ETF Volatility Targeting

May 30, 2012 in Drawdown | Volatility

A new book in the Market Wizards series (by author Jack Schwager) has come out this month.  While these books are a collection of interviews of great money managers -- Schwager himself also does a nice job of summarizing some of the themes he personally has gleaned by incorporating his decades of experience into a series of observations.   

He also recently summarized a few of these observations on his twitter account (@jackschwager).  

A few of his takeaways from interviewing top money managers:


* It is not about predicting what will happen -- but rather recognizing what is happening


* Many go wrong by failing to adjust exposures to changing market volatility


This all conveniently ties into ETFreplay, using Relative Strength to help recognize what is happening is foremost.   But on the second point, we recently added a module to help think about how to adjust exposures to changing Market Volatility.  Let's look at one example of the latter.

Let's think about the Russell 2000 Index, the most popular index for small cap U.S. stocks, which is one of many important market segments we can access at ultra low-cost (never any redemption fees or lock-ups with ETFs) and it of course has total transparency and is deeply liquid.

Let's look back at 2010 for an interesting example of how this segment has traded.  

2010 was a very good year -- but you wouldn't have said that during the summer of 2010 when there was a large drawdown following a flash crash in May and yes, continual negative newsflow from... drumroll... Europe.    The final IWM return was very strong +27% but masks the mid-year washout and pain many investors felt.


Here is 2010 as full year snapshot. 

Go forward one year to 2011, the IWM final return of -4% for IWM also greatly masks the 'path-taken.'  Another large drawdown, this time -29% and about the same actually as the European index loss (VGK was off -30% from peak to trough).

This is very important and something that investors must study a great deal --- the long-term return of the markets is not all that great in relation to the often wild path taken to get that return.  That is, a long-term return of say +7% might have huge drawdowns along the way that cause investors to actually end up losing money if they don't learn how to deal with this.    

(In modern portfolio theory terms, you describe this situation (low return relative to high volatility) by saying simply that the Sharpe Ratio is not very good.)   

If the short-term S&P 500 sharpe ratio gets really high, just wait -- it's coming back in at some point.   This is what happened in Q1 2012 when the S&P 500 YTD sharpe ratio was over 3.00 at one point.   We noted this as an unsustainable figure on our Allocations board timeline.   And now we see the inevitable washout that occurs with assets that don't have good long-term sharpe ratios.   If you want a more efficient equity curve, then don't buy and hold stocks --- unless its part of a well thought-out allocation that adjusts to prevailing conditions.

On the Tools page is a new module 'Volatility Target Test.'  This module executes a convenient, clean performance backtesting report for you complete with detailed period-by-period weightings and return.   

It combines any ETF of your choosing (such as IWM) with a cash-like ETF (SHY) and allows you to therefore approximate a level of volatility for the combination based on changing (dynamic) market conditions.  It continually adapts to the current environment and records the performance of such a mechanical targeting approach.

It should be clear that if you target low volatility and the market goes up a lot -- then obviously it will underperform.   But if you target lower volatility and the market goes down a lot, it will obviously then outperform.    The point of the application is not to be an optimal weighting, it is to help us all understand how volatility targeting is working and how to avoid one of Schwagers main points repeated here

* Many go wrong by failing to adjust exposures to changing market volatility

Below is a single view screenshot of the new volatility Tool:

Follow us on Follow etfreplay on Twitter


Comments (5) -

May 30, 2012 16:04 #

Due to extremely high levels of computer-automated spam that hit blogs, the comments section may close promptly - as always, you can use the Contact tab is you would like to reach us.

etfreplay United States

May 31, 2012 00:47 #

Great new tool we can test the impact of volatility with, thanks!

ps: in your screenshots above, I see that the ENTIRE user interface of the website is using a  "black text on white background" style. But users get the "white text on black background" style by default. How can we also get the "black on white" style? (is there a pref setting for this somewhere?)

Turnleaf Belgium

May 31, 2012 01:26 #

To change the website color theme:
1) Login
2) Click 'Account Settings'
3) Click 'Website Color' on the menu on the left hand side
4) Choose 'White' and click 'Save'

etfreplay United States

May 31, 2012 18:13 #

Thanks for adding the ability to view the website in white- huge improvement for someone with eye challenges!

Isaac United States

Jul 20, 2012 13:16 #

would it be useful/possible on multiple sustem back test to have a volatility function that would reduce and expand allocations based upon changes in market volatitliy , ex if spy was one model and agg was the other  adjust porfolio weight based upon spy volatitly?

broark United States

Comments are closed