By: Steve Sosnick Chief Strategist at Interactive Brokers

In the spirit of recycling, we will offer the opening sentences from a piece I wrote almost exactly one year ago today:

I frown upon the use of intoxicants during the trading day, but I would like to propose a drinking game for bored, late-summer traders. Have each one tune into a different financial network and take a drink when the phrase “Friday’s key non-farm payrolls report” or similar is mentioned. 

At that time, we were in the thick of a bull market amidst relative complacency, as indicated by a 16 level on the CBOE Volatility Index (VIX). There were some concerns that Federal Reserve rhetoric might move away from being solidly accommodative, but few expected an imminent change to actual policy. It is understandable why traders might have been somewhat bored.

What a difference a year makes. The S&P 500 Index (SPX) is about 10% lower and VIX about 10 points higher. The Fed Funds target rate is 2.5% compared to 0%. The Federal Reserve’s balance sheet is now shrinking after peaking in April. And perhaps most importantly, Fed Chair Jerome Powell offered a speech on Friday designed to eliminate any ambiguity about the central bank’s inflation-fighting resolve. Nonetheless, employment statistics remain a key indicator for a data-driven Fed.

Today’s release of the ADP National Employment Report for August was awaited by many who hoped that it could offer a clue to Friday’s Nonfarm Payrolls Report from the Bureau of Labor Statistics (BLS). The number was disappointing, coming in at a gain of 132,000, well under the 300,000 consensus estimate. In the “bad news is good news” world of liquidity-addicted markets, the weaker report should offer the sort of data that would cause the Fed hike rates less aggressively. The tepid market reaction tells us that ADP data is having little impact on traders’ thinking.

Perhaps traders have come to realize that ADP is at best an imperfect predictor of the BLS report, nor does it strive to be. This is from the FAQ section on ADP’s website:

The NER [National Employment Report] uses ADP payroll data to provide a nationally representative measure of employment. This new approach differs from the former NER’s model-based methodology, which sought to forecast changes in the Current Employment Statistics monthly survey conducted by the Bureau of Labor Statistics.

 When we look at the data graphically, we see that the two reports don’t correlate particularly well:

ADP National Employment Report (white) vs. BLS Nonfarm Payrolls Report (blue), monthly data

(Click on image to enlarge)

Source: Bloomberg

We know that ADP’s numbers will be lower than BLS’s by definition. ADP measures its customers, while the BLS attempts to be a broad survey of households. Obviously, not everyone works for a company whose payrolls are handled by ADP (I do, btw). The hope is that the changes in the ADP report would foreshadow those in the BLS report. Unfortunately, they don’t.

Last month, ADP showed a smaller increase while BLS rose unexpectedly. In April 2021 we see ADP rising while BLS showed fewer jobs created. A close look at the graph above shows several divergences in either direction. Economists are predicting that the payrolls report will show the creation of 300,000 new jobs (yes, the same as their ADP estimate) on Friday. 

That would be a significant decline from last month’s 528,000. If the estimates are correct, that would indeed show a large drop, just like ADP showed. Quite frankly, it is hard to imagine another upside surprise of the magnitude that we saw last month; and even so, the strong data did little to quash market hopes for a Fed pivot. Those hopes were obliterated last week by Mr. Powell, but one can’t blame traders if they remain hopeful for the sort of weak data that could provide a softening of Fed rhetoric. On the other hand, be careful what you wish for….

More By This Author: