John P. Urban and I were discussing the potential for a new, higher volatility regime now that the Fed has abandoned forward guidance as a policy tool. But does forward guidance serve its intended purpose of lowering volatility by removing uncertainty?

Obviously, this is a very nuanced question, but superficially, I'd say the answer is no. Although the Fed began using forward guidance as a policy tool in as early as 2003, it became standard in 2008 when the Bernanke Fed lowered FFR to their zero bound (federalreserve.gov/econres/feds/f…).

Considering the December 16, 2008 meeting as a change-point, there are two potential vol regimes: pre-forward guidance (1990-2008) and post-guidance (2008-2022). There were 207 meetings before Dec. 2008 and 125 since (not including last week's).

The choice of 1990 is subjective. I chose it because it coincides with the earliest published values of $VIX. If you are only interested in realized volatility, the St. Louis Fed has published meeting dates going back to the 1930s.

Forward 30-day realized vol has been significantly higher AFTER the Fed began providing guidance. Before 2008, the mean realized volatility for the 30 days following an FOMC meeting was 15.9%; after it began providing forward guidance, it has been 18.3%.

More surprising, the standard deviation of realized vol also increased by over 25%. Even acknowledging the limitations of hypothesis testing and the certainty that distributional assumptions will be violated, this is a very significant difference in the statistical sense.

The same phenomenon to a lesser degree is noted for implied volatility. The average $VIX close on FOMC days was slightly over 20 before forward guidance began and nearly 22 afterwards. Realized vol of $VIX has been 50% higher post-guidance.

$VVIX may be an interesting quantity to examine, however, since options on $VIX were only listed in 2006, $VVIX can only be calculated back to that point, and therefore has very little pre-forward guidance data associated with it.