, , , , , , , , , , ,

Even though (thankfully) few claims have been heard lately that American domestic manufacturing is either enjoying a renaissance or on the verge of one, it’s still commonplace to insist that the sector’s output is at an all-time high. Invariably, the purpose is to argue that U.S. industry is in perfectly acceptable shape, and that anyone believing the contrary is mistakenly obsessing about the industry’s long years of job losses – which are attributed to its ever faster adoption of labor-saving technologies that are encouragingly boosting productivity.

Let’s (charitably) leave aside the reality that manufacturing’s productivity performance lately has been historically feeble. Instead, let’s express thanks to the Commerce Department, which has just published its latest reminder of how complicated – and uninspiring – manufacturing output has been lately.

As known by RealityChek regulars, the Federal Reserve’s monthly industrial production numbers have made sadly clear that, when adjusted for inflation, American manufacturing output is anything but at an all-time high. Indeed, its peak came at the very end of 2007, just as the Great Recession was beginning according to the quasi-official scorekeepers at the National Bureau of Economic Research. Since then, real manufacturing production has shrunk by 4.43 percent.

But as the Commerce Department’s new quarterly report on gross domestic product (GDP) by industry once again describe in detail, the U.S. government measures manufacturing output in several different ways. Unfortunately, however, none of them tell a story especially flattering to the sector.

Commerce gives pride of place to a gauge called real value-added – which tries to eliminate the double-counting that often marks the simpler production numbers mentioned above. In other words, real value-added (along with value-added that’s not adjusted for inflation) tries to ensure that the government isn’t counting an auto part once when it comes out of a parts factory and again when it comes out of a vehicle factory as one piece of a finished good.

The real value-added data certainly make manufacturing’s recent production growth look better than the real output figures. But that’s not to say they make this performance look good, for they show that from its 2007 peak through the end of last year, domestic manufacturing has shrunk by 0.81 percent.

It’s only when you don’t adjust for inflation that you start getting genuinely encouraging manufacturing output numbers. According to the pre-inflation value-added data, American manufacturing production was indeed at an all-time high as of the end of 2016. In fact, it’s 17.31 percent higher than the 2007 level.

Don’t get too excited by this number, however. If we switch time-frames and just look at the current economic recovery, we see that from the time the economy resumed growing in 2009 (and unfortunately, this data set doesn’t contain quarterly figures, so we have to use full years) through the end of 2016, current dollar manufacturing value-added advanced by 25.97 percent. That’s only a little faster than manufacturing’s 25.42 percent value-added growth during the previous (2001-2007) expansion – which was a year shorter. Moreover, no one ever considered those years a manufacturing golden age – or even close.

Yet another version of the manufacturing story is told by the gross output numbers compiled by the Commerce Department. They show that manufacturing production rose after the last recession began – by 12.94 percent. But that growth ended in 2014. Over the next two years, it dropped back – by 4.41 percent. So record manufacturing output came – and then went.

Adjusting gross output for inflation serves up a less volatile picture of manufacturing production, but it’s not much of an improvement. According to this measure, output peaked in 2007, and since then has dipped by 0.27 percent. That’s not much. But it doesn’t mark 2016 as an all-time high, either.

To me, a reasonable bottom line is this: The U.S. government uses five different measures of manufacturing output. Both its timeliest (the Fed industrial production figures) and arguably the one it likes best (real value-added) show that real manufacturing production is off its recent records. So do two other measures. The only data set that tells a reasonably good story is one that’s doesn’t take inflation into account. That’s a pretty flimsy reed on which to base claims that domestic manufacturing output has never been stronger.