Worst Case Housing Needs 2007 Kö¡“Kö× Kö× BOOKMOBI ª > > *> :> J> Z> j> z> Š> š> ª> º> Ê> Ú> ê> ú> > > *> :> J> Z> j> z> Š> š> ª> º> Ê> Ú> ê> ú> > > *> :> J> Z> j> z> Š> š> ª> º> Ê> Ú> ê> ú> > > *> :> J> Z> j> z> Š> š> ª> º> Ê> Ú> ê> ÛO = MOBI ýé@–zÖ
U.S. Department of Housing and Urban Development | Office of Policy Development and Research
Visit PD&R’s website www.huduser.org to find this report and others sponsored by HUD’s Office of Policy Development and Research (PD&R). Other services of HUD USER, PD&R’s research information service, include listservs, special interest reports, bimonthly publications (best practices, significant studies from other sources), access to public use databases, and a hotline (800-245-2691) for help accessing the information you need.
A REPORT TO CONGRESS Prepared for U.S. Department of Housing and Urban Development Washington, D.C. Prepared by David L. Hardiman Carolyn Lynch Marge Martin Barry L. Steffen David A. Vandenbroucke Yung Gann David Yao May 2010
iii Foreword I am pleased to transmit to the U.S. Congress this report from the Department of Housing and Urban Development, “Worst Case Housing Needs 2007.” The report, the twelfth in the longstanding series on worst case housing needs, provides national data and information on critical problems facing low-income American renting families. The report draws on data from the American Housing Survey (AHS), which is funded by HUD and conducted by the Census Bureau. The AHS has been conducted every 2 years since 1973 and is a key source of national data on housing markets, conditions, and dynamics. It is important to note that the report is based on data from 2007, before the full repercussions of the U.S. mortgage market crisis were felt across the broader economy. This year’s report shows that the number of worst case needs households remained effectively flat, with levels still 18 percent higher than in 2000. Worst case needs remains an equal opportunity problem, with significant incidence across races, family types, geographic regions, and boundaries of cities, suburbs and rural areas. In addition, the report highlights the critical role that unit availability, and not just market supply alone, plays in determining burdens. For the very lowest income groups, there is an insufficient and shrinking supply of affordable rental housing. Moreover, even though there are sufficient affordable units on a national basis for very low-income renters, many of the lower rent units are occupied by renters with higher incomes. This leaves many of the very low-income renters unable to find affordable housing and forced to take on greater burdens in higher rent units. Further, the report studies the urban geography of the incidence of worst case needs among families. We discovered two distinct patterns. In cities, most worst case needs families lived in higher poverty neighborhoods. By contrast, most worst case needs families living in suburbs and non-metro areas were located in low poverty neighborhoods. Finally, the report demonstrates that HUD programs providing rental assistance and spurring the production of affordable housing play a critical role in creating stable housing for low-income renters. However, the evidence is clear: the scope of the problem demands that we do more to ensure that all Americans are suitably housed. Raphael W. Bostic Assistant Secretary for Policy Development and Research Worst Case Housing Needs
Table of Contents Foreword Executive Summary..................................................................................................vii Findings................................................................................................................vii Chapter 1. Introduction and Key Findings ..............................................................1 Chapter 2 – Demographics of Worst Cas e Needs ................................................2 Chapter 3 – Availability of Affordable Housing Stock..................................................5 Chapter 4 – Worst Case Needs and Neighborhoods.............................................6 Chapter 2. Demographics of Worst Case Needs.......................................................7 Worst Case Needs in 2007......................................................................................7 Types of Worst Case Housing Problems ..............................................................7 Worst Case Needs by Income Groups ..................................................................10 Worst Case Needs by Race/Ethnicity Group and Across Central Cities, Suburbs, and Non-Metro Areas ........................................................................12 Worst Case Needs by Family Types .....................................................................13 Geography of Worst Case Needs..........................................................................16 Transportation and Worst Case Needs..................................................................18 Working Households With Worst Case Needs .....................................................19 Housing Assistance Mitigates the Effect of Very Low Income...........................21 Summary ..............................................................................................................22 Chapter 3. Availability of Affordable Housing Stock ............................................23 Why Housing Supply Is an Issue..........................................................................23 Measures of Affordability and Availability ..........................................................24 Summary ..............................................................................................................38 Chapter 4. Worst Case Needs and Neighborhoods ................................................40 Poverty and Neighborhood Distress.....................................................................40 Worst Case Needs Households and Neighborhood Poverty Rates.......................41 Summary ..............................................................................................................46 Worst Case Housing Needs v
Chapter 5. Background and Methodology .............................................................47 Worst Case Needs by Definition Are Below the “Very Low-Income” Threshold .......49 References..................................................................................................................52 Appendix A. Data on Housing Problems and Supply of Affordable Housing.....54 List of Tables ........................................................................................................54 Appendix B. The American Housing Survey: Terms and Data Sources.............82 American Housing Survey....................................................................................82 Exclusions From the AHS Data............................................................................84 Household and Family Types ...............................................................................84 Housing Problems ................................................................................................85 Income Measurement ...........................................................................................87 Income Categories................................................................................................88 Housing Assistance Status....................................................................................90 Location................................................................................................................90 Mismatch of Supply and Demand for Affordable Rental Housing .....................90 Race and Ethnicity. ...............................................................................................92 Appendix C. Changes to the AHS Survey Questions, 2005 to 2007, Regarding Receipt of Housing Assistance............................................................93 Appendix D. Additional Examination of the 2005-to-2007 Trend in Worst Case Needs...................................................................................................97 National Trend in Poverty and Income Decline in the Number of Extremely Low-Income Renter Households, .................................................................97 2005 to 2007......................................................................................................99 Additional Examination of the 2005-to-2007 Changes......................................100 Worst Case Housing Needs vi
Executive Summary The report is the twelfth in the series of worst case housing needs reports that have been issued by the Department of Housing and Urban Development (HUD) since 1991. Worst case housing needs (WCN) are experienced by unassisted very low-income renters who either (1) pay more than one-half of their monthly income for rent; or (2) live in severely inadequate conditions, or both. HUD defines “very low-income” as below 50 percent of the local area median income (AMI) and “extremely low-income” as below 30 percent of AMI. Findings Demographics of Worst Case Needs • The number of households with worst case housing needs in 2007 was 5.91 million households, comprising 12.97 million individuals. This number was a slight (and statistically insignificant) decrease of 1.5 percent from the 5.99 million worst case needs in 2005. • The small decrease in worst case needs from 2005 to 2007 occurred following a large and statistically significant increase of 19.6 percent during the 2001- to-2005 period. The number of households with worst case needs increased from 5.01 million in 2001, to 5.18 million in 2003, to 5.99 million in 2005 and then dropped slightly to 5.91 million in 2007. • The primary cause of worst case needs is severe rent burden. Of the 5.91 million households with worst case housing needs, 5.48 million had severe rent burden as their sole problem, 190,000 households lived in severely inadequate housing, and 240,000 households had both problems. • Income Levels. In general, worst case needs are more prevalent among extremely low-income families. A majority of WCN households (4.33 million) had extremely low incomes, and a smaller but substantial portion (1.58 mil- lion) were in the 30 to 50 percent of area median income group. • Family Types. Worst case needs are found across different types of families. Among the 5.91 million WCN households were: 2.19 million families with children; 1.21 million elderly households; 602,000 non-elderly disabled house- holds; and 1.91 million “other” households. • Disabilities. In addition to the 602,000 non-elderly disabled households, there were an additional 404,000 families with children that also had an adult with disabilities present – bringing the combined total of these two types of house- holds with disabilities and worst case housing needs to more than 1 million. Disabled households were found to have the highest likelihood of having WCN among the four main family types (families with children, elderly, disabled and “other” households). Among unassisted very low-income households with Worst Case Housing Needs vii
disabilities, the likelihood of having worst case needs was 65.6 percent. This exceeds the second highest rate of elderly households (58 percent) and is far above the rates among families with children (47.8 percent) and “other” households (47.7 percent). • Race and Ethnicity. Worst case needs were found across racial and ethnic groups. There were 2.92 million no n-Hispanic white households, 1.35 million non-Hispanic black households, and 1.23 million Hispanic households with worst case housing needs. • Geography. Worst case needs were found across national regions and across central cities, suburbs and non-metropolitan areas. In the West, 40.0 percent of very low-income renters had worst case needs in 2007, while the incidence in the South was 37.4 percent and 36.2 percent in the Northeast. The Midwest had the lowest incidence, yet in this region too, more than one-third of very low-income renters, 34.3 percent, had worst case needs. In central cities, 37.7 percent of very low-income renters, or 2.80 million households, had worst case housing needs. Very low-income renters were most likely to have worst case housing needs in the suburbs (39.2 percent) although overall fewer worst case needs households (2.05 million) lived in these areas. In non-metropolitan areas, 32.1 percent of very low-income renters, or 1.05 million households had worst case needs. • Working Families. During 2007, almost one-half (46 percent) of households with children that experienced worst case needs were working full-time (with earned incomes at least at the minimum wage for 40 hours per week). Housing Supply • Although there were a sufficient number of affordable units for very low-income renter families, there were an insufficient number of these units available for these families. Units are counted as available if they were either occupied by very low-income renters or vacant and available for rent (the remainder being occupied by higher income families). There were an insufficient number of both affordable and available rental units for extremely low-income renters. Neighborhoods • Relative to all renters, families with worst case housing needs were only some- what more likely to live in high poverty neighborhoods. • More than 40 percent of worst case needs renters in cities lived in higher poverty neighborhoods (indicating both a lack of housing options in cities; and a lack of affordable housing even in high poverty areas). • Most worst case needs renters in suburbs and non-metro areas lived in low poverty neighborhoods (reflecting both the overall proportion that such neigh- borhoods comprise in these areas and the lack of affordable housing options throughout these areas). Worst Case Housing Needs viii
Chapter 1. Introduction and Key Findings This introduction provides a brief background on worst case housing needs and a sum- mary road map of the Report. It also includes a list of 24 key findings from the main body of the Report. What are worst case housing needs? Households have worst case housing needs if they: are renters with very low-incomes; do not currently have housing assistance; and either have a severe rent burden and/ or live in severely inadequate housing. Severe rent burden means a family is paying more than one-half their income for rent. 1 Severely inadequate housing includes a variety of serious physical problems related to heating, plumbing, electric, or maintenance. Homeless persons are not included in estimates of worst case needs in this and earlier reports because the American Housing Survey counts only persons living in housing units. 2 Worst Case Needs Households With Worst Case Housing Needs: Are very low-income renters; Do not have housing assistance; And either: (1) Have a severe rent burden (pay more than one-half their income for rent); and/or (2) Live in severely inad- equate conditions Homeowners are also not included in worst case needs, although they may have very low-incomes and have high housing costs consuming more than one-half of their incomes. Two common terms used throughout this report refer to the income levels of renters: “very low-income” and “extremely low-income.” “Very low-income” households are below 50 perc ent of area median income; and “extremely low-income” are those with incomes below 30 percent of area median income. Department of Housing and Urban Development (HUD) programs use “area median incomes” calculated on the basis of local family incomes, with adjustments for household size. However, as a general rule of thumb, the nationwide median income 1 Rent-to-income ratios are calculated on the basis of gross rent, which is the sum of rent plus tenant-paid utilities. 2 HUD is assessing the potential of supplementing future estimates of worst case needs with esti- mates of people who experience homelessness from local Homeless Management Information Sys- tems (HMIS). The fourth Annual Homeless Assessment Report, transmitted to Congress in July 2009, reported that during the 12-month period of October 1, 2007, to September 30, 2008, about 1.6 million people used an emergency shelter or transitional housing, including 1,092,600 individuals (68 percent) and 516,700 people in families (32 percent). Worst Case Housing Needs 1
(for HUD programs) in 2007 was $59,000, with the “very low-income” level being at $29,500 per year, and the “extremely low-income” level at $17,700 per year. All of these income levels are for a family of four. 3 Families with fewer than four people or who are living in areas with lower family incomes can have far smaller incomes than these national figures of 30 and 50 percent of median income. • Chapter 2 focuses on the demographics and geography of worst case needs households. • Chapter 3 focuses on housing supply issues. • Chapter 4 explores poverty levels of the neighborhoods where worst case needs households live. Chapter 5 and various appendices include more detailed discussion of methodology, terms, sources and related topics. Chapter 2 – Demographics of Worst Case Needs Chapter 2 examines various demographic characteristics of households with worst case housing needs, including income, race and ethnicity, family types, geography and other factors. The chapter also includes a short examination of transportation costs, by looking at commuting patterns, of worst case needs households, included in this report for the first time. The chapter then takes a closer look at working families, including families with children, that still have worst case needs and concludes with a short examination of the effect of housing assistance to reduce the level of worst case needs. Unmet needs for decent, affordable rental housing Finding 1: In 2007, There Were 5.91 Million Households (Comprising 12.97 Million People) With Worst Case Housing Needs. Finding 2: Severe Rent Burden Remains the Key Housing Issue. Of the 5.91 million households with worst case needs, 5.48 million had severe rent burden as their sole priority problem. 190,000 households lived in severely inadequate housing, and 240,000 households experienced both of these problems. Finding 3: No Statistically Significant Change From 2005 to 2007. The 5.91 million households reporting worst case problems in 2007 represent a slight and statistically insignificant decrease of 1.5 percent from the 5.99 million worst case needs in 2005. 3 These income levels are based on “national median” income as calculated for HUD programs by inflating Census 2000 and ACS 2005 survey data to mid-FY 2007 to distinguish this median from the Census published 2007 median family income. Chapter 5 includes a chart with examples of specific area median income limits in selected major cities Worst Case Housing Needs 2
Finding 4: Longer Term Trend Consistent With Increase in Worst Case Needs From 2001 to 2005. The relative stability during the 2005-to-2007 period followed a large and statistically significant increase of 19.5 percent during the 2001-to-2005 period. The number of households with worst case needs increased from 5.01 million in 2001, to 5.18 mi llion in 2003, to 5.99 million in 2005 and dropped slightly to 5.91 million in 2007. Worst case needs by income group Finding 5: Most Worst Case Needs Households Were Extremely Low-Income. In 2007, of the 5.91 million households with worst case needs, a majority (4.33 mil- lion) fell into the extremely low income group, and a smaller but substantial portion (1.58 million) were in the 30 to 50 percent of area median income group. Finding 6: Extremely Low-Income Renters Had a Higher Likelihood of Worst Case Needs. Among all extremely low-income renter households (those with less than 30 percent of area median income), 46.8 percent had worst case housing needs. Among renters in the 30 to 50 percent of area median income range, 23.6 percent had worst case housing needs. Race and ethnicity Finding 7: Worst Case Needs for Housing Cut Across Lines of Race and Ethni- city. In 2007, there were 2.92 million non-Hispanic white households, 1.35 million non-Hispanic black households, and 1.23 million Hispanic households who had worst case needs. The incidence of worst case needs among non-Hispanic whites who are very low-income renters was 39.0 percent in 2007, compared with 33.3 percent for non-Hispanic black households, and 37.4 percent for Hispanic households. Family types Finding 8: Over One-Third of Worst Case Needs Households Were Families With Children. In 2007, 2.19 million households (34.6 percent) of the 6.33 million very low-income renter households with one or more child had worst case housing needs. These families with children represented a large share, 37.0 percent, of house- holds with worst case needs. Finding 9: About One in Five Worst Case Needs Households Were Seniors. In 2007, 1.21 million households (34.6 percent) of the 3.49 million very low-income renter households with either a head or spouse at least 62 years of age had worst case housing needs. These 1.21 million households made up 20.5 percent of house- holds with worst case housing needs. Finding 10: There Were More Than 1 Million Non-Elderly Worst Case Needs Households in Which at Least One Adult Had a Disability. In 2007, 602,000 dis- abled households had worst case housing needs (35.3 percent of all such households). A household is counted in the disabled category if it is non-elderly, without children and received some form of government assistance indicating a payment for disability. An additional 404,000 families with children had an adult with disabilities present. Together, these two households types therefore comprised more than 1 million worst Worst Case Housing Needs 3
case housing needs households in which at least one adult had a disability. Disabled households made up 10.2 percent of all worst case needs households. When families with children that also have an adult with disabilities are included, this proportion increases to 17.0 percent. Finding 11: Disabled Households Had the Highest Incidence of Worst Case Housing Needs Among Unassisted Households (of the Four Types of Fami- lies). The likelihood of having worst case housing needs among unassisted very low-income renters is highest among disabled households, with rates higher than those of the other family types (the elderly, families with children and other house- holds). Among unassisted very low-income households with disabilities, the likelihood of having worst case needs is 65.6 percent. This exceeds the second highest rate of elderly households (58.0 percent) and is far above the rates among households with children (47.8 percent) and “other” households (47.7 percent). Geography of worst case needs Finding 12: All Regions of the Country Had High Levels of Worst Case Housing Needs. Very low-income renters do not escape high levels of worst case needs in any region of the country—Northeast, Midwest, South, or West. In the West, 40.0 percent of very low-income renters had worst case needs in 2007, while the incidence in the South was 37.4 percent and 36.2 percent in the Northeast. The Midwest had the low- est incidence, yet in this region too, more than one-third of very low-income renters, 34.3 percent, had worst case needs. Finding 13: Substantial Worst Case Housing Needs Were Found in Central Cities, Suburbs, and Non-Metro Areas. In central cities, 37.7 percent of very low-income renters, or 2.80 million households, had worst case housing needs. In suburban areas, very low-income renters had a higher incidence of worst case hous- ing needs (39.2 percent) although overall fewer worst case needs households (2.05 million) lived in these areas. Non-metropolitan areas had 1.05 million households with worst case needs - with almost one-third (32.1 percent) of all very low-income renters in these areas falling into this situation. The overall distribution of worst case needs households generally reflected the distribution of very low-income renters, of which 46.5 percent were living in central cities, 32.8 percent were in suburbs, and 20.5 percent resided in non-metropolitan areas during 2007. Transportation and worst case needs Finding 14: Many Working Worst Case Needs Families Also Experience Lengthy Commutes to Work. Of the 2.92 million worst case needs households that reported they were working at a fixed location, 485,000 households had commutes of longer than 30 minutes to get to work, and 348,000 households had to commute more than 20 miles to get to work (these two commute types overlap and are not exclusive of each other). Taking into account the overlap between those two groups, there were a total of 583,000 households who had either one or both of these types of lengthy commutes and had worst case housing needs. Worst Case Housing Needs 4
Working families Finding 15: Working Households Had High Rates of Worst Case Needs. During 2007, of 1.56 million renter households with both extremely low incomes and earnings consistent with full-time employment, 860,000 households (55 percent) had worst case housing needs. Among the 3.39 million extremely low-income renter households with earnings as their primary source of income, 2.07 million (61 percent) had worst case needs. Among renters in the 30 to 50 percent of area median income range with earnings as their primary source of income, over one in five (21.6 percent), or 1.07 million households, had worst case housing needs. Finding 16: Worst Case Needs Among Wage Earning Families With Children Were Also High. Almost one-half (46 percent) of households with children that experienced worst case needs were working full-time (with earned income at least at the minimum wage for forty hours per week). Among extremely low-income renters with children, about 1.04 million (34 percent) were working full time in 2007, and 540,000 of these had worst case housing needs. Effect of housing assistance on worst case needs Finding 17: Without Housing Assistance, Lower Income Households Are Far More Likely To Experience Worst Case Needs. Among unassisted renters with extremely low incomes, 73.4 percent had worst case needs during 2007. Incidence of severe problems is far lower, but still substantial at 27.8 percent, among unassisted households with incomes from 30 to 50 percent of area median income. Finding 18: Housing Assistance Mitigates the Overall Number of Worst Case Needs. If current government housing assistance were withdrawn, and those receiv- ing it experienced worst case needs at incidence rates of those currently without such assistance, there would be an additional 2.74 million worst case needs households (2.46 million extremely low-income and 280,000 in the 30 to 50 percent of area median income range). Chapter 3 – Availability of Affordable Housing Stock Chapter 3 analyzes housing supply issues, looking specifically at the number of afford- able and available rental units for different income level groups. Finding 19: There Is an Insufficient Supply of Affordable and Available Rental Housing for the Lowest Income Groups. For every 100 extremely low-income renter households, there were only 76 rental units that would be affordable to them at 30 percent of income for rent, and of these only 44 such units were actually available to them (that is, either occupied by renters in this income group or vacant and avail- able for rent). Finding 20: Shortages of Affordable and Available Rental Units Are Not Confined to Extremely Low-Income Renters. For somewhat higher low-income groups, the af- fordable stock is nominally sufficient to house every household above 45 percent of AMI. However, for rental units that are both affordable and available, the available stock does not match the number of renters until household incomes reach 70 to 75 percent of AMI. Worst Case Housing Needs 5
Finding 21: There Are Insufficient Affordable and Available Rental Units for Renter Households Between 30 and 50 Percent of Area Median Income. For every 100 renter households in the 30 to 50 percent of median income range, there were 112.9 rental units that were affordable to them. However, of these units only 74 rental units were available to this income group (that is either occupied by such households in this income group and not occupied by higher income households, or vacant and available for rent). Chapter 4 – Worst Case Needs and Neighborhoods Chapter 4 is an in-depth analysis of neighborhood locations of worst case needs rent- ers, and also a comparison of neighborhood types (by poverty level) across central city, suburb, and non-metropolitan areas. This analysis was first added in the last Worst Case Needs Report (Affordable Housing Needs 2005) and is expanded in this report. Finding 22: Worst Case Needs Households Were More Likely To Live in Poorer Neighborhoods. Nationwide, 29.8 percent of worst case needs households resided in neighborhoods with poverty rates that are greater than 20 percent. This figure com- pares to 24.3 percent of all renter households and 20 percent of all households that resided in these poorer neighborhoods. Only 34.9 percent of worst-case households lived in low-poverty neighborhoods in contrast to 40.3 percent of all renters and 50.0 percent of all households. Finding 23: Worst Case Needs Households in Central Cities Were Far More Concentrated in Higher Poverty Neighborhoods. In central cities, 43.6 percent of worst-case households lived in neighborhoods with poverty rates of 20 percent or more. Despite living in neighborhoods where housing is likely to be less expensive, a high percentage of very low-income renters in central cities were still unable to locate affordable housing. Finding 24: In Suburban and Non-Metro Areas, Worst Case Needs Households Were Less Likely To Live in High Poverty Neighborhoods, but Could Not Access Affordable Housing. In suburban areas, approximately one-half of both worst-case needs households (52.7 percent) and renters with very low-income (48.9 percent) lived in neighborhoods with the lowest poverty rates. These high proportions reflect the fact that numerous suburban areas have few neighborhoods with high poverty rates. The only suburban residential options that very low-income renters have are likely to leave them without sufficient affordable housing options. Likewise, in non-metro areas, about half of both worst-case need households (49.6 percent) and renters with very low-incomes (50.9 percent) resided in neighborhoods with moderate poverty levels of 10 to 20 percent—reflecting both the prevailing neighbor - hood characteristics and lack of affordable housing in such areas. Worst Case Housing Needs 6
Chapter 2. Demographics of Worst Case Needs Worst Case Needs in 2007 This chapter examines various demographic characteristics of households with worst case housing needs, including income, race and ethnicity, family types (families with children, seniors, and persons with disabilities), geography, and other factors. The chapter als o includes a brief analysis of transportation costs, by looking at commuting patterns of worst case needs households, included in this report for the first time. Finally, the chapter takes a closer look at working families, including families with children, that still have worst case needs and concludes with a short examination of the effect of housing assistance to reduce worst case needs. In 2007, 5.91 million very low-income renter households had worst case housing needs because they were unassisted and they had severe rent burdens exceeding one-half of their income or lived in severely inadequate housing conditions. These 5.91 million households included 12.97 million individual people. During the 2005-to-2007 period, the level of worst case needs remained basically unchanged. The Department of Housing and Urban Development (HUD) reported a total level of 5.99 million worst case needs households in 2005. There was a slight decline of 87,000 households (a 1.5 percent change), from that level in 2007, but this decline is well within the margin of error and thus is not statistically significant. 1 Types of Worst Case Housing Problems Severe rent burdens Severe rent burden constitutes by far the largest portion of worst case needs. Families with severe rent burdens pay more than one-half of their monthly income for rent. Out of 5.91 million households with worst case needs, 5.72 million (96.8 percent) had severe rent burdens. Of these 5.72 million households, 4.25 million were extremely low-income (below 30 percent of area median income) and the remaining 1.47 million households were in the 30 to 50 percent of area median income group. The low incomes of these renter households means that, after paying one-half of their income for rent, very little would be left for necessities such as food, medical care, transportation expenses, education and childcare. 1 The 90 percent confidence interval for the 2007 estimate is 5.675 million to 6.135 million. Worst Case Housing Needs 7
uate Severely inadequate housing conditions Compared with rent burdens, physical housing problems account for a much smaller portion of worst case needs. In 2007, 429,000 households with worst case needs lived in severely inadequate housing. Severely inadequate housing includes a variety of serious physical problems related to heating, plumbing, electric, or maintenance (a full description is provided in Appendix B). More than one-half of these households (56 percent) also had severe rent burdens, and a smaller portion (44 percent) had severely inadequate conditions as their only priority housing problem. Exhibit 2-1. Worst Case Needs in 2007 0–30% of Area Median Income 30–50% of Area Median Income Total (below 50% Area Median Income) Total Unassisted Renter Households 5,893 5,861 11,754 Worst Case Needs Households 4,327 1,578 5,905 Percent of Households with WCN 73.4% 27.8% 50.2% Numbers are in thousands of households. Exhibit 2-2. Worst Case Needs 2001–2007 Both priority problems 4.1% Severely inadequate housing only 3.2% Severe rent burden only 92.7% Both priority probl q .2% Severe rent burden Trend in Worst Case Needs 2001 to 2007 Although the level of worst case needs remained relatively stable between 2005 and 2007, this stability followed a longer period of substantially worsening conditions for very low-income renters. As Exhibit 2-3 illustrates, the small (statistically insignificant) decrease in worst case needs from 2005 to 2007 followed a very large increase in worst case needs during the 2001-to-2005 period. Between 2001 and 2005, worst case needs increased by nearly one-fifth (19.5 percent). Over the longer 2001-to-2007 period, the estimated increase in worst case needs was 17.8 percent. Thus, it is critical to observe trends in worst case needs over a longer term period. Indeed, when the report on the 2003 Worst Case Housing Needs 8
Exhibit 2-3. W orst Case Needs 2001-2007 data was issued, it showed a much smaller increase in housing needs than was found in other private research studies. Seen in this context, the overall level of worst case needs in 2007 is still consistent with a longer term trend. Exhibit 2-3. Worst Case Needs 2001–2007 4.5 5 5.5 6 6.5 Households (millions) 4 2001 2003 2005 2007 # Households (millions) Exhibit 2-4 shows the same trend in worst case needs from 2001 to 2007 in compari- son with the total number of all U.S. households during the same period. Exhibit 2-4. Worst Case Needs 2001–2007 2001 2003 2005 2007 All households (millions) 105.44 105.87 108.90 110.72 Renter households with worst case needs (millions) 5.01 5.18 5.99 5.91 Worst case needs as percent of all households 4.76% 4.89% 5.50% 5.33% Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Looking back a bit further, worst case needs increased by 22.0 percent between 1991 and 2007, a rate of growth exceeding the 13.8 percent increase in very low-income renters and the 18.9 percent increase in total U.S. households during the same period. These changes imply a long-term increase in incidence as well, so that 51.0 percent of unassisted very low-income renters had worst case needs in 2007, the same incidence observed in 1997, but higher than the 48.1 percent incidence in 1991. It should be noted that year-to-year comparisons are not always comparable, due to changes in the American Housing Survey (AHS) in some years and because the pre-2000 AHS figures are weighted using 1990 Census data. Worst Case Housing Needs 9
Exhibit 2-5. Overall Trend in the Number of Very Low-Income Renters and Those with Worst Case Needs, 1991–2007 0 5,000 10,000 15,000 20,000 4,842 14,002 14,738 14,549 14,801 14,803 14,903 15,658 16,072 15,940 5,198 5,203 5,379 4,856 5,014 5,175 5,992 5,905 1991 1993 1995 1997 1999 2001 2003 2005 2007 VLI renters (1990 weights) VLI renters (2000 weights) unassisted VLI renters with priority problems (1990 weights) unassisted VLI renters with priority problems (2000 weights) VLI = very low-income The slight decline in worst case needs from 2005 to 2007 was accompanied by a decline in the total number of extremely low-income renters, from 9.73 million households to 9.24 million households during the same period. This phenomenon is explained in greater detail, along with other possible causes of the small decline in worst case needs, in Appendix D, “Additional Examination of the 2005-to-2007 Trend in Worst Case Needs.” Worst Case Needs by Income Groups By definition, households with worst case needs for affordable housing are below the very low-income threshold (50 percent of median income), but households with extremely low incomes constitute nearly three-fourths of those with worst case needs. In 2007, of the 5.91 million households with worst case needs, 73.3 percent (4.33 million households) fell into the extremely low-income group, and a smaller but substantial portion (26.7 percent, or 1.58 million households) were in the 30 to 50 percent of median income group. Because severe rent burden, rather than inadequate housing, accounts for most worst case needs, income is obviously a critical factor. The predominance of extremely low-income renters among worst case households is explained by two factors. First, 58 percent of very low-income renters are also extremely low-income renters, having incomes below the 30 percent of median in- come threshold. Second, unless they receive housing assistance, the poorer subset of renters is much more likely to experience severe housing problems. Extremely low-income renters were much more likely to have worst case housing needs. As previously shown in Exhibit 2-1, 73.4 percent of unassisted renters with extrem ely low incomes had worst case housing needs in 2007. Incidence of severe problems is far lower, but still substantial at 27.8 percent, for the remaining households making up worst case needs—those with incomes from 30 to 50 percent of median. Worst Case Housing Needs 10
Race and Ethnicity Worst case needs for affordable housing cut across lines of race and ethnicity. Minority status depends on definitions that are fluctuating as the nation’s population becomes more diverse. 2 Analyzing housing needs in terms of these demographics shows both similarities and differences among the three largest groups defined by race and ethnicity. Exhibit 2-6 shows that in 2007, there were 2.92 million non-Hispanic white house- holds with worst case needs, 1.35 million non-Hispanic Black households with worst case needs, and 1.23 million Hispanic households with worst case needs. Together, these three race/ethnicity groups account for 93 percent of worst case needs in 2007. 3 Exhibit 2-6. Race/Ethnicity Groups and Worst Case Needs, 2005–2007 In terms of incidence, non-Hispanic Whites had the highest likelihood of having worst case housing needs, with 39.0 percent of all such very low-income renter households falling into this category. Hispanic households had a 37.4 percent likelihood, and non- Hispanic black households had a 33.3 percent likelihood of having worst case housing needs. Note that these incidence levels, and those in the next three sections (family types, regions and metro types) are calculated among the total number of very low-income renters (including assisted households) in each category unless otherwise noted. For further detail, see Table A-9 in the Appendix. Severely inadequate housing varies somewhat by demographic group. Focusing again on very low-income renters, only 2.6 percent of non-Hispanic whites experienced severe physical problems with their units in 2007, compared with 4.4 percent of both non-Hispanic blacks and Hispanics. 2 Beginning in 2003, the AHS used revised Census Bureau categories of race and ethnicity that are not directly comparable with prior surveys. Survey respondents now are allowed to select more than one racial group, causing small but significant decreases in the size of the single-race categories. 3 The estimated changes in worst case needs are not statistically significant for these groups. The 2005 estimates fall within the 90 percent confidence intervals of the 2007 estimates: ± 257 for non- Hispanic whites, ±174 for non-Hispanic blacks, and ±192 for Hispanics. Worst Case Housing Needs 11
Among very low-income renters with severe problems, 6 percent of white households, 9 percent of black households, and 10 percent of Hispanic households lived in severely inadequate housing. Hispanic immigration is an important factor in the U.S. demographic picture, because 843,000 Hispanic residents were added between 2005 and 2007. 4 Despite hardships that may be associated with immigration, HUD’s examination of immigration for all groups indicates that it does not contribute significantly to worst case needs. In 2007, householders 5 who are naturalized citizens accounted for 6.3 percent of worst case needs, proportional to their 6.0 percent share of unassisted very low-income households. Likewise, non-citizen householders constituted 14.3 percent of worst case needs, slightly less than their 16.5 percent share of unassisted very low-income households. Among unassisted very low-income renters, incidence of worst case needs is 44.2 percent for non-citizen householders, and 53.3 percent for naturalized citizens, averaging 46.6 percent for all immigrant householders. By comparison, 52.3 percent of U.S. native householders have worst case needs. These data show that, on a proportional basis, unassisted very low-income renters who are immigrants are slightly less likely than U.S. native householders to have worst case housing needs. Worst Case Needs by Race/Ethnicity Group and Across Central Cities, Suburbs, and Non-Metro Areas Exhibit 2-7 shows the distribution of worst case needs by the three main race/ ethnicity groups covered and their distribution across central cities, suburbs, and non-metro areas. Non-Hispanic White households with worst case needs lived most often in central cities and suburbs (with roughly the same distribution in each type of area), and somewhat less commonly in non-metro areas. Both non-Hispanic Black and Hispanic households with worst case needs lived most commonly in central cities, somewhat less commonly (though still with a considerable share) in suburban areas and much less commonly in non-metro areas. Exhibit 2-7 also shows that non-Hispanic White households with worst case needs were more numerous than the other two largest race/ethnicity groups in all three metropolitan location types—central cities, suburbs, and non-metro areas. 4 This estimate is based on Census Bureau figures for U.S. resident population in July 2007, relative to age cohorts that existed in July 2005. Births occurring within the U.S. thus are not counted but deaths are netted out. 5 In the American Housing Survey, “householder” refers to the reference person who is responding to the survey, not necessarily to the head of household. Worst Case Housing Needs 12
Exhibit 2-7. Worst Case Needs by Race/Ethnicity and Metro Type, 2007 Metropolitan Location Race Central Cities Percent Suburbs Percent Non- metro Areas Percent Totals Percent Non-Hispanic White (thousands) Households with Worst Case Needs 1,089 37.3% 1,064 36.5% 766 26.2% 2,919 100% Non-Hispanic Black (thousands) Households with Worst Case Needs 789 58.7% 417 31.0% 139 10.3% 1,345 100% Hispanic (thousands) Households with Worst Case Needs 690 55.9% 466 37.8% 78 6.3% 1,234 100% Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Needs by Family Types The composition of different households reflects variations in their stage of life, income, other resources, and housing needs. Exhibit 2-8 shows the differences among four household types and illustrates how the number of very low-income renters of these households relates to worst case needs for each group. Exhibit 2-8. Household Type and Worst Case Needs, 2005–2007 Worst Case Housing Needs 13
Families with Children In 2007, 6.33 million very low-income renter households included one or more children under 18 years of age, and 2.19 million of these households had worst case housing needs. The resulting incidence of 34.6 percent is a non-significant change from the 35.9 percent rate in 2005. Both a lower incidence and a smaller number of very low-income renters contributed to the reduction of 140,000 in worst case needs among families with children. Families with children represented a large share (37.0 percent) of households with worst case needs in 2007. Another substantial fraction of very low-income renters with children (27.7 percent) have rental assistance, and, by definition, these 1.75 million households do not have worst case needs. Elderly Households HUD defines elderly households as those with either a household head or spouse who is at least 62 years of age and includes no children under 18 years of age. During 2007, elderly households constituted 1.21 million, or 20.5 percent, of households with worst case needs. Between 2005 and 2007, the estimated number of elderly households with worst case needs declined by 80,000 and those that reported receiving housing assistance increased by 50,000. Among elderly very low-income renter households, 34.6 percent had worst case needs and 40.3 percent reported housing assistance in 2007. Households Including Persons with Disabilities In 2007, there were 602,000 non-elderly disabled households with wo rst case housing needs. Disabled households thus made up 10.2 percent of all worst case needs households (when families with children that also have an adult with disabili- ties are included, this proportion increases to 17.0 percent). In terms of incidence, 35.3 percent of all non-elderly disabled very low-income renter households had worst case housing needs. This incidence level on its own is roughly comparable to the incidence of worst case needs among families with children (34.6 percent) and elderly households (34.6 percent) and somewhat below that for “other” households (43.2 percent). However, when examining incidence of worst case needs among unassisted very low-income renters, the rate for disabled households is the highest of the four family types considered. Almost two-thirds (65.6 percent) of unassisted very low-income renter households with disabilities have worst case housing needs. This is the highest incidence level, exceeding the second highest rate of elderly households (58 percent), and far above the rates among households with children (47.8 percent) and “other” households (47.7 percent). Between 2005 and 2007, the number of non-elderly disabled households with worst case needs declined from 694,000 to 602,000. However, the apparent decline in worst case needs among these households was not statistically significant and was Worst Case Housing Needs 14
accompanied by two other changes in the data: (1) a decline in the total number of non-elderly disabled households from 1.767 million to 1.707 million; and (2) an increase in the number of such households that reported receiving housing assistance (thus removing them from those who could be considered as worst case needs) from 703,000 to 790,000. It is likely that both of these changes were due to either changes in the AHS Survey between those two years (such as the change in the question on housing assistance) or other factors in the data, rather than overall population changes or an actual increase in the receipt of assistance. In 2008, PD&R issued a separate report on worst case needs among non-elderly disabled households, entitled, Housing Needs of Persons With Disabilities: Supple- mental Findings to the Affordable Housing Needs 2005 Report (February 2008). This report was made possible by improved survey questions in the AHS about the receipt of various government supplemental income programs. The improved AHS Survey allowed the report to include better detail on housing needs among the disabled. 6 The report found that a larger portion of all households with worst case needs were in the non-elderly disabled household category. 7 Based on these improvements, a household is counted as a “disabled household” if it (1) is non-elderly; (2) is without children present; and (3) reports having received assistance from Social Security, Supplemental Security Income (SSI), public welfare, or “worker’s compensation or other disability payment.” A key finding of the supplemental report was that in 2005 there were an additional 365,000 households within the group of households with children that had worst case needs that also had an adult with a disability present. 8 In 2007, this number increased to 404,000 households with children that also included an adult with disabilities. Note that these families are already counted within the families with children category dis- cussed above. Together, these two groups—the non-elderly disabled households and households with children that also had an adult with disabilities present—represent more than 1 million households with worst case needs. This figure is 17.0 percent of all households with worst case housing needs. 6 The improved 2005 estimate of 694,000 was an increase over the estimate of 540,000 households using the old methodology. The new method added a fourth income source to the proxy for disabil- ity, using a new questio n about “worker’s compensation, SSDI [Social Security Disability Income] or other disability payment,” which was added to the AHS beginning in 2005. See a discussion of the disability proxy in Appendix B, and in Housing Needs of Persons With Disabilities: Supplemental Find- ings to the Affordable Housing Needs 2005 Report (HUD, 2008). 7 A review of past Affordable Housing Needs reports identified a mislabeled column heading in sev- eral appendix tables. Specifically, Tables A-6, A-7, and A-8 of Affordable Housing Needs 2005 and previous reports include a heading, “Nonfamily Reporting SSI Income.” This shorthand heading would be more accurately titled, “Nonelderly Disabled Household Without Children,” because it does not reflect that (1) families consisting of related individuals but no children are included, and (2) income types other than SSI were included in the three-source proxy. The mislabeled category contains those households that have nonelderly heads and no children present and reported income from Social Security, SSI, or welfare (variables QSS, QSSI, or QWELF for 2005, and QSS or QWELF for 2003). When no children are present, welfare income is likely to be associated with disability rather than with children, as the Temporary Assistance to Needy Families program notably would be. 8 See Exhibit 4 in Housing Needs of Persons With Disabilities: Supplemental Findings to the Afford- able Housing Needs 2005 Report (HUD, 2008). Worst Case Housing Needs 15
Other Households The final household category shown in Exhibits 2-8 and 2-9 is other households, These are renter households that are not elderly, not disabled, and do not have children in the household. This group may be further broken down into “other families” (related persons without children) and “other non-family households” (single persons, or unrelated persons sharing housing), as shown in Appendix Table A-6a. As shown in Exhibit 2-9, of the 5.91 million households with worst case housing needs, 1.91 million were “other households.” These 1.91 million households included 352,000 “other families” (related persons without children) and 1.56 million “other non-family households” (single persons or unrelated persons sharing housing). Between these two household types, the “other non-family households” had a higher incidence of worst case needs (46.0 percent of the 3.38 million total number of very low-income renter households) than the “other families” households (34.0 percent of the 1.03 million total very low-income renter households). Exhibit 2-9. Incidence of Worst Case Needs Among Unassisted Very Low-Income Renters, By Family Type, 2007 Disabled Households Elderly Households Families with Children “Other Households” Total VLI Renter Households 1,707 3,492 6,329 4,413 Number of Unassisted VLI Renters 917 2,083 4,578 3,997 Number of WCN Households 602 1,209 2,187 1,907 WCN as percent of VLI renters 35.3% 34.6% 34.6% 43.2% WCN as percent of Unassisted VLI Renters 65.6% 58.0% 47.8% 47.7% Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Geography of Worst Case Needs Worst Case Needs by Region High levels of worst case needs are prevalent among very low-income renters in all of the four major regions—Northeast, Midwest, South, and West. Exhibit 2-10 shows that in the West, 40.0 percent of very low-income renters had worst case needs in 2007. Incidence was 36.2 percent in the Northeast and 37.4 percent in the South. The Midwest had the lowest incidence, yet in this region too, more than one-third of very low-income renters, 34.3 percent, had worst case needs. The greater prevalence of worst case needs in the West to some extent reflects higher housing costs in the region. The population of very low-income renters reported by the AHS declined somewhat in much of the country between 2005 and 2007. Only in the Midwest did the number of very low-income renters increase (by 260,000). This addition was balanced by a comparable 250,000 decrease in the number of very low-income renters in the South Worst Case Housing Needs 16
and exceeded by reductions of 50,000 in the Northeast and 90,000 in the West. As the most populous region, the South continues to provide homes for one-third (33 percent) of very low-income renters, while the other three regions each have between 21 and 24 percent of very low-income renters. Exhibit 2-10. Worst Case Needs and Housing Assistance by Region 2003 2005 2007 Northeast Very low-income renters (thousands) 3,444 3,538 3,487 Worst case needs 1,146 1,354 1,263 Percent with worst case needs 33.3% 38.3% 36.2% Midwest Very low-income renters (thousands) 3,327 3,331 3,587 Worst case needs 1,009 1,152 1,230 Percent with worst case needs 30.3% 34.6% 34.3% South Very low-income renters (thousands) 5,294 5,444 5,192 Worst case needs 1,649 1,987 1,942 Percent with worst case needs 31.1% 36.5% 37.4% West Very low-income renters (thousands) 3,592 3,759 3,673 Worst case needs 1,371 1,500 1,470 Percent with worst case needs 38.2% 39.9% 40.0% Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Needs by City, Suburb, and Non-Metro Area Worst case needs are distributed not only across all regions of the country but also throughout central cities, suburbs, and non-metropolitan areas. Central cities had the highest number of worst case needs, with 2.80 million households living there. However, families living in the suburbs had a higher likelihood of experiencing worst case needs—with 39.2 percent of very low-income renter households falling into that category compared to 37.7 percent in central cities. Even in non-metropolitan areas, almost one-third (32.1 percent) of very low-income households had worst case housing needs. Worst case needs households in non-metro areas made up 17.8 percent of all worst case needs. The changes in incidence for the 2005-to-2007 period are statistically insignificant for all three geographies. Worst Case Housing Needs 17
Exhibit 2-11. Worst Case Needs and Housing Assistance in Central Cities, Suburbs, and Non-Metro Areas 2003 2005 2007 Central Cities Very low-income renters (thousands) 7,446 7,505 7,420 Worst case needs renters (thousands) 2,532 2,909 2,800 Percent with worst case needs 33.9% 38.8% 37.7% Suburbs Very low-income renters (thousands) 5,506 5,417 5,239 Worst case needs renters (thousands) 1,987 2,092 2,052 Percent with worst case needs 36.1% 38.6% 39.2% Non-Metropolitan Areas Very low-income renters (thousands) 2,685 3,149 3,281 Worst case needs renters (thousands) 657 991 1,053 Percent with worst case needs 24.5% 31.5% 32.1% Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Transportation and Worst Case Needs The issue of affordable housing is inherently linked to a host of other critical issues, including access to jobs, schools, businesses, public services, and amenities. House- holds may choose to pay higher housing costs for better access to these necessities. Access to jobs in particular is a critical issue for working families because higher transportation costs will further erode already limited budgets of low-income families. The AHS provides some basic data on commuting times and distances of renter families, and these data are summarized here for households with worst case needs and other renters. Exhibit 2-12 provides estimates of renter households having persons working at a fixed location, including the number of these households with lengthy commutes. Those with lengthy commutes either needed mor e than 30 minutes or had to travel more than 20 miles to get to work. A substantial number of worst case needs house- holds that were working had lengthy commutes. The data show that, of the 2.92 million worst case needs households that were working at a fixed location, 583,000 households, or 38 percent, had commutes that either took more than 30 minutes to get to work, or had a distance of more than 20 miles to do so, or both. These long commutes likely add a substantial burden to these households, who already experience high housing costs (and/or severely inadequate conditions). The data show that there is not always a tradeoff between housing and transportation times and costs, but that many families experience both high housing and transportation costs. Worst Case Housing Needs 18
Exhibit 2-12. Commuting Times and Distances of Renter Households Working at a Fixed Location, by Income as Percent of AMI (thousands of households) 0–30% 31–50% 51–80% All Renters All Renters Total 3,296 4,541 5,904 23,212 Commute >30 minutes 500 790 1,046 4,593 Commute distance >20 miles 341 570 906 3,670 Commute >30 minutes and/or >20 miles** 616 926 1,298 5,565 Worst Case Needs/Severe Problems* Total 1,963 958 439* – Commute >30 minutes 286 199 88 – Commute distance >20 miles 217 131 61 – Commute >30 minutes and/or >20 miles 358 225 98 – Assisted Total 773 563 358 – Commute >30 minutes 123 74 52 – Commute distance >20 miles 58 49 40 – Commute >30 minutes and/or >20 miles 146 85 62 – Table includes only renter households having persons who were working at a fixed location. Assisted households are included for comparison purposes. * Over 50% of median income is by definition not WCN – the households in this column are unassisted renters with a priority problem (but are over the income threshold for inclusion in worst case needs). ** This row shows the number of renters with either type of long commute, or both of them (without double counting), and not an additional category of renters. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data The AHS data for 2007 show that about two-thirds of homeowners who were working had shorter commutes (under 30 minutes or under 20 miles to get to work) while one-third had longer commutes (in both categories). About three-fourths of all renters who were working had shorter commutes by time (less than 30 minutes) and about 85 percent had shorter commutes by distance (under 20 miles), with the remainder having longer commutes in both categories. Working Households With Worst Case Needs Among households with working-age, non-disabled family members, both their work participation and severe housing problems are issues of substantial policy interest. Numerous federal and state policies and programs focus on helping families move to- ward long-term self-sufficiency while providing short-term assistance for daily needs. The success of these policies, as well as the substantial impact of housing problems, including severe rent burdens, is a key issue and the analysis of worst case housing needs can shed substantial light on the issue. Worst Case Housing Needs 19
Very Low-Income Renters With Significant Work Participation In 2007, 1.91 million very low-income renter households with worst case needs had earnings equivalent to a full-time job at the minimum wage. 9 By this measure, worst case needs are found among 29 percent of the 6.48 million very low-income renters with significant work participation. These working, worst case needs households included 860,000 with extremely low incomes and 1.05 million with incomes between 30 and 50 percent of area median income. Well over one-half (55.5 percent) of extremely low-income renter households with earnings exceeding a full-time minimum wage job nevertheless had worst case needs. A smaller but still significant portion (21.2 percent) of renters with significant work participation providing incomes of 30 to 50 percent of area median income had worst case housing needs. A more inclusive proxy for estimating workforce participation counts households that had earned income as their primary source of income. 10 Based on this measure, out of 8.35 million very low-income renter households with earnings as their primary source of income, 3.14 million had worst case housing needs, with 2.07 million having extremely low incomes and 1.07 million having incomes of 30 to 50 percent of area median income. The incidence of worst case housing needs among these two groups was 60.9 percent for the extremely low-income group and a smaller but still substan- tial 21.6 percent for the higher income subset. These statistics show that lack of access to affordable housing is a substantial problem for the nation’s very low-income workforce. As discussed above, the overwhelming majority of these worst case needs are caused by severe rent burdens. The impact of severe rent burdens on the limited earnings of these very low-income households obviously is crucial in the lives of these families, accounting for a vast share of income and leaving little in disposable income for essential costs such as medical care, educa- tion, transportation, child care or other necessities. Worst Case Needs Among Families With Children and Earnings A substantial number of working worst case needs households also were families with children. Of the 1.91 million worst case needs households with earnings at least at full-time minimum wage, 1.0 million were families with children, including 540,000 with extremely low incomes and 460,000 with incomes between 30 and 50 percent of area median income. 11 9 The AHS does not directly measure work participation. HUD uses household earned income of $10,300 as a proxy for full-time employment. This level of annual income is equivalent to 40 hours per week for 50 weeks at the national minimum wage of $5.15 per hour. The minimum wage in- creased to $6.55 in 2008. 10 “Primary source of income” means their earnings represent 50 percent or more of household income. 11 Low-income working families with children are eligible to receive a cash benefit through the federal Earned Income Tax Credit. The AHS does not capture this form of income. Berube (2006) reports that the average claimant in 2004 received roughly $1,800, or more than 10 percent of his or her annual income, from the federal credit, and that 19 states plus the District of Columbia offer their own earned income tax credits, matching the federal credit at rates ranging from 5 to 35 percent. Worst Case Housing Needs 20
The number of worst case needs households that both were working and had children is even larger when defining “working” on the basis of earnings as being the primary source of income, with a total of 1.42 million families meeting all of these criteria. These 1.42 million families included 950,000 with extremely low incomes and 470,000 with incomes between 30 and 50 percent of area median income. Similar to the pattern for working households overall, the incidence of worst case needs is higher for those very low-income families with children that have extremely low incomes. The 950,000 extremely low-income families with children having worst case housing needs represent 30.0 percent of the 1.80 million working extremely low-income families with children. For the 30 to 50 percent of area median income group, worst case needs are found in 18.5 percent of the 2.57 million working families with children. Housing Assistance Mitigates the Effect of Very Low Income By definition, households receiving housing assistance do not have worst case needs. Housing assistance from various sources plays a substantial role in reducing worst case housing needs. Among the 5.16 million renter ho useholds that reported receiving housing assistance in 2007, 84.6 percent had incomes below the very low-income threshold. 12 There were 3.35 million assisted households with extremely low incomes, and another 1.02 million assisted households with incomes in the 30 to 50 percent of AMI range. Thus, 65 percent of assisted households had extremely low incomes placing them at a very substantial risk of experiencing worst case housing needs if they did not receive assistance. Applying the incidence estimates of Exhibit 2-1 to assisted renters with similar incomes (as reported in the AHS, see Table A-1) suggests that in the absence of assistance, an additional 2.74 million households would have worst case needs, including 2.46 million extremely low-income renters and 0.28 million renters with incomes of 30 to 50 percent of area median income. This simple estimate does not take account of any secondary effects such as changes in rents that might result from shifts in demand for unsubsidized housing. In addition, without rental assistance many households would become homeless. 12 By comparison, 69.5 percent of 6.48 million households reporting rental assistance in 2005 had very low incomes. The 2007 estimate is closer to the targeting and totals for HUD’s assisted housing programs, a change that reflects improvements in the 2007 AHS that primarily excluded misreported assistance by higher-income households. See Appendix B for more detail on the changes in the 2007 AHS Survey Questionnaire. Worst Case Housing Needs 21
Summary In 2007, a total of 5.91 million very low-income renters had worst case housing needs, representing 5.3 percent of U.S. households and 37.0 percent of all very low- income renter households. The number of worst case needs declined from 5.99 million households in 2005, but this change was statistically insignificant (i.e., within the margin for error). The level of 5.91 million households should also be considered in the context of a large overall increase in worst case housing needs from 5.01 million in 2001—a 17.9 percent increase during this longer period. Severe housing cost burdens, rather than severely inadequate housing, continue to be the predominant cause of worst case needs. Of the 5.91 million households with worst case needs, 97 percent had an excessive rent burden, while 7 percent lived in severely inadequate conditions. Of this latter group of 429,000 households, 242,000 households faced both problems. Worst Case Housing Needs 22
Chapter 3. Availability of Affordable Housing Stock Why Housing Supply Is an Issue Chapter 2 of this report focused mainly on describing who has worst case housing needs. This chapter examines the question of housing supply. Affordability, Availability, and Adequacy This chapter uses three concepts to assess the rental housing stock: affordability, availability, and adequacy. • Affordability measures the extent to which there are enough rental housing units of different costs to provide each household with a unit it can afford (based on the 30 percent of income standard). Affordability is the broadest measure of housing stock sufficiency, addressing whether there would be sufficient housing units if allocated solely on the basis of cost. The affordable stock includes both vacant and occupied units. • Availability measures the extent to which affordable rental housing units are available to households within a particular income range. Some households choose to spend less than 30 percent of their incomes on rent, occupying housing that is affordable to households of lower income. These units are thus not available to the lower income households. 1 A unit is available at a given level of income if it is affordable at that level and either (1) occupied by a household with that income or less or (2) vacant. • Adequacy extends the concept of availability by considering whether sufficient rental units are physically adequate as well as affordable and available. 2 In this chapter, these concepts are examined in terms of data at the national level, and do not take into account important variations in local or metropolitan housing markets, although regional data are examined later in the Chapter, along with other variations such as those across central cities, suburbs and non-metropolitan areas. 1 The availability measure also removes units from consideration if they have artificially low rents because they are occupied as a benefit of employment (for example, by caretakers) or because they are owned by relatives or friends of the occupants. The 2007 AHS data indicate that 2.4 million renter households (6.2 percent) occupied their units while paying no rent. The AHS does not provide estimates of the number of households paying a positive but below-market rent because of employ- ment or other reasons. 2 The AHS rates housing units using a three-level measure: adequate, moderately inadequate, and severely inadequate. For additional detail, see the entry for the variable ZADEQ in the Codebook for the American Housing Survey, Public Use File: 1997 and Later (U.S. Department of Housing and Urban Development 2006). Worst Case Housing Needs 23
Measures of Affordability and Availability Exhibit 3-1 describes the U.S. rental housing stock in 2007 using data from the Ameri- can Housing Survey (AHS). 3 For purposes of this analysis, income, and affordability are divided into intervals representing 5 percent of area median income (AMI). Exhibit 3-1. Three Measures Characterize the Sufficiency of the U.S. Rental Housing Stock, 2007 The point at which the Affordable line crosses 100 represents the income level at which there is an affordable rental unit for every household. This point occurs at 45 percent of AMI, meaning the number of rental housing units is sufficient—with ideal allocation—to provide affordable housing to households with incomes above 45 per- cent of area median income. This is unchanged from 2005. Affordable units peak at an income level of 85 percent of AMI. Beyond this, more households than housing units are being added. The downward slope beyond 85 percent of AMI represents a reduction in housing need, because the households with incomes greater than each successive threshold are more and more likely to spend less than 30 percent of their incomes on housing. The Affordable and Available line shows a different story. Its position below and to the right of the Affordable line indicates that availability is a substantial additional con- straint. For example, the Affordable and Available line indicates that about 76 percent 3 Measures of affordability and availability do not reflect small-scale geographic detail. The results presented in this chapter reflect large-scale measures that compare the entire housing stock with the entire rental population. Although this chapter presents more geographically restricted measures, they are still too large to fairly represent housing demand and supply as owners and renters expe- rience them because these are local phenomena. Thus, these results should be viewed with some caution as national or regional indicators based on underlying local housing markets. More severe shortages or generous surpluses can occur in specific housing markets, despite these national and regional findings. For an overview of issues related to local markets, see Khadduri, Burnett and Rodda (2003). Worst Case Housing Needs 24
of households with incomes below 30 percent of AMI could be housed in affordable units if such units somehow could be perfectly allocated. The 76 percent figure is an increase from 68 percent in 2005 and about the same as the 78 percent in 2003. The Affordable and Available line shows that a considerable portion of the most af- fordable housing stock is occupied by households who could afford to spend more. As a result, many units that are affo rdable to lower income renters are not available to them. The affordable stock is nominally sufficient to house every household above 45 percent of AMI, yet the available stock does not match the number of renters until household incomes reach 70-75 percent of AMI. 4 This 70-percent balance point is about the same as it was in 2005 but is higher than the 65 percent of AMI that suf- ficed in 2003. The Affordable, Available, and Adequate line in Exhibit 3-1 shows that excluding physi- cally inadequate units further reduces the sufficiency of the rental housing stock. Even for renters with low incomes up to 80 percent of median, only 95 adequate units are available per 100 renters. The adequate stock is not fully sufficient for demand until those units affordable only above 115 percent of AMI are included. 5 Rental Stock by Income Fewer affordable units are available to households with the lowest incomes. Exhibit 3-2 illustrates this by presenting the housing stock measures for the standard income groups used in this report. There is a mismatch between the number of extremely low-income renters and the number of affordable units available to them. There are only 76.2 affordable units for every 100 extremely low-income households. The ratio of available units is about three-fifths as great, at 44.2 units per 100 households. If physically adequate units are required, only 37.4 are available per 100 extremely low-income households. 6 4 This statement interprets the horizontal difference between the Affordable and the Affordable/Avail- able line, which can be understood as showing the income levels of families who are “displaced” by higher income households. The preceding example reflects the vertical difference between the lines, which represents the difference between nominal and available supply of affordable units for house- holds of a given income level. 5 The 2007 AHS dropped certain questions related to the state of common hallways, which had previ- ously been included in the overall adequacy measure. Consequently, it is likely that some units are rated as adequate in 2007, which would not have been so rated in 2005. This is particularly true for apartments in larger structures. 6 Research based on the Residential Finance Survey indicates that 12 percent of units with gross rents of $400 or less produced negative net operating income, suggesting they are heading for de- molition or conversion to nonresidential use (Joint Center for Housing Studies 2006, 24). Worst Case Housing Needs 25
Exhibit 3-2. Rental Housing Stock by Income Category, 2007 Housing Units per 100 Households Income Affordable Affordable and Available Affordable, Available, and Adequate Extremely low-income renters (0–30% AMI) 76.2 44.2 37.4 Very low-income renters (0–50% AMI) 112.9 73.9 66.5 Low-income renters (0–80% AMI) 139.8 105.5 95.4 Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data At the very low-income level there are enough units overall to house all renters, but there is a mismatch of available units here as well. There are only 74 available units for every 100 very low-income households, and fewer than 67 that are also physically adequate. At the higher levels of income, the available rental stock is sufficient to house all renters, though a small proportion of units have physical problems. Exhibit 3-3 illustrates that the supply of affordable housing stock rose between 2005 and 2007 relative to extremely low-income households but fell relative to very low-income households. The overall supply of affordable units per 100 extremely low-income renters increased by more than 8 units, from 67.6 to 76.2. For the same population, the deficiency in available units eased by more than 4 units per 100 renters, from 39.9 to 44.2. On the other hand, very low-income renters experienced decreases in the stock u nder both the affordable measure (4 units) and the available measure (almost 3 units). Taking a somewhat longer view, the changes from 2003 are generally in the direction of reduced supply, although the number of affordable and available units for extremely low-income households is basically unchanged. Very low-income households experienced a larger reduction in supply than did extremely low-income households during the 2003-to-2007 period. Exhibit 3-3. Trend in Rental Housing Stock by Income Category, 2003–2007 Housing Units per 100 Households Change 2003 2005 2007 2003–07 2005–07 Extremely low-income renters (0–30% AMI) Affordable 78.2 67.6 76.2 – 2.0 + 8.6 Affordable and available 44.0 39.9 44.2 + 0.2 + 4.3 Very low-income renters (0–50% AMI) Affordable 127.0 117.1 112.9 – 14.1 – 4.2 Affordable and available 81.4 76.7 73.9 – 7.5 – 2.8 Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 26
Rental Stock by Location Deficiencies in the affordable and available stock are less severe in nonmetropolitan areas, as Exhibit 3-4 illustrates. The available stock is larger in nonmetropolitan areas at all levels of income above 15 percent of AMI, reaching the one-unit-per-household ratio around 60-65 percent of AMI, compared with 70-75 percent of AMI for cities and suburbs. The similar profiles of the city and suburb lines in Exhibit 3-4 show that on average, central cities and suburbs have about the same proportion of units available for every 100 renters at all income levels. Exhibit 3-4. Nonmetropolitan Areas Have More Available Rental Units Than Cities and Suburbs Exhibit 3-5 summarizes the sufficiency patterns among cities, suburbs, and nonmet- ropolitan areas. Notably, although cities and suburbs display comparable available- unit-ratios—with slightly more than 40 units for every 100 extremely low-income renters and 68–72 units for every 100 very low-income renters—the underlying supply of affordable units is more constrained in central cities than in suburbs. The difference between the Affordable and the Available estimates implies that in cities, higher income households occupy about 30 units that would be affordable at very low incomes. Comparable displacement figures are 43 units in suburbs and 53 units in non-metropolitan areas. Worst Case Housing Needs 27
Exhibit 3-5. Rental Housing Stock by Income Groups in Central Cities, Suburbs, and Nonmetropolitan Areas, 2007 Housing Units per 100 Households Income Affordable Affordable and Available Affordable, Available, and Adequate Central Cities Extremely low-income renters (0–30% AMI) 60.7 41.4 36.0 Very low-income renters (0–50% AMI) 102.7 72.3 64.1 Low-income renters (0–80% AMI) 133.9 105.5 94.3 Suburbs Extremely low-income renters (0–30% AMI) 79.1 42.7 39.5 Very low-income renters (0–50% AMI) 110.6 68.1 62.8 Low-income renters (0–80% AMI) 146.5 104.3 96.1 Nonmetropolitan Areas Extremely low-income renters (0–30% AMI) 108.4 53.0 48.7 Very low-income renters (0–50% AMI) 139.7 86.7 78.0 Low-income renters (0–80% AMI) 141.5 107.4 96.7 AMI = Area Median Income. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Exhibit 3-6 examines how affordability ratios and availability ratios changed across central cities, suburbs, and non-metropolitan areas during the 2003-to-2007 period. The key finding is that while the sufficiency of affordable and available units improved during the 2005-to-2007 period for every category of urbanization and income level, the opposite is true over the longer 2003-to-2007 period. Nonmetropolitan areas show smaller improvements over during the 2005-to-2007 period and larger declines during the 2003-to-2007 period than do central cities and suburbs. Thus, although nonmetropolitan areas continue to have more favorable supply conditions, their relative advantage has declined. The very low-income category shows the smallest improvements during the 2005-to- 2007 period and the largest declines during the 2003-to-2007 period compared with the other two income categories. Although the extremely low-income category can hardly be said to have a satisfactory supply, its relative dearth has eased somewhat, particularly in the suburbs, where it shows the only improvement in affordable and available stock during the 2003-to-2007 period. Worst Case Housing Needs 28
Exhibit 3-6. Trend in Rental Housing Stock by Income Groups in Central Cities, Suburbs and Nometropolitan Areas, 2003–2007 Affordable Housing Units per 100 Households Affordable and Available Housing Units per 100 Households Difference Difference 2003 2005 2007 2003–07 2005–07 2003 2005 2007 2003–07 2005–07 Central Cities Extremely low- income renters (0–30% AMI) 65.6 52.2 60.7 Very low- income renters –4.9 8.5 43 31.8 41.4 –1.6 9.6 (0–50% AMI) 120.3 106.1 102.7 Low-income renters –17.6 –3.4 81.5 64.1 72.3 –9.2 8.2 (0–80% AMI) 137.4 133.9 133.9 Suburbs Extremely low- income renters –3.5 0.0 107.7 93.6 105.5 –2.2 11.9 (0–30% AMI) 74.5 66.7 79.1 Very low- income renters 4.6 12.4 39.1 32.4 42.7 3.6 10.3 (0–50% AMI) 121.2 113.9 110.6 Low-income renters –10.6 –3.3 73.3 64.6 68.1 –5.2 3.5 (0–80% AMI) 149.6 145.3 146.5 Nonmetro- politan areas Extremely low- income renters –3.1 1.2 106.9 96.7 104.3 –2.6 7.6 (0–30% AMI) 121.4 106.1 108.3 Very low- income renters –13.1 2.2 56.7 49.4 53.0 –3.7 3.6 (0–50% AMI) 160.3 148.9 139.7 Low-income renters –20.6 –9.2 97.6 82.6 86.7 –10.9 4.1 (0–80% AMI) 155.1 140.9 141.5 –13.6 0.6 115.6 98.3 107.4 –8.2 9.1 AMI = Area Median Income. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Exhibit 3-7 details the affordable, available, and physically adequate stock relative to renter populations in the four regions for each of the standard income categories. The West has the greatest mismatch, with considerably fewer units for every 100 house- holds than the other three regions have. Worst Case Housing Needs 29
Exhibit 3-7. Rental Housing Stock by Region and Income Category, 2007 Housing Units per 100 Households Income Affordable Affordable and Available Affordable, Available, and Adequate Northeast Extremely low-income renters (0–30% AMI) 73.1 47.8 42.2 Very low-income renters (0–50% AMI) 103.8 71.4 64.1 Low-income renters (0–80% AMI) 132.7 100.0 89.7 Midwest Extremely low-income renters (0–30% AMI) 75.8 45.3 42.9 Very low-income renters (0–50% AMI) 139.8 92.6 85.9 Low-income renters (0–80% AMI) 146.8 115.4 106.4 South Extremely low-income renters (0–30% AMI) 84.3 46.3 40.5 Very low-income renters (0–50% AMI) 118.1 74.7 65.4 Low-income renters (0–80% AMI) 145.6 108.6 97.6 West Extremely low-income renters (0–30% AMI) 68.2 36.0 32.5 Very low-income renters (0–50% AMI) 88.1 56.9 51.6 Low-income renters (0–80% AMI) 131.1 96.3 86.8 Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Sufficiency Relative to Fair Market Rent HUD establishes, for every housing market, a Fair Market Rent (FMR) that is intended to represent the cost of decent existing housing that is neither new, nor luxury, nor subsidized. 7 The FMR is used in the largest h ousing assistance program, Housing Choice Voucher Program, to determine the maximum level of subsidy for assisted households. It is also used in other contexts as an indicator of reasonable housing costs in a given area. A natural question is whether the stock of housing renting for less than the FMR is adequate to meet the needs of households that can afford to pay no more than the FMR. Exhibit 3-8 illustrates that the rental stock is insufficient using the FMR standard as well. Although enough affordable units exist in each region, the number of available units in each region is sufficient to house only 83–90 percent of the households that can afford rents no higher than the FMR. 7 In general, the FMR is the 40th percentile rent paid by recent movers for standard-quality units within each region. For further information, see http://www.huduser.org/datasets/fmr.html. Worst Case Housing Needs 30
Exhibit 3-8. Rental Stock of Below-FMR Units, 2007 Households (thousands) Housing Units (thousands) Housing Units per 100 Households Affordable Affordable and Available Affordable, Available, and Adequate Affordable Affordable and Available Affordable, Available, and Adequate All 20,097 23,627 17,017 15,329 117.6 84.7 76.3 Northeast 4,478 5,153 3,912 3,497 115.1 87.4 78.1 Midwest 3,842 4,826 3,451 3,182 125.6 89.8 82.8 South 6,641 7,879 5,499 4,878 118.6 82.8 73.5 West 5,135 5,769 4,155 3,773 112.3 80.9 73.5 Cities 9,246 10,312 7,876 6,980 111.5 85.2 75.5 Suburbs 6,946 8,316 5,754 5,300 119.7 82.8 76.3 Nonmetropolitan areas 3,905 4,998 3,387 3,049 128.0 86.7 78.1 FMR = Fair Market Rent. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Trends in Rental Stock Through 2007, the availability of the affordable rental stock has been relatively stable for more than two decades. Exhibit 3-9 shows the available rental units for every 100 households for the four standard income categories during the 1985-to- 2007 period. 8 Exhibit 3-9. Availability of Affordable Rental Units, 1985–2007 8 This figure is based on custom tabulations of the American Housing Survey national data sets for odd-numbered years in the period. Worst Case Housing Needs 31
Availability turned downward in 2005 for renters of all income groups below 80 per- cent of AMI, but most of the losses were recovered in 2007. Middle-income renters continued to experience stable supply during the 2003-to-2007 period. Exhibit 3-10 examines the factors responsible for the change in the availability of affordable units from 2005 to 2007. The HUD income limits rose by approximately 5 percent during the period, although median household income (as measured in the AHS) rose about 5.25 percent. 9 Other things being equal, this implies that a larger proportion of households would be in the higher income categories in 2007 than in 2005. Table 3-10 supports this implication; it shows decreases in the number of extremely low-income and very low-income households, although the very low-income decrease is much smaller in percentage terms. At the same time, the number of housing units affordable and available to extremely low-income households rose by 6 percent, but the proportion affordable and available to very low-income households Exhibit 3-10. Factors Explaining Changes in Rental Housing Availability Rate, 2005–2007 Extremely Low- Income Units (30% AMI) Very Low-Income Units (50% AMI) Low-Income Units (80% AMI) All Units Cumulative households (thousands) 2005 9,982 16,324 23,812 33,951 2007 9,555 16,251 23,902 35,054 Percent change – 4.28 – 0.45 + 0.38 + 3.25 Cumulative affordable and available housing units (thousands) 2005 3,982 12,539 25,400 37,924 2007 4,224 12,011 25,207 39,330 Percent change + 6.08 – 4.21 – 0.76 + 3.71 Income limit (median, current dollars) 2005 14,804 24,665 39,402 NA 2007 15,500 25,850 41,200 NA Percent change + 4.70 + 4.80 + 4.56 NA Median household income (all renters, current dollars) 2005 26,000 2007 27,364 Percent change + 5.25 Median monthly housing cost (all renters, current dollars) 2005 672 2007 722 Percent change + 7.44 AMI = Area Median Income. NA = data is not available. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data 9 Note that AMI is calculated on the basis of all households, not just renters. Worst Case Housing Needs 32
fell by 4.21 percent. As a consequence, the relative supply of affordable and available units for extremely low-income households rose but that of affordable and available units for extremely low-income households fell, as exhibit 3-3 illustrated. The reason for the difference between changes in extremely low-income and very low-income units is not clear. The median monthly housing cost for all rental units rose 7 percent between 2005 and 2007, which would imply a general reduction in the affordable stock. Preliminary analysis of the dynamics of the rental housing stock during the period, however, suggests that the extremely low-income stock gained from rental units filtering downward in cost, while the stock of strictly very low-income units 10 experienced a net decline from filtering. In addition, the very low-income stock experienced a net loss of about 5 percent from units entering and leaving the rental stock, 11 while the net change in the extremely low-income stock from these sources was essentially zero. 12 Changes in Rental Stock, 2005 to 2007, by Affordability Categories Additional evidence of the supply of affordable rental housing can be drawn from the HUD report, Rental Market Dynamics: 2005–2007 (June 2009). 13 Exhibit 3-11 shows the change in rental units affordable to various income groups. Units are grouped into eight categories of affordability. At the lower end of the rent scale these groups include: non-market (either no cash rent or a subsidized rent), extremely low rent (affordable to renters with incomes that are less than 30 percent of the area median income), very low rent (affordable to renters with incomes between 30 and 50 percent of area median income), low rent (affordable to renters with incomes between 50 and 60 percent of area median income), and moderate rent (affordable to renters with incomes within 60 to 80 percent of area median income). Between 2005 and 2007, there was an increase in the overall U.S. rental stock of 1.4 million units (3.7 percent increase). However, at the lower end of the affordability scale, there was a large net loss of 1.14 million affordable units (10.8 percent) in the “very low rent” group—those rental units affordable to families between 30 and 50 percent of area median income. These data show that overall changes in rental hous- ing do not always directly benefit all segments of the stock itself. 10 That is to say, units affordable at the very low income threshold, but not affordable at the extreme- ly low income threshold. 11 This would include new construction, demolition, conversion to and from owner-occupied status, conversion to and from nonresidential use, and so on. 12 See “Rental Market Dynamics, 2005–2007,” (June 2009), prepared by Frederick J. Eggers and Fouad Moumen http://www.huduser.org/portal/datasets/cinch/cinch07/Rental_Mrkt_05-07.pdf 13 “Rental Market Dynamics, 2005–2007,” (June 2009). It is important to note that the weights used in the Rental Market Dynamics report may differ from those used elsewhere in this report and thus unit counts cited here may not always be directly comparable to this report’s unit counts. Worst Case Housing Needs 33
Exhibit 3-11. Rental Units by Affordability Categories, 2005–2007 (thousands) Affor dability Categories Rental Units in 2005 Rental Units in 2007 Change Percent Change Nonmarket 8,363,000 8,186,000 –177,000 –2.1 Extremely low rent 2,262,000 2,201,000 –61,000 –2.7 Very low rent 10,566,000 9,429,000 –1,137,000 –10.8 Low rent 5,914,000 6,309,000 395,000 + 6.7 Moderate rent 6,526,000 7,608,000 1,082,000 + 16.6 High rent 1,916,000 2,367,000 451,000 + 23.6 Very high rent 899,000 1,385,000 486,000 + 54.0 Extremely high rent 1,473,000 1,840,000 366,000 + 24.9 Total 37,919,000 39,324,000 1,405,000 + 3.7 Source: American Housing Survey Rental Market Dynamics: 2005–2007 (HUD PD&R, 2009) Crowding While crowding (defined as more than one person per room) is not a component of the definition of worst case housing needs, 14 it can be a symptom of affordability problems and housing-related stress. Households may double up, and young adults or couples may delay forming new households, because of an inability to afford their own units. This section examines the extent of crowding by income and location as well as the supply of large units relative to the number of large households. 15 Overall about 4.3 percent of renter households are crowded, as shown in Exhibit 3-12. The incidence of crowding is significantly lower in non-metropolitan areas, 3.2 percent, and higher in central cities, 5.1 percent. On a regional basis, the Midwest has substantially less crowding, 2.2 percent, and the West has substantially more (7.0 percent). These regional and metropolitan patterns match the ordering of the affordable and available stock ratios by metropolitan status and region, as illustrated in Exhibits 3-5 and 3-7. These parallels reinforce the hypothesis that crowding is a strategy for coping with reduced affordability. These patterns are slightly improved from 2005, when the overall incidence of crowd- ing was 4.8 percent. The incidence of crowding by region and metropolitan status has fallen about in proportion for all categories, leaving the relative incidence unchanged. 14 Crowding is classified as a moderate problem rather than a severe problem. 15 For a more thorough discussion of crowding and its measurement using AHS data, see Blake, Kev- in S. and Kellerson, Rebecca. Measuring Overcrowding in Housing. Prepared for the U.S. Department of Housing and Urban Development, Office of Policy Development and Research by ICF Consulting. September 2007. http://www.huduser.org/publications/pdf/Measuring_Overcrowding_in_Hsg.pdf . Worst Case Housing Needs 34
Exhibit 3-12. Geographic Pattern of Crowded Renter Households, 2007 Households (thousands) Incidence per 100 Households All areas 1,511 4.31 Metropolitan status Central cities 772 5.12 Suburbs 516 3.99 Nonmetropolitan areas 222 3.16 Region Northeast 290 4.10 Midwest 157 2.21 South 445 3.68 West 619 7.04 Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Larger households are much more likely than smaller households to be crowded. Exhibit 3-13 shows a substantially greater incidence of crowding among households with five or more persons, with 41 percent of such large households being crowded. Indeed, a renter household with five or more members is about eight times more likely to be crowded than a renter household with four persons. Despite the inverse relationship between income and crowding, extremely low-income households show less crowding than very low-income households do. The explanation is that extremely low-income households are more likely to be one-person house- holds, which by definition cannot be crowded. Exhibit 3-13. Characteristics of Crowded Renter Households, 2007 Households (thousands) Incidence per 100 Households All 1,511 4.31 Income* Extremely low-income renters (≤30% AMI) 515 5.57 Other very low-income renters (30–50% AMI) 437 6.52 Other low-income renters (50–8 0% AMI) 318 4.15 Middle-income renters (>80% AMI) 241 2.11 Household size One person NA NA Two persons 69 0.74 Three persons 29 0.55 Four persons 205 5.04 Five or more Persons 1,207 40.94 AMI = Area Median Income. NA = data not available. *Income categories in this table exclude lower income subsets. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 35
Crowding experienced by large families is not caused simply by a lack of large units. The number of affordable large units is abundant relative to the number of large households. In Exhibit 3-14, the Affordable line is entirely above the 100-units-per- large-household line that denotes sufficient stock; in fact, the number of affordable units with five or more rooms is two to five times larger than the number of house- holds with five or more persons. Exhibit 3-14. Insufficiency of Large Units is Primarily a Problem of Availability for Large Families, Not Affordability The main cause of crowding must be the lack of available affordable units. Thus, crowding does not appear to be caused by a lack of large units, but by the fact that smaller households prefer these units as well and keep them off the market. In addi- tion, large units may be concentrated in certain areas, so that they are not available to large households in other areas. Exhibit 3-15 summarizes the supply of rental units with five or more rooms relative to households with five or more persons. Even at 30 percent of AMI, there are 396.4 large units available for every 100 large renter households, and this increases to 602.6 per 100 for middle income renters; however, only 64.3 units are available per 100 extremely low-income households that need large units. Even for very low-income renters, there are only 86.4 units per 100 households. In terms of crowding experienced by families of different racial and ethnic types, Hispanic renter households are more likely to be crowded. In 2007, 9.5 percent of very low-income Hispanic households have no severe problems yet are overcrowded according to the one-person-per-room benchmark (Appendix, Table A-9). The cor- responding values are 1.2 percent for Non-Hispanic White households and 2.8 percent for Non-Hispanic Black households. Hispanic families are more likely to live in multi- Worst Case Housing Needs 36
Exhibit 3-15. Metropolitan Patterns of Supply of Large Units for Large Households (five or more persons and five or more rooms only), 2007 Units per 100 Large Households Affordable Affordable and Available Affordable, Available, and Adequate Nation Extremely low-income renters (0–30% AMI) 396.4 64.3 61.5 Very low-income renters (0–50% AMI) 479.5 86.4 81.9 Low-income renters (0–80% AMI) 602.6 122.9 116.2 Central Cities Extremely low-income renters (0–30% AMI) 236.7 48.2 44.8 Very low-income renters (0–50% AMI) 366.0 76.2 71.6 Low-income renters (0–80% AMI) 497.4 115.5 108.6 Suburbs Extremely low-income renters (0–30% AMI) 477.0 78.0 76.5 Very low-income renters (0–50% AMI) 481.7 82.0 79.8 Low-income renters (0–80% AMI) 646.7 119.6 115.6 Nonmetropolitan areas Extremely low-income renters (0–30% AMI) 705.1 87.9 84.8 Very low-income renters (0–50% AMI) 768.8 120.3 112.2 Low-income renters (0–80% AMI) 775.7 146.4 135.3 AMI = Area Median Income. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data generational households in non-caregiving relationships (U.S. Census Bureau, 2003). Non-Hispanic Blacks, and especially Whites, are less likely to live in this arrangement 16 and much less likely to be overcrowded. The importance of this association is compounded by the fact that Hispanics represent the fastest-growing segment of the U.S. populat ion. The Hispanic population increased by 58 percent between 1990 and 2000, and is projected to increase another 34 per- cent by 2010 (U.S. Census Bureau, 2004). Thus the availability of large units in regions where the Hispanic population is large is worthy of consideration. Hispanics are concentrated in the West, as 43 percent of the Hispanic population lived there in 2000 (U.S. Census Bureau, 2001). The West also is the region where the incidence of crowding is two-thirds greater than any other region (Exhibit 3-11) and it has the most severe shortage of affordable units for very low-income households, with only 56.9 units available for every 100 very low-income households (Exhibit 3-7). 16 While 8 percent of both black and Hispanic adults over 30 were grandparents living with their grandchildren in 2000 (compared with 2 percent of non-Hispanic whites over 30), only 35 percent of these Hispanic grandparents lived there to care for the grandchildren, compared with 52 percent of the black grandparents (U.S. Census Bureau 2003). Worst Case Housing Needs 37
Exhibit 3-16 shows that for every 100 large very low-income renter households in the West, only 43.1 large units are available. The availability of large affordable units is substantially more limited in the West than in other regions. The number of large units available for very low-income renters is fully sufficient in the Midwest and nearly so in the South, at least on a regional basis. Exhibit 3-16. Regional Patterns of Supply of Large Units for Large Households (five or more persons and five or more rooms only), 2007 Units per 100 Large Households Affordable Affordable and Available Affordable, Available, and Adequate Northeast Extremely low-income renters (0–30% AMI) 379.9 65.4 57.2 Very low-income renters (0–50% AMI) 520.2 80.5 74.2 Low-income renters (0–80% AMI) 651.2 104.9 95.8 Midwest Extremely low-income renters (0–30% AMI) 469.5 71.6 69.5 Very low-income renters (0–50% AMI) 710.7 139.2 133.3 Low-income renters (0–80% AMI) 817.9 181.8 173.6 South Extremely low-income renters (0–30% AMI) 415.8 73.7 71.7 Very low-income renters (0–50% AMI) 498.8 93.6 88.5 Low-income renters (0–80% AMI) 647.1 143.9 136.8 West Extremely low-income renters (0–30% AMI) 323.6 43.1 43.1 Very low-income renters (0–50% AMI) 284.5 47.6 46.0 Low-income renters (0–80% AMI) 396.1 72.7 68.7 AMI = Area Median Income. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Summary One way to assess the condition of the market for affordable rental housing is to compare, on a cumulative basis, the number of affordable units with the number of renters relative to their income levels. This chapter has presented three such ratios, based on (1) affordable units, (2) affordable units that also are occupied by such renters or are available for rent, and (3) affordable and available units that also are in physically adequate condition. Nationwide in 2007, there existed a rental unit that would have been affordable for every renter household with income above 45 percent of area median income. This is about what it was in 2005. However, the comparable figure in 2003 was 40 percent, implying that the affordable rental stock decreased relative to the number of house- holds in this critical very low-income range. Worst Case Housing Needs 38
The number of affordable units that are actually available to households with the low- est incomes was substantially fewer than the number of apparently affordable units because many affordable units were rented to higher income families. Employing the Availability measure, the national supply of affordable units per renter household does not become sufficient until household incomes reach 70 percent of AMI. Available units become progressively harder to obtain at lower income levels. While there were 113 affordable units available per 100 low-income renters in 2007, the availability ratio was only 74 per 100 very low-income renters and 44 per 100 ex- tremely low-income renters. National measures of affordability and availability do not fully reflect the regional concentrations or shortages of units relative to households. Nonmetropolitan areas have more affordable and available rental units than cities or suburbs. However, the immobility of housing makes such units inaccessible for very low-income renters who need them near their families and employment. In addition, larger households are much more likely than smaller households to be crowded, and large units also are distributed in areas where they are unavailable to large families. A substantial proportion of available units are physically inadequate. Employing the additional criterion of physical adequacy, the number of available units in 2007 was reduced from 74 to 67 units per 100 very low-income renters, and from 44 to 37 units per 100 extremely low-income renters. Occupancy by higher-income households restricts the supply of units renting for less than the Fair Market Rent to only about 85 percent of households who can afford only such units. This is about the same as the 84 units for every 100 households that were available in 2005 and an increase from the 79 units per 100 households that were available in 2003. The mixed change in availability of affordable units between 2005 and 2007 can be attributed to a modest decrease in renter households that interacted with differential filtering of the housing stock for extremely low-income and very low-income households. For very low-income renters, both affordability and avail- ability ratios declined from 2005 to 2007, and they declined even more from 2003 to 2007. For extremely low-income renters, affordability and availability ratios improved from 2005 to 2007, although affordability is still below the levels recorded in 2003 and availability is no better than it was in 2003. Worst Case Housing Needs 39
Chapter 4. Worst Case Needs and Neighborhoods Poverty and Neighborhood Distress This chapter examines worst case needs renters and the neighborhoods where they live. The analysis examines the poverty levels of neighborhoods where worst case needs renters live, and compares their residential patterns with all U.S. households and other very low-income renters. The comparison includes a breakdown of residen- tial living patterns across central cities, suburbs, and non-metropolitan areas. Concentrated neighborhood poverty can have a significant impact on a household’s well-being. For example, studies indicate that living in areas with poverty rates of 20 to 30 percent can have high social costs in terms of safety, quality of schools, munici- pal services, and access to employment opportunities. Evidence suggests that neigh- borhoods characterized by high concentrations of poverty of more than 40 percent can actually lead to a downward spiral of poverty for residents living there. 1 This chapter divides neighborhoods into four basic categories based on the percentage of people in poverty: 0 to 10 percent in poverty, 10 to 20 percent in poverty, 20 to 30 percent in poverty, and 30 percent or greater in poverty. 2 Nationwide, one-half of all neighborhoods (census tracts) fall into the first category with low poverty rates and 30 percent of neighborhoods fall into the second category with moderate poverty rates of 10 to 20 percent. The other two categories with high poverty rates (20 to 30 percent and 30 percent or greater) each contain 10 percent of the nation’s neighborhoods. 3 In this chapter, the prevalence of worst case needs households in each of the four neighborhood categories is compared with other household residential patterns: for all households in general, for renters, and for all very low-income renters. This analysis sheds some light on the question of whether the severe rent burden of worst case needs households may be due to a lack of affordable housing or a decision on their part to accept severe rent burdens as the price of locating to better neighborhoods when lower cost options are available. 1 See, for example, Jargowsky (1997), Kingsley and Pettit (2003), “Areas of Concentrated Poverty: 1999” U.S. Census Bureau (2005), and Galster (2006). 2 “Neighborhoods” in this chapter are defined as census tracts. Census tracts generally have a popu- lation of between 1,500 and 8,000 people, with an optimum size of 4,000 people. Poverty rates in this chapter are calculated excluding college students, because the presence of students in some ar- eas may result in a tract appearing to be a high-poverty area while the poverty rates of year-round residents may be quite low. 3 The census tracts and appropriate neighborhood types are merged with the American Housing Sur- vey (AHS) file. Weighting of the AHS sample provides estimates of housing units for each neighbor- hood and other specified categories Worst Case Housing Needs 40
Worst Case Needs Households and Neighborhood Poverty Rates National Neighborhood Characteristics and Residential Patterns The first two exhibits in this chapter highlight basic characteristics of the different neighborhood types and enable comparison of residential patterns of worst case needs and other types of households nationally. First, Exhibit 4-1 shows the four basic categories of neighborhoods by poverty rate and key socio-economic factors. These factors include percent of owner-occupied housing, unemployment, educational achievement of residents, income levels, and reported neighborhood crime. As shown in the Exhibit, the factors are closely related with the overall poverty rate of the neighborhood, providing a good indication that neighborhood poverty rate often reflects a whole host of other socio-economic issues. This outcome is supported by a substantial amount of research that shows important correlations between neighborhood well-being and poverty rates. Exhibit 4-1. Neighborhood Characteristics < 10% 10–20% 20–30% 30% or > Average Neighborhood Characteristics a Population density b 3,906 6,825 11,868 17,005 6,233 Percent in owner-occupied housing 75.3 58.0 45.1 33.9 65.4 Percent unemployed c 3.7 6.4 9.9 15.5 5.7 Percent bachelors degree or higher 36.8 21.1 15.9 11.8 28.8 Average median family income 65,795 41,920 32,141 23,547 53,239 Central cities 68,120 43,537 32,026 23,096 49,050 Suburbs 68,681 43,274 32,912 24,660 61,175 Non-metropolitan areas 54,559 39,728 31,779 24,307 44,379 Percent of households that responded: Neighborhood has crime 12.0 16.3 24.2 32.0 15.5 Public elementary school is unsatisfactory 1.3 1.8 2.6 3.7 1.7 a Average values of neighborhood characteristics are given for each neighborhood type. The statistics on educational attainment, neighborhood crime, and satisfaction with elementary schools are from the 2007 American Housing Survey (AHS). The remaining variables are from the 2000 Decennial Census. b Population per square mile. c The percent unemployed excludes college towns. It is from the 2000 Decennial Census and may vary from the Consumer Population Survey (CPS) measure of unemployment. The unemployment rate in 2007 was 4.6 percent. Source: Census Bureau tabulations of 2007 AHS and 2000 Decennial Census data performed under contract to Department of Housing and Urban Development Worst Case Housing Needs 41
Exhibit 4-2 shows the residential distribution of all households nationally across the four types of neighborhoods. It provides information on where renters and homeown- ers reside across the neighborhood types and whether these households are located within a central city or suburb of a metropolitan statistical area (MSA), or in a non- metropolitan area outside an MSA. This information will be useful for comparison with the chapter’s later findings on the neighborhood distribution of worst case needs households. These data will be used to compare the neighborhood quality of worst case needs households with that of other households and renters. Generally, the Exhibit shows that most of the poorer neighborhoods are located in central cities, and relatively few exist in suburban and non-metropolitan areas. The results show that 28 percent of all housing units in central cities are located in high- poverty neighborhoods. In comparison, only 6 percent of housing in the suburbs and only 14 percent of housing in non-metropolitan areas are located in neighborhoods with high levels of poverty. Nationwide, out of an estimated 16.6 million housing units located in poorer neighborhoods (greater than 20 percent poverty), more than one-half (8.7 million) are situated in central cities. In contrast, less than one-fifth (2.9 million) of housing units in high poverty neighborhoods are located in suburban areas and less than one-fourth (4.0 million) are found in non-metropolitan areas. Exhibit 4-2. Distribution of Renter Households by Neighborhood Poverty and Metropolitan Status Census Tracts with Non-College Poverty Rates of: (Numbers in 1,000s) < 10% 10–20% 20–30% 30% or > Total Central Cities Renters in central cities 4,311 5,172 2,653 2,955 15,091 percent of renters in central cities 28.6 34.3 17.6 19.6 100 All occupied housing units 12,677 9,980 4,500 4,242 31,399 percent of occupied units in central cities 40.4 31.8 14.3 13.5 100 Suburbs Renters in suburban areas 7,606 3,811 1,035 474 12,926 percent of renters in suburban areas 58.8 29.5 8.0 3.7 100 All occupied housing units 36,502 10,226 2,004 939 49,671 percent of occupied units in suburban areas 73.5 20.6 4.0 1.9 100 Non-Metropolitan Areas Renters in non-metropolitan areas 2,220 3,439 926 453 7,037 percent of renters in non-metropolitan areas 31.5 48.9 13.2 6.4 100 All occupied housing units 12,000 13,677 2,832 1,140 29,649 percent of occupied units in non-metro areas 40.5 46.1 9.6 3.8 100 National Total All renters 14,137 12,422 4,614 3,882 35,054 percent of all renters 40.3 35.4 13.2 11.1 100 All occupied housing units 61,179 33,883 9,336 6,321 110,719 percent of all occupied housing units 55.3 30.6 8.4 5.7 100 Source: Census Bureau tabulations of 2007 American Housing Survey and 2000 Decennial Census data performed under contract to Department of Housing and Urban Development Worst Case Housing Needs 42
Distribution of Worst Case Needs Households By Neighborhood Type Exhibit 4-3 illustrates where worst case needs households reside by neighborhood type, both nationwide and across central cities, suburbs and non-metropolitan areas. Nationally, 70.2 percent of worst case needs households live in neighborhoods with less than 20 percent poverty: 34.9 percent live in low poverty neighborhoods and 35.3 percent live in neighborhoods with moderate poverty rates. Nationally, 29.8 percent of worst case needs households reside in poorer neighborhoods with poverty rates greater than 20 percent, including 13.6 percent in neighborhoods with poverty rates greater than 30 percent. Exhibit 4-3. Distribution of Worst-Case and Very Low-Income Renters by Neighborhood Poverty and Metropolitan Status Census Tracts with Non-College Poverty Rates of: (Numbers in 1,000s) < 10% 10–20% 20–30% 30% or > Total Central Cities Worst-case renters 670 908 605 617 2,801 as percent of all worst-case renters 23.9 32.4 21.6 22.0 100.0 VLI renters 1,504 2,422 1,484 2,010 7,420 as percent of all VLI renters 20.3 32.6 20.0 27.1 100.0 Gap: % WCN > % all VLI renters 3.7 –0.2 1.6 –5.1 Suburbs Worst-case renters 1,081 652 203 115 2,051 as percent of all worst-case renters 52.7 31.8 9.9 5.6 100.0 VLI renters 2,563 1,796 569 310 5,238 as percent of all VLI renters 48.9 34.3 10.9 5.9 100.0 Gap: % WCN > % al l VLI renters 3.8 –2.5 –1.0 –0.3 Non-Metropolitan Areas Worst-case renters 310 522 150 72 1,053 as percent of all worst-case renters 29.4 49.6 14.2 6.8 100.0 VLI renters 806 1,670 515 293 3,282 as percent of all VLI renters 24.6 50.9 15.7 8.9 100.0 Gap: % WCN > % all VLI renters 4.9 –1.3 –1.4 –2.1 Total, All Areas Worst-case renters 2,061 2,082 958 804 5,905 as percent of all worst-case renters 34.9 35.3 16.2 13.6 100.0 VLI renters 4,873 5,887 2,568 2,613 15,940 as percent of all VLI renters 30.6 36.9 16.1 16.4 100.0 Gap: % WCN > % all VLI renters 4.3 –1.7 0.1 –2.8 Source: Census Bureau tabulations of 2007 American Housing Survey and 2000 Decennial Census data performed under contract to Department of Housing and Urban Development Worst Case Housing Needs 43
Further analysis by central cities, suburbs, and non-metropolitan areas provides some interesting findings. Worst case needs households are much more concentrated in high poverty neighborhoods in central cities and much less so in both suburbs and non-metropolitan areas. Although most of the 56.3 percent of worst case households in central cities still reside in neighborhoods with less than 20 percent poverty, the 43.6 percent of the distribution that live in high poverty areas is much higher than the national average of 29.8 percent of all worst case needs households that live in such areas nationally. The suburban and non-metropolitan areas have much lower proportions of worst case households, 15.5 and 21.0 percent, respectively, in their high poverty neighborhoods. These lower rates in both suburban and non-metropolitan areas reflect the fact that far fewer housing units overall are located in high poverty neighborhoods in these geographic areas, as shown in Exhibit 4-2. The remainder of this chapter compares the residential patterns of worst case needs renters with those of all U.S. households, all renters, and very low-income renters overall. Comparison With All U.S. Households and Renters The comparison of worst case needs renters shown in Exhibit 4-3 with all renters and homeowners shown in Exhibit 4-2 illustrates that a considerably higher por- tion of worst case needs households live in high poverty neighborhoods. The 29.8 percent of worst case needs households that reside in neighborhoods with poverty rates greater than 20 percent is disproportionately higher than the 24.3 percent of all renter households and the 14.1 percent of all households. In contrast, the low- poverty neighborhoods include only 34.9 percent of worst case households, despite containing 40.3 percent of all renters and 55.3 percent of all households. These results might be expected because the concentration of worst case needs households in neighborhoods with high poverty rates reflects the lower rents that typically exist in poorer neighborhoods. Comparison With All U.S. Very Low-Income Renter Households How the neighborhood quality of worst case needs households compares with that of all very low-income renters is also of great interest. Because worst case needs households are very low-income renters themselves, one might expect their choice of neighborhoods within each locality to be roughly consistent with very low-income renters overall. Indeed, the distribution of very low-income renters presented in Exhibit 4-3 shows a very similar distribution to worst case needs households among the different neighborhood types. The residential patterns of worst case needs renters also track very similarly to those of very low-income renters overall in each of the three types of geography considered: central cities, suburbs, and non-metropolitan areas. In central cities, 43.6 percent of worst case needs households live in neighborhoods with poverty rates of 20 percent or more. This rate is only slightly below the 47.1 percent rate of all very low-income Worst Case Housing Needs 44
renters who live in these areas. Housing is relatively less expensive in neighborhoods with high rates of poverty, but worst case needs households still experience severe rent burden living there. As discussed previously, the suburbs and non-metropolitan areas present a very different picture. Much lower percentages of worst case needs households reside in high-poverty neighborhoods in the suburbs and non-metropolitan areas because few of these types of neighborhoods exist in these geographic areas. In suburban areas approximately one-half of both worst case needs households and renters with very low incomes live in neighborhoods with the lowest poverty rates. These high percentages of very low-income renters in low-poverty neighborhoods reflect the fact that numer- ous suburban areas have few neighborhoods with high poverty rates. As a result, the only suburban residential options that very low-income renters have are likely to be located in low-poverty neighborhoods, but cause them severe rent burdens. In non-metropolitan areas, nearly 30 percent of worst case needs households live in low poverty neighborhoods and about one-half of both worst case needs households and all very low-income renters live in neighborhoods with moderate poverty levels. Similar to suburban areas with low-poverty neighborhoods, numerous non-metropoli- tan areas consist predominantly of neighborhoods with moderate poverty rates and few high poverty ones. These areas, therefore, present very low-income renters with few choices about whether to lease housing that would result in a very high rent burden. One interesting difference between worst case needs households and very low-income renters overall is the small national gap of 4.3 percentage points between the propor- tion of worst case needs households (34.9 percent) and that of all very low-income renters (30.6 percent) residing in low poverty areas. Although very small, this dif- ference seems to show that worst case needs renters are slightly more likely to live in low poverty neighborhoods than very low-income renters overall. However, when renter households receiving government housing assistance are taken into account, the 4.3 percentage difference is reduced to only 1.4 percent (which is not statistically significant). This is because assisted households tend to be more concentrated in higher poverty neighborhoods for a variety of reasons, including the locations where public housing was constructed decades ago (Table A-15 in Appendix A shows more detail). This analysis thus does not provide support for the contention that worst case needs households in the aggregate are incurring severe rent burdens as a result of “strategic behavior” to locate in better neighborhoods. As stated, a goal of this chapter is to explore the question of whether worst case needs households may be incurring severe rent burdens as a strategic decision to “buy up” into better quality neighborhoods. Overall, the data presented in this chapter do not provide much support for that contention. Rather, there is significant evidence that the lack of affordable housing, regardless of location, is the reason why households with worst case needs experience severe rent burden. Worst Case Housing Needs 45
Summary This chapter examined the extent to which worst case needs households are living in distressed neighborhoods. It also explored the related question of whether worst case needs households incurred severe rent burdens in order to live in low-poverty neighborhoods. A majority of worst case needs households live in lower poverty areas. However, a considerably higher percentage of worst case needs households live in high-poverty neighborhoods compared with all U.S. households overall. Nearly one-third of households with worst case housing needs were living in high-poverty neighbor- hoods in 2007 compared with less than one-fourth of all renters and less than 15 percent of all households. More than 40 pe rcent of worst case needs households in central cities live in high- poverty neighborhoods. Even in high-poverty neighborhoods, where housing is less expensive, worst case needs households still incur severe rent burdens living there. The prevalence of worst case needs households in the four neighborhood types classi- fied by poverty level corresponds very closely to that of very low-income renters. This was true even though worst case needs households share the same income group as very low-income renters but sustain severe rent burdens. In very low-poverty neigh- borhoods nationally, only a 4.3-percentage point disparity was found in the concentra- tion of worst case needs households compared with very low-income renters. For worst case needs households in low-poverty neighborhoods, no strong evidence supported the hypothesis that, overall, they incurred severe cost burdens in order to live in those more favorable neighborhoods. Worst Case Housing Needs 46
Chapter 5. Background and Methodology Since 1991, the Department of Housing and Urban Development (HUD) has issued regular reports to Congress on worst case needs for housing assistance among the na- tion’s very low-income renters. These reports developed from requests from Congres- sional Committees in the 1980s for information on housing needs. In 1990, the U.S. Senate Appropriations Committee directed HUD to “resume the annual compilation of a worst case housing needs survey of the United States... [to estimate] the number of families and individuals whose incomes fall 50 percent below an area’s median in- come, who either pay 50 percent or more of their monthly income for rent, or who live in substandard housing.” 1 This report is the twelfth in a series of Worst Case Needs reports to Congress, prepared by HUD’s Office of Policy Development and Research. 2 Households with worst case needs are defined as unassisted renters with very low incomes (below 50 percent of the area median income) who pay more than one-half of their income for housing or live in severely substandard housing. HUD originally developed the definition of worst case needs in consultation with the Office of Man- agement and Budget and Congressional Committees. It was based on the federal preference rules that prioritized admissions for housing assistance programs in the 1980s and early 1990s. To assess changes over time, HUD has retained this consistent definition of worst case needs for affordable housing. Although federal preferences for housing assistance have been repealed, the current definition of worst case needs is still highly useful. As stated, a consistent definition makes it possible to assess changes over time, and severe rent burden and physical adequacy of living conditions remain key indicators of the overall need for safe and affordable housing. 1 Committee Report to accompany H.R. 5158, The VA-HUD Appropriations Act for FY 1991 (S. Rpt. 101-474). 2 HUD’s previous reports to Congress about worst case housing needs are as follows: • Priority Problems and “Worst Case” Needs in 1989 (June 1991, HUD-1314-PDR). • The Location of Worst Case Needs in the Late 1980s (December 1992, HUD-1387-PDR). • Worst Case Needs for Housing Assistance in the United States in 1990 and 1991 (June 1994, HUD-1481-PDR). • Rental Housing Assistance at a Crossroads: A Report to Congress on Worst Case Housing Needs (March 1996). • Rental Housing Assistance—The Crisis Continues (April 1998). • Rental Housing Assistance—The Worsening Crisis: A Report to Congress on Worst Case Housing Needs (March 2000). • A Report on Worst Case Housing Needs in 1999: New Opportunity Amid Continuing Challenges, Executive Summary (January 2001). • Trends in Worst Case Needs for Housing, 1978–1999 (December 2003). • Affordable Housing Needs: A Report to Congress on the Significant Need for Hou sing (December 2005). • Affordable Housing Needs 2005: Report to Congress (May 2007). • Housing Needs of Persons With Disabilities: Supplemental Findings to the Affordable Housing Needs 2005 Report (February 2008). These publications are available on line at http://www.huduser.org. Worst Case Housing Needs 47
The American Housing Survey This report uses data from the latest available American Housing Survey (AHS) in 2007. The AHS is sponsored by HUD and conducted by the U.S. Census Bureau and is the only detailed periodic national housing survey in the United States. It provides nationally representative data on a wide range of housing subjects, including apart- ments, single-family homes, mobile homes, vacant homes, family composition, income, housing and neighborhood quality, housing costs, equipment, fuel type, size of housing units, and recent moves. National data are collected every 2 years from a sample of about 53,000 housing units. The survey, which started in 1973, has sampled the same housing units since 1985; it also samples newly constructed units to ensure both continuity and timeliness of the data. Information from the Worst Case Needs reports has helped inform public policy decisions, including decisions on target- ing of existing resources, determining the need for additional resources, and the form housing assistance should take. Explanation of Household Income Categories Many HUD programs and other federal housing programs use specific income limits to determine whether households qualify for those programs. HUD has developed a useful means of establishing these income limits so they reflect area income levels. Income limits are set on the basis of area median incomes for each metropolitan area and non-metropolitan county. Area median incomes are also adjusted for family size before income limits are determined. The terms “low-income,” “very low-income,” and “extremely low-income” used in this report follow the specific meanings of those terms as used in several of HUD’s afford- able housing programs: • Low-income. Not more than 80 percent of area median income. Defined by the United States Housing Act of 1937 and used as an income limit for many rental and homeownership programs. • Very low-income. Not more than 50 percent of area median income. Defined by the United States Housing Act of 1937 and used as an income limit for many rental programs. • Extremely low-income. Not more than 30 percent of area median income. Although “extremely low income” is not a defined term in the U.S. Housing Act of 1937, the income threshold itself is used for the purpose of establishing admissions standards in HUD’s major rental assistance programs. Exhibit 5-1 illustrates income limits for selected U.S. cities. For the detailed methodology on setting HUD income limits for fiscal year 2007, see http://www.huduser.org/datasets/il/il07/IncomeLimitsBriefingMate- rial.pdf. Worst Case Housing Needs 48
Exhibit 5-1. HUD Income Limits in Selected Cities, Fiscal Year 2007 Annual Income for a Four-Person Household 30% AMI ($) 50% AMI ($) 80% AMI ($) AMI ($) New York City 21,250 35,450 56,700 56,800 Los Angeles 22,200 37,000 59,200 56,500 Chicago 22,600 37,700 59,600 69,800 Houston 18,300 30,500 48,800 57,300 Philadelphia 21,650 36,050 57,700 71,600 Phoenix 18,050 30,050 48,100 59,100 Jacksonville 18,100 30,150 48,250 59,700 Washington, DC 28,350 47,250 60,000 94,500 Denver 21,500 35,850 57,350 71,400 Atlanta 21,350 35,600 56,950 67,100 Seattle 23,350 38,950 59,600 75,600 Worst Case Needs by Definition Are Below the “Very Low-Income” Threshold Exhibit 5-2 shows how the two severe problems (severely inadequate housing and severe rent burden), represented by the smaller ovals, combine with the very low- income limits, represented by the large oval, to constitute worst case housing needs . The large oval represents the total universe of 11.57 million unassisted very low- income renter households, as reported by the AHS in 2007. Of these very low-income renter households, 5.67 million do not have either of the severe problems, while 5.91 million households have one or both of the severe problems constituting worst case housing needs. Exhibit 5-2. Bird’s Eye View of Worst Case Housing Needs in 2007 Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 49
Although the United States has a total of 35.05 million renter households, the diagram does not show the 19.11 million renters who are outside the largest oval, because they have incomes above the very low-income threshold or receive some form of housing assistance. A substantial number of these households—those represented by the overhanging portions of the small ovals—also face severe problems, especially se- vere rent burden. Among all renters, 8.45 million have one or both severe problems. Housing Assistance and Affordable Housing Programs HUD provides rental housing assistance through three key programs: • Public housing. This program provides affordable housing to 1.1 million families. The units are owned and managed by local public housing agencies (PHAs). Families are required to pay 30 percent of their income for rent. • Project-based assisted housing. The project-based Section 8 program provides assistance to 1.3 million families living in privately owned rental housing. The assistance is attached to the units, which are reserved for low-income families that are required to pay 30 percent of their income for rent. • Tenant-based rental assistance. Administered through local PHAs, the Section 8 voucher program assists more than 2.0 million families to help pay their rent in the private rental market. Although 30 percent is a baseline, families often pay more than 30 percent of their income in rent and may use these portable subsidies to locate housing of their choice. A number of other federal housing programs produce affordable housing, typically with shallower subsidies. Although these units are often more affordable than market-rate units, without additional rental subsidies (such as vouchers), extremely low-income families would often have to pay well over 30 percent of their income to occupy for units in these programs. These programs include the following: • Low Income Housing Tax Credit (LIHTC) program. This tax credit program sub- sidizes the capital costs of units that will have rents affordable to households with incomes at or below 60 percent of area median income. Between 1995 and 2006, approximately 16,800 projects with 1.23 million units were placed into service through the LIHTC program. 3 3 For more information on the LIHTC program, see: U.S. Department of Housing and Urban Develop- ment, Office of Policy Development and Research, January 2009. Updating the Low-Income Housing Tax Credit (LIHTC) Database: Projects Placed in Service Through 2006. Available at: http://www. huduser.org/intercept.asp?loc=/Datasets/lihtc/report9506.pdf. Worst Case Housing Needs 50
• HOME Investment Partnership (HOME). This program provides annual formula grants to state and local governments that can be used to assist homeowners, first-time homebuyers, or renters. Between 1992 and 2007, HOME provided subsidies for 307,000 affordable rental units (138,000 new construction, 153,000 rehabilitation, and 16,000 acquisition). Qualifying rents must be affordable to households with incomes at or below 65 percent of area median income or must be below local Fair Market Rents, whichever is less. • Housing Opportunities for Persons with AIDS (HOPWA). HOPWA funds have been available to state and local governments and non-profits by annual for- mula and competitive gra nts since 1992. Currently, 67,000 low-income house- holds receive housing assistance that serves as a base for participating in care and HIV treatment. Assistance is targeted to a special needs population. Grantees report that 77 percent of recipients have extremely low incomes, and another 16 percent have very low incomes. • Older rental subsidy programs. Programs named for sections of the National Housing Act, primarily the Section 221(d)(3) Below Market Interest Rate (BMIR) Program and the Section 236 mortgage assistance program were active from the early 1960s through the early 1970s. They were designed to produce housing affordable for families with incomes above the public housing income limits. For further detail on HUD program requirements, see “Programs of HUD, 2006: Major Mortgage, Grant, Assistance, and Regulatory Programs” (HUD 2006a). Worst Case Housing Needs 51
References Berube, Alan. 2006. Using the Earned Income Tax Credit To Stimulate Local Econo- mies. Washington, DC: Brookings Institution. http://www.brook.edu/metro/pubs/ Berube20061101eitc.pdf. Blake, Kevin S., and Rebecca Kellerson. 2007 (September). Measuring Overcrowding in Housing. Prepared for the U.S. Department of Housing and Urban Development, Office of Policy Development and Research by ICF Consulting. http://www.huduser. org/publications/pdf/Measuring_Overcrowding_in_Hsg.pdf. Eggers, Frederick J., and Fouad Moumen. 2007. Rental Market Dynamics: 2003–2005. U.S. Department of Housing and Urban Development, Office of Policy Development and Research. http://www.huduser.org/datasets/cinch.html. ______. 2009. American Housing Survey Rental Market Dynamics: 2005–2007. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. http://www.huduser.org/datasets/cinch/cinch07/ Rental_Mrkt_05-07.pdf. Galster, George,2002. “An Economic Efficiency Analysis of Deconcentrating Poverty Populations” Journal of Housing Economics 11:303-29. Jargowsky, Paul A. 1997. Poverty and Place; Ghettos, Barrios, and the American City. New York: Russell Sage Foundation. Joint Center for Housing Studies of Harvard University. 2006. State of the Nation’s Housing: 2006. Cambridge, MA: Joint Center for Housing Studies of Harvard Univer- sity. www.jchs.harvard.edu. Khadduri, Jill, Kimberly Burnett, and David Rodda. 2003. Targeting Housing Production Subsidies: Literature Review. U.S. Department of Housing and Urban Development, Office of Policy Development and Research. http://www.huduser.org/publications/pdf/ targetinglitreview.pdf. Kingsley, G. Thomas, and Kathryn L.S. Pettit. 2003. “Concentrated Poverty: A Change in Course.” In Neighborhood Change in Urban America, No. 2. Washington, DC: Urban Institute. www.urban.org/nnip. Shroder, Mark. 2002. “Does Housing Assistance Perversely Affect Self-Sufficiency? A Review Essay,” Journal of Housing Economics 11: 381–417. U.S. Census Bureau. 2001. The Hispanic Population: 2000. Census 2000 Brief C2KBR/01-3. Washington, DC: U.S. Census Bureau. ______. 2003. Grandparents Living With Grandchildren: 2000. Census 2000 Brief C2KBR-31. Washington, DC: U.S. Census Bureau. ______. 2004. U.S. Interim Projections by Age, Sex, Race, and Hispanic Origin: 2000- 2050. http://www.census.gov/ipc/www/usinterimproj. Washington, DC: U.S. Census Bureau. Worst Case Housing Needs 52
______. 2005. American Housing Survey for the United States: 2003. Series H150/03, Housing and Household Economic Statistics Division. http://www.census. gov/hhes/www/housing/ahs/ahs.html. Washington, DC: U.S. Census Bureau. U.S. Department of Housing and Urban Development (HUD), Office of Policy De- velopment and Research (PD&R). 2003. Trends in Worst Case Needs for Housing, 1978–1999: A Report to Congress on Worst Case Housing Needs—Plus Update on Worst Case Needs in 2001. http://www.huduser.org/publ ications/affhsg/worstcase03. html. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. ______. 2005a. Affordable Housing Needs: A Report to Congress on the Significant Need for Housing. http://www.huduser.org/publications/affhsg/affhsgneed.html. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. ______. 2005b. “FY 2005 HUD Income Limits Briefing Material.” http://www.huduser. org/datasets/il/il05/index.html. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. ______. 2006a. Programs of HUD: Major Mortgage, Grant, Assistance, and Regulatory Programs. http://www.huduser.org/resources/hudprgs/ProgOfHUD06.pdf. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Develop- ment and Research. ______. 2006b. U.S. Housing Market Conditions. http://www.huduser.org/periodicals/ ushmc.htm. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. ______. 2006c. Fair Market Rents. http://www.huduser.org/datasets/fmr.html. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. ______. 2006d. Codebook for the American Housing Survey, Public Use File: 1997 and Later. http://www.huduser.org/datasets/ahs/ahsprev.html. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. ______. 2008a. The Third Annual Homeless Assessment Report to Congress. http:// www.hudhre.info/documents/3rdHomelessAssessmentReport.pdf. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. ______. 2008b. Housing Needs of Persons With Disabilities: Supplemental Findings to the Affordable Housing Needs 2005 Report.” (February 2008). http://www.huduser. org/publications/affhsg/affhsgneedsdis.html. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. ______. 2009. Updating the Low-Income Housing Tax Credit (LIHTC) Database: Projects Placed in Service Through 2006. http://www.huduser.org/intercept.asp?loc=/ Datasets/lihtc/report9506.pdf. Washington, DC: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. Worst Case Housing Needs 53
Appendix A. Data on Housing Problems and Supply of Affordable Housing List of Tables Table A–1a. Housing Conditions of Renter Households by Relative Income, 2005 and 2007 .....................................................................57 Table A–1b. Housing Conditions of Owner Households by Relative Income, 2005 and 2007 .....................................................................58 Table A–2a. Housing Conditions of Renters and Owners, 1995–2007—Number of Households.................................................................59 Table A–2b. Housing Conditions of Renters and Owners, 1995–2007—Percentage of Households ............................................................ 60 Table A–3. Housing Conditions of Unassisted Renter Households by Relative Income, 2005 and 2007 .....................................................................61 Table A–4. Incidence of Housing Problems Among Renters by Relative Income, 2005 and 2007—Number and Percentage ................................. 62 Table A–5. Incidence of Housing Problems Among Very Low-Income Renters by Household Type, 2005 and 2007—Number and Percentage.................................. 65 Table A–6a. Housing Problems and Characteristics of Very Low-Income Renters by Household Type, 2007 .................................................. 67 Table A–6b. Housing Problems and Characteristics of Extremely Low-Income Rente rs by Household Type, 2007 .................................................. 69 Table A–7. Housing Problems and Characteristics of Worst Case Renters by Household Type, 2007....................................................................................71 Table A–8. Housing Problems and Characteristics of Extremely Low-Income Worst Case Renters by Household Type, 2007..................................73 Table A–9. Incidence of Housing Problems Among Very Low-Income Renters by Race and Ethnicity, 2005 and 2007—Number and Percentage ..............................75 Table A–10. Incidence of Housing Problems Among Very Low-Income Renters by Region, 2005 and 2007—Number and Percentage ..............................................77 Table A–11. Incidence of Housing Problems Among Very Low-Income Renters by Metropolitan Location, 2005 and 2007—Number and Percentage..........................79 Table A–12. Households Occupying U.S. Rental Units by Affordability of Rent and Income of Occupants, 2005 and 2007 ..............................................................81 Table A–13. Renters and Rental Units Affordable and Available to Them by Relative Income, 1995–2007...........................................................................82 Table A–14. Average Income and Average Gross Rent of Renter Households by Relative Income, 2005 and 2007 .....................................................................83 Worst Case Housing Needs 54
Table A–1a. Housing Conditions of Renter Households by Relative Income, 2005 and 2007 Household Income as Percentage of HUD-Adjusted Area Median Family Income 2007 0–30% 30–50% 50–80% 80–120% >120% All Incomes Total households (1,000) 9,243 6,697 7,650 5,518 5,946 35,054 Unassisted with severe problems a 4,327 1,578 681 245 162 6,993 Unassisted with non-severe problems only a 828 2,935 2,960 1,048 674 8,445 Unassisted with no problems a 738 1,168 3,541 4,059 4,949 14,455 Assisted b 3,350 1,016 468 166 161 5,161 Any with severe problems 5,732 1,711 703 253 165 8,564 Rent burden >50% of income 5,588 1,579 486 98 42 7,793 Severely inadequate housing 372 198 224 156 123 1,073 Any with non-severe problems only c 1,778 3,375 3,095 1,077 690 10,015 Rent burden 30–50% of income 1,536 3,062 2,524 684 293 8,099 Moderately inadequate housing 324 393 494 308 307 1,828 Crowded housing 196 369 305 99 127 1,096 Any with no problems 1,733 1,611 3,853 4,187 5,090 16,476 Household Income as Percentage of HUD-Adjusted Area Median Family Income 2005 0–30% 30–50% 50–80% 80–120% >120% All Incomes Total households (1,000) 9,729 6,342 7,488 5,449 4,943 33,951 Unassisted with severe problems a 4,643 1,349 472 264 132 6,860 Unassisted with non-severe problems only a 981 2,604 2,363 799 557 7,303 Unassisted with no problems a 816 1,127 3,661 3,832 3,804 13,240 Assisted 3,289 1,262 992 555 450 6,547 Any with severe problems 6,151 1,548 532 290 145 8,665 Rent burden >50% of income 5,999 1,361 321 147 61 7,891 Severely inadequate housing 430 228 215 142 84 1,100 Any with non-severe problems only c 1,804 3,253 2,727 909 593 9,286 Rent burden 30–50% of income 1,535 2,934 2,091 452 226 7,238 Moderately inadequate housing 340 417 488 325 289 1,859 Crowded housing 229 346 347 170 109 1,200 Any with no problems 1,775 1,542 4,229 4,249 4,204 16,000 a See table A–3 for housing problems experienced by unassisted renters. b The American Housing Survey (AHS) survey questions on housing assistance changed between 2005 and 2007; see Appendix B. c See Table A–2 for estimates of the incidence of non-severe problems without regard to whether severe problems are also present. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 55
Table A–1b. Housing Conditions of Owner Households by Relative Income, 2005 and 2007 Household Income as Percentage of HUD-Adjusted Area Median Family Income 2007 0–30% 30–50% 50–80% 80–120% >120% All Incomes Total households (1,000) 6,192 7,630 12,213 13,195 36,436 75,665 Unassisted with severe problems 3,667 2,136 1,991 1,101 1,057 9,951 Unassisted with non-severe problems only 1,185 2,264 3,370 3,312 4,175 14,307 Unassisted with no problems 1,339 3,230 6,851 8,783 31,204 51,407 Assisted — — — — — — Any with severe problems 3,667 2,136 1,991 1,101 1,057 9,951 Cost burden >50% of income 3,595 2,077 1,858 979 838 9,347 Severely inadequate housing 143 86 149 123 231 732 Any with non-severe problems only 1,185 2,264 3,370 3,312 4,175 14,307 Cost burden 30–50% of income 1,019 2,035 2,945 2,983 3,710 12,691 Moderately inadequate housing 189 199 304 234 410 1,336 Crowded housing 54 140 290 190 145 820 Any with no problems 1,339 3,230 6,851 8,783 31,204 51,407 Household Income as Percentage of HUD-Adjusted Area Median Family Income 2005 0–30% 30–50% 50–80% 80–120% >120% All Incomes Total households (1,000) 7,473 7,614 11,820 15,230 32,812 74,950 Unassisted with severe problems 4,444 1,836 1,481 850 671 9,282 Unassisted with non-severe problems only 1,670 2,183 3,555 3,300 2,838 13,546 Unassisted with no problems 1,360 3,596 6,784 11,080 29,302 52,122 Assisted — — — — — — Any with severe problems 4,444 1,836 1,481 850 671 9,282 Cost burden >50% of income 4,360 1,744 1,347 685 405 8,542 Severely inadequate housing 201 130 156 170 266 923 Any with non-severe problems only 1,670 2,183 3,555 3,300 2,838 13,546 Cost burden 30–50% of income 1,508 1,919 3,099 2,876 2,378 11,780 Moderately inadequate housing 195 224 352 260 353 1,384 Crowded housing 56 142 255 237 146 837 Any with no problems 1,360 3,596 6,784 11,080 29,302 52,122 Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 56
Table A–2a. Housing Conditions of Renters and Owners, 1995–2007—Number of Households 1995 1997 1999 2001 2003 2005 2007 Total households (1,000) 97,694 99,487 102,802 105,435 105,868 108,901 110,719 Unassisted with severe problems 11,744 12,206 12,203 13,494 13,398 16,142 16,944 Unassisted with non-severe problems only 17,693 17,900 18,237 19,217 19,790 20,849 22,752 Unassisted with no 63,023 63,682 66,163 66,445 66,468 65,362 65,862 problems Assisted 5,230 5,697 6,168 6,279 6,211 6,547 5,161 Cost burden >50% of income 11,158 12,223 12,141 13,330 13,188 16,433 17,140 Cost burden 30–50% of income 15,481 15,115 15,862 16,923 17,856 19,403 21,153 Severely inadequate housing 2,022 1,797 2,056 2,108 1,971 2,023 1,805 Moderately inadequate housing 4,348 5,191 4,821 4,504 4,311 4,177 3,954 Crowded housing 2,554 2,807 2,570 2,631 2,559 2,621 2,529 Renter households (1,000) 34,150 34,000 34,007 33,727 33,614 33,951 35,054 Unassisted with severe problems 5,777 6,024 5,591 5,758 5,887 6,860 6,993 Unassisted with non-severe problems only 7,651 7,451 7,560 7,283 7,557 7,303 8,445 Unassisted with no 15,492 14,827 14,657 14,407 13,958 13,240 14,455 problems Assisted 5,230 5,697 6,203 6,279 6,211 6,547 5,161 Rent burden >50% of income 6,236 6,686 6,301 6,412 6,477 7,891 7,793 Rent burden 30–50% of income 7,424 6,778 7,141 6,916 7,468 7,502 8,340 Severely inadequate housing 849 1,072 1,183 1,168 1,038 1,100 1,073 Moderately inadequate housing 2,277 3,021 2,768 2,508 2,525 2,542 2,400 Crowded housing 1,673 1,891 1,666 1,658 1,615 1,635 1,511 Owner households (1,000) 63,544 65,487 68,795 71,708 72,254 74,950 75,665 Unassisted with severe problems 5,967 6,182 6,604 7,736 7,511 9,282 9,951 Unassisted with non-severe problems only 10,042 10,449 10,684 11,934 12,233 13,546 14,307 Unassisted with no 47,531 48,855 51,507 52,038 52,510 52,122 51,407 problems Assisted — — — — — — — Cost burden >50% of income 4,922 5,537 5,841 6,918 6,711 8,542 9,347 Cost burden 30–50% of income 8,057 8,337 8,716 10,007 10,388 11,901 12,813 Severely inadequate housing 1,173 725 867 940 933 923 732 Moderately inadequate housing 2,071 2,170 2,064 1,996 1,786 1,635 1,554 Crowded housing 881 916 894 973 944 986 1,018 Worst Case Housing Needs 57
Table A–2b. Housing Conditions of Renters and Owners, 1995–2007—Percentage of Households 1995 1997 1999 2001 2003 2005 2007 Total households 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Unassisted with severe problems 12.0% 12.3% 11.9% 12.8% 12.7% 14.8% 15.3% Unassisted with non- severe problems only 18.1% 18.0% 17.7% 18.2% 18.7% 19.1% 20.5% Unassisted with no problems 64.5% 64.0% 64.4% 63.0% 62.8% 60.0% 59.5% Assisted 5.4% 5.7% 6.0% 6.0% 5.9% 6.0% 4.7% Cost burden >50% of income 11.4% 12.3% 11.8% 12.6% 12.5% 15.1% 15.5% Cost burden 30–50% of income 15.8% 15.2% 15.4% 16.1% 16.9% 17.8% 19.1% Severely inadequate housing 2.1% 1.8% 2.0% 2.0% 1.9% 1.9% 1.6% Moderately inadequate housing 4.5% 5.2% 4.7% 4.3% 4.1% 3.8% 3.6% Crowded housing 2.6% 2.8% 2.5% 2.5% 2.4% 2.4% 2.3% Renter households 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Unassisted with severe problems 16.9% 17.7% 16.4% 17.1% 17.5% 20.2% 19.9% Unassisted with non- severe problems only 22.4% 21.9% 22.2% 21.6% 22.5% 21.5% 24.1% Unassisted with no problems 45.4% 43.6% 43.1% 42.7% 41.5% 39.0% 41.2% Assisted 15.3% 16.8% 18.2% 18.6% 18.5% 19.3% 14.7% Rent burden >50% of income 18.3% 19.7% 18.5% 19.0% 19.3% 23.2% 22.2% Rent burden 30–50% of income 21.7% 19.9% 21.0% 20.5% 22.2% 22.1% 23.8% Severely inadequate housing 2.5% 3.2% 3.5% 3.5% 3.1% 3.2% 3.1% Moderately inadequate housing 6.7% 8.9% 8.1% 7.4% 7.5% 7.5% 6.8% Crowded housing 4.9% 5.6% 4.9% 4.9% 4.8% 4.8% 4.3% Owner households 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Unassisted with severe problems 9.4% 9.4% 9.6% 10.8% 10.4% 12.4% 13.2% Unassisted with non- severe problems only 15.8% 16.0% 15.5% 16.6% 16.9% 18.1% 18.9% Unassisted with no problems 74.8% 74.6% 74.9% 72.6% 72.7% 69.5% 67.9% Assisted — — — — — — — Cost burden >50% of income 7.7% 8.5% 8.5% 9.6% 9.3% 11.4% 12.4% Cost burden 30–50% of income 12.7% 12.7% 12.7% 14.0% 14.4% 15.9% 16.9% Severely inadequate housing 1.8% 1.1% 1.3% 1.3% 1.3% 1.2% 1.0% Moderately inadequate housing 3.3% 3.3% 3.0% 2.8% 2.5% 2.2% 2.1% Crowded housing 1.4% 1.4% 1.3% 1.4% 1.3% 1.3% 1.3% Worst Case Housing Needs 58
Table A–3. Housing Conditions of Unassisted Renter Households by Relative Income, 2005 and 2007 Household Income as Percentage of HUD-Adjusted Area Median Family Income 2007 0–30% 30–50% 50–80% 80–120% >120% All Incomes Total unassisted households (1,000) 5,893 5,681 7,182 5,352 5,785 29,893 Any with Severe Problems 4,327 1,578 681 245 162 6,993 Rent burden >50% of income 4,246 1,472 470 96 42 6,326 [And rent > fair market rent] 1,458 1,043 463 96 42 3,102 Severely inadequate housing 268 161 215 149 120 913 Any with non-severe problems only 828 2,935 2,960 1,048 674 8,445 Rent burden 30–50% of income 701 2,693 2,440 661 283 6,777 Moderately inadequate housing 191 330 457 305 302 1,586 Crowded housing 132 326 286 96 124 964 Any with no problems 738 1,168 3,541 4,059 4,949 14,455 Household Income as Percentage of HUD-Adjusted Area Median Family Income 2005 0–30% 30–50% 50–80% 80–120% >120% All Incomes Total unassisted households (1,000) 6,440 5,081 6,496 4,894 4,492 27,404 Any with s evere problems 4,643 1,349 472 264 132 6,860 Rent burden >50% of income 4,545 1,184 295 134 56 6,214 [And rent > fair market rent] 1,454 829 290 134 56 2,763 Severely inadequate housing 318 203 182 129 76 908 Any with non-severe problems only 981 2,604 2,363 799 557 7,303 Rent burden 30–50% of income 833 2,375 1,836 397 216 5,657 Moderately inadequate housing 203 339 421 273 273 1,509 Crowded housing 153 271 285 158 97 963 Any with no problems 816 1,127 3,661 3,832 3,804 13,240 Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 59
Table A–4. Incidence of Housing Problems Among Renters by Relative Income, 2005 and 2007—Number and Percentage (1 of 3) 2005 2007 2005 2007 Renter households (1,000) 33,951 35,054 100.0% 100.0% Unassisted with severe problems 6,860 6,993 20.2% 19.9% Unassisted with non-severe problems only 7,303 8,445 21.5% 24.1% Unassisted with no problems 13,240 14,455 39.0% 41.2% Assisted 6,547 5,161 19.3% 14.7% Any with severe problems 8,665 8,564 25.5% 24.4% Rent burden >50% of income 7,891 7,793 23.2% 22.2% Severely inadequate housing 1,100 1,073 3.2% 3.1% Rent burden only a 6,883 6,918 20.3% 19.7% Any with non–severe problems only 9,286 10,015 27.4% 28.6% Rent burden 30–50% of income 7,238 8,099 21.3% 23.1% Moderately inadequate housing 1,859 1,828 5.5% 5.2% Crowded housing 1,200 1,096 3.5% 3.1% Rent burden only 6,363 7,234 18.7% 20.6% Any with no problems 16,000 16,476 47.1% 47.0% Income 0–30% HAMFI (1,000) 9,729 9,243 100.0% 100.0% Unassisted with severe problems 4,643 4,327 47.7% 46.8% Unassisted with non-severe problems only 981 828 10.1% 9.0% Unassisted with no problems 816 738 8.4% 8.0% Assisted 3,289 3,350 33.8% 36.2% Any with severe problems 6,151 5,732 63.2% 62.0% Rent burden >50% of income 5,999 5,588 61.7% 60.5% Severely inadequate housing 430 372 4.4% 4.0% Rent burden only a 5,160 4,892 53.0% 52.9% Any with non-severe problems only 1,804 1,778 18.5% 19.2% Rent burden 30–50% of income 1,535 1,536 15.8% 16.6% Moderately inadequate housing 340 324 3.5% 3.5% Crowded housing 229 196 2.4% 2.1% Rent burden only 1,263 1,302 13.0% 14.1% Any with no problems 1,775 1,733 18.2% 18.7% Worst Case Housing Needs 60
Table A–4. Incidence of Housing Problems Among Renters by Relative Income, 2005 and 2007—Number and Percentage (2 of 3) 2005 2007 2005 2007 Income 30–50% HAMFI (1,000) 6,342 6,697 100.0% 100.0% Unassisted with severe problems 1,349 1,578 21.3% 23.6% Unassisted with non-severe problems only 2,604 2,935 41.0% 43.8% Unassisted with no problems 1,127 1,168 17.8% 17.4% Assisted 1,262 1,016 19.9% 15.2% Any with severe problems 1,548 1,711 24.4% 25.5% Rent burden >50% of income 1,361 1,579 21.5% 23.6% Severely inadequate housing 228 198 3.6% 3.0% Rent burden only a 1,233 1,429 19.4% 21.3% Any with non-severe problems only 3,253 3,375 51.3% 50.4% Rent burden 30–50% of income 2,934 3,062 46.3% 45.7% Moderately inadequate housing 417 393 6.6% 5.9% Crowded housing 346 369 5.5% 5.5% Rent burden only a 2,535 2,655 40.0% 39.6% Any with no problems 1,542 1,611 24.3% 24.1% Income 50–80% HAMFI (1,000) 7,488 7,650 100.0% 100.0% Unassisted with severe problems 472 681 6.3% 8.9% Unassisted with non-severe problems only 2,363 2,960 31.6% 38.7% Unassisted with no problems 3,661 3,541 48.9% 46.3% Assisted 992 468 13.2% 6.1% Any with severe problems 532 703 7.1% 9.2% Rent burden >50% of income 321 486 4.3% 6.4% Severely inadequate housing 215 224 2.9% 2.9% Rent burden only a 289 465 3.9% 6.1% Any with non-severe problems only 2,727 3,095 36.4% 40.5% Rent burden 30–50% of income 2,091 2,524 27.9% 33.0% Moderately inadequate housing 488 494 6.5% 6.5% Crowded housing 347 305 4.6% 4.0% Rent burden only 1,930 2,327 25.8% 30.4% Any with no pr oblems 4,229 3,853 56.5% 50.4% Worst Case Housing Needs 61
Table A–4. Incidence of Housing Problems Among Renters by Relative Income, 2005 and 2007—Number and Percentage (3 of 3) 2005 2007 2005 2007 Income 80–120% HAMFI (1,000) 5,449 5,518 100.0% 100.0% Unassisted with severe problems 264 245 4.8% 4.4% Unassisted with non-severe problems only 799 1,048 14.7% 19.0% Unassisted with no problems 3,832 4,059 70.3% 73.6% Assisted 555 166 10.2% 3.0% Any with severe problems 290 253 5.3% 4.6% Rent burden >50% of income 147 98 2.7% 1.8% Severely inadequate housing 142 156 2.6% 2.8% Rent burden only a 143 95 2.6% 1.7% Any with non-severe problems only 909 1,077 16.7% 19.5% Rent burden 30–50% of income 452 684 8.3% 12.4% Moderately inadequate housing 325 308 6.0% 5.6% Crowded housing 170 99 3.1% 1.8% Rent burden only 428 679 7.9% 12.3% Any with no problems 4,249 4,187 78.0% 75.9% Income >120% HAMFI (1,000) 4,943 5,946 100.0% 100.0% Unassisted with severe problems 132 162 2.7% 2.7% Unassisted with non-severe problems only 557 674 11.3% 11.3% Unassisted with no problems 3,804 4,949 77.0% 83.2% Assisted 450 161 9.1% 2.7% Any with severe problems 145 165 2.9% 2.8% Rent burden >50% of income 61 42 1.2% 0.7% Severely inadequate housing 84 123 1.7% 2.1% Rent burden only a 57 38 1.2% 0.6% Any with non-severe problems only 593 690 12.0% 11.6% Rent burden 30–50% of income 226 293 4.6% 4.9% Moderately inadequate housing 289 307 5.8% 5.2% Crowded housing 109 127 2.2% 2.1% Rent burden only 207 272 4.2% 4.6% Any with no problems 4,204 5,090 85.0% 85.6% HAMFI = HUD-adjusted area median family income. a The estimates for “rent burden only” exclude households with any non-severe problem. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 62
Table A–5. Incidence of Housing Problems Among Very Low-Income Renters by Household Type, 2005 and 2007—Number and Percentage (1 of 2) Number Percentage 2005 2007 2005 2007 Elderly (1,000) 3,587 3,492 100.0% 100.0% Unassisted with severe problems 1,291 1,209 36.0% 34.6% Unassisted with non-severe problems only 528 505 14.7% 14.5% Unassisted with no problems 409 369 11.4% 10.6% Assisted 1,358 1,409 37.9% 40.3% Any with severe problems 1,722 1,607 48.0% 46.0% Rent burden >50% of income 1,664 1,572 46.4% 45.0% Severely inadequate housing 110 101 3.1% 2.9% Rent burden only a 1,484 1,397 41.4% 40.0% Any with non-severe problems only 937 947 26.1% 27.1% Rent burden 30–50% of income 888 892 24.8% 25.5% Moderately inadequate housing 97 118 2.7% 3.4% Crowded housing 9 3 0.3% 0.1% Rent burden only a 837 828 23.3% 23.7% Any with no problems 928 938 25.9% 26.9% Families with children (1,000) 6,465 6,329 100.0% 100.0% Unassisted with severe problems 2,324 2,187 35.9% 34.6% Unassisted with non-severe problems only 1,674 1,767 25.9% 27.9% Unassisted with no problems 676 624 10.5% 9.9% Assisted 1,791 1,751 27.7% 27.7% Any with severe problems 3,012 2,909 46.6% 46.0% Rent burden >50% of income 2,904 2,788 44.9% 44.1% Severely inadequate housing 238 234 3.7% 3.7% Rent burden only a 2,533 2,491 39.2% 39.4% Any with non-severe problems only 2,301 2,311 35.6% 36.5% Rent burden 30–50% of income 1,948 1,964 30.1% 31.0% Moderately inadequate housing 311 304 4.8% 4.8% Crowded housing 548 530 8.5% 8.4% Rent burden only a 1,501 1,542 23.2% 24.4% Any with no problems 1,153 1,108 17.8% 17.5% Worst Case Housing Needs 63
Table A–5. Incidence of Housing Problems Among Very Low-Income Renters by Household Type, 2005 and 2007—Number and Percentage (2 of 2) Number Percentage 2005 2007 2005 2007 Non-elderly disabled (1,000) b 1,767 1,707 100.0% 100.0% Unassisted with severe problems 694 602 39.3% 35.3% Unassisted with non-severe problems onl y 267 218 15.1% 12.8% Unassisted with no problems 102 97 5.8% 5.7% Assisted 703 790 39.8% 46.3% Any with severe problems 989 882 56.0% 51.7% Rent burden >50% of income 946 845 53.5% 49.5% Severely inadequate housing 93 78 5.3% 4.6% Rent burden only a 780 703 44.1% 41.2% Any with non-severe problems only 443 492 25.1% 28.8% Rent burden 30–50% of income 407 477 23.0% 27.9% Moderately inadequate housing 78 57 4.4% 3.3% Crowded housing 0 3 0.0% 0.2% Rent burden only a 366 432 20.7% 25.3% Any with no problems 334 332 18.9% 19.4% Other households (1,000) 4,253 4,413 100.0% 100.0% Unassisted with severe problems 1,683 1,907 39.6% 43.2% Unassisted with non-severe problems only 1,116 1,273 26.2% 28.8% Unassisted with no problems 757 816 17.8% 18.5% Assisted 698 416 16.4% 9.4% Any with severe problems 1,975 2,043 46.4% 46.3% Rent burden >50% of income 1,846 1,962 43.4% 44.5% Severely inadequate housing 217 157 5.1% 3.6% Rent burden only a 1,597 1,729 37.5% 39.2% Any with non-severe problems only 1,376 1,403 32.4% 31.8% Rent burden 30–50% of income 1,226 1,265 28.8% 28.7% Moderately inadequate housing 271 239 6.4% 5.4% Crowded housing 18 30 0.4% 0.7% Rent burden only a 1,094 1,154 25.7% 26.2% Any with no problems 902 966 21.2% 21.9% a The estimates for “rent burden only” exclude households with any non-severe problem. b Non-elderly disabled represents childless households with adults identified as disabled, using a proxy based on source of income. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 64
Table A–6a. Housing Problems and Characteristics of Very Low-Income Renters by Household Type, 2007 (1 of 2) Total Elderly, No Children Families With Children Other Families Nonelderly Disabled b Other Nonfamily Renter households (1,000) 15,940 3,492 6,329 1,033 1,707 3,380 Number of children 12,801 0 12,801 0 0 0 Number of persons 36,521 4,299 23,250 2,451 2,279 4,242 Children/household 0.80 0.00 2.02 0.00 0.00 0.00 Persons/household 2.29 1.23 3.67 2.37 1.34 1.26 Unassisted with severe problems 5,905 1,209 2,187 352 602 1,555 Unassisted with non-severe problems only 3,763 505 1,767 324 218 949 Unassisted with no problems 1,906 369 624 234 97 582 Assisted 4,366 1,409 1,751 123 790 293 Any with severe problems 7,442 1,607 2,909 379 882 1,664 Rent burden >50% of income 7,167 1,572 2,788 364 845 1,598 Severely inadequate housing 570 101 234 29 78 128 Rent burden only a 6,320 1,397 2,491 313 703 1,416 Any with non-severe problems only 5,153 947 2,311 363 492 1,040 Rent burden 30–50% of income 4,598 892 1,964 335 477 930 Moderately inadequate housing 718 118 304 51 57 188 Crowded housing 565 3 530 11 3 19 Rent burden only 3,957 828 1,542 302 432 852 Any with no problems 3,345 938 1,108 291 332 675 Worst Case Housing Needs 65
Table A–6a. Housing Problems and Characteristics of Very Low-Income Renters by Household Type, 2007 (2 of 2) Total Elderly, No Children Families With Children Other Families Nonelderly Disabled b Other Nonfamily Other Characteristics One person in household 6,863 2,771 132 0 1,267 2,693 Husband-wife family 3,045 436 1,928 511 169 0 Female head 9,894 2,373 4,437 548 951 1,585 Minority head 8,464 1,287 4,282 612 790 1,493 AFDC/SSI income 3,019 685 1,370 3 961 0 Social security income 3,941 2,887 484 33 538 0 Income below 50% poverty 3,649 600 1,658 163 349 878 Income below poverty 8,298 1,626 3,709 386 1,176 1,402 Income below 150% of poverty 12,600 2,634 5,547 676 1,516 2,226 High school graduate 10,880 1,986 4,077 730 1,175 2,911 Two+ years post high school 2,392 381 669 182 245 916 Earnings at minimum wage: At least half time 7,815 296 4,136 820 186 2,377 At least full time 6,474 189 3,590 706 104 1,885 Earnings main source of i ncome 8,348 257 4,371 850 133 2,737 Housing rated poor 973 86 471 66 152 197 Housing rated good+ 11,622 2,904 4,388 734 1,217 2,380 Neighborhood rated poor 1,280 123 676 76 151 254 Neighborhood rated good+ 11,233 2,838 4,153 721 1,182 2,338 Central cities 7,420 1,471 2,830 510 805 1,805 Suburbs 3,281 835 1,283 180 392 591 Non-metropolitan areas 5,239 1,186 2,216 343 509 984 Northeast 3,587 841 1,243 175 423 905 Midwest 3,487 1,021 1,226 220 472 549 South 5,192 910 2,332 376 520 1,054 West 3,673 721 1,528 262 291 872 a The estimates for “rent burden only” exclude households with any non-severe problem. b Non-elderly disabled represents childless households with adults identified as disabled using a proxy based on source of income. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 66
Table A–6b. Housing Problems and Characteristics of Extremely Low-Income Renters by Household Type, 2007 (1 of 2) Total Elderly, No Children Families With Children Other Families Nonelderly Disabled b Other Nonfamily Renter households (1,000) 9,243 2,336 3,452 393 1,320 1,743 Number of children 7,160 0 7,160 0 0 0 Number of persons 19,797 2,717 12,396 935 1,612 2,136 Children/household 0.77 0.00 2.07 0.00 0.00 0.00 Persons/household 2.14 1.16 3.59 2.38 1.22 1.23 Unassisted with severe problems 4,327 901 1,602 203 518 1,103 Unassisted with non-severe problems only 828 175 381 48 51 174 Unassisted with no problems 738 154 188 71 54 271 Assisted 3,350 1,105 1,282 71 696 195 Any with severe problems 5,732 1,272 2,259 225 787 1,188 Rent burden >50% of income 5,588 1,257 2,195 222 759 1,155 Severely inadequate housing 372 68 145 15 59 86 Rent burden only a 4,892 1,125 1,948 186 627 1,006 Any with non-severe problems only 1,778 501 703 66 294 213 Rent burden 30–50% of income 1,536 466 556 57 289 169 Moderately inadequate housing 324 76 150 11 20 69 Crowded housing 196 3 176 6 0 12 Rent burden only a 1,302 426 408 49 275 144 Any with no problems 1,733 562 490 102 238 341 Worst Case Housing Needs 67
Table A–6b. Housing Problems and Characteristics of Extremely Low-Income Renters by Household Type, 2007 (2 of 2) Total Elderly, No Children Families With Children Other Families Nonelderly Disabled b Other Nonfamily Other Characteristics One person in household 4,605 1,985 110 0 1,087 1,422 Husband-wife family 1,222 213 762 158 88 0 Female head 6,177 1,684 2,632 233 744 884 Minority head 4,963 975 2,384 226 634 745 Afdc/ssi income 2,420 562 1,060 0 798 0 Social security income 2,572 1,862 316 20 374 0 Income below 50% poverty 3,649 600 1,658 163 349 878 Income below poverty 7,717 1,618 3,205 346 1,162 1,386 Income below 150% of poverty 9,129 2,283 3,418 389 1,317 1,721 High school graduate 5,985 1,234 2,093 282 880 1,497 Two+ years post high school 1,248 233 274 87 170 484 Earnings at minimum wage: At least half time 2,621 62 1,495 226 29 809 At least full time 1,555 22 1,032 130 5 365 Earnings main source of income 3,391 88 1,803 270 41 1,190 Housing rated poor 583 59 300 22 110 93 Housing rated good+ 6,677 1,939 2,291 290 934 1,223 Neighborhood rated poor 803 92 436 31 117 127 Neighborhood rated good+ 6,395 1,868 2,150 290 892 1,196 Central cities 4,508 1,068 1,662 195 639 944 Suburbs 1,925 502 718 66 307 332 Non-metropolitan areas 2,810 766 1,072 132 373 468 Northeast 2,152 515 696 66 344 532 Midwest 2,229 740 722 86 382 298 South 2,883 610 1,260 146 382 485 West 1,979 471 774 95 211 429 a The estimates for “rent burden only” exclude households with any non-severe problem. b Non-elderly disabled represents childless households with adults identified as disabled, using a proxy based on source of income. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 68
Table A–7. Housing Problems and Characteristics of Worst Case Renters by Household Type, 2007 (1 of 2) Total Elderly, No Children Families With Children Other Families Nonelderly Disabled b Other Nonfamily Renter households (1,000) 5,905 1,209 2,187 352 602 1,555 Number of children 4,336 0 4,336 0 0 0 Number of persons 12,966 1,477 7,857 833 793 2,006 Children/household 0.73 0.00 1.98 0.00 0.00 0.00 Persons/household 2.20 1.22 3.59 2.37 1.32 1.29 Unassisted with severe problems 5,905 1,209 2,187 352 602 1,555 Unassisted with non-severe problems only — — — — — — Unassisted with no problems — — — — — — Assisted — — — — — — Any with severe problems 5,905 1,209 2,187 352 602 1,555 Rent burden >50% of income 5,717 1,190 2,123 336 573 1,495 Severely inadequate housing 429 75 149 27 59 119 Rent burden only a 5,038 1,043 1,903 290 472 1,329 Any with non-severe problems only — — — — — — Rent burden 30–50% of income — — — — — — Moderately inadequate housing — — — — — — Crowded housing — — — — — — Rent burden only a — — — — — — Any with no problems — — — — — — Worst Case Housing Needs 69
Table A–7. Housing Problems and Characteristics of Worst Case Renters by Household Type, 2007 (2 of 2) Total Elderly, No Children Families With Children Other Families Nonelderly Disabled b Other Nonfamily Other Characteristics One person in household 2,685 962 70 0 444 1,208 Husband-wife family 995 134 612 179 71 0 Female head 3,587 815 1,539 191 273 768 Minority head 2,986 413 1,439 197 277 660 AFDC/SSI income 917 131 461 0 325 0 Social security income 1,408 1,025 209 14 160 0 Income below 50% poverty 1,768 301 738 76 166 488 Income below poverty 3,754 637 1,592 193 471 861 Income below 150% of poverty 5,226 978 2,088 287 565 1,308 High school graduate 4,184 756 1,407 252 417 1,352 Two+ years post high school 1,085 152 269 82 103 478 Earnings at minimum wage: At least half time 2,627 74 1,242 281 45 984 At least full time 1,907 44 1,000 203 16 644 Earnings main source of income 3,138 77 1,419 305 33 1,304 Housing rated poor 376 25 176 15 65 95 Housing rated good+ 4,279 999 1,498 256 422 1,104 Neighborhood rated poor 421 32 199 28 52 110 Neighborhood rated good+ 4,193 994 1,479 240 414 1,066 Central cities 2,800 533 965 172 267 865 Suburbs 1,053 225 394 50 144 240 Non-metropolitan areas 2,052 451 828 130 191 450 Northeast 1,230 273 382 48 142 386 Midwest 1,263 338 410 79 183 254 South 1,942 333 819 136 174 480 West 1,470 266 576 89 103 435 a The estimates for “rent burden only” exclude households with any non-severe problem. b Non-elderly disabled represents childless households with adults identified as disabled, using a proxy based on source of income. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 70
Table A–8. Housing Problems and Characteristics of Extremely Low-Income Worst Case Renters by Household Type, 2007 (1 of 2) Total Elderly, No Children Families With Children Other Families Nonelderly Disabled b Other Nonfamily Renter households (1,000) 4,327 901 1,602 203 518 1,103 Number of children 3,232 0 3,232 0 0 0 Number of persons 9,335 1,068 5,746 474 656 1,391 Children/household 0.75 0.00 2.02 0.00 0.00 0.00 Persons/household 2.16 1.19 3.59 2.34 1.27 1.26 Unassisted with severe problems 4,327 901 1,602 203 518 1, 103 Unassisted with non-severe problems only — — — — — — Unassisted with no problems — — — — — — Assisted — — — — — — Any with severe problems 4,327 901 1,602 203 518 1,103 Rent burden >50% of income 4,246 895 1,577 200 499 1,076 Severely inadequate housing 268 49 84 12 42 80 Rent burden only a 3,703 789 1,402 169 405 936 Any with non-severe problems only — — — — — — Rent burden 30–50% of income — — — — — — Moderately inadequate housing — — — — — — Crowded housing — — — — — — Rent burden only — — — — — — Any with no problems — — — — — — Worst Case Housing Needs 71
Table A–8. Housing Problems and Characteristics of Extremely Low-Income Worst Case Renters by Household Type, 2007 (2 of 2) Total Elderly, No Children Families With Children Other Families Nonelderly Disabled b Other Nonfamily Other Characteristics One person in household 2,095 752 62 0 406 875 Husband-wife family 618 83 394 96 46 0 Female head 2,689 619 1,163 120 236 551 Minority head 2,219 343 1,076 112 243 445 AFDC/SSI income 807 112 404 0 291 0 Social security income 1,059 756 167 8 128 0 Income below 50% poverty 1,768 301 738 76 166 488 Income below poverty 3,617 635 1,483 177 468 854 Income below 150% of poverty 4,273 880 1,591 203 515 1,084 High school graduate 2,977 542 984 145 352 955 Two+ years post high school 712 99 166 48 82 317 Earnings at minimum wage: At least half time 1,494 24 756 142 8 565 At least full time 861 7 538 72 0 243 Earnings main source of income 2,066 39 945 174 16 891 Housing rated poor 266 17 133 6 57 53 Housing rated good+ 3,111 734 1,073 158 358 788 Neighborhood rated poor 316 30 167 14 45 60 Neighborhood rated good+ 3,039 723 1,049 144 354 769 In central cities 2,126 446 745 92 228 616 Suburbs 787 138 313 28 128 179 Non-metropolitan areas 1,414 317 544 84 162 309 Northeast 1,013 201 321 34 126 330 Midwest 949 274 292 44 162 177 South 1,382 247 593 82 149 310 West 984 179 396 43 80 286 a The estimates for “rent burden only” exclude households with any non-severe problem. b Non-elderly disabled represents childless households with adults identified as disabled, using a proxy based on source of income. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 72
Table A–9. Incidence of Housing Problems Among Very Low-Income Renters by Race and Ethnicity, 2005 and 2007—Number and Percentage (1 of 2) Number Percentage 2005 2007 2005 2007 Non-Hispanic White (1,000) 7,907 7,477 100.0% 100.0% Unassisted with severe problems 3,098 2,919 39.2% 39.0% Unassisted with non-severe problems only 1,700 1,696 21.5% 22.7% Unassisted with no problems 1,175 1,099 14.9% 14.7% Assisted 1,934 1,763 24.5% 23.6% Any with severe problems 3,764 3,469 47.6% 46.4% Rent burden >50% of income 3,596 3,374 45.5% 45.1% Severely inadequate housing 306 197 3.9% 2.6% Rent burden only a 3,187 3,007 40.3% 40.2% Any with non-severe problems only 2,300 2,271 29.1% 30.4% Rent burden 30–50% of income 2,098 2,101 26.5% 28.1% Moderately inadequate housing 310 289 3.9% 3.9% Crowded housing 97 87 1.2% 1.2% Rent burden only 1,904 1,917 24.1% 25.6% Any with no problems 1,843 1,737 23.3% 23.2% Non-Hispanic Black (1,000) 3,989 4,040 100.0% 100.0% Unassisted with severe problems 1,336 1,345 33.5% 33.3% Unassisted with non-severe problems only 746 752 18.7% 18.6% Unassisted with no problems 388 362 9.7% 9.0% Assisted 1,519 1,581 38.1% 39.1% Any with severe problems 1,969 1,960 49.4% 48.5% Rent burden >50% of income 1,880 1,853 47.1% 45.9% Severely inadequate housing 181 177 4.5% 4.4% Rent burden only a 1,562 1,644 39.2% 40.7% Any with non-severe problems only 1,176 1,253 29.5% 31.0% Rent burden 30–50% of income 1,046 1,082 26.2% 26.8% Moderately inadequate housing 208 203 5.2% 5.0% Crowded housing 94 114 2.4% 2.8% Rent burden only 886 945 22.2% 23.4% Any with no problems 843 827 21.1% 20.5% Worst Case Housing Needs 73
Table A–9. Incidence of Housing Problems Among Very Low-Income Renters by Race and Ethnicity, 2005 and 2007—Number and Percentage (2 of 2) Number Percentage 2005 2007 2005 2007 Hispanic (1,000) 3,167 3,297 100.0% 100.0% Unassisted with severe problems 1,168 1,234 36.9% 37.4% Unassisted with non-severe problems only 928 1,049 29.3% 31.8% Unassisted with no problems 267 301 8.4% 9.1% Assisted 805 713 25.4% 21.6% Any with severe problems 1,496 1,504 47.2% 45.6% Rent burden >50% of income 1,423 1,446 44.9% 43.9% Severely inadequate housing 151 145 4.8% 4.4% Rent burden only a 1,234 1,249 39.0% 37.9% Any with non-severe problems only 1,256 1,279 39.7% 38.8% Rent burden 30–50% of income 1,050 1,130 33.2% 34.3% Moderately inadequate housing 169 152 5.3% 4.6% Crowded housing 328 313 10.4% 9.5% Rent burden only 793 857 25.0% 26.0% Any with no problems 415 513 13.1% 15.6% a The estimates for “rent burden only” exclude households with any non-severe problem. Source: Department of Housing and Urban Development, Office of Policy Development and Research, tabulations of American Housing Survey data Worst Case Housing Needs 74
Table A–10. Incidence of Housing Problems Among Very Low-Income Renters by Region, 2005 and 2007—Number and Percentage (1 of 2) Number Percentage 2005 2007 2005 2007 Northeast (1,000) 3,538 3,487 100.0% 100.0% Unassisted with severe problems 1,354 1,263 38.3% 36.2% Unassisted with non-severe problems only 600 624 17.0% 17.9% Unassisted with no problems 369 421 10.4% 12.1% Assisted 1,215 1,179 34.3% 33.8% Any with severe problems 1,829 1,692 51.7% 48.5% Rent burden >50% of income 1,732 1,620 49.0% 46.5% Severely inadequate housing 227 166 6.4% 4.8% Rent burden only a 1,457 1,412 41.2% 40.5% Any with non-severe problems only 964 1,003 27.2% 28.8% Rent burden 30–50% of income 888 893 25.1% 25.6% Moderately inadequate housing 106 110 3.0% 3.2% Crowded housing 91 106 2.6% 3.0% Rent burden only 781 793 22.1% 22.7% Any with no problems 745 792 21.1% 22.7% Midwest (1,000) 3,331 3,587 100.0% 100.0% Unassisted with severe problems 1,152 1,230 34.6% 34.3% Unassisted with non-severe problems only 704 857 21.1% 23.9% Unassisted with no problems 507 495 15.2% 13.8% Assisted 968 1,005 29.1% 28.0% Any with severe problems 1,450 1,554 43.5% 43.3% Rent burden >50% of income 1,385 1,494 41.6% 41.7% Severely inadequate housing 109 99 3.3% 2.8% Rent burden only a 1,232 1,340 37.0% 37.4% Any with non-severe problems only 1,025 1,143 30.8% 31.9% Rent burden 30–50% of income 912 1,052 27.4% 29.3% Moderately inadequate housing 135 127 4.1% 3.5% Crowded housing 60 56 1.8% 1.6% Rent burden only 838 968 25.2% 27.0% Any with no problems 856 889 25.7% 24.8% Worst Case Housing Needs 75
Table A–10. Incidence of Housing Problems Among Very Low-Income Renters by Region, 2005 and 2007—Number and Percentage (2 of 2) Number Percentage 2005 2007 2005 2007 South (1,000) 5,444 5,192 100.0% 100.0% Unassisted with severe problems 1,987 1,942 36.5% 37.4% Unassisted with non-severe problems only 1,320 1,331 24.2% 25.6% Unassisted with no problems 728 628 13.4% 12.1% Assisted 1,410 1,291 25.9% 24.9% Any with severe problems 2,570 2,420 47.2% 46.6% Rent burden >50% of income 2,477 2,319 45.5% 44.7% Severely inadequate housing 175 186 3.2% 3.6% Rent burden only a 2,156 2,047 39.6% 39.4% Any with non-severe problems only 1,759 1,746 32.3% 33.6% Rent burden 30&nd