Low risk outperforms high risk

Traders seek to find the next BIG think ala Apple, Bidu and co. Often times this desire is founded by the underlying assumption that taking higher risk is rewarded with higher returns. Furthermore many traders believe in an efficient market where as one can outperform the market only by taking above average risk. However this hasn’t been the case for the US stock market over the last couple of decades or so. A lot has been written about the low-volatility anomaly in financial markets. With this post I want to share some insight how this works out on index stocks over various time frames.

The setup

For my research I define risk in terms of historical volatility. I looked into S&P100 and S&P500 stocks (survivorship bias free) from Jan. 1990 until Nov. 2011. Stocks are ranked by historical volatility of various time-frames (20, 50, 100, 252 and 500 days). I tested buying the top or bottom decile (long only). So for the S&P500 index the strategy is always invested in top/bottom 50 stocks where as for the S&P100 the strategy is always invested in top/bottom 10 stocks. The test and all calculations are done using daily bars, re-ranking/rotation happened weekly. No trading cost considered.

The result

(click on the picture to enlarge it)


Thoughts and observations

  • Low volatility (low risk) outperforms high volatility (high risk) on an absolute as well as risk adjusted basis.
  • The anomaly is consistent among all volatility time-frames (20,50, 100, 252 as well as 500 days).
  • Buying bottom decile (low volatility) outperforms index buy and hold (without considering trading cost).
  • The same test has been conducted for Nasdaq 100 stocks, with confirming results (and conclusions).


For me this is eye-opening. It refines my way of thinking about risk (and reward). Tell me what you think and how you gonna apply that to your trading (-strategies).

# Frank

About these ads


  1. Frank, great research as always. I am very surprised about this simple but effective approach. Also the large cap top decile is interesting for longer periods, I think. It may be diversifying in combination with the bottom decile? Is it maybe some other factor for selecting growth and value stocks?
    Could be also an input for long/short and hedged approaches, I guess.

    • Hello Andreas,

      I think low volatility stocks are an excellent addition for a portfolio that is maybe trading NDX100 momentum stocks. However, looking into a min covariance portfolio of low volatility stocks volatility can be further reduced.

      In my opinion one reason why low vola outperforms high vola is the compound effect. Just think of a 80% draw-dawn that many of the high vola portfolios had. The high vola portfolio needs a couple of houndred percent (~500%) to make up for that draw-down. During that that low vola stocks are also raising (not as fast but from a higher level).


  2. What was your measure of volatility?

  3. Very interesting — thanks.

    A number of low-volatility ETFs have been issued recently (examples: EEMV, LVOL, SPLV, HILO, …). Unfortunately, it seems like when ETFs recognize a hot industry, theme, or investment anomaly, it is about to go cool!

  4. I have been following Your blog for some time and like your focus on risk. Everything I do is based off the following maxim: By properly measuring, controlling, and limiting risk, returns will naturally accumulate. Basically, a focus on risk is more profitable than a focus on return. I have been running experiments (link) in which I test this theory.

    My only thoughts on your post are that declining stocks fall faster than they rise, and are therefore more volatile. Would your system above be in actuality a type of relative strength system?

  5. Have you tried defining risk by downside volatility? Upside seems to be nice, it’s just the down side that hurts.


  6. Joshua Schultz says:

    I understand, but in looking at the lowest vol stocks using CRSP, they are also usually stocks that are “not declining” since volatility increases as a stock falls. Also, stocks that crash will be eliminated in a down market, and stocks with lower vol (implying that they did not fall as fast) will pass the test. In a bar graph comparison, the similarities of RS decile components is similar to the components of a test ran ranking by vol. I found it very interesting. The more we can understand the interconnectedness of the numbers we use, the more we can understand the risk associated (and which environments we should use certain methodologies)

  7. May I suggest you read “Re-Thinking Risk: What the Beta Puzzle Tells Us about Investing” from http://www.gmo.com/America/MyHome/default ? The best study on that topic so far.

  8. Hi Frank,

    very nice test. I think adding a simple market filter like Close>/<SMA200 (on SPY or stock level) might show addition insights. I would guess that during bull markets you want to choose volatility stocks (up-volatility) over low volatility. That might turn around during bear market conditions, cause strong stocks will get hit harder during a crash (lots of people are invested in highflyers and they all want to get out of the stock market at once). As your numbers show, its the Draw Down that makes the differences.

  9. Run a test where you adjust for the possible relative strength-component and see if it’s significant. You could also check if they are correlated or if the pure volatility ranking really ads value as a signal.

    By the way, isn’t a high VIX bullish for the market. How would that fit into the context?


    • Hi Emil, a high VIX is not bullish for the market as far as trend following is still working somehow. In the past 10 years it made sense to stay away from stocks in a high VIX, VaR, Hist Vol environment. Simple risk based portfolio protection was an useful strategy. Cheers, Andreas

      • Thanks. Haven’t really done any testing myself, but I did a test on the standard deviation of daily return over the last 21 days and found that when this is at its highest for the last 2 to 30 days the market tends to rise and vice versa. Haven’t done any testing on less extreme readings. Since this measurement is pretty darn correlated to the VIX I made the conclusion that it a well would indicate a bullish market at extremely high readings.

        How are you testing it? I would imagine that longer look back would isolate bear markets from bull markets and thus be trend following.

        • Hello Emil, Hello Andreas,

          i think to a certain degree you are both correct:

          - When VIX is raising (from a low level) that’s typically negative for the market.
          - When VIX is very high (after the stock market sinked) both are about to mean-revert.


  10. prem nath says:

    .I have one question.For a given stock in a given time how often is occurence of M size move compared to 2M size move.My studies over 20 years conclude that M size move should be 4 times more common than 2M size move—based on square root of time principal for option pricing/odds of move of a given size in the underlying in the BLACK SCHOLES formula. BLACK SCHOLES may not be perfect–but the fact remains that trillion dollars of derivatives are traded each day based on this formula & option market makers make money year after year & laugh all the way to the BANK.Similar to relationship between price of one month option versus price of 4 month option,If it takes one month to get M move, it would take 4 months for 2M move in the same stock in the same time period.This tells me that if human brain is kept out of the equation & in SYSTAMATIC(automated) investing in stocks, take profit be 1/2 the size of stop loss( & we already know that markets are RANDOM) then over thosands of RUNS investor shall make a whole lot of money.Out of four trials one makes one dollar 3 times & loses 2 dollars one time with net profit of one dollar–all one has to do is that when take profit is HIT, cancell the unfilled stop loss order & repeat the process over & over.Investor stays direction neutral at all times with no bias long/short.This is not just a theoretical question–if you spend few hours pondering this question I bet your next 5 generations can make tons of money over the next 500 years,just clicking on the LAP TOP & would never have to look for A JOB.Any criticism would be appreciated.Thank you for your help in advance.
    Dr Prem Nath MD cell# 845 641 6778 email indus68@gmail.com

  11. What is the holding period for this test, i.e exit criteria? Are you holding for 5 days, 1 month, 1 year? This could have a significant impact if it is incorporated in a system. High volatility for a short time period is a lot different when holding for a long time (long term investment).

  12. Hey! This is my 1st comment here so I just
    wanted to give a quick shout out and say I truly enjoy reading your articles.
    Can you recommend any other blogs/websites/forums
    that deal with the same topics? Thank you!


  1. [...] Low risk outperforms high risk [...]

  2. [...] Low risk outperforms high risk [...]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

Join 306 other followers

%d bloggers like this: