Quantifying The "Epidemic of Violence" Against Trans People
“2020 is now the most violent year for transgender people in America for five years.”
- Forbes, 10/2/2020
“‘Horrific spike’ in fatal violence against transgender community”
- CBS, 7/14/2020
“Philadelphia murder is latest in ‘epidemic of violence’ against trans community, city says”
- The Independent, 9/30/2020
The messaging around transgender civil rights is very often framed in terms of public safety from fatal anti-trans violence. Most recently, employees at Netflix protested content that mocks or invalidates transgender messaging, arguing that it contributes to violence against trans people. While it’s probably difficult to establish a direct causal link between transphobic media content and actual violence, the claim that trans people (particularly black trans people) face disproportionately high odds of being murdered is rarely contested. Media outlets have published countless articles reporting on the murder of transgender individuals,1 consistently framing each murder as part of an “epidemic” of violence. The phrase “epidemic of violence” is quoted or repeated with remarkable consistency across multiple organizations, including CNN2, Forbes3, the New York Times4, the ACLU5, the Biden campaign6, and various other news outlets7. The origin of this phrasing appears to be the Human Rights Campaign (HRC), which consistently uses this phrase in its annual reporting on transgender murder victims.8
The claim that trans people face a disproportionate risk of being murdered is generally accepted as fact by pretty much every reputable media outlet. Surveys of the transgender population reveal disproportionately high rates of abuse, discrimination and unemployment,9 but the claim that trans people are disproportionately murdered is rarely explored beyond a simple count of annual transgender murder victims. In order to determine whether these murders really constitute an “epidemic” - that is, whether trans people actually face a disproportionate risk of fatal violence - we at least need to compare the risk level facing trans people to some baseline risk level.
Since transgender people make up a small percentage of the population, the absolute number of transgender murder victims is comparatively small. But we are interested in the probability of being murdered faced by a randomly selected transgender person living in the United States. In order to make this probability meaningful, we must at the very least compare it to the probability of being murdered faced by a random member of the US general population. Comparisons with the risk facing cisgender white people or other identity groups may also provide useful insights. If the phrase “epidemic of violence” is accurate, we should at least expect trans people to face a significantly higher probability of being murdered than the general population. In this post, I use multiple data sources to estimate the risk of murder facing trans people in the US, and then compare this to the risk faced by the general population and other identity groups.
Transgender Murder Victims
The FBI Uniform Crime Reporting program does not track the number of transgender murder victims, so we must rely on data from other organizations such as the Human Rights Campaign. According to the Human Rights Campaign, 44 “transgender or gender non-conforming” individuals were murdered in the United States in 2020.10 The HRC does not specify whether all of these individuals were targeted because they were transgender. However, the HRC does provide the name of each murder victim along with links to associated media coverage. In some cases, the authorities claimed there was no evidence to suggest the victim was targeted because they were transgender.11 For example, in one case a gender non-conforming person was murdered by a burglar during a home invasion.12 In other cases, it’s very clear that the victim was murdered because they were transgender, such as in the tragic case of Neulisa Luciano Ruiz, a trans-woman who was murdered after using a public women’s restroom.13 There are also cases where the transgender identity of the victim was a factor, but transphobia was probably not a motive, such as one case in Harlem, New York where a trans-woman was likely murdered by another trans-woman.14
Of the 44 murders in 2020 recorded by the HRC, 6 of them were almost certainly motivated by transphobia15, 5 of them were almost certainly NOT motivated by transphobia16, and the remaining murders are unsolved or involve an unclear motive. Regardless, my analysis will be agnostic to motives. It’s possible that trans people are disproportionately murdered for reasons unrelated to transphobia; perhaps being trans correlates with other factors that increase the likelihood of being murdered. We also must consider that, according to the HRC, the actual number of transgender murder victims may be significantly higher than the reported figures, because local news outlets often misgender or “deadname” transgender murder victims (i.e. report the name used by the victim before they transitioned), decreasing the chances that any particular victim shows up on the HRC’s radar.17
We’ll assume for now that the figure of 44 murders in 2020 is accurate. An estimated 0.6% of the US adult (ages 18 and over) population in 2020 was transgender18, so using data from the 2020 census19 we can estimate that there were about 1,550,060 transgender adults in the US in 2020. I limit this analysis to people 18 and over because the available data on the transgender population is mostly limited to adults. While there is a 2017 UCLA study that indicates a significant percentage of trans people are under 18,20 the median age of transgender murder victims in 2020 was 27, and only one victim was under 18. So limiting our analysis to transgender adults controls for age, allowing a more accurate comparison. With that said, the probability of a randomly selected transgender adult in the US being murdered in the year 2020 was 0.003%.
This may seem like a very low probability, but in order to draw any useful conclusions we have to put this into context by comparing it against the probability of being murdered faced by the general population, the white majority, and other minorities. Using 2020 census data21 and homicide data provided by the FBI22, we get the following probabilities of being murdered in 2020 for a random adult member of the US general population, along with a random adult white, black, and Latino person respectively: 0.007%, 0.002%, 0.029%, 0.007%.23
Hmm… these results are highly counter-intuitive. It seems transgender people face less of a risk of being murdered than the general adult population, and only slightly more of a risk of being murdered than white people as a group. Trans people also appear to face significantly less of a risk of being murdered than blacks or Latinos as a group.
We can also look at this another way: given approximately 18,194 adults murdered in the US in 2020,24 and a total population of 258,343,281 adults in the US in 202025, the probability of a random adult being murdered is about 0.007%. Thus, given that around 1,550,060 adults in the US are transgender, we could reasonably expect at least about 0.007% of them to have been murdered in 2020, simply by pure chance. That comes out to about 108 transgender murder victims, which is way more than the 44 transgender murder victims reported in 2020. This indicates that transgender people, as a group, are actually less likely to be murdered than a random member of the general adult population. This certainly doesn’t seem like an “epidemic”. So what the hell is going on here?
One possibility is that, as the HRC claims, transgender murder victims are widely under-reported. Yet we’d have to assume that over 90% of transgender murder victims are unreported if the risk of murder facing trans people is to even approach the risk facing black people as a group. Regardless, the above analysis is probably much too coarse-grained of a comparison. While the HRC and the media often imply the “epidemic of violence” afflicts all trans people, they also consistently clarify that it most harshly afflicts black trans people.
Quantifying the Epidemic of Violence Part 2 (Where I Embrace Intersectionality)
I find it difficult to believe that trans people face nearly the same risk of being murdered as the mostly cisgender white majority, given the disproportionately high rates of abuse, discrimination and unemployment reported in surveys of the transgender population.26 Perhaps a more intersectional analysis is required here. Notably, the vast majority of transgender murder victims in 2020 were trans-women (84.1%), and 72.7% were black or Latina trans-women. Yet among the general population, the majority of white, black and Latino murder victims in 2020 were cisgender males (73%, 85.4%, and 83.5% respectively). Therefore, we probably want to compare risk of being murdered while breaking down the trans population by gender and race. Data providing break-downs of the transgender population by gender or race is less recent and probably has greater margins of error than total population estimates27, but the 2015 US Transgender Survey (USTS) breaks down the transgender population by gender, giving 33% trans-women, 29% trans-men, and 35% non-binary.28
Wait… 35% non-binary? Reading over the relevant literature, it quickly becomes clear that the definition of “transgender” is still kind of fuzzy. A 2020 Gallup poll estimates that 0.6% of the US adult population identifies as transgender, but this poll doesn’t even mention non-binary as a category.29 A 2016 UCLA study estimates 0.58% of the adult population is transgender,30 and a 2017 study estimates 0.6%.31 Neither uses the term “non-binary”, but both define transgender as “male-to-female, female-to-male, or gender non-conforming”. The CDC defines transgender the same way,32 and doesn’t use the term “non-binary” either. But the USTS survey uses transgender as an umbrella term that includes non-binary or “genderqueer” individuals,33 and a 2021 study treats non-binary as synonymous with genderqueer.34
Regardless, of the 44 transgender people murdered in 2020, only 3 of them were non-binary or gender-non-conforming. A large majority (84.1%) were male-to-female (trans-women). Clearly, trans-women face a substantially greater risk of being murdered than other types of transgender people, and should not be statistically combined with other transgender individuals when analyzing the risk of murder.
The percentage breakdown provided by the 2015 USTS survey (estimating 35% of transgender people to be non-binary) is supported by a 2021 UCLA study of the non-binary population, which estimates that 32.1% of transgender people are non-binary.35 A different 2021 UCLA study breaks down the transgender population as 38% trans-women, 31% trans-men, and 31% non-binary, which also more or less supports the USTS percentages.36
Averaging out these estimated percentages, we’ll break down the transgender population as 36.1% trans-women, 30.6% trans-men, and 33.3% non-binary. This gives 559,246 trans-women over the age of 18, and thus a 0.007% probability that a random trans-woman would be murdered in 2020. This result seems more plausible - it’s more than twice the risk faced by white people as a group. But it’s still equal to the risk faced by the general population, and utterly negligible compared to the risk of murder faced by black people as a group.
We’ll need to further break down the trans-woman population by race in order to get a clearer picture. A 2021 UCLA study on transgender women37 allows us to break down trans-women by race, giving 67.63% white, 11.03% black, and 15.86% Latina. Applying these to our total estimated population of trans-women, we get the following probabilities of being murdered in 2020 for a random white, black, or Latina trans-woman respectively: 0.001%, 0.037%, 0.01%. Comparing these probabilities to the general population, we get:
When we break down trans-women by race, it becomes clear that black trans-women face a highly disproportionate risk of being murdered compared with other trans-women, and even a higher risk than all black people as a group. This at least appears to validate the HRC’s claims about an “epidemic of violence” afflicting black trans-women. Similarly, Latina trans-women face a higher risk of being murdered than the Latino/a population as a whole. Yet this comparison still counter-intuitively demonstrates that white trans-women, who constitute the majority (67.6%) of all trans-women, are actually less at risk of being murdered than the white population as a whole.
Furthermore, this comparison is not entirely fair either, because it breaks down trans people by gender and race, but breaks down the general population by race alone. If we further break down our comparison, this time comparing the risk of murder facing trans-women with the risk of murder facing cisgender males, broken down similarly by race, we get the following:
This reveals that cisgender men face a higher risk of murder than trans-women of the same race. The disparity is particularly obvious with cisgender black males, who are about 46% more likely to be murdered than black trans-women. And the risk of murder for Latino males is slightly higher than the risk for Latina trans-women. Even the risk of murder to white males is greater than the risk of murder to white trans-women.
At this point, two major problems with the “epidemic of violence” narrative have emerged. Firstly, claiming that this epidemic afflicts transgender people in general is very misleading. As we have seen, it specifically afflicts black trans-women, a very tiny subset of trans people - a minority within a minority. Most trans people are not trans-women, and most trans-women are not black. Black trans-women make up only 4% of transgender people, comprising only around 61,721 people in the entire United States. The remaining 96% of trans people, consisting mainly of white trans-women, trans-men and non-binary people, face a similar or lesser risk of being murdered compared with the general population. This seems to discredit, or at least partially discredit, the widely reported “epidemic of violence” narrative. Perhaps it would be accurate to claim that an “epidemic of violence” specifically afflicts black trans-women, but claiming that trans people in general are afflicted is simply not true. While the HRC does consistently emphasize that black trans people are most harshly affected by this epidemic,38 general media reporting very often falsely implies the epidemic afflicts all trans people.39
Secondly, while black trans-women clearly face a disproportionate risk of being murdered, they still face less of a risk than cisgender black males. Generally, the concept of intersectionality is invoked to explain the high murder rate of black trans-women.40 That is, multiple oppressed identities (black, female and trans) compound to produce a uniquely destructive form of oppression intersecting with racism, transphobia and sexism.41 This puts black trans-women at the “top” of the intersectional hierarchy, pronouncing them to be the most oppressed of the oppressed, with the disproportionately high risk of being murdered serving as justification. Yet the homicide data we have doesn’t necessarily suggest this conclusion. Since men (especially black men) are murdered more often than trans-women when we control for race, we could also interpret the data to suggest that maleness is the strongest identity-based determinant of the risk of murder. Well… we could interpret it that way if it weren’t for the fact that the risk of murder facing trans-men is negligible. But more on that later. First…
Some Caveats…
I present this conclusion with some caveats. The HRC claims that the actual number of transgender murder victims is probably much higher,42 presumably because law enforcement or local media often mistakenly misgenders or deadnames the victim.43 The 2015 USTS report lends credibility to this claim, indicating that 49% of transgender survey participants did not have any ID with their preferred name on it.44 Additionally, a Media Matters analysis found that of the 37 transgender victims murdered from January to November of 2020, 23 of them (62%) were initially deadnamed or misgendered by the news media.45 This may suggest that a significant percentage of transgender murder victims are never counted.
On the other hand, a closer analysis of these 23 misidentified victims reveals that news outlets sometimes deadname the victim but nonetheless correctly identify the victim as transgender.46 Secondly, even when the victim is misgendered or deadnamed and not identified as transgender initially, it takes 5 days on average before at least one local media outlet corrects the initial reporting or publishes a report that correctly identifies the victim.47 (It takes only 3 days on average if we remove a single outlier case where it took a month to identify the victim.) LGBT and black media outlets often correctly identify the victim much earlier.48 Thus, initial misgendering or deadnaming of trans victims may be a problem of journalistic etiquette, but there’s no evidence that this systematically prevents or even significantly delays transgender murder victims from being identified as trans.
In many cases the police initially have no name to report other than the name on the victim’s ID, which is often the victim’s pre-transition name. In some cases the police can successfully identify a transgender victim without any ID.49 In other cases, the victim’s family asks the media to use the victim’s birth name, or deadnames the victim when speaking to the media.50 But there is no evidence that police or medical examiners are inexplicably incapable of determining that a victim was trans (unless the victim doesn’t present as their preferred gender). Police departments often communicate with local LGBT advocacy groups or provide details about trans victims to a local Office of LGBT Affairs.51
Finally, if a murder victim is not identifiable as transgender to medical or law enforcement personnel, then we have to wonder if the actual murderer perceived the victim as transgender. If the victim did not clearly present as their preferred gender - that is, if the victim was not easily identifiable as trans in some way - then we can’t easily conclude that the murderer was motivated by transphobia. So if some huge percentage of transgender murder victims are never identified as trans, we might conclude that a similarly large percentage of these murders were probably not motivated by transphobia. Of course, my analysis is meant to be agnostic towards motive, and transphobia is not the only factor producing violence against trans people. But a significant absence of explicit transphobia in homicide offenders would at the very least weaken the case for the standard intersectional argument and suggest that other variables better explain anti-trans violence. In any case, it’s possible that some percentage of transgender murder victims are unreported, but ultimately the available data is what it is; speculating about missing victims is pointless without additional evidence.
Comparing women to women…
We’ve seen that adult trans-women face less of a risk of being murdered than adult cis males of the same race. This is at least partially intuitive, because after controlling for race, women in general face less risk of being murdered than men. On the other hand, trans-women face a significantly greater risk of being murdered than cis women of the same race.
The probability of a random black adult trans-woman being murdered is over five times the probability of a random black adult cis woman being murdered. Combining this with the previous comparison showing that cis males face a higher risk of murder than trans-women of the same race, one possible conclusion is that the most significant predictor of being murdered is being “assigned male at birth”. A bizarre and counter-intuitive corollary of this is that a randomly selected adult male living in the US is likely to face less risk of being murdered if he becomes a trans-woman.
We’ve seen that cis females face much less of a risk of murder than cis males or trans-women, but so far we haven’t discussed trans-men. The number of trans-men murdered each year is so incredibly small, yet trans-men comprise almost one third of all transgender people. The probability of a randomly selected trans-man being murdered in 2020 is negligible (approximately 0.0008%). Trans-men face less risk than any other identity group we’ve examined. The sample size of trans-man homicide victims is so tiny (there were 4 in 2020) that breaking it down by race is pointless. (If you’re curious, it’s 2 black, 1 white and 1 Latino.) Assuming accurate reporting, trans-men seem to be more or less murder-proof. We might conclude from this that biological females as a group, including both cis females and trans-men “assigned female at birth”, face the lowest risk of murder when controlling for race.
The Puerto Rican Problem
We’ve seen that after controlling for race, cisgender males face a higher risk of being murdered than trans-women (or transgender people in general). Yet the comparisons I’ve presented so far are still not entirely fair; they slightly inflate the risk of murder faced by trans-women. This is because of the 44 transgender murder victims reported by the HRC, a highly disproportionate number of them (13.6%) were murdered in Puerto Rico. Puerto Rico is a US territory, not a state, and all other data used to calculate probability of murder (including the size of the transgender population, census data breaking down populations by age/race/gender, and FBI homicide statistics) accounts only for the 50 US states. Thus, the number of transgender homicides is inflated in this analysis.
A more fair comparison would exclude the transgender murder victims from Puerto Rico, or account for the Puerto Rican population in the general census and homicide data. Including Puerto Rico in my analysis requires a larger degree of estimation because the FBI does not provide a racial breakdown of homicide data for Puerto Rico and all estimates of the transgender population size in the US that I have found do not account for Puerto Rico. Regardless, using demographic data for Puerto Rico provided by the Census Bureau and homicide data from the 50 states, we can estimate the racial breakdown of homicide victims in Puerto Rico.52 If we further assume that the 0.6% national transgender relative population size estimate is not significantly affected by the additional 3.3 million residents of Puerto Rico, we can produce the following comparison of the risk of murder facing trans and other identity groups in the 50 US states and Puerto Rico:
Accounting for residents of Puerto Rico in the census and homicide data slightly increases the risk of murder faced by cisgender black and Latino males. Alternatively, if we remove Puerto Rico entirely (i.e. if we remove homicide stats and population data for Puerto Rico), we get:
Removing Puerto Rico from our analysis halves the risk of murder faced by Latina trans-women, and slightly decreases the risk to cisgender black and Latino males. But regardless of whether or not we account for Puerto Rico in our analysis, the overall conclusion remains the same: the risk of murder facing cisgender males is higher than the risk facing trans-women of the same race.
2020: A Most Violent Year
It has been widely reported by the HRC and the media that 2020 saw a particularly sharp increase in the number of transgender murder victims in the US.53 In fact, 2020 saw the highest number of transgender murder victims ever recorded (until 2021 surpassed it.54) The number of transgender murder victims remained relatively stable from 2015 through 2019, but there was a sharp increase of about 76% between 2019 and 2020.55
Despite the fact that cisgender males are more likely to be murdered than trans-women of the same race, this sharp increase between 2019 and 2020 could potentially justify the use of a phrase like “epidemic of violence” towards trans-women. The HRC and the media have been using this phrase for years at this point,56 but in 2020 the phrase may be justified due to the unusually high number of transgender people that were murdered.
On the other hand, the overall number of homicides in the US also saw a substantial increase in 2020:57
Yet, the overall increase in homicides can only partially explain the spike in transgender homicide victims in 2020, since the number of transgender victims increased by 76% in 2020, whereas the overall number of homicide victims increased by only 25.4%.
Of course, most transgender murder victims are black trans-women, so presumably most of the 2020 increase occurred due to an increase in black trans-woman murder victims. Yet strangely, when we look at the number of black trans-women murdered each year from 2015 through 2020, we find something very counter-intuitive:
Hmm… the number of black trans-woman murder victims barely increased at all between 2019 and 2020. So what explains the insane 76% increase in transgender murder victims between 2019 and 2020? If we break down the number of transgender murder victims each year by race, a very revealing pattern emerges:
It appears that between 2019 and 2020, almost every group of transgender murder victims either increased very slightly (e.g. black trans-woman murder victims increased from 21 to 23) or stayed within its historical range, except for Latina trans-women. The number of Latina trans-woman murder victims went from 1 in 2019 to 9 in 2020. Prior to 2020, the highest number of Latina trans-woman murder victims was 3 (recorded in 2018, 2016, and 2015). A sudden jump to 9 murdered Latina trans-women in 2020 is pretty alarming and requires some sort of explanation.
It just so happens that 2020 is also the first year that the HRC ever recorded any transgender murder victims from Puerto Rico. Looking through the HRC victim reports from 2015 through 2019, there is not a single murder victim from Puerto Rico recorded. Suddenly, in 2020, there are 6. What appears to have happened is that in 2020 the HRC expanded its reach to cover murders in Puerto Rico. As we’ve seen earlier, Puerto Rico also has an unusually high number of transgender murder victims. This caused a substantial spike in the total number of reported Latina trans-women murder victims, which, combined with the overall increase in homicides throughout the United States, could potentially explain the 2020 spike in transgender murder victims. Apart from the increase in murders reported from Puerto Rico, the murder rate for each transgender group increased only very slightly, with a percentage increase that is explainable entirely by the overall increase in homicides across the US in 2020. Additionally, when we’re dealing with such small numbers, tiny fluctuations in the number of murder victims can substantially effect the percentage increase. For example, the number of non-binary victims increased by an “astonishing” 200% from 2019 to 2020, because it increased from 1 to 3 people. We can more easily see this if we remove the victims murdered in Puerto Rico from the total number of transgender murder victims in 2020 for each identity group:
Without the spike in Latina trans-women caused by the inclusion of Puerto Rico, we can see that every group stayed close to its historical range, except for black trans-woman victims, which increased from 21 to 23, and non-binary victims, which increased from an historical high of 1 victim to 3 victims in 2020. Yet because of the “small number effect” these slight increases between 2019 and 2020 add up to a seemingly alarming percentage increase. If we add in the 2020 murders from Puerto Rico, the increase appears even more alarming. Regardless, these small increases can likely be explained by the increase in overall homicides in 2020, and the large increase in Latina victims can be explained by the addition of data from Puerto Rico in 2020.
What does it all mean??
The widely reported claim that trans people are suffering through an “epidemic of violence” due to an alarmingly high murder rate is partially true, but extremely misleading without further qualifying statements. Firstly, the reality is that only a very small subset of transgender people actually face any statistically abnormal risk. Recall that the word “transgender” includes a wide variety of people, including trans-men and non-binary or “gender queer” individuals. Trans-women make up approximately 36.1% of all trans people, and among trans-women, an even smaller percentage (approximately 11.03%) are black - meaning that in total, only 4% of transgender people face any abnormally high risk of being murdered. This amounts to approximately 61,721 people throughout all 50 states (and 62,372 if we include Puerto Rico) - a tiny minority within a minority.
The other 96% of trans people, consisting mostly of white trans-women, trans-men and non-binary individuals, face a similar or lesser risk of being murdered as the general population, and less of a risk of being murdered than cisgender males of the same race. Even black trans-women, who indeed face a highly disproportionate risk of being murdered compared with the general population, actually face less of a risk than black cisgender males (0.037% versus 0.054%).
Finally, the sharp increase in transgender murders reported in 2020 appears to be a result of including transgender murder victims from Puerto Rico in the annual HRC report for the first time, combined with the overall increase in homicides during 2020 and the “small number effect” (where even 1 or 2 additional victims will produce a seemingly alarming percentage increase). These factors enable attention-grabbing alarmist headlines reporting huge percentage increases in transgender murder victims, often misleading the reader to believe that unprecedented levels of transphobia have been unleashed upon America.58
Of course, it is certainly true that transgender people report higher levels of discrimination, abuse, and threats of violence than the general population.59 And while black trans-women face less of a risk of being murdered than cisgender black males, they face a much greater risk than cisgender black females, validating the intersectional claim that something about being black and male-to-female transgender draws violence in a statistically unique manner.
The disproportionately high risk of being murdered faced by black trans-women cannot entirely be explained by race or gender, whether we go by gender assigned at birth or preferred gender identity. It appears that among black people, “male-to-female trans-ness” produces a statistically unique pattern of risk. I can only speculate what causes this pattern, but one possible clue is found in a 2021 UCLA study on transgender women.60 This study found that only 8% of lesbian/bisexual trans-women (trans-women who are attracted to women) are black, whereas 26.7% of straight trans-women (trans-women attracted to men) are black. In contrast, 72% of lesbian/bisexual trans-women are white, whereas only 45% of straight trans-women are white. In other words, among straight trans-women, black people are over-represented and white people are under-represented. The cause of this non-uniform racial distribution is anyone’s guess, but it could help to explain (at least partially) the disproportionate risk of violence faced by black trans-women: for obvious cultural reasons, it is probably more dangerous for a trans-woman to seek (cisgender) male sexual partners than female sexual partners, particularly among poorer communities.
In any case, while black trans-women face a disproportionately high risk of murder, this does not justify making exaggerated claims about the safety of transgender people in general. Again, black trans-women make up only 4% of transgender adults in the US. The other 96% face a similar or lesser risk of being murdered as the general population. It is therefore highly misleading to imply that trans murder victims point towards a broader “epidemic of violence” afflicting trans people, when 96% of trans people face no abnormal risk of being murdered. Finally, it seems grossly disproportionate for society at large to spread fear about fatal violence towards some vague notion of a greater “transgender community”, just so that this 96% can vicariously adopt the “victim cred” of a tiny minority within a minority (the approximately 61,721 black trans-women in the US), all the while mostly ignoring the fact that adult cisgender black men face a substantially higher risk of being murdered than any trans person, and comprise over 16 million people.
https://www.cnn.com/2020/11/20/us/transgender-deaths-2020-trnd/index.html
https://www.nbcnews.com/feature/nbc-out/anti-trans-violence-surges-advocates-demand-policy-reform-n1260485
https://www.cbsnews.com/news/transgender-community-fatal-violence-spike/
https://www.nbcnews.com/feature/nbc-out/biden-calls-anti-trans-violence-epidemic-needs-national-leadership-n1243932
https://www.usatoday.com/story/news/nation/2020/10/14/transgender-murders-reach-least-32-year-surpassing-record/3639313001/
https://www.independent.co.uk/news/world/americas/philadelphia-murder-trans-violence-epidemic-b724964.html
https://www.insider.com/30-transgender-people-have-been-killed-in-2020-highest-record-2020-10
https://web.archive.org/web/20200401200950/https://www.wfxg.com/story/41944277/family-of-john-scott-devore-speaks-with-fox-54-after-richmond-county-sheriffs-office-calls-off-the-search
https://web.archive.org/web/20210131233055/https://www.wfxg.com/story/41911220/two-people-arrested-for-burglary-at-home-of-missing-person
https://www.nbcnews.com/news/latino/transgender-woman-killed-puerto-rico-after-using-women-s-bathroom-n1142661
Archived twitter link
These include:
Neulisa Luciano Ruiz: https://www.hrc.org/blog/hrc-mourns-neulisa-luciano-ruiz-trans-woman-killed-in-puerto-rico
Serena Angelique Velázquez Ramos and Layla Pelaez Sánchez: https://www.cbsnews.com/news/transgender-women-killed-puerto-rico-serena-angelique-velazquez-ramos-layla-pelaez-sanchez-men-detained/
https://www.justice.gov/opa/press-release/file/1273056/download
Selena Reyes-Hernandez: https://web.archive.org/web/20200625001805/https://www.chicagotribune.com/news/breaking/ct-homicide-arrest-061620-20200616-2jmtylvohngyjcm4ya7aowok7a-story.html?outputType=amp&__twitter_impression=true
Brayla Stone: https://web.archive.org/web/20210814122217/https://www.arkansasonline.com/news/2021/aug/14/man-says-he-killed-trans-girl-last-year/
Merci Mack: https://web.archive.org/web/20210324062633/https://www.wfaa.com/article/news/crime/man-faces-murder-charge-in-shooting-death-of-black-transgender-woman/287-0a49c212-9f20-451c-84f5-e2ae84cf2168
These include:
Scottlyn Devore: https://web.archive.org/web/20200401200950/https://www.wfxg.com/story/41944277/family-of-john-scott-devore-speaks-with-fox-54-after-richmond-county-sheriffs-office-calls-off-the-search
Riah Milton: https://web.archive.org/web/20210125101409/https://www.wlwt.com/article/sheriff-2-arrested-including-14-year-old-girl-in-liberty-township-slaying/32827639
Jayne Thompson: https://web.archive.org/web/20210120073933/https://www.denverpost.com/2020/05/11/man-shot-state-trooper-mesa-county-identified/
https://da.mesacounty.us/globalassets/district-attorney/cirt-letters/may-19-2020.pdf
Summer Taylor: https://www.cnn.com/2020/07/09/us/seattle-protesters-car-death-man-charged/index.html
https://www.hrc.org/news/hrc-mourns-alexandria-winchester-transgender-latina-killed-in-new-york
HRC has now tracked at least 44 deaths in 2020 of transgender and gender non-conforming people. We say “at least” because too often these deaths go unreported — or misreported.
Deadnaming by police and media can result in many transgender and gender non-conforming fatalities being unreported or reported inaccurately.
https://news.gallup.com/poll/329708/lgbt-identification-rises-latest-estimate.aspx
A 2016 UCLA study by Flores et al. estimates that 0.58% of the US adult population is transgender: https://williamsinstitute.law.ucla.edu/wp-content/uploads/Trans-Adults-US-Aug-2016.pdf
A 2017 study estimates 0.6%, matching the Gallup poll: http://williamsinstitute.law.ucla.edu/wp-content/uploads/Age-Trans-Individuals-Jan-2017.pdf
The Census Bureau estimates a total population of 331,449,281, and a total adult (ages 18 and over) population of 258,343,281 in 2020:
https://www.census.gov/programs-surveys/decennial-census/decade.2020.html
https://www2.census.gov/programs-surveys/decennial/2020/data/redistricting-supplementary-tables/redistricting-supplementary-table-package.pdf
https://data.census.gov/cedsci/all?y=2020&d=DEC%20Redistricting%20Data%20%28PL%2094-171%29
Raw data:
https://data.census.gov/cedsci/all?y=2020&d=DEC%20Redistricting%20Data%20%28PL%2094-171%29
Summarized tables:
https://www.census.gov/data/tables/2020/dec/2020-redistricting-supplementary-tables.html
https://www2.census.gov/programs-surveys/decennial/2020/data/redistricting-supplementary-tables/redistricting-supplementary-table-package.pdf
The FBI provides homicide data broken down by race as part of their “expanded homicide tables” package available via the FBI crime data explorer app. (Direct link to expanded homicide tables).
However, the expanded homicide table data is usually out of date and incomplete, and does not separate Hispanic white/black from non-Hispanic white/black. In my analysis I use the raw data contained in the master file Supplementary Homicide Report provided by the FBI at https://crime-data-explorer.fr.cloud.gov/pages/downloads. Click on “Master File Downloads” and select “Supplementary Homicide Report”.
The Supplementary Homicide Report is also incomplete, because it does not include homicide data for Florida (Florida does not participate in the FBI reporting program). Homicide data for Florida is provided separately by the State of Florida at https://www.fdle.state.fl.us/FSAC/Uniform-Crime-Report/Individual-Crime/Others/Supplemental-Homicide. (Direct link to 2020 Florida homicide data).
In my analysis, I combine the FBI Supplementary Homicide Report with the Florida Supplemental Homicide data.
The 2020 census data provides a more fine-grained racial break down than previous years, now including categories for mixed raced individuals. Additionally, both the Census Bureau and the FBI Uniform Crime Reporting (UCR) program consider Hispanic/Latino to refer to an ethnicity, not a race. This means there can be white Hispanics, black Hispanics, Asian Hispanics, etc. In my analysis, I use the “social justice” definition of “white” and “black”, where “white” usually excludes Hispanics or mixed-race people, and “black” follows the “one-drop” rule (meaning a black person can also be Hispanic or mixed-race). Therefore, when computing risk percentages for white people, I use the “White Alone” Non-Hispanic category from the 2020 census (meaning non-mixed race white people who do not identify as Hispanic), along with the non-Hispanic white category from the FBI Supplementary Homicide Report. (The census and FBI categories do not entirely overlap; the FBI data may include non-Hispanic whites who the FBI identified as white but are also mixed race, and would therefore not be classified as “White Alone” by the 2020 census. Unlike the 2020 census data, the FBI racial classification system does not accommodate mixed-race individuals in a non-ambiguous manner. This could have the effect of slightly inflating the risk percentages I calculate for white people.)
2020 Census data is available at: https://data.census.gov/cedsci/all?y=2020&d=DEC%20Redistricting%20Data%20%28PL%2094-171%29
The FBI Supplementary Homicide Report is available at https://crime-data-explorer.fr.cloud.gov/pages/downloads. Click on “Master File Downloads” and select “Supplementary Homicide Report”. Homicide data for Florida is not included in the FBI report, and must be obtained separately at http://www.fdle.state.fl.us/FSAC/Documents/Annual/Offense/Supp_Homicide_1996-2020.aspx
The homicide statistics used here are based on the FBI Supplementary Homicide Report, available at https://crime-data-explorer.fr.cloud.gov/pages/downloads. The Supplementary Homicide Report is more up to date and accurate than the “Expanded Homicide Tables”. FBI homicide data does not include data for the State of Florida. Homicide statistics for Florida must be obtained separately at http://www.fdle.state.fl.us/FSAC/Documents/Annual/Offense/Supp_Homicide_1996-2020.aspx
The figure of 18,194 adults murdered in 2020 is computed using the FBI Supplementary Homicide Report and the Supplemental Homicide data from the State of Florida.
See the 2020 census data:
Raw data:
https://data.census.gov/cedsci/all?y=2020&d=DEC%20Redistricting%20Data%20%28PL%2094-171%29
Summarized tables:
https://www.census.gov/data/tables/2020/dec/2020-redistricting-supplementary-tables.html
https://www2.census.gov/programs-surveys/decennial/2020/data/redistricting-supplementary-tables/redistricting-supplementary-table-package.pdf
Some examples:
A CNN article reports a spike in murders of “transgender and gender nonconforming people”, but does not mention this primarily affects black trans-women until the 5th paragraph: https://www.cnn.com/2020/11/20/us/transgender-deaths-2020-trnd/index.html
Similarly, an NBC affiliate reported on the murder of a trans-woman, quoting a spokesperson from an LGBT advocacy group as saying “we are also furious because every week there is a transwoman who is murdered across this country and it really needs to stop”. This implies at least that all trans-women are equally likely to be afflicted by this wave of violence.
https://www.nbc12.com/2020/11/24/richmond-police-continue-investigate-homicide-advocates-call-change/
A CBS News article reporting the murder of a trans-woman repeats the phrase “epidemic of anti-LGBTQ violence”, and quotes a transgender activist saying “they are hunting us and they are killing us”. The article implies all trans-women are equally likely to be afflicted by this violence.
https://www.cbsnews.com/news/transgender-women-killed-puerto-rico-serena-angelique-velazquez-ramos-layla-pelaez-sanchez-men-detained/
An ACLU article includes the phrase “Addressing An Epidemic of Violence Against Trans and Non-Binary People” in the title. In the second paragraph, it mentions that this “especially [affects] Black trans and non-binary communities”, which (1) still implies non-black trans people are more harshly affected than the general population, and (2) incorrectly states that non-binary people are affected more than the general population.
https://www.aclu.org/news/lgbtq-rights/remembrance-resilience-and-response-addressing-an-epidemic-of-violence-against-trans-non-binary-people/
See https://www.hrc.org/news/hrc-mourns-alexandria-winchester-transgender-latina-killed-in-new-york
It is clear that fatal violence disproportionately affects transgender women of color, especially Black transgender women. The intersections of racism, transphobia, sexism, biphobia and homophobia conspire to deprive them of necessities to live and thrive, so we must all work together to cultivate acceptance, reject hate and end stigma for everyone in the trans and gender non-conforming community.
https://www.hrc.org/news/hrc-mourns-alexandria-winchester-transgender-latina-killed-in-new-york
HRC has now tracked at least 44 deaths in 2020 of transgender and gender non-conforming people. We say “at least” because too often these deaths go unreported — or misreported.
Deadnaming by police and media can result in many transgender and gender non-conforming fatalities being unreported or reported inaccurately.
See the following examples:
https://web.archive.org/web/20200725021236/https://www.thedesertreview.com/news/brawley-police-investigate-homicide/article_54d3abaa-c6b5-11ea-b7bf-f715d7c5c850.html
https://web.archive.org/web/20200506195301/https://www.kfvs12.com/2020/05/04/semo-major-case-squad-activated-sikeston-death-investigation/
https://web.archive.org/web/20210117085908/https://www.katc.com/news/lafayette-parish/victim-of-wednesday-shooting-on-evangeline-thruway-has-died
See my analysis of the 23 deadnamed or misgendered victims identified in the Media Matters report: https://docs.google.com/document/d/e/2PACX-1vSMWrD60PFfKKQomwkj6v7FOStosCU7eywSeE-vDj-y5HFK193VOkGdIPFp35QgojXkEM24iQqKCxHH/pub
See for example this case in New York, where the victim’s identity was determined by interviewing witnesses or people from the surrounding neighborhood: https://web.archive.org/web/20210120232100/https://www.nydailynews.com/new-york/nyc-crime/ny-woman-found-knifed-to-death-in-harlem-20200328-rlrv4welp5cipglo5pfmrmaxii-story.html?fbclid=IwAR0uUzYtmG-qvwjGKMIOMvKR1SWpJBH_IAk_9kg9HleDgK5loBFOgGSgxFo
See also the extremely tragic case of Neulisa Luciano Ruiz, also known as Alexa Negrón Luciano, a homeless woman who very likely did not carry ID with her preferred name: https://web.archive.org/web/20210205121928/https://www.nytimes.com/2020/02/26/us/puerto-rico-alexa-transgender-killing.html
Examples:
https://web.archive.org/web/20200725154201/https://www.kark.com/news/local-news/sherwood-police-investigating-homicide-off-of-gap-creek-drive/
https://web.archive.org/web/20200728194520/https://www.nydailynews.com/new-york/nyc-crime/ny-bronx-suspect-stabbing-death-transgender-woman-20200727-vsazb5iwcjdprokdmrqltgvstm-story.html
https://web.archive.org/web/20200702084302/https://www.advocate.com/crime/2020/6/29/brayla-stone-17-year-old-transgender-girl-found-dead-little-rock
The FBI provides a count of total annual homicide victims in Puerto Rico at https://crime-data-explorer.app.cloud.gov/pages/downloads (under Additional Datasets > US Territory Data), but does not break this down by age, race or gender.
Using demographic data for Puerto Rico from the 2020 census and a breakdown of homicides in the 50 US states by age, race and gender provided by the FBI, I use Bayesian ridge regression to estimate the racial/gender breakdown of homicide victims in Puerto Rico in 2020. My estimate is:
Hispanic Black: 362, M=313, F=49
Non-Hispanic Black: 8, M=5, F=3
Hispanic White: 196, M=165, F=31
Non-Hispanic White: 11, M=6, F=5
Hispanic Asian: 2, M=0, F=2
Non-Hispanic Asian: 12, M=5, F=7
Hispanic Indigenous: 13, M=11, F=2
Non-Hispanic Indigenous: 2, M=1, F=1
Total: 606, M=506, F=100
https://www.nbcnews.com/feature/nbc-out/killings-transgender-americans-reach-all-time-high-rights-group-says-n1242417
https://www.cbsnews.com/news/transgender-community-fatal-violence-spike/
https://www.cnn.com/2020/11/20/us/transgender-deaths-2020-trnd/index.html
https://www.hrc.org/resources/violence-against-the-trans-and-gender-non-conforming-community-in-2020
HRC annual reports on transgender murder victims (2015 to 2020):
https://www.hrc.org/resources/violence-against-the-trans-and-gender-non-conforming-community-in-2020
https://www.hrc.org/resources/a-national-epidemic-fatal-anti-trans-violence-in-the-united-states-in-2019
https://www.hrc.org/resources/a-national-epidemic-fatal-anti-transgender-violence-in-america-in-2018
http://assets2.hrc.org/files/assets/resources/A_Time_To_Act_2017_REV3.pdf?_ga=2.24739770.1039888331.1637318740-325184809.1634928372
https://assets2.hrc.org/files/assets/resources/A-Matter-of-Life-and-Death-2016.pdf?_ga=2.24739770.1039888331.1637318740-325184809.1634928372
http://assets2.hrc.org/files/assets/resources/HRC-AntiTransgenderViolence-0519.pdf?_ga=2.24739770.1039888331.1637318740-325184809.1634928372
HRC report on transgender murder victims so far in 2021:
https://www.hrc.org/resources/fatal-violence-against-the-transgender-and-gender-non-conforming-community-in-2021
The victim totals here use the FBI Supplementary Homicide Report (SHR) data for each year (2015 thru 2020), since it is more accurate than the “Expanded Homicide Tables”. The SHR can be found at https://crime-data-explorer.fr.cloud.gov/pages/downloads. Click “Master File Downloads” and select “Supplementary Homicide Report”.
The victim totals are also supplemented with Florida SHR data. The FBI SHR does not include homicide data for Florida, so Florida homicide data is obtained separately at http://www.fdle.state.fl.us/FSAC/CJAB-Home/Uniform-Crime-Report/Data-Archives/UCR-Offense-Data