Logout | View Home Page
  • PIT 2023 reported over 7,935  (1.2%)  transgender, non-singular, or questioning homeles.

    This represents a 19% rise from the 6,678 total in 2022.

  • Tell us your opinions of HUD homeless policy and progress towards reducing homelessness. 



Boston CoC best in Housing Proficiency among Major Cities

(April 2024) A review of the 2023 PIT count data show that 4 Major City CoCs including Boston, NYC, Milwaukee and Baltimore reported greater than 90% percent of homeless individuals being housed in facilities defined as homeless by HUD.  Below are the Top 5 and Lowest 5 among the major cities. (sheltered/total homeless).  A full list of all 49 Major Cities and full list of All CoCs ranked by shelter efficiency are available.

Top 5 (Grade A-)
1.  Boston -  97% (5,033/5,202 )
2.  New York City - 95% (83,940/88,025)
3.  Baltimore - 93% (1,514/1,627)
4.  Milwaukee - 91% (623/685)
5.  Memphis - 87% (1,127/1,292)

Bottom 5:  (Grade F)
1. San Jose - 25%  (2,502/9,903)
2. Los Angeles - 27% (19,013/71,320)
3. Oakland - 27% (2,624/9,759)
4.  Long Beach - 28% (965/3,447)
5.  Sacramento - 28% (2,617/9,281)

While cold weather regions tended to have higher housing rates than warm weather climates, this relationship is not perfectly correlated.  In addition, since PIT counts are not done in the summer months in cold weather climates,  it is not possible to determine housing efficiency during the entire year.  Previous data produced by the NHIP also have shown that most warm weather CoCs receive significant less funding than colder climate CoCs, inequity that HUD has failed to address since the NHIP highlighted the issue in 2016.

The primary job of homeless services providers is to move persons from the "streets" (unsheltered) into housing.  This is what the public wants to see. Presenting data in this fashion is a straightforward way of presenting performance in handling the issue.  These numbers are the ones that should be emphasized in the PIT count results.  

The fact that people may move from HUD-defined homeless housing to other housing that offers more privacy is a separate issue.  Remember, what they gain in privacy, they lose in amenities.  Many persons find comfort living in congregate setting where they are provided security and social interaction, if only with staff.  The NHIP continues to warn not to judge people by your own values of what you consider "appropriate housing".  With an exception to night-only shelters, the NHIP continues to press for a redefinition of homeless to exclude those housed in "shelters".

* * * * * *

HUD Funding Inequities Persist in 2023 Awards

(February 2024) An analysis of McKinney Vento Homeless Assistance funding awarded recently for FY2023 continues to show enormous inequities between states especially compared to homeless prevalence.  Comparing total funding of $3.15 billion with the most recent PIT count of  653,104 yields a funding per capita (count) of $4,384.  The top 5 and bottom 5 states in funding per PIT capita are as follows:

A good example of the massive inequities can be seen comparing the State of Oregon which receives $60 million in homeless assistance and has reported an estimated PIT2023 total of 20,000+ persons with Ohio which receives $153 million and has an estimated PIT of 11,000+ persons.  These inequities are further compounded by the fact that many of the CoCs with high per capita funding have lower housing costs.

It is interesting that HUD seems so fixated on addressing inequities, but it does not address the most fundamental inequity that resides in its own house.  The VA also does not address inequities in its GPD funding - although its staff receive their COLA.  CLCK HERE for the full 2023 Funding Per Capita report.

* * * * * * *

PIT 2023: More errors and humbug

(December 2023) The NHIP has compiled a chart comparing PIT 2022 with the recent PIT 2023 numbers for the following totals by CoC :Total Homeless, Total Emergency, Total Transitional, Total Safe Haven, and Total Unsheltered.  Key highlights from this chart and the overall PIT data include:

  • The NHIP appears to have identified a potential error in the official PIT count comparison.  The 2022 PIT count included a homeless count of 1,205 unsheltered individuals from the Northern Mariana Islands (NMI).  The 2023 PIT count does not include any information from the NMI CoC.  Any comparison between 2022 and 2023 must take into account this missing data.  Taking out the NMI 2022 PIT count, the number of unsheltered homeless went up 10.3% not 9.7% as documented in the AHAR report and publicized in the media.  Similarly, the overall PIT increase would be 12.4% not 12.1%.  If HUD or AHAR officials have any comment about this issue, please contact the NHIP. Or otherwise update your reports.
  • 26 CoCs including 17 CoCs from CA did not complete a 2023 unsheltered count.  Key large CoCs including Bay Area CoCs of San Fran, Oakland, Sacramento, Santa Ana, Marin and Humboldt Counties.  Outside of CA, the cities of Seattle and Detroit also did not complete an unsheltered count. The Bay Area CoCs not completing an annual unsheltered count is pretty miserable considering UCSF is home to the Benioff Institute for Homeless and Housing Initiatives whose mission is focused on empirically-driven research on homelessness.  
  • Failure to complete 2023 unsheltered counts at 26 CoCs likely further underestimated the rise in unsheltered homelessness.
  • HUD continues to provide no information on the number of two-parent couples with children experiencing homelessness.  This data is important because multi-adult households with or without children remain the best prevention for homelessness by a factor of over 100x. 
  • A total of 2,028 persons were reported as residing in Safe Havens.  It should be explained that over the last couple years, the VA has requested that its GPD-Low Demand program track be categorized as a Safe Haven program.  The VA low demand track is a much different type of program than the original Safe Haven program which the NHIP editor has worked with in the past.  The Safe Haven paradigm provides a significantly higher amount of support (higher staff ratios and greater funding) than the GPD-Low Demand program in addition to many other differences.  If there any providers who would like to chime in about their thoughts on GPD Low Demand as a Safe Haven, please contact the NHIP.  Some anecdotal inquiries by the NHIP seem to support the fact that many GPD programs are not aware that these beds are considered as Safe Haven beds.
  • Females represented 30% of the overall unsheltered count.  Four CoCs (FL-517=62% ,FL-506=56%, NY-501=53%, SC-503=50%) reported females representing 50% or greater of their unsheltered counts.  Among large CoCs, DC, Atlanta and Houston reported the highest male unsheltered proportion between 78% to 81% of the total unsheltered count.
  • The number of persons residing in housing defined as "homeless transitional" crept up by 4,104 to 65,521.  The AHAR notes that a significantly number (15% to 20%) of transitional housing beds were open/unused at the time of the count.  The NHIP has maintained that people living in transitional housing should not be defined as homeless since their accommodations do not differ in any meaningful way than people living in rental housing.

If persons are interested in the 2022-23 PIT data comparison file in EXCEL format so they can do their own sorting for analysis purposes, please contact the NHIP via email.

* * * * * * *

USICH/HUD misinterpret homeless mortality data

(January 2024) The USICH is currently misinterpreting results from reports on mortality among homeless persons.  The figure below shows what the USICH is publishing on its website stating that the average age of "people who experience homelessness die nearly 30 years earlier than the average American".  There is no citation with their graph. A search of the internet appears to show that this information may have come from the 2019 report from Sacramento that showed a shortened life expectancy of people who were homelessness living on the street at the time of their death which was later cited in the publication Homeless Mortality Data Toolkit in 2021 by the National Health Care for the Homeless Council which cited other reports from cities with similar data.

The key information in the Sacramento report can be found below which states in the lower left that the average age of homeless male deaths was 51.9 years and homeless female deaths was 43 years.  Since males outnumbered females 3:1, the average of all deaths was approximately 50 years. 

A close review of the Sacramento report and the NHCHC publication by the Dr. Michael Ullman, NHIP Editor, who is a research methodologist, yielded the understanding that these data reflect persons who died while living on the streets, not all homeless persons or all persons experiencing homelessness. The information itself comes from corner's reports. This is, of course, a very important distinction.  While the data is very important, it is a world of difference than stating the average longevity of persons who experience homelessness is 52 years for men and 43 years for women. 

To understand the data presented, one would need to compare the "street deaths" of those non-homeless with those homeless which would likely be much lower than the average age of all deaths.   Or compare the deaths by injury or overdose among housed and non-housed people.  While not being housed is a problem, clearly the former is the greater issue.  Most studies including mortality data reviewed by Dr. Ullman show an average shortened life expectancy of closer to 10 to 15 years among persons with a significant period of homelessness.  Certainly, persons with a short stint of homelessness do not lose 25 to 30 years of their life.  These data are not even collectible since questions about history of homelessness is not gathered at the time of death for people.

The inability or unwillingness of these government reports and the USICH and HUD partners to clearly make this distinction is very troubling. Does the USICH understand the difference?  The lack of a citation in the USICH website for such a important statistic is also unacceptable.  You don't have to convince people that living on the streets is a problem - no spins necessary.  

The NHIP hopes the USICH for publish a clarification on their website and appropriately citations.  If anyone - especially HUD and USICH readers - can explain further the mortality data presented, the NHIP welcomes their response for clarificatio



NHIP Analysis finds Oahu Continuum PIT report greatly underestimated Homeless Veterans in PIT 2022 and 2023

(December 18,2023) In a review of the recent release of Annual Homeless Assessment Report 2023 Part 1 by HUD which reported a large 12.1% increase in homelessness across the country and a 7.4% increase in veterans homelessness, the National Homeless Information Program (NHIP) has found that the number of homeless veterans (active duty and non-active duty) reported by the Oahu Continuum of Care point-in-time (PIT) homeless data is likely underestimated by as much as 100% due to errors and/or exclusions in the PIT 2022 and 2023 reports.

            The Oahu PIT 2022 reported a total of 228 homeless veterans including 101 unsheltered veterans, while the more recent 2023 report found 261 homeless veterans including 108 unsheltered veterans.  A review of the data by the NHIP found that the unsheltered veterans total was taken from a 37% sample in 2022 and a 49% sample in 2023 without any extrapolated estimate for the entire unsheltered population. The 2023 sample questioned 1,161 unsheltered persons about veteran status representing 49% of the total 2,365 unsheltered population.  The 2022 sample questioned 874 out of 2,355, or 37% of the total unsheltered population.

            An estimation of unsheltered veterans that assumes an equal number of veterans among those surveyed compared to those not surveyed would yield 269 unsheltered veterans in 2022 and 220 unsheltered veterans in 2023. Total homelessness among veterans would rise to 396 and 373 in 2022 and 2023, respectively.  The true figure would likely vary from this linear extrapolation but would be significantly higher than currently reported.

            The NHIP had alerted the HUD field director as well as the Oahu Continuum of Care about these data problems in August 2023. There is currently no quality review performed by HUD on local homeless point in time counts. The NHIP finds errors in a majority of reports it reviews around the country that are made publicly available.  The NHIP has written extensively about how much of the reported large decrease in homeless veterans and unsheltered persons during the 2007 to 2015 was due in large part by poor quality homeless counts including large overestimates in several cities.

            In addition, the VA and HUD began in 2020 to omit non-active duty military personnel as veterans in the homeless count which is why HUD further decreased the total number of homeless veterans as reported by the Oahu Continuum of Care from 261 to 226 in 2023 and from 228 to 198 in 2022.

The NHIP also noted in its review of the Oahu Homeless Count reports that only 24% of all unsheltered persons surveyed had an active record in the Statewide Continuums Homeless Management Information System.  This low percentage is reflective of poor implementation of the Coordinated Entry System which has the objective of engaging and prioritizing all persons experiencing homelessness especially those living on the streets.

            The National Homeless Information Project (NHIP) is a volunteer-run initiative that disseminates information nationally about homelessness and advocates for a better understanding of modern homelessness.  Modern homelessness cannot be explained as a problem of affordable housing, but rather is singularly fueled by the rise of single-adult households, changes in societal family norms since the 1960s, and anti-family HUD housing subsidy policies. 



Modern Homelessness:  Product of societal changes, political definitions, bad policy and anti-family government regulation


(Honolulu Star-Advertiser, July 9, 2023)  GUEST COLUMNIST  -  Michael  Ullman, Ph.D.

The most recent homeless point in time count by the Continuum of Care program on Oahu and the neighbor islands continues to show high levels of unsheltered homelessness, as documented now for more than 30 years since the first official Hawaii homeless census in 1991. These counts that appear to demonstrate little to no progress vex both those working in the field, and perhaps even more so, regular citizens who have to live in a community with such numbers of visible homeless people. This frustration could be ameliorated if a better and more holistic explanation of the phenomenon of modern homelessness was understood.

There are two challenges with the homeless population: 1) housing those who are homeless — which homeless providers have done very well over the last 30 years; and 2) stopping the flow of new people into the ranks of the homeless. Addressing the second issue is beyond the ability of government, nonprofit agencies, or anyone else that tries. A hospital can mend broken bones from car crashes, but it cannot impact how many crashes occur. The tired conventional wisdom centering on the lack of affordable housing continues to offer a poor explanation since it lacks a wider ecological perspective.

A simple look at the different rates of homelessness among whites and Native Hawaiians compared with Japanese, Chinese and Filipinos in Hawaii and on the mainland reveals another factor in the homelessness equation. People who cannot afford their own housing are supposed to live with (or return to live with) their families. Homelessness is much lower (three to 10 times) among populations that still maintain these large family support networks.

Understanding homelessness demands a more historical understanding of the trends of household formation and composition. The key change in household formation since the 1960s is the rise of single adult households either via divorce, never marrying, or simply out of choice due to the relaxation of strict family and gender norms which previously offered a safety net. The more single adult households you have, the more homelessness will exist. It is really that simple. The last few decades have seen the rise of freedom from previous norms. Freedom begets chaos in a neutral sense. Homelessness is just one of the characteristics of this freedom from norms.

Another issue that confuses people is the definition of homelessness itself, which defines people living in housing ranging from a dormitory at the Institute for Human Services, a four-person barracks room at U.S. VETS, or an apartment with a two-year lease at Catholic Charities’ Maili Land transitional housing complex as sheltered homeless. Such housing offers everything and more than an apartment does, and often at a fully subsidized price. The only real distinction between this housing and regular nonhomeless housing is limited privacy.

The other issue that gets lost in the details of homeless programs is the anti- family policies of the U.S. Department Housing and Urban Development (HUD), which effectively deters the ability of housing-voucher holders to allow family members or friends to live in their HUD-supported unit. It is normative to share housing with others, and it is culturally appropriate to allow people to let their ohana come live with them for a month, a year or however long needed.

It is extremely difficult for people to give up the belief that affordable housing is the key driver of homelessness. Providers see it because they try to house homeless people in their own apartments instead of urging them to go back to family or putting them in shared housing at half or sometimes a third of the cost. The vast majority of people who are homeless would not benefit from affordable housing, because they essentially need free housing since they can barely pay $200 a month in rent.

Until people understand how changes in household formation and relaxation of societal norms are the primary drivers of homelessness, and until advocates, politicians and regular citizens recognize that homeless agencies and local governments are powerless against these characteristics of modern society, frustration will loom unimpeded.


* * * * * * * * 

HUD Funding Disparities continue unabated - NHIP Report

(June 2023)  The NHIP has revised its CoC 2022 Funding Per Capita Disparities Analysis previously completed in 2016 and 2019 using the latest complete data from the PIT 2022 and the CoC PY2022 funding awards.  Results show similar trends from previous analyses revealing wide disparities in funding when compared to homeless levels ranging from $48,316 in MI-502 Dearborn/Dearborn Heights to $17 in CA-524  Yuba/Sutter County.  Dearborn receives $8M on 166 PIT homeless, while Yuba receives $18,000 (essentially planning funding) for over 1,000 PIT homeless.  Dearborn also has reported zero unsheltered homeless during 2022 and less than 25 in years prior to COVID restriction.

While these are extreme examples, the trend in funding continues to be heavily influenced by the original formulae which allocated funding based on arcane measures that favors old NE cities and rust belt areas. That said, in theory if CoCs have received more funding over the years, results should see a decrease in homeless and therefore an increase in funding per capita since HUD funding today essentially covers Permanent Supportive Housing and costly infrastructure for CES implementation.  While this may have occurred in a few CoCs, there has generally been no linkage found between homeless funding and homeless prevalence due to new inflows which complicate the analysis. 

Two additional factors have impacted the levels of funding at individual CoCs.  During the late 1990s and 2000s, CoCs with superior applications were awarded bonus projects.  This practice essentially ended in the early 2010s.  In addition, CoCs that sought funding for housing voucher programs like Shelter Plus Care received large increases due to increases in FMR over the years.  

Overall, the average HUD CoC funding per capita is $4,741.  The Los Angeles CoC received over $157M - the most of any CoC - which translated into only $2,424 per capita.  Other large cities include San Francisco ($6,838), Chicago ($22,331), Miami ($12,686) and New York City ($2,387).  The NHIP points out that disparities are actually underrepresented by this analysis since it does not take into account the cost of housing in each CoC which can be 2 to 3 times higher than in many areas of CA, DC, and CT compared to Midwest states like Oklahoma, Missouri and Southern states like Tennessee, Alabama, and Mississippi.

The NHIP has also recommended over the years to use bonus or new funds to mitigate some of the extreme disparities to no avail. HUD has had many chances to reduce these disparities but has chosen not to address the issue. Even the most recent new Outreach funding did not target these disparities effectively with states like Illinois receiving over 12% of the funding while accounting for only 1.3% of the nation's unsheltered count. The table below shows the Top 10 CoC by Funding per Capita.



* * * * * * * * * * * * 

What Bill Walton doesn't understand about homelessness (and it's not his fault)

(January 2023) NBA legend Bill Walton and George Mullen, CEO of the Sunbreak Ranch, offered an approach to help end homelessness in an editorial published in the Times of San Diego this past weekend. They called for 2,000 acres of government land outside of San Diego to be used for a new "tent city" area with amenities for homeless people to relocate and live relatively unfettered. This approach was to be replicated outside of many cities. 

While the effort is laudable, although immediately attacked by some, Walton's approach is ba sed on the same wrong-headed information that still dominates the national homeless dialogue. Bill - you need a new coach to explain the fundamentals of homelessness that the media and "experts" will not provide - and I am here to help!

"It's like this, Bill.  The offense is scoring 120 per night, but the defense is giving up 125+."

Mr. Walton and Mr. Mullen, the public, politicians, policy makers and the homeless industry still confuses the difference between offense (helping currently homeless people find a place) and defense (preventing new people from entering the ranks of homeless living). While the pick n' roll is a great offensive strategy, it will not stop Steph Curry on the other end from getting an open three and winning at the buzzer. The offense is fine. Homeless services is great at housing people who are experiencing homelessness - it is not a failure – it houses 350,000+ individuals plus each year and keeps 500,000+ from going back to the streets. .

Increasing the salary cap could house more people. But it can't house everyone of course, because just like in basketball, there are rules like the Bill of Rights and the Constitution. In the NBA, you can't just charge into your defender to score or grab a scorer driving to the bucket. That's a flagrant foul.  America also has rules (laws) that prevent you from doing just that to homeless people in many public spaces.  And if you add an expansion team like your proposed Sunbreak Ranchers, even if you are willing to pay KD or Joker way big money, they ain't gonna come play for your team at your new arena.  They like it where they are.  Well, KD not sure.

It's like this, Bill. The offense is scoring 120 per night, but the defense is giving up 125+.   The answer is not scoring 130 since the opponent can score 135-140 especially since some of the plays give an easy transition to your opponent. Yes, we have a losing record, but we not the 1972-73 76ers. it’s not because we don't have a good inside game. San Diego had 3,000 dunks last year (people housed). But we need to block more shots (i.e. prevent new people from entering homelessness), and a lot of new norms, bad policies, and poorly understood definitions of homelessness are hurting us on defense and making us look bad in the press.

Improving our defense and stopping the other team from scoring, that is, preventing new people entering homelessness is a completely separate issue and it is one that is beyond the scope of any one team, non-profit or homeless coalition. You see there is a new kind of game. College ballers think they are all one and dones. Sure, you and Kareem could have left college early had the norms been different, but today too many recruits leave college (people leave family homes) and can't even get picked up by the G league!  

You see society has moved from a concept of multiple adult households - since most adults cannot pay the rent alone which is why most people share housing with family or friends or just roommates they don't even know - to one where single adult households are almost the predominant form. This began in the 60s/70s and is the new norm - through divorce, never marrying, and not wanting to return to the family home when they run out of money, and for many not wanting to share a locker room with others.. People without means are supposed to go back to family or friends and obey a few rules as a house guest. Luckily, many still join a team, but too many don't and too many of those become homeless.

Worse yet, the "game" now has rules that limit teams to only 5 players and only one center. In policy terms, the housing subsidies that are given to millions of very poor and homeless households prevent them from allowing other friends and family who are or maybe homeless from staying with them. It's called the Brooke Amendment.  You can look it up.  Can you imagine when the Celtics had offered you a spot on the roster in 1985 and the league said "No, you have your five!"  Like Parish could have gone all 48 minutes!

Lastly Bill, remember when there was an ABA and NBA.  Okay, the NBA was probably superior overall, but the ABA had great players (Erving, Gilmore, Barry) and good teams. Imagine if the NBA said that the ABA was not basketball and refused to allow these players to add those points to their career totals. Well, that is essentially what the federal government does. About 60% of "homeless" people are housed. They just don't have the "digs" of the better housing. They do live in housing, sometimes shared, or congregate or even some private units with a key. In San Diego, about 55% of the homeless population are housed including about 20% who have their own apartment. We just call them homeless cause "they have to share a locker room" basically. Some people complain about the condition of the stadium (house) these people live in - but they don't seem to want to make it nicer. Do they deserve a better arena?  That's up to the fans (community) to demand it and owners (government and other funders) to build it.

You see, Mr. Walton.  I really am sorry for the poor coaching on homelessness you have received. I'm sorry we don't have instant replay to see all these miscues the refs are making. As one of the all-time great defensive centers, I know you will now better understand the problem and want to know more.  Maybe you can help me talk to the commissioner (HUD).  And although I'm not Riley, Auerbach, and certainly not Mr. Wooden, I can teach you some fundamentals of homelessness. My next camp starts soon!

* * * * *

Whites show lower shelter accessibility than Blacks

(January 2023)  A review of the sheltered and unsheltered numbers for individuals in the PIT2022 census shows that White persons who are homeless have 63% lower odds of using emergency shelter services compared to Black persons who are homeless.  While 59% of white persons are unsheltered, only 47% of Black homeless individuals are unsheltered. The analysis did not include persons in transitional housing.  Virginia Beach, VA offers a good example of this accessibility issue.

Among the 133 shelters individuals, 65 were White and 68 were Black for a near even proportion.  Among unsheltered persons, the PIT reported 40 Whites and 8 Blacks.  In Virginia Beach, Black persons had 5 times greater odds of utilizing shelter place compared to White persons.  Shelters in both New York City (2.13 greater odds) and Seattle/King County (1.60 greater odds) also were more accessible to Black individuals compared to White individuals experiencing homelessness. Some cities showed a reverse trend like Chicago, Salt Lake City, and Los Angeles where Blacks had lower shelter accessibility odds (less than 1.0) than Whites..   A complete listing can be found under the Homeless Data reports menu.

While publications have focused on the over-representation of Blacks in the homeless population compared to their general population representation in the U.S., almost no coverage has pointed out the disparities in shelter accessibility by race. More attention should be paid to ensuring equal shelter access for all race groups.

Location of shelters (urban vs. rural) may be a factor in the Black to White disparities.  There is not enough data on the PIT to perform this type of analysis.  Balance of State CoCs were represented as both race accessible and inaccessible.    Some accessibility differences may not be salient as in Washtenaw, Michigan which showed a 4x lower likelihood of whites utilizing shelters, but the total unsheltered White population was only 5 persons compared to 1 Black person.

* * * * * * 

New gender listings proposed by HUD

(November, 2022) HUD is requesting feedback on the data elements under the proposed changes for the HMIS race, ethnicity and gender questions.  Below are HUD's suggested changes for the gender element. 


From a statistical and research point of view, all responses other than singular woman or man are not needed, since the responses make up only 0.5% of all responses (AHAR 2020) and are too heterogenous for any analysis.  Any meaningful analysis would just lump them as an other category. This type of complex gender identification information should be collected through case management discussion as it relates to the person's needs, psycho-social stability and unique challenges.  The additional data burden on the system compared to the value of the data points have yet to be shown. Meanwhile, the HMIS still does not require collection of information on the parents, siblings, spouses, or children not with the homeless adult and the family members' ability to support this family member in need.

In comparison to the unlimited combination of responses for the multi-tiered gender questions, the question on race/ethnicity continues to lump a massively heterogeneous mix of people the government calls Hispanic/Latino - of whom refer to themselves as Hispanic/Latino - and have only adopted the label as a convenience for these census questions since they don't give any additional choices for their actual ethnicity.  The same is true for Whites.  Try telling a person of Greek heritage that they are similar to the French. 


HUD awards 2.6 billion for CoC Homeless Assistance

(March 2022) HUD has released the list of project grantee awards that total $2.6 billion under the McKinney-Vento Homeless Assistance Act.  A complete list of awards from the 2021 grant process can be found HERE.

The 5 CoCs with the largest awards were Los Angeles ($155M), New York City ($145M), and Chicago ($86M), Seattle ($53M), and San Francisco ($51M).  The average CoC award was $13.7M with a median award of $2.5M. 


Why the PIT does not tell the full story

(June 2021) Each year, HUD releases the PIT homeless count as if this is a measure of how well Homeless Housing Continuums are performing.  This could not be farther from the truth.  First, the PIT includes people who are living in housing (what is called emergency or transitional) and people living in unsheltered locations.  These situations are completely opposite housing situations - one is housing, one is not.

People in housing labelled by the government as emergency or transitional receive more amenities than people living in government subsidized housing. The only basic difference is the lack of private space.  Lack of privacy is a late 20th century value-laden characteristic of housing in most, but not all, developed countries that has little in common with the notion of a permanent and safe dwelling. A full discussion of the need for privacy as a requirement for permanent housing is beyond the scope of this brief.

The second problem of the PIT is the failure to include the bulk of housing services provided by Continuums and funded by the Federal Government, namely, Permanent Supportive Housing, Rapid Re-Housing, and Other Permanent Housing situations.  These three categories of housing nearly equal the total of PIT-reported housing.  From the 2021 PIT Count, we have the following totals:

Total Permanent Housing                     511,809 (47%)
    PSH                                 346,389
    RRH                                 122,908
    OPH                                   42,512

Total "Homeless" Housing                       354,386 (32%)
    Transitional Housing            72,484
    Safe Haven                                  1,986
    Emergency Housing            279,916 
TOTAL HOUSING                                  866,195
UNSHELTERED                                     226,080 (21%)
TOTAL                                                  1,092,275
% HOUSED                                               79.3%

The full story of Continuum Housing shows that 79.3% of persons served are currently housed including 47% living in government-funded Permanent Housing.  Only 21% of persons identified by Continuums are living unsheltered.  This total also includes an unverified number of persons living in RVs.  The %Housed number for each of the geographic categories of Continuums are:

Major City                    78.5% 
Largely Urban            84.2%
Largely Suburban     82.8%
Largely Rural              73.8%

While Major Cities contain more than half of the individuals defined as homeless including over half of those defined as unsheltered homeless, they actually outperform Largely Rural areas with respect to the percent of persons who are receiving some form of housing (78.5 vs. 73.8%).  Largely Urban CoCs (84.2%) and Largely Suburban CoCs (82.8%) have the two highest housing rates.  Major City CoCs outperform Rural CoCs primarily due to the significant number of Permanent Housing options funded by HUD and local governments.  Rural CoCs have few homeless persons but have much fewer resources.

The NHIP reminds that unsheltered homeless prevalence is largely unrelated to the efficacy of Continuums.  Key determinants of unsheltered homelessness include:  1) Winter weather conditions, 2) Number of multiple-adult families, 3) Presence of restricted HUD or VA housing subsidies, and 4) Other socio-cultural characteristics of geographic areas.

CoC Housing Rankings:  The Full Picture

The NHIP has developed two types of CoC housing performance rankings:  1)  Percentage Housed - as described above which is calculated by one minus the number of persons unsheltered as estimated by the PIT divided by the total housed (ES,TH,SH,PSH,RRH,OH) plus the unsheltered count, and 2)  Housing Score which allocates points based on the type of housing setting with 10= Permanent Housing (PSH,RRH,OH); 9 = Safe Haven, 7=Transitional Housing, 3=Emergency Housing, and 0=Unsheltered.  The total score is divided by the total possible score if all persons were in permanent housing. 

Theses rankings provide a more comprehensive understanding of the homeless housing system in Continuums as opposed to the limited focus on persons unsheltered or in Continuum "shelter" housing.  The full results for PERCENTAGE HOUSED and HOUSING SCORE is available.

* * * * * *

New Orleans ranked #1 in SPM performance

(December 21, 2020)  The New Orleans CoC led the SPM rankings among the 48 Big City Continuums for the 3rd straight year with an overall score of 17 out of 18 representing high performance in 5 of 6 categories and average performance in one SPM category (income increases).  The NHIP compiled the 2019 ranking from SPM data submitted by CoCs to HUD for the period Oct 1, 2018 to Sept 30, 2019.  The top ranked (in order) CoCs include:

1.  New Orleans (17)
2.  Virginia Beach (16)
3.  Denver (15)
4.  Miami-Dade (15)
5.  Detroit (15)
6.  Seattle (15)
7.  Portland (15)
8.  Long Beach (15)
9.  Jacksonville (15)

The NHIP scores CoCs on six key SPM in three tiers (High, Avg, Low Performance).  Delineation between raw scores is based on adjustments by program stay length and empirical totals from the 5 percentage measures.  The many weaknesses including incomplete HMIS participation, large variation in Outreach efforts among CoCs, and general data errors precludes the wisdom of further refinements of the rankings at this point. 

The NHIP provides these general measures with an emphasis on the 3 tiers not on individual rankings.  Whether Detroit is doing a better job than Portland is not really an answer that the measures can accurately give.  The ranking do provide comparisons that can giveimpetus for CoCs to improve performance in areas and see how their performance compares with other CoCs. 

Tier scoring for each of the six measures uses the following cut-offs which have remained relatively stable for the last three years.

Measure 1:  Median Program Days

  • 3= 90 days or fewer
  • 2 = 91 to 180 days
  • 1= More than 180 days

Measure 2:  Homeless Recividism (24 months)

  • 3 = 15% or less
  • 2= 16 to 29%
  • 1 = 30% or greater

Measure 3:  Income Increases (both stayers and leavers)

  • 3 = 50% or higher
  • 2=  30 to 49%
  • 1 = Below 30%

Measure 4:  Outreach successful discharges

  • 3 = 50% or higher
  • 2= 30 to 49%
  • 1 = Below 30%

Measure 5:  Permanent Housing Placement

  • 3 = 70% or higher
  • 2= 31 to 69%
  • 1 = Below 30%

Measure 6:  Permanent Supportive Housing retention

  • 3 = 95% or or greater
  • 2= 90% to 94%
  • 1 = Below 90%

The NHIP welcomes any discussion or feedback on the SPM measures via email at

* * * * * 

2019 SPM Rankings of Smaller Urban Areas 

The NHIP also completed rankings of 49 Smaller Urban Areas that had complete data on all measures.  Thirteen CoCs did not report any Outreach data and were excluded from the analysis.   The Top Tier Smaller Urban CoCs included:

#1 Newport New/Hampton VA
#2 Sioux City ND
#3 Fort Collings CO
#4 Tallahassee FL
#5 Portsmouth VA
#6 Alexandria VA
#7 Oxnard CA
#8 Kalamazoo, MI
#9 Burlington VT
#10 Durham NC
#11 Cambridge MA
#12 Bakersfield CA
#13 Lansing MI
#14 Grand Rapids MI
#15 Bridgeport CT


* * * * * 

EDITORIAL:  Racism and Homelessness

In a recent request for input to guide a new committee to address homelessness among Black San Diegans, the Regional Task Force on Homelessness wrote:

"In the 2020 Point-In-Time Count, Black people made up 21% of the unsheltered population and 30% of the sheltered population, while only making up 5.5% of the County’s general population.  This is the result of systemic racism and a history of racial oppression in our nation and community. "

The NHIP feels it is necessary to address these type of statements of causality because they intertwine cold hard facts (racism against Black Americans) with  oversimplified explanations of outcomes.  The NHIP realizes it is difficult to attempt to clarify - let alone argue - such statements, because they are often read as denials or minimizations of historical and current racism.  It is important, however, because if the end goal is to reduce the disparity, in this case homelessness, then a need to understand the proximal factors that create the situation must be illuminated.  One can never prove that racism is not a factor, even if a singular Black person believes that to be the case for herself, because the condition existed and is temporally associated with the outcome.  

NHIP research identifies the following primary factors that overwhelming cause the overrepresentation of Black Americans among the homeless:

High rates of single adult families with children - between 2 to 3 fold compared to non-Black households.  Single adult households, with or without dependent, are the primary causal factor for homelessness.  HUD states in the 2019 AHAR that one-person households have a 5-fold over-representation. Homeless statistics need to better illuminate this issue for all ethnicities.
Perverse restrictions in housing subsidy programs (Public Housing, Section 8, Shelter Plus Care, SHP, VASH) that effectively prevent persons with these vouchers from allowing other family members or friends to live with them.  The more people of any ethnicity with a restricted voucher, the more persons of that ethnicity or community will be cut-off or face insurmountable barriers to natural support networks. The 1968 Brooke Amendment which brought these rules into existence has likely caused more homelessness than any other federal housing policy.
Draconian regulations in federal housing that do not allow persons with criminal backgrounds into federal housing programs.  This has had a particular pernicious effect on males and males spouses in the Black community over the past 50 years.
Black Americans are over-represented in housing subsidy programs (3 fold) and thus more Black households are impacted by these perverse housing regulations - of which problems #2 and #3 work to impact condition #1.

Other factors such as substance abuse, mental illness, veteran status are equally problematic among the White populations, so while these are important covariates, they are not necessarily unique to Black Americans. Tackling Condition #1 cannot be addressed directly since it is a socio-economic byproduct heavily influenced by the new norms that prioritize having your own place. The NHIP argue that Conditions #2 and #3 could be changed and would have a large effect on reducing homelessness. 

Tackling the problem of income inequality is also beyond the scope of homeless and housing services.  One must remember that there is not a direct correlation between income and homelessness.  Women have far less income than men, yet they experience unsheltered homelessness at 1/3 the rate.  Women access natural supports more often.  Ultimately , that is what men, both Black and White, need to be able to achieve to permanently reduce the race and gender gap in homelessness.

The NHIP realizes the necessity of having statements like the one by San Diego as more of a statement of solidarity concerning the issue, rather than a scientific explanatory model.  That said, it is also important to tackle current policies that continue to impact Blacks and cause too many to experience homelessness.

People must also realize that  living homelessness today, even more so than 20 years ago, has become a normative housing option - especially for people in warm weather climates.  When behaviors become normative, it is difficult to change these habits in a society with protected civil rights.  In a strict Marxist sense, one could argue that some Blacks and Whites are exercising their civil rights to be free from any societal rules  (zoning, housing codes, paying rent)  - their homelessness merely being a form of civil disobedience.

With this in mind, the NHIP offers a different statement that San Diego or others could employ that does not minimize racial oppression but underscores factors that must be addressed:

"Addressing the over-representation of Black Americans among the homeless population requires prioritization due to the systemic racism and history of racial oppression in our country.  Policies and economic impacts that have weakened too many households in the Black community must be changed to reduce this grave disparity."


* * * * *

2018 CoC System Performance Measures Rankings show high performers, improvers and strugglers

(May 20, 2020) The NHIP has completed its 2018 Rankings of CoCs based on key HUD System Performance Measures.  These data are the most recent publicly released data and cover the period from Oct 1, 2017 to Sept 30, 2018.  Six measures were included:

  1. Median Stay in Days (Emergency Shelter/Safe Haven)   (CoC Median 29 days)
  2. Homeless Recidivism - 24 months  (CoC Avg 19.4%)
  3. Increased Income (Stayers and Leavers) (CoC Avg 35%)
  4. Successful Outreach Exits  (CoC Avg 36%)
  5. Successful Placements in Permanent Housing (all programs) (CoC Avg 42%)
  6. Permanent Supportive Housing retention (CoC Avg 96%)

Rankings are divided by HUD's three CoC types:  Big Cities, Balance of Sheet/Statewide, and Smaller Cities/Rural/Suburbs.  Scoring for each measure uses thresholds with scoring of high, medium or low (2,1,0).  Scores range from 0 to 12.  For more on the methodology, please read the full 2017 CoC Ranking Report.  

 Due to the limitations of the HMIS data, this summary grouping is thought to be the best approach. Many CoCs do not operate programs in certain areas (Outreach, Permanent Supportive Housing, Emergency Shelter) that negatively impact their scoring.  The NHIP recommends using this rankings to identify areas of needed improvement and providing a general understanding of how well their fare compared to peers. 

Below are the top performers for each CoC class and the full ranking of all CoCs for each rank.

BIG CITIES  RANKING LIST:  Top CoCs: New Orleans, Miami; Big Improvers:  Denver, Detroit

BALANCE OF STATE LIST:  Top CoCs: Mississippi, Kentucky, Nevada;  Big Improvers: Alabama, Oklahoma, Michigan

SMALL CITIES:  Top CoCs:  Slidell/Southeast LA, Overland Park KS Big Improvers: Lynchburg, VA, Roseville/Placer, CA, San Luis Obispo, CA.

A couple important notes include:

1) Outreach Exit performance is heavily dependent on the extent of outreach programs in a CoC.  The collection of more outreach data (unsheltered persons) may also bias recidivism upward compared to CoCs that do not enter much or any outreach discharges in the HMIS.  Similarly, those with little outreach or outreach that is not linked to placement will score low on this measure.

2)  Recidivism is also impacted by the type of housing programs offerred, since emergency shelters will have high rates of recidivism compared to transitional and permanent facilities.  This is another reason why ES are not an effective solution.  In addition, the lack of substantial outreach that documents unsheltered episodes may bias recidivism downward since data on persons living unsheltered is not collected in their HMIS.

3)  HUD notes that the use of SPM results should be used with discretion due to many changes including changes in HMIS providers and data transfer issues, 

4)  Most CoCs do not have 100% HMIS participation with their housing facilities listed on the Housing Inventory Chart.  CoCs with a higher participation rate may be bias upward or downward due to the exclusion of key facility data.

You can contact the NHIP at for any request concerning these rankings.


* * * * * 

Dear Marc Benioff, CEO Salesforce:

You have donated $30 million for a Research Institute on Homelessness at the University of California at San Francisco.  You have personally lobbied for hundreds of millions of dollars in additional state and local funding for housing and homeless services for San Francisco, the Bay Area and the State of California.  Those efforts are appreciated.

The NHIP offers the following action steps needed if you want to help the public better understand the problem of homelessness and improve services with the hope of reducing, but not ending, street homelessness.

1.  Improve the accuracy and increase the frequency of street counts.
How effective could you be in leading Salesforce if you only received financial statements every two years?  Currently, San Francisco, Oakland and several other Bay Area Cities only conduct a homeless street census every two years without any check for accuracy.  This is below the standard of most cities who complete an annually count.  Given the urgency of the problem, unsheltered counts should be done quarterly (if not monthly), including recounts for accuracy, especially in key areas with a high density of street dwellers (e.g. Mission, SoMa and Tenderloin) with results disseminated widely.  These counts also need to be completed by trained professionals - not volunteers receiving 45 minutes of training or overworked case managers who already have a full-time job. 
2.   Improve your monthly HMIS reporting

Currently, the reporting from the HMIS is below standard. Only 56% of the emergency and transitional beds and 73% of Permanent Supportive Housing units are reporting data into the HMIS system.  How could you make decisions about Salesforce with only half of your departments reporting?  The monthly reports produced by the homeless department continue to be lacking in necessary data needed for decisions.  I challenge you personally to improve this aspect of homeless reporting for SF.  Here is the latest January 2020 report.  If you read it, you will see. If you need assistance, the NHIP can explain what is missing. I do not blame the current staff,  because they do not have the skill and training to produce rigorous reports.  Salesforce is all about timely data analytics. Let's make that a goal for SF homeless reports.

3.  Stop opening Emergency Shelters (aka Navigation Centers)

The Emergency Shelter, call it what you want, is the failed solution of the 1980s and 90s.  This is why HUD provides zero Continuum funding for these shelters.  All Emergency Shelters should be closed or converted to permanent housing.  If you read the monthly report, you will see that the success rate from these Navigation Centers is less than 15% with most clients just returning to the streets.  The fundamental flaw of homeless services is asking a person who has come in from the streets to leave the shelter and go to a "better" place.  Upper-class people may not think a shelter or congregate living is permanent, but for many people it is good enough.   Do not kick people out just because they stay a long time. This is a good solution for them until THEY decide to move on.

4.  Explain to the public that the lack of affordable housing does not cause homelessness

Please understand that affordable housing does NOT cause homelessness.  The primary engine for homelessness is the formation of one adult households without sufficient resources.  Most adults without resources live with other adults.  This is why homelessness is not significantly higher.  Any functioning household who cannot afford housing will move somewhere else or move in with family or friends.  Laws, federal regulations and societal trends that fuel the creation of these one-adult "deformed" households need to be addressed.  Currently, an individual with a housing subsidy cannot allow another family member or close friend to come live with them.  These types of regulations are anti-family and anti-community and a big part of why we have such high levels of homelessness..

5.  Prioritize shared-housing - the normative housing for most SF residents - for current homeless persons.

SF has 8,600 permanent supportive housing vouchers/subsidies that each support one adult.  The norm for middle-to-lower income persons in San Francisco and the world is shared housing - whether that is 4 Ivy League grads sharing a house or 3 tech coders sharing a 3 bedroom apartment.  SF and other Bay Area homeless services needs to prioritize the development of shared-housing opportunities for any new or re-opened voucher.  This can create 50 to 100 percent more housing opportunities with the same amount of funding.  Only a very small percentage of homeless persons clinically require their own apartment, most can and many thrive in shared-living situations. Shared-living is how most people make it in San Fran. It is both appropriate and beneficial for social animals like humans.

6. Explain to the public you cannot end homelessness unless you want a police state

Stalin and Hitler ended homelessness.  In the US, we have civil rights that allow persons to occupy public spaces. How much space is between the community and the constitution. These same civil rights allow people to own a gun, marry same sex partners, practice their religion, and seek justice for crimes.  To a large extent, persons living in public urban space is a social movement that says "F-You" to the wealthy who also enjoy the same city.  We can do better, but we should not even attempt to end homelessness and people need to understand the reasons. 

You have made a good start by hiring two medical professionals to take on the leadership of the Center.  Homelessness is co-morbid with a high prevalence of serious disabilities and chronic health conditions.  As we know, living on the streets is associated with a shorten life expectancy of 10 to 15 years. 

The Center needs to add to its arsenal diversity in the understanding of the roots causes of homelessness (not affordable housing), the analytics required to monitor progress, and the creativity to apply normative housing solutions to quicken the pace of action.  Unfortunately,  the Science of Homelessness is currently at a Second Grade level. We need to improve it enough to earn a High School diploma, and by doing so, improve the homeless situation and the public's understanding of homelessness in both SF, the greater Bay Area and the Nation.

Thank You 
Michael Ullman, Ph.D. 
National Homeless Information Project

* * * * *

Transitional Housing masks true rise in homelessness

(January 2020)  The recent release of 2019 PIT Count Data show a total one-day estimate of 567,715 residing at shelters or living unsheltered.  This total includes 77,095 persons living in transitional housing (including Safe Haven facilities).  The NHIP no longer defines people living in  Transitional housing  as homeless since it needs to be understood and defined as permanent housing.  The continued inclusion of persons in transitional housing has also worked to mask significant increases in homelessness over the past several years due to increases in emergency shelter counts and unsheltered estimates.  The graph belows shows the HUD totals and the NHIP homeless totals adjusted without persons living in transitional facilities (including Safe Havens)>

Two differences can be seen comparing the two definitions.  First, the reported decrease in homelessness from 2013 to 2016 was fueled by the reduction of transitional housing facilities.  HUD efforts over the last decade to eliminate transitional housing funding has worked to reduce the transitional housing bed capacity from over 200,000 to less than 100,000 currently.  Similarly, the PIT count from transitional housing has more than halved during the period.

The second trend from the graph is the rapid increase in homelessness since 2016.  The last three years have seen the NHIP-defined homeless count rise 14% from 430,663 to 490,620.  This trend has been fueled by the steady increase in new emergency shelters throughout the country and the rise in the unsheltered estimate especially over the past two years.

The NHIP has compiled a state-by-state comparison between HUD and NHIP homeless counts for 2019 for each of 56 states and territories.

* * * * * 

Unsheltered Count rises 8 percent to highest level since 2012

CLICK HERE for a state-by-state unsheltered count comparison between 2018 and 2019.

* * * * *

Over 6,000 homeless veterans counted in communities that "Ended Veterans Homelessness"

(Novembe 2019)  An analysis of the 80 communities that have been designated by HUD and promoted by the USICH as having ended homelessness show a total of 6,029 homeless veterans including 1,492 unsheltered veterans found in 64 communities.  There are 16 CoCs that report no homeless unsheltered veterans including Alexandria, VA, Bergen County, NJ, Reading, PA and Nashua, NH.

The analysis performed by the NHIP did show that these 80 communities achieved better progress comparing 2018 with 2019 with an average decline of 4.3% comparing a decrease of 1.7% of the other 320 CoCs.

The NHIP continues to advocate for the elimination of the label of "Ending Veteran Homelessness" since it is a false notion that is contradicted by the data and represent a slap in the face to any veteran who may be experiencing homelessness in these communities.  It is fine to acknowledge communities that have built effective systems that minimize homelessness and the length of homelessness by veterans.  Calling these efforts "Ending Homelessness" is akin to G.W. Bush's infamous "Mission Accomplished"call that the Iraq war had ended in 2003.

CLICK HERE for a complete list of the 80 communities analysis.

CLICK HERE for a complet list of 2019 Veterans Homeless by State.



* * * * *

HUD releases PIT 2019 veteran counts

(November 2019)  HUD released veterans data from the most recent PIT 2019 homeless one-day count completed by the 400 CoCs around the country.  Results showed a small decrease from 2018 to 2019 in total one-day homelessness from 37,878 to 37,085.  Thirteen CoCs reported drops of 50 veterans or more, while 10 CoCs reported increases of 50 veterans or more.

The TOP TEN CoCs reporting the greatest decreases from 2018 to 2019 included:

San Diego (-244)
Texas BOS (-127)
Pasco County, GL (-122)
Santa Ana/Anaheim (-108)
Chicago (-95)
Santa Cruz/Watsonville (-94)
Seattle (-91)
Colorado BOS (-75)
Imperial County, CA (-69)
Houston/Harris County (-69)

Top 5 Veteran Homeless PIT Increases (2018-2019)

Sacramento  (+175)
Oakland/Alameda County (+166)
Dallas (+111)
Louisville (+98)
Georgia BOS  (+96)

The one-day homeless totals include over 10,000 veterans residing in VA-funded housing programs including the Grant and Per Diem transitional program and VA Emergency or Respite Bed programs. 

Click for a full listing of Veteran PIT 2019 by CoC sorted by greatest decrease. 

* * * * *

Help with Racial Equity Analysis for Exhibit 1

(August 2019)  Below are the two excel files that can run racial equity analyses:

1)  HUD Racial Equity Tool - Using PIT information to compare overall population prevalence and poverty prevalance with homeless service utilitization prevalence.

Copy of Sample Racial Equity Tool for DuPage County formatted to fit one page.  You will need to UNPROTECT the spreadsheet to manipulate formatting.  Click here for sample narrative analysis.

2)  NAEH Racial Equity Outcomes Tool - Using HMIS annual utilization and outcomes data to measure equity of positive outcomes by race/ethnic groups. CLICK HERE for sample data and narrative analysis.

3)  NHIP Staffing Equity Tool - Using a modified version of the NAEH tool, CoCs can calculate a staffing equity analysis.  

* *  * * *

San Fran, Oakland, San Jose Report Large Homeless Rises

(May 2019)  Preliminary reports released by Continuums in San Francisco, Oakland, and San Jose showed signficant increases in the one-day estimate of homelessness in comparison to their previous 2017 counts.

San Francisco:  17% from  6,858 (2017) to 8,011 (2019)

Oakland/Alameda County:  43% from 5,629 (2017) to 8,022 (2019)

San Jose/Santa Clara County:  35% from 7,294 (2017) to 9,706 (2019)

Full reports on the breakdown of the 2019 PIT count have not yet been released.  It is important to understand the differential changes among people living in shelters, living in RVs, cars, or tents and encampments.  These represent different populations with the need for difference interpretations.  HUD continues to lump all unsheltered persons together - which is not helpful in understanding the trends in the situation.  The media also does not tend to differentiate between those living in shelters which is generally a function of shelter capacity and those not living in shelters.

Persons living in RVs should not be considered homelessness according to the NHIP.  Assistance can be provided to this group, but defining them as homeless is another large stretch of the definition which makes it meaningless. 

* * * * * * *

Orange, Riverside and San Bernardino Report Significant Increases in Homeless Estimates                          

(April 2019)  The California counties of Orange, Riverside and San Bernardino have all reported significant increases in the estimates of one-day homelessness from their 2019 counts as compared to their counts in 2018 and 2018.  A common theme among counties is the implementation of mobile apps that have increased count accuracy by reducing the time recording homeless encounters allowing for increased coverage and reducing arithmatic mistakes.  Preliminary data from the NHIP 2019 PIT Survey show that an increasing number of CoCs are utilizing Mobile Apps although problems still exist in reaching all geographic areas and having sufficient capacity to complete the count.  Highlights from the three counts include:


  • Orange County reported an one-day estimate of 6,860 homeless persons representing a 38% increase over the 2018 estimate of 4,955. 
  • Officials report adding 1,390 additional emergency shelter beds over the past two years within the county.
  • The unsheltered estimate rose from 2,584 in 2017 to 3,961 in 2019.
  • Officials feel the switch from an estimation procedure to a physical count using the mobile apps has produced a better estimate than in previous years.
  • For more information, read the full 2019 ORANGE COUNTY Report.


  • Riverside reported a 22% increase finding a total of 2,811 homeless persons compared to 2,316 in 2018 and 2,406 in 2017.
  • A record 747 volunteers assisted with the one-day count.
  • Estimates of unsheltered homelessness rose from 1,685 in 2018 to 2,045 in 2019, while sheltered persons increased from 625 to 766.
  • The Continuum has not released a full report detailing the numbers.


  • San Benardino officials reported a 23% increase in the 2019 PIT count tallying 2,607 persons, nearly 500 more than the previous count estimate of 2,118 in 2018.
  • 73% of the total, or 1,920 persons, were found living unsheltered, a jump of more than 500 over 2018.
  • The City of San Bernardino has witnessed a near doubling of homelessness rising from 491  in 2017 to 890 in 2019.
  • People aged 55 and older saw a 71% increase in homelessness.

The other large cities of San Francisco, Los Angeles, Oakland, Long Beach, San Diego and Santa Clara/San Jose have not publicly released any data.


* * * * *

National Performance Rankings of Continuums

(February 2019)  Performance rankings of 364 Homeless Continuums of Care have been tabulated by the National Homeless Information Project (NHIP) using System Performance Measure data submitted to HUD for the 2017 fiscal year.  Thirty (30) Continuums lacking complete SPM data are excluded from the rankings.

The rankings provide a three tiered-rating system (Higher, Average, Lower) with a second index score to differentiate within each tier.  Rankings are listed by each of three geographic types designated by HUD:  1) Big Cities, 2) Small cities/Suburban Areas, 3) Balance of State/Statewide Continuums.

Among big cities, six (6) Continuums ranked highest (in order): 
1)  New Orleans
2)  Virginia Beach
3)  Boston
4)  Nashville/Davidson County
5)  Miami/Dade County
6)  Sacramento

Lower performing big cities included:
1)  Dallas
2)  Oklahoma City
3)  Kansas City
4)  San Jose
5)  Wichita
6)  Omaha
7)  Minneapolis
8)  Houston

The two highest scoring CoCs in the country are Jackson, Mississippi and Patterson/Passaic, New Jersey, both of which are small city/suburban areas.  High performing Balance of State CoCs include Ohio and North Carolina.

The rankings are provided to give an overall assessment of performance and do not reflect challenges and successes in each community or the resources available .  The primary goal is to identify Continuums that appear to be performing at a high level (High Performers) and Continuums that continue to face challenges (Lower Performers).

Data collected by Continuums from client self-report still contain many sources of error.  The rankings are provided as a baseline.  As more detailed information is publicly-available, the rankings can improve over time.  The NHIP understands that there is always subjective element with any ranking system.  HUD has a rating system to determine High Performing Communities but has not publicly released these rankings.

The Full Report is available HERE for review.  The NHIP welcomes comments and suggestions to improve future rankings including suggestions on performance thresholds.  The 2018 performance data will likely be available in summer/fall 2019.  Since there are no quality checks on HMIS data, the NHIP suggests using multiple methods to judge the effectiveness of a homeless continuum.  The NHIP also reminds readers that increases or decreases in total homelessness in a communityi is not necessarily correlated to effective services systems since larger macro-economic and social forces trends drive the continued flows into homelessness.

* * * * * 

2018 Chronic rates vary widely across CoC

(January 2019)  The NHIP analysis of chronic rates among unsheltered and sheltered persons shows wide variation in the estimated prevalence running between 0% and 100% depending on the Continuum.  The national estimate based on all CoC 2018 PIT data is 31% for unsheltered, 13% for sheltered including 16% for sheltered individuals and 4% for sheltered families individuals.

The Analysis of Chronic Prevalence among Unsheltered persons reveals several Continuums reporting a 90%+ rate including two CoCs estimating 100% of unsheltered as chronic.   A total of 16 CoCs reported a rate of 70% or higher, while 42 CoCs reported a rate of 5% or lower.  Twenty-seven CoCs reported zero chronically homeless persons. 

The Analysis of Chronic Prevalence among Sheltered persons found 6 CoCs reporting a prevalence of 50% or higher, while 69 reported 5% or lower.  While it is expected that CoCs with more family emergency shelter beds will have a lower chronic rate, the variation reported supports the difficulty in operationalizing the current federal definition.

Results from this analysis are similar to previous analyses performed on data from other PIT years.  The NHIP advocates for a change in the definition using only HMIS information to calculate chronic prevalence.  This would allow a more rigorous scientifically-sound approach that would give incentive to enter more frequent information on unsheltered persons.

* * * * *

HMIS participation rates slide in 2108

(January 2019) The national HMIS participation rate for emergency, transitional, and safe haven homeless shelter facilities fell from 82.6% in 2017 to 80.8% marking the first significant decline in participation rates since the HMIS mandate began in 2003.  A total of 65,792 beds out of the national total of 341,939 year-round, non-domestic violence beds did not enter data into community HMIS which provides the basis for the national data.

Participation among Permanent Supportive Housing programs fared better with an increase from 73.7% in 2017 to 75.3% among the 350,000+ PSH beds, although non-participation is greater with almost 90,000 not reporting data, including a large number ( of VASH beds in communities where the VA refuses to participate in the HMIS process.

A total of 95 Continuums, or approximately one-quarter, reported that 100% of shelter homeless beds were reporting to the HMIS.  Seventy-four Continuums reported less than 60% participation including Denver, Los Angeles, and Dallas.

CLICK HERE for a summary of HMIS shelter participation rates 


* * * * * *

Rise in Emergency Shelter Usage Drives Increase in PIT 2018 estimate

(December 17,2018) The U.S. Department of Housing and Urban Development released its PIT 2018 results on December 17th showing an uptick in persons living in emergency shelters that drove the overall count sightly up 0.3% to 552,850 from 550,996 in 2018.  This marked the third straight year of increases or stagnant progress on reducing the one-day count estimate.

The NHIP has completed a 2017 to 2018 PIT estimate comparison for each of the nearly 400 CoCs including changes in counts at emergency and transitional shelters and enumerated through unsheltered counts.  This is the first year that HUD disaggregated its shelter counts in the HUD PIT Summary Files released to the public making it easy for researchers and advocates to follow key changes in transitional housing usage.

Emergency shelter utilization on the 2018 PIT date rose 5% to an all-time high of 275,705 persons, while unsheltered homeless rose 2% to 194,467 persons.  The unsheltered count includes approximately 3,000 persons sheltered temporarily in hotels who were evacuated from hurricane-hit areas including Puerto Rico.

Continued closures of transitional housing facilities helped prevent larger overall increases since persons defined as homeless and living in these transitional facilities by 16,263 which countered nearly all of the 17,613 increase in shelter counts of those in emergency shelters and living unsheltered.  Previous reports have shown that the nation's emergency shelter capacity continues to increase, despite evidence that shelters also exacerbate homelessness in the long-run.

Evidence continues to build on how shelters maintain high levels of homelessness and the reduction of shelter capacity does not appear to influence unsheltered prevalence.  The NHIP has advocated the conversion of all shelters into permanent facilities, which can result in massive decreases in homeless prevalence throughout the country.

* * * * * * * * 

Nov 15, 2018

Homeless Vets PIT drops 5.4% in 2018

70 Communities that "Ended Homelessness" tally 5,273 homeless vets

HUD released the 2018 PIT estimates of homeless veterans from each of 400 Continuums of Care collected this past January 2018.  The overall national estimate dropped 5.4% from 40,020 to 37,878.  Communities that have been recognized as "Ending Homelessness" tallied a total 5,273 homeless veterans down 4% from 5,491 in 2017.  The NHIP has put together an Analysis comparing 2018 to 2017 with denotations of Continuums that have been recognized as "Ending Veteran Homelessness." CoCs recognized nationally that still have high PIT veteran totals include:

1.  Las Vegas - 594
2.  Portland - 448
3.  Houston - 443
4.  Atlanta - 399
5.  Philadelphia - 239

A review of the veteran count by shelter status shows a 14% decrease in veterans residing at transitional facilities including VA-funded transitional veterans-only facilities dropping from 13,876 to 11,986.  Unshelterd homelessness dropped 5% from 15,330 to 14,566.  Emergency shelter usage increased slightly by 185  veterans.  Emergency shelter capacity continues to rise across the nations as communities continue to open emergency shelters despite the HUD emphasis on Permanent Supportive Housing and Rapid ReHousing.


* * * * *

Advocate Requests HUD to de-certify San Diego PIT Count

See Full Letter Below

(May 29, 2018)  Michael McConnell, advocate and former Vice-Chair of the San Diego Regional Task Force on Homelessness  (RTFH), has submitted a written request to Norm Suchar, Director of the Office of Special Needs, to de-certify the 2018 San Diego PIT count submitted by the RTFH due to key methodological problems.

The primary reason for the de-certification request stems from the exclusion of persons living in recreational vehicles (RVs) in the 2018 report.  The 2017 PIT count included this population.  The 6% drop in unsheltered homelessness is likely due in part to this significant change.  San Diego used an estimate of 1.66 persons per RV to calculate a total number of persons (e.g. 300 RVs equals 500 persons). 

The NHIP notes that most media outlets reported on the "progress" achieved by San Diego without any caveat about the counting change.  The lack of a consistent definition prevents any year-to-year comparison.  The NHIP has found additional significant inconsistencies (see brief following letter) between the two years that support Mr. McConnell's argument.  To read the entire two-page letter, please click below:


Multiple PIT inconsistencies found in recent San Diego homeless report

(May 29, 2018) The NHIP has found multiple inconsistencies in the inventory of shelter programs included in the 2017 count compared to the 2018 count.  These differences do not invalidate the report - but they make comparison and evidence for progress  in reducing homelessness highly problematic.

Among the key problems is the exclusion of Recreational Vehicles (as discussed in the above brief).  The NHIP completed a comparison of census tracts for San Diego to determine which ones experienced large increases or decrease.  Below are the 14 Census Tracts with large decreases.

In,addition, there were seven (7) tracts that registered increases of 25 persons or more.  Large increases  or decreases should be routinely examined and confirmed through multiple stakeholders.  Duplication or re-counting of the tract may be necessary.

CLICK HERE for a full listing of San Diego's Tract-by-Tract comparison between 2017 and 2018.  Small rounding or transcription errors may be present.  Please notify the NHIP is you see any error or misrepresentation.

* * * * * *

HUD Budget set to increase

(May 21, 2018) Despite fears of large reductions in the HUD budget as proposed by the Trump Administration, the FY2018 $1.3 trillion Omnibus Bill passed in late March contains significant increases for HUD programs including homeless services, Housing Choice Vouchers, CDBG and Home programs. One of the largest increases is targeted for public housing repairs with a 41.6% increase in the Public Housing Capital Fund.  It is estimated that public housing repairs total in excess of $25 billion.  Homeless Assistance Funding is slated to increase approximately $130 million which will fund up to 25,000 new Permanent Supportive Housing units.

Additional provisions in the Low-Income Tax Credit program will reduce some of the impact caused by lower corporate tax rates which reduce the leverage of the program.  Below is a brief detail from Novogradac:

* * * * * * * * * * * * 

HMIS participation slightly lower; BOS CoCs struggle

Data from the recently-released 2017 Housing Inventory Charts show a small decrease in the overall HMIS participation rate among non-DV emergency, transitional and safe haven facilities.  A total of 290,151 out of 351,168 beds (84.6%) were reported to be entering data in CoC HMIS around the country.  This reflects a small, but important, drop from the 2016 rate of 84.8%.  Participation had increased every year since implementation began in the mid-2000s.

An analysis of the 34 Balance of State CoCs reveals significantly lower participation rates.  BOS CoCs typically receive less federal homeless planning funding despite having a larger and more difficult area to enumerate.  The chart below lists participation rates sorted from highest to lowest.  

* * * * * * * 

Housing and Homeless Bill of Rights and Obligations

The NHIP proposes a new Bill of Rights and Obligations that addresses some of the long-time pathologies and perversions in U.S. Housing policy:

Right and Obligation 1:  All Housing is Permanent
All facilities or shelters will not have any length of stay restrictions.  Persons staying in a shelter are considered to have a home.  They may stay at this home until they desire to move to another location. Facilities will charge a modest fee for long-term stays like any SRO, hostel or hotel.

Right and Obligation 2:  All Housing facilities must provide a  raised bed or cot for sleeping
All housing/shelter facilities will provide a minimum of a raised cot or bed that is above the floor level.  If facilities are unable to provide this basic level of housing comfort, they should close.

Right and Obligation 3:  Rent will not change due to Income changes
While initial eligibility for a housing facility may have general income eligibility rules, rents do not change based on changes in income or addition of family members.  This is currently a discriminatory practice under the Brooke Amendment which has caused much of homelessness and household fragmentation.

Right and Obligation 4:  Sheltered persons are not defined as homeless 
Persons living in a facility are no longer defined and/or called "homeless persons".  Amendment 1 allows these persons and household to stay as long as needed.  The practice of labelling person as "homeless" when they have a place to sleep is discriminatory.  

Right and Obligation 5:  Persons using Domestic Violence shelters are not defined as homeless
Persons seeking assistance at domestic violence facilities are not defined as homeless since Domestic Violence is a separate social problem related to an on-going crime.  The domestic violence housing utilization should be reported separately from the number of persons in unsheltered situations and other housing facilities. Household characteristics including family income are significant different than households experience general homelessness.

Right and Obligation 6: Single persons needing housing will accept shared living apartments and houses
Shared living (2 to 4 people in an apartment or house) is a normative standard for most people who are not married and not wealthy.  Providing an apartment to a person with housing stability issues is an expensive option that should be limited to persons who need solitary living due to diagnosis, typically less than 10% of population.  Shared living placement can create thousands of additional placements with the same amount of public funding.


More Rights and Obligations to come !


Published 11.20.17

20 Most Expensive Cities and a History of High FMR rates

The NHIP has completed an analysis of the Top 20 most expensive cities in the U.S. using the recently released Fair Market Rent standards for 2 bedroom units.  The chart below shows the Top 20 cities with both their 2018 and 1983 FMR rates and the annualized rate of rent increase over the past 35 years.

Here is a top 10 list for selected years since 1983.  Interestingly, cities in Alaska used to be the most expensive along with Chicago.  Chicago has seen only average rental increases over the last three decades.


* * * * * * * * *

Built for Zero vs. Federal Benchmarks

Continuums have been inundated in recent years with national initiatives by government entities and technical assistance groups to achieve various empirical benchmarks in homeless services.  Over the last couple years, both Community Solutions and the USICH have introduced veteran-focused initiatives aimed at showing cities are "ending homelessness" and have needed infrastructure and capacity in place to assist all veterans experiencing homelessness.

Confusion between these two initiatives has been exacerbated by the changing names of the campaigns (Functional Zero, Vets Zero, Zero 2016, Built for Zero, Less than Zero (oh, that was a book).  The new Federal Benchmarks begun by USICH in colloboration with the VA is the more recent.  The facf that HUD has not mandated participation or otherwise assigned points in its annual application for participation in either campaigns also lends confusion about the important of whether communities need participate.

The NHIP has been providing technical assistance to CoCs about the benchmarks and has found understanding the differences and explaning the differences to providers quite challenging.  The table below summarizes some of the key differences in characteristics.

Both initiative share important problematic similarities:

1.  Neither has any stated empirical follow-up to justify continued "Zero compliance" or evidence that federal benchmarks are being maintained.  Maintaining the progress is critical or the achievement is a little hollow.

2.  Both use the phrase "ending homelessness" which is affront to veterans who remain homeless in a jurisdiction making the claim.  Can we claim DV shelters "end DV"?

3.  Neither has a systematic external review by an outside evaluator especially a review of the annual PIT count which should reinforce outreach efforts have engaged all veterans.

If these initiatives are to continue, more rigor is needed to support the pronouncements of "ending homelessness."  The NHIP continues to monitor homeless counts of cities and states that have claimed to end veteran homelessness.

* * * * * * *

(published September 2016)

State of Oregon PIT counts rises 5% since 2015

Recently released 2017 PIT Statewide Data from Oregon revealed a 6% increase in homelessness from 2015 to 2017 with the total count rising from 13,176 to 13,953.  Oregon PIT Data are posted on public tableau and available to view by County.  Oregon's increase follows the general trend of increases in west coast states.

The rise in Oregon's PIT estimates were led by an 8% increase in unsheltered persons rising from 7,357 to 7,967.  Reports from the Balance of State Continuum covering 28 counties reported an overall  7%  increase with 8 BOS counties  experiencing significant increases of between 40% to over 300%.  In addition,  four medium sized BOS counties (Coos, Columbia, Josephine, Klamath) reported decreases between 24% to 50% from 2015 to 2017.  

The NHIP continues to advise people that large one or even two year count differences are often due to methodological differences in the counts not actual changes unless specific interventions can be identified that caused such an increase or decrease.  The Oregon study points out problem with completing unsheltered counts throughout its many urban and rural counties.

* * * * * * *

Data from 32 CA Continuums show 14% increase

Data compiled by the NHIP show a 14% increase among the 32 California Continuums who have publicly released their 2017 Point In Time data.  Although a majority of CoCs report decreases, increases often quite large from heavily-populated urban areas particularly in the Bay area and Los Angeles County fueled the increase from 110,794 in 2016 to 126,856 in 2017.  The remaining eight (8) CA CoCs not reporting are relatively small and would likely not impact the final difference.

The State of Washington has also reported a 5% increase since 2016 bringing their total close to 22,000.  Reports from Portland and Eugene, Oregon also report substantial increase.  It is likely that all three Western states will show increases when totals are finalized.

* * * *

Updated: Results to Date from 2017 Point-in-Time Count

The following communities have publicly released their 2017 PIT count estimates.  The deadline has passed for submission of PIT and HIC information  into the HUD Exchange.  Communities appear more reticent to release information publicly compared to last year.  HUD now weighs progress toward reduction of the PIT count in its annual competition for McKinney Vento Homeless Assistance Funding.  Communities that have released counts include the following:.

Decreasing Homelessness

  • The state of Kentucky reported a 4% statewide decrease to 4,025 in 2017.
  • Chicago FINAL PIT report reported a 3% decrease in total homeless although unsheltered rose 22%
  • Contra Costa PIT 2017 report found a 7% decrease  with a total of 1,607 compared to 1,730  in 2016.  Veterans count declined from 136 to 99.
  • Charlotte/Mecklenburg County decreased 12% from 2016 although unsheltered homelessness increased from 185 to 217.  Download Charlotte 2010-2017 Report.
  • Roanoke, VA (Blue Ridge Area) - Reports a 19% decrease from 331 in 2016 to 267 in 2017.  2017 Blue Ridge Report and Five-Year Blue Ridge Report available.
  • Tarrant County (Ft. Worth, TX) - Slight decrease of 0.7% from 1,938 to 1,924.
  • Topeka, KS - Down from 417 to 365 in 2017.
  • City of Detroit down from 2,335 in 2016 to 2,078 in 2017.
  • State of Connecticut down 13% from 3,911 to 3,387.
  • Washington DC - Down 10.5% from 8,350 to 7,473.  Read the Summary PIT 2017 Fact Sheet.
  • Santa Rosa/Petaluma/Sonoma County CoC. CA-504 totalled 2,835 down 2% from 2,906 in 2016.
  • Ventura County, CA - Down from 1,271 to 1,152 in 2017.
  • New Bedford, MA - Down 25% primarily due to closure of transitional shelters.
  • Pierce County, WA - Decreased from 1,762 to 1,321
  • Ann Arbor, MI - Down from 342 in 2016 to 300 in 2017.
  • Fairfax County, VA - Down from 1059 in 2016 to 954 in 2017.
  • San Bernardino, CA - Down 1% from 1,887 to 1,866.
  • Barnstable County (Cape Cod MA) - Down 70 persons.
  • Hawaii (Neighbor Islands) - 31% reduction in unsheltered homelessness.  Possible methodological changes may have augmented arge decreases in "street" counts.
  • Austin, TX - Overall 5% decrease in one-day count.


Increasing Homelessness:

  • Texas Balance of State 2017 PIT report shows an overall 18% rise in homelessness including a 49% rise in  one-day unsheltered homeless.
  • Tulsa reported a 2% increase in from 811 to 824.  Unsheltered homelessness rose from 150 to 174 person.
  • New York City reported a 39% rise in unsheltered homelessness.
  • Santa Cruz - Up 14.5% from 2015.  Veterans rising over 50%.  See Santa Cruz PIT report.
  • Sacremento - Up 30% from 2015 to 3,665 from 2,822 in 2015 including over 2,000 living unsheltered.
  • Oakland/Alameda County -  Up 39% from 4,040 to 5,629. See 2017 PIT Results Brief.
  • San Francisco - Increase in veterans/chronic homeless/single adults.  Decrease in youth/families. Overall count nearly identical to 2015.  See 2017 PIT Results Brief.
  • Dallas, TX - Increase in unsheltered from 739 to 1,087.  Overall small decrease of 0.6% from 3,810 to 3,789.
  • Orange County - Up from 4,319 in 2016 to 4,792 in 2017.
  • Wichita/Sedgwick, KS - Up 0.7% from 571 to 575.  Veterans up from 45 in 2016 to 57 in 2017.
  • Seattle/King County - Overall increase of 8.5% with 22% rise in unsheltered persons.
  • State of Washington - Statewide count of 21,843 up from 20,827 in 2016.
  • Los Angeles County - Up 23% to 57,794 persons. Unsheltered increase 8,000+ and sheltered increased 2,500+ since 2016.
  • Maricopa Couny - Unsheltered Homeless increased to 2,059 up from 1,646 in 2016.  Veterans homeless up 9% to 489 persons.  Chronic homelessness up 26%.
  • Pasadena, CA up 575 up from 530 in 2016.
  • Central Oregon count up 31% from 2015  to 778 persons
  • Statewide Vermont count up 19% to 1,225 persons
  • Naples, FL (FL-606) up from 545  in 2016 in 621 in 2017.
  • Riverside County, CA  - Up from  2,165 to 2,413 in 2017
  • Snohomish WA - Unsheltered count rises from 471 to 515.
  • Grand Rapids, MN - Increase over 2016 from 71 to 155.
  • Hawaii (Oahu) - Increase of 0.5% over 2016 including a 7% increase in unsheltered homelessness.
  • Merced County (CA) - 12.5% decrease in total homelessness.  Only 17 unsheltered veterans total.


Survey Results:  Transitional Housing Conversion popular; 73% say Chronic Homeless definition problematic

The NHIP recently asked readers their opinion on different aspects of homeless and housing services on a 15 question survey.  The full SURVEY RESULTS Part 2 are available.  Highlights of the results include:

  • 71% of readers agree that most or some of a CoCs Transitional facilities should be converted to Permanent Housing.
  • 23% of readers feel the Chronic Homeless definition is ridiculous to implement. An additional 50% believe it misses chronic people since it is too difficult to implement.
  • 82% of readers believe individuals living doubled-up should be considered homeless either always (21%) or depending on circumstances (61%).
  • 59% believe emergency shelters should never charge fees.
  • 63% believe emergency shelters should not be required to have cots/bed frames in order to eliminate the practice of people sleeping on mats on the floor.
  • 43% believe homelessness can be ended with sufficient resources and collaboration.
  • 21% believe agree that there should be a category of "super-chronic" to denote people homeless for 5 or more year.

Updated findings from the initial SURVEY RESULTS Part 1 are also available.  A total of 255 persons completed Part 1 and 98 persons completed Part 2. 

Survey Part 2 is still open for persons wanting to respond.


* * * * * *

Published 03.25.17

NHIP estimates over $10 billion spent annually on homeless facilities

The NHIP has completed a State-by-State and Continuum-by-Continuum analysis of the estimated total spending on homeless facilities including emergency shelters, transitional housing, safe havens, permanent supportive housing, and rapid rehousing units.  The estimates use the total number of facilities as listed on the 2016 Housing Inventory Chart for each Continuum.  Facility bed costs are estimated per bed or unit and use general estimate from a review of existing literature and experience at several Continuums throughout the country.  The NHIP analysis estimates that more than $10 billion is spent from federal, state, local and private sources to operate these facilities 

NHIP took estimates of the annual cost of each type of facility and the number of facilities (beds or units) listed in the 2016 Housing Inventory Charts. Some CoC costs are hand-adjusted using a cost-of-living factor ranging from 0.85 (e.g. rural Louisiana, Tennessee) to 1.35 (e.g.New York City, San Francisco).  Cost of facilities vary widely, especially emergency and transitional shelters. There is no comprehensive literature available that contains accurate unit costs by Continuum area and facility.  

The NHIP used the following annual cost assumptions of the major types of HIC facilities based on existing estimates and experience working in several Continuums:

1)  Emergency Shelter Bed for single adult ($18,000)
2)  Emergency Shelter Unit for Family ($30,000)
3)  Transitional Housing Unit ($15,000)
4)  Save Haven Unit ($35,000)
5)  Permanent Supportive Housing Unit for single adult ($12,000)
6)  Permanent Supportive Housing for Family ($18,000)
7)  Rapid Rehousing unit assistance for single ($4,000)
8)  Rapid Rehousing unit assistance for family  ($6,000)

The NHIP does not make any claim of accuracy especially at the Continuum level.  The NHIP does feel that the true number is likely between $8 to $12 billion annually. This estimate provides a starting point for more comprehensive estimates using actual budget data.  Validation could be determined if a Continuum or State wishes to add budgets for homeless residential services in their jurisdiction.

There are many issues to think about when determining costs including the fact that many services and costs are often provided free such as free rent or low cost building expenses and free or low-cost meal provision through in-kind services.  For readers interested in changing some of the cost assumptions and re-running the analysis for their Continuum or State, NHIP can provide the full spreadsheet upon request. 

This total does not include non-residential homeless services that are not listed on the Continuum Housing Inventory Chart Including federally-funded PATH Outreach funding ($65 millIon, Health Care for the Homeless funding ($400), non-residential HOPWA funding, and other initiatives that would put the total amount of federal, state, and local spending over the $11 billion mark. 

The NHIP welcomes feedback on this analysis.

* * * * * 

Survey Results:  1/3 feel homelessness is getting worse; Only 25% think street counts within 10% of accuracy

A total of 235 respondents (thank you!) completed the first NHIP 10-item survey covering opinions on important subjects within homeless services.   The full report contains a complete breakdown by question. Click on survey (right)  to read and download.  Key highlights of the survey include:

  • 43% feel the unsheltered count should occur with greater frequency than annually.  The current HUD mandate is every two years.
  • 20% agreed that persons living in Transitional Housing should not be considered homeless.  
  • 31% feel there is too much emphasis on chronic homelessness funding.
  • 65% feel that HUD is definitely (17%) or somewhat (48%) on the right track.
  • 11% think that Transitional Housing should receive no federal funding.

The intent of the NHIP is two-fold: 1) To track attitudes from key stakeholders working in the homeless field, and 2) To use questions to elicit different ways of thinking about key axioms guiding homeless services (e.g. Should people in DV shelters or Transitional Housing be defined as homeless?)

While the survey sampling is not methodologically rigorous, the NHIP does feel that given the number of respondents and the distribution of readership of the NHIP across the country, there is a good likelihood that results reflect the attitudes of people working at the management level of homeless services in the U.S.

Published 02.23.17

* * * * * * * * 

Published 01.23.17

Will the Senate Housing Appropriations Committee house their 44,000+ homeless constituents?

Homeless estimates in the home states of the Senate Appropriations Subcommittee that controls funding for homeless services shows that the one-day prevalence of the Senators constituents with serious mental illness who are homeless is over 44,000.  The list includes:

Additional funding of $450 million is needed to fund 22,000 new Housing vouchers.  The chart above shows an estimate amount of funding that would be distributed to each of the Senator's states if the needed funding were approved.  The State of California has tremendous need due to continuing growing population of homeless persons with serious mental illness. 

In FY16, a $250 million increase in funding was coupled with a transfer of $200 million in CDBG million to cover the needed HUD funding increase for chronically homeless persons, the majority of whom have serious mental illness.

The NHIP estimates that a minimum of 40,000 new vouchers for SMI homeless experiencing chronic homelessness are needed using the same successful strategy that has cut chronic homeless among veterans by more than 50% in the past 6 years.

* * * * *

Vouchers for SMI Homeless critical to ending long bouts of unsheltered homelessness


Homeless estimates from the annual HUD application reported a total of 45,000+ persons with serious mental illness (SMI) who were experiencing unsheltered homelessness on any given day.  A total of 14 CoCs reported more than 500 SMI homeless persons.  See table below. For a completed breakdown, CLICK HERE for the full report.

A total of 9 CoCs from California were among the 14 with 500 or more SMI homeless.  Both CoCs in Hawaii were also represented as well as CoCs from New York City, Las Vegas, and BOS Georgia.

The NHIP has estimated a minimum need of 40,000 vouchers specifically for chronically homeless persons with serious mental illness.  In comparison, the VA-HUD VASH initiative has provided over 87,000 vouchers for homeless veterans.  When the VASH initiative began in 2009, estimates of one-day veterans homeless prevalence ranged between 80,000 and 100,000.  Using this simple comparison, the NHIP estimates the need for 40,000 new vouchers and reassignment of additional 40,000 vouchers as they turnover in current PSH programs.

A review of the SMI prevalence also revealed a large variation between CoCs in the percentage of unsheltered homelessness that are estimated to be seriously mentally ill.  Below are the select CoCs with very high and very low SMI percentages.

High Prevalence                                       Lowest Prevalence
1.  CT-503 Bridgeport - 86.5%             1.  WA-500 Seattle 4.4%
2.  CA-504 Santa Rosa - 69.8%            2.  FL-509  Fort Pierce 5.8%
3.  KY-501 Louisville -  68.8%               3.  CA-503 San Diego 13.6%   

The average estimated prevalence of SMI across all Continuums is 25.7% for unsheltered homeless persons and 16.5% for sheltered persons.  NHIP recommends that Continuums who have a prevalence much higher or much lower than the national average review their estimation procedures.  The use of untrained volunteers or staff to collect data during Point In Time counts reduces the ability to do estimation procedures with any accuracy.  

* * * * * * * * * * *

Published Jan 11, 2017 

NHIP "Purchasing Power" Analysis:  High FMRs reduce True Homeless Funding level for high-cost areas

The NHIP has completed an Analysis on 50 Select Communities that adjusts HUD funding for the cost of housing using the 2016 studio FMR rate.  Since HUD funding essentially pays for rental assistance, adjusting the funding by the cost of housing produces a "Purchasing Power" level which can allow for a better comparison of the ability of funding to reduce homelessness.  

Top Five Communities impacted
by Purchasing Power reductions:

1.  New York City (-$57.6 million)
2.  Los Angeles (-$38.5 million)
3.  Chicago (-$19.6 million)
4.  San Francisco (-$18.3 million)
5.  Seattle (-$14.7 million)

Adjusting Homeless Per Capita Funding (Funding divided by PIT Count) using the new Purchasing Power funding levels yield even greater disparities between high cost areas with high homeless prevalence.  The table below shows select communities with the highest and lowest funding levels after controlling for purchasing power.  The average level of Funding Per Capita is $3,522.

Lowest PP-adjusted Funding Per Capita
1.  New York City ($678)
2.  Honolulu ($854)
3.  District of Columbia ($1,165)
4.  San Diego ($1,214)
5.  San Jose ($1,366)

Highest PP-Adjusted Funding Per Capita
1.  Wilkes-Barre ($31,826)
2.  Cincinnati ($18,453)
3.  Cleveland ($18,113)
4.  Pittsburgh ($16,605)
5.  Ann Arbor ($13,081)

Currently, the HUD funding formula does not take into account the cost of housing. There is also no effort to connect persons living homeless in high cost areas to lower cost areas, a strategy used by many non-homeless households.  HUD new FMR standards will actually allow many communities to use even higher FMR levels to allow for increased deconcentration of PSH units.

Currently, HUD rent rules do not provide any incentive to recipients to choose lower cost housing since recipient rent is based on their income not on the cost of housing.  The NHIP feels this is a one of the fundamental flaws of the rent regulations.

The NHIP recommends HUD rule changes to incorporate normative alignment of rental housing regulations including great portability of subsidies, non-linkage of rent to income, elimination of the 30% rent rule which provides a great disincentive to form multi-earner households, and specifically allowing HUD housing subsidy recipients the right to share their housing with family or friends without impact to rent or compliance.

The NHIP believes that HUD policies since the 1960s - particularly the 30% rent rule and its regulations - are responsible for much of modern homelessness and housing instability. 

Analysis of 2016 Homeless Funding compared to Homeless Prevalence

A new NHIP analysis comparing HUD homeless funding awards with the most recent estimates of homelessness shows continued large disparities among Continuums reinforcing the need for HUD to adjust funding for communities significantly underfunded.

The NHIP created two adjusted funding measures:  1)  Measure 1 represents Funding Per Homeless Person by dividing the HUD homeless funding award for the CoC by its PIT homeless count; 2) Measure 2 represents  Funding Per Homeless Services by dividing the HUD homeless funding award for the CoC by its PIT homeless count plus its PSH count.  Since more than 70% of homeless funding supports PSH units, a measure incorporating these totals is needed to give a more balanced picture.  Taken together, these two measures offer the ability to identify communities with sigificant underfunding compared to its peer Continuum.

Measure 1 produced an average funding level of $3,531 with a high of $23,180 in PA-503 Wilkes-Barre/Hazelton/Luzerne and a low of $3.55 in CA-524 Yuba/Sutter. Measure 2 produced an average funding level of $2,180 with the high of $8,027 in PA-503 and a low of $3.55 in CA-524.  The report table is sorted by Measure 1.  Below are the top 20 most "underfunded" large Continuums (CoCs with more than 1,000 homeless):

A review of the top twenty "most underfunded" is populated with many CoCs from California (8) and Florida (6) in addition to two other West Coast/Pacific Continuums.

Continuums that continue experience increases in homelessness also showed Measure 2 levels well below average including Hawaii BOS ($625), District of Columbia ($1,421), and San Diego ($1,423) . 

The NHIP will provide additional analysis on funding levels in the coming weeks.

* * * * * * * * * * * * * * * * * * * * * * * *

Published 12.20.16

Homelessness among Adults with Serious Mental Illness rises

In the latest 2016  estimates for one-day homelessness collected in January 2016, data show a rise of 3.6 percent in the estimated prevalence of homelessness among adults with serious mental illness (SMI).  The 2016 estimates from 402 Continuums across the country totaled 107,801 compared to 104,083 in 2015. Increases in sheltered and unsheltered homelessness among SMI persons were reported with a total of 61,846 in emergency or transitional shelters and 45,955 living in the streets, parks and other places not fit for human habitation.

The NHIP has compiled a state-by-state prevalence report from the individual subpopulation reports.  The report also calculates the SMI prevalence as a percentage of total one-day homelessness.  HUD has not provided a comparison chart for homeless subpopulation estimates.

Eight states led by California, Florida, Hawaii, Nevada, New York, Oregon, Texas, and Washington reported a one-day SMI unsheltered homeless count of 1,000 or great.  California had by far the highest number of SMI unsheltered homeless with a one-day estimate of 24,269 followed by Florida with 2,791.

The State of Maine reported the highest percentage of SMI homeless among all homeless persons at 42.2 percent, while South Dakota had the lowest estimated prevalence rate of 7.7 percent.  Overall, the national average is estimated at 19.6 percent of all homeless, with a sheltered rate of 16.6 percent and unsheltered rate of 26.1 percent.

Estimates of seriously mentally ill are generally thought to undercount the true prevalence since PIT estimates generally rely on self-reports.  Serious mental illness includes schizophrenia, schizo-affective disorder, bipolar depression, post-traumatic stress disorder, and major depression.

The NHIP has called for the creation of 40,000 new Permanent Supportive Housing vouchers at an annual cost of $500 million to reduce by half the prevalence of SMI homelessness.

Published 12.05.16

Homeless Increases Skew toward the West Coast/Pacific

A new NHIP Report using recently released HUD data on 2016 Point-in-Time sheltered and unsheltered homeless estimate show that 13 of the 15 Continuums with the largest increases in unsheltered homeless are all from West Coast/Hawaii states or those neighboring (Idaho or Arizona). The 15 areas include:

These fifteen areas reported a combined 9,186 increase from 51,550 in 2015 and 60,736 in 2016 for an overall increase of 17.7%.

The report revealed that among all 402 Continuums, 150 CoCs reported increase with 251 CoCs showing decreases in overall one-day homeless estimates.  Unsheltered estimates which experienced an overall increase of 1.8% featured 163 CoC with increases and 173 with decreases.  Sixty-six (66) reported no change, with almost all using 2015 data due to the lack of count in 2016.

Many CoCs like Honolulu and San Diego have been ata the top of lists of as areas receiving significantly less funding compared to their homeless prevalence.  The NHIP plans to do additional analysis comparing funding per capita and homeless estimate changes.


Published on November 22, 2016

2016 PIT Skinny:  Sheltered, Vets, Chronic Down; Unsheltered Non-Vet Homeless Up 4%

The U.S. Department of Housing and Urban Development has published comprehensive tables on the 2016 PIT count estimates from each of the 400+ CoCs around the country.

Data reported show an overall 2.6% decrease in one-day homeless estimates (564,708 to 549,928) from 2015 to 2016 with a 4.6% decrease in sheltered homelessness and 1.8% increase in unsheltered homelessness.  The NHIP notes that the HUD-reported 2.6% decrease is identical to the decrease reported by NHIP in mid-summer after collection of a 30% sample of CoC reports. 

Overall, 15 states led by Idaho (14%) and the District of Columbia (14%) reported an increase in total overall homelessness.  Other states reporting large increases included Delaware, Oklahoma, Colorado, Washington, Rhode Island, and Hawaii.  Among the 39 states and terrorities that reported decreases, North Dakota (-29%) and Vermont (-27%) and West Virginia (-24%) led the way. 

Homeless veteran counts continue to decrease with a 17% drop from 47,725 in 2015 to 39,471 in 2016. The massive funding through VASH and VA service connected increases has been a key factor in addition to the lower overall prevalence of veterans in the nation.  

The NHIP reports that Persons who are Not Veterans experienced only a 1.3% decrease from 516,983 in 2015 to 510,457 in 2016, less than half the total national increase. Unsheltered prevalence of non-veterans rose by 4% from 157,048 in 2015 to 163,290 in 2016. 

The NHIP will be providing additional reports and analyses using the newly released data at the State and CoC level in the upcoming months.  As always, the NHIP welcomes suggestions on reports.

Originally published 11.21.2106

Chronic Homeless Prevalence Estimates Report shows wide State-by-State Variation 

In the first analysis using newly published 2016 PIT Homeless , the NHIP compared the percentage of chronically homeless persons as a percent of the report one-day homeless counts.

Overall, 16% of homeless persons were identified as chronic with chronically homeless persons representing 8% of sheltered persons and 32% of unsheltered persons.  Variation in the chronic prevalence among the 54 states and U.S. Terrorities showed a range between 4% to 40%.  Sheltered chronics ranged from 0% to 21% and unsheltered chronic prevalence ranged from 4% to 75%.

Variation in chronic prevalence can be a marker for several issues: 1) Lack of consistent methodology used by CoC across the country, 2) Differential mix in the types of homeless facilities in a state, 3)  Real differences in population characteristics, and 4) General unspecific errors in tabulations by CoCs and States.

Prevalence rates that fall outside one standard deviation of the average (total  9% to 23%; unsheltered 18% to 46%) should be reviewed for all the above potential threats to accuracy.  The NHIP will complete comparisons at the CoC level in the next week.

It is critical for CoCs to correctly identify persons who are chronic as Coordinated Entry Systems using the HUD Orders of Priorities are being implemented across the country.

 Originally published 10.21.16

NHIP publishes State of Florida Homeless Report

In partnership with the Florida Coalition for the Homeless, the NHIP has published the "State of Florida Report:  2016 Homeless Census Estimates and Funding Need to End Chronic Homelessness".  The report was presented at the recent FCH Conference in Orlando, FL on October 5th.   

The report contains data from each of 27 FL Continuums and presents census estimates on all major demographic groups  including total homeless, sheltered, unsheltered, veterans, and chronically homeless persons.  Comparisons with 2015 data are included for many data tables.  The report also presents population prevalence rates of homelessness.

Data show a 7 percent decrease in total homelessness from 35,900 in 2015 to 33,466 in 2016.  Veterans homeless fell 35% from 3,926 to 2,902  

A highlight of the report is an estimate of the need for additional permanent supportive housing beds for each of the 27 Continuums in order to end chronic homelessness in their community.  The report estimates a need for 2,991 beds at a cost of $34 million annually.


Updated Sept 15, 2016

NHIP:  It's the Funding Level, Stupid !

Response to Open Comment on funding formula


The NHIP has formulated a response to the HUD Open Comment period concerning the funding formula. The response contains four key recommendations:

1)  HUD should solicit comments on modifications to ARD funding not PPRN funding which has only limited impact.

2)   While VA funding has increase 350% since 2001, HUD funding has increase only 27%.  Substantial funding increases must be advocated including the creation of 40,000 vouchers for chronically homeless individuals.

The NHIP reminds that the systemic problem is a lack of funding. The figure belows shows the increase in VA funding that has been the primary engine to reduce veteran homelessness.

3)  Allocation of Continuum Homeless Assistance funding should use the following five (5) indicators:
           30%  Current ARD
           20%  HIC bed count (ES,TH,SH,PSH,OPH)
           20%  Unsheltered count
           15%  FMR
           15% Level of new homeless entry

4)  Allocation of Continuum Planning dollars should provide a minimum of $25,000 with a reduced percentage for CoC ARD amounts above $10 million.


HMIS Vendor Distribution for CA and FL

The NHIP has collected information on HMIS choices for two (2) of the larger states - California and Florida.  A big thank you to Joe Colletti at Urban Initiatives for providing the CA listing and the Florida Homeless Coalition for its state listing.  A written request  to HUD in June 2016 for the complete HMIS vendor listing for all 400+ CoCs was not answered.  The NHIP welcomes other states (especially large states) and individual CoCs to send their vendor distribution.  More competition for HMIS software helps make better choices for agencies.


Updated 08.10.16

More Scores Posted!  CoCs share scores and debriefs

The NHIP received several responses to its request last week asking CoCs to share HUD scoring of their 2015 application. Three high-scoring Continuums - Philadelphia (PA-500), Springfield-Hampden (MA-504), Omaha-Council Bluffs (NE-501) - have given the NHIP permission to publicly release their scores (coupled with their applications).  NE-501 has also made available the detail of their scoring.  Two additional CoCs also provided their application score.  

1)  Kings/Tulare (CA-513) - Score of 160.75  (not bad!)
2)  Philadelphia (PA-500) - Score of 168.5  (good going!)
3)  Springfield-Hampden (MA-504) - Score of 179.5  (impressive!)
4)  Omaha-Council Bluffs (NE-501) - Score of 184 (debrief) (silver medal?). NE-501 is featured under "Know Your Continuum" on the NHIP website..
5)  BOS Hawaii (HI-500) - Score of 149.5 (And its Hawaii!)
6)  Charlotte/Punta Gorda (FL-602) - Score of 130.5 (it is a great vacation spot !)
7)  BOS West Virginia (WV-508) - Score of 153 (debrief)
8)  Dallas /Irving(TX-600) - Score of 139 (debrief) 
9)  Louisville (KY- 501) -  Score of 157
10)  Santa Ana/Anaheim/Orange (CA-602) - Score of 160 (debrief)

We encourage more Continuums - both high and lower performers - to share their scores and applications from 2015.  We will add to this posting as we receive additional scoring information.

The Houston Coalition also confirmed to the NHIP that they did receive the high score of 188.  The NHIP could not find a copy of the application on the the Houston Coalition website. HUD posted that the median score was 150. 

The NHIP wishes all CoCs success on their upcoming applications! As always, if the NHIP can be of any help, please contact.


HUD solicits Comments on Homeless Funding Formula

On Thursday, July 21st, HUD released its OPEN COMMENT solitation  on the Continuum Homeless Funding Formula. The following set of documents accompanies the announcement:

  1. Federal Register Explanation Narrative - MUST READ!
  2. Excel-based Alternative Funding Tool -  Designed by HUD to show funding scenarios based on weighting of HUD-selected funding factors.
  3. HUD-produced  Alternative Funding Scenarios with amounts for each of the 400+ Continuums.

Initial NHIP comments include:

Critique 1:  HUD is requesting comments on PPRN amounts NOT the Annual Renewal Demand (ARD) which is the effective CoC funding level.  It is unclear how modifications of an adjustment to the PPRN will significantly impact funding for the vast majority of Continuums. See example below:

EXAMPLE 1: CIty and County San Diego
1.  Current PPRN:        $13,323,923
2.  HUD Alternative Scenario #1:  $16,840,372
3.  HUD Alternative Scenario #2:  $16,673,326
4 . HUD Alternative Scenario #3:  $15,738,952
5.  HUD Alternative Scenario #4:  $16,628,640
6.  Current 2016 ARD: $17,363,919

Critique 2:  HUD specifically in the federal register states that it does not believe  current PIT counts (either sheltered/unsheltered or a combination or transformation) are reliable enough to be used for funding allocation purposes.  The NHIP specifically believes this to be untrue and short-sighted and that PIT counts can be incorporated given simple data transformations to adjust for the fact that unsheltered counts are known to be estimated (with error increasing as estimates rise) with varying degrees of quality.

EXAMPLE: CIty and County San Diego
Funding based on Homeless Prevalence
1)  NHIP estimate: $22,267,919
2)  HUD estimate:  $24,661,182

Critique 3:  HUD's excel-based scenarios are very limited since only HUD selected factors are included.  Impacts for cities with significant underfunding are not significantly altered, magnitude-wise or in comparison with other cities.  By the doing, it is limiting the ability of advocates to provide strong evidence for use of different factors.

Critique 4:  While HUD states that PIT counts are not reliable enough, it employs these very rates in its statistical correlation analyses to justify its own factor choices. 

In the coming weeks, the NHIP will provide additional information and analyses for Continuums and communities to help them craft responses to the Funding Formula. The OPEN COMMENT period is for 60 days or approximately September 20th (pending confirmation of publication date).

The NHIP acknowledges the work of Rep.  Scott Peters (D-CA-52) and his staff on advocating for changing the formula over the last three years.  


(posted 07.19.16)

What is the true cost of reducing Veteran Homelessness?

As communities push for achieving Functional Zero for veterans, it is important to understand and document the true total cost of achieving these reductions. Underreported in the media and undeniably an important factor is the tremendous increase in benefits compensation awarded by the VA since the campaign to end veteran homelessness began in 2009.  

The recent NHIP veterans compensation analysis reports that from 2010 to 2015 the number of veterans receiving disability compensation rose from 3.2 million to 4.2 million including over 250,000 Vietnam veterans receiving benefits for the first time.  In addition, the average annual disability compensation has risen from $9,400 in 2006 to $14,400 in 2015.  Currently, nearly 1 in 5 veterans receive service-connected disabilities.

In 2015 alone, approximately 95,000 veterans were newly awarded service-connected benefits at the 70% to 100% degree level.  A total of 313,000 veterans received first time benefits in 2015 totalling $3.62 billion,  The VA disability compensation  budget has more than doubled over the past six years from $50 billion to over $100 billion.

This voluminous increase in spending on veterans disability benefits must be understood within the context of reducing homeless prevalence among other important subpopulations. While SSI benefits to non-veterans have increased over the years, the rate is far less (single digits) and the federal SSI awards currently $733 per month equal only a 50% service-connected combined degree.

The direct VA homeless budget, listed at $1.6 billion annually, is as large as nearly two-thirds of the recently announced 2016 $2.5 billion HUD homeless services budget.  If the veterans population represents only 15% of the total homeless population, then common sense would dictate that funding to reduce homelessness among the severally mentally ill and other key populations would need a budget at least three-fold and as much as six-fold the current HUD Homeless Assitance budget.  The VA and HUD created and maintains 70,000+ vouchers for veterans at a cost approaching $1 billion alone.  Wouldn't at least 200,000 new vouchers be required for non-veterans?  Isn't a minimum increase of $3 billion needed?

The silence among leading advocacy groups for a significant increase in funding for non-veterans has always been troubling. The CoC budget has barely doubled in the past 15 years. Needless to say, homeless service workers pay is relatively flat as well hurting the system further by making it ever difficult to attact the talent needed. Taking away the reduction in veterans homelessness, homeless prevalence is relatively flat over the past few years. 

If success to reduce veterans homelessness is to be proclaimed, it must be understood that TREMENDOUS  INCREASES in funding that are as big as the entire $48 billion HUD budget have been a driving force.  If significant gains among the non-veterans population are to be seriously targeted, then serious commitments are a prerequisite.

Michael Ullman, Ph.D.
NHIP Coordinator


PIT UPDATE: Count down 2.6% with 144 CoCs reporting

With information collected from 136 Continuums, results show an estimated national decrease of approximately 6,500 compared to the 2015 count. Results from publicly-released one-day counts held in January 2016 report a total of 240,797 homeless persons compared to 247,171 from the same communities in 2015 for an overall 2.6% decrease.

The Homeless Action of Detroit (HAND) reported a 10% overall decrease from 2,597 in 2015 to 2,335 this past January.  Detroit experienced an increased in unsheltered from 151 to 193 while sheltered persons decrease 12% from 2,446 to 2,142.

Other recent reports include the South Carolina 2016 report from the SC Coalition for the Homeless representing the state's four (4) Continuums estimates an overall reduction from 5,354 to 5,050 including an 11 percent reduction in unsheltered homelessness from 1,897 in 2015 to 1,689 in 2016.  

Data from the Institute for Community Alliances (ICA), the HMIS lead for the State of Alaska, showed little change in statewide totals (1,956 in 2016 compared to 1,940 in 2015) although the Anchorage CoC total fell 9% while the BOS rose 12%.  The ICA become the lead agency during the past year and did not conduct the 2015 PIT count. Conclusions about trends using the 2015 data are not recommended by the ICA.

For a complete listing of the PIT 2016 results collected to date CLICK HERE.  


Originally posted 06.06.15

FMR rates rise further in already high-priced cities

An NHIP analysis of the change in FMR rates from 2015 to 2016 showed an estimated 4% average increase across the 4,500+ areas in the nation.  A review of the 20 most expensive metro areas revealed very large increases among a majority, including San Francisco, New York City, and Honolulu. FMR rents for a few cities including Santa Cruz and the suburbs of NYC were adjusted downward.  The table below provides details on the 2BR FMR rate change for the 20 areas.

Among all areas with HUD FMR rates, the estimated average 2BR FMR is $800 with two-bedroom rates below $600 in non-urban areas of Louisiana, Alabama, West Virginia, and Tennessee.  The  middle two quartiles (from the 25th to 75th percentiles) ranged from $659 to $934.  

The 20 fastest rising FMR rates in 2016 (chart below) included multiple areas in North Carolina and growing areas in the State of Washington and Lousiana.  FMR rate adjustments typically lag actual rent change by one year or longer. Increases in FMR rates also require increases in HUD funding for grants for lease costs like Permanent Supportive Housing to ensure that subsidies can be maintained as rents rise.  Due to the HUD rent formula, clients are not required to pay more in rent since payments are tied to client income not actual rental costs.

The NHIP would like to thanks Matt Simmonds of SimTech Solutions for sending us the FMR comparison file for review. 


And the HUD Funding Winner is. . .CA-507 Marin County!    

Using a ratio of the current Continuum ARD funding level and its PPRN amount, the NHIP has developed a measure to assess the success of Continuums in securing funding through the annual McKinney-Vento homeless assistance awards.

In the past decade, HUD has initiated a more competitive process for Continuum funding.  The ARD/PPRN ratio provides an empirical tool to assess how successful Continuums have been in securing bonus awards and priotizing housing projects that can increase its ARD.  In general, the higher the ratio, the better the Continuum has performed in the funding competition.  The PPRN essentially serves as a baseline funding level, while the ARD is the current funding renewal level of previously awarded projects.  The NHIP realizes that additional mitigating circumstances may exist at the Continuum level to explain some of the ratio variance.

The current 2015 ARD total is $1.83 million, while the PPRN total is $1.42 million. The overall ARD/PPRN ratio is 1.28.  Continuums with a ratio exceeding the national average have fared better over the past decade in securing funding.

1. Marin County (CA-507) Ratio =8.91
2. Mendocino County (CA-509) Ratio=7.79
3. Nashua/Hillsborough (NH-502) Ratio=6.48
4. Charles/St. Mary's/Calvert (MD-508) Ratio= 6.18
5. Garrett County (MD-510) Ratio = 4.83
6. Eaton County (MI-523) Ratio = 4.55
7. Eugene/Springfield/Lane (OR-500) Ratio = 4.37
8. Alexandria County (VA-603) Ratio = 4.14
9. Washtenow County (MI-509) Ratio = 4.09
10. Newburg/Middletown/Orange (NY-602) Ratio =  3.77 

For a complete list of Continuum Ratios, CLICK  HERE

1. Connecticut  -  Ratio = 2.68
2. Washington  -  Ratio = 2.29
3. District of Columbia = Ratio = 2.26
4. Maine - Ratio = 2.20
5. Alaska - Ratio = 2.09

For a State-by-State Ratio comparison, CLICK HERE

Note:  The Louisiana Balance of State Continuum was not included in this analysis.  A preliminary calculation yielded a ratio above 15.0.  The NHIP wanted to confirm the information on this Continuum before adding it to the analysis.


Originally Posted 04.01.16

NHIP Special Update: State-by-State Ranking of Homeless Assistance  "Per Capita" Funding 

In response to reader requests, the NHIP has tabulated Homeless Assistance Funding "Per Capita" Rates by State.  Per capita funding is a measure developed by the NHIP to assess disparities in federal homeless assistance funding.  Per Capita funding is calculated by dividing the Annual Renewal Demand and Planning funding (numerator) by the Point-in-Time Homeless Count (denominator).

The analysis ranks the 50 states, District of Columbia and U.S. Territories from highest-to-lowest homeless assistance per capita funding.  Data reveal that the State of Connecticut receives the highest amount of funding per capita with $8,464, while the State of Wyoming receives the lowest with $369.  The average funding per capita for the U.S. and  its territories is $2,822.  

Western/Pacific states, especially those with with low population densities, tend to receive very low per capita funding.  Rust belt and Northeastern states represented most of the top funded states.

The report also calculated the amount necessary to achieve an equalized level of funding using the U.S. average of $2,822.  The State of Illinois with a per capita rate of $6.675 currently receives $50 million more than it would receive with an equalized funding formula.  Ohio would require a reduction of $47 million, while Connecticut would lose $22 million.

Among states with below average per capita funding, New York would require an additional $79 million to achieve equalization, while California and Florida would need an additional $50 million and $27 million, respectively.  Wyoming, who has the lowest per capita rate, would stand to gain an additional $2 million in homeless assistance funding.

Many factors impact the level of federal McKinney-Vento homeless assistance funding including the performance of local communities in planning services and reducing homelessness.  This analysis offers information that can address wide disparities that may be hindering states’ ability to reduce homelessness, especially when compared to their peers.


Originally Posted 03.27.16

NHIP Special Report: Large Disparities in Federal Continuum Funding compared to Homeless PIT Count Prevalence

A comparison between Federal McKinney-Vento funding and Continuum Point-in-Time estimates of one-day homelessness revealed large inequalities in funding "per capita" received by the 400+ Continuums. The report entitled "An Analysis of the Allocation of the Federal Homeless Funding" aims to provide Continuums with information on potential federal funding barriers that may be hindering their progress to end homelessness.

Using 2014 data, the NHIP calculated an average of $2,822 in funding dollars per homeless person enumerated through the PIT across the nation. Continuum "per capita" funding ranged from less than $500 to more than $18,000 with 12 Continuums receiving more than $10,000  in funding. while 60 Continuums received less than $1,000 in funding per capita.  

Continuums from West Coast (Hawaii, California) and Sun Belt (Florida) tended to be over-represented among the lower funded Continuums, while Continuums from Northeastern states (New York, Ohio, Pennsylvania) tended to receive high funding levels. The report contains "per capita" rates for all 400+ Continuums.

The NHIP calculated that an equalization of funding would require movement of approximately $500 million in annual funding from high to low funded Continuums.  The report strongly recommends that the annual $300 million in "bonus" funding be employed to address large funding inequities between Continuums.

The report develops several diagnostics including a ratio of homeless-to-permanent housing beds to better understand and compare the structure of homeless bed inventories across the nation and its relationship to funding and success in reducing homelessness.

The report recognizes the limitations of the analysis and the problematic nature of all homeless services data. 


Originally posted 03.14.16

NEW REPORT:  Domestic Violence Beds within Continuum

The NHIP has published a new report on the prevalence of Domestic Violence beds within Continuums (as listed on the HIC) and at the State-level.  The report provides comprehensive data on the number of DV beds and the percentage of beds designated as DV.  A total 50,267 or 11.9% of all Continuum Emergency, Transitional and Safe Haven beds as designated for DV survivors.

Key recommendations from the study include requiring Continuum to report separately the domestic violence ned census and utilization in Continuum Point-in-Time counts.  The PIT DV counts can augment data currently collected the National Network to End Domestic Violence.  Additional recommendations and findings can be found in the report.

The new 2015 NNEDV Domestic Violence shelter count report can be found HERE for comparison.  Their annual DV count occurs in mid-September of each year and tallies utilization of shelter and services at over 1,900 domestic violence programs in the U.S.


Originally posted 03.01.16

Who can use the NHIP website?

One of the purposes of the NHIP is to provide a central place for persons working in homeless services to find and share important data concerning homelessness in their own community, neighboring communities, and national trends.

Many types of people can find the website helpful including:
-  Agency management
-  Grant writing staff or consultants
-  Foundations and private funders
-  Media Outlets (local or national)
-  Legislators and policymakers
-  Researchers and students

The website may especially be helpful to agencies and communities that have limited funding or capacity to find and write data that empirically describes the problems in their communities.

If you or your agency is looking for data and cannot locate it on the web or on our website - please feel free to contact us at and we will gladly help you out - NO CHARGE !


Originally posted 02.09.16


We are very proud and excited to launch this new collaborative grassroots effort - the National Homeless Information Project (NHIP). The mission of the NHIP is to increase knowledge and understanding of homeless services by working collaboratively across the nation.

For too long, the skills of hundreds of homeless service professionals lay underutilized as an engine to propel a greater understanding of homeless services and to foster an increased knowledge base of professional data collection and data analysis skills. The NHIP will provide that bridge among peers in an environment of openness and support.

The last decade, especially the last five years, has witnessed tremendous changes in the  strategies of homeless services and increased burdens on homeless services data systems. The NHIP will specifically focus on supporting all individual professionals and Continuums of Care to build a stronger foundation in the area of homeless data collection and to facilitate sharing homeless data from across the country to better inform both decision-makers, advocates, and homeless line staff workers.

The NHIP will also engage in original research that can illuminate disparities  across the nation and build knowledge in important topics impacting homeless services. We hope to build on data reported from the Exhibit 1 Continuum and engage professionals to contribute additional data and analysis that can provide an additional voice in reporting on progress towards reducing homelessness.

There is no cost to joining the NHIP effort. You can join by simply reading the information we will share through our website and emails. For those professionals interested in working to direct and disseminate the content, we invite you to join one of our Guidance Team

We look forward to getting started, and look forward to hearing from you and building collaborations across the country.


Thank you -

National Homeless Information Project 

P.S.  Please excuse any problems in the website! And let us know about any you see.



NEW! 2021 CoC Funding by PIT count report



Faith-Based Report:  Baylor Institute

* * * * * 

LA COVID stats at homeless shelters

(as of Aug 8, 2020)

* * * *

NEW! PIT 2020 Reporting Template for Comprehensive Homeless Housing Data


* * * *

2020 HUD Data Standards Cheat Sheet


* * * * * 

White House Homeless Report (Sept 2019);  NAEH Response


* * * * * 

HUD 2019 NOFA Homeless Assistance


* * * * *

HUD for Urban Initiatives report on California 2019 Homeless Count Results by Continuum


* * * * *

HUD 2017 CoC Awards - Tier 1 and 2 - Complete List


* * * * *

NHIP Editor presentation "10 Things You May NOT Know about Homelessness" to Los Gatos Lion's Club (January 30, 2019)


* * * * *

Chapin Study: 4 million homeless youth?   More details in Journal of Adolescent Youth


* * * * *


Sonoma County 2018 PIT Report


* * * * *

CA Appropriates $500 million to combat homelessness

The California Legislature has allocated $500 million in Emergency AId that will  be apportioned to Continuums and Communities based on their level of homelessness.  CLICK HERE for a detail of the allocation by city/community for the first $349 million.  Funds are a one-time emergency appropriation and must be spent down by June 30, 2021.

* * * * * 

LA reports decline


Results from the 2018 LA Point-In-Time Homeless Count showed a 3% decrease from 55,038 to 53,195.  A large decrease in homelessness among African-Americans accounted for much of the decline. Read the Data Summaries on the LAHSA website for more details.

* * * * * 

Houston shows increases in total and Veterans homelessness persons

The 2018 PIT report released by Houston/Harris County showed significant increases in total homelessness, unsheltered homelessness and veterans homelessness.  CLICK HERE to read the full report.

* * * * *

2017 HUD Homeless Funding per Capita by CoC 


* * * * *


as of Dec 31, 2017

83% Utilization Rate

73,069 Vouchers in use
14,628 Vouchers NOT in use

* * * * * * * *





Hear songs from the amazing musical about homelessness!

"Shelter Plus Care"

"This Park is No Place"

"Axis I"

"Just Like Me"

or  20 minute filmes highlights       on YOUTUBE !

To learn more about Truly Dually, contact Michael Ullman at


Check Out the San Diego HMIS Dashboard



Easy Steps to

Email Congress!


One-step connection to all your senators and representatives.  Just enter your ZIP CODE and your Senate and House representatives will come up.  Type your message and one-click will send to all !!

* * * * * * * * * * * * * * * * * * *

Guide to Advocacy

"Indivisible" brief guide to effective advocacy written by ex-staffers. Highly Recommended !! The Guide is focused on Fighting Trump - but it is a good general guide to advocacy.


* * * * * * * *


Read Comments submitted on HUD formula

A total of 143 comments (108 individuals documents posted) were submitted to HUD concerning the homeless funding formula. Advocates from San Diego (18)) and Chicago (32) submitted about a large portion of all comments.  Comment were also submitted by other cities/states including Portland, Maine, Los Angeles, New Orlands, Sonoma, Santa Barbara, Montana, Austin, and Florida. Commenters from Chicago were particularly concerned that proposed formulas could reduce CoC funding by $22 million, or about 50% of the current level.  San Diego commenters proposed formula that could increase their funding which has shown to be very low compared to their homeless prevalence. Only 18 states were represented in the comments.



Congress fails to approve chronic homeless funding

At the Florida Homeless Conference, USICH Director Matt Doherty and SNAPS Director Norm Suchar reported that Congress did not approve the additional amount requested by HUD for expansion of vouchers for chronically homeless persons. The NHIP estimates a need for 40,000 vouchers at an annual cost of $500 million.

The annual costs of the VASH program is estimated to be over $1 billion providing 84,000 vouchers to homeless veterans, a portion of which have been chronically homeless.


CA-504 Sonoma HUD Comment Response

Sonoma County receives a very low amount of CoC funding  ($606 per homeless) compared to its homeless prevalence.  For more information about Sonoma County's initiatives, contact Continuum  Director Michael Gause at

Final San Diego RCCC Funding Open Comment Letter

San Diego RCCC drafts Funding Formula response 

The San Diego Regional Continuum of Care has drafted a response to  the HUD Open comment period on the modifying the Homeless Federal Funding period.  

The Board Action item recommends advocating for a formula using three factors with the following weights: 1) Renter Occupanied Units  (65%), 2) Affordability Gap (30%), and 3)  Rent-burdened Eli Housingholds (5%).  

NHIP encourages CoCs to use the San Diego formula to see its impact on your CoC funding.  The deadline for submission of comments to HUD ends September 23, 2016.

HUD rejects CoC FOIA for scoring detail

A recent Freedom of Information Request (FOIA) submitted on June 10, 2016 by the Lousiville Coalition for the Homeless requesting release of the full scoring of their 2015 application for homeless assistance was denied by HUD in a letter signed by Helen Goff Foster, HUD FOIA Office. 

In her letter, CoC Director Natalie Harris detailed that HUD provided explanation for only 59 of the 203 points available.  In order to improve the Louisville application, the CoC wanted more information on areas they needed to improve upon for this year.  The Louisville/Jefferson application receive a score of 157 out of 203.

The Louisville CoC also alerted Kentucky Senator Mitch McConnell and Rep. John Yarmuth about the denial and requested assistance in pursuing the matter further.


CoC files FOIA for 2015 HUD scoring

The Charlotte/Punta Gorda (FL-602) Continuum has filed a Freedom of Information Act request with HUD to provide a full list of all 2015 Homeless Assistance application scoring. The FL-602 CoC also desires that greater information be produced by HUD to illuminate scoring criteria. 

In an interview with the CoC  Head Angela Hogan, the CoC said that it is concerned about the accuracy of the scoring given discrepancies that FL-602 has identified with their application scoring. 

ARD 2015/2016 comparison

An NHIP Brief Analysis of the 2016 ARD shows that 23 Continuums received increased of more than $1 million, while 11 CoCs saw decreases of $1 million or greater.

CoCs with large increases included Los Angeles, Houston, Portland, Denver, Connecticut BOS, and San Francisco.  Large decreases hit Miami, Baltimore, Boston, New York City, and Washington DC.  The 2016 ARD  totalled $1.88 billion, up $37 million from 2015.


TIER 2 Awards

FL, MA, IN, KY, HI, PR suffer big funding losses

An NHIP ANALYSIS of Tier 2 awards reveal the winners and losses in the HUD Homeless Assistance competition.  A total of $354 million funded both project renewals and new permanent housing programs.

Among states receiving increased funding, California and New York led the way receiving $104 million in renewal and bonus projects. Both states received more in Tier 2 awards than their ARD 15% reallocation amount.  Vermont received funding equal to 324% of the 15% ARD, the highest of any state or territory.

A total of 38 States and Territories received funding in excess of their 15% ARD amount, while 16 states were awarded less than the 15% ARD.  Florida, Indiana and Massachusetts each lost over $2 million in funding.  Indian, Hawaii, and North Dakota lost more than 50% of their ARD reallocation amount.  South Dakota did not receive any Tier 2 funding.

An analysis of FLORIDA Continuum awards show that the $3.2 million decrease, the largest in the nation, was led by a $4 million loss in funding by Miami/Dade (FL-600). Fourteen (14) of the 27 Florida Continuums received fewer Tier 2 funds compared to their 15% ARD reallocation amounts. Pasco County did not receive any Tier 2 awards.

The State of Florida received funding for 11 new projects totalling $2.8 million.  Sarasota and St. Petersburg and Sarasota earned  three (3) new project awards, respectively.  Orlando was awarded the single biggest new project in the State, a $1.08 million Rapid Rehousing award.

Florida is already a state that receives below average Homeless Per Capita  funding.  The 2016 funding losses will move Florida further down the ranks compared to other states.

The NHIP does not have access to individual Continuum project priority lists.  Assumptions were made in the analysis based on the current public information available.



Continuums interested in emailing the Senate Banking, Finance and Urban Affairs committee concerning H.R. 3700, please visit the Committee website.

Key amendments have been successfully added to H.R. 3700 that call for revising the funding formula for the HUD Continuum of Care and HOPWA (AIDS/HIV) programs.

Key to this effort has been the work of Rep. Scott Peters (D-CA-52) who has reached out to HUD Secretary Castro urging him to update HUD's funding formula that allocates homeless funding to 400+ Continuums throughout the country.  Peters serves the San Diego area which receives low per capita allocation ($1,711) compared to the national average ($2,822). Equalized funding based on homeless prevalence would bring San Diego an additional $9+ million annually. 

H.R. 3700 Housing Opportunity through Modernization Act was passed by the House on February  2, 2016 and now sits in the Senate Banking, Housing and Urban Affairs committee.  Rep. Peters' office expects the comment period for the HUD homeless funding formula to be open in the next few months after passage by the Senate.

Key amendments in H.R. 3700

Sponsor: Rep. Scott Peters (D-CA-52)

"Directs the Secretary of Housing and Urban Development to reopen the period for public comment for the "Homeless Emergency Assistance and Rapid Transition to Housing: Continuum of Care Program" to allow stakeholders the opportunity to provide input on how HUD's resources can be most equitably used to end homelessness in our country."

H.R. 3700
Sponsor: Rep. David Price (D-NC-4) 

"To update and modernize HUD's funding formula for the Housing Opportunities for Persons With AIDS (HOPWA) program so that funding is distributed to jurisdictions based on living cases of HIV/AIDS."

H.R. 3700
Sponsor: Rep. Keith Ellison (D-MN-5)

"Amendment sought to provide affirmative permission for housing providers who administer U.S. Department of Housing and Urban Development funds to report on-time rental payment data for their tenants to credit reporting agencies without requiring and managing individual written consent agreements; and to direct HUD to retain tenant privacy so the furnished information would not specifically note that tenants receive HUD assistance."


New Chronic Homeless Reports

See our Homeless Data page for Chronic Homeless Estimates by Continuum 2013 to 2015. Data available in PDF and .XLS for sorting.

* * * * *

Coming this June!

Homeless U

One of the first initiatives from NHIP will be HOMELESS U - an effort to build knowledge about homeless services among all homeless services staff — from CEOs to Case Managers to Data Specialists to Resident Assistants.

Understanding homelessness requires a great amount of knowledge in a specialized field generally not covered in any formal education.  HOMELESS U will help to fill this specialized training gap.

Homeless 101 –            

Sample Test Question

Q: Which of the following persons(s) are considered "Homeless" according to the Federal Definition?

  1. A person sleeping in a car at a rest stop.
  2. A family living in a camper for six months out of the year
  3. A person residing in jail
  4. A family living in a transitional apartment
  5. None of the above

The HOMELESS U initiative will disseminate curriculum and test instruments that can be incorporated into your employee training. More knowledgeable workers means better workers!