PRRAC Poverty & Race Research Action Council
Home About PRRAC Current Projects Publications Newsletters Resources Contact Us Support PRRAC Join Our Email List

"The Ethno/Racial Context of Poverty in Rural and Small Town America,"

by Calvin L. Beale March/April 2003 issue of Poverty & Race

Everyone with an interest in rural and small town poverty in the United States is aware that it frequently occurs in an ethno/racial context, just as in cities. This article is intended primarily to look at nonmetropolitan (nonmetro) areas having high incidence of poverty, as revealed by the 2000 Census, and: (1) to determine the extent to which such poverty is that of minority populations; and (2) identify ways in which these areas vary by their ethno/racial makeup in poverty-relevant characteristics of education, health, dependency per worker, steadiness of work, family structure, transportation and language proficiency. It will also review trends in poverty levels from 1990-2000.

The standard used to denote a high-poverty incidence is 20% or more of the population. This is better than a third higher than the nonmetro average of 14.7% and, by representing a fifth or more of all residents of an area, is surely a conservative indicator of the presence of serious income problems. (Poverty income thresholds vary by number and age of persons in households and families. For a family of two adults and two children, the Census poverty measure was annual income less than $16,895 in the preceding year, 1999). Nonmetro counties are those not in metro areas, which contain urbanized cores of 50,000 or more people, with boundaries generalized out to county lines and including any fringe counties that meet certain tests of metro character and job commuting into the core counties.

Of all nonmetro counties, 444 of them (a fifth) had poverty rates of 20% or higher in the 2000 Census, based on income received in the prior year (Fig. 1). In three-fourths of this group, the high-poverty proves to be primarily that of racial or ethnic minorities. In such counties, either: (1) a majority of the poor are Black, Native American or Hispanic; or (2) it is only the high incidence of poverty among these minority groups that brings the overall county rate above 20%. In the remaining fourth of the high-poverty counties, the problem is predominantly among non-Hispanic Whites, with most such areas being in the Southern Highlands -- centered on Eastern Kentucky, West Virginia and parts of Missouri and Oklahoma.

Black high-poverty areas

Two-hundred-and-ten of the high-poverty counties were characterized by poverty among Blacks. They lie in the old plantation belt of the southern coastal plain, especially from Southern North Carolina through Louisiana. Here, 39% of the Black population lived with poverty-level income (Table 1), a proportion well above that of Blacks in other nonmetro counties (28.1%) or in metro areas (23.7%). Among measures relevant to poverty that can be identified from the Census, the counties of high Black poverty stand out most prominently in the fact that 32.7% of all children under 18 lived in female-headed households with no husband present. This is a much higher proportion than that found in other types of high-poverty areas, and double that in nonmetro counties that have less than 20% poverty. Poverty in female-headed families with children, but no husband present, is dramatically higher everywhere than is true of other household types. In nonmetro America as a whole, persons in such families had a poverty rate of 41.6%, whereas in all other families with minor children the rate was just 9.6%.

The Black high-poverty counties also have the highest rate of households without a motor vehicle (12.5%), thus inhibiting access to employment and essential services. In addition, they have the highest self-reported incidence of disability (27% of the population 21-64 years of age) of any of the minority high-poverty county blocs. In this respect, though, the high-poverty areas of the Southern Highlands -- where the population is overwhelmingly non-Hispanic White -- report a higher disability rate (31%) than any of the minority groups, a pattern consistent with earlier censuses.

Hispanic high-poverty areas

In 74 counties, high-poverty derived from conditions among Hispanics. Although these counties are still concentrated in the Southwest, especially in Texas and New Mexico, they have begun to appear elsewhere, with examples now in Florida, Georgia, Missouri and Washington as Hispanics have both dispersed and grown rapidly from immigration. Within the 74 counties, Hispanic poverty averaged 32.3%, a lower incidence than that of Blacks in Black poverty areas, and a major decline from the 41% level in the 1990 Census. This drop was achieved despite the fact that Hispanics rose as a percentage of the entire population in these areas (from 52.8% to 58.5%), while the higher income non-Hispanic Whites became a smaller proportion and often decreased in absolute numbers as well from outmigration. Despite this rising dominance of Hispanics in high-poverty areas where the poor are mostly Hispanic, a declining proportion of all nonmetro Hispanics live in such areas. Their growth in other areas was so rapid in the 1990s that the percentage of all nonmetro Hispanics living in the Hispanic high-poverty counties fell from 34.1% to 25.6%. This contrasts with nonmetro Blacks and Native Americans, who showed little shift away from high-poverty areas beyond their 1990 level.

The Hispanic high-poverty counties are very different from others in one social measure -- the percentage of people who reported they do not speak English "very well" (21.7%). The lack of English proficiency is especially prevalent in areas where recent immigration has been large, such as along the Mexican border, but also is far above average in longer-settled areas. Hispanic poverty counties have a large minority of adults who have not completed high school (36.8%), a condition partly derived from the frequency of recent immigration. This is well above the corresponding figure of 21.3% for Hispanics in nonmetro counties that do not have high-poverty, although it is no worse than is found in the non-Hispanic Southern Highlands poverty areas. Both the Hispanic and Black high-poverty county groups have more than double the ratio of high school dropouts to persons with a four-year college degree that is found in nonmetro counties that do not have high-poverty.

Native American high-poverty areas

Forty nonmetro counties with high-poverty incidence reflected low income among Native Americans, including Alaskan natives. They are all in areas of either historic tribal presence or 19th Century Indian removal, especially in the Northern Plains, coastal Alaska, the Southwest and Oklahoma. The poverty level of Native Americans within these counties was 40.7%, a level greater than that of the dominant minorities in other high-poverty county groups. Furthermore, the Native American counties did not simply have a higher incidence of poverty, they had the highest occurrence of severe poverty. A full fifth (20.5%) of the total population in these areas lived in households with income less than three-fourths the poverty standard. Thus, substantial increases in income --whether from earnings, retirement or assistance -- would be required to lift this segment of the population to a minimally sufficient level of living.

The Native American high-poverty counties also had the highest ratio of total population to employed people of all county groups, with 288 persons of all ages for every 100 with jobs. By comparison, in nonmetro counties with poverty lower than 20%, there were just 214 persons for each 100 workers. The high dependency ratio in Native American counties stems partly from the age composition of the population, but its impact on income adequacy is worsened by the fact that only 36% of all males 16 years old and over had full-time, year-round work, compared with 47.5% in nonmetro counties without high-poverty. Low labor force participation, high unemployment within the labor force and lack of steady work for those employed all contribute to the situation.

Children and the elderly are typically regarded as the most vulnerable classes in society. With a comparatively young population, Native American persons in poverty consist much more of children (and, perforce, the parent or parents with whom they live) than of elderly people, in comparison with other minorities or the general population. In the Native American high-poverty counties, there were 5.9 poor children (under age 18) for each poor person 65 and over. This compares with ratios of 4.2/1 in Hispanic counties, 3.2/1 in the Black counties, and only 2.6/1 in nonmetro counties with poverty incidence of below 20%. Thus, alleviation of the conditions produced by poverty has to be more focused on children and their parents than it does in most other areas.

In many of the Native American high-poverty areas, especially in the Northern Plains, the non-Hispanic White proportion of the population has been dwindling steadily as the Native American population has grown and the number of local White farmers and ranchers has fallen. The non-Hispanic White share of population in these areas fell from 44.5% in 1990 to 39.8% in 2000. Thus, the overall 1990-2000 reduction in poverty rate in the Native American areas, from 34.0% to 28.3%, was achieved despite the diminished presence of the racial group with the highest income

Asian and Pacific Islander high-poverty areas

The Asian and Pacific Islander population is comparatively small in nonmetro locations and is the main component of the poor in only two high-poverty counties. One is Aleutians East Borough, Alaska, where a large group of Filipino men lived in 2000 who worked in seafood processing but lacked good income. The other instance was Kalawao County, Hawaii, where most residents are elderly Hawaiian and Asian survivors of the former colony for persons with Hansen's disease (leprosy). Nationally, Asian and Pacific Islander nonmetro poverty stood at 15.6%, well below that of other ethno/racial minorities.

Non-Hispanic White poverty in minority areas

In all three types of minority high-poverty areas, the poverty rates of non-Hispanic Whites are somewhat higher than they are in nonmetro counties without 20% poverty. They range from 13.3% in the Black counties to 14% in Hispanic areas, compared with a 10.8% incidence in counties with overall poverty of less than 20%. Thus, non-Hispanic Whites are not altogether immune from the forces that lead to high-poverty in minority areas. However, the difference in these areas between White poverty and minority poverty is still stark. Black and Native American poverty rates run nearly three times that of the non-Hispanic White population in the respective Black and Native American poverty areas and better than double that of non-Hispanic Whites in Hispanic areas.

Trends from 1990-2000

Minority nonmetro poverty rates declined from 1990-2000, as did those in the nation as a whole. In the South, poverty among nonmetro Blacks dropped from 40.8% to 33.4%, a reduction of better than a sixth. All states saw declines. In the Deep South states of Arkansas, Louisiana, Mississippi and Texas, with the highest prior rates of nonmetro Black poverty (45% to 53%), the incidence fell by more than a fifth. The relative amount of decline was more limited (about one-ninth) in states that had already achieved levels below 30% by 1990 (Delaware, Maryland, Virginia). Thus, some leveling of differences within the South has occurred, but the total nonmetro Black poverty rate is still so high that at the recent pace of change it would take 2 1/2 decades to bring it below 20%, much less to a more acceptable level. The worst conditions now remaining are in Louisiana, where there are still eleven high-poverty parishes in which over half of the Black population was still in poverty in 2000. This is a level rarely seen elsewhere now, even in Mississippi.

Nonmetro Hispanic poverty nationwide fell from 33.4% to 27.2% in the 1990-2000 decade (Table 2). This was very similar to the rate of decline among Blacks, but is notable given the fact that the Hispanic group -- alone among the three major minorities -- was substantially enlarged by rapid infusion of immigrants of below-average education and immediate earning capacity, whose presence would be expected to retard a lowering of the poverty rate in the short term.

Areas of high non-Hispanic White poverty are not discussed here. It is worth noting, though, that in such areas of the Southern Highlands the pace of White poverty decline in the 1990s (12.8%) was distinctly below that of the principal minorities in each of the high-poverty county groups where minority poverty predominates. In part, this may stem from the fact that Southern Highlands poverty levels are lower than those of minorities -- even though high by national standards -- and high-poverty levels generally fell at a faster rate than lower levels in the 1990s.

Although poverty-level income is more common in the nonmetro population than in metro areas, most of the progress made in the United States between 1990 and 2000 in reducing national poverty occurred in nonmetro areas. This was true both among minority populations and among non-Hispanic Whites.

In nonmetro counties, there was a strong correspondence between the 1990 poverty rate and the relative change by 2000 -- that is, the greater the poverty, the greater the rate of decline. Thus, there is now a good bit less disparity in poverty rate between nonmetro areas than was the case earlier, both for minorities and non-Hispanic Whites. There was a modest convergence among metro counties as well, but in part this resulted from counties that had below-average poverty in 1990 reverting to slightly higher levels in 2000, rather than from strong reductions in higher-poverty areas.


The year 1999, for which the 2000 Census income data were collected, was a banner year for earnings -- with the lowest unemployment rate in 30 years -- and also a prime period for nonearnings income, such as capital gains. Two findings would appear to deserve equal emphasis in this context. The first is that while nonmetro areas more than shared in the reduction of poverty rates reported for that year in comparison with the 1990 Census, this was particularly true of Blacks, Native Americans and Hispanics of any race. Secondly, however, despite this improvement, the nonmetro population is still more subject to low income and high-poverty than are metro residents. And within nonmetro areas, minorities have poverty rates that are more than double those of non-Hispanic Whites.

Although the minority high-poverty counties vary in the kinds of poverty-related measures by which they are most affected, without exception they show multiple types of social and economic characteristics on which they differ in a problem context from the mass of nonmetro counties that have less poverty.

Where high rural and small town poverty characterizes entire counties, it reflects historic geographic concentrations of minority populations in three cases of every four. It limits the tax base and, where chronic, as it typically is, imposes a poverty of services. It is usually in such concentrations that conditions among minorities are the worst. But each of the major minorities has its unique history and its own signature characteristics that are poverty-related and essential to recognize if income problems are to be addressed successfully.


Calvin L. Beale (, 202/694-5416) is Senior Demographer, Economic Research Service, U.S. Department of Agriculture, Washington, DC. The author is grateful for the assistance of Kathleen Kassel in preparation of this article.


Join Our Email List
Search for:             
Join Our Email List