By Madeline Schnapp, Director of Macroeconomic Research
Today we are going to talk about jobs. Occasionally, we get to pat ourselves on the back. Does anybody recall the big difference between our employment estimates and the Bureau of Labor Statistics' estimates in August and September? More importantly, did you check out the revisions to the BLS employment data? That's what we are going to talk about today. Then we will talk about why the BLS' October estimate is wrong again!
In August, we reported the economy created a healthy 185,000 new jobs. Meanwhile the BLS reported a paltry 96,000 new jobs, nearly half of our estimate. We received lots of skeptical emails, primarily from the financial press. In the past two months the BLS revised its August estimate up a whopping 100% to 192,000. So for all the skepticism, who turned out to be right here? Yours truly, thank you very much! Ok, what about September? In September, we reported the economy created an even healthier two 210,000 new jobs, while the BLS reported only 114,000 new jobs, also nearly half our estimate. We received lots more skeptical emails, once again from the financial press. The BLS' first revision to the September employment estimate was up 20%, to 142,000. While 142,000 is still a far cry from our 210,000, it is definitely moving in the right direction. In addition, I am very confident, that in December, the BLS will revise their September estimate even higher. The hefty BLS revisions beg the question of what the BLS missed.
Beginning in June we observed a steady increase in year-over-year growth in real-time income tax withholdings. That means wages and salaries were growing due to an increase in jobs. In fact, wage and salary growth began to accelerate in July, continued in August, peaked in September and then slowed in October. The BLS completely missed the June through September acceleration in growth. Worse still, the BLS is missing the deceleration in growth in October.
We postulated that the increase in tax withholdings over the summer was due to job growth in interest rate sensitive sectors of the economy, such as single family and multi-family construction, home sales, mortgage refinancing, and vehicle sale. But, we had one big caveat. In our August and September reports, we stated that growth was probably temporary and would end in October due to the seasonal nature of construction and home selling activities. Our predictions have been correct. Of particular concern, however, is that beginning in October, year-over-year growth in real-time income tax withholdings has slowed. That means there has been a slowdown in job growth. Our tax based employment model, therefore, reported a 33% pullback in October job growth to 140,000, down from our September estimate of 210,000.
How does that square with what the BLS just reported? The BLS reported that 171,000 jobs were created in October compared to our estimate of 140,000. But what is really important here is the fact that the BLS is reporting increasing job growth while we are reporting a decrease in job growth. The BLS completely missed the job growth pop this past summer that is now becoming apparent two months late. Worse still, the BLS is also going to miss the slowdown in employment that is going on right now and will likely continue through the end of the year
So what, in our view, is driving the current slowdown?
We believe the slowdown is due to uncertainty surrounding the massive new taxes associated with Obamacare, expiring federal stimulus programs, and likely cuts in military spending all beginning January first. Without additional clarity concerning taxes and spending priorities, we expect the negative impact on economic growth to continue.