Nonmetro Poverty: Assessing the Effect
of the 1990s
Dean Jolliffe
The 1990s ushered in many changes in America that may have either
aided or burdened the poor in nonmetropolitan (nonmetro) areas.
The Personal Responsibility and Work Opportunity Reconciliation
Act (PRWORA) of 1996 was designed to move poor people from welfare
to work by imposing work requirements and a 5-year lifetime limit
on Federal benefits. The law also limited who was eligible to receive
assistance. These changes in welfare policy affected the poor both
adversely (by reducing benefits) and positively (by providing stronger
incentives toward achieving self-sufficiency). The overall effect
of these two opposing forces on poverty is the subject of contentious
debate.
Unprecedented economic growth and demographic shifts during the
1990s formed the backdrop for welfare policy changes. Between 1993
and 2000, the economy grew by 4 percent annually, versus 2.7 percent
during the 20 years prior to 1993. As the U.S. economy boomed in
the 1990s, so, too, did nonmetro population growth—over 10
percent during the 1990s, compared with 3 percent in the 1980s.
What effect did these major demographic and economic changes as
well as government policy have on poverty? During the 1990s, the
nonmetro poverty rate declined fairly steadily from a high of 17.1
percent in 1993 to a record low of 13.4 percent in 2000. However,
with the end of the economic expansion, the nonmetro poverty rate
crept back up to 14.2 percent in 2001. Poverty in metro areas followed
a similar pattern, declining from a high of 14.6 percent in 1993
to a low of 10.8 percent in 2000, and edging up to 11.1 percent
in 2001.
Degree of Urbanization Aligned With Degree of Poverty
What's
Nonmetro?
The
data from the Current Population Survey (CPS) used
in this analysis identify metro and nonmetro areas
according to the 1993 designation by the Office of
Management and Budget (OMB). Metro areas are defined
to include core counties with one or more central cities
of at least 50,000 residents or with an urbanized area
of 50,000 or more and total area population of at least
100,000. "Fringe" counties that are economically
tied to the core counties are also considered as metro
areas.
Nonmetro areas are all areas outside the boundaries
of metro areas, and contain no cities with populations
over 50,000. OMB is currently revising its metropolitan
area classification system using new definitions
and the 2000 Census data. A discussion of the revision
underway is provided in "Behind the Data." Unfortunately, the new nonmetro designations are
not yet available in the CPS data and are therefore
not incorporated in this article.
Metro counties are commonly characterized as densely populated
central cities and suburbs, and nonmetro counties as sparsely populated
small towns and open countryside. This distinction oversimplifies
the many differences across metro and nonmetro areas. Some metro
counties have relatively small populations and are adjacent to
rural areas, and some nonmetro counties contain urban areas but
still qualify as nonmetro (see “What’s Nonmetro?”).
A more comprehensive classification—separating metro areas
into highly and less urbanized counties (using 1 million population
as the cutoff) and nonmetro areas into somewhat urbanized (with
urban population of 20,000 or more) and more rural counties (with
smaller or no urban population)—reveals important differences
in poverty. Throughout the 1990s, highly urbanized metro areas
had the lowest incidence of poverty and the more rural nonmetro
areas had the highest, indicating that poverty is higher on average
in the least populated areas. The greatest reduction in poverty
in the 1990s occurred in the least populated rural areas. Poverty
declined from 17.9 percent in 1989 to 14.9 percent in 1999 in the
more rural nonmetro areas while it increased slightly in the more
urban non-metro areas over this period.
Nonmetro West Grows, As Do Its Poor
In 2001, 7.5 million nonmetro people were poor (14.2 percent of
the nonmetro population), as were 25.4 million metro people (11.1
percent) (see “Who’s Poor?”). Nonmetro and
metro poverty rates differ substantially across U.S. regions. In
the Midwest, nonmetro and metro poverty rates differ by less than
1 percentage point. On the other hand, nonmetro poverty is more
than 5 percentage points higher than metro poverty in the South,
where more than 40 percent of the U.S. nonmetro population live.
More than one in six persons in the nonmetro South are poor, while
less than one in eight persons living in the metro South are poor.
Over the last 10 years, the regional pattern of nonmetro poverty
has changed significantly. At the beginning of the 1990s, nonmetro
poverty in the West, Northeast, and Midwest was at or below 15
percent, while poverty in the South was around 20 percent. Throughout
the rest of the decade, the nonmetro poverty rate declined on average
in the South and Midwest, while the rate remained about the same
in the West and Northeast. By the end of the 1990s, the gap between
the South and the West in the level of nonmetro poverty had significantly
narrowed, and their poverty rates were higher than the rates in
the Northeast and Midwest. The relative deterioration of the economic
well-being of the nonmetro West is noteworthy because its population
grew significantly during most of the 1990s, fed largely by Hispanics.
Between 1990 and 1997, the population of the nonmetro West increased
15.5 percent while the rest of the nonmetro U.S. increased 5.2
percent.
Dimensions of Poverty
Race and Ethnicity Poverty is more prevalent among some racial and ethnic
groups than others. The nonmetro poverty rates in 2001 for non-Hispanic
Blacks (31.4 percent) and Native Americans (28.8 percent) were
more than twice the overall nonmetro poverty rate. The nonmetro–metro
disparity is also greatest for these two groups, with nonmetro
poverty more than 10 percentage points higher. The nonmetro poverty
rate for Hispanics was lower (25.4 percent) than for Blacks or
American Indians, but still more than twice the rate for non-Hispanic
Whites (11.1 percent) in 2001. High Hispanic poverty is particularly
compelling because Hispanics are the fastest growing minority group
in the U.S., increasing 70 percent between 1990 and 2000 in nonmetro
areas. Hispanic population growth results from both high birth
rates and high migration rates into nonmetro areas during the 1990s.
According to the 2000 Census, racial and ethnic minorities comprise
17 percent of nonmetro residents and are growing more dispersed
geographically. Because more than one out of every four nonmetro
Hispanics, Blacks, and Native Americans live in poverty, understanding
racial differences in poverty is becoming increasingly important
in designing nonmetro programs and policies. Federal programs
target assistance to these groups through outreach and community
programs.
For example, the Food Distribution Program on Indian Reservations
provides food to low-income households living on Indian reservations
as well as to Native American families living in designated areas
near reservations. Additionally, the Food Stamp Program distributes
informational brochures in Spanish and 19 other languages.
Family Structure
Family structure has a significant bearing on poverty. Over three-quarters
of all nonmetro families are headed by a married couple, and about 15 percent
are headed by a single female. Nonmetro families headed by a married couple
have the lowest incidence of poverty (7.6 percent), while more than one
out of every three nonmetro persons living in female-headed
families is poor.
In contrast, metro family structure is comprised of more female-headed
families (18 percent) and a lower percentage of married-couple
families (76 percent).
There are certainly many reasons for the differences in poverty
rates by family structure, but one reason for the low poverty rates
for married-couple families is the greater likelihood of having
at least one wage earner. However, that factor alone does not fully
explain the lower incidence of poverty in married-couple families.
When only working families (at least one working adult present)
are considered, female-headed families still have a poverty rate
that is more than four times greater than the poverty rate for
families headed by a married couple.
This stark difference in poverty
rates across these two types of families may even understate
differences in economic well-being.
Consider two families, both with three people, where one family
is headed by a married couple with one child, and the other is
headed by a single mother with two children. Both have approximately
the same poverty threshold. However, the married couple heading
a poor family can share the responsibilities, anxieties, and
scheduling burdens of raising their child under difficult financial
conditions.
The single mother, who needs to care for twice as many children,
will likely bear the difficulties alone or with help from relatives
and friends.
Now assume that no one is employed in either of the
two families. Access to even a low-paying job might well lift
the poor married-couple
family out of poverty, as one adult could work while the other
could tend to the child and housework. In contrast, if the single
mother were to become employed, she would then need to manage
her housework during nonworking hours and incur the costs of child
care during her working hours. The costs of child care might
well
prevent this family from escaping poverty. Hence, programs to
alleviate poverty must be mindful of these circumstances.
Ability To Work
Different age groups require different types of assistance and/or
services. The elderly poor are more likely to need assistance
with nutrition, health
care, and medical expenses, including elder care and medications. Poor
working-age adults are the most likely to benefit from job
training programs, food stamps,
and tax credits. Nonmetro poverty rates for adults and the elderly have
been similar throughout the last decade, and in fact were the
same in 2001 (12.2
percent).
Nonmetro child poverty rates, in contrast, stood at 20.2 percent
in 2001 and eclipsed adult and elderly rates throughout the 1990s.
Federal programs targeted to assist poor children include free
school breakfasts and lunches and larger tax credits for households
with children (the Earned Income Tax Credit). In addition, educational
programs like Head Start are intended to help poor children attain
a quality education and increase the likelihood of high school
graduation and college attendance. Such programs may pay long-term
dividends since children from poor families are less likely to
graduate from high school, and low educational attainment increases
the chance of their remaining poor as adults. In 2002, 41 percent
of poor adults in nonmetro areas had not completed high school,
and only 23 percent had any schooling after high school. In comparison,
only 18 percent of nonmetro adults above the poverty line had
not completed high school while 43 percent had some schooling after
high school.
Depth of Poverty Often Hidden From View
Up to this point, all poor people have been grouped together without
regard to their relative level of poverty. This ignores the fact
that a poor family with income equal to half the poverty line has
more extreme needs than a poor family just a few dollars short
of the poverty line. The latter family is more likely to have sufficient
assets and personal skills to right itself with a modest amount
of assistance. On the other hand, a family living in severe poverty,
with income less than half the poverty line, might require a more
significant infusion of help to acquire work and socialization
skills, child care, and care with daily activities like transportation
in addition to financial assistance. In 2001, 37.7 percent of the
nonmetro poor had incomes less than half the poverty line, versus
41.8 percent of the metro poor.
Another way to examine the relative
well-being of the poor is to measure their average income shortfall
(or the average difference
between income and the poverty line). Since the poverty line is
adjusted by family size, the income shortfall is expressed as a
percentage of each family’s poverty line. In 2001, the nonmetro
poor had an average income shortfall equal to 44.8 percent of the
poverty line, while the average shortfall for the metro poor was
greater at 47.1 percent. This gap persisted throughout the 1990s
and widened at times, suggesting that the metro poor are worse
off on average than the nonmetro poor.
Still, throughout the history of recording poverty rates, the
incidence of nonmetro poverty has been consistently higher than
metro poverty rates. As such, poverty reduction programs and policies
would be well served to include components that focus on nonmetro
areas. An additional focus suggested by data would be on people
living in the South and West, racial/ethnic minorities, children,
and female-headed families.
Who's
Poor?
Any individual with income less than that deemed
sufficient to purchase basic needs of food, shelter,
clothing, and other essential goods and services
is classified as poor. The income necessary to purchase
these basic needs varies by the size and composition
of the household. Official poverty lines or thresholds
are set by the Office of Management and Budget. The
2001 poverty line for an individual under age 65
is $9,214. For a three-person family with one child
and two adults, it is $14,255. For a five-person
family (two adults and three children), the poverty
line is $21,135. Income includes cash income (pretax
income and cash welfare assistance), but excludes
inkind welfare assistance, such as food stamps and
Medicare. Poverty lines are adjusted annually to
correct for inflation.
Comparisons of metro-nonmetro poverty rates pose
some measurement difficulties. For example, U.S.
poverty rates do not adjust for differences in cost
of living across areas. If, as assumed, purchasing
basic needs costs less in nonmetro areas, then the
nonmetro poverty rate would be lower. However, some
costs—such as transportation to work—are
likely to be higher in nonmetro areas. The poverty
line also does not account for access to other "public
goods," such as health care, schooling, and
communication networks. And of course, indicators
of quality of life, such as noise and air pollution,
are altogether overlooked in measures of poverty.