Talking about Jobs and the US GDP

The Jobs That Were Never There

Incompetence? Deceit? Who knows. Am I the only one who still cares? Sure does feel like it. The Bureau of Labor Statistics released their Q4 2023 BED (Business Employment Dynamics) Report very quietly on Wednesday. As we did three months ago, we compare the jobs created for that period to what had been reported in near real time as Non-Farm Payroll job creation in the monthly BLS Establishment Survey that the financial media regularly takes as gospel. Why is the BED Report understood by economists to be far more accurate than the monthly surveys? Simple. It comes out with a seven-month lag and is based on data provided by 9.1M establishments. The monthly survey involves just 670K establishments and the rest is then extrapolated through often incorrect seasonal and birth/death model adjustments.

Well, the results are in. For the fourth quarter of 2023, the BED report shows 344K jobs created in the US. The Non-Farm Payrolls prints for those months, even after revisions came to 656K. That's an implied net overstatement of 312K jobs created over three months. Shall we move on? For Q3 2023, the BED Report shows job losses of 192K versus the revised NFP prints totaling a gain of 681K jobs. That's an implied overstatement of 873K jobs created.

For the first quarter of 2023, the two reports were very close, just 19K apart. The sharp disagreement in the numbers really began in the second quarter of 2023, when the revised Non-Farm Payrolls prints totaled job creation of 785K while the BED report shows just 332K jobs created. All in all, for the full year of 2023, the monthly Non-Farm Payrolls numbers in aggregate (after revisions) totaled job creation of 3.117M positions. The far more accurate BED Report shows that number as 1.46M jobs. The implied overstatement in job creation for the full year 2023 is 1.657M positions. Why even report monthly survey results if the models used are that far off?

Who is to blame for either their incompetence (best case) or worse when job creation comes to just 46.8% of what was reported in close to real-time and devoured by a financial media all too eager to support a narrative?

You do know this means that GDP for 2023 likely has to be restated, and revised significantly lower, right? For 2023, full year GDP printed at growth of 2.5%. Full year GDI (a just as worthy measurement of economic growth as the Fed recommends averaging the two when they are not close) came to growth of just 0.4%. The US economy was probably very close to recession in 2023. Very, very close. Those economists who predicted a tough economy in 2023 and were openly mocked for it? Yeah... turns out they pretty much nailed it. The "stro.ng economy" crowd? They were just plain old wrong.

Oh, by the way.. This morning, the BEA posted its first estimate for US Q2 GDP. The number printed at +2.84%. (0.82 of that is inventory building, 0.53 of that is government spending). Note: GDI is not released with the advance estimate so there is no check on the headline number as of yet