How to Get Started Picking Individual Stocks:
This article explains how to get started picking individual winning stocks
based on fundamentals.
There are basically two main ways to pick stocks:
The first way is based on Charting or price patterns. This method looks at
charts or profiles of how the stock's price has risen and follow. Some
practitioners have elaborate methods of looking for patterns and believe that
they can pick the highs and lows. One variation is to simply try to buy whatever stocks seem to be rising and then sell if the price begins to fall. Trading
stocks based on price patterns is known as technical analysis. I find that
name to be an unfortunate misnomer. I don't consider it to be technical in that
it is usually not based on mathematics but is based on patterns. Almost all short-term
traders use some form of price pattern method.
Academic types mostly dismiss Charting out of hand, although they have lately admitted that momentum strategies have tended to work
to some degree.
Overall, while charting might work it makes little or no sense to me. I don't have the stomach to
trade rapidly in and out of the market. I will have nothing further to say about
charting or momentum strategies.
The second big category of stock picking methods is called fundamental
analysis and this is the method that I follow and recommend.
Fundamental analysis attempts to identify stocks that are under-valued based
on their current and future expected earnings and or cash generating capacity.
Fundamental analysis can be pursued on a number of levels from very simple
methods to quite complex methods.
Level 1 - "Mechanically" Buy stocks with low Price to Earnings ("P/E")
ratios. P/E ratios are published in newspapers and other sources. Many financial
Web Sites have a "screening" facility that allows you to easily identify stocks
with the lowest P/E ratios.
Advantages: I understand that various academic studies have proven that this
methods works (or at least that it works based on historical data, there is no
way to prove that it will work in future). A major advantage of this method is
that it is very easy to apply. It is also very intuitively appealing, using this
method you get more earnings for each dollar spent on a stock.
Disadvantages: When applied mechanically to lists of low P/E stocks it will
result in the purchase of some stocks that have a a low P/E for good
reason. For example if a stock has unusually high earnings in one year that are
not expected to be repeated in future, then it will usually trade at a low P/E.
When earnings return to normal, the P/E rises as the earnings fall and you find that you actually
don't have a low P/E stock. In other cases the stock may be the subject of
negative news. For example as ENRON collapsed it sported an increasingly
attractive looking P/E. The problem was that the historic "E" part of that ratio
was now thought to have been a fraud and in any event it became clear that the
earnings were never going to return to historic levels.
Overall, this can be a reasonable way to get started picking stocks.
However, it is crucial to pick at least 6 and preferably 10 or more low P/E
stocks in order to diversify away the risk that at least some of your low P/E
stocks are on their way to bankruptcy.
Level 2: Apply discrimination in buying low P/E stocks. At this level the
stock picker begins to look at how representative the earnings are that drive
the P/E. For example P/E ratios that are based on analyst estimates of future
earnings rather than mechanically calculated based on the previous 4 quarter
earning can eliminate most of the problems with unrepresentative historic
earnings.
The stock picker may also decide to restrict purchases to low P/E stocks with
good earnings potential due to being in more profitable industry segments.
The stock picker begins to realize that higher growth can justify a higher
P/E. The PEG ratio divides the P/E by the expected growth rate to attempt to "normalize"
for growth. Low PEG ratio stocks have a low P/E relative to their expected
growth.
The stock picker may also begin to look at other ratios such as the dividend
yield and the price to book value ratio to assist in the stock picking decision.
This level has infinite variations but generally consists of picking
relatively low P/E stocks with good growth prospects relative to the P/E.
Advantages: Compared to level 1, this level should weed out some low P/E
stocks that are probably on their way to bankruptcy or where the apparent low P/E is
based on an unusual blip in earnings and is therefore not really valid.
Disadvantages: When using the PEG ratio, a problem arises as to which growth
rate to use. Last year's growth or next year's expected growth is not likely
representative of the longer term growth rate. The PEG ratio only very
imperfectly normalizes for growth. Investors can be fooled by a low P/E or low
PEG that is associated with a cyclic stock. A cyclic stock, when it is near
the top of its earnings cycle, should and usually does trade at a much lower P/E
or PEG compared to a non-cyclic stock.
Level 3: Buy stocks that are priced low compared to their true value
based on expected future cash flows to the investor. This method sounds very
daunting if not impossible. The investor is required to forecast the future cash
flows, perform a "present value" analysis and then compare that to the stock
price. I call this the Price to Value Ratio and it is the ultimate way of
picking bargain stocks. Warren Buffett follows this method and talks about a
goal of "buying dollar bills for 50 cents".
Fortunately there are some simplifications that can make this seemingly
impossible calculation reasonably straightforward. First we must start with a
normalized level of earnings per share for the most recent year or the upcoming
year that adjusts for unusual gains and losses. This earnings per share level
provides a base from which we can project growth. The easiest way to obtain this
figure is to use analyst forecasts of future earnings per share. I'm generally
loath to trust those estimates but they do have the advantage of usually not
containing any forecast for unusual gains or losses, so in theory they do
represent a normalized level of earnings. The other way to obtain this figure is
to start with actual net income and make necessary adjustments based on
information in the financial reports. Also many companies report their
"recurring" earnings per share after adjusting for unusual items.
The next step is to forecast the longer term (5 to 10 years) average earnings
per share growth rate. This can be based on the past growth rate and should
take into consideration the economic outlook for the company and ideally its
industry.
My policy is to be conservative on the growth rate assumption and to never
assume a growth rate higher than 20% and rarely higher than 15%. (And, for more
mature or stable companies, 5% might be more realistic). Assuming a very
high growth rate is essentially saying that you are willing to pay-up now for
that growth rate and leaves too much down-side risk and little up-side risk. By
assuming a conservative growth rate you leave yourself plenty of up-side risk.
Next we assume that we will cash out of the stock after the 5 or 10 year
growth period by selling the stock at an assumed P/E ratio. My policy is to
assume that the P/E will revert to some conservative value such as 12 to 15 or
sometimes lower. Again this assumption leaves room for an up-side risk.
Finally we calculate the present value by calculating the value of the dividends
received each year and proceeds from selling the stock at the end all
"discounted" at some reasonable return rate such as 9%.
Voila, you now have a calculation of the amount that the share is worth based
on your conservative (but not overly conservative) estimate. If this value is
sufficiently higher than the stock's current market price, you buy.
There are many variations of the method, but the above steps should be
common. Additional steps would include attempting to understand which industries
have the best growth prospects in order to fine-tune growth rate assumptions.
One variation on this method is to restrict the analysis only to stocks that
have shown a long and consistent history of consistent earnings and earnings
growth. It is a lot easier to predict the growth of stocks that have exhibited a
high level of consistency in the past. In fact it amy not make much sense to
attempt to apply this method to any stock that does not have a consistent
history.
Advantages: This method is well grounded in fundamental finance theory. It
"should" work on average as long as you can find under-valued stocks.
Disadvantages:
For cyclic and commodity linked stocks there is essentially no such thing
as a "normal" level of earnings so it is probably best to not attempt this
method on resource and commodity based stocks (oil, gas, mining etc). This
method cannot be applied to early stage companies that have yet to earn a
profit. This method is also a fair amount of work.
Summary:
Getting started picking your own stocks based on fundamentals is very easy if
you start with the simplest method of picking low P/E stocks. Enhanced methods
are then available as you gain skill, understanding and experience. An
experienced and knowledgeable fundamental analyst can eventually get to the
point of being comfortable calculating the "true" value of a stock and
selecting low "P/V" - Price to Value ratio stocks.
If you are comfortable with the above material, then you are ready to
graduate to my more detailed and advanced article on
Stock Picking.
Shawn Allen, CMA, MBA, P.Eng.
InvestorsFriend Inc., (c) February 27, 2002