Tyler Cowen is not optimistic about the future of the American economy. And that is reason to worry, for the man is usually spot-on. Cowen’s 2011 book on the state of the economy, the popular Great Stagnation, set forth why the economic growth we saw in the decades prior to the 2008–2009 recession is unlikely to return anytime soon.

Average Is Over picks up on that cheery message and suggests that even if steady 3 percent annual growth in U.S. gross domestic product were to magically reappear, it probably wouldn’t help Americans all that much anyway, unless they happen to be members of the intellectual elite whose skills are enhanced—and not replaced—by computers. That class of Americans is smaller than what most people seem to realize, Cowen avers.

Here’s hoping he isn’t as prescient as he was with Great Stagnation.

In reality, the new book isn’t a litany of solely ill tidings. It focuses on explaining how he sees the world changing and how his readers can make sure that they’re not the ones punished by the economy’s transformation. But taken together with Great Stagnation, Average Is Over seems to say that only a few of us have careers that will survive the culmination of the information technology revolution. The college degrees that so many of us assumed would keep us in good stead 20 years ago may be millstones around our necks, it turns out.

Or maybe not. Cowen is taking a trend that’s been occurring for decades—an increasing demand by employers for skilled, talented workers—and positing that this “taste for talent” has become much more refined of late, meaning that only some skilled workers will be in strong demand in the future. Despite his marshaling of an impressive litany of anecdotes manifesting this trend, I’m not yet convinced that his dire thesis is right. After all, it’s a thesis that’s been advanced in previous downturns in the business cycle, but then dismissed when those cycles turned upward.

Job-displacing technology isn’t new /​My first college class—back in 1983—began with my economics professor asking the class why a majority of people managing shoe stores had college degrees when, a generation before, degreed shoe-store managers were virtually unheard of. He argued that the job hadn’t changed all that much over the intervening 30 years and that the high school graduates of the 1950s managed to hire workers, keep the books, and sell wingtips just fine. What skills did four years of college confer on today’s shoe store manager to make him more productive?

Yet the use of degreed employees to fill jobs formerly performed well by high school diploma holders has continued unabated in the era of the personal computer. Today nearly all retail managers have college degrees. Is the post-computer trend different than what my professor discussed? I think Cowen would say that these shoe-store managers are simply victims of a stagnant labor market and of technological transformations that have rendered their training worthless. Worse, they’ll soon be joined by more of their baccalaureate-holding ilk, doomed to be under-employed for the duration of their career.

I have a qualm with this supposedly grim picture: who’s to say what is and isn’t an occupation appropriate for a college-educated worker?

Much of what has happened in the last five years is that some college graduates have taken jobs that, in earlier decades, were taken by their less-educated brethren. It’s unclear whether this trend will reverse—not because higher-skill jobs in this economy never fully absorb the ranks of the educated, but because the owners of such businesses may conclude that college graduates are sufficiently more productive and thus worth paying the higher wages necessary to keep them around. In other words, these jobs become higher-skill jobs and employers are willing to pay the premium for a college graduate when a better labor market imposes one.

A few years ago, a friend of mine abandoned his nascent Chicago business career and returned to his small-town home to take care of a dying parent. As part of the move, he took a low-paying job near his family home for a relative pittance, managing the loading dock at a big box store. After his parent passed away the following year, he made plans to return to Chicago and resume his better-paying career, but his manager realized that having an “overeducated” worker had worked out quite well. Negotiations ensued and a formerly blue-collar job was turned into something that paid akin to a white-collar salary. My friend remained with the company. So in Cowen’s world, is this outcome bad or good?

Underachieving college grads / Of course, there are plenty of sad stories about college grads “slumming it” in today’s labor force. The head of physical plant maintenance for Indiana University, located in the idyllic slacker paradise of Bloomington, Ind., once told me that he could fill every janitorial job on campus with a college graduate and every supervisory position with someone with an advanced degree and still have plenty of applicants to choose from. That was 20 years ago, incidentally.

Cowen believes that any job formerly done by a non-degreed worker but now filled by a degreed worker is an unfortunate effect of technological change.

The plight of the overeducated is not a new concern. When Arjo Klamer, while preparing his book Conversations with Economists, asked Nobel laureate Robert Lucas his thoughts on the apparent under-employment of an accountant who drove the cab that had delivered Klamer to Lucas’s office, Lucas remarked that if he was driving a cab he was a cab driver, not an accountant. It’s facile to suggest that the growth in the number and variety of jobs for which a college degree is now preferred is solely due to a faltering economy that can’t make “suitable” use of such well-educated workers. Cowen is correct that there are both cyclical and structural forces pushing college graduates into a wider variety of jobs; however, we’re just arguing over the relative weights. I think cyclical factors dominated while Cowen suggests the opposite.

Cowen’s perspective in the book is that any job formerly done by a non-degreed worker but now filled by someone with a college degree is an unfortunate side-effect of technological change. I’m not so sure that it is a side-effect or that it is necessarily unfortunate.

Back in my hometown near Peoria, Ill., one of my childhood friends told me, from kindergarten onward, that he aspired to be a sheet metal worker like his father. When he graduated high school, however, Caterpillar—the heavy equipment manufacturer that employed my friend’s father (as well as the father of every other kid in my kindergarten class)—had fallen on hard times and had no jobs available on the line for newbies. My friend spent the next two decades working—earning workmen’s wages—for a Caterpillar supplier while taking night classes at the nearby junior college. Twenty years after high school, Caterpillar offered him what was essentially his father’s job, but it had morphed into something entirely different over the interim. Instead of crafting sheet metal by standing on the assembly line wearing jeans and a t‑shirt, my classmate worked in an office at a computer screen, wore a collared shirt, and spent his nights finishing up an engineering degree.

The high union wages of my youth led to absurd outcomes, such as college engineering students postponing completing their degrees because they had gotten work on the assembly line, and then later discovering that the jobs they found with their newly minted engineering degrees earned them less than union workers. For three decades I lived in a time and place where unskilled blue-collar workers made more, on average, than their typical college-educated brethren. It is slightly amazing that this wage disparity lasted as long as it did; had it endured much longer (it took a long and bitter strike-cum-lockout in the early 1990s to end it), Caterpillar likely would have endured the same fate as Chrysler and General Motors and been whittled down to size during the Great Recession.

Productivity over the business cycle /​In the book, Cowen is at his best when exploring precisely how computers can accentuate the skills of workers to increase productivity. His argument belies intuition at first, but he marshals considerable evidence to both explain and prove his thesis, especially when discussing (a little too in-depth, in fact) the advent of “freestyle chess,” which he submits as an analogy for the new new economy.

He’s on less firm ground, however, when making his argument that the Great Recession suddenly rendered “zero marginal product” workers unemployable forever. For instance, he interprets the jump in productivity in 2008, when firms began shedding employees wholesale, as evidence that the laid-off workers were completely unproductive (or actually pulled productivity down by their mere presence). He claims that these workers won’t be getting their jobs back now that firms are newly constrained to be profit-maximizers—which apparently wasn’t the case before.

However, I think there’s a different story for the productivity spike: When an economic downturn begins, managers presumably don’t know whether the decline in demand for their output is a temporary or long-term phenomenon. Since it is costly both to lay off and rehire a worker, the managers’ first response to a diminution in demand usually isn’t to lay off workers; rather, the remaining workload gets spread across more people and periods of idleness ensue. We call this “labor hoarding.” This, I submit, was happening in 2007–2008. Less output spread over the same number of people results in faltering productivity. When it became clear that the recession was real and likely to be severe and long-lasting, surplus workers were laid off. Maybe the laid-off were zero productivity workers, but it’s also possible (and practically speaking, far more likely) that they are capable workers who are no longer being hoarded during an enduring period of decreased demand.

When a company lays off one-fourth of all workers while producing the same amount of output, its productivity skyrockets. But that doesn’t mean the now-dismissed workers had low capability. Because of labor hoarding, worker productivity is a poor measure of worker capability at various points in the business cycle. Yet, productivity is what Cowen examines when formulating his story. He’s not the first economist to make that mistake—Nobel laureate Fin Kydland once wrote a paper blaming recessions on “negative technological shocks,” one that he later repudiated once he realized the error of his assumptions.

Keep the shades on? / There are now very few unskilled workers getting primo wages because of union power. Some of those jobs have simply disappeared, having been replaced by machines. Or they are being done by a machine-cum-highly trained worker, like my friend the sheet metal guy. Cowen warns white-collar workers not to think we are immune from the technological revolution making our skills irrelevant to the new economy, too—unless we have the skills that the revolution complements rather than replaces. This is a message that economists of various ilks have delivered for centuries, but history repeatedly proved them wrong. However, Cowen is a far better thinker than those predecessors, and he makes a compelling (although not airtight) case that this time things are truly different.

The economy is going to grow more slowly than it used to, and your share of its riches is likely going to shrink, Cowen tells us. Here’s hoping he’s wrong, for a change.