Trends and Correlates of Poverty and Inequality

In this section, data are presented and discussed on general trends in poverty and inequality since the War on Poverty began. In addition to general trends, data are also presented on the levels of poverty and inequality among various segments of the population. This is essential because certain segments of the population have always experienced higher poverty rates than other segments. Further, over the decades there have been significant and unforeseen economic structure transformations and changing social conditions that have differentially affected some segments of the population. These economic and social changes have also made reducing poverty levels much more difficult than once expected. To be effective, poverty policies and programs must take these trends and differences into account.

In Figure 1, data are presented showing the number of people living in poverty and the percent of the population living in poverty from 1965 to 2003. It is evident from this figure that while there have been variations in the proportion of the population living in poverty over the years; poverty rates have been remarkably consistent (Danziger and Weinberg 1994). This is despite a massive input of resources intended to reduce or eliminate poverty. In 1965, 17.3 percent of Americans were living in poverty. With the aid of "War on Poverty" resources, this proportion declined steadily to 11.1 percent in 1973. Poverty rates have never been that low since. In the 30 years since 1973, the proportion of the population living in poverty has increased during times of economic recession (reaching a high of 15.1 percent in 1993), and decreased during times when the economy was robust (reaching a new low of 11.3 percent in 2000), but the range of this variation has been limited. The number of Americans living in poverty reached a low of about 23 million in 1973 and has been as high as 39.3 million in 1993.

Table 1 provides data on income inequality from 1967 to 2003. In general, this table shows a slight decline in inequality levels from 1967 to 1975, and then since 1975, income inequality has increased steadily and substantially. The first segment of Table 1 shows mean household income by quintile using constant 2003 dollars. These data make it apparent that average incomes have increased significantly since the 1960s. However, most of this increase has gone to the affluent, while the incomes of the poor have changed little. Between 1975 and 2003, the incomes of the 20 percent of the population at the bottom of the wage scale increased by only 9 percent. At the same time, the incomes of the 20 percent of the population at the top of the wage scale increased by 64 percent. Over the past 30 years, persons in higher income categories have consistently had greater income growth than persons in lower income categories. As a result, the gap between the poor and the wealthy has increased extensively.

The second segment of Table 1 shows the shares of household income by quintile. Again, since 1975, the percent of income going to the lowest 20 percent has steadily declined, while the proportion going to the highest 20 percent has steadily increased. The share of household income going to the bottom 20 percent has declined from 4.4 percent in 1975 to 3.4 percent in 2003. At the same time, while the top 20 percent had 43.2 percent of the income in 1975, this share has increased to 49.8 percent in 2003.

Further evidence of increased levels of inequality is found when looking at the income ratios and the gini coefficient. Since 1975, both the 90/10 and the 80/20 income ratios have steadily increased. By 2003, individuals at the 90th income percentile made 11.22 times more than individuals at the 10th income percentile. Further, individuals at the 80th income percentile made 4.83 times more than individuals at the 20th income percentile. The gini coefficient has also steadily increased and was .464 by 2003. Data was not available to directly compare inequality levels in metro and nonmetro communities. Research by others, however, indicates that inequality levels are even higher in nonmetro areas than in metro areas (McLaughlin 2002). In sum, the data in Table 1 show that the economy of the United States is growing, but this increased wealth is disproportionately going to the wealthy while the poor have benefited little.

Poverty, Inequality and Residence

From the time of the initial settlement of the Americas by people of European descent, income levels have always been higher and poverty rates lower in metropolitan compared to nonmetropolitan communities (Albrecht et al. 2000). As a consequence of these metro economic advantages, there has been a near steady migration stream from nonmetro to metro areas throughout U.S. history (Johnson 1989). The data presented in Table 2 show that during the past four decades, poverty rates have always been higher in nonmetro counties than in metro counties. However, the gap is substantially smaller in 2003 than in the 1960s. In 1967, 20.2 percent of the residents of nonmetro counties were living in poverty, compared to only 10.9 percent in metro counties. Further, in 1967, the total number of people in poverty (13.9 million) in nonmetro counties exceeded the total number of people living in poverty in metro counties (13.8 million). In addition, in 1967, the proportion of nonmetro residents in poverty exceeded the proportion of people in poverty in both central cities (15.0 percent) and suburbs (7.5 percent). By 2003, circumstances had changed considerably. While nonmetro poverty rates (14.2 percent) still exceed total metro poverty rates (12.1 percent), the gap is much smaller than in previous years. Also, since 1980, poverty rates in the central city have been higher than nonmetro poverty rates. Suburban poverty rates remain lower than nonmetro poverty rates. Finally, in 2003, nearly 80 percent of all Americans living in poverty are in metro counties. Of course, a critical reason for this is that the U.S. population is much more metropolitan than in the 1960s. The data in Table 2 show that since the 1960s, the population of metro counties has increased significantly, while the population of nonmetro counties has remained remarkably consistent.

Poverty, Inequality and Race

From the time our nation was founded, a number of racial and ethnic groups have experienced extensive levels of prejudice and discrimination. As a consequence of this prejudice and discrimination, minority people have had fewer employment opportunities and obtained lower levels of education. As a result, minority incomes have always been lower than white incomes, and minority poverty rates have always been higher than white poverty rates (Albrecht et al. 2005). Minority disadvantage has consistently existed in both metro and nonmetro communities (Albrecht et al. 2005).

Continued minority disadvantage is very apparent in the data presented in Tables 3 and 4. The year 1973 was chosen as the base for these tables because definitions of race and ethnicity have been relatively consistent since that time. Table 3 presents data showing the number and percent of people living in poverty by race from 1973 to 2003. Data are shown for the three most numerous racial and ethnic groups in the United States, non-Hispanic whites and blacks and Hispanics. Throughout the 30 year time period represented in this table, poverty rates for whites have been much lower than poverty rates for both blacks and Hispanics. White poverty rates have ranged from a low of 7.4 percent in 2000 to a high of 9.7 percent in 1985. In comparison, poverty rates for blacks ranged from a low of 22.5 percent in 2000 to a high of 32.5 percent in 1980, while Hispanic poverty rates ranged from a low of 21.5 percent in 2000 to a high of 30.3 in 1995. Another important piece of information in Table 3 concerns the number of people in poverty by race. For both white and black populations, the numeric increase in the number of people in poverty has not been especially large (from 12.9 million in 1973 to 15.9 million in 2003 for whites and from 7.4 million in 1973 to 8.8 million in 2003 for blacks). In comparison, the number of Hispanic people in poverty increased from 2.4 million in 1973 to 9.1 million in 2003. This increase reflects the tremendous growth in the Hispanic population of the United States in recent decades (from 10.8 million in 1973 to 40.3 million in 2003). The fact that minority populations are growing much more rapidly than white populations has important fundamental implications for poverty in this country. From 1973 to 1995, the absolute number of white individuals in poverty exceeded the absolute number of black and Hispanic individuals combined in poverty. However, as a result of growing minority populations, and especially the growing numbers of Hispanics, and continued high poverty rates among minorities, the absolute number of black and Hispanic individuals in poverty has exceeded the number of white individuals in poverty since 1995.

Table 4 provides additional data on the household incomes of whites, blacks and Hispanics by showing the percent distribution of incomes and median incomes (in constant 2003 dollars) from 1973 to 2003. The data in this table is based on households rather than individuals. The census defines a household as consisting of all people who occupy a housing unit regardless of relationship. A household may consist of a persons living alone or multiple unrelated individuals or families living together.

Again, the economic disadvantages of minorities are apparent in this table. The median household incomes of whites have always been significantly higher than the median household incomes of blacks and Hispanics. In 2003, the median household income for whites was $47,777, compared to $29,645 for blacks and $32,997 for Hispanics. When looking at the percent income distribution, Table 4 shows that the percent of white households in the very low income categories is much smaller than the percent of minority households in these same categories. In 2003, 13.4 percent of white households had incomes of less than $15,000, compared to 18.9 percent of Hispanics and 27.4 percent of blacks. Likewise, the percent of white households with incomes of $100,000 or more (17.3 percent) was much greater than the percent of blacks (6.7 percent) or Hispanics (7.5 percent). The overall increase in incomes during this 30 year period is apparent in Table 4 as there has been a decline in the percent of households in the low income categories, an increase in the percent of households in the high income categories, and an overall increase in median household incomes.

Family Structure, Poverty and Inequality

Among of the more profound changes occurring in the United States in recent years are changes in the family (Albrecht and Albrecht 2004; Giddens 2003). Prior to the last third of the 20th century, most Americans lived in a family consisting of a husband and wife and their biological children. Both divorce and nonmarital childbearing were relatively rare, especially in nonmetro communities (Popenoe 1996). Consequently, during the 1950s, about 80 percent of American children under age 18 were living with both biological parents who were married to one another (Bumpass and Sweet 1989), and the vast majority of adult men and women were married (Saluter and Lugaila 1998). Beginning in the 1960s, the structure of American families changed rapidly and dramatically (Wu and Wolfe 2001). Among the more prominent of the changes include much higher divorce rates (Cherlin 1992), more nonmarital births (Wu et al. 2001), the postponement of first marriage, and an increase in nonmarital cohabitation (Bianchi and Casper 2000; Bumpass and Lu 2000). Most estimates indicate that about 50 percent of all current marriages will end in divorce (Popenoe 1996), while about 1 in 3 births now occur out of marriage (McLanahan et al. 2001). On average, the first marriage for both men and women now occurs about four years later than in the 1950s (Fields and Casper 2001), while there has been a sevenfold increase in the number of cohabitating couples since 1970 (Casper and Cohen 2000). Of relevance to this discussion is that these family structure changes have important implications for poverty and inequality (Garfinkel and McLanahan 1994). While nonmetro families remain more traditional than metro families, poverty levels among single-parent families are higher in nonmetro counties than in metro counties (Albrecht et al. 2000).

Table 5 presents data on family structure and poverty. This table is based on data for families rather than for individuals or households. The Census Bureau defines a family as two or more people related by birth, marriage or adoption and residing in the same housing unit. This table shows that nonmarried-couple families (and especially female-headed families) are much more likely to be in poverty than married-couple families. In 2003, 28 percent of the female-headed families were in poverty, compared to 13.5 percent of the male-headed families and only 5.4 percent of the married-couple families. Part of the problem, of course, is that single-parent families generally only have one adult who can potentially be employed. This is especially problematic for female-headed families where the generally higher male income is missing. In 2003, the median earnings of full-time, year-round female workers was only 75.5 percent as high as similarly employed male workers ($40,668 for males compared to $30,724 for females). The rapidly changing family structure means that an increasingly large proportion of individuals in poverty are living in single-parent and especially female-headed families. In 1965, 28.5 percent of all families in poverty were female-headed; by 2003, over half of the families in poverty were female-headed.

Age, Poverty and Inequality

Historically, poverty rates have been significantly higher among children and the elderly than among those in the prime working ages. The obvious reason is that those in the prime working ages can generally obtain employment and support themselves, while the old and the young tend to be dependent on the employment of others. This is problematic for nonmetro areas where the proportion of the population that is either older or younger exceeds similar proportions in metro areas.

In Table 6, data are presented showing the number of people in three different age categories and the number and percent of these people in poverty from 1967 to 2003. This table shows that in 1967 poverty rates were much higher among those 65 and older (29.5 percent) and among children under 18 (16.6 percent) than among persons between 18 and 64 (10.0 percent). Since 1967, the age structure of the U.S. population has changed considerably. In 1967, 36.0 percent of U.S. residents were under 18, 54.7 were between 18 and 64, and 9.3 percent were 65 or older. By 2003, the proportion of the population that was under 18 had declined to 25.4 percent, while the proportion from 18 to 64 had increased to 62.6 percent and the proportion of elderly had increased to 12.0 percent. Between 1967 and 2003, the absolute number of young people changed very little, while there were substantial increases in the number of people in the other two age categories. Throughout the time period covered in the table, poverty rates have changed little for working age people (ranging from a low of 9.0 percent in 1970 to a high of 11.4 percent in 1995). Poverty rates for children have always been much higher and have ranged from 15.1 percent in 1970 to 20.8 percent in 1995. Meanwhile, poverty rates for the elderly have declined extensively. By 2003, elderly poverty rates were only 10.2 percent, and were actually lower than for working age people. No doubt the reduction of poverty rates among the elderly is a result of successful War on Poverty programs. The implications of these age structure changes for the distribution of poverty is important. A decline in the proportion of the population under age 18, and a decline in elderly poverty rates have resulted in poverty being much more concentrated among those who are working aged than in the past. In 1967, 38.6 percent of the population in poverty was working age; by 2003, more than one-half (54.2 percent) of those in poverty were of working age.

In sum, in this section on trend and correlates of poverty, the data show that poverty is remarkably persistent as the percent of people in poverty has only ranged from between 11.1 percent and 15.1 percent since 1973. Since the War on Poverty began in the 1960s, however, the correlates of poverty have changed considerably. Since the 1960s, poverty has become much more concentrated among metro residents than among nonmetro residents; persons in poverty are much more likely to be minority; persons in poverty are much more likely to reside in female-headed households; and persons in poverty are much more likely to be between the ages of 18 and 64.

back     next