I am in the process of evaluating my portfolio for my second quarter update. The first step in this process is deciding whether to make new purchases, or whether to just reinvest in the stocks I already own. I presently have 50 positions, and if I continue to find stocks that meet my criteria, I will continue to add them to my portfolio.
A previous article I wrote discusses my thoughts about the number of positions I wish to have in my portfolio. Basically I am happy to have as many as 100 positions, because the more positions I have (as long as they meet my buying criteria) the more I will be protected from a dividend cut. But eventually, I will get to the point where I am just looking to either add more shares of the stocks I already own, or rebalancing them, rather then adding any new positions.
When I get to that point, I have wondered what is the best way to go about it. Some argue that a straight DRIP system is the best way to go. Reinvest the dividend directly back into the stocks that paid them. This may work fine for dividends I receive (although I am going to choose NOT to do this, as you'll see later), but I also receive pension contributions during the year. So I have new money coming into the account. How should I allocate this money? Should I simply buy or sell shares to maintain an equal dollar weighting? Should I attempt to buy and sell shares to maintain or achieve a certain yield? Or should I look to sell overvalued shares and buy undervalued ones? In this article, I am going to discuss a method I have chosen which allows me to do the last of these option, buying more shares of the stocks i own that are undervalued.
First, I'll refer you to an article by Chowder where he discusses dividend reinvestment. He makes a very strong argument as to why you should use dividend reinvestment directly back into the stock that paid you the dividend. Setting up DRIP's for each of your stocks is the best way to do this. Unfortunately for me, this option is not available to me. Except for a few stocks held in a brokerage account: General Electric (GE), Annaly (NLY), and Wells Fargo (WFC), my portfolio is in a pension trust account, and they do not offer automatic dividend reinvestment as an option. I could manually keep track of the dividends paid by each stock, and put in buy orders in that amount for each stock, but for 50 or more different stocks, that would be labor intensive, and I would get charged a commission for each order. I have decided that It is simply not worth it for me. I believe there is a way that in the long term will serve me better.
PAAY: What I have chosen to do is collect all the dividends paid into my account and selectively reinvest them into the particular stocks in my portfolio which appear to be undervalued. As a dividend growth investor, I am always looking to achieve a safe, steady and rising string of dividend payments. If I have added certain stocks to my portfolio, then they must have met my criteria for inclusion, including, obviously, their dividend payment history and their yield. As time goes on, I will look for opportunities to buy more of these stocks when I think they are undervalued.
There are many ways to try to determine if a stock is undervalued. Looking at the PE ratio, Price/book ratio or Price/sales ratio are a few of them. I believe that another way to determine if a stock is undervalued is to look at its present yield as compared to its average yield in the past. If a stock has a yield of 2.5% over the past few years, but now has a yield of 3.0%, that, to me, is an indication that it may be undervalued. As long as it still meets my criteria for owning the stock, as long as the company still has a good earnings and dividend history, and no indications that the dividend may be cut in the near future, then I will consider adding to my position. So, when it is time for me to reinvest the dividends I have collected over the past quarter, and invest new money that has come in, I calculate the Percentage Above Average Yield [PAAY] for each of the stocks I own.
To do this, I use the weekly adjusted close for each stock. I calculate the present yield based on the weekly close and the most recent regular dividend payment (special dividend payments should not be included), and then determine the average yield for the past year. Basically, I calculate a 52 week moving average using the weekly closing yield. I then divide the most recently weekly yield by the 52 week average yield to see what percentage the stock is above or below its average. In this way, I am comparing each stock to its own average yield. It does not compare one stock's yield to any other stock's yield. Just because one stock has a 5% yield and another has a 3% yield does not mean the 5% yielder is undervalued compared to the 3% yielder. If the stock with a 3% yield usually has a 2% yield, then it's PAAY will be 50% (50% above its average yield). A high PAAY could be an indication that it is undervalued. A stock with a 5% yield, that usually has a 6% yield, would have a PAAY of -16%. This would be an indication that it is overvalued. This is, of course, always with the caveat that the company itself is still healthy, and that the price is not dropping because the company (and dividend) is in trouble.
Just to show one example, I will use CSX Corp (CSX). On 9/30/2011, when I would have been doing my quarterly review, it had fallen to $18.09. The dividend was .12 quarterly, and the yield was 2.65%. At that time, the 52 week average yield for CSX was only 1.78%. Therefore the PAAY for CSX was 49.09%. So the yield for CSX was almost 50% higher then usual, which to me is an indication that it was undervalued. CSX is now at $24.63 (as of 3/29), so had It been bought at that time it would have returned 36.15%.
Now back to the portfolio. Once I have calculated the PAAY for each stock in my portfolio, I take the dividends that I have collected over the past quarter, and new money that has come into the account, and divide it amongst the ten stocks with the highest PAAY, buying an equal dollar amount of each.
I ran a backtest to see if this method had a reasonable chance of working. I took the 50 stocks that are presently in my portfolio and used them for the backtest. I went back to January of 2008 and assumed I purchased $10,000 of each on January 4th (the weekly close). I then ran two scenarios. In the first, I ran a DRIP plan for each stock, but also divided my $12,000 quarterly pension contribution equally amongst each stock ($240 per stock), using it to purchase more shares. So, each quarter, more of every stock was purchased using the $240 shares of the pension contribution, plus whatever that particular stock had paid in dividends during the quarter.
In the other scenario I took the $12,000 pension contribution, plus all the dividends collected during the quarter, and used that money to make equal dollar purchases only in each of each of the ten stocks with the highest PAAY.
A total of $752,000 was put into each portfolio over the 5 year period of the back test. I ran both scenarios through March 29th, and here are the resulting portfolio values:
DRIP scenario --- $1,484,125
PAAY scenario --- $1,813,133
As you can see the system in which I use PAAY to determine which stocks are undervalued, and reinvest dividends and pension contributions only in those stocks, gives significantly higher value then a straight forward DRIP plan. Of course I understand the risks of survivorship bias when doing back tests. Back test results can not guarantee future results. But, to me, it shows that the method may, at least, have some merit. And it seems to make sense. If you are buying stocks that are undervalued based on their PAAY then not only will you buy shares that may appreciate more (to return back to their usual yield), but you also buy them when they are at a higher yield, thereby increasing the over all yield of your portfolio, and therefore the dividends you will collect in the future. I want to point out that PAAY is not meant to be used for initial purchases. It is only for adding shares or rebalancing.
As I update my portfolio on a quarterly basis, and show these updates on SA, I will show the calculations I make, the stocks in my portfolio that have the highest PAAY, and how I use it to buy more shares. In this way there will be a real time, real world example of how it can be used, and whether or not it can be successful.
Disclaimer: I am not an investment advisor. Nothing I write should be considered to be a recommendation to buy any particular stock. I am simply discussing and explaining the methods I use. Every investor should do their own due diligence.