– by a New Deal Democrat
Here’s another detailed look at some important data from last Friday’s employment report. In this post I will look at some leading and matching indicators.
First, about the unemployment rate. It is a lagging indicator coming out of recession but leading on an entry path. While it’s interesting that it’s down -0.1% for the month, for forecasting purposes the year-over-year change is more important.
Here’s another look at the “jobless claims lead to the unemployment rate” metric. The charts below measure the year-over-year change in jobless claims, averaged monthly, /10 per scale (blue). The red line is the year-over-year change in the unemployment rate, from which I subtract -0.1% for reasons I’ll explain after these first two graphs:
What they show is that going back 50+ years, initial claims always rise first on an annual basis while the unemployment rate rises +0.2% or more within 6 months prior to 2 months after the recession started. There are only two exceptions: one month in 1985 and the near double dip in 2002.
Here is the same chart for the last 24 months:
The unemployment rate was +0.1% higher year-over-year in May and unchanged year-over-year in June. This is close but not yet signaling the beginning of a recession. But if jobless claims remain as high year-over-year in July and August as they were in June, that will almost certainly mean a trigger in the jobless rate is soon to follow.
Next, let’s look at two leading employment sectors that have yet to fully reverse due to supply bottlenecks in the Fed’s rate pass-through mechanism.
Motor vehicle production was severely constrained in supply all the way through 2022. As a result (not shown), sales regularly reached or exceeded 15 million year over year in the first half of that year alone.
As a result, employment in vehicle manufacturing (blue in the graph below) continues to rise sharply. Typically, manufacturing employment (in red) declines well before a recession begins. But although it flattened, it did not turn down:
As shown above, employment in vehicle manufacturing is about 5-8% of all manufacturing employment.
The next chart below compares total manufacturing employment, with manufacturing employment excluding motor vehicles (blue):
Ex-motor vehicle manufacturing employment has indeed declined since the start of this year.
The situation is similar in housing construction. The first graph below compares housing units under construction (red) with housing construction employment (blue). Not surprisingly, the former leads the latter:
As housing starts have declined slightly since the peak, housing employment has barely budged.
Finally, let’s update one of my big coincident recession indicators, the year-over-year change in real aggregate wages. Below I’ve broken it down into nominal wages (blue) vs. CPI (red) not yet reported for June:
You can see that going back 50+ years, when annual inflation exceeded nominal annual wage growth, the recession was just beginning.
Here is the same chart from the beginning of the pandemic:
Nominal wage growth, while slowing, is slightly slower than inflation, meaning that real aggregate wages continue to rise. That means more purchasing power for average Americans.
An important question is whether, with year-over-year comparisons that now include gas prices falling from $5 (which means they are less positive), the casual comparison between wages and inflation will continue. And a big part of that problem is also that now that the supply bottlenecks in vehicle manufacturing and housing have eased, those two employment sectors will finally turn around.