Reporting
Down for the Count: Women Veterans Likely Underestimated in Federal Homelessness Figures
“Homeless women are something of a sociological mystery,” wrote University of Virginia social scientist Ted Caplow, Ph.D. While he wasn’t talking about homeless women veterans, he might as well have been.
As a phenomenon, they manage to be starkly invisible – rarely included in discussions about homeless veterans as a whole, ignored in the popular media and under-researched in the academic literature. Further adding to their marginalized status is the fact it seems almost impossible to come up with a reasonably accurate count of how many homeless women veterans there actually are. For that count, the federal government relies on several notoriously imprecise methods that are particularly inaccurate when attempting to estimate this population, because of behaviors and preferences that leave them largely uncounted. Yet being accurately represented in annual federal estimates is essential for homeless women veterans – because everything from funding to resources to services that address this unique population depends on it. Being undercounted – or uncounted – perpetuates the very invisibility of these women, the fastest growing segment of homeless veterans today.
And the numbers don’t really add up. If you ask a handful of experts, you get a handful of different estimates, from smaller to larger, with no real sense of which one – if any – is most likely to be accurate. (That there is a range is obvious, because women veterans are so easily overlooked in a count of the visibly homeless.)
The U.S. Department of Housing and Urban Development (HUD), which is responsible for the federal count, said in 2016 that there were 3,328 women veterans who were homeless – out of a population of more than two million women veterans alive today, according to their 2016 Annual Homeless Assessment Report to Congress. (The number of women veterans will continue to increase, even as the overall population of veterans declines over the next several decades, according to VA.) But the figure for homeless women veterans? That figure works out to represent fewer than two-tenths of one percent of the women veterans alive today in the U.S.. Is that figure reasonable?
Separately, when more than 1,600 women veterans from every era and every branch of service were recently asked about periods of unstable housing or homelessness of a week or more after military service, fully one-half answered that they had spent at least a week or more at risk for – or actually – homeless.
And figures published by the U.S. Department of Veterans Affairs (VA) over the past few years in research journal articles as well as estimates proposed by other subject matter experts offer a range of other estimates.
An every-ten-years census was established in the Constitution, but the census did not ask men about their veteran status until 1840, and women not until (cough, cough) 1980. Separately, the first major study of homelessness was done around the turn of the 20th century, and obviously focused on men. Public attention to homelessness waxed and waned, becoming stronger in the 1980s, which is also when some of the first important research about male veterans and homelessness was produced. Five years ago, according to VA researchers, a complete scan of the research literature revealed only two studies about female veteran homelessness, although there were many about male veterans. According to the U.S. Department of Housing and Urban Development (HUD), which collects the data and creates the crucial estimates of veteran homelessness, it was not until 2013 that they started collecting data about women veterans who were homeless for the first time. So asking women about their veteran status, doing research on female veteran homelessness, and/or estimating how many women veterans homelessness affects are all relatively new developments.
Now back to why are there several difference estimates in play about how many women veterans are homeless.
Part of the problem is that homeless veterans are notoriously hard to count – especially female veterans who are homeless. Because of the frequency with which they experienced trauma, especially military sexual assault, and/or because of their status often as single mothers of dependent children, women veterans who are at risk for homelessness very often choose doubling up with family or friends, a practice known as “couch-surfing,” which the federal government a few years ago eliminated from their definition of homelessness. Sadly, this one change profoundly affected the status of women veterans much more so than men, because of how frequently this status precedes seemingly more dire or drastic forms of homelessness.
To recap material from previous articles in this series about women veterans and homelessness, the all-important federal count, administered by HUD, comes from two main sources: first, from an annual, single-night “point in time” count performed by volunteers at cities across the country, during which unsheltered people, sleeping outdoors, are eyeballed and added to the count, usually without verbal confirmation. (“Veterans” among this count are guessed at, in part by their attire: Are they wearing anything camouflage? Could be a veteran. Or they could have just picked up that attire at the local Goodwill store.) Separately, a count is made of those who move through services designed to support the homeless, such as soup kitchens and shelters, over the course of the year – and these two counts are combined to come up with an overall count of how many homeless there are, and furthermore, how many homeless veterans.
Clearly, there are multiple problems with this count. One is functional: women veterans, whether due to their trauma histories from military sexual assault, also known as military sexual trauma (MST) or from their status often as single parents with dependent children in tow, are unlikely to sleep outdoors or stay in shelters because of the risk of violence or being separated from their kids. The other problem is philosophical: the count has always been considered imprecise for reasons longtime homelessness expert Kim Hopper laid out in Reckoning with Homelessness:
“Any attempt to arrive at an accurate number of homeless people in a given area is subject to a host of difficulties. The estimates that surface from time to time have proven notoriously unreliable, subject to wild discrepancies depending on methods of estimation used, sources relied on, the season of the year, and (it may not be too cynical to suggest) the intended purpose of the count.“
In other words, the results could not have been more different.
Original Research – Survey I
Of the 400 respondents to the earlier survey administered in 2015 whose results have been analyzed, more than seven in 10 reported periods of homelessness after military service. Of those, more than one in four (26.3 percent) reported being homeless for at least a year, 17.5 percent between six months and a year, 14.8 percent between one month and three months, and 12.3 percent between three and six months. Across every age range 65 years old or younger, more than 40 percent of women veterans surveyed affirmed being homeless for at least a month or more, with more than one in every five reporting having been homeless for more than a year.
And this experience of homelessness after military service was not confined to women veterans from any particular era. More than a third of Gulf War I-era respondents reported being homeless for a year or longer, followed by a quarter of Vietnam-era and peacetime veterans (the largest group), and more than one in five of post-9/11 veterans.
Perhaps most surprisingly, even among women veterans who had served for 20 or more years in the military, only slightly more than half (52.3 percent) said they experienced no problems, with almost one in five (18.5 percent) reporting periods of homelessness for a year or longer.
Original Research – Survey II
In the current, still ongoing survey of those respondents reporting homelessness, almost one in every three (31.75 percent) reported having been “continually homeless” – meaning without a break of at least seven days between episodes – for a year or more. Additionally, respondents were asked if they had been homeless “four or more times within the past three years, which when added together would equal 12 months or more.” While more than half (53 percent) said they had not, more than one every in ten said yes; while another six percent had been homeless at least four times but for less than 12 months, and three in 10 had been homeless during the past three years, but not for the volume of time specified.
When you consider a topic like female veteran homelessness and the seemingly resounding lack of attention, interest and resources that so far with few exceptions have been devoted to it specifically, you do start to wonder if this lack of urgency has anything to do with an underlying suspicion it affects so few people that it’s either not of particular concern or that it’s OK to mostly overlook it until proven otherwise.
The latest federal (HUD) estimate of 3,328 homeless women veterans is a mere 0.162% of the total population of 2,051,484 women veterans in the United States, as calculated by VA. Now, slightly over three thousand anythings is a fairly low number – maybe the number of people hospitalized in a year for stepping on Legos, I would jest. Is this very low figure at least partially to blame for the disconnect whereby most Americans, even most veterans, have never heard of female veteran homelessness and associate a homeless veteran mostly with older white males who panhandle at intersections, or sleep under bridges and highway overpasses?
Ironically, even if you ask women veterans – including those who have been homeless themselves – to describe a homeless veteran, that stereotype is what you hear; the image is that rooted in our collective national consciousness. But if you ask women veterans instead whether they’ve ever lacked permanent of stable housing since they’ve left the military, you learn how prevalent this experience can be. About half the women veterans who responded to the two surveys I’ve conducted since 2015 report at least a week of homelessness – and often not just months, but years.
So here we have the crux of the dilemma: Is female veteran homelessness ultra-rare – as the HUD figures would indicate, or much more common – as the independent research among women veterans indicates?
My own introduction to the problem of female veteran homelessness came about ten years ago, in conversations with a well-known military sexual trauma (MST) advocate, Susan Avila-Smith, who was describing how few of the MST survivors she knew among women veterans were able to maintain gainful employment. She described how only a handful of the more than 5,000 MST survivors whose claims she had helped settle, pro bono, were able to work, and consequently, many of these traumatized survivors were staying with friends and relatives, sleeping on living room floors or couches, or if they were lucky, in someone’s spare room, a phenomenon known as “couch-surfing,” or more technically, “doubling up.” She told me that she’d tell women veterans who were couch-surfing that according to the U.S. Department of Veterans Affairs (VA) they were homeless, and to get to VA and avail themselves of the resources and services intended for homeless veterans. It was always a bit difficult for Avila-Smith to convince women veterans that they were homeless, she said, but it was important because the services VA could provide were important for them to access, and would help them get back on their feet.
So women veterans are: a) unlikely to be counted in federal estimates of the homeless; b) frequently choose a method of accommodating unstable housing that the federal government excludes from its definition of homelessness; and c) in general keep such a low profile on this topic that few people, including other veterans, even realize there are homeless women veterans – let alone that homelessness among women veterans is much more common than previously understood. And we know that from VA’s own research literature. We’ll get to that in a minute, but first some backstory about where the various estimates come from.
The lowest number in the range of estimates, as we’ve mentioned, is HUD’s current 2016 official estimate of fewer than 3,400 homeless women veterans – out of a population of more than two million women veterans. But other estimates have also been proposed.
In 2014, two of the key half-dozen or so principal investigators for VA on homelessness among women veterans, Ann Elizabeth Montgomery, Ph.D. and Thomas Byrne, Ph.D., writing in an article published in Military Medicine (2014) had cited a higher figure, along with their description of why they used this estimate:
“Women veterans experiencing homelessness comprise a discrete population, estimated to number around 14,000 persons over the course of a year.”
“Although a fairly small population,” compared with a larger population of male veterans who are homeless, they continued, women veterans are “over-represented within the homeless population” and apparently become homeless at a rate faster than male veterans. Additionally, citing recent reports from the U.S. Department of Housing and Urban Development (HUD) and the Government Account Office (GAO) as references they wrote:
“The number of female Veterans experiencing homelessness is likely to rise as the overall female veteran population is projected to increase by 17% over the next two decades.”
Then in 2015, retired Army Col. Carl Castro, Ph.D., Anthony Hassan, Ed.D., and Suzanne Wenzel, Ph.D., in their white paper, ” Call to Action: Toward Ending Female Veteran Homelessness,” introduced a slightly higher estimate. They wrote:
“There are an estimated 17,000 female homeless veterans, many with children who share in their homeless situation.”
And, in 2016, frustrated with what she saw as unrealistically low numbers, and knowing that she’d seen several women reservists, mothers of children, deploy to the war zone from her transitional housing program, Final Salute, Inc. in Washington, DC, Army Reserve Maj. Jas Boothe decided to come up with some figures of her own. She took a look at U.S. Department of Labor (DOL) statistics about women veterans and homelessness, because they also track the issue, and from the figures she observed, created a white paper and an infographic that both asked the question:
“Are we to assume that 18,622 women veterans, most of them with children, are able to maintain housing and care for their children – without employment?”
So now we have a range of figures, provided by various sources, estimating that the number of homeless women veterans in the course of a year may be as few as well under 4,000 to close to 19,000. But what if we tossed out all the estimates and just calculated back into the known population of women veterans in the U.S., how many were likely to be homeless, given relationships that have already been established in the research literature?
Although various factors are known to influence women veterans becoming homeless including unemployment, poverty, physical disability, mental health disorders such as anxiety or post-traumatic stress disorder, and a history of military sexual trauma (MST), according to various sources in the peer-reviewed research literature, several have prevalence factors associated with them. From a systematic review of the literature on homeless female veterans, VA researchers have estimated that at any one time between one and two percent of all women veterans are homeless – or that 13 to 15 percent of women veterans living in poverty (separately estimated to be ten percent of all women veterans) are homeless. Even without looking at any other factors that influence risk of homelessness, that leaves us with two potential models for calculating a more accurate number of homeless women veterans.
- In the first model, between 20,515 (one percent) and 41,030 (two percent) of women veterans would be considered homeless.
- In the second model, between 13 percent and 15 percent of the 10 percent of women veterans who live in poverty would be considered homeless, or between 26,660 (13 percent) and 30,773 (15 percent) of the 205,148 (10 percent) women veterans living in poverty.
Clearly, these figures would not be additive – many women veterans who are homeless suffer from poverty as well as other conditions – but if we supposed they were, just hypothetically, that would mean that somewhere between almost 30,000 and more than 70,000 women veterans were homeless, without accounting for any other factors. Compare this to the extremely low figure arrived at by HUD’s count, 3,328, which presumably mostly misses women veterans precariously housed or homelessness who are not sleeping outdoors (and found by volunteer counters, self-identify as women veterans and are finally confirmed to be women veterans by volunteer counters – unlikely) or staying in shelters or using other services designated for the homeless over the course of a year (there are many reasons this is an unlikely way as well to encounter many homeless women veterans, other than the chronically homeless). The figure represents less than two-tenths of one percent of all the women veterans in America. As a contrast, there are likely to be far more homeless women veterans in just the top five most populous states for women veterans – estimated using either of the models described earlier.
(For a more in-depth look at estimates of homeless women veterans in all 50 states based on just a single risk favor – poverty – see the data visualization below, produced by the author using the most recent VA figures available for women veterans by state.)
As we’ve said in many and various ways here in this series, the all-important count and the inclusion of or exclusion from the federal definition of homelessness for women veterans according to how they experience homelessness are essential to how women veterans in need can ensure that adequate funding, research, and resources – including suitable housing – are available when they need them. Gratifyingly, from conversations this past week with both VA and HUD, it appears that even if the count is unreliable, the mandate to “end veteran homelessness, period” is how both agencies understand their mission, according to HUD Deputy Assistant Secretary for Special Needs, Ann Oliva.
While different programs exist to address veteran homelessness, the question is whether they have sufficient capacity, infrastructure and/or funding to serve all the women veterans they need to. In this series, we’ve been taking a look at “just” women veterans who are homeless, because they’ve rarely benefited from research or resources matched to their specific needs, such as an emphasis on trauma-informed housing since many are survivors of sexual assault. And as the population of women veterans varies by state, so too does the amount of the available resources such as emergency shelters, transitional housing, permanent supportive housing that are matched to this population. Additionally, poverty rates vary from state to state, so while we use a standard rate of 10 percent of women veterans as living in poverty, per VA research, it’s also likely that economic conditions are worse and poverty is higher for women veterans in some states and lower in better/lower in others. (There’s also a yet-fairly-unexplored difference in urban vs. rural for where male and female veterans who are homeless live. Between 2009 and 2015, according to HUD figures, the percentage of male veterans moving away from rural areas to cities is growing; in female veterans that trend is reversing, and more women veterans are living further away from urban centers.)
and then in a data visualization that’s interactive (you can click on it and learn the details about the state) below.
Want a closer look? See how that looks for just the top five states by women veteran population:
I was compelled to write this article out of a desire to replace the too-low estimate of women veterans who are homeless, which I feared would keep being repeated otherwise, with what I hoped would be at least one way of looking at the same problem, which could potential produce more reasonable numbers, based on what I was finding from the survey results – just how common and almost “ordinary” women veterans’ frequent experiences with unstable housing and homelessness were. I was very much operating out of a belief that “it all depends on the count.” And in some ways, it does.
But what was tremendously reassuring to learn nonetheless from reporting completed this week and last was that high up within both VA and HUD there is an awareness that the numbers are really just a start – and that the mandate to end homelessness among veterans, whether those homeless veterans are counted and identified, or somehow evade the count completely – is their dedicated mission.
The discussion of how women veterans who are homeless and are uncounted or undercounted “is an important one,” says VA researcher Ann Elizabeth Montgomery, Ph.D., whose work was mentioned earlier. But to solve the problem, she added, “we take data from multiple sources and extrapolate it” when it comes to policy. “We combine the lessons learned and the epidemiological research,” she said. According to her, the HUD number is most useful as an indicator of is the overall number of homeless veterans heading up or down, from the previous year. But VA also shares its own experience of how many homeless veterans are using services it provides, from health care through housing, to inform policy discussions about funding and other issues.
Over at HUD, the sense was the same. “We don’t conflate veterans being counted as homeless with eligibility for homeless services,” said HUD’s Ann Oliva. “There’s a recognition that this is a population that needs to be served,” she stated, adding that what may be needed are more creative ways to do outreach to the populations like women veterans who aren’t well-represented in the count.
Far from public view, but awarded a variety of “best-in-government” awards in recognition of their efforts, HUD and VA are working together, to share the best information they have, along with best practices and a sense of the next steps needed, to reduce and eliminate veteran homelessness. This isn’t something the casual observer would be aware of; in fact, it was only two years into my own investigation of this subject that I learned about it, and then only because Keith Harris, Ph.D., the acting executive director of the VA’s Homeless Program Office, was kind enough to suggest that I might be surprised to learn that HUD was already aware of the limitations of the count, and that they were working in close partnership to address the gaps together.
Some of the most evocative comments from the current survey of women veterans, which has had an almost 600% increase in responses, have been turned into awareness-raising “memes” and circulated on Facebook, Twitter, Instagram and even Pinterest. You can see a whole album of these memes here, each of which is also linked to the actual survey:
As far as going forward goes, several improvements suggest themselves regarding attaining a more accurate count of how many women veterans are experiencing precarious housing and homelessness.
One would be a concept like predictive modeling, which dips into population estimates such as the one for women veterans and assigns numerical values to individual risk factors – the ones that have so far been determined to be associated with homelessness in women veterans – could go far towards providing a more likely to be accurate, nuanced and sophisticated sense of how many women veterans are at risk for homelessness in any given state. Barring that, looking harder at poverty levels of women veterans by state – since there will be differences there as well – and connecting that back to estimates of women veterans who may become homeless will also be valuable. But we do know about risk factors other than poverty which exacerbate the chance women veterans may become homeless. One of the more recent important associations was military sexual trauma (MST), which JAMA Psychiatry reported last year was independently associated with homelessness in both male and female veterans at all the intervals they surveyed: at one month, at one year and at five years. Adding MST to other known risk factors for female veteran homelessness such as poverty could likely produce much more realistic/accurate estimates, which in turn could lead to more appropriate levels of funding, service provision and housing options – including trauma-informed housing – in which homeless women veterans have been shown to thrive.
(It might also be beneficial from a research perspective to examine any possible connection between ACEs scores and housing instability for women veterans after service, given the trauma connection. You can learn more about the Adverse Childhood Experiences Study (ACEs) here.)
Going back to 2009, then-President Obama and then-VA Secretary Eric Shinseki proposed the radical plan of completely “eradicating veteran homelessness” by the end of 2015. The deadline was extended a year and later discarded entirely. But in the meantime, great progress was made. During that time period, homelessness among veterans declined by 46 percent, or 33,896 veterans overall, according to HUD figures. It also rose among women veterans, but the manner in which the data was being captured I feared would leave most women who were not not “visibly” (aka, chronically) homeless out of the count.I’ve since come to realize that problems with the count – how women veterans are estimated, in manners unlikely to find them, and the rollback of couch-surfing, a prevalent way women veterans accommodate periods of unstable housing, from the federal definition – are even more troubling. But from conversations last week and this week with HUD and VA, I’m convinced that they too know the issues with missing these women veterans who are homeless, and the need to take steps to address this, as well as to continue to offer appropriate services, based on the objective of the mission, not a pure reliance on the numbers themselves.
So, from the research literature and from the anecdotal, lived experience of women veterans who were interviewed for this “Coming Out of the Shadows: Women Veterans and Homelessness” series and their advocates, several key concerns emerge:
- Women veterans continue to be under-counted and under-represented in the current estimates of homeless veterans, perhaps severely. There are under-explored gender differences in how male and female veterans accommodate periods of unstable or precarious housing. Because of several factors, including experience of military sexual trauma (MST) and their frequent status as single mothers of dependent children, women veterans frequently stay off the grid and away from places (sleeping outdoors, staying in shelters, etc.) that might allow them to be counted in greater numbers as part of these conventional counts.
- Women veterans, like their civilian female counterparts, frequently choose “couch-surfing” during periods of functional homelessness. When the federal definition of homelessness changed several years ago, the language excluded doubling up, disproportionately affecting women veterans.
- Women veterans struggle with feeling invisible in a system not set up to recognize them. This invisibility takes several forms, from failing to identify themselves as veterans, out of a sense that the term itself has been co-opted by male veterans, or, in the case of women veterans who served in previous eras, being told that they were not eligible for VA healthcare services when they left the military. For many older women veterans, a gap of time existed between leaving the military and realizing that they were eligible to access VA benefits and services.
- Many estimates of how many women veterans are homeless are calculated in part by VA recording how many women veterans they see who affirm unstable housing. We’ve earlier addressed some issues with that model, including women veterans who become homeless apparently having difficulty, as a group, with predicting that in advance, and VA’s lack of a consistent coding system for recording homelessness among its patients. A larger issue is of course that not all women veterans use VA for healthcare, and so VA figures on how many homeless women veterans they encounter represent only a portion – no guesses as to how large a proportion – of how many women veterans actually are homeless.
- The trend in recent years to emphasize reduction and elimination of “chronic homelessness” among veterans – the most visible form of homelessness, characterized by many years of relatively intractable long-term/chronic mental health and/or substance abuse issues – also fails to reflect a common lived experience of women veterans who are homeless. It’s not unreasonable to suggest that here, too, this model disproportionately reinforces the stereotype of older males as the face of veteran homelessness and excludes a growing population of younger women veterans, with and without children.
- Finally, most official figures for veteran homelessness do not readily distinguish between male and female veterans, and so the fact that female veteran homelessness may actually be on the rise while male and overall veteran homelessness is falling is obscured by the lack of attention paid to women veterans as a separate population. When greeted with good news that veteran homelessness is falling, unless we really have reason to be more aware of the subject, we’re liable to assume the best and move on – rather than realizing there’s actually a problem here that’s poorly understood, under-researched and under-resourced, and likely to getting worse.
Because everything about veteran homelessness – the funding to reduce it, the number and types of housing and programs to accommodate it – starts with and at least provisionally relies on an estimate of how many women veterans a community can expect to be homeless, with or without their children, it’s essential that we take steps to improve this estimate, to bring it more in line with reality. We can do this by estimating back into the population of women veterans how many we can assume are likely to be homeless, based on known associations with various risk factors. We can do this by separating out, finally and fully, this population away from male veterans who experience homelessness as its own distinct population, with unique needs and challenges. And we can do this by purposefully finding women veterans and asking them about their lived experiences with homelessness and unstable housing. (If you’re a woman veteran, or an advocate who can get this survey to women veterans in your life, by email or via your social network, you can take or share this brief survey right now about your own experiences with unstable housing after military service.)
We can all be part of the path forward, helping women veterans who are homeless or struggle with being at risk for homelessness, be brought into the picture and finally “seen.” They also served. They shouldn’t have to be invisible.
NOTE: This article is the final element in the series entitled Coming Out of the Shadows: Women Veterans and Homelessness. The reporting for the series was conducted under a Howard G. Buffett inaugural grant from the International Women’s Media Foundation. The articles include ” G.I. Jane Needs a Place to Sleep,” ” Into the Gap: Women Veterans Describe Homelessness,” ” Camaraderie Offsets Trauma for Women Veterans,” ” The Path Home for Women Veterans,” and ” Homeless Women Veterans Struggle to Be Seen.” There is also an interactive timeline of how we got to this point, a radio show/podcast featuring the author, Lily Casura, as well as veteran advocates BriGette McCoy and Rosie Palfy talking about female veteran homelessness, and a data visualization showing the comments of 400 women veterans describing their experiences of unstable housing after military service. Finally, there’s a website of state-specific housing resources for women veterans at risk for homelessness, where you can recommend and/or suggest additional resources.
This post was published on the now-closed HuffPost Contributor platform. Contributors control their own work and posted freely to our site. If you need to flag this entry as abusive, send us an email.