The unemployment rate in the United States has now dropped to 3.7 percent—lower than it has been in 50 years. But that glowing report card obscures a complicated reality: Real wages are still stagnant; inequality is sky-high; racial gaps persist in the labor market, with unemployment for black workers almost double that of white counterparts; and many parts of the country are still locked out of prosperity. Many older, disabled, less educated, and formerly incarcerated people are not able to participate in the labor force at all. And for many employed at the lower end of the income distribution, wages are abysmally low and working conditions, very unstable.
In other words, many people around the country still desperately need good jobs. And among the fixes gaining traction at the moment, particularly on the left, is the idea of a federal jobs guarantee: that the government could provide work to every American who needs it, either in the public sector or by subsidizing it in private sector.
It’s a big idea with several logistical and conceptual hurdles in its path. But a number of proposals are already injecting some specifics. Senator Cory Booker, a Democrat from New Jersey, for example, has introduced legislation that seeks to offer $15 per-hour jobs in 15 chosen pilot districts across the country. Various left-leaning think tanks have also outlined their own iterations of a federal plan that are geared toward staffing infrastructure and public works projects as well as child and elder care positions.
But what would be the effect of a national jobs guarantee? A new analysis out of the Brookings Institution’s Hamilton Project takes a stab at the answer. It finds that if such a plan offers $15 an hour, it could affect up to 44 million already-employed workers, a maximum of 5.9 million unemployed workers, as well as a portion of the tens of millions of working-age people currently outside the labor force. Although such a “sweeping” program would cost a lot of money—hundreds of billions a year—it could improve employment rates by 2 to 4 percent, depending on how it’s received, the report finds.
“A full national job guarantee at a relatively high wage would be a radical transformation of today’s labor market,” says Jay Shambaugh, director of the Hamilton Project at Brookings. “So it it would not just be something affecting today’s unemployed, it would be affecting a much wider swath of today’s labor market.”
What jumps out in the report is the incredible uncertainty. A federal jobs program could have dramatic effects not just for unemployed individuals but across the U.S. economy—for better and for worse. Much of the impact of this proposal depends on how exactly it is implemented and how it is seen by workers and private companies.
Take the unemployed population, for example, which is the target group of a national jobs guarantee. At 5.9 million, it will likely make up a small share of the total population of workers attracted by a job guarantee. But even then, not all unemployed people will be champing at the bit to sign up. The ones who were at the mid- to high-range of the income spectrum before they lost their jobs may want to keep looking for a comparable opportunity, instead of downgrading to a $15 government job. Those who are in very long spells of unemployment, however, may not have a choice.
How employed workers react to such a program depends, in large part, on the kind of economic ripples a federal jobs guarantee creates. Currently, 27.9 million full-time workers earn below $15 in the country, and would clearly benefit from switching over to a government job. So would 15.9 million part-time workers.
But one potential impact of a jobs guarantee, often hailed by proponents, is that it may compel private companies to raise their wages so they can compete for workers. If that happens, many workers may stay put at their now-higher-paying jobs, lowering the overall dependency on the government jobs program. It’s also possible, however, that this upward pressure on wages could adversely affect some smaller private employers, meaning they won’t be able to afford workers and might even close up shop, leading to job losses in some cases.
There could be many other spillovers. A positive one: More money in the pockets of people who didn’t have it before could mean more spending in the economy. That, in turn, could mean more jobs. On the flip side: More jobs could mean more people start dropping out of high school and college to work. This decrease in what economists call “human capital investments” could affect long-term economic growth.
If there’s a stigma attached to a federal jobs program, workers who take up these jobs may have difficulty re-entering the private market. Would that create a permanent class of government-reliant workers? It’s not entirely clear. And what would a jobs guarantee mean for efficiency? Steering workers toward jobs that they’re not necessarily best suited to may turn out to be wasteful (although it’s “totally fair to question whether the market [currently] allocates optimally”—especially at the bottom end of the spectrum right now, Shambaugh says.)
Then, there’s also the fact that workers are humans—not perfect economic actors. They may not always go for the option that earns them maximum wages. Part-time workers would trade off lower wages for more flexible hours, for example, if they have to take care of a child or an elderly relative. Full-time workers may forgo a higher wage if they enjoy their current job.
In other words, a national jobs proposal could trigger a number of economic effects in opposite directions, and those effects may intersect with the economic realities and preferences of Americans in ways that are not currently certain. What is abundantly certain from this analysis is that the wage offered by the job guarantee program matters. A lot. A government job at around $10 an hour would affect only about five million workers, dramatically fewer than a program offering $5 more per hour.
“What it comes down to is: The lower the wage you set the more it is affecting people who are currently out of a job as opposed to affecting how the labor market works, and really depends on what you’re trying to do. Are you trying to deal with the fact that some people aren’t employed or are you trying to actually really truly reshape the low end of the wage market in the United States?” Shambaugh said. “That second one is a much bigger task and a much more uncertain task. The first one is a little bit more narrow and targeted in scope … and effect.”
Some plans out there try to address certain concerns. Shambaugh cites the recent jobs guarantee proposal by his colleague David Neumark, for example, in which skills development is a central tenet. The goal of this proposal—apart from giving good jobs to those who need them—is to mitigate the losses in human capital investment that may happen as a result of creating additional jobs, and ease the transition of workers back to private industry. In the first phase of his proposal, Neumark recommends a 100 percent subsidy for positions with non-profits that create public goods in depressed areas. In the second phase, the workers would be given jobs in the private sector, subsidized at 50 percent by the government. Both phases focus explicitly on skills-building, so workers are ultimately independent of the government.
It’s a good sign that policymakers and academics are considering what a federal jobs guarantee could look like now—before the effects of automation kick in. But the key is to gather more data, and hone in on a plan that minimizes the side effects and maximizes the benefits.
“I think learning and experimentation are really important aspects of this,” Shambaugh said, “and highly targeted pilot programs might be a very sensible way to start thinking about what can we do to help people at the bottom edges of the labor market.”
This story originally appeared on CityLab, an editorial partner site. Subscribe to CityLab’s newsletters and follow CityLab on Facebook and Twitter.
