America’s ‘islands of poverty’ need targeted policies

Ed Finkel | Jun 29, 2018

Sections Economics

US regions converged economically for a few decades after World War II, but this dynamic has stalled, if not reversed, with particularly pernicious effects on the eastern half of the American heartland—roughly between the Mississippi River and the Appalachian Mountains.

This should prompt economists and policy makers to consider the merits of geographically targeted policies, argue Harvard’s Edward L. Glaeser and Lawrence H. Summers, who was the 71st US Treasury secretary, and Harvard PhD candidate Benjamin A. Austin

Economists traditionally have eschewed basing antipoverty policy on geography, holding that it should be directed to poor people rather than poor places, but Austin, Glaeser, and Summers say social problems are increasingly linked to location. In the past 40 years, coastal states have experienced an aggregate 342 percent economic growth and the western heartland 475 percent, they note, but growth has been more sluggish in the eastern heartland, a traditional center of manufacturing, at 187 percent. 

Men aged 25 to 54 have seen the most notable rise in nonemployment in the US during the past few decades, although the impact of this has been wildly disproportionate, the researchers note. (Unemployment refers to people not working but also searching for work, while nonemployment refers to people simply not working.) The nonemployment rate for men in this age range was more than 35 percent in Flint, Michigan, but less than 5 percent in Alexandria, Virginia, according to 2016 figures. In past eras, US residents moved from job-poor to job-rich areas, but higher housing prices, an increase in the geographic segregation of skills, and rising opioid addiction mean “America appears to be evolving into durable islands of wealth and poverty.”

The researchers see “market failures that can be most plausibly addressed at the local level” as the best rationale for location-based policy. A one-size-fits-all employment subsidy might have a large impact in West Virginia, but minimal effect in San Francisco. Targeted infrastructure spending, stronger employment subsidies, and more incentives involving community colleges may be more promising.

Attempting to address poverty by hiring more people to work at larger federal agencies such as the Departments of Defense and Veterans Affairs would be of limited use, despite the positive impact these agencies have on local economies, according to the research. The costs of moving these agency outposts to locations of high joblessness would make increasing employment practical only when military bases or medical facilities otherwise needed to be opened or closed. And other agencies that would be easier to relocate are much smaller so wouldn’t provide the jobs boost, the researchers note.

Targeted infrastructure projects might seem like another way to encourage local economic development. However, the most glaring infrastructure needs aren’t necessarily where nonworking rates are highest, given that demand for infrastructure projects tends to be greatest in large, busy, urban areas that are also major job centers, the researchers note.

Employment subsidies such as the Earned Income Tax Credit (a tax credit tied to income) or public subsidies on top of employee wages can be effective. Prices for goods tend to be lower in areas where a lot of people aren’t working, especially when these subsidies come in the form of a fixed dollar amount rather than a percentage increase. Another approach would involve lowering marginal taxes in areas that need a boost.

Lastly, the researchers considered the viability of place-based education reform. “For example, community colleges could receive bonus payments for admitting students from distressed regions, who are then employed for some number of years post-graduation,” write the researchers, who also support geographically targeted training programs.