In a few minutes, analysts of America’s manufacturing sector and its economy are going to start pouring over the newest monthly gauge of U.S. industry released by the Institute for Supply Management (ISM). And literally, with each passing month, the evidence keeps mounting that it’s a massive waste of time.
Recently, I’ve showed that correlations between ISM’s headline reading on American manufacturing’s health historically has had little to do with the sector’s actual growth rate, as measured by the Federal Reserve’s industrial production index. I’ve also explained that surveys like the ISM’s inevitably suffer from survivorship bias – they may arguably say something useful about what exists of a sample at any given moment, but they’re unable to measure directly how the size of that sample has changed over time.
Of course, the ISM’s headline incorporates more than just manufacturing production. It also includes findings about indicators ranging from employment to new orders to prices paid and received to exports and imports. Nonetheless, when you look at the ISM’s readings on production specifically, it becomes clear that they stack up no better than the headline with the Fed’s figures on manufacturing’s inflation-adjusted growth and shrinkage.
Here’s the comparison for this year so far:
In January, the Fed reported a sharp, weather-related, 1.08 percent decline in manufacturing’s real output. But the ISM’s 54.8 production reading indicated expansion.
In February, the industrial production index revealed a strong 1.34 percent rebound in manufacturing production. The ISM’s production reading fell all the way to 48.2 – contraction territory
In March, according to the Fed, the growth of inflation-adjusted manufacturing output slowed – to 0.89 percent. But the reading was still strong. The ISM’s production results matched up better with the Fed’s – changing from a contraction reading to a solid 55.9 growth number.
Over the next four months, the ISM and Fed figures were even more closely matched, with both revealing faster and slower production increases in the same months.
In August, however, the gap returned. According to the Fed, real manufacturing output fell by 0.40 percent that month. (The figure is preliminary, like all the Fed’ initial findings, and could be revised later this month.) But the ISM’s production figure not only grew, but it’s highest level for the year so far (64.5).
Since I have no crystal ball, I can’t say what today’s ISM report will show. What I can say with great confidence is that there’s little reason to care.