The goal here is to determine, for each stock previously qualified, the most that should be paid when accumulating, and the price level above which the stock is not worth what it is currently selling for, and is at least a candidate for selling. These price levels then dictate when to buy or sell, or at least consider doing so. Other than picking stocks on their way to significant under-performance, with dividend cuts, acquisition by a competitor at a distress price, huge dilution, or even bankruptcy, the most serious challenge in achieving reasonable returns is to avoid over-paying for stocks. On the other hand, one should avoid over-attachment to a position, and have a methodology that at least considers selling some or all of a position if it becomes substantially over-priced.

Before looking at valuation methods for individual stocks, one needs to consider the history of stock market prices. Two excellent books on this topic are "Unexpected Returns" by Ed Easterling and "Irrational Exuberance" by Robert Shiller. (A substantial list of excellent books is presented in the essay on resources.) It is often said that earnings drive stock prices, and this is true to some extent. However, history has shown that stock prices can range widely for a given level of earnings. That is, the "multiple" assigned to a stock, defined as the price/earnings (PE) ratio, varies greatly from one time period to another. Studies of average PEís across groups of stocks, such as all stocks in the S&P 500 index, over varying time periods reveal that stocks can trend higher or lower for many years, beyond what earnings variations alone would indicate. These trends are defined as secular bull or bear markets. The best returns, certainly for the "buy and hold" investor, have occurred when stocks have been purchased at a time when PEís are low (bear market), while poor returns are likely if stocks are purchased when PEís are high (bull market). There is not much an individual investor can do about where we are in the current long term cycle, even if all the experts agree (which they donít) on what the current cycle is and where we are in it. Who wants to wait 5 or 10 or even 20 years for the PEís to get down to where they are extremely low, such as 1932 or 1981? Still, an investor should be aware of these cycles, if for no other reason than to be encouraged that an active, opportunistic strategy is the best way forward considering where PEís are today. For comparison purposes, the Shiller 10 year inflation-adjusted S&P 500 PE has ranged from less than 5 (1920) to more than 44 (1999) over the last 100 years. The average is around 16. The most recent buying opportunity of a lifetime was in 1982 (PE around 7). Today (October 2011), the PE is around 21. Thus, in spite of the market decline since 2007, based on the Shiller PE research, stocks overall are still not cheap.

While that is all very interesting for background purposes, we are not that interested in the total market, just our list of qualified stocks. The PE at the individual stock level is the most commonly recognized value statistic. The PE shown most commonly on web sites is based on the trailing 12 months earnings as reported using GAAP (Generally Accepted Accounting Principles). To really get to true value based on on-going earnings capability, S&P introduced the concept of "core" earnings, which excludes one-time items that muddy the picture. Ideally, this is the PE the dividend investor should look at. Since it is not commonly available, PE based on GAAP can be used, just be aware that it could be distorted if significant one-time items have occurred. PEís for large-cap dividend stocks are usually lower than the overall market. While there is no absolute rule for using PE as a trigger for buying or selling, it is useful to review it and compare to other stocks to get a feel for how a stock is currently being valued by the market. The lower the better, as far as obtaining value is concerned, but if a stock has a PE considerably lower than its peers, there is probably a reason. Find out before buying. Conversely, if PE is higher than comparable companies, the stock may not be much of a bargain at this time.

Probably the most important valuation metric for the dividend investor is the yield. That has a direct impact on the potential return on investment. While there is much more to consider in evaluating a stock besides yield, if the yield is not sufficient, there is no reason to consider it further. I have some yield ranges that I need to see before I will consider certain types of companies. A yield in the 3% range is reasonable for a "blue chip" industrial company. A utility or REIT (Real Estate Investment Trust) should pay 4% to 5%. An MREIT ( mortgage REIT) would need to be in the double-digits to justify the risk. A BDC (Business Development Company) would need to be 7% or more to be attractive. An MLP (Master Limited Partnership) would need to pay 7% or more. For smaller (non-IRA) accounts, an MLP holding is probably more trouble than it is worth, considering the tax complications for the investor. While most advisors say to avoid MLPís in IRA accounts, I believe the risk that "Unrelated Business Taxable Income" will exceed $1000 for a given year, thus requiring tax to be paid even though the account is an IRA, is negligible for small IRA accounts with one or two MLP positions of 100 or fewer shares. Other than this issue, the onerous MLP tax reporting requirements for taxable accounts do not apply to IRAs.

Professionals have substantially more sophisticated techniques for valuing stocks, most of which are based on cash flows provided by dividends or earnings, levels of interest rates, inflation expectations, and dozens, if not hundreds, of other factors. I would encourage any investor to learn more, not that you can expect to quickly become proficient at valuing stocks like the professionals employed by institutions, but so you can at least be aware of the universe of methodologies out there.

Stock prices fluctuate. Depending on the overall market, recent earnings, investor perceptions, and a million other things, the price for a stock can range widely over the short term even though the intrinsic value of the company as an on-going business logically does not change over this same time period. To get a feel for how the current price compares to the past, the investor should review a stock chart looking back 20 years, 10 years, 5 years, 3 years, 2 years, 1 year, 6 months, 3 months, 1 month, and intraday for the most recent trading sessions. How did the stock do in the 2000-2003 bear market? How high did it get when the overall market reached the 2000 and 2007 highs? How low did it go during the financial crisis of 2008 and 2009? How does the current price relate to how it has been trading recently, over the past year? You certainly donít want to buy at a high, particularly if the rise has been sudden and extreme. If the stock has been on a steady decline over a long time period, it may look like a bargain, but be wary Ė ask yourself, is there any reason to think the decline is coming to an end? As a general rule, most stocks will move with the market. If the stock has been vetted as a quality issue and is down along with the general market as a decline occurs, that is the time to buy or add to your position.

Now a word about ex-dividend dates. You should be aware of the next anticipated ex-dividend date (either declared or estimated based on the dividend frequency) of any stock being considered for purchase or sale. Numerous web sites give this information, particularly E*Trade. But be aware that occasionally web sites provide erroneous data regarding dividends. The most reliable source is the dividend announcement by the company in a press release. Now, all else being equal, I would rather be paid sooner rather than later. Thus I will be more eager to buy when the ex-dividend date is only a few days away rather than months away. But avoid the trap of being too eager to get the dividend and ignoring the guidelines to avoid overpaying. Never let yourself be pressured into buying. One would think that stocks would tend to rise over the last few days preceding an ex-dividend date, and for very high-yielding stocks, that is often true. So usually if you want the next dividend, you should not wait until the last day or two. But this tendency is a weak force, and easily is more than offset if the overall market is declining substantially. This means that you should not give up on a buy if the stock is too expensive with only a few days to go, it can still decline, even on the last day. What about after the ex-dividend date has arrived? Now, as a dividend player, you realize the opening price on the ex-dividend day is reduced from the previous dayís close by the amount of the dividend. But sometimes it seems like investors donít realize this, and upon seeing the stock opening lower, proceed to sell it off as if something is suddenly wrong with the company. This can be a good time to buy at a discount if the drop is 2 to 3 times the amount of the dividend. Remember, another dividend will roll around soon. To summarize what the point is of all these ex-dividend factoids -- be aware of ex-dividend dates, but donít over-react to them.

After considering all of the above, and the buy-under prices specified in any advisory services subscribed to, set your buy price. This is the most you will pay. It will usually be lower than the current market price. Now comes the hard part; waiting for a buy opportunity.

Now, about position size. My guideline is no more than 5% in any one stock. For a $100,000 portfolio, that would imply a limit of 100 shares for a $50 stock. As a simplification, I have set 100 shares as my full position limit for all stocks. That means I will be under 5% for stocks under $50, and over 5% for stocks over $50. My acquisition size limit for a single purchase is 25 to 50 shares. To start a new position, if the stock in my judgment represents an outstanding value, I will buy 50 shares. If I believe a better opportunity will come along sometime, but probably not soon, and the current price is at least reasonable and an ex-dividend date is approaching, I will only buy 25 shares. Either way, I have plenty of room to buy more before I max out. For stocks under $20, I usually double the position size and acquisition limits. After the initial buy, the attractiveness of subsequent opportunities is in relation to where I initially bought. Baring extreme market movements, such as a 100 point drop (or more) in the S&P 500 Index over a few days, I would not expect to buy again until some time later, at least several days to several weeks. If the stock drops by 2% to 3% or more from my initial buy, I will consider adding to my position. This assumes nothing has occurred to upset my original premise regarding the stock. After two buys, I require an even better bargain to buy again. Do another review before buying more, to be sure nothing fundamental has changed. Once I get to 100 shares, I consider myself maxed out on that stock. I will only buy more if the stock is a top-quality holding and has fallen to a level where it is too good of a buy to pass up. I will allow myself to get as high as 150 shares, with the excess over 100 being considered "inventory" to be sold upon a rebound.

Now, what if the vetting process failed, and the stock keeps on declining, and the decline is not due to a general market sell-off. If all stocks are dropping, as occurred in the fall of 2008, donít sell out in a panic. Selling because of a decline only applies to a stock dropping in the absence of a generalized sell-off. Compare the current price to the average cost per share of the position. Once the decline exceeds 20% of cost, it is time to put the stock on "stop loss alert". If the decline continues, sell. Another reason to sell at a loss is if the dividend is eliminated. Usually if that happens, it wonít take long for the price to decline more than 20% anyway. If the dividend is cut but not eliminated, the stock may be worth keeping. Evaluate anew considering the new dividend, especially considering if it is sustainable. Take into consideration what is going on with the overall market, along with cash levels and alternative uses of capital that may be available. One final thought on stop-losses. Consider the case of the Arkansas high-school football coach who never punts or kicks a field goal. The entire field is 4-down territory. His record is remarkably successful. If you hate taking losses as much as I do, maybe you want to refuse to punt, ever. You only get out when the company goes bankrupt or otherwise disappears (is acquired). If you follow this approach (which is anathema to probably all investment advisors and Iím not really recommending it, just entertaining the thought), you must have a very high bar for vetting stocks in the first place, and limit your exposure to 3% or less of your portfolio even if a total loss occurs. Or you could take a hybrid approach, say divide your stocks into two tiers. The top tier stocks are those that you will not sell at a loss (until the stock is acquired at a distress price or goes bankrupt), and the second tier are stocks to sell if they decline a set per cent. Or maybe take a less extreme approach for the top tier stocks, just give them more room. For instance, only sell stocks at a loss in the top tier after a larger decline, say 30%, 40%, or even more, and have a lower threshold for stop-loss selling of the lower tier stocks. The point is, define a strategy that suits your risk tolerance and follow it, donít wait until disaster strikes (be assured it will at some point, for some stock you own) to start thinking about what to do.

If stocks are on a tear, a stock can increase in price such that it is above the "full value" price. This price will be per advisory services subscribed to, price targets per brokerage analystsí reports, or possibly your own analysis based on PE or yield. Donít sell in a knee-jerk reaction thinking you can buy it back right away when the price drops. That opportunity may not come for a long time. Factors to consider are cash levels and alternatives available, the over-all market level, how long the stock has been held, and more. Sell only after considering all aspects. Just as in buying, a partial sell may be the best strategy, to at least take advantage of the price gain without sacrificing the entire position. If a small initial position was entered that in hindsight was a great buy, and no opportunities ever came up to buy at a better price, I consider it an "orphan" position, and will be more inclined to sell than if I have a full position. Also, if the stock has run up quickly, I am more inclined to sell, because it is more likely to run down just as quickly. When the market gives you a gift, you need to take it. It doesnít happen often. How much of a gain is needed to tempt me to let go of a good dividend stock? At least $10, sometimes $20 or more. A rule of thumb I use is, if I can capture a gain (after commissions) that gives me as much as two or more yearís worth of dividends, I will consider it. Four or more yearís worth is hard to pass up. But it also depends on what else is available for the funds that are freed up, and whether my cash level is high or low at that point.

With the strategy of slowly increasing stock holdings during declines and scaling out of positions as prices rise, cash levels should decrease during declines to as low as 10% to 15% after a lengthy decline, and increase to as much as 30% to 40% after a lengthy advance. Use these levels to restrain yourself both ways to avoid buying too much or selling too much at any one time. Stocks can always go lower than the lowest price imaginable, and higher than any price level that makes sense, based on fundamentals, on the upside.

Now, what about transaction costs of trading around in small lots? First, always consider the brokerage commission when deciding whether to make a trade. With the highest commissions at $10 per trade, and many brokerages offering less (direct access brokers offer much less), this consideration is not the inhibiting factor regarding frequent trading it once was. A 25 share trade involving a $50 stock is $1,250, so a $10 commission represents .8%, or .008. A $5 gain on 25 shares is $125, with two commissions (buy and sell) totaling $20, leaving a net profit of $105. For a position held a short time, say 2 or3 months or less, that may be a profit worth taking, again considering all factors.

The next essay considers the "nuts and bolts" of implementing the strategy, Resources and Account Management.