Sen. Bernie Sanders (I-Vt.) in Las Vegas in 2015. (John Locher/AP)

Sen. Bernie Sanders (I-Vt.) wants the federal government to guarantee a job for every American willing and able to work. The proposal sounds compassionate and enlightened, but in practice, it would almost certainly be a disaster. The fact that it’s taken seriously is evidence that many Democrats, like Republicans before them, embrace loony economic agendas that are more public- ­relations gestures than sensible policy.

Just precisely how Sanders’s scheme would work is unclear, because he hasn’t yet submitted detailed legislation. However, the website of the Sanders Institute endorses a job-guarantee plan devised by economists at Bard College’s Levy Economics Institute. This suggests how a job guarantee might function.

Under their plan, anyone needing a job could get one at a uniform wage of $15 an hour, plus health insurance (probably Medicare) and other benefits (importantly: child care). When fully deployed, the program would create 15 million public-service jobs, estimate the economists. This would be huge: about five times the number of existing federal jobs (2.8 million) and triple the number of state government jobs (5 million).

Although the federal government would pay the costs, the program would be administered by states, localities and nonprofit organizations, which would design jobs and enroll beneficiaries. Some jobs mentioned by the economists: cleaning up vacant properties; overseeing programs for new mothers and at-risk youths; tree planting; and weatherizing homes.

To be sure, there is a real problem here. Even when reported unemployment is low — as now, at 3.9 percent — “millions of Americans remain unemployed or underemployed.” They often have poor skills, wrestle with drug or alcohol problems, or are so discouraged that they’ve dropped out of the labor force. The job guarantee’s appeal is obvious. A recent Civis Analytics poll for the Nation magazine found 52 percent of respondents in favor.

The trouble is that there is a vast gap between rhetoric and reality. Indeed, some leftish commentators recognize this. Here’s Kevin Drum, a blogger for Mother Jones:

“Even our lefty comrades in social democratic Europe don’t guarantee jobs for everyone. It would cost a fortune; it would massively disrupt the private labor market; it would almost certainly tank productivity; and it’s unlikely in the extreme that the millions of workers in this program could ever be made fully competent at their jobs.”

Many problems are unavoidable. The proposal would add to already swollen federal budget deficits. The Bard economists put the annual cost at about $400 billion. Although some of this might be recaptured from savings from food stamps and other welfare programs, overall spending is likely underestimated.

The reason is Medicare. If it’s provided for those making $15 an hour, there will be pressures to provide it for most workers. Otherwise, uncovered workers might stage a political rebellion or switch from today’s low-paying private-sector jobs to the better-paid public-service jobs, as the Wall Street Journal’s Greg Ip notes. The same logic applies to child-care subsidies.

Then there’s inflation. The extra spending and higher wages might push prices upward. The Bard economists profess to be unworried — mainly because their economic “model” predicts a negligible inflation effect. But models are often unreliable, and the Federal Reserve is unlikely to be so complacent.

Other practical problems loom. On his always useful and strictly nonpartisan blog, Conversable Economist, Timothy Taylor poses difficult questions.

Does the federal government have the managerial competence to oversee the creation of so many jobs? Taylor is skeptical. (The 15 million added jobs would equal about 1 in 10 existing jobs.)

Is there a skills mismatch between what the jobless can do and what actually needs doing? Probably. (Remember: The candidates for the public-sector jobs are among the least-skilled workers.)

Is there a similar geographic mismatch — say, the jobless are in Michigan and the jobs are in Arizona? This, too, seems probable.

Can the new workers be disciplined? Good question. “The problem with a job ‘guarantee’ is that you can’t fire people,” notes Taylor.

Finally, would state and local governments substitute federally funded jobs for existing jobs that are supported by local taxes? This seems inevitable. It, too, would limit the overall effect on employment.

Americans are suckers for great crusades that make the world safe for the pursuit of happiness. In this context, Sanders’s job guarantee seems a masterstroke. The chronically unemployed need jobs; and states and localities have large unmet needs for public and quasi- ­public services. It’s a bargain made in heaven.

Back here on Earth, the collaboration looks less noble. The object is to appear good and buy political support. Many of the suggested jobs seem best described as make-work. The irony is that, by assigning government tasks likely to fail, the advocates of activist government bring government into disrepute.

Read more from Robert Samuelson’s archive.