, , , , , , , , , , ,

Since Janet Yellen is a leading labor economist as well as Federal Reserve Chair, and she closely follows the monthly so-called JOLTS reports, so do I. So should you if you’re interested in how the U.S. labor market is faring and therefore (to a great degree) whether the central bank will move to stimulate the economy or cool it off.

My main interest in these data has focused on what light they shed on job quality – and specifically on whether the job openings reported in these surveys of employment turnover have come mainly in low-wage or high-wage sectors. (My work shows it’s the former.) But this morning’s JOLTS numbers from the Labor Department contained such astonishing results for manufacturing that they deserve special attention – and not simply because the April data were so exceptional, but because since the last recession began, they have contrasted so strikingly with other measures of manufacturing’s performance.

According to the new JOLTS report, America’s manufacturers reported 415,000 job openings at their companies in April. That’s the second highest figure on record (the data go back to the end of 2000), which is newsworthy enough. But it also represents a 23.15 percent jump from the March total of 337. Just as interesting, the year-on-year increase is 23.15 percent, too.

Logically, this surge means that manufacturers became much more optimistic about their prospects in April. Why else would they be looking for so many new workers? Yet nothing else we know about domestic manufacturing in April would seem to justify this optimism.

For example, in inflation-adjusted terms, manufacturing production inched up only by 0.33 percent in April over the March levels – a decent performance by recent standards, but no standout. Since April, 2015, manufacturing output rose by only 0.54 percent.

Do future-oriented gauges of manufacturing signal the appearance of great expectations? New orders for manufactured goods in April did rise by 1.92 percent on month (these are not price adjusted), but that kind of improvement is nothing exceptional. Moreover, year-on-year, this measure of incoming work is down 1.80 percent.

But this disconnect between the job openings data and other manufacturing statistics doesn’t just stem from one month that could be a classic outlier. (Also, the data will be revised next month.) It’s been the case since the recession began.

During the downturn itself, JOLTS trends did follow the other gauges way down. Between the slump’s onset, in December, 2007, and manufacturing’s employment bottom, in March, 2010, industry’s job openings plunged by 45.11 percent. During this period, real manufacturing output sank by 14.92 percent, and factory orders dropped by 16.73 percent. So far so good.

But since March, 2010, the number of job openings reported by American manufacturers has skyrocketed by 184.25 percent. This increase has left in the dust the rise in constant-dollar industrial production (12.68 percent) and manufacturing orders (15.40 percent). And even if you take out the unusual April manufacturing job openings number, the gap is still enormous.

In fact, since the recession began more than nine years ago, manufacturing job openings are up by 56.01 percent, even though real output is down by 4.13 percent and factory orders have fallen by 3.91 percent.

This gap suggests that the conventional wisdom about the relationship between manufacturing employment and manufacturing output need some big rethinking. After all, it’s become a commonplace that manufacturing has no chronic output problem – it does, however, have a serious jobs problem (which is rarely described in this context as a problem) because technology makes it possible to turn out more products with fewer workers.

But the picture created by combining the JOLTS, production, and orders statistics indicates that modest gains in production and orders have been spurring a tremendous increase in the demand for workers. How can that be if the sector is so increasingly capital-intensive? Further, standard economic theory teaches that when businesses find themselves short of labor, they either boost wages in order to attract more applicants, substitute technology (machinery, equipment, software, etc.) if they don’t feel like offering higher pay, or respond with some combination of these steps.

Yet as I’ve exhaustively documented, until very recently, manufacturing wages have been going nearly nowhere. And businesses overall have displayed few signs of significantly increasing their capital spending (on that new machinery, etc.) – at best.

Of course, it’s possible that all of these government statistics are wrong. But I suspect it’s more likely that the so-called experts know much less about manufacturing than they think.