Tuesday, May 31, 2011

Stock Allocation: A Way of Increasing Returns

If you are an investor, chances are you would have heard of the term "asset allocation".  The idea is to maintain a certain percentage of your portfolio in different asset classes, for example, stocks, bonds, etc.  Why would you want to do that?  The answer is simple: it increases your return.

Suppose you have Asset Class A and Asset Class B, and you want to maintain the percentage of each to 50%.  Say you had invested $10000 in each and Asset Class A rose 50% in the first year while Asset Class B dropped 30%.  At the end of the first year, you would have $15000 in A and only $7000 in B.  You now have an imbalance.  What you do is sell A and buy B so that you have equal amounts of both (50% of each).  At the start of year 2, you would have $11000 of each of A and B.  If B then rises 50% in year 2 while A falls 30%, you would have $7700 in A and $16500 in B.  This would total to $24200.  If you had not rebalanced your portfolio, you would have had $10500 in A and $10500 in B, totaling only $21000.  Asset allocation works very well when your asset classes are not correlated (i.e. they do not move in the same direction all the time).

The really curious thing is this: your portfolio returns is actually greater than the average of the returns of both investments.  Back to that example above.  A had risen 50% in the first year and dropped 30% in the second year.  Its returns was 5% over 2 years.  B had dropped 30% in the first year and risen 50% in the second year.  Similarly, its returns was 5% over 2 years.  However, the rebalanced portfolio ended with $24200, which works out to be 21% gain over 2 years.  Both only returned 5% over 2 years, but somehow, your portfolio returned 21%! Isn’t that cool?

The key lies in correlation, the lack of it actually, between the assets.  What we are effectively doing is buying low and selling high.  Now, you all know that I am a stock kind of guy (I only hold cash or stocks).  Therefore, instead of asset allocation, I propose doing a stock allocation.  I currently have 4 stocks in my portfolio.  Chances are, they will be slightly correlated.  That is, they tend to move in the same general direction, but some times do deviate.  In Figure 1, you can see the performance of the 4 stocks over the past 2 years.  As you can see, they definitely follow different paths.  The goal is to allocate a certain percentage of your portfolio to each stock.  In this case, I may want to allocate 25% of my portfolio in each stock equally.  It doesn't necessarily have to be an even split.  I may have a belief that True Religion would outperform the other 3 stocks.  In that case, I might want to give it a 40% weighting and the rest 20% each.  It is entirely up to your discretion.

Figure 1: The Stocks in My Portfolio: Google, True Religion, First Solar, Synaptics

Dumb Stock Allocation
So, you now get the general idea of stock allocation.  You can either use a "dumb" stock allocation strategy or a "smart" stock allocation strategy.  The dumb strategy is to simply set an allocation target percentage and every few months (or weeks or years), rebalance your portfolio to that percentage.

Let’s look at a real life scenario to see how that works out.  Instead of 4 stocks, let's keep things simple and use 2 stocks.  The table below shows quarterly price data of Google (Ticker: GOOG) and True Religion (Ticker: TRLG), starting from January 2005.  If you had bought and held the stocks, you would have gotten 193% returns over the 25 quarters.  If you had rebalanced your portfolio every quarter using dumb stock allocation, you would achieve 297% returns over the 25 quarters!  It works alright!

Smart Stock Allocation
"Smart" stock allocation, on the other hand, requires some smarts from you.  I would definitely say that this is a more advanced strategy with a combination of Rule # 1 Investing and pure gut feel.  I'm starting to employ this strategy...So, we're into uncharted territory here!  Ok, ok, I'm exaggerating...investors have done this many, many times already.  In fact, many people do this without knowing it.

Here's a real life example.  If you have followed my trading history, you would have noticed that I started buying True Religion around this time last year and began to sell it just recently.  I have also bought some First Solar recently.  This is me doing some smart stock allocation.  True Religion has run up to the high $20s in the last couple of months.  My cost was $22.55.  Because of this run up, the holding has grown in proportion.  Meanwhile, First Solar spiked up for a little while, but is now trading in the $120s, off its 52-week highs in the $170s.  It has decreased in proportion.  Therefore, I sold some True Religion and bought some First Solar.  The likelihood of First Solar moving back up to $170 is greater than True Religion moving up the same amount to $42; the True Religion stock had already gone up from $22 to $29 in a short amount of time.  If things work out as they should, we should see First Solar bounce back up while True Religion takes a breather and stays steady or retreats a little.  Hopefully, this will work out.  Pray for me!!

Some Major Major Caveats

Whether it's dumb or smart stock allocation you are using, you must have good reason to own any particular stock.  Please do not, I repeat, do not use stock allocation blindly by just picking whichever stocks you like and hope that you will make money.  If you pick 5 overpriced stocks, as the prices tumble 50%, stock allocation won't be of much help.

You need to know the intrinsic value of the stock.  My advice is to use Rule #1 to help you determine that.  Never buy a stock that is ridiculously overpriced.  For that matter, never buy a stock that is fairly priced.  Always buy a stock that is underpriced.

If you are using smart stock allocation, use technical analysis to help you figure out when to buy and sell.  Rule #1 talks a bit about the "3 tools", but from my own experience, I prefer using supports and resistances.  However, the 3 tools in Rule #1 are more objective and require less experience.

Lastly, know that stock allocation might not increase your returns.  If your one stock does really, really well for an extended amount of time compared to the others, you would end up selling more and more of it prematurely.  Therefore, study the movement of the stock.  If it tends to gyrate between supports and resistances, then it would be wise to rotate in and out of the stock as it moves up and down.  However, if the stock tends to be steadier and rises constantly, stock allocation may not end up helping you.  I would recommend doing some back testing to see how stock allocation works out.  By back testing, I mean running a simulation of the past few years of data and see if the returns are magnified or decreased (similar to what I did above with Google and True Religion).

Let me know how this works for you!