Employment and Poverty Among the Homeless

views updated



There is an undeniable connection between homelessness and poverty. People in poverty live from day to day with little or no safety net for times when unforeseen expenses arise. If a family's resources are very small, expenditures on such necessities as food, shelter, or health care have to be carefully decided and sometimes sacrificed. Should one spend money on food, a visit to the doctor, buying necessary medicines, or paying the rent? In 2005 a full-time job paying minimum wage for forty hours per week provided an income of just $10,712 annually. (The federal poverty guideline for 2005 for one person was $9,570; for two people, $12,830.) Being poor often means that an illness, an accident, or a missed paycheck could be enough to cause homelessness.

Housing costs for such a family may be out of reach, costing from 50%–75% of the family income. According to Ralph da Costa Nunez and Laura M. Caruso in "Are Shelters the Answer to Family Homelessness?" in USA Today Magazine (January 1, 2003), low income and high rent payments often result in substandard housing accommodation, doubled-up living, or living on the street or in a public shelter. The necessity of basic sustenance and medical care usually leaves little money left to meet housing needs. People in poverty have further difficulties finding housing if they have previously defaulted on their rent payments or perhaps their house payments, with the end result of homelessness.


Defining poverty and counting the poor is a difficult task. The official poverty measure used in the United States was developed during the 1960s by Mollie Orshansky of the Social Security Administration. Called the poverty index, this measure was based on the Department of Agriculture's 1955 Household Food Consumption Survey, which had determined that a family of three spent approximately one-third of its income on food. The poverty threshold for a family of three was therefore set at three times the cost of the economy food plan, an amount seen as necessary to cover minimal living expenses. A family whose annual before-tax income was below this poverty threshold was "poor." The government has since revised the poverty threshold regularly to account for inflation and changes in the economy.

The Official Poverty Threshold

Table 3.1 shows the 2004 poverty thresholds for families by size and number of children. These amounts include income before taxes but do not include any capital gains or noncash benefits such as public housing, Medicaid, or food stamps. For example, in 2004 a family of five consisting of a father, mother, two related children under age eighteen and an aunt to those children could jointly earn up to $23,108 and still be considered "poor" by the official poverty measure. If, however, all the adults in the family were employed and their annual incomes were as follows: father, $12,000; mother, $8,000; and aunt, $4,000, then the family would have a joint income of $24,000 which is higher than the 2004 poverty threshold figure for a family of five. In 2004 the poverty thresholds ranged from $9,060 for an elderly person living alone to $36,520 for a family of nine or more members with at least one child. The poverty threshold for a family of four was $19,157 (two adults and two related children under eighteen years of age). The threshold for a typical single-parent family with two children was $15,219, almost 50% more than one person would earn working full-time at the federal minimum wage of $5.15 per hour.

Concerns about the Accuracy of the Official
Poverty Rate

Social scientists have for years debated about the best and most accurate means of establishing a poverty threshold.

Related children under 18 years
Size of family unitNoneOneTwoThreeFourFiveSixSevenEight or more
One person (unrelated individual)
Under 65 years9,827
65 years and over9,060
Two persons
Householder under 65 years12,64913,020
Householder 65 years and over11,41812,971
Three persons14,77615,20515,219
Four persons19,48419,80319,15719,223
Five persons23,49723,83823,10822,54322,199
Six persons27,02527,13326,57326,03725,24124,768
Seven persons31,09631,29030,62130,15429,28528,27127,159
Eight persons34,77835,08634,45433,90133,11532,11931,08230,818
Nine persons or more41,83642,03941,48041,01040,24039,17938,22037,98336,520

The central question that arises in debates about measuring poverty is whether to use an absolute or a relative means of updating the poverty rate on an annual or periodic basis. Once established, an absolute poverty measure is updated to account for price changes (inflation) only. A relative poverty measure is one that is updated based on changes in the median or mean income or compensation of the general population. The relative poverty measure adjusts for changing standards of living. The official poverty measure used by the United States is an absolute measure.

In 2001 the "Conveners of the Working Group on Revising the Poverty Measure"—a group of economists, lawyers, professors, and social academics—wrote An Open Letter on Revising the Official Measure of Poverty to the Director of the Office of Management and Budget. They wrote to express their concerns over the inadequacy of the official poverty level measurement and proposed a set of guidelines for a revised standard. The letter stated that the current system was one that was established in the 1960s and that it had not been meaningfully adjusted in the years since, despite decades of major changes in the social safety net for low-income families.

Three of the items that were specifically listed in the letter as examples of areas not well accounted for in determining the official poverty rate were:

  • Noncash benefits (food stamps, housing assistance, free school lunch programs) that are not included in the calculation of income
  • Out-of-pocket medical expenditures that are not included in the calculation of costs
  • Out-of-pocket child care costs that are not included in the calculation of costs

The letter criticizes many aspects of the methodology used to determine the official poverty threshold, as did a report published by the National Academy of Sciences in 1995. The debate about how best and most accurately to determine who is and who is not poor has gone on for decades and will likely continue. It is, therefore, worth-while when reviewing statistics about poverty to keep in mind that they may be skewed by the methods used in calculating them.


Based on the Official Poverty Rate

An official count of the poor includes all those whose family incomes fall below the poverty threshold figure for that particular size family, as seen in Table 3.1. In 2003, based on those criteria, there were 35.9 million poor in the United States, 12.5% of the population. (See Table 3.2.) Children were poor at a higher rate (17.6%) than were adults aged eighteen to sixty-four (10.8%) or aged sixty-five and older (10.2%).

In 2003 whites had the lowest rate of poverty (10.6%), while African-Americans had the highest rate (24.4%). (See Table 3.2.) People of Hispanic origin also had high poverty rates; more than one in five (22.5%) lived below the poverty threshold. Asian people had a poverty rate slightly below the national rate (11.8%).

In 2003, 7.6 million families, or one in ten of all families, were poor. (See Table 3.2.) Only one in twenty married couple families were below the poverty threshold (5.4%). Male-headed, single-parent families had a rate of poverty of 13.5%, less than half that experienced by female-headed, single-parent families (28%). In other

2002 below poverty2003 below poverty
Family status
In families24,53410.425,68410.8
Related children under 1811,64616.312,34017.2
Related children under 64,29618.54,65419.8
In urelated subfamilies41733.746438.6
Reference person16731.719137.6
Children under 1824135.427141.7
Unrelated individual9,61820.49,71320.4
Racea and Hispanic origin
White alone or in combination24,07410.324,95010.6
White aloneb23,46610.224,27210.5
White alone, not Hispanic15,5678.015,9028.2
Black alone or in combination8,88423.99,10824.3
Black alonec8,60224.18,78124.4
Asian alone or in combination1,24310.01,52711.8
Asian aloned1,16110.11,40111.8
Hispanic origin (of any race)8,55521.89,05122.5
Under 18 years12,13316.712,86617.6
18 to 64 years18,86110.619,44310.8
65 years and older3,57610.43,55210.2
Foreign born5,55816.65,89717.2
Naturalized citizen1,28510.01,30910.0
Not a citizen4,27320.74,58821.7
Inside metropolitan areas27,09611.628,36712.1
Inside central cities13,78416.714,55117.5
Outside central cities13,3118.913,8169.1
Outside metropolitan areas7,47414.27,49514.2
Work experience
All workers (16 years and older)8,9545.98,8205.8
Worked full-time year-round2,6352.62,6362.6
Not full-time year-round6,31812.46,18312.2
Did not work at least one week14,64721.015,44621.5
Type of family
Female householder, no husband present3,61326.53,85628.0
Male householder, no wife present56412.163613.5
aData for American Indians and Alaska Natives, and Asian, Native Hawaiian and other Pacific Islanders are not shown separately.
b The 2003 and 2004 Current Population Survey (CPS) asked respondents to choose one or more races. White alone refers to people who reported white and did not report any other race category. The use of this single-race population does not imply that it is the preferred method of presenting or analyzing data. The Census Bureau uses a variety of approaches.
About 2.6 percent of people reported more than one race in Census 2000.
c Black alone refers to people who reported black and did not report any other race category.
d Asian alone refers to people who reported Asian and did not report any other race category.

words, more than one in four female-headed, single-parent families lived below the poverty threshold in 2003.

Based on Alternate Poverty Measures

The size of the poverty-stricken segment of the U.S. population varies depending on which method of determining the poverty threshold is used to identify the poor population. The U.S. Census Bureau has experimented with ways to measure the numbers of people and families categorized as poor. The experimental methods differ from the official methods in two primary ways. First, they take into account out-of-pocket medical expenses. Second, they make adjustments for different housing costs based on geography, by region and/or metropolitan versus rural residences.

In 2001 the Census Bureau used both official and experimental methods to calculate poverty rates in response to the 1995 National Academy of Sciences recommendations. Figure 3.1 presents poverty rates by age group as calculated by different methods. The official rate of poverty for children under eighteen was higher than the rates produced using experimental methods. Nonetheless, all methods for determining poverty showed a higher rate of poverty for children than for adults aged eighteen to sixty-four years.

The rates for senior citizens showed the most variation between the experimental and official methods for measuring poverty. (See Figure 3.1.) The experimental methods used to study poverty rates make adjustments for medical out-of-pocket expenditures and because the elderly have high out-of-pocket medical expenses, methodologies that adjusted income accordingly resulted in much higher numbers of elderly falling into the ranks of the poor.

Trends in the Poverty Rate

For the most part, the poverty rate is linked to the performance of the U.S. economy. At times when the economy is in recession, the poverty rate increases. In fact, the poverty rate often begins to increase somewhat before a serious economic downturn, and may not begin to decline until some time after a recession ends.

Figure 3.2 demonstrates that during the recession of 1980-82, when many Americans lost their jobs and the economy performed poorly, the poverty rate increased dramatically. In 1979, the year before the recession began, the poverty rate was 11.7%; by 1983 it stood at 15.2%. After the recession of the early 1980s ended, the poverty rate gradually declined. By 1989 it was down to 12.8%, its lowest level since 1979. The United States then endured another recession in 1990-91, and by its end the poverty level was up to 14.2%. While the recession that began in March 2001 ended the following year, the poverty rate and the number of individuals living in poverty continued to rise into 2003.

Poverty Rates by Category


The Census Bureau's 2004 report on poverty data included poverty rates by family structure. (See Table 3.2.) The recession that started in March 2001 was reflected in rising numbers of people living in poverty across all categories of family structure. The rate of poverty for all families in 2000 was 8.7%; by 2002 it had risen to 10.4% and by 2003 to 10.8%. (For 2000 poverty data, see "Table 1. People and Families in Poverty by Selected Characteristics: 2000 and 2001," in Bernadette D. Proctor and Joseph Dalaker, Poverty in the United States: Current Population Reports, P60-219, U.S. Census Bureau, September 2002.) The poverty rate for married couple families in 2000 was 4.7%; by 2002 it had risen to 5.3% and by 2003 to 5.4%. Households headed by single females had a poverty rate of 25.4% of 2000; in 2002, the rate had risen to 26.5% and by 2003, to 28%. The poverty rate of households headed by single males rose from 11.3% in 2000 to 12.1% in 2002 and 13.5% in 2003.


Almost thirteen million children are in poverty. In 2003 a higher proportion of children lived in poverty (17.6%) than any other age group. (See Figure 3.3.) Over one-third (35.9%) of the people living in poverty were children even though they made up only one-quarter (25.4%) of the population (Carmen DeNavas-Walt et al., Income, Poverty, and Health Insurance Coverage in the United States: 2003, U.S. Census Bureau, August 2004).

According to DeNavas-Walt et al., over half (52.9%) of children under six years of age and living in a female-headed, single-parent family lived in poverty. These young children were at particularly high risk of poverty compared to their peers who lived in two-parent households. The poverty rate for children under the age of six living in married-couple households was 9.6%. Overall, children under the age of eighteen who lived in single-parent households were over five times more likely to live in poverty than were children living in married-couple households.

Over the past few decades the child poverty rate has fallen, from a high of 27.3% in 1959 to 17.6% in 2003. (See Figure 3.3.) The decline in poverty for children was not steady but mirrored the fluctuations seen in the poverty rate of adults aged eighteen to sixty-four. The age group that had the most improvement in poverty rates was senior citizens; their rate in 1959 was 35.2% and in 2003 it was 10.2%.


In 2003 Hispanics and African-Americans were much more likely to be poor than whites. Members of these two populations also had higher rates of single female-headed households and lower rates of high school graduation, two indicators of increased likelihood of poverty. (See Figure 3.4.) In 2003 the poverty rate for Hispanics (who may be of any race) was 22.5%; 18% of Hispanic families were headed by a single mother, and 43% of the Hispanic population aged twenty-five and older had not graduated from high school. The poverty rate for African-Americans was 24.3%; 32.4% of black families were headed by a single mother, and one in five African-Americans aged twenty-five and older had not graduated from high school. Non-Hispanic whites and Asian Americans (as reported by DeNavas-Walt et al.) had the lowest rates of poverty by racial group (8.2% and 11.8%, respectively). Only 8.8% of white families were headed by a single mother, and only 10.6% of the white population aged twenty-five and older had not graduated from high school.


The poor are often assumed to be the unemployed. People who are unemployed are more likely to be poor but many people who work can be poor as well. People who worked full-time at any point during 2003 had a 2.6% poverty rate. People who worked part-time at any point during 2003 had a 12.2% poverty rate. Those who did not work at all during 2003 had a 21.5% poverty rate. (See Table 3.2.)


Poverty rates vary from one part of the United States to another. All regions saw an increase in poverty rates between 2002 and 2003. (See Table 3.2.) The South had the highest rate of poverty in 2003, 14.1%, up from a rate of 13.8% in 2002. DeNavas-Walt et al. pointed out that four out of ten (40.6%) poor people lived in the South in 2003, while only 35.9% of Americans lived there. The poverty rate increased in the other regions as well. In the Northeast, the poverty rate increased from 10.9% in 2002 to 11.3% in 2003. In the Midwest, the rate increased from 10.3% in 2002 to 10.7% in 2003. In the West, the rate increased from 12.4% in 2002 to 12.6% in 2003.

Poverty rates also vary by residential area. Inner cities tend to have the highest rates. In 2003 the poverty rate for people living in central cities was 17.5%, up from 16.7% the year before. Rural areas also have a high poverty rate; in 2003 nonmetropolitan areas had a poverty rate of 14.2%, unchanged from the previous year. Suburban poverty rates tend to be the lowest, but even there the rates were rising in the early 2000s; in 2003 the poverty rate in the suburbs was 9.1%, up from 8.9% the year before.


The poverty rate is established by counting the number of families with before-tax income that is lower than the poverty threshold set for a family that size, so employment and wages are an essential part of the determination.

Household Income

Median annual household income in 2003 was $43,318. Half of all American households earned less than this amount and half earned more. (See Table 3.3.) Not surprisingly, median household income varied greatly according to the composition of the household. The median income for married-couple families in 2003 was $62,405, but for female-headed families with no husband present, it was only $29,307, less than half that of a couple-headed family. Race and ethnicity were also factors. In 2003 Asian households had the highest median income ($55,699). The median income for white, non-Hispanic households was $47,777; for people of Hispanic origin it was $32,997; and for blacks it was $29,689. The median income for all groups except for Asians had fallen from the previous year.

Educational Attainment and Income Level

A strong correlation exists between income level and educational attainment. The poor tend to have less education, and a lower proportion of well-educated people are poor. Table 3.4 presents average earnings by level of highest degree. Generally, the higher the degree a person earns, the higher their average income.

When educational attainment is added to the comparison of family income distribution, a clear and predictable pattern emerges. In 2001 families whose householder had

20022003Percentage change
income (dollars)
(2003 less 2002)
Median money income (dollars)Median money income (dollars)
All households111,27843,381112,00043,318− 0.1
Type of household
Family households75,59653,91176,21753,9910.1
Married-couple57,32062,65757,71962,405− 0.4
Female householder, no husband present13,62029,66513,78129,307− 1.2
Male householder, no wife present4,65642,6674,71741,959− 1.7
Nonfamily households35,68225,98835,78325,741− 1.0
Female householder19,66221,39219,64721,313− 0.4
Male householder16,02032,12316,13631,928− 0.6
Racea and Hispanic origin
White alone or in combination92,74045,99493,19645,572f − 0.9
White aloneb91,64546,11991,96245,631f − 1.1
White alone, not Hispanic81,16647,97481,14847,777− 0.4
Black alone or in combination13,77829,84513,96929,689− 0.5
Black alonec13,46529,69113,62929,645− 0.2
Asian alone or in combination4,07953,4834,23555,2623.3
Asian aloned3,91753,8324,04055,6993.5
Hispanic origin (of any race)11,33933,86111,69332,997f − 2.6
Age of householder
Under 65 years88,61950,64488,95150,171f − 0.9
15 to 24 years6,61128,4666,61027,053f − 5.0
25 to 34 years19,05546,36819,15944,779f − 3.4
35 to 44 years24,06954,74723,22255,0440.5
45 to 54 years22,62360,37323,13760,242− 0.2
55 to 64 years16,26048,28416,82449,215f 1.9
65 years and older22,65923,68223,04823,7870.4
Nativity of the householder
Foreign born13,91238,84914,15937,499f − 3.5
Naturalized citizen6,42346,4716,56746,049− 0.9
Not a citizen7,49034,7587,59232,806f − 5.6
Northeast21,22946,91321,01746,742− 0.4
South40,10740,42740,74239,823f − 1.5
Inside metropolitan areas90,07546,29490,61346,060− 0.5
Inside central cities33,54337,70833,71737,174f − 1.4
Outside central cities56,53251,87956,89651,737− 0.3
Outside metropolitan areas21,20335,44821,38735,112− 0.9
Shares of household income quintiles
and gini index
Lowest quintile22,2563.522,4003.4f − 2.9
Second quintile22,2568.822,4008.7− 1.1
Third quintile22,25614.822,40014.8
Fourth quintile22,25623.322,40023.40.4
Highest quintile22,25649.722,40049.80.2
Gini index of income inequality111,2780.462112,0000.4640.4
Earnings of full-time year-round workers
Women41,87630,89541,90830,724− 0.6

the least education earned the least, while the majority of families with a highly educated householder earned the highest incomes. (See Figure 3.5.) Of heads of households with less than a ninth-grade education, one-half earned under $25,000, and 24.5% earned under $15,000 in 2001. On the other hand, more than half the families headed by a person with a doctoral degree (54.4%) earned $100,000 or more.

20022003Percentage change
income (dollars)
(2003 less 2002)
Median money
income (dollars)
Median money
income (dollars)
CharacteristicNumber (thousands)ValueNumber (thousands)ValueEstimate
Per capita income
White alone or in combination235,03624,511236,87524,442− 0.3
White aloneb230,80924,695232,25424,626− 0.3
White alone, not Hispanic194,42126,727194,87726,7740.2
Black alone or in combination37,35015,61937,65115,583− 0.2
Black alonec35,80615,79536,12115,775− 0.1
Asian alone or in combination12,50423,78512,90523,654− 0.5
Asian aloned11,55824,68411,86924,604− 0.3
Hispanic origin (of any race)39,38413,79640,42513,492− 2.2
Note: "—" represents zero or rounds to zero.
a Data for American Indians and Alaska Natives, and Asian, Native Hawaiian and Other Pacific Islanders are not shown separately.
b The 2003 and 2004 CPS asked respondents to choose one or more races. White alone refers to people who reported white and did not report any other race category. The use of this single-race population does not imply that it is the preferred method of presenting or analyzing data. The Census Bureau uses a variety of approaches. About 2.6 percent of people reported more than one race in Census 2000.
c Black alone refers to people who reported black and did not report any other race category.
d Asian alone refers to people who reported Asian and did not report any other race category.
Level of highest degree
CharacteristicTotal personsNot a high school graduateHigh school graduate onlySome college, no degreeAssociate'sBachelor'sMaster'sProfessionalDoctorate
All personsb36,30818,82627,28029,72534,17751,19460,445112,84589,737
25 to 34 years old32,52719,23526,27828,87930,66242,62348,59875,24762,190
35 to 44 years old41,96322,32430,25937,53337,44058,26763,758123,81188,818
45 to 54 years old45,39221,23131,25140,22539,16760,68067,096126,230112,538
55 to 64 years old42,38124,76130,89337,45034,84855,05762,640132,37281,166
65 years old and over36,61118,94927,51929,80934,33151,61261,151114,98191,771
aBase figure too small to meet statistical standards for reliability of a derived figure.
b Includes other races, not shown separately.
c For persons who selected this race group only. The 2003 Current Population Survey (CPS) allowed respondents to choose more than one race. Beginning 2003 data represent persons who selected this race group only and exclude persons reporting more than one race. The CPS in prior years only allowed respondents to report one race group.
d Persons of Hispanic origin may be of any race.

A lack of education leaves a person ill equipped to support him- or herself, or a family, comfortably. Education opens doors and opportunities that are simply not available to the undereducated, especially in an economy that is transitioning from a reliance on manufacturing to a reliance on the information and service industries.

Growth of Jobs in Low Wage Industries

During the 1970s manufacturing industries began closing down plants and moving their production facilities to cheaper labor markets at home and abroad. For much of the twentieth century the United States had been primarily an industrial economy. The last two decades of the century saw the beginning of the nation's transition to what has now been dubbed the information or service economy.

Figure 3.6 shows the change in employment between 1980 and 2001 by industrial sector. The industry that grew the most was the service industry. In 1980 the service sector employed 28.8 million people; by 2001 it had grown to employ 50.5 million, a 75.6% increase. Industries that pay higher wages than those offered in the services, like mining and manufacturing, shrank during this same period, by 42.1% and 13.5% respectively.

The shift in employment from production work to service work has shifted workers from higher to lower paying jobs. According to the U.S. Bureau of Labor Statistics, average hourly wages earned by people employed in the service industries are much lower than average hourly wages earned in the fields of manufacturing, mining, and construction.

Minimum Wage Jobs

The Bureau of Labor Statistics provides data on which industrial sectors employ the highest percentage of people at minimum wage. Table 3.5 shows the total number of wage and salary workers by occupation and by industry. It also shows the number and percent of those workers who, in 2004, earned at or below the federal minimum wage of $5.15 per hour.

Almost one in ten service workers (9.1%) earned at or below the federal minimum wage, and almost one in five food preparation and service workers (19%) earned

Workers paid hourly ratesTotal at or below prevailing federal minimum wage
Occupation and industryTotalBelow prevailing federal minimum wageAt prevailing federal minimum wageNumberPercent of hourly-paid Number workers
Management, professional, and related occupations13,7433927660.7
Management, business, and financial operations occupations3,75014216.4
Management occupations2,12012113.6
Business and financial operations occupations1,630213.2
Professional and related occupations9,993252550.5
Computer and mathematical occupations76212.2
Architecture and engineering occupations952123.3
Life, physical, and social science occupations414112.5
Community and social services occupations717214.5
Legal occupations281
Education, training, and library occupations2,024811191.0
Arts, design, entertainment, sports, and media occupations77786131.7
Healthcare practitioner and technical occupations4,065437.2
Service occupations16,3811,2382561,4949.1
Healthcare support occupations2,4602112321.3
Protective service occupations1,823204251.3
Food preparation and serving related occupations6,2361,0361471,18319.0
Building and grounds cleaning and maintenance occupations3,60863391022.8
Personal care and service occupations2,25497541516.7
Sales and office occupations20,6501041462501.2
Sales and related occupations7,51266761431.9
Office and administrative support occupations13,1393769107.8
Natural resources, construction, and maintenance occupations9,368242650.5
Farming, fishing, and forestry occupations6031011223.6
Construction and extraction occupations5,4307614.3
Installation, maintenance, and repair occupations3,3356915.4
Production, transportation, and material moving occupations13,79678661441.0
Production occupations7,708361855.7
Transportation and material moving occupations6,0874247891.5
Private sector64,7081,4234671,8892.9
Agriculture and related industries623910193.1
Nonagriculture and related industries64,0851,4134571,8702.9
Durable goods6,39720829.4
Nondurable goods3,991181532.8
Wholesale and retail trade12,45685961811.5
Wholesale trade2,03871016.8
Retail trade10,41878861651.6
Transportation and utilities2,91513721.7
Transportation and warehousing2,43112720.8
Publishing, except Internet32823.8
Motion pictures and sound recording20437104.8
Broadcasting, except Internet228
Financial activities3,4531320331.0
Finance and insurance2,42981018.7
Real estate and rental and leasing1,025610151.5
Professional and business services5,4613724611.1
Professional and technical services2,2495914.6
Management, administrative, and waste services3,2123215471.5
Education and health services11,02276621371.2
Educational services1,2631321342.7
Health care and social assistance9,75962411031.1
Workers paid hourly ratesTotal at or below prevailing federal minimum wage
Occupation and industryTotalBelow prevailing federal minimum wageAt prevailing federal minimum wageNumberPercent of hourly-paid Number workers
Leisure and hospitality8,0951,0661761,24215.3
Arts, entertainment, and recreation1,1984520655.4
Accommodation and food services6,8971,0211561,17717.1
Food services and drinking places5,8609861471,13319.3
Other services*2,95971311013.4
Other services, except private households2,4703627632.5
Other services, private households489354397.9
Public sector9,23160531141.2
Federal government1,781811201.1
State government2,3471517321.4
Local government5,1033725621.2
*Includes other industries, not shown separately.
Note: The prevailing federal minimum wage was $5.15 per hour in 2004. Data are for wage and salary workers, excluding the incorporated self-employed. They refer to a person's earnings on their sole or principal job, and pertain only to workers who are paid hourly rates. Salaried workers and other nonhourly workers are not included. The presence of workers with hourly earnings below the minimum wage does not necessarily indicate violations of the Fair Labor Standards Act, as there are exceptions to the minimum wage provisions of the law. In addition, some survey respondents might have rounded hourly earnings to the nearest dollar, and, as a result, reported hourly earnings below the minimum wage even though they earned the minimum wage or higher. Beginning in January 2004, data reflect revised population controls used in the household survey.

minimum wage or below. (See Table 3.5.) Almost 8% of service workers who worked in private households earned minimum wage or less. Almost one in seven workers (15.3%) in the leisure and hospitality industry earned this little. In contrast, less than 1% of managers and professionals earned minimum wage or less.

The shift from a primarily industrial economy to one primarily engaged in providing services has been one of the leading factors in a shift in the distribution of wealth in the United States.


While most discussions of poverty focus simply on people who are below the poverty line versus those who are above it, it is important to keep in mind that even among the poor, some people have fewer resources than others. An analysis of changes in the distribution of wealth helps to explain why homelessness and poverty can remain level, even when the economy performs well.

During the 1980s and 1990s the poverty rate declined for the most part. In the first years of each of these decades the number of people in poverty rose but during the rest of the twenty-year period the numbers of people in poverty declined. (See Figure 3.2.) However, the patterns seen in the distribution of wealth underwent a change. Table 3.6 breaks the U.S. population down into five groups based on income, ranging from the fifth of the population with the lowest incomes to the fifth with the highest incomes. It then displays the percent of total income earned by each fifth of the population in a particular year, as well as the amount earned by those in the top 5% of all Americans (the highest fifth of the highest fifth).

From 1991 to 2001, the highest fifth or quintile of the population increased its income 7.9%, bringing its share of total income from 44.2% to 47.7%. (See Table 3.6.) In each of the other, lower income quintiles the share of aggregate income received declined over the same time period, 1991—2001. The lowest fifth went from 4.5% of all income in 1991 down to 4.2% in 2001. The top 5% of Americans saw their percentage of aggregate income rise 23% during the 1990s, increasing their share from 17.1% to 21% of the nation's income.

As the U.S. economy went through a period of historic growth during the 1990s, every fifth of the population also saw its income rise. (See Figure 3.7.) The richest fifth of the population, however, gained the largest share of the new wealth and increased their overall share of aggregate income while each of the other fifths lost some of its share of the aggregate income. Although it is not accurate to say that the poor got poorer, since their incomes grew, it is true that they became poorer relative to the rich. The 1990s resulted in increased disparity between the rich and the poor.

Lowest fifthSecond fifthMiddle fifthFourth fifthHighest fifthTop 5 percent


One of the risk factors for becoming homeless is the lack of a financial safety net upon which individuals can fall back should something unexpected occur. The loss of a job, and in many cases the concomitant loss of health insurance, is one such occurrence. An accident resulting in the need for expensive medical treatment is another example of the sort of incident that requires a financial safety net. Financial experts stress the importance of maintaining savings that can help see people through this sort of crisis. Yet in the United States, the rate at which Americans save has been declining since the 1980s.

Figure 3.8 provides an overview of the personal savings rate over the period 1970-2004. The Bureau of Economic Analysis defines personal savings as the rate at which we save our disposable income (the income remaining after taxes have been paid). Disposable income grew over this period from $13,563 in 1970 to $27,237 in 2004 (based on constant, inflation-adjusted 2000 dollars). The savings rate in 1970 was 9.4% of disposable income, but by 2004 it had dropped to just 1.2%. As people save less, they become more vulnerable to unplanned expenditures that periodically arise in life.

Exacerbating the problem of low savings is rising consumer credit debt. Kim Khan, in "How Does Your Debt Compare?" reported that not only is debt rising, but 43% of American families spend more than they earn each year and average $8,000 in credit card debt (MSN Money, http://moneycentral.msn.com/content/SavingandDebt/P70581.asp, accessed July 2, 2005). Mortgages also skyrocketed in the early 2000s. Because of historically low interest rates, many homeowners over-borrowed. According to the American Housing Survey for the United States in 2003 (Census Bureau, September 2004), more than 855,000 owners had three or more mortgages on the homes they occupied, and 2.5 million owners had loans equal to or greater than the value of their homes. As a result, bankruptcies and foreclosures were on the rise.

The rise in the number of bankruptcy filings during the late 1990s, a period of strong economic growth, was likely the result of low savings and high credit debt. According to the Federal Reserve, the typical family filing for bankruptcy in 1997 owed more than one and a half times its annual income in short-term, high-interest debt. For example, a family earning $24,000 had an average of $36,000 in credit card and similar debt. Between 1980 and 2002, the number of personal bank-ruptcy filings rose by an astonishing 435%, from 287,580 in 1980 to 1,539,111 in 2002.

Controversial new bankruptcy regulations were signed into law by President George W. Bush in April 2005, making it more difficult for many to file personal bankruptcy. The new law addresses the uninformed and uneducated use of consumer credit by large numbers of Americans and includes a requirement that the successful conclusion of any personal bankruptcy case be accompanied by credit counseling for the filer, as well as a requirement that some filers repay some of their debt over a period of years, based on means.


The official unemployment rate has been the subject of considerable scrutiny and criticism over the years. Many social and economic researchers believe that the rate misrepresents the actual number of people who cannot find work to support themselves and their families. For example, the official unemployment figures do not count those who have given up searching for work because of failure to find work over a long period of time. The figures also leave out those who are underemployed, such as college graduates who take low-paying jobs or part-time jobs until adequate employment becomes available. It is likely, then, that the true number of people who cannot find adequate employment is higher than official statistics indicate.

Unemployment is a permanent feature of the economy. It can never be entirely eliminated since there will always be people who lose their jobs for various reasons. The transition between a lost job and the next job takes time even in the best of economic times.

The 1990s were a period of economic growth. The decade began with a seasonally adjusted unemployment rate in 1990 of 5.6%. The rate rose to a high of 7.5% in 1992 and then fell to a low of 4% in 2000. A rise in unemployment between 2000 and 2002 (31%) reflected the downturn in the economy that began in 2001 and was exacerbated in the uncertainty that followed the September 11 attacks.

By the end of 2002 the unemployment rate was 5.8%. (See Figure 3.9.) The first two quarters of 2003 saw the unemployment rate continue to rise. By June 2003 it had risen to a high of 6.5%. By August 2003 the unemployment rate had fallen slightly to 6%, but there were signs of a growing number of people dropping off the unemployment registers, having given up the search for a new job. The U.S. Bureau of Labor Statistics reported that in February 2005, eight million people were officially unemployed, or 5.4% of the labor-aged population wishing to work, down from the high of June 2003, but up from the previous month.


Some segments of American society have experienced more unemployment than other segments. According to the U.S. Bureau of Labor Statistics in Employment Characteristics of Families (April 2004), 8.1% of the nation's 75.3 million families reported having an unemployed member at some time during 2003. The proportion of black families with an unemployed member (13.7%) was higher than the proportion for Hispanic (11.1%), Asian (9.4%), or white families (7.1). (See Figure 3.10.)

Underemployed and Discouraged Workers

The U.S. Census Bureau's Current Population Survey regularly reports unemployment figures as well as figures for adults who are not in the labor force and a subgroup of people not in the labor force who are discouraged and have stopped looking for work. Dis couraged workers give up looking for work specifically because they believe no jobs are available for them. In January 2005 the Bureau of Labor Statistics reported in Labor Force Statistics from the Current Population Survey, that 515,000 workers had given up looking for work, up 83,000 over the number reported one year earlier.


In 1996 the passage of the Personal Responsibility and Work Opportunity Reconciliation Act (PL 104-193), the most sweeping welfare legislation since the 1960s, ended Aid to Families with Dependent Children (AFDC) and gave the states control over the administration of benefits in the form of Temporary Assistance for Needy Families (TANF) grants. In addition, the law made significant changes to Supplemental Security Income (SSI) and Medicaid.

Returning Local Control

The TANF block grant legislation (the welfare reform initiative) was designed to fulfill two primary objectives:

  • To return more control of relief assistance programs to the state governments—this granted the states great flexibility to design whatever mix of services and benefits they thought would reduce dependency and provide for the needy
  • To limit the amount of time a person spends on public assistance—specifically, the 1996 law contained a five-year lifetime limit on receipt of federally funded cash assistance and authorized states to impose shorter time limits at their discretion; and required that, by the year 2002, 50% of all recipients who had received cash aid for two years work at least thirty hours a week in order to continue receiving benefits

Many people argued that these were worthy goals. First, they believed local authorities would be better equipped to determine the needs of the people in their area than a distant federal government. Second, they believed the focus on encouraging the poor to work rather than receive public assistance was important because the old system unintentionally discouraged welfare recipients from working, since most welfare recipients could not meet the increased expenses incurred while working (transportation and child care) with their low-paying jobs. Supporters of the reform also pointed out that the welfare-to-work initiative had another, residual benefit: the likelihood that the children of the recipient would be impoverished as an adult were lowered.

However, critics of the reform bill pointed out that local responsibility for the poor has historically proven to result in discriminatory practices and lends itself to subjective and/or punitive practices. The second objective—to put people to work—resembled the deserving/undeserving moral judgments of the poor in the nineteenth century. It was in applying these same qualifications that the poorhouses were born.

These attitudes and their results continue to provide a basis for concern regarding welfare reform. While local control makes it easier to address the specific needs of the poor in particular areas, it also makes it possible for localities to restrict benefits in ways that a nationwide program would never be able to do. The welfare reform laws also contain an inherent assumption that anyone who really wants work can find it, and that anyone who cannot find work within a specified period must not really be trying and therefore does not deserve government aid. This assumption ignores the fact that many people on welfare, and poor people in general, lack the skills or education necessary for most jobs. The homeless, who do not have a fixed address or telephone number and may not have clean clothes for an interview, are especially vulnerable in this system.


Welfare reform proved to be more successful at removing individuals and families from welfare rolls than its many critics expected. Welfare rolls had declined by roughly half since the early 1990s and employment rates had risen for most former (and many current) welfare recipients. How much of these positive results were due to the economic boom of the 1990s, with the accompanying increase in jobs, remains to be determined. Since the economic downturn in 2001, when the most recent recession began, the number of welfare recipients continued to decline from 5.3 million in 2001 to 4.7 million in June 2004, according to the Administration for Children and Families of the U.S. Department of Health and Human Services.

Researchers know that there have been declines in the welfare rolls, but as of yet they do not know what happened to the people who are no longer aid recipients. Major studies are now underway. To get a look at what information is available and what the welfare reform act meant to people whose lives were directly affected, Sheila Zedlewski and Donald Alderson of the Urban Institute analyzed current data and presented the following preliminary results of the AFDC-to-TANF reform measures (Families on Welfare in the Post-TANF Era: Do They Differ from Their Pre-TANF Counterparts? Paper presented at the American Economics Association meeting, New Orleans, January 2001). They based their analysis on a comparison of 1,831 families on TANF in 1997 and 850 families in 1999, representing 2.2 million and 1.5 million families, respectively. The characteristics and work activities presented were obtained from interviewing the adult most knowledgeable about the children in the family, usually the mother.

The researchers found:

  • The proportion of single mothers on welfare who reported living with partners increased.
  • The proportion of African-Americans on welfare who reported living with partners increased.
  • The proportion of adults on welfare who worked for pay rose.
  • The proportion of recipients who were new entrants to the welfare system was about the same, despite some new state programs that attempt to divert adults from enrolling in TANF.
  • Adults on TANF in 1999 were no more disadvantaged than those on TANF in 1997.
  • Longer-term welfare recipients were significantly more disadvantaged than the new entrants.
  • The proportion of recipients with less than high school education, and therefore less employable, was the same (2%) in 1997 and 1999.


It is extremely difficult for the homeless to escape their condition without a job. Yet it is equally difficult for the homeless to find and keep good jobs. Health Care for the Homeless Information Resource Center's newsletter "Employment for Homeless People: What Works, Fall 2000" listed the following barriers:

  • Lack of transportation
  • Medical and dental problems
  • Housing instability
  • Mental illness
  • Substance abuse
  • Domestic violence
  • Lack of job skills
  • Criminal record
  • Lack of child care
  • Lack of education
  • Diminished self-confidence

In addition, the homeless, like other workers, are subject to the state of the labor market. The availability of jobs and the wages paid for the available jobs often determined whether or not people could remove themselves from homelessness.

Wage Barriers to Exiting Homelessness

Of the homeless respondents from a 1996 study by the Urban Institute, 44% reported working in the previous month. Two percent earned income as self-employed entrepreneurs—by peddling or selling belongings. Forty-two percent of the homeless respondents worked for, and were paid by, an employer.

Advocates for the homeless are concerned that this dependency on wages, combined with the unfavorable labor market conditions, actually supports continued homelessness. Since the majority of homeless people do not have more than a high school education, and since a majority of the low-paying jobs go to those with at least a high school education, advocates worry that the available job opportunities for homeless people provide an insufficient base for exiting homelessness.


It costs money to live. Even homeless people have needs that can only be met with money. From needing something as simple as a toothbrush or a meal, to money for a newspaper or a phone call to a job prospect, homeless people need money to begin to improve their lives. Out of the need to survive, homeless people have come up with a number of ways to earn money, weaving their ventures among local ordinances, public opinion, and the labor market.

Street Newspapers: Bootstrap Initiatives

In the United States, as well as overseas, homeless people are writing, publishing, and selling their own newspapers. The 2003 North American Directory of Street Newspapers, published by the National Coalition for the Homeless in September 2003, reported there were over fifty street newspapers across the United States and Canada.

Many street newspaper publishers belong to a professional organization, the North American Street Newspaper Association (NASNA), organized in Chicago in 1996. NASNA holds an annual conference, offers business advice and services, and supports street newspaper publishers in the same way that any professional organization supports its membership. They also lobby the government on homeless issues.

Generally, the street newspapers are loaned on credit to homeless vendors who then sell them for $1 or $2 each. At the end of the workday, the vendor pays the publisher the agreed-upon price and pockets the remainder as profit. For example, Boston's Spare Change newspaper publishes 10,000 copies every two weeks. Vendors purchase newspapers for $0.25 each and resell them for $1.00, pocketing $0.75 for each paper sold. New York City's Street News gives the vendors thirty free copies, allows them to buy additional copies for $0.25 each, and the papers are sold for $1.25.

This cooperative arrangement among publishers, vendors, and consumers has many benefits:

  • Creation of jobs
  • Supports the work ethic
  • Accommodates the mobility of homeless people
  • Provides reliable employment despite crisis living conditions
  • Informs the public about homelessness
  • Erases stereotypes of the drunken, illiterate, "unworthy" homeless person
  • Gives the writers and vendors a sense of accomplishment
  • Provides immediate cash to people who desperately need it

Most of the homeless newspaper vendors have not been able to earn enough just from selling newspapers to move themselves from homelessness, but as the quality and availability of these publications grow, homeless people envision the street newspaper industry becoming a means of moving tens of thousands from homelessness.


The earliest known homeless publication, Street News, was launched in 1989 by the New York Times and the New York Metropolitan Transportation Authority. By 1991, when the publication suffered from a scandal involving misuse of funds, Street News was boasting some 25,000 copies per issue, which then fell to 6,000 by 2005. The paper is known for its radical political stance and unusual stories.


Begun in 1992 as one of the nation's first street newspapers to benefit the homeless, Spare Change has been published every other week by the Homeless Empowerment Project (HEP) in Cambridge, Massachusetts. Its stated mission is to "play a role in ending homelessness in our community by providing income, skill development and self-advocacy opportunities to people who are homeless or at risk of homelessness." The newspaper provides a forum for creative expression and advocacy for homeless individuals. Along with the production, distribution, and sale of the street newspaper, HEP operates a training center for teaching computer skills to the homeless. They also sponsor a writer's workshop and promote a speaker's bureau. In June 2005, however, the paper was facing an uncertain future, with an operating loss of $4,000 per issue, according to Jay Fitzgerald in the Boston Herald (June 16, 2005).

Day Labor

Regular work, characterized by a permanent and ongoing relationship between employer and employee, does not figure significantly in the lives and routines of most homeless, as it is usually unavailable or inaccessible. Homelessness makes getting and keeping regular work difficult due to the lack of a fixed address, communication, and, in many cases, the inability to get a good night's sleep, clean up, and dress appropriately. Studies have found that the longer a person is homeless, the less likely he or she is to pursue wage labor and the more likely that person is to engage in some other form of work. For those who do participate in regular jobs, in most cases, the wages received are not sufficient to escape from living on the street.

Day labor, wage labor secured on a day-to-day basis, typically at lower wages and changing locations, is somewhat easier for the homeless to secure. Day labor may involve unloading trucks, cleaning up warehouses, cutting grass, or washing windows. Day labor often fits the abilities of the homeless because transportation may be provided to the work site, and appearance, work history, and references are less important. Equally attractive to a homeless person, day labor usually pays cash at quitting time, thus providing immediate pocket money. Day labor jobs are, however, by definition, without a future. They can provide for daily survival on the street but are not generally sufficient to get a person off the street. Consequently, many homeless turn to shadow work.

Shadow Work

Shadow work refers to methods of getting money that are outside the normal economy, some of them illegal. These methods include panhandling, scavenging, selling possessions, picking up cans and selling them, selling one's blood or plasma, theft, or peddling illegal goods, drugs, or services. A homeless person seldom engages in all of these activities consistently but may turn to some of them as needed. Researchers estimate that 60% of homeless people engage in some shadow work. Shadow work is more common for homeless men than homeless women. Theft is more common for younger homeless persons.

A mixture of institutionalized assistance, wage labor, and shadow work is typical of those who live on the streets. Studies have found that many homeless people are very resourceful in surviving the rigors of street life and recommend that this resourcefulness be somehow channeled into training that can lead to jobs paying a living wage. Some observers suggest, however, that homeless people who have adapted to street life may likely need transitional socialization programs as much as programs that teach them a marketable skill.

Institutionalized Assistance

Institutionalized assistance refers to "established or routine monetary assistance patterned in accordance with tradition, legislation, or organizations" (David Snow et al., Material Survival Strategies on the Street: Homeless People as Bricoleurs, Homelessness in America, edited by the National Coalition for the Homeless, Phoenix, AZ: Oryx Press, 1996). This would include institutionalized labor, such as that provided by soup kitchens, shelters, and rehabilitation programs that sometimes pay the homeless for work related to facility operation. The number of people employed by these agencies is a small percentage of the homeless population. In addition, the pay—room, board, and a small stipend—tends to tie the homeless to the organization rather than providing the means to get off the street.

Institutionalized assistance also includes income supplements provided by the government, family, and friends. According to Material Survival Strategies on the Street, while a considerable number of the homeless may receive some financial help from family or friends, it is usually small. Women seem to receive more help from family and friends and to remain on the streets for shorter periods of time than men. Cash from family and friends seems to decline with the amount of time spent on the street and with age.


According to the 1996 Urban Institute study, homeless people say that the primary reason they cannot exit homelessness is insufficient income. Of those clients surveyed, 54% cited employment-related reasons for why they remained homeless. Nearly a third (30%) cited insufficient income and nearly a quarter (24%) cited lack of a job.

The data from the Urban Institute study showed how little income the homeless earn. Eighty-one percent of the "currently" homeless had incomes of less than $700 in the thirty days before the study; the average income was $367. Most of the homeless in the study were receiving their income from Aid to Families with Dependent Children (now TANF). Of formerly homeless people surveyed, the median income of $470 would amount to an annual income of $5,640, an amount well below the poverty level for a single person ($7,740 in 1996).

These income levels clearly demonstrate the financial difficulty a homeless person encounters in trying to permanently exit homelessness or poverty. However, exiting homelessness—especially by the chronically homeless—requires more than income. Persistent medical assistance, sometimes for an entire lifetime, has to be available for the mentally ill, or people with addiction and substance abuse problems. Furthermore, without programs such as job training, assistance with general education, help with socialization skills, and in many instances counseling, the maintenance of a degree of independent life for the long term can be very difficult for the chronically homeless.

About this article

Employment and Poverty Among the Homeless

Updated About encyclopedia.com content Print Article