Performance After Payrolls
Thursday, March 4th, 2004 8:35pm EST
Jobs Report Effect
Bottom Line: A month after payroll report surprises, the market has most often traded in the opposite direction of the surprise.
In an intraday note today, I mentioned that the day prior to payroll reports tends to be more muted than average. The last five “pre-report” days saw an average range of 8 points on the S&P (about ˝ the normal range) and 9 of the last 10 had closed positively (now we can make that 10 of the last 11). I’m no big fan of extrapolating too much from past reactions to economic releases, because investor behavior towards similar outcomes can often be drastically different.
In any event, it can still be fruitful to look at instances of how the market reacted after prior jobs reports. Over the past two years, on the day of the release, the S&P 500 cash index showed an average range of 17 points (from high to low), compared to 16 points for any other day, so there was not unusual volatility according to that measure. The day closed positively 42% of the time (11 out of 26 occurrences), but notably only 3 of the last 10 closed higher.
When the actual payroll number fell short of estimates by a large amount, not surprisingly the day closed higher only 36% of the time, although the average return was measly 0.1% loss. After 3 days, the S&P was higher only 3 out of 13 times, with an average loss of 1.2%. Within 10 days, however, the S&P was back to being higher a majority of the time, at 62%. For these purposes, I am using the consensus estimate of the payroll number as reported by Briefing.com, and I am considering anything more than 20,000 jobs away from the estimate as being a “large” difference.
When we saw positive surprises in the payroll number, meaning those times when the number exceeded estimates by 20,000 or more jobs, then the day closed higher 50% of the time with an average gain of 0.1%. This neutral performance continued for up to 10 days later.
An interesting observation is what happened after 20 days. When the payroll number surprised to the upside, the S&P was up 20 days later only 33% of the time, and with an average loss of 2.3%. However, when the number surprised to the downside, then the S&P was higher 20 days later 69% of the time, and with an average gain of 1.2%. These are opposite reactions to each other, and from what one would normally expect. Logically, one would think that if the payroll number surprises to the upside, then the market will rally. This gets down to the perverse and fickle nature of economic analysis – when is “good” considered “bad” and when is “bad” considered “good”? These perceptions change constantly, and it’s nearly impossible to compare one release to another. Depending on where we are in the economic cycle and how the market has performed leading up to a release, the same surprise in the payroll number could have completely different reactions.
If we were at more of an extreme sentiment-wise, then there may have been some edge in approaching the reaction to the numbers, but we are neutral for the most part. Recent history suggests that if we see a large negative surprise, then the day will likely close lower, but that is a short-term phenomenon. It is taken for granted that a large upside surprise will spur a rally, but over the past couple of years that has not necessarily been the case.
Conclusion
I’ve spent a great deal of time today going over everything that we follow, and there is a real dearth of new information to go over. We are completely stuck at the moment, and I don’t see any solid edge to give a high-odds, low-risk opportunity in either direction. My preference for weeks now has been to fade (i.e. trade against) the extremes we see on short-term oscillator-type indicators (such as our intraday indicators), and I continue to prefer that strategy. At some point, obviously, this range will break but for the moment I don’t see the sense in becoming aggressive in anticipating it. Our shortest-term measures are very slightly overbought, but once again I am not seeing anything that would want me to be aggressive about shorting because of them.
Jason Goepfert
President and CEO
Sundial Capital Research, Inc.
Disclosure: no positions
This disclosure is not intended as trading advice in any form. It is meant as a note to subscribers that the author may have a position directly affected by the market outlook reflected in the commentary. Although the author takes great pains to remain objective in any commentaries, it is only fair that readers should know that the author may have taken positions in accordance with his market outlook. Positions can and do change at any time, without notice to the reader.
© 2004 Sundial Capital Research, Inc. All Rights Reserved.