America, you have a workaholism problem.

A Gallup poll that made the rounds in August of last year showed that half of full-time workers in the U.S. clocked in more than 40 hours a week. Almost 40 percent logged over 50 hours, while only 8 percent worked less than 40 hours a week.

The trend is actually worse among salaried workers: 50 percent work over 50 hours a week, and 25 percent work over 60. And they don't get paid more for those extra hours. Back in 1975, over 65 percent of those salaried workers would have qualified for overtime pay under federal law. But as it stands today, it's 11 percent.

Here's the fun part: It also looks like a fair number of us may be faking that workaholism. The biggest increase in long hours has actually been among elite workers. And a recent study by a Boston University business professor, looking at an unidentified high-powered consulting firm, found that 31 percent of the men and 11 percent of the women managed to moderate their work schedule while still looking like they were cranking out 80 or 90 hours a week.

They rejiggered their client load so it was mainly local. They didn't tell anyone when they skipped work to spend time with family. And they even arranged informal networks with colleagues to cover for one another.

As Neil Irwin reported inThe New York Times, they got away with it. The "fakers" didn't get worse performance reviews or less income in comparison to the actual workaholics at the firm. But the employees who actually requested for lighter or more flexible schedules from the higher-ups were punished for it. (There was a heavy gender component here, with women and mothers doing most of the explicit asking.)

"Heavy workloads may be more about signaling devotion to a firm than really being more productive," Irwin mused. "The person working 80 hours isn't necessarily serving clients any better than the person working 50." Shane Ferro at Business Insider was even more blunt: "The business world isn't actually very good at rewarding productivity (or measuring it for that matter)."

It's impossible right now to know how widespread the practice is. But Erin Reid, the professor who conducted the study, has collected anecdotal evidence since its publication that this sort of thing goes on at lots of companies in lots of industries.

Okay, you might say, so Americans have some crazy cultural predilections when it comes to work-life balance, and some are beating the system. Good for them, what else is new?

But this cuts way deeper than that.

The first assumption at work here is another popular one in economic theory: that human beings are rational economic actors, i.e. they know how to get what they want (higher income, more productivity from their workforce, etc.) and can reliably identify what they need to do to achieve those ends. That undergirds the marginal product of labor theory, which states that workers are paid for the added value they bring to their firm. Which of course requires the a priori assumption that employers are rational actors who can correctly identify who is adding what value to their firm and how much, and can distribute rewards from the firm's revenue accordingly.

As Seth Ackerman and Mike Beggs pointed out at Jacobin a while back, the marginal product of labor theory is often a normative idea masquerading as a descriptive one. It's a value-based attempt to justify why the economy works a particular way, rather than just an empirical account of how it works. Conservatives love it for this exact reason, as it serves as a readymade excuse for inequality: Some people are paid way more because they bring more value.

You see where this is going. If Irwin and Ferro are correct, then employers are actually really bad at getting rewards to line up with actual value creation. They're probably really bad at even identifying actual value creation in the first place. In point of fact, research does show that past a certain point, longer hours don't make workers any more productive for their firms, for rather obvious reasons of fatigue and stress and burnout. American employers are literally working their employees harder to no discernible benefit, even in terms of their own self-interest.

So what's going on here? The idea that the revenue firms generate can be broken up into discrete chunks, which can then be attributed to individual actors, is largely bogus. In truth, that revenue is basically one big pot of money that's generated by the cooperative interaction of the firm as a whole. Employers may want to add another worker to that cooperative whole, and that worker may want to be added. But what cut of the pot both are willing to accept to make that happen is heavily dependent on a whole host of internal and structural factors — from social relationships to status to worker bargaining power — and just where that equilibrium winds up can vary enormously across circumstances and times.

Which brings us back to our work-hour fakers. Every firm is just a little miniature society. And if, across all those firms, inequality between workers and their executives is growing; if unions have collapsed, leaving workers to fend for themselves; and if jobs are harder to come by, making each individual worker more desperate; well, then all those little societies are going to become more oligarchical, with workers more desperate and willing to take smaller cuts of the pot of money.

On top of that, if the oligarchs of these little societies are irrational — if they can't tell the difference between actual respectable productivity and the empty social signaling of workaholism as tribal devotion to the firm — then workers are going to have to be equally irrational in order to stay in the oligarchs' good graces.