States spend billions of dollars every year on economic development incentives and programs—yet many regions and neighborhoods continue to face economic distress. Tim Bartik is a senior economist at the Michigan-based nonprofit W.E. Upjohn Institute for Employment Research and co-director of the Institute’s research initiative on place-based policies; for decades, his research has focused on local labor markets and state and local economic development policies. His new report, “How State Governments Can Target Job Opportunities to Distressed Places,” and an accompanying brief outline how state governments can significantly—and affordably—boost employment rates in distressed places.
This interview has been edited for length and clarity.
Q: What’s the scale of the “distressed areas” problem?
A: About two-fifths of all Americans live in distressed local labor markets, which are multicounty areas with an employment rate—that is, an employment-to-population ratio—of at least 5 percentage points below the maximum employment rates that seem economically feasible. And one-fifth of all Americans live in distressed census tracts—these average 4,000 people and are one definition of a “neighborhood”—with an employment rate more than 5 percentage points below their local labor market’s average. These low employment rates cause numerous social problems: substance abuse, crime, and poorer child development.
Q: How are the needs of distressed local labor markets distinct from those of distressed neighborhoods?
A: Distressed local labor markets would benefit from simply having more jobs. By their very definition, local labor markets—such as metro areas or rural commuting zones—are multicounty regions that encompass most residents’ work commutes. Since many people are able to travel to jobs throughout these areas, plopping jobs down anywhere in a distressed local labor market can provide job opportunities for residents who live throughout the labor market. And research suggests that, in the long term, this strategy does substantively increase the employment rate in the labor market. Some of these benefits will even go to residents of distressed neighborhoods, though not enough to solve their employment problems.
What residents of distressed neighborhoods need is access to jobs throughout the broader labor market. Most jobs in a neighborhood aren’t held by people who live there, and most people don’t work in the same neighborhood where they live. So when we see neighborhoods with particularly low employment rates, it suggests the problem is not a lack of jobs within those communities but rather that residents likely face barriers accessing jobs outside their neighborhoods. This means that simply putting jobs in a distressed neighborhood is neither necessary nor sufficient for increasing the employment rate of its residents. Instead, the best way to achieve this is to improve job access, broadly defined, to include not only better transportation but also job opening information, job training, and child care.
Q: How can states actually create jobs in a local labor market?
A: States should invest in customized business services, such as production advice and worker training programs, which are more cost-effective than typical state policies that stress tax incentives and other cash subsidies as ways to encourage job creation and business investment.
Cash subsidies are tempting because they’re easy to implement and have many political friends. But providing services, which help overcome the real problems impeding job creation and neighborhood job access, is more cost effective. For example, the evidence suggests that the cost-per-job-created for customized business services is less than a third that of handing out cash to create jobs.
Q: Can states do more to ensure that funding for their job creation and access programs targets the areas that need it most?
A: State efforts to geographically target their funding for these programs are limited by politics: The easiest thing to do politically is to provide a similar level of per-resident funding to all local labor markets and neighborhoods—and even try to take credit for job creation in the areas that would have experienced booms without any intervention. To combat these tendencies, states can structure their funding rules so that the amount provided to different areas is based on clear and measurable indicators of residents’ needs. For example, states can allocate dollars to local labor markets based on the gap between a statewide “target” employment rate and the actual employment rate in each local labor market. My report illustrates how an approach like this could successfully provide at least some funding to almost all areas while also ensuring that the most distressed places receive much higher benefits. In other words, it targets need within a nearly universal program.
Q: How much would it cost to implement these policies at a large enough scale to have a meaningful impact?
A: The budget costs of “place-based” polices depend in part on how tightly the aid is targeted to distressed places and how quickly you want to boost these areas’ employment rates. But even the most cost-effective approaches aren’t necessarily cheap. For example, boosting the employment rate by one job will probably cost around $100,000, but this one-time cost of permanently boosting employment is far less than the combined benefits to the employed individual and to society—including social benefits, such as lower rates of substance abuse and crime, and better child development.
To significantly boost the employment rate in distressed places, simulations suggest that the required costs will frequently be in the range of $100 to $300 per resident annually for at least 10 years. This spending level is similar to that of past programs designed to help distressed places, such as the Tennessee Valley Authority and the Appalachian Regional Commission. In my report, I came up with the cost for a comprehensive place-based program: $30 billion per year nationally, which is slightly less than what state governments currently devote to business tax incentives. But more targeting could reduce costs somewhat. For example, if one targeted only the most distressed 15% in the United States, which is around 50 million people, and provided funding at the lower end of the range—say, $100 per person—the annual cost would be about $5 billion. It’s unclear which is more politically feasible: a targeted program that provides at least a modest level of funding to nearly all places, or a tightly targeted program that benefits fewer places but would reduce the overall costs for states.