It’s puzzling to observe the market and media fascination with the monthly Employment Report released by the Bureau of Labor Statistics (BLS) on the first Friday each month.
There’s no doubt this report can move markets; however, historically it’s an overrated economic indicator that’s not particularly useful in gauging market trends.
First up, let’s look at the most oft-cited statistic from each monthly report – the growth (or decline) in payrolls (jobs):
Source: Bloomberg
This chart shows the total change in payrolls from each monthly BLS report between the end of 1992 and the end of 2002.
As you can see, payrolls growth looked strong through the spring of 2000; indeed, in March when the stock market peaked ahead of the 2000-02 bear market, payrolls grew at the fastest pace in almost three years.
In the second half of 2000, payrolls did weaken a bit, but the slowdown in jobs growth looked roughly comparable to the 1995-96 period (circled in red) when the US economy did not slip into recession. Indeed, the 1994-95 experience, when the Fed hiked rates to bring down inflation without tipping the economy into a new recession is often cited as one of the only examples of a successful “soft landing” in US history.
One could have credibly argued in the summer of 2000 the Fed’s 175 basis points of rate hikes between June 1999 and May 2000 were once again quelling inflation without tipping the economy into contraction, an echo of the 1994-95 experience.
In fact, many did argue just that partly due to widespread respect at the time for then-Fed Chair Alan Greenspan’s skill in micro-managing the economy through the historic US economic expansion of the 1990s.
Jobs growth didn’t turn outright negative until the spring of 2001 when the S&P 500 was already mired in a bear market and the economy had already slipped into recession.
At best, the monthly payrolls releases were a coincident indicator of US economic health in the 1999-2001 period.
And the picture looks even worse when you consider this chart:
Source: Bloomberg
The first payrolls chart in today’s issue represents the final revision for payrolls data from 1992 through 2002, not the initial payrolls growth estimates released in real time through this cycle.
Unfortunately, I don’t have initial estimates for payrolls data prior to 1996; the chart above shows initial estimates for monthly payrolls growth released by BLS from late 1996 through late 2006.
As you can see, the initial estimates of US jobs growth looked very strong in early 2000 and remained solid for months after the stock market peaked in March 2000. Strength in employment data was undoubtedly one factor behind the Greenspan Fed’s decision to hike rates by 50 basis points in mid-May 2000 despite a surge in stock market volatility that saw the Nasdaq Composite plummet more than 37% from its intraday peak on March 10, 2000 to its intraday low on April 17th of the same year.
Initial estimates of monthly payrolls turned negative for two consecutive months over the summer, only to strengthen once again in the fall – average monthly growth in payrolls from September 2000 through February 2001 was 165,000 based on initial estimates from BLS, a healthy level.
When the final numbers came available months later, BLS reduced those average monthly payrolls estimates over this period by more than 70,000 per month.
In short, there are two glaring flaws with using BLS data as an economic indicator.
First, BLS releases its first estimate for monthly payrolls on the first Friday of every month. However, these first estimates are routinely subject to major subsequent revisions that can entirely change the implications and outlook for the US labor market.
And, second, even if we use the final release from BLS, payrolls are at best a coincident indicator of the state of the US economy – typically the jobs market, measured on this basis, weakens just as the economy enters recession or even with a significant lag.
Take a look at the 2004-2010 experience for a second example of these flaws:
Source: Bloomberg
This chart shows the initial monthly payrolls estimates from BLS for the period from December 2004 to December 2010 as a blue line; the orange bars show the subsequent revisions to each monthly number. In particular, when the orange bars are below zero, that means the BLS ultimately revised the monthly payrolls data lower while positive bars mean jobs growth was ultimately revised higher.
In this cycle, initial estimates from BLS didn’t show job losses until January 2008, three months after the stock market peak and a month after the recession started in December ‘07.
And the first month of significant job losses this cycle (more than 100,000 monthly payrolls decline) was September 2008, by which time the stock market had been falling for nearly a year.
Even worse, look at those orange bars – BLS dramatically and consistently overestimated the strength of the economy from April 2008 through to April 2009 with an average subsequent downside revision of more than 175,000 jobs per month. In other words, if you were looking at BLS initial estimates of jobs losses in early 2008, the economic downturn still looked fairly mild with monthly jobs losses that were lower than the 2001 recession.
Subsequent revisions revealed a much more dangerous and sinister trend underway; however, these final estimates from BLS weren’t available until months later, far too late to represent a real-time read on the economy.
My point isn’t to tell you the monthly BLS Report is worthless or unimportant – employment is the bedrock of consumer spending, which accounts for roughly 70% of the US economy.
However, trying to draw sweeping conclusions about the health of the US economy based on shifts in initial BLS payrolls estimates is dangerous.
In the next issue of The Free Market Speculator, I’ll take a quick look under the hood of the December 2022 report released on Friday January 6th and why it suggests a major decline in payrolls, consistent with a US recession, is imminent.
DISCLAIMER: This article is not investment advice and represents the opinions of its author, Elliott Gue. The Free Market Speculator is NOT a securities broker/dealer or an investment advisor. You are responsible for your own investment decisions. All information contained in our newsletters and posts should be independently verified with the companies mentioned, and readers should always conduct their own research and due diligence and consider obtaining professional advice before making any investment decision.