EMBARGOED FOR RELEASE/SHARING UNTIL 2 PM, MONDAY, JUNE 19
Memorandum

Date: June 19, 2017
Re:

2017 Point in Time Count

To:

A Home for Everyone Executive Committee

From: Marc Jolin, Director, Joint Office of Homeless Services
___________________________________________________________________

I.

Introduction

The U.S. Department of Housing and Urban Development (HUD) requires all communities to
count people living in emergency shelters, transitional housing or places not meant for human
habitation (collectively known as “HUD Homeless”) on a single night (a “point in time” count) at
least once every two years. Portland and Multnomah County have thus far conducted the count
every other year, including this year; our intention is to conduct a count annually going forward.
This enumeration is one way of understanding the levels and trends in unmet need for shelter
and permanent housing within the community. The method is inherently limited, even to
understand unmet need, and importantly it does not answer the critical question of why we are
seeing the need levels and trends we’re seeing. To answer those questions we have to look to
other data sources.
HUD sets the date for the Point in Time (PIT) count as the last Wednesday in January, unless
unforeseen circumstances require a community to delay the date. Because of severe weather in
January 2017, HUD granted our community’s request to conduct this year’s PIT count the last
Wednesday in February. The PIT count was overseen by the Joint Office of Homeless Services
and implemented in partnership with Portland State University’s Population Research Center
and Survey Research Lab (PSU).
PSU is preparing the full 2017 PIT Report, and we expect that to be available in early July.
Understanding this Committee’s and the public’s desire to see key results from the 2017 PIT
count as soon as possible, and given that other cities around the country that conducted their
counts in January are now releasing data, we have prepared this Memorandum with some of
the essential, high-level PIT count data. Much of the more specific information, and information
requiring more in depth analysis, will be available with the full report.

 II.

Methodology

HUD sets broad guidelines and definitions for PIT counts, but does not prescribe a uniform
methodology for conducting the PIT count. Over the past decade, our community has used a
consistent and comprehensive methodology to maximize accuracy and integrity over time.
The PIT count is comprised of two elements. Information about residents in shelter and
transitional housing on the night of the count (the “shelter count”) is retrieved from the
community’s social services data system, the Homeless Management Information System
(HMIS), and through resident surveys. Information about the unsheltered population (the “street
count”) is collected primarily through the in-person administration of surveys, and to a much
lesser extent from client data in HMIS.
Street count surveys are administered by professional outreach workers and volunteers over the
course of the week that includes the official night of the count. This year, surveys were
administered at more than 120 sites providing services to people experiencing homelessness,
including meal programs, day centers, libraries, and clinics. They are also administered directly
in encampments, in parks, in vehicles, and on sidewalks. Enough information is collected on
each respondent to ensure we don’t double-count respondents. Special attention is paid to
working with agencies and volunteers that can help us ensure communities of color and
immigrant and refugee communities are being counted as accurately as possible; surveys are
available in six languages.
Like any point in time count methodology, ours has limitations. These will be spelled out in much
greater detail in the final 2017 PIT Report, and the numbers in this Memorandum should be
used with caution until all the critical methodological caveats are understood. Some of the key
limitations include:
(a)
Because of the inherent difficulty of counting people experiencing unsheltered
homelessness, we are confident, as we have been in all previous counts, that the street
count numbers reported here are an undercount. This is particularly true for some
communities of color, immigrant and refugee communities, unaccompanied youth and
for families with children. Because the same street count methodology was used this
year as in years past, we have no reason to believe that the overall rate of undercount is
any higher or lower this year.
(b)
HMIS does not contain information responsive to all the questions on the street
count survey form. Some of the data we report, therefore, describes only the street
homeless population, not necessarily the HUD Homeless population as a whole. Where
this Memorandum presents such data, this limitation is called out.

 (c)
Responding to questions on the street count survey is voluntary. And, in some
cases, street count data is pulled from HMIS as well, where certain information is not
available. Therefore, some questions have high non-response rates, which makes
drawing conclusions and certain year-over-year comparisons about the street homeless
population challenging. Because these questions require more careful and in-depth
analysis, they are not presented in this Memorandum but will be in the full 2017 PIT
Report.
(d)
HUD’s definition of homelessness does not include people who are involuntarily
doubled up, sleeping on couches, in garages, or in overcrowded situations. While we
lack comprehensive data on this population, we know that families with children and
people of color make up a higher percentage of this population of people experiencing
homelessness. The data presented here pertain only to the HUD Homeless population.
III.

Declining Unsheltered Population, Increased HUD Homeless Population

The number of people counted sleeping outdoors or in places not meant for human habitation
(e.g. vehicles and abandoned buildings) in Multnomah County on Wednesday, February 22,
2017, was 1,668. The number of people living in emergency shelter and transitional housing on
that night was 2,509. This represents an 11.6% decrease in the unsheltered population since
2015, and a 31% increase in the sheltered population. The increase in the sheltered population
reflects the fact that while we continued to convert transitional housing units to permanent
housing options, we have roughly twice as many publicly funded emergency shelter beds
available in 2017. Taken together, the HUD Homeless population increased by 376 individuals
to 4,177 people, a 9.9% increase over 2015.
The unsheltered population count is the lowest it’s been since 2009 (when it was 1,591). The
decline in the count of unsheltered individuals, despite ongoing challenges such as rapidly
increasing housing costs and stagnant incomes for low-income households, likely reflects our
community’s significant expansion of prevention, housing placement, and emergency shelter
capacity over the past two years. Because of the expansion of emergency shelter, the number
of people counted in shelter beds increased from 872 in 2015 to 1,752 in 2017, a 100.9%
increase.
It is important to note that HUD’s definition of “emergency shelter” continues to require us to
count the people sleeping at Dignity Village, Right 2 Dream Too, and Hazelnut Grove as
“unsheltered,” even though locally we have recognized that these “villages” can be an
appropriate emergency shelter solution if adequately planned and supported. Including the
residents of these villages in our count as sheltered would reduce the unsheltered count and
increase the emergency shelter count by approximately 90 individuals.
Within these overall numbers, we see important trends in specific populations:

 (a) Families: ​Our community has made a commitment not to turn any family away from
shelter, and has significantly expanded family shelter capacity over the past two years.
This may explain why the trend from unsheltered homelessness toward sheltered
homelessness is most pronounced among families. Although the number of homeless
people in families remained essentially unchanged between 2015 and 2017 (rising from
653 to 654), the unsheltered count dropped from 152 to 77, a 49% reduction. Overall in
Oregon, HUD’s most recent report found 59% of homeless families living unsheltered. In
Multnomah County, according to PIT data, only 11.7% of families experiencing
homelessness are unsheltered.
(b) Chronically Homeless:​ While the number of chronically homeless people rose 24%, a
rate higher than the overall increase in homelessness, the number of unsheltered
chronically homeless individuals increased by less than 5%, rising from 875 to 917
between 2015 and 2017. We instead observe the larger increase in the number of
chronically homeless people in shelter; that number more than doubled from 158 to 373,
reflecting our commitment to creating low-barrier shelter that better meets the needs of
our disabled and long-term homeless population. A significant majority of chronically
homeless people remain unsheltered, at 71%, but that’s down from 86% in 2015.
IV.

Racial Disparities

One of our core commitments is addressing the disproportionately high rates of homelessness
among many communities of color. As stated above, the definition of homelessness used by
HUD reduces the visibility of communities of color overall. But even using that definition,
significant disparities remain.
People of Color made up 40.5% of this year’s HUD Homeless count, up from from 38.8% in
2015. By contrast, people of color make up 29% of Multnomah County’s population. Within the
unsheltered population, 35% were people of color.
Within the overall percentage increase shown for people of color are some important differences
by racial/ethnic group.
●
●

●
●

Latinos saw a small numeric increase (39 people), but no change in their percentage of
the HUD Homeless population at 10.2%.
There were 185 fewer African Americans in the HUD Homeless population, compared to
2015, resulting in their percentage of HUD Homeless falling from 22.6% to 16.2%. That
included a 57% decrease in African Americans experiencing unsheltered homelessness,
from 396 in 2015 to 169 in 2017.
Asians and Native Hawaiian and Pacific Islanders each saw their percentage of the HUD
Homeless population change by less than half a percentage point.
The only group where we counted a substantial increase in both the number and
percentage of the overall HUD Homeless population was Native Americans. The count

 rose from 82 in 2015 to 424 in 2017, and Native Americans went from 2.2% to 10.2% of
the HUD Homeless population. It’s very unlikely that this change reflects an actual
four-fold increase in the number of Native American people experiencing homelessness
over just the past two years. Rather, it’s likely a correction of an unexplained issue with
the 2015 count. The 2015 PIT Report called out its extremely low tally of Native
Americans and noted that no service providers believed the count accurately reflected
reality for Native Americans at that time. The 2017 number of 424 much more closely
mirrors the 2013 number of 386, which was 9% of the HUD Homeless population.
Compared to 2013, Native Americans have seen an increase in both their real numbers
(38) and their percentage of the HUD Homeless population (+1.2%).
Overall, we continue to see disparities in rates of HUD homelessness for African Americans,
Native Americans and Native Pacific Islanders compared to their overall share of Multnomah
County’s population. Over the last two years, the disparity has declined for African Americans,
consistent with the priority placed on increasing services to this group by A Home for Everyone.
However, African Americans, like Native Hawaiian and Pacific Islanders, continue to experience
homelessness at a rate more than twice as high as their percentage of the population in
Multnomah County. The greatest documented disparity is for Native Americans, whose rate of
HUD homelessness is about four times higher than their percentage of the population.
V. Gender
Reducing the number of adult women experiencing unsheltered homelessness was another
priority for A Home for Everyone coming out of the 2015 count. Overall, 1,355 women were
counted as homeless in the 2017 PIT, making up 36.2% of the total HUD Homeless population.
This was up about 16% from the 1,161 women counted in 2015. As we’d hoped, however, the
number of unsheltered adult women dropped from 594 in 2015 to 471 in 2017, a more than 20%
decrease, while the number of women in emergency shelter nearly doubled.
Another important change from 2015 to 2017 is in the number of individuals who identified as
transgender. While the total number remains small, 44, it is more than double the number who
identified as transgender in 2015 (20). Also notable is that, at 27.3%, those who identified as
transgender had a lower unsheltered rate than those who identified as men or women.
VI. Veterans
Over the past two years, more than 1,300 Veterans have moved from homelessness into
permanent housing. And through a combination of federal, state, and local investments, we
have developed enough housing placement capacity to help Veterans who become homeless
and who seek assistance to rapidly return to permanent housing. This led to our federal
recognition last year. But as with homelessness in general, even as as one Veteran escapes the
streets, another is likely becoming homeless. We were thus not particularly surprised to see 446
individuals who reported being Veterans in the 2017 count, compared to 422 in 2015. While the

 total number is up by 24, the question in 2015 defined Veteran status less clearly than the
question in 2017 -- and thus the two years are not comparable for purposes of showing a trend.
But this year’s number closely mirrors the number of Veterans counted as “Active” (e.g.
engaged in housing process) and “Inactive” (not currently engaging) on our community’s
“Veterans By Name Registry.” This data primarily illustrates the urgency of continuing to fully
support the housing placement capacity for Veterans that we have built over the past two years.
VII.

Disability

Of the 4,177 people counted, 2,527 (60.5%) reported living with one or more disabling
conditions , including a mental disability, chronic physical condition, and/or a substance-use
disorder. The rate was higher among the unsheltered population and among Veterans, at nearly
72% for each. Among those identifying as disabled, 47% reported being unsheltered. Compared
to 2015, 350 more people reported being disabled, an increase of 16%. The count of the
disabled population is thus growing faster than the HUD homeless population as a whole. Data
broken down by disability type is not yet available.
VIII.

Domestic Violence

Because of significant confusion created by the domestic violence question on the 2015 survey,
the 2017 survey used a differently worded question designed to better capture the experience of
intimate partner violence. As a result, a meaningful contrast between results from 2015 to 2017
isn’t possible. Consistent with national data, however, a very high percentage of women
experiencing HUD homelessness report having a history of domestic violence. In 2017, 743
women reported being victims of domestic violence. This represents 55% of the women
counted.
IX.

Duration of Homelessness

For the subset of the population of homeless people who participated in the street count (as
opposed to the shelter count), we have data on the duration of their current episode of
homelessness. We lack responses for about 18% of those counted in the street count, but 36%
report having been homeless less than a year, a decrease from 2015, when 41% of those
counted had been homeless less than a year. We also see a change in the percentage of
respondents in 2017 who report having been homeless two years or more. That percentage is
32% in 2017, as compared to 23% in 2015. Both trends are consistent with reports from
community-based organizations that it’s taking longer than ever in the current housing market to
find permanent housing placements, especially given the widening gap between what homeless
households have as income and the costs of rent.
X.

Additional Analysis Needed

 There are many other potentially important pieces of information that we expect to be able to
provide from the 2017 PIT that are not available yet. Some of these include more specific
breakdowns of overall trends by population or demographic group. For example, we are
interested in knowing more about the rates of different types of disability and whether they differ
by gender and race.
There are also questions that were asked only of a subset of street count participants about, for
example, the duration of their residency in Multnomah County, their employment status, and
what part of the community they slept in on the night of the count. Those questions present
methodological and analytic challenges that we are continuing to work through with our partners
at Portland State University. We expect this additional information to be released in conjunction
with the final 2017 PIT report.
XI.

Conclusion

Portland and Multnomah County, like similar communities up and down the West Coast,
continue to struggle with overall increases in HUD Homelessness. Recently released data from
King County (Seattle) showed a 16% increase since 2015. In Los Angeles County, the increase
was 30%, and in Alameda County (Oakland), it was 39%.
Common to these communities are growing disparities between incomes and rents, record
low-vacancy rates, a critical lack of deeply affordable housing, and insufficient access among
very low-income households to quality employment, adequate benefit levels, and health care,
especially mental health and addiction-related care.
There are some important differences, however, for Portland and Multnomah County. Not only
did our HUD Homeless count increase at a lower rate than other big West Coast communities,
but our reported decline in the unsheltered population also is unusual. Since 2015, while our
unsheltered count declined nearly 12%, Los Angeles County’s went up 38%, Seattle’s count
rose 45% and Alameda County’s climbed 61%.
It is too soon, without the full 2017 PIT Report in hand, to re-evaluate A Home for Everyone’s
goals, priorities, and policies, but that work will happen. And, as in years past, that process will
draw on the experience and expertise of our broad array of community stakeholders, especially
those who are themselves experiencing homelessness in our community.
Attachments: ​Executive Committee PIT Briefing Slide Deck