All of a sudden, you can’t turn around in progressive D.C. policy circles without bumping into a jobs guarantee proposal. The most prominent version is a bill introduced this week by Sen. Cory Booker (D-N.J.), with heavy hitting  Democratic co-sponsors including Sens. Jeff Merkley (Ore.), Kirsten Gillibrand (N.Y.), Kamala D. Harris (Calif.) and Elizabeth Warren (Mass.). The bill is not for a national job guarantee but for a three-year pilot program in 15 different places, wherein anyone older than 18 would be eligible for a guaranteed job paying at least $15 an hour, with paid family and sick leave and health benefits.

Given the scope of this idea — its depth of intervention in the labor market — giving it a pilot run is the right way to go.

Obviously, the idea isn’t going anywhere in this Congress, but the fact that it has captured the interest of many policymakers on the left suggests that if and when the pendulum swings back, some version of public job creation or subsidy will probably be one the big ideas with which they hit the ground running. With that in mind, and with the admission that I’ve long advocated for variants of this idea (and even helped to implement one during the Great Recession), here’s some Q&A.

With unemployment at a 17-year low, why do we need a job creation program?

First, even as we close in on “full employment,” there are people and places without enough job opportunities (the Booker plan targets those areas for his test cases). Moreover, recent research finds that even long-term jobless people are not moving to opportunity the way they once did, implying the need to bring jobs to people. Second, many of the jobs being created are of low quality, in terms of compensation: From 2016 to 2017, jobs paying less than $13 an hour grew twice as fast of jobs paying more than $45 an hour. Finally, some workers face steep barriers to labor market entry, including discrimination, criminal records, health and/or skill deficits. As such, they are immune to the pull of even a hot labor market and need the training that often accompanies these programs (Booker’s plan includes a two-month, paid training program).

How does it work?

These ideas exist on a continuum with Booker-type plans on one end — I’d call it the left end — and at the other end, far less interventionist programs that provide, for example, small subsidies to private employers to hire disadvantaged workers for a short period.

The Full Employment Project (FEP) that I run at the Center on Budget and Policy Priorities commissioned a paper by Mark Paul, William Darity and Darrick Hamilton for a highly ambitious job guarantee program (CBPP doesn’t endorse this particular idea; it’s one of many our FEP wanted to get in the mix). Their program creates a new national agency, the National Investment Employment Corps, housed in the Labor Department, which provides employment grants to localities “to engage in direct employment projects. These projects should be designed to address community needs and provide socially beneficial goods and services to communities and society at large.” In other words, public jobs maintaining and improving public goods and providing educational and caring services (one problem here is a potential collision with public sector unions).

Advocates of this level of intervention claim, correctly in my view, that if it could be pulled off — the program’s price tag rivals that of the Defense Department — it would totally transform the U.S. labor market, particularly the low end, where poor job quality and exploitative, discriminatory employment relationships abound. Simply put, why would anyone work in a lousy, low-wage job when they could be guaranteed a much better job through the program?

That realization implies one of two outcomes: The low-wage, private sector evaporates, and public jobs grow by the tens of millions, or low-wage employers meet the terms of the government program, meaning higher compensation (and prices). Either way, we’re talking about massive labor market disruption, but it’s the disruption of persistent unemployment, wage/employment discrimination and low job quality.

Now you see why I endorse Booker’s idea for a pilot. We need to see how this works before we consider nationalizing it.

What we do know, however, is that the other end of the continuum works badly. Subsidize someone in a poor-quality job for a few months, and too often, when the subsidy ends, nothing changes. But the record also shows that better programs — longer lasting, with a serious training component tied to actual jobs — have better outcomes. During the Great Recession, the Obama administration implemented a subsidized jobs program through the TANF program, and in some instances, poor workers managed to lastingly overcome steep barriers to work.

Who pays for it?

Notably, Booker’s plan has yet to be scored, but as noted, the Paul et al. plan costs more than $500 billion a year, more than 2 percent of gross domestic product. And if my prediction that more people than they think would flock to these better public alternatives, that number may be low. Something in the middle of the continuum — a program targeting geographical pockets of unemployment where the government subsidizes workers wages for many months, along with robust, job-specific training — might clock in at around $50 billion to $100 billion a year. If even that price tag sounds unrealistic to you, I strongly urge you to consider that the Republican tax cut costs adds two to three times that much to the deficit every year between 2018 and 2025.

What other economic purposes does a jobs program solve?

One important and useful function of a public jobs program is that it creates a funding stream that can ramp up and down with the business cycle. That is, in a recession, with these programs in place across the country, they can quickly scale up, creating a double safety net, one that catches the unemployed, while simultaneously supporting consumer spending.

Are we “fetishizing” work?

The right is talking work requirements (a policy I’ve argued will do much more harm than good); the left, subsidized jobs. Why is so much policy focused on work in the paid labor market? One answer is that U.S. anti-poverty policy is increasingly conditioned on work as a ladder out of poverty. And yet, while minimum wages and wage subsidies go part of the way to improve job quality, given the elevated role of work in poverty reduction, we do far too little to ensure low-income people can get good jobs. In this regard, jobs programs can be viewed as filling an important hole in our policy framework.

So, let the debate proceed, and, once the politics evolve, let the pilots take off. Amid all the wonkery on where we should best settle on the interventionist continuum, let’s not lose sight of how important it is to nurture this development. Democrats are finally contemplating big ideas targeting people who need their help. Sounds simple, I know, but it’s been a long time coming.