Association for Public Policy Analysis & Management

In this paper I consider whether and how poverty measures should reflect rising inequality at the top of the income distribution. While most scholars of poverty in developed countries adopt a relative definition that identifies someone as poor if they lack sufficient resources to participate fully in “mainstream” society, the definition of “mainstream” has remained unchanged at the national median over the last half-century despite a massive upward redistribution of resources. Further, the standard measure is typically calculated for the nation as a whole, but the economic situations of different regions of the United States have diverged massively over the last four decades. Today, a person who is quite poor by the standards of San Francisco might have an income at the median of many rural areas.

I argue, based on studies of consumption and firm behavior, that US economic life is no longer centered on the median person. This is reflected in the sets of goods and services provided by the market, which are increasingly targeted at higher income consumers. I then propose two modifications to the standard relative poverty metric that allow it to reflect the way the economy has shifted due to rising income inequality in general and income divergence between regions of the country in particular.

Trends using the updated measure diverge sharply from those using standard poverty metrics. While the standard relative poverty rate has been fairly stable over time, trends using the updated measure show a dramatic rise in economic exclusion. Further, when updated to account for income differences across the country the new measures show much higher rates of poverty in high income areas.