Media Resource: Trends in Homelessness


Fact Sheets | November 19, 2015

Files: Media Resource: Trends in Homelessness (PDF | 125 KB | 2 pages)

Is homelessness in the U.S. increasing or decreasing? Are there more homeless children today than this day last year; this year than three years ago? There are two primary data sources on U.S. homelessness: the U.S. Department of Housing and Urban Development (HUD) and the U.S. Department of Education (Education). The measures implemented by these agencies do not enumerate the same populations, include the same living situations, or cover the same time periods.

Populations Enumerated

HUD releases two annual reports on homelessness: the Annual Homeless Assessment Reports (AHAR) to Congress – Volumes 1 and 2. Volume 1 of the AHAR presents point-in-time counts, which detail the numbers of sheltered and unsheltered people on a night.1 Volume 2 of the AHAR presents an estimate of the number of people who enter homeless shelters or transitional housing throughout the year. The National Center for Homeless Education reports publically on the data collected by Education. This report includes number of students who are enrolled in local and state school districts who are identified by personnel as living unsheltered, in shelters, in hotels or motels, or doubled up with family and friends over the course of the school year. See Table 1.

Table 1. Measures of Official Reports and Data Sources of Federal Agencies on Homelessness


Despite the variation­­­s in populations enumerated, data sources and methodology, these reports can show fairly consistent trends if the data is dissected to examine distinct populations:

  • The number of people sleeping on the streets is declining. This is good: people are presumably safer indoors than when exposed to the elements outdoors. See Figure 1.
  • The number of people sleeping in emergency shelters and transitional housing is increasing slightly in some measures and decreasing slightly in others. Presumably, people living in shelters otherwise would be living on the streets which may mean communities are doing a better job of getting people off the street and under a roof. See Figure 2.
  • The number of people living doubled up, considered homeless by some agencies but not by others, has grown substantially over the last several years. In some measures, the doubled population has started to decrease with overall improvements in the economy, but in others, that trend has not developed as yet. These are a group of people who are housed, but not living independently in their own homes. This not a positive development and is a symptom of the affordable housing crisis in this country, as is homelessness. See Figure 3.
Figure 1. Unsheltered Homelessness Figure 2. Sheltered Homelessness

Figure 3. Doubled Up Households6


All three reports are limited by the scope of what they are designed to measure and the capacity a community has to execute the enumeration. Point-in-time counts (HUD AHAR Volume 1) are limited to those people who are able to be identified as homeless on a given night. Annual sheltered counts (HUD AHAR Volume 2) are calculated using a sample of communities from across the country and may be limited by the number of programs that participate in a community’s HMIS. Education data only captures enrolled students who are identified by school personnel. Many school districts do not have dedicated staff to identify students so it is likely there are students who go unidentified.

All of these reports may also count people or students multiple times. Low-income and homeless populations are highly mobile and may move from state to state, city to city, or school district to school district. As the HUD AHAR Volume 2 and the Education data are collected over longer periods of time, they are more susceptible to this type of duplication. Point-in-time counts (HUD AHAR Volume 1) are the measure least likely to contain duplications, as communities often execute a one-night count and, nationally, counts are required to take place the last week of January, minimizing the likelihood of a person being counted in multiple communities. 

1 A one-night, unduplicated count of people experiencing homelessness in a Continuum of Care (CoC). HUD requires that CoCs count their sheltered homeless population every year and their unsheltered homeless population every other year, on odd numbered calendar years.

People counted as living unsheltered are those living on the street and in places not meant for human habitations such as cars and abandoned buildings.

People counted as living in sheltered locations are those staying in emergency shelters, transitional housing, and in hotels and motels not paid for by themselves.

People counted as living in motels are those living in hotels and motels paid for on their own.

People counted as living in doubled up situations are those that are not lease holders or home owners living with family and friends or other non-relatives.

National Alliance to End Homelessness’ The State of Homelessness in America is an annual report that examines national and state trends in homelessness. The Alliance reports on the point-in-time counts covered in HUD’s AHAR Volume 1, using U.S. Census Bureau data to calculate rates of homelessness. This report also uses U.S. Census Bureau data to determine the number of poor people living in doubled up situations. This report considers people living doubled up to be at-risk of homelessness, along with those with severe housing cost burden, those living in poverty, and those who are unemployed.