The Congressional Budget Office’s latest budget projections, released last week, estimated that 2.5 million people would leave the job market as a result of Obamacare. Immediately the two political sides engaged in verbal bickering, with Republicans saying the program would cause 2.5 million to lose their jobs, while Democrats claimed that 2.5 million people choosing leisure over work was a net increase in human welfare. Actually, both sides were wrong. It’s worth examining why, and what the skewed incentives at modest earnings levels mean for our future.

The Democrats are right that the direct loss of jobs due to Obamacare is likely to be fairly limited. Although it imposes substantial costs on some employers, and makes the healthcare system overall less efficient, employers always have the option of restricting pay rises for employees whose healthcare costs have been increased, or of raising their healthcare premiums. There is a likely to be a certain squeezing of hours worked by part-timers, to keep them below the 30-hour week level, but direct job losses should be limited, according to the CBO. And of course, if the number of people with health insurance increases, and to the extent that the population covered by Medicare increases, there will be jobs created in the healthcare system to cover the newly insured people.

Nevertheless, the Republicans are correct that the 2.5 million people whose incentives are so changed by Obamacare that they will choose not to work are a problem not a side-benefit. If they do not work, the 2.5 million people will not contribute to the tax and benefits system, imposing greater costs on the rest of us. The 2.5 million themselves may value increased leisure time sufficiently to give up their income from work, they may receive enough in unemployment, social security and disability benefit that they are little worse off or (without being too cynical about it) they may feel they can earn nearly as much from working “off the books” on odd jobs, landscaping or some other activity for cash, thus avoiding costly interaction with the tax system.

But from society’s point of view, we are much less well-off for the loss of the labor of those 2.5 million people. Their output would presumably have been worth more than their pay, so losing it is a blow to the economy. Further, if they had worked they would normally have contributed, possibly modest payments of income tax, certainly rather less modest payments of payroll tax. Without working, they will contribute nothing in direct tax to the general coffers, though they will still of course pay sales taxes on their purchases and property taxes if they own a home. What’s more, as unemployed they will likely benefit from welfare, disability and other benefits. Thus the scales, which may be close to balanced from the individual points of view of the 2.5 million people themselves, are heavily unbalanced from the point of view of the U.S. economy as a whole and its tax base.

This is one of the reasons the U.S. budget is still so severely out of whack, with a projected deficit of $514 billion in the year to September 30. The labor force participation rate is now 63%, compared with 66.4% at its peak in December 2006. The unemployment rate at 6.6% is only 2.2 percentage points above its December 2006 level, so an additional 3.5 million more people are officially unemployed. However there are an additional 8.4 million people, over and above those 3.5 million, who would have been in the labor force if participation was at its December 2006 level, but who have dropped out of the labor force altogether. Some of those are early-retiring baby boomers, but by no means all of them; participation rates have also declined for young and prime-age workers.

It is thus not surprising that the United States is still running a $500 billion deficit, in spite of substantial tax increases since 2006, a reining back of military spending, and some moderation in the giant increases in domestic spending pushed through by the Democrat-controlled Congress in 2007-10. With 11.9 million fewer people than there should be paying for the costs of government, and not providing economic output, we should expect government to be further from being paid for than it was in 2006.

There are a lot of factors providing disincentives to work for those with modest incomes. From their point of view, if the social “safety net” is sufficiently strong, and especially if the opportunities for working “off the books” are fairly plentiful, there is little incentive to joining the formal economy. That’s why Spanish unemployment among those under 25 is currently 57.7%. Spanish wages are low, the construction business (which pays modest-skill people well) is in the doldrums, the Spanish income tax authorities are not very efficient and the Spanish welfare state is relatively bountiful.

Apart from Obamacare, there are a number of other U.S. programs that provide incentives not to work. Student loans are one example. Contrary to the massive propaganda from the higher education industry, many college qualifications do little to propel even workers with 4-year degrees out of the low-income trap. Degrees in sociology, art history, et cetera et cetera – you can supply you own favorite useless degree subjects here – do little to raise the incomes of their recipients sufficiently to pay for their cost. Of course many graduates even in art history have sufficient other qualities that propel them fairly quickly into the ranks of the affluent, while a degree in the most useless subject from the Ivy League is probably sufficient to keep you off the breadline unless you have a serious alcohol or drugs problem.

Still, there are an awful lot of mid-level students for whom 2014’s levels of student debt weigh them down like an anvil, preventing them from ever rising above the indigent. Stories of graduates repaying student loans until they draw social security are already extant but consider: anybody now close to drawing social security went through college in the 1970s when it was much cheaper. Today’s generation will have a much higher proportion of graduates who will not have paid off student debt by the time they retire in the 2050s. For such people, dropping out of the tax, debt and wage system altogether is an almost irresistible alternative.

Unemployment insurance for too long a period similarly incentivizes non-work, and thus damages the U.S. economy than its already substantial out-of-pocket cost. An unemployed person of modest income-generating capacity may not lose too much by a lengthy period before returning to the workforce, but society loses his output, the profits made through that output and the taxes paid on it. The termination of extended unemployment benefits, harsh though it was in its effect on the long-term unemployed, was a significant boost to the economy, whose effect should be seen in the employment figures in coming months.

Not all state payments to those of modest means are debilitating. The Earned Income Tax Credit, for example, provides an additional supplement to those with jobs earning modest wages. It thereby encourages work rather than discourages it, and to some extent finances itself by the additional output that the wage-earner produces and the taxes thereon. I am also less opposed to the minimum wage than classical economists; minimum-wage jobs are heavily concentrated in sectors such as fast food and hospitality that are local in nature, so there is little job loss if the cost of employees in those sectors is modestly increased. To the extent a higher minimum wage makes it worthwhile for a worker to pay his Obamacare premium and participate in the workforce, it increases output and thus provides more tax revenues.

As I discussed a few weeks ago, there is one technological change that will make this problem much more acute: the advent of fully functional robots, able to take over many modestly skilled jobs in the service sector. It is not enough to replace human output with robot output and provide welfare for the displaced humans. A society in which only 20% of the population is gainfully employed and the remainder lives on welfare is a society divided as was H.G. Wells’ time machine dystopia between Eloi and Morlocks – with the difference that the Morlocks will generally be considerably better off than the useless Eloi. There will still be useful occupations for most people – but probably not especially lucrative ones.

We must consequently construct a system in which welfare is mostly structured like the Earned Income Tax Credit, so that the modest earnings received by those of moderate abilities – producing “long tail” crafts or video games, for example – are supplemented by an additional payment, and work is thus subsidized rather than discouraged. A society of highly productive robots, highly paid robot professionals, artistic and sporting “stars” and doubtless lawyers and modestly paid craftsmen whose earnings are supplemented by the state is tolerable and probably stable. A society in which the majority live on welfare is highly unstable and destined for ultimate collapse.

However voluntary the loss, losing 2.5 million jobs is a major blow to the U.S. economy and to U.S. society, even if those who lose the jobs are individually little worse off. We must design future programs to pull the modestly-skilled into work, not push them out of it. It is a problem that will become increasingly critical to our economic and societal future.

-0-

(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)