disability (15% of sexual orientation charges vs. 11% gender identity). Overall, sexual orientation charges contained more additional bases of discrimination than did gender identity charges.
• Intersectionality across states: Other bases of alleged discrimination varied somewhat for charges filed in states with SOGI protection versus those without protection. In particular, race bases were slightly more common in charges in states without SOGI nondiscrimination laws (19%) compared to those with such laws (16%). As previously noted, this could be attributable to differences in racial demographics in states with and without SOGI nondiscrimination laws.

3.4 Issues

The charge of employment discrimination includes allegations related to specific employer practices that constitute discrimination, such as termination, unequal compensation, or harassment. The EEOC calls these alleged experiences “issues.” The EEOC and FEPAs match up a basis for discrimination (such as sexual orientation) with an issue to create an allegation. However, other ongoing research on this data suggests that it is often difficult for charging parties to identify which basis is linked to a particular employer practice. For example, if an employee alleges a discharge issue, the discharge could be attributable to sexual orientation, sex, race, or a combination of these factors. Our approach is to assume that the intersection of identities could be related to any particular issue. Therefore, here we report the issues contained in a charge of discrimination that includes at least one allegation of SOGI discrimination, given that any or all of these issues could be tied to SOGI identity or the way that identity intersects with another protected characteristic. In SOGI charges, the most common issue alleged was discharge (54%), followed by harassment (47%), terms and conditions (29%), discipline (15%), and sexual harassment (14%).¹¹  In addition, approximately 42% of charges alleged retaliation.¹²

• Issues across charge type: Some differences emerged in the issues alleged for charges based on sexual orientation and those based on gender identity. Approximately 48% of the charges based on sexual orientation contained allegations of harassment, in contrast to 40% of those charges based on gender identity. A higher proportion of charges of sexual orientation also included issues of sexual harassment (15%) as compared to gender identity (9%). In addition, a higher proportion of charges of sexual orientation discrimination, compared to gender identity charges, included issues of discharge (56% vs. 48%) or constructive discharge (10% vs. 7%). A higher proportion of charges of gender identity discrimination, on the other hand, include issues of hiring (10% vs. 4%) and benefits (5% vs. 2%). In addition, retaliation was a much more common allegation in sexual orientation charges (44%) than in gender identity charges (31%).
• Issues across states: Two primary differences emerged in the types of issues contained in SOGI charges in states without SOGI nondiscrimination laws compared to those with such laws. Approximately 52% of charges in states without SOGI nondiscrimination laws included harassment as an issue, in contrast to 41% of charges in states with SOGI nondiscrimination laws. In addition, discharge was alleged as an issue in 58% of charges in states without SOGI nondiscrimination laws, in comparison with 51% of charges in states with SOGI nondiscrimination laws. Further, charges in states without a SOGI law were more likely to include retaliation (46%) as a basis than those with a SOGI nondiscrimination law (39%). In states with SOGI nondiscrimination laws, in comparison to those without SOGI nondiscrimination laws, there was a slightly higher proportion of charges with terms and conditions as an issue (31% vs. 27%) and sexual harassment as an issue (15% vs. 13%).

 

Outcome of Charges

We were also able to measure whether a benefit accrued to the charging party during the EEOC or FEPA’s handling of the case. Monetary benefits include items such as back pay, fringe benefits, compensatory damages, punitive damages (primarily under state law), or attorney’s fees. Nonmonetary benefits include items like policy changes, training, or accommodations. The EEOC reports that its charges have resulted in $16.1 million in monetary benefits for charging parties.¹³ This figure would not reflect any benefits accrued through private discrimination lawsuits filed by charging parties following receipt of a Right to Sue letter from the EEOC. Overall, approximately 15% of closed SOGI charges resulted in a benefit, with no notable differences between sexual orientation and gender identity charges or between charges handled by FEPAs or the EEOC. This rate is slightly lower than the 17.5% overall average rate of merit resolutions seen for all EEOC charges resolved in fiscal years 2013-2016.¹⁴ Charges in states without SOGI nondiscrimination laws were slightly more likely to result in a meritorious outcome at 17%, whereas 13% of charges in states with SOGI nondiscrimination laws resulted in a meritorious outcome for the charging party.

 

Conclusion

Between 2012 and 2016, the EEOC’s policy that considers sex discrimination to include SOGI discrimination has dramatically increased employees’ access to legal recourse when they believe they have experienced SOGI discrimination. More than 9,100 charges have been filed.

The overall filing trends demonstrate that people in states without a SOGI nondiscrimination law incurred the greatest benefit from the EEOC’s policy change, with a 10-fold increase in SOGI filings during this period. The EEOC has handled the majority of charges in these states, emphasizing that the change at the federal level provided previously unavailable remedies for persons in these states. In addition, many more charges have been filed with the EEOC in states that do have SOGI protections, so access appears to be greater in both kinds of states since 2013. 

Some differences have emerged in filing patterns. There have been notable increases in charge filings alleging discrimination based on sexual orientation and gender identity, but the increase has been most dramatic for sexual orientation. This larger increase could be due in part to transgender individuals already filing charges under the earlier recognition of gender stereotyping discrimination, and some of the difference is likely related to the fact that there are many more LGB people than transgender people.

The rate of filing in states without SOGI protection, 83 charges per 100,000 LGBT people, is much lower than in the states with SOGI protections, 102 charges per 100,000 LGBT people. Thus our findings suggest that SOGI charges are underreported in the states without SOGI protections. A federal or state law that increases the visibility of the right to file a discrimination charge could motivate people to file even larger numbers of charges in the 28 states with no explicit SOGI protections.

Our analysis of the characteristics of charging parties and employers reveals other important patterns. Sex and race-based differences were evident in SOGI charges, and these varied across sexual orientation and gender identity charges. For sexual orientation charges, a disproportionate number of charges were filed by men and by Black individuals. This could reflect less accepting attitudes toward gay men than toward lesbians, resulting in more negative workplace experiences for gay men, or a higher likelihood of filing a charge for men and African Americans.

In contrast, for gender identity charges, a disproportionate number of charges were filed by women and White individuals. Given the lack of clarity in the sex categorization, it is unclear what this might mean for gender identity claims. If investigators are entering individuals’ sex as reflecting their lived identity, this could suggest that more claims are filed by transgender women; if so, this pattern could also reflect the lower status of perceived femininity or being female in the workplace. Ongoing research on the details of the charges also suggests the possibility that some charging parties are misgendered by investigators so it is likely that sex categorization for those filing gender identity claims is more ambiguous than that for lesbian and gay charging parties.

The disproportionate number of SOGI charges filed by Black individuals could reflect differences in the overall legal awareness of this group, with African Americans more likely to perceive discrimination and to file charges as a result of experiences with race discrimination. This is supported by the fact that approximately 17% of the charges included race as a basis. This pattern also varied geographically, with a greater proportion of charging parties being Black and a greater proportion of race-based charges in states without SOGI nondiscrimination laws, perhaps because of large populations of African Americans in states without SOGI protections. These findings related to sex and race suggest that intersectionality of stigmatized characteristics could play an important role in SOGI discrimination.

Variation in bases and issues between states with and without SOGI nondiscrimination laws suggests that the workplace experiences of LGBT people (and those perceived to be LGBT) might be shaped by the sociopolitical context within a state. In particular, a higher proportion of charges in states without SOGI nondiscrimination laws included a retaliation basis, as well as issues of discharge and harassment; the presence of these types of issues could explain the slightly higher proportion of charges resulting in a meritorious outcome in these states. These differences in rates of filing and issues alleged could be attributable to prior lack of legal enforcement, as well as to a lack of internal employer policies and practices, to reduce SOGI discrimination. Further, those living in these states are likely to encounter more negative attitudes about LGBT individuals that could manifest as harassment or retaliation. These findings suggest that the availability of a federal remedy has the potential to improve workplace experiences for LGBT individuals residing in states without SOGI nondiscrimination laws. 

 

Acknowledgements: We thank the Gill Foundation for funding for this publication and our online visualizations. The U.S. Department of Labor’s Employment and Training Administration and Chief Evaluation Office also provided funding for the larger project. We are grateful to the EEOC for making the data available to us, and to Mo Willow, Ron Edwards, and Patrick Davis for their assistance and feedback. We thank EEOC Commissioner Chai Feldblum, Naomi Goldberg, Sara Santos, and Don Tomaskovic-Devey for their comments on an earlier draft. We thank Kerith Conron from the Williams Institute for making state-level Gallup data available to us. Carly McCann and Eunjung Jee provided excellent research assistance. The analysis and interpretation of the data represent the views of the authors and not those of funders or the EEOC.

_{1. Movement Advancement Project, “Percent of LGBT Population Covered by Laws,” http://www.lgbtmap.org/equality-maps/non_discrimination_laws (last accessed May 15, 2018).}

_{2. The EEOC ruled in two cases related to federal employment that discrimination based on gender identity (Macy v. Department of Justice, 2012) or sexual orientation (Baldwin v. Department of Transportation, 2015) is discrimination based on sex and is illegal under Title VII of the Civil Rights Act. See also U.S. Equal Employment Opportunity Commission, “What You Should Know About EEOC and the Enforcement Protections for LGBT Workers,” https://www.eeoc.gov/eeoc/newsroom/wysk/enforcement_protections_lgbt_wor... (last accessed May 15, 2018).}

_{3. Altitude Express v. Zarda from the 2nd Circuit and R.G. & G.R. Harris Funeral Homes vs. EEOC from the 6th Circuit.}

_{4. The EEOC reports data on SOGI charges they received that also shows a leveling off in fiscal year 2017. U.S. EEOC, “LGBT Based Sex Discrimination Charges,” https://www.eeoc.gov/eeoc/statistics/enforcement/lgbt_sex_based.cfm, accessed July 9, 2018.}

_{5. New Hampshire passed a law in 2018 that extended their state antidiscrimination protections to include gender identity.}

_{6. Christy Mallory and Brad Sears, “Evidence of Employment Discrimination Based on Sexual Orientation and Gender Identity: An Analysis of Complaints Filed with State Enforcement Agencies, 2008-2014,” Williams Institute, 2015, https://williamsinstitute.law.ucla.edu/wp-content/uploads/Employment-Dis.... The denominator in that study was somewhat smaller, since it multiplied the percentage of LGBT people in the population times the state’s labor force.}

_{7. U.S. Bureau of Labor Statistics, “Labor Force Characteristics by Race and Ethnicity, 2016,” https://www.bls.gov/opub/reports/race-and-ethnicity/2016/pdf/home.pdf, accessed July 12, 2018.}

_{8. Calculated from U.S. Census Bureau “Annual Estimates of the Resident Population by Race,” https://factfinder.census.gov/faces/tableservices/jsf/pages/productview...., accessed July 12, 2018}

_{9. In the January Current Employment Statistics, the Bureau of Labor Statistics reports that the average hourly wage in the U.S. was $26.71 (https://www.bls.gov/web/empsit/ceseesummary.htm) compared with $18.45 in Retail (February 2018; https://www.bls.gov/iag/tgs/iag44-45.htm) and $14.89 in Accommodation and Food Services (January 2018; https://www.bls.gov/iag/tgs/iag72.htm).}

_{10. In January 2018, average hourly earnings for all employees in this industry was $20.55 (https://www.bls.gov/iag/tgs/iag56.htm).} 

_{11. Harassment is unwelcome conduct, such as offensive jokes, slurs, insults, physical assaults, threats, or ridicule, that must be endured to continue working for an employer or that creates a hostile work environment. EEOC, “Harassment,” https://www.eeoc.gov/laws/types/harassment.cfm, last accessed July 13, 2018. In these charges, the harassment might be related to SOGI status or to some other protected characteristic such as race or sex. Sexual harassment includes unwelcome sexual advances and harassment of a sexual nature. EEOC, “Sexual Harassment,” https://www.eeoc.gov/laws/types/sexual_harassment.cfm, last accessed July 13, 2018.}

_{12. Although the EEOC treats retaliation has a basis rather than an issue, we report retaliation statistics in the issues section of this report.}

_{13. U.S. EEOC, “LGBT Based Sex Discrimination Charges,” https://www.eeoc.gov/eeoc/statistics/enforcement/lgbt_sex_based.cfm, accessed July 9, 2018.}

_{14. U.S. EEOC, “All Statutes (Charges Filed With EEOC) FY 1997-FY2017,” https://www.eeoc.gov/eeoc/statistics/enforcement/all.cfm, accessed July 10, 2018.}