Tuesday, October 23, 2012

The Modern Portfolio Theory Cliff

I have been reading a lot about correlations of asset classes increasing since 2007ish.  Of course this is something I can easily see with my own eyes as it seems like every day in the market is either a risk on day (buy any kind of stock or commodity) or a risk off day (buy Treasuries).  Increasing correlations present another problem for Modern Portfolio Theory (MPT).  Most practioners assume that asset classes like large stocks, small stocks, international stocks, value stocks, growth stocks, commodities, and high yield bonds are not that correlated (meaning combining them in a portfolio actually gives you diversification) and that correlations will remain constant over the investment period.    This presents a couple of problems:

1. Correlations are not constant, as evidenced by increasing correlations lately.

2. Taking a straight line correlation ignores the most important type of correlation there is, underwater correlation, or how correlated asset classes are during a drawdown.

Two assets may appear to be uncorrelated when you look at a long time period, for example stocks might appear to be relatively uncorrelated with oil.  However, during an economic decline stocks will generally suffer, as will oil because of a decrease in demand.  So having these two assets could help some during bull markets but could just give me two different ways to lose money during a down market. 

One time when increasing correlations could help is during a bull market.  If my entire portfolio is correlated to stocks then my entire portfolio might be increasing during a bull market.  This may seem to be helpful but my 9 year old son, and just about anyone else, can make money in a bull market.  Where diversification is supposed to help is in a down market.   Increasing asset class correlations plus a down market is a recipe for disaster.

The solution:  Diversify by methodology, time frame, and security basket, basically have different return streams that are uncorrelated on an underwater basis.

Another Study Showing Trend Following Works

A Century of Evidence on Trend Following

Great study done by AQR showing that trend following has produced consistent profits since 1903.   Some powerful points made by the study:

1. If you look at the ten largest drawdowns in a classic 60/40 portfolio since 1903, a trend following portfolio (as designed in the study) would have had positive returns in 9 out of 10.

2. Trend following works very well in Bear Markets because:
"The intuition is that the majority of bear markets have historically occurred gradually over several months, rather than abruptly over a few days, which allows trend-followers an opportunity to position themselves short after the initial market decline and profi t from continued market declines."
The paper also rightly points out the many trend following strategies have had trouble over the past few years.  While this is not out of the norm looking back to 1903 it still makes sense to explore the reasons why.
Trend following does best when there are consistent longer term trends, the investor can get in after a trend has been verified and hold on until the trend reverses.  If you look at just the monthly returns of the S&P 500 over the past 3 years, the risk on/risk off markets created by the European crisis have made it difficult to follow a trend following strategy.  For example in 2010 February through April the market had a great run, only to be down 7.99% in May and down 5.23% in June.  We see the same pattern in 2011 and 2012 where it looks like an uptrend is forming, only to be followed by a quick reversal.   Depending on the time frame the trend follower is looking at, he or she might get into the market because of  a large up move just in time for the reversal (of course this argues for varying your time frames). 
Have markets fundamentally changed so that trend following is no longer viable?  Not at all.  It is human nature to create trends.  Markets are not rational or efficient.  When investors see something going up they pile in.  It is only the complexities of the European situation over the past couple of years that has caused trendless markets.  This will either pass, paving the way for a market rally, or drag us down into a massive market decline.  Either way, trend following wins.

Friday, October 5, 2012

Combining Factor Investing & Relative Strength

I get Morningstar Advisor magazine free every month.  I usually find little value in it as most articles are buy and hold drivel but this month they had some interesting articles devoted to factor investing.  Numerous studies have been done that identify certain factors that have outperformed the market:

1. Value investing
2. Momentum
3. Smaller stocks
4. Low Beta Stocks

Momentum is obviously something that has always interested us.  As has value, since it is not correlated with momentum.  We are constantly studying ways to apply value across asset classes and see how it can be combined with momentum strategies (perhaps the subject of another post one day).  The others have never really had any interest to us as they may outperform the market but they also will get creamed when the market goes down.  Momentum is the only one of these factors that can be tweaked to avoid large losses.  However, the articles got me thinking about combining relative strength (buying the one asset out of a basket of assets that has the strongest performance over a period of time) with factors.  So I decided to study whether this would work.  I tried to find ETFs or mutual funds to represent each of the factors that have been around a long time (I had some trouble with this and more work needs to be done to find stuff that has been around longer).  I settled on the following:

Momentum- Powershares Tactical Leaders
Low Beta- SEI Managed Volatility
Value- iShares Large Cap Value
Small Cap- iShares Russell 2000
Tactical- PIMCO All Asset (Threw this in there as well to provide a little diversification)

I designed a simple relative strength system that would rotate monthly among the best performing fund and would be in cash if nothing was in an uptrend.  I was able to go back to 3/2007 for a common history but since we needed time to determine relative strength my first trade wasn't until 2/25/09.  Here are the results from 2/25/09 to 10/4/12:

Average Annual Return: 21.11%
Maximum Drawdown: -15.37%
Worst Month: -7.16%
Sharpe Ratio: 1.32
MAR Ratio: 1.37

The performance numbers and MAR are very attractive but the drawdowns and worst month are a little high for my taste.  I also really would have liked to see what this would have done in 2008.  That being said, there is definitely enough here for further study and refinement.