In a post last week, Jessica Flanigan takes me to task for my opposition to a universal basic income. Because I worry that a UBI would further encourage mass idleness, a serious and worsening social blight among the less educated and less skilled, I favor instead social policies that promote engagement in the work force – in particular, through wage subsidies for low-skill work. Flanigan says this makes me a paternalist.

There are serious reasons for favoring a UBI, and I applaud Flanigan for raising important issues. But I don’t think the paternalism charge really gets us anywhere. After all, the purpose of both a UBI and wage subsidies is to help people who are failing to support themselves adequately. In one sense, then, both policies are paternalistic, since in both cases the state is assuming a paternal role of providing for dependents. Viewed from another angle, though, neither policy is properly considered paternalistic. Paternalism, after all, is about reducing people’s choices for their own good. But either a UBI or wage subsidies would expand the choices of their intended beneficiaries relative to what they would be in the absence of any government provision at all.

Nevertheless, it is true that the choices of UBI recipients are less constrained than those of workers who receive wage subsidies. With a UBI, you get a check every month no matter what, whereas to benefit from wage subsidies you have to get a job.

The great virtue of a UBI is its directness and simplicity: people need help, so just give them money. Don’t worry about providing food stamps, public housing, job training, etc. – instead just cut a check and let people figure out for themselves how best to use the money. As the cleanest social policy option, the UBI sets an appropriate benchmark for judging any alternative form of assistance. Supporters of any program of specific government-provided benefits – including wage subsidies – need to be able to show that such in-kind aid helps its beneficiaries more than simply writing checks for the equivalent amount of money.

I think a good case can be made that a UBI would be more helpful to the disadvantaged than the patchwork of frequently intrusive, infantilizing, bureaucratic, and wasteful means-tested programs that presently constitutes the American social safety net. So if I could wave a magic wand and replace the policy status quo with a UBI, I would do so. That said, my reading of the available evidence convinces me that a social policy that channels benefits through work and thereby encourages paid employment has important advantages over a UBI in helping the disadvantaged to live full, happy, productive, and rewarding lives.

What evidence? Let’s start with the well-established finding that unemployment has major negative effects on well-being, including both mental and physical health. And the effects are remarkably persistent. A study using German panel data examined changes in reported life satisfaction after marriage, divorce, birth of a child, death of a spouse, layoff, and unemployment. All had predictable effects in the short term, but for five of the six the effect generally wore off with time: the joy of having a new baby subsided, while the pain of a loved one’s death gradually faded. The exception was unemployment: even after five years, the researchers found little evidence of adaptation.

Evidence even more directly on point comes from the experience of welfare reform – specifically, the imposition of work requirements on recipients of public assistance. Interestingly, studies of the economic consequences of reform showed little or no change in recipients’ material well-being. But a pair of studies found a positive impact on single mothers’ happiness as a result of moving off welfare and finding work.

Flanigan is certainly correct that it’s possible to have an enjoyable and satisfying life without working for pay. Employment’s psychic benefits come from engaging us in challenges to overcome, encouraging us to develop and realize our inborn talents, and involving us in projects and purposes larger than ourselves. But we can obtain these benefits just as well through hobbies, volunteer work, and family life. And indeed, there is a real tension between the demands of a job and these other pathways to happiness, as all of us who struggle for that elusive work-life balance can attest.

So you might think that not having to work would free people to spend more time on these other, potentially more rewarding activities. But life doesn’t seem to work that way. Consider the most recent results from the American Time Use Survey, compiled annually by the Bureau of Labor Statistics. In 2013, employed men averaged 6.43 hours a day on work and related activities (like commuting). So how did men without jobs fill up all that free time? Well, compared to employed men they spent 19 extra minutes a day on housework, 11 more minutes on socializing, 9 more minutes on exercise and recreation, 8 more minutes on childcare, and 6 more minutes on organizational, civic, and religious activities. The really dramatic differences in time use, though, came in two areas: jobless men spent an extra hour sleeping (for a total of 9.25 hours a day!) and two extra hours watching TV (4.05 hours a day!). The evidence is quite clear: people who don’t work can’t be counted on to fill that void with other forms of productive, engaged, goal-oriented activity.

Yes, there are plenty of happy students and stay-at-home parents, and retirement apparently improves well-being. You don’t need a paycheck to thrive. But for most working-age people, paid employment is the most reliable path to commitment, engagement, and a sense of purpose. For most people, joblessness means not only a lack of income, but also lack of status, lack of identity, and lack of direction. It is the path, not to nonpecuniary forms of fulfillment, but to anomie and despair.

Over the past few decades, there has been a steady deterioration in American men’s commitment to work. For so-called prime-age males aged 25-54, the labor force participation rate has fallen from 96 percent in 1970 to 88 percent today. This drop-off in participation is concentrated among the less educated and less skilled. Among all adult men in 2010, the labor force participation rate for college grads was 81 percent – compared to 71 percent for high school grads, and only 59 percent for high school dropouts.

What is going on? The emergence of the postindustrial information economy is reducing relative demand for, and the relative wages of, less skilled workers. Meanwhile, eligibility for the dole has expanded (especially disability insurance) while the cultural stigma against idleness has faded. This pincer movement is squeezing less skilled men out of the work force.

And in turn, the reduced availability of “marriageable” (i.e., gainfully employed) men is contributing to family breakdown at the bottom of the socioeconomic scale. In 2011, 87 percent of kids who have a parent with a college degree lived with both of their parents – compared to only 53 percent of kids of high school grads, and 47 percent of the kids of high school dropouts. Unstable single-parent families can then be expected to produce another generation of unmarriageable less skilled men, thus perpetuating a vicious circle.

The rise of mass joblessness among the less skilled is a catastrophe, plain and simple. Work and family, the two great cornerstones of life satisfaction, are both under assault, and declining commitment to one is feeding declining commitment to the other.

Under these circumstances, a UBI cannot be recommended as sound social policy. The great challenge at present is to arrest and reverse the slide of less skilled Americans into a permanent underclass – even as automation and globalization continue to marginalize the role and value of low-skill work. But as the celebrated negative income tax experiments of the late 1960s and early 1970s made clear, unconditional income support reduces labor supply. Perhaps not dramatically, but still the impact is going in the wrong direction. By contrast, wage subsidies in the form of graduated payments to employers of low-skill workers can increase the attractiveness of work and boost labor force participation.

It’s entirely possible that the continued progress of automation will eventually make paid employment the exception rather than the rule. We must hope that by then we are up to facing what Keynes called man’s “permanent problem – how to use his freedom from pressing economic cares, how to occupy the leisure, which compound interest and science will have won for him, to live wisely and agreeably and well.”

But we’re not ready yet. We cannot maintain our current living standards without a high degree of labor participation, and against that backdrop mass joblessness is a recipe for dysfunction and misery. We may someday enjoy a post-work society of productive, creative leisure, but maintaining and expanding the underclass aren’t the way to get there.