Outsized Volatility And Market Bottoms

By in
Outsized Volatility And Market Bottoms

Every stock trader knows the adage “never short a dull market”, but now that COVID-19 has sparked remarkable equity volatility we got to wondering if the opposite is true. Can you make money shorting highly volatile markets, or avoid future losses by selling down long positions during these periods?

To answer that question, we first need to define price volatility and determine what level constitutes truly stressed market conditions. Here is our methodology and answer:

  • We took daily closing S&P 500 prices back to 1999, calculated one-day returns and the rolling 30-day standard deviations of those day-to-day price moves.
  • The average standard deviation of daily returns is 1.0%, which means any move +/- 1% in a day is in the fat part (67%) of a normal distribution.
  • This 1% mean result has its own standard deviation observation, which is 0.6%.
  • That means that when we reach daily price volatility that results in a +4.1% standard deviation of 30-day price moves, we are in the realm of the truly unusual (5 standard deviations from the mean).
  • See the graph below for the entire 1999 – present timeline of trailing 30-day return standard deviations.

Lots of math, but here is the upshot: current S&P 500 price volatility is just as bad as the 2008 Financial Crisis, and (importantly) much higher than other recent market/geopolitical events like the dot com bubble bursting, 9-11, the second Gulf War, and the 2011 Greek debt crisis. The numbers:

  • Current trailing 30-day S&P 500 price volatility (1 standard deviation): 4.4%.

    It has been above the 4.1% level for 3 days (not including today’s close).
  • At the start of Q4 2008, S&P price volatility rose above 4.1% on October 15th. It stayed above that level for 46 consecutive trading sessions.

    There was an investable low on November 20th, 27 trading days after October 15th. It was not the absolute low (that came in March 2009), but it was the low for the year 2008.

For comparison:

  • Peak volatility after the 9-11 terror attacks: 1.8% (October 11th, 2001).
  • Peak volatility after the dot bubble burst/run up to Gulf War II: 2.7% (August 15th, 2002).
  • Peak volatility during the 2011 Greek debt crisis: 2.7%.

All this supports our “2008 Playbook” thesis that the current COVID-19 Crisis has only one useful market analog – the 2008 Financial Crisis – and that means we need to consider the following issues in timing the depth/duration of the downturn in stock prices:

  • American political cohesion. In Q4 2008 market volatility remained high for +40 trading days in part because the November elections changed the balance of political power in Washington and delayed DC’s policy response.
  • The magnitude and duration of COVID-19’s spread and impact on American communities. Living in NYC as we do, we’re seeing this unfold in real time.
  • The speed and scope of currently considered and future fiscal stimulus. Also in the mix: what trillions of dollars of deficit spending will do to the structural level of interest rates.
  • The ability for central bankers around the world to address stress in the global financial system.

Some of these issues are in better shape just now than in 2008 (DC knows it owns this crisis and is moving faster, as are central bankers) and some are still unknowable (virus spread, political appetite for further rounds of stimulus/bailouts). That is why we’re not inclined to move off our 2008 Playbook approach even though the situation then-to-now may seem different. In truth, it is not. At least not yet.

Before we finish up, there are 2 other charts below with the same data but specific to the Financials and Technology sectors. They show that:

  • Financials sector volatility is actually not back to 2008 levels, which makes sense to us. As much as this group is suffering with economic volatility and disrupted capital markets, they are not under the same systemic threat as 2008.
  • Tech sector volatility at present is on par with both dot com bubble bursting levels as well as those during the Financial Crisis. Again, that seems reasonable to us.

Summing up: the data shows that highly volatile markets (2008 and now) are not buying opportunities, at least during the front part of the storm. In 2008 the lows did not come until 5-standard deviation volatility had been raging for over a month. Using the same math, today was just Day 4. We continue to search for other market signals that will show why this period will see a faster trip to the lows. As of yet we have not found any.