Poverty in Appalachia

Applied Research

Community Development Applied Research Seminar Series

Lexington, Kentucky
October 1, 2009

This event was held in partnership with the University of Kentucky’s Center for Poverty Research and featured a special presentation on a study released late last year by the Federal Reserve System and the Brookings Institution titled The Enduring Challenge of Concentrated Poverty in America.

We believe that collaboration between researchers and practitioners benefits their work and our communities. To that end, this seminar aimed to facilitate the dissemination of current research on poverty and development in Appalachia to practitioners, while also providing researchers with reciprocal expertise, knowledge and insight from practitioners. The audience included representatives from governmental, financial and academic institutions, as well as from community-based organizations.


Poverty is back on the rise. In 2008, the poverty rate climbed to 13.2 percent of the U.S. population—or 40 million people—the highest it’s been since 1997. Though substantial attention has been given to the problem of urban poverty, often overlooked is the crisis in rural areas, in particular those across the swath of territory known as Appalachia.

Appalachia stretches from upstate New York to Mississippi, intersecting parts of 13 states. It is 40 percent rural, compared with 20 percent of the United States. It is also very poor when compared with the rest of the United States.

The Federal Reserve Bank of Cleveland, in partnership with the Center for Poverty Research at the University of Kentucky, recently hosted its third Applied Research Seminar of 2009 in Lexington. As with all seminars in this series, academic researchers discussed their findings with on—the—ground practitioners. The languages of the researchers and the practitioners sometimes differ, but their goals are identical: to find effective ways to reduce poverty and raise standards of living for millions of Americans.

James Ziliak—A Research Agenda for Poverty in Appalachia

Ever since President Lyndon Johnson declared a war on poverty in 1964 from Inez, Kentucky, standards of living have improved across the United States—but Appalachia remains stubbornly poor. The persistence of poverty in Appalachia continues to vex policymakers. In this region, earnings are lower and transfers are higher. (The government defines “persistently poor” as a county with a poverty rate exceeding 20 percent since 1970.)

Since 2002 the Center for Poverty Research at the University of Kentucky has conducted research on the causes and consequences of poverty and inequality in America, with a thematic emphasis on the South. A recently added focus of the center, directed by economist Ziliak, is to perform research addressing the “pockets of deep, systemic poverty” in Appalachia. Per capita incomes are consistently $4,000 lower than the national level. Another alarming trend: government assistance to Appalachian residents has grown from about $300 (inflation—adjusted) a month in 1979 to $600 in 2005.

The source of Appalachia’s economic malaise can be traced, at least in part, to the educational attainment levels of its working—age people, Ziliak said. During his term as a visiting scholar for the Clevleand Fed, Ziliak studied examine the evolution of wages of men and women in Appalachia compared to the rest of the U.S. Although more Appalachians are completing high school in the past decade, fewer—compared to the national average—are moving on to college. In a labor market that values higher skills, that’s a recipe for economic decline. The economic return on a high school degree in Appalachia is falling. But even more problematic, however, is that the return on higher degrees in Appalachia have only slightly increased in the past two decades in comparison to non—Appalachia trends. Ziliak’s conjecture is that Appalachia suffers from “missing markets.” It is in a double—jeopardy situation of having lower skill levels and lower returns to skills. The Center for Poverty Research has commissioned scholars to examine the roles of poverty traps, place—based policies, family structure, socioeconomic status, health and human capital on the economic status of Appalachians.

In response to a question, Ziliak rejected the idea of paying people to move as a solution to Appalachia’s poverty problem. Some scholars have floated a version of that idea. Ziliak said he favored a “grow your own” strategy that would stem the tide of out—migration and bring in few more immigrants.

Bruce Weber—Education, Migration, and Local Labor Markets: Pathways out of Rural Poverty?

The textbook explanation for why some people are poor include three sets of causes: 1) personal inadequacies, such as low levels of education or bad health; 2) restricted opportunities, such as a shortage of jobs or barriers to employment; and 3) government policies that reduce incentives to earn money.

Oregon State University economist Weber asks whether migration is a pathway out of poverty for rural households. How do employment and education relate to migration patterns, and how do these factors ultimately affect poverty? And is encouraging rural to urban migration a strategy for poverty reduction? Research shows that people who grow up in rural areas are less likely to migrate than those who didn’t grow up rural. So this raises the question: Do rural people who move to urban areas tend to have lower poverty rates?

Weber used a model to estimate the effect of migration on poverty, taking care that this measured effect would be free of other factors affecting both migration and poverty. For instance, people who migrate from rural to urban areas may be different from those who stay: they may have traits uncaptured by the data (such as motivation) that may also affect their risk of being in poverty. After this careful analysis , Weber found that migration does “not necessarily reduce the chances of remaining poor.

In a sample of 700 rural American adults, only 8 percent moved. Among the factors found to slow migration were homeownership and low levels of education.

Having more education certainly does reduce poverty, but moving in and of itself does not. “If you have an education, you’re more likely to move [out of a rural area, according to Weber’s sample] but you’re not more likely to get out of poverty just by moving,” Weber said.

An ongoing research question is how changes in the social safety net have contributed to the persistence of poverty in Appalachia. Since 1997, cash assistance has diminished as a focus of the social safety net, replaced with more social services tied to work. In rural areas in particular, access to services is a dominant area of anti—poverty strategies. Barriers to services abound in rural areas—such as access to child care or substance abuse counseling — compared with the simplicity of providing cash transfers. These barriers also have an effect on people’s decisions about how much to work.

Stephan Goetz—Migration vs. Communting as Poverty reduction Strategies: Applications of Social Network Analysis to U.S. County Data

Researchers have already spent many years trying to isolate factors that affect the poverty rate. A general consensus is that the existence of big—box retailers puts upward pressure on the poverty rate, while self—employment, social capital, and political competition drive down the poverty rate. Uncertain has been the effect of migration on poverty.

Goetz, an economist with Pennsylvania State University, is interested in the importance of social networks and their influence on poverty rates. His research has tried to measure social networks arising from in and out migration at the county level, and understand their effect on poverty. One working hypothesis is that the greater a county’s “centrality”—its geographic interconnectedness with other neighboring counties—within migration and commuting patterns, the stronger its effect in reducing poverty.

The results have been mixed so far. The number of people migrating into a county has an upward effect on the poverty rate, for example, and so too does the number of people commuting out of a county. Networks do seem to matter, but not always as expected. “One county’s solution is another’s problem,” Goetz said.

Bruce Weinberg—Social Interactions and Poverty

Social interactions models are a way to think about how people’s behaviors and outcomes are affected by the people around them. Groups influence behaviors by showing what’s appropriate (normative behaviors), by information transmission, and by knowledge spillovers. Weinberg, an economist with Ohio State University, noted that selection bias is important in social interactions research. To be able to measure the effect of neighbors and place on an individual, it is necessary to account for the fact that individuals may to some extent self select into groups and places.

The degree to which social groups affect people’s behavior has important anti—poverty policy implications. For instance, one would expect that helping people move out from neighborhoods of concentrated poverty to higher income neighborhoods may increase their chances of coming out of poverty. If so, how much of this effect is due to the change in social networks they experience? One way to overcome selection problems in order to measure social interactions is by conducting experiments where people are randomly assigned into different neighborhoods. An example of such experiment in urban areas has been the Moving to Opportunity (MTO) program. Results from this experiment point to small social effects, however. Some of Weinberg’s latest research provides a possible explanation, suggesting that once assigned to new groups, people select within groups. More to the point, larger groups tend to facilitate sorting, Weinberg said.

While there has been substantial application of social interactions models on urban poverty research, there hasn’t been as much attention to the role of social interactions in rural poverty. But if Weinberg’s findings hold, the implication for rural settings may be that it’s important to think about how people interact once they’re inside a group—and that sorting within groups may be more difficult in smaller, rural communities. Perhaps poverty persists because people never learn behaviors that would help them move up the income ladder. “As a consequence, you’d expect that in smaller groups, people’s behaviors are much more affected by the average person in the community,” Weinberg said. However, this “low population” effect may be partially offset by a “low density” effect, wherein people’s ability to keep an eye on each other and mimic behaviors is more difficult.

The upshot may be that policy efforts should be focused on helping people within communities, as opposed to attempting to move people out of their groups, Weinberg said. A program that offers people housing vouchers in other communities may not be effective if people simply re—segregate into familiar social groups, for example.

Alan Berube—The Enduring Challenge of Concentrated Poverty in America

In 2008 the Brookings Institution and the Federal Reserve System reported on the factors associated with concentrated poverty across the United States in The Enduring Challenge of Concentrated Poverty in America. Berube, senior fellow and research director with the Brookings Institution Metropolitan Policy Program, laid out some of the common challenges in Appalachian and other high—poverty communities. First, they tend to be isolated geographically. Second, the region’s traditional industries have declined, taking jobs with them. Third, education levels are well below national averages, lowering the stock of human capital. Finally, local governments and organizations often have insufficient capacity to handle the enormous challenges in their communities.

Unique to Appalachian communities with high poverty are high rates of poor health. In a case study of Martin County, Ky., 40 percent of working age men reported a disability. Obesity is rampant, and the economic depression in areas of rural poverty seems to be causing physical depression in residents. Additionally, housing is substandard in many rural poor areas, even as most people are homeowners. Finally, rural areas must overcome their own histories. In Martin County, for example, 80 percent of residents are natives of either Kentucky or West Virginia. People do move, but they tend to go from one unstable situation to the next, Berube said.

Moderated Discussion (Led by Justin Maxson, president, Mountain Association for Community Economic Development

Maxson urged the closing of the gap between academic research and the needs and understandings of practitioners. He asked seminar attendees for a list of ideas to be included in a research agenda for attacking poverty in Appalachia. Ideas offered for examination included: