Reality Check – VIX, SPX, and Non-Farm Payrolls

The first Friday of the month means different things to different people.  For instance, at the Rhoads household that’s taco night!  However, in the world that is the financial markets the first Friday of the month is (typically) the day that the market digests the Non-Farm Payrolls number.  This is often the first number to give us insight into the economy from the previous month.  For example this coming Friday’s Non-Farm Payroll report discusses changes in employment during October.

This report is usually met with lots of discussion in the financial media, leading up to and then in response to the report.  I decided to see if all the hubbub around this report was justified.  I was actually a bit surprised by all but one of the results from some basic data analysis.

First I took a look at the S&P 500 performance on the day of the report versus any other trading day.  My assumption going into this exercise was that the S&P 500 would be more volatile on days that the employment report was announced relative to any other trading day.  To check this theory out I took the S&P 500’s reaction to the employment report and looked at the average move regardless of direction, from January 2007 through October 2015.  The result was that the average move for the S&P 500 on employment day is a gain or loss of 0.89%.  I then took the average S&P 500 price change on all trading days over the same time period.  The average S&P 500 price change, on any trading day, was a gain or loss of 0.88%.  It appears the trading action around the Non-Farm Payroll report is just like any other day, at least in the S&P 500.

The next two tests involved my favorite index, VIX.  The findings here were a little bit different and also not all as much of a surprise.  First I ran the same absolute average change in VIX test as was run above for the S&P 500.  The average VIX change on employment day has been up or down 5.23% while an average day for VIX involves a gain or loss of 5.36%.  That’s right, VIX is usually a little less volatile on employment day compared to average trading days.

Here’s where something I always thought was validated quantitatively.  Everyone thinks of VIX as being reactive to what goes on in the stock market.  However, I’ve always argued there is a bit of an anticipatory element to VIX in front of and after big economic reports.  Much like the implied volatility for options on a stock that is preparing to report earnings, but not nearly to the same extent.  I took a look at what VIX has done on a directional basis in reaction to the employment number.  It turns out about 71% of employment reports resulted in VIX dropping.  On any trading day VIX has about a 54% chance of losing value, however, there is a bit of a day of week impact on VIX where the index is more likely to drop on a Friday.  Knowing that I broke Friday out and found that 61% of Fridays VIX loses value.  Therefore I’m going with my hunch that VIX reacts a bit to the downside post Non-Farm Payrolls.

So in summary –

  • The S&P 500 price reaction to the Non-Farm Payroll number is about the same as any other trading day
  • VIX actually moves a little less on average in reaction to Non-Farm Payrolls that on an average trading day
  • There is an anticipatory element to VIX which results in the index having a higher chance of moving lower on Non-Farm Payroll day than any other day, including Fridays.