During several of its meetings, the Commission heard testimony from many citizens who currently live in poverty and many who have received public assistance in the past. Most were alarmed at the number of myths about the poor that presently pervade public discourse about welfare reform. For example, while those receiving government assistance are often considered "lazy," AFDC recipients who testified before the Commission expressed a strong willingness to work, but reported that they either could not find a job with "livable" wages or could not find a job at all. Drawing largely on the demographic data presented in the previous chapter, this chapter reviews and challenges some common myths, both about the measurement of poverty in Kentucky and about the individuals who are poor.

Even before discussing the characteristics of the poor in Kentucky, it is important to understand that the definition and measurement of poverty is a complex issue with serious policy consequences. Too frequently, policy decisions reflect an over-reliance on measures of poverty which give an incomplete description of the poor.

Finding: The county poverty rate tends to highlight areas with large proportions of poor people among small populations. Use of the poverty rate as the sole measure of poverty obscures the fact that many poor live in counties that are also home to a larger number of nonpoor. Reliance on the county poverty rate alone to allocate government resources implies that the poor who live among other poor are of greater concern than the poor who live amid relative affluence.

Finding: If the number of poor in each county is considered, then it is clear that the greatest numbers of poor Kentuckians are found in the urban counties, particularly Jefferson, Fayette, Boone, Kenton, and Campbell Counties. At the subcounty (or block-group) level, small areas of concentrated poverty exist throughout the state. A measure of poverty which reflects both the poverty rate and the number of poor in small areas indicates that Jefferson and Fayette Counties, together, account for the same number of block groups with an overrepresentation of at least 400 people in poverty as do all of the counties of eastern Kentucky combined. (See Chapter II.)

Finding: As defined by the Bureau of the Census, the federal poverty line is a useful, but limited, definition of poverty. This guideline has been criticized as both an underestimate and an overestimate of poverty. The value of the federal poverty line, however, lies in its being the only widely available measure of poverty which is consistent across time, consistent across geographic areas and available at a relatively disaggregated level. In terms of understanding the demographic and geographic characteristics of the poor in Kentucky, no superior measure is available. It should be understood, however, that having an income below the federal poverty line does not fully qualify individuals for poverty programs. Other eligibility criteria come into play in determining which low-income individuals receive government assistance. So it should not be assumed that the characteristics of the poor, as defined by the federal poverty line, are the same as the characteristics of those who receive government assistance.

Finding: Like the problem of poverty itself, federal poverty data are complex and sometimes difficult to interpret. For example, the statistic that 47 percent of the poor own (or are buying) their own homes, while accurate, is easily misinterpreted without further examination. First, it might be of interest to know that 31 percent of poor homeowners are elderly. While assets such as a home could be converted into income, state policymakers have generally been unwilling to force the elderly out of their homes as a qualification for government assistance. Of those poor homeowners who are younger than 65, the median home value is less than $20,000, and nearly one-third of the homes owned are mobile homes. For the poor who are paying a mortgage, the median monthly payment is $187, compared to the median monthly rental payment of $245 for the poor who are renters. Thus, the conclusion that home ownership statistics indicate that poverty is overestimated is likely a misinterpretation of the data.

Common misconceptions persist regarding those people living in poverty as well. Policies which are based on such misconceptions are less likely to assist the poor in moving out of poverty than policies which more accurately reflect their true circumstances.

Finding: The characteristics of the poor vary among regions of the state. The poor in eastern Kentucky are more likely to be adults, more likely to be living in married-couple families, and less likely to have a high school diploma than the poor in general. The poor in urban areas are more likely to have a high school diploma and more likely to be black than the poor in general. The elderly poor tend to be overrepresented in south-central and far-western Kentucky. Also, the poor are not uniformly poor and, in many cases, not substantially different from the near-poor. Over half of all poor children live in families with incomes 50 percent of the poverty line or below, and nearly half of all children in Kentucky live in families with incomes 200 percent of the poverty line or below. It is likely that the problems facing each of these subgroups are different.

Finding: Research indicates that poverty is strongly associated with characteristics, such as race, gender, and age, that individuals cannot change. While programs designed to help individuals change characteristics that are alterable, such as educational attainment or work-preparedness, can affect the prospects of poor individuals, such programs alone may not significantly affect the way poverty is distributed among the population.

Finding: State programs other than poverty programs can also be categorized as antipoverty programs. Poverty is a multifaceted problem which requires a multi-level policy response. Programs that merely target the characteristics of individuals in poverty are unlikely to fully succeed, since individual demographic characteristics were shown to account for only 34 percent of the variation in family income. At some point, the ability of the local economy to provide jobs will determine the number of people who have them. However, it is clear that jobs and income are not distributed on a random basis and that the characteristics of individuals determine their share of the distribution. It is also clear that jobs and income are distributed partly on the basis of characteristics that individuals cannot change--such as race and sex. Additionally, poverty has been shown to be cyclical--those poor as children are more likely to be poor as adults and to raise poor children of their own. Also, poverty is associated with problems, such as crime and drug abuse, that exacerbate the problems of the poor and impinge negatively on the nonpoor as well. Therefore, it is believed that policy responses to poverty can include programs besides those which provide direct assistance to the poor. Examples are those which target economic development efforts; those which reduce the effects of poverty on children; those which affect the distribution of jobs and income on the basis of characteristics which individuals cannot change (race, gender, and age); and those which attempt to address problems associated with poverty (such as crime and substance abuse).

Finding: While the addition of jobs in a community may act to reduce the community's poverty rate, it is not necessarily true that the addition of jobs will likewise reduce the number of poor individuals in that community. If the new jobs are taken by in-migrants to the community, then the situation of the individual poor will remain unchanged. Analysis of the 1990 Census data indicated that people who moved within Kentucky and across state borders had significantly more education and were much more likely to have worked in the previous year than people who had not moved. Therefore, it is not unreasonable to expect that those who have more marketable work skills might be drawn to an area where new jobs were being created. Those already living in an area who are poor may not have the education or skills to compete with in-migrants, and may therefore be unable to benefit from the economic growth of the community.

Finding: Nearly 70 percent of adults living with poor children live in families with one or two children and 90 percent live in families with three or fewer children. Thus, large families account for a very small percentage of poor families in Kentucky.

Finding: Over half (51%) of poor children live in married-couple families. This is significantly less, however, than the proportion of nonpoor children (87%).

Finding: Nearly 60 percent of poor children are urban. This is a significantly greater proportion than among nonpoor children, of whom 52 percent are urban. This is also in contrast to the adults who live with poor children, only 34 percent of whom are urban.

Finding: Only one fourth of the adults living with children in poverty reported receiving any income from public assistance, according to the 1990 Census. The source of income reported by the largest proportion of these adults was wages and salary (42%).

Finding: Poor adults without children are no more likely to be women than men. In contrast, 61 percent of the poor adults with children are women. This would appear to indicate that the overrepresentation of women in poverty is at least partly explained by the fact that they are more likely than men to be single adults bearing responsibility for children in a family. Another fact which can partially explain the overrepresentation of women in poverty is that 72 percent of the poor elderly are women. Thus, among women, being poor is more likely to be associated with having children or being elderly.

Finding: Only 33,000, or 6 percent, of adults who work more than 75 percent of full-time annual hours live in families with incomes below the poverty level. Larger family size was a common characteristic of many in this group. For most adults, the ability to find and keep a full-time job was sufficient to keep a family above the federal poverty line.

Finding: A summary of demographic characteristics is instructive in showing who is poor, but is not sufficient to explain why certain individuals are poor. Together, the set of demographic variables available from the 1990 Census explained only 34 percent of the variation in family income. Analysis which incorporates additional data, both on the non-demographic characteristics of individuals and on the characteristics of the communities in which they live, is necessary to begin to understand why particular individuals, in particular situations, appear to be unable to obtain adequate financial resources.