You've heard the well-worn saying “Don't put all your eggs in one basket.” That's why most financial professionals recommend you diversify your investments across a variety of assets.

Normally, you don't want to be 100% in stocks or 100% in bonds. You want a good mix of assets so that if one asset class is underperforming, there's a good chance another one is outperforming.

Typically, you want to own a mixture of stocks, bonds, real estate, precious metals, and maybe some commodities or other investments. The recent financial crisis is a good example of how this type of portfolio can balance things out.

While stocks and housing were crashing in 2008 and early 2009, bonds, gold, and commodities performed well. An investor who was well diversified lost less than one who was primarily in stocks and real estate. I knew plenty of people who lost everything because all of their money was tied up in real estate—the very same people who told me just two years earlier that “Real estate is the only way to make money.”

Next time someone tells you that one specific way is the “only way to make money,” figure out a way to short that person's net worth, because it's heading south within a few years. I guarantee it. I don't care if the way is gold, real estate, stocks, or Italian trattorias. When people are so arrogant about their investments (even if they're usually smart investors) that they've concluded it's infeasible for there to be a better way to make money, that means that investment is likely at a top and heading lower.

Within any asset class, it makes sense to diversify as well.

If you own a portfolio of rental properties, you wouldn't want to own houses that were all on the same block. If that block suddenly becomes undesirable, your portfolio will take a big hit.

You'd want to have houses spread out all over town or maybe even all over the country. If your house in Florida takes a big hit in price, perhaps the apartment in California will hold its value. If rental prices slide in New Jersey, maybe they're going up in Colorado.

It's the same with stocks and mutual funds. In fact, the Oxford Club, where I am the Chief Income Strategist, has an asset allocation model consisting of stocks, bonds, precious metals, and real estate.

Within the stocks asset class, we further sort (and diversify) them into large caps, small caps, international (further categorized into Pacific Rim and European), real estate investment trusts (REITs), and so on.

Bonds are diversified as well. Our bond recommendations include short-term corporates, high-yield corporates, and Treasury Inflation-Protected Securities (TIPS).

A portfolio of dividend stocks should be the same. Although it may be tempting to load up on dividend payers with 10% yields, that's likely a recipe for disaster. There's nothing wrong with sprinkling a few of those into a well-diversified portfolio to boost yield, but if all you are holding are stocks with double-digit yields, you are taking on way too much risk.

Generally speaking, you want to diversify your dividend-paying stocks across different yields and sectors.

You'll want industrials, technology, energy (often master limited partnerships [MLPs]), REITS, health care, consumer staples, and a host of other sectors.

You'll constantly have some group in the market outperforming and another underperforming. So by diversifying, you are trying to ensure you always have exposure to a group that is performing well.

If consumer stocks are weak, perhaps health care will remain strong. When the economy is starting to show signs of recovery, industrials should work.

There will always be a group on the rise, either because of the cyclical nature of stocks and the economy or because a certain sector gets hot.

If Warren Buffett suddenly announces that he is buying large pharmaceutical companies, you'll want to have already bought into that sector because they are likely going to take off for a few weeks or months. Deciding to get in once Warren Buffett proclaims that he likes those stocks on CNBC will be too late. The market will have already reacted, and the price of those stocks will be significantly higher half a second after the Oracle utters those words.

But if you have a diversified portfolio, where you already own some large pharmaceutical stocks, when Buffett says he likes big pharma, your shares of Bristol-Myers Squibb (NYSE: BMY) and Abbott Laboratories, which you bought two years ago, 15% lower, will take off.

And important for dividend investors, your yield will remain the same. It doesn't matter what the stock price is today; it matters what you paid for it.

That's an important difference for the investor who either needs the income today or is trying to build a wealth-creating portfolio for the long term.

The Oxford Income Letter Portfolio—An Example

I manage the three portfolios for the Oxford Club's Oxford Income Letter. The Instant Income Portfolio uses my 10–11–12 System to generate a high level of income today and even more tomorrow. The Compound Income Portfolio also uses my 10–11–12 System and is for investors who don't need the income today and instead want to use the power of compounding to grow their wealth by reinvesting the dividends. The third portfolio is the Retirement Catch-Up/ High Yield Portfolio, which does not use my 10–11–12 System and is for investors who can handle a higher degree of risk in exchange for higher yield.

I'm not going to mention what stocks are in the portfolios because by the time you read this, the portfolios may very well have changed. For more information on the portfolios, please visit www.oxfordincomeletter.com or www.oxfordclub.com.

However, I will tell you how the portfolios are currently diversified. Again, this may change by the time you read this, so don't take this as gospel. But as you'll see, it's a good mix of stocks that should let us participate in strong markets and keep us from getting badly hurt if any one sector or stock blows up.

As of September 2014, the Instant Income Portfolio consisted of eight positions. The stocks are in these sectors:

The Compound Income Portfolio has 15 stocks from the following sectors:

And the Retirement Catch-Up/High Yield Portfolio has eight stocks in these sectors:

We have a diversity of yields as well. As I mentioned, we can't just load up on stocks paying 10% dividends, for reasons I'll explain later in this chapter. Although we want the yield as high as possible, we need to take into account risk and the growth of the dividend.

I would rather own a stock paying a 4% yield that grows its dividend every year by 10% than one with a 6% yield and a dividend growth of 3%.

If I'm holding these stocks for the long term, a stock with a current 4% yield but 10% growth will yield more than the 6% stock with 3% growth in seven and a half years.

By year 10, the stock that started out with the lower dividend payment will yield 9.4%; the one that started out higher will deliver only 7.8%.

So, for long-term holders, dividend growth is just as, if not more, important than current yield.

The Oxford Income Letter portfolios' stocks have these yields (based on the price when they entered the portfolio)

You can see we've got a few stocks with high yields in there: In the Instant Income Portfolio we have one that yields 7.3%. In the Compound Income Portfolio we have two stocks that yield 6.8%. These high-dividend payers help us get the average yield over 5%, which is very healthy in today's market. But we also have some lower-yielding stocks, such as a defense contractor that yields 3.6%, an insurance company that yields 3.1%, and a technology company that yields 3.5%.

If something should happen to our 7.3% stock and the company has to cut the dividend, the share price will slide considerably. However, our 3.6% defense contractor and 3.5% technology company should stand up over the years, as they have for decades.

Now that I've established the importance of diversification, let's go ahead and talk about how to pick dividend-paying stocks.

Setting Up the Portfolio

The first thing you need to do is answer these questions:

1. What is your time frame?
2. What is the purpose of the portfolio: income or wealth creation?

If the answer to question 1 is three years or less, put down this book, and look at something less risky than stocks. Really, the only thing you should be looking at are money market accounts, certificates of deposit, Treasuries, and maybe very highly rated corporates that mature in three years.

If you need the money back in three years, you shouldn't be taking much risk with it. Of course, the bonds that are available will pay you practically nothing in today's interest rate environment, but at least you'll be sure that your money will be there when you need it.

Even blue chip stocks with 50-year track records of raising the dividend will fall in a bear market.

One of my favorite Perpetual Dividend Raisers, Genuine Parts, which has been hiking its dividend every year since 1956, saw its stock price cut in half from its peak in 2007 to the lows of 2008.

Granted, the financial collapse of 2008 and early 2009 was a rare event, but for anyone who needed his or her capital back in 2008, the reason the market fell or the uniqueness of the selloff didn't matter. The fact was, investors' money was not available when they needed it. It was gone.

By the way, patient investors who were able to ride out the storm of 2008 (and, I hope, reinvest those dividends at low prices) saw Genuine Parts' stock come roaring back, more than doubling in two years. As of September 2014, it was trading 70% higher than its high in 2007, before it started to slide. Shareholders who reinvested their dividends in March 2009 were able to buy shares as low as $27.05. The stock hit $90 in early 2014.

In fact, the 2008 collapse taught many investors a valuable lesson—to hang on despite a nasty bear market. Even investors who bought at the very top in 2007 and survived a 57% haircut were made whole five years later and had significant gains within seven years.

But if you need your funds now or in the near future, to live on in retirement, to pay for school, or for a host of other reasons that mean the money can't be at risk, stocks, even stable dividend stocks, are not the answer.

However, if your time horizon is at least five years, this portfolio should work out great.

Ultimately, this portfolio works best with a 10-year or longer time. The compounding nature of the rising dividends really kicks into gear starting around year 8 or 9. The longer you can go without touching the principal, the better.

If you can go beyond 10 years, that's where significant wealth starts to get created.

If you buy a stock with a 4.5% dividend yield and the company raises its dividend by 10% each year, in 10 years, your stock will be yielding 10.6%.

Assuming a straight dividend payout (no dividend reinvestment), after 10 years you'll have collected 71% of your principal back in dividends.

But watch what happens because of the power of compounding as the years go on.

After 12.5 years, your investment will have been fully paid for by the dividends you've collected, and you'll be earning a 13% yield.

In year 15, the yield on your original investment will be 17%.

In year 18, you'll have collected dividends equal to double your original investment. Notice how it took 12.5 years to capture 100% of your capital back in dividends and fewer than six years to do it again. Compounding is a powerful tool.

At the end of year 20, you'll be earning 27% yield every year and will have earned dividends equal to 250% of your original investment.

Keep in mind that the yield I just discussed has nothing to do with the stock price. The stock could have tripled during this time, or it could have been cut in half. As long as the company is paying and raising its dividend by 10% per year, those are the yields you would have enjoyed.

The numbers get even more astounding when you reinvest those dividends, as I'll show you in the next chapter.

Remember that most companies do not raise their dividend by the same percentage every year. But some companies do have a target range for their dividend growth rate. And a number of companies have averaged 10% per year dividend hikes over 10 years. It might not have been 10% every year. One year might have been 5%; the next, 15%. But over the course of the 10 years, the average was 10%.

In a perfect world, we're going to have Warren Buffett's desired holding period, which is for life. If we can hang on to these investments forever, they should continue to generate increasing amounts of income for us every year.

Of course, not everyone has Buffett's flexibility. Many investors need to eventually sell stock to fund their retirement. But if you can put off selling for as long as possible, it will help ensure the additional income is there when you need it.

Last, whether you need the income today or you're trying to create wealth for tomorrow will determine what you do with the dividends.

Those who need income today will collect the dividends when they are paid, usually every quarter. Investors who rely on dividends for income typically keep track of when their dividends will arrive.

Some investors, particularly retirees, may be tempted to factor when the dividends will arrive in their decision as to which stocks they're going to buy. They like the idea of checks coming in regularly every week or so. With a portfolio of 10 to 20 stocks, you probably could structure it so that you are receiving dividends as regularly as you wish.

Companies that pay monthly dividends rather than quarterly are particularly popular with those on a fixed income.

However, I wouldn't invest that way. Deciding which stocks you're going to buy based on which week of the quarter they happen to pay out their dividend is not a smart thing to do.

You want to pick the very best stocks that offer the juiciest yield with the greatest degree of safety and opportunity for dividend growth. These three factors should be your main criteria.

The company isn't taking your schedule into account. It could delay the dividend by a week or two in a certain quarter, which could mean you don't receive the dividend when you are counting on it.

If you focus on exactly when you will receive a dividend check, you'll limit yourself and possibly miss the best opportunities in the market at that time.

If you're only looking for a stock with a dividend payout in January, April, August, and October, for example, you might not invest in one with a safer dividend, a higher yield, and better growth opportunities.

Of course, once you own the stocks, you can set up your calendar so that you know when the dividends are expected to arrive, but don't buy the stocks according to when the payouts are due.

Yields

If you need the income now, do not figure out how much dividend income you need and pick the stocks that will deliver. That's a recipe for disaster. You're too likely to cut corners and choose stocks that may not meet your otherwise stringent criteria. You may focus only on how much money you'll get today and not enough on growth and safety.

Instead, find the very best stocks, and see whether they meet your income goals. If they don't meet your objectives, go through your proposed portfolio, and see which stocks you can substitute without sacrificing safety or growth.

Perhaps you'll be able to replace a stock with a 4% yield with another that has a 4.7% yield with only a slightly higher payout ratio and similar growth.

But saying, “I need to earn 7%” and looking only for stocks that can generate 7% yields is going to be a catastrophe. Why? You take on too much risk to obtain those higher yields.

You know the expression “There's no such thing as a free lunch.” That applies especially to Wall Street. If a stock is paying a yield way above average, there is usually a good reason for it. The reason might be that management believes it must pay a high yield to attract investors. You don't want to buy a stock where management dangles that yield in front of investors like a carrot on a stick. Especially if that yield is not sustainable.

Rather, you want a company with management that pays out a respectable dividend because it believes it should return some shareholders' cash every quarter and it has the funds to do so.

In September 2014, the S&P 500's dividend yield was 1.86%. Over the past 50 years, the average yield has been 1.98%.¹ Generally speaking, I look for companies whose yield is at least one and a half times that of the current S&P 500's and preferably at least two times.

Again, growth and safety of the dividend are more important than the yield, so I may opt for a yield of 3.7% rather than 4.5% if I think it will make for a better investment over the next 10 years.

You also want to be sure your yield will keep up with expected inflation.

Currently, inflation is very low, about 2%. To ensure that your buying power will remain the same or grow in the future, your yield should be above the rate of inflation.

No one knows where inflation will be in five or 10 years, but we can look at historical averages as a guide. Since 1914, inflation has averaged about 3.4% per year, so ideally, you'd like to start your search with a stock paying a 4% yield or more—even higher if it's in a taxable account. (More on taxes in Chapter 12.)

REITs and MLPs often pay significantly higher yields because of their corporate structure, as I explained in Chapter NaN. But for now, keep in mind that while they may have a place in your portfolio, you should avoid the temptation of adding too many REITs and MLPs just because of their attractive yields. As we discussed earlier, you want to diversify your portfolio and not get too heavy into any one or two sectors.

In the current environment, I would almost automatically reject anything with a double-digit yield. I say almost because there can be a situation where a good stock gets beaten up because of its sector (a baby being thrown out with the bathwater), or perhaps it deserved to get a thrashing but said thrashing was a bit overdone.

But for the most part, a stock yielding 10% should be a warning sign rather than a come-hither sign. If you're going to invest in a stock with that kind of yield, be sure to look at it very carefully.

The first thing you should look at is . . .

Payout Ratio

The payout ratio is the ratio of the dividends paid versus net income. For example, if a company makes $100 million in profit and pays out $30 million in dividends, its payout ratio is 30%.

Notice that the payout ratio has nothing to do with yield or dividends per share. We can figure out the payout ratio by looking at the financial statements—the statement of cash flow, to be exact.

Figure 7.1 is the statement of cash flows for Meredith Corp. (NYSE: MDP). Meredith is the publisher of magazines, such as Ladies' Home Journal, Parents, and Family Circle.

You can see that in the fiscal year ending June 30, 2014, Meredith Corp. paid out $75,392,000 in dividends against $113,541,000 in net income, for a payout ratio of 66%.

The year before, the company paid $70,527,000 in dividends versus $123,650,000 in net income, or a payout ratio of 57%. So the dividend went up, but so did the payout ratio. If net income were much higher, let's say $150,000,000, then the payout ratio would have declined to 50%. So you can (and often do) have a situation where the dividend payment increases, but because net income climbed even more, the payout ratio goes down.

The payout ratio tells you whether the company has enough profits to maintain (or grow) the dividend. If a company has a payout ratio of 66%, as Meredith Corp. did, that means it is paying shareholders $0.66 in dividends of every $1 in profit.

That's a sustainable number and one that has room to grow. If you're considering this company and know that earnings are expected to rise, you could make the assumption that dividends should increase as well, since the payout ratio is only 66%. Considering that Meredith has raised its dividend every year for the past 21 years, it's a safe assumption that as net income climbs, so should the dividend.

The lower the payout ratio, the more room there is to grow the dividend.

If a company's payout ratio is 90%, any decrease in earnings may cause a dividend cut as the company will not be able to afford to pay the full dividend, unless it dips into its capital, which sometimes occurs.

Occasionally you will see companies with payout ratios of over 100%, meaning all of their earnings and some of their cash on hand is going toward the dividend.

That is not sustainable for the long term, and you should avoid investing in those companies.

Often that is the scenario when you see a stock with a yield above 10%. The company is pouring every dollar it can into the dividend to attract investors, but likely it will not be able to continue on that track for too long.

Going back to our example with Meredith Corp., back in 2009, the company paid out $39,730,000 even though it lost $107,084,000 during the year.

You may be asking, How could the company pay nearly $40 million in dividends when it lost boatloads of money?

And that would be a very good question.

The answer is because Meredith Corp. was still cash flow positive.

There is a very big difference between earnings and cash flow. Regulators allow all kinds of noncash deductions that can lower a company's profits.

For example, when a company buys a piece of machinery, it takes depreciation off its profits. However, that depreciation does not affect the cash that the company's operations generated.

Let's create a very simplified income statement to illustrate what I mean, using my Authentic Italian Trattoria. (See Table 7.1.)

Let's assume because of my incredible baked ziti recipe (it really is very good), the restaurant brought in $1 million in revenue. Our cost of goods sold was $500,000, giving us a gross profit of $500,000.

We paid out $300,000 in operating expenses, leaving us with a $200,000 operating profit.

When we opened, we bought a bunch of equipment that depreciates every year. We're allowed to take that depreciation as an expense, which lowers our profit.

Finally, we pay no taxes—not because we have a creative accountant, but because we have losses that we carried forward.

As you can see from the table, the depreciation lowered our net income to $100,000 from what would have been $200,000. But did we really make $100,000, or did we make $200,000?

If we create a statement of cash flow, we add back in all noncash items, like depreciation. Remember, depreciation doesn't represent any actual cash that was laid out this year. We paid for the equipment in previous years but now claim depreciation as an expense against our operating profit.

Let's create a very simplified statement of cash flow where we add back in the depreciation. (See Table 7.2.)

For simplicity's sake, I didn't include other variables that can alter cash flow, so let's just assume that the cash flow from operating activities is the total cash flow from the business.

You can see that while the net income that will be reported to the government for tax purposes is $100,000, the cash flow—the amount of cash the business actually generated—is $200,000.

Going back to our real-life example with Meredith Corp., while it was unprofitable in 2009, it was able to pay the $39,730,000 in dividends because its cash flow from operating activities was $180,920,000.

Even though the company lost over $107 million during the year, its business generated $181 million in cash, which enabled it to pay the dividend.

Calculating the payout ratio based on the cash flow from operations gives us a ratio of just 22%.

When I look at the payout ratio, I calculate it using free cash flow or cash flow from operations. It's a more accurate representation of whether a company will be able to pay its dividend than using earnings.

Because of the myriad accounting rules, earnings can be (and often are) manipulated to tell the story that management wants to tell.

CEOs are frequently paid bonuses and stock options based on earnings. Stocks tend to follow earnings, so if the CEO has a lot of stock or options, it's in his or her interest to make sure the stock price is high. One surefire way to increase your stock price is to grow your earnings at a rapid clip.

So, CEOs often have a direct financial incentive to make their earnings as high as possible, whether they reflect the truth or not.

Cash flow is a bit harder to fudge. Of course, a motivated executive who wants to commit outright fraud probably can do so, but manipulating cash flow numbers is more difficult as it represents the actual amount of cash the company generated.

Think of it as all of the cash coming in the door minus all of the cash that went out.

Net income is something accountants dreamed up. Cash flow is something businesspeople rely on.

As I mentioned, since stock prices follow earnings over the long haul, you, of course, want to be invested in a company with earnings growth. But for the purpose of analyzing the dividend and its likelihood of being cut or growing in the future, cash flow is a more reliable indicator.

A company can't pay dividends with earnings. It has to pay it with cash.

For that reason, I prefer to use cash flow when determining the payout ratio. Similar to earnings, I generally want to see a payout ratio of 75% or less; if it's a utility, BDC, REIT, or MLP, the payout ratio can be higher.

A payout ratio of less than 75% gives me the confidence that management can continue not only to pay the dividend but also to increase it, even if the business slumps.

A company with a 50% payout ratio (based on cash flow) and a 20-year history of raising dividends, for example, should have no problem raising the dividend next year, even if cash flow slips 10%.

Remember, companies with long histories of raising dividends want to continue to raise them, even if it's just by a penny, to keep their record intact. Management knows that investors are watching closely and that any change in policy will be perceived as a change in outlook.

Dividend Growth Rate

At this point, I'm assuming that any stock you're looking at is one that raises its dividend every year. But a company that inches the dividend half a penny higher each year, simply to make the list of companies that raise dividends, isn't one that will likely help you achieve your goals.

What you need to look at is the dividend growth rate.

There are two ways of doing this. The first way is to go to the DRiP Resource Center (http://dripinvesting.org), which publishes a list of all the stocks with a minimum of five consecutive annual dividend raises.

The Excel spreadsheet that is published every month and is available for you to download for free contains a group of columns headed DGR, which stands for dividend growth rate. The spreadsheet shows the percentage growth over the past 1, 3, 5, and 10 years.

Take a look at the spreadsheet in Figure 7.2. You can see that Becton, Dickinson and Company (NYSE: BDX) raised its dividend by 10% in the last year. Over the past three years, the average annual increase was 11.1%. Over five, it was 12.2% and over 10, 17.6%.

That's a very strong record of raises. Unfortunately, at this time, the stock yields only 1.9%, which is why the company may be able to raise the dividend so much each year.

Contrast that with Black Hills Corp. (NYSE: BKH), which raised its dividend only 2.7% last year, 1.8% over 3 years, 1.7% over 5 years, and 2.4% over 10 years.

Ultimately, you want to find a company with a yield you can live with today but that also has a record of meaningful dividend raises so that it will get you to your goals over the years.

There's another column here that may be useful: the column with the header 5/10. This is the ratio of the average annual dividend raise over five years versus 10 years. This shows whether a company has been raising the dividend more over the past five years than it has on average over 10.

Think of it as a momentum indicator for dividend raises.

So, in Becton, Dickinson's case, if you divide the 5-year average of 12.7 by the 10-year average of 17.6, you get 0.694. Anything over 1 signifies a 5-year average higher than the 10-year average. Below 1 and the 5-year average is below the 10-year—perhaps signifying that the dividend increases are slowing down.

CLARCOR Inc. (NYSE: CLC), in contrast, has a ratio of 1.327, which tells us that the momentum of the dividend raise has increased in the past five years. And by looking at the one- and three-year figures, we see the increases are continuing to get larger.

I don't have a hard-and-fast rule about this ratio. I'm willing to accept a slower growth rate if it's still meaningful. For example, Colgate-Palmolive raised its dividend by 11.3% annually over the past 5 years and 11.4% over the past 10. But when you look at the one-year growth rate, you can see it slowed to 9.0% over the past year. That's a raise I can live with, because 9% will still outpace the rate of inflation (at least today). Considering after the Great Recession, the economy is still not full steam ahead, 9% growth doesn't seem too bad.

If I were considering Colgate-Palmolive, that ratio wouldn't scare me off if I liked the other attributes of the stock. Now, if next year the raise were only 2% and stayed low for another year, I might have to seriously consider whether this stock belongs in my portfolio.

If you prefer to calculate the dividend raises yourself, you can go to companies' websites—particularly those that have a long history of dividend increases. (They like to boast and give investors as much positive information as they can.) You will usually find a history of the company's dividends.

Simply calculate the rate that the company increased the dividend every year and average it out. You'll come up with the average growth rate. It might be helpful to see that dividend raise every year so that you can figure out what the numbers would have meant to you had you bought the stock X number of years ago.

For example, Table 7.3 is Brady Corp.'s (NYSE: BRC) dividend history (adjusted for a 2:1 stock split) over the last 10 years.

You can see that in 2005, the company raised its dividend from $0.425 to $0.46, or 8.2%. Then in 2006, the dividend was increased to $0.53, or 15.2%, and so on. Over the course of 10 years, the average raise was 6.6%.

You can also further see that the rate of the dividend increase slowed dramatically after 2009. From 2010 to 2013, the dividend boost never went above 3%. If you'd owned the stock for a number of years, somewhere around 2012, you might've seriously taken a look at what was going on and why the annual dividend raise had slowed to a crawl.

Is it a new policy? Is the payout ratio too high? Has cash flow dried up? Once you understand the various pieces of information, you'll be better equipped to decide if you're going to keep the stock in your portfolio or punt it and move on to something else.

Companies don't always post their entire history of dividends; sometimes they choose to just show the past few years. But you can call the company's investor relations department to get the full data.

A free website, www.dividata.com, offers dividend history data as well. It has the dividend histories for most companies, although it doesn't go all the way back for every company.

For example, the site has data on Genuine Auto Parts (NYSE: GPC) going back to 1983, although the company has raised its dividend every year since 1956.

In Johnson & Johnson's (J&J) case, the data goes back to 1970, whereas J&J's own investor relations page on its corporate site goes back only to 1972.

If you want to see all of the dividends going back 30 or 40 years just for fun, knock yourself out. But it's not really relevant to whether the stock is an appropriate investment today. It doesn't matter that the company raised its dividend 11% in 1971. What we're most interested in is the past few years because that is likely the best indicator of what we can expect in the near future.

Of course, things can change. A company can find itself with a hot product and see a meaningful increase in cash flow, which might spur management to grow the dividend more than it has in the past. Or the opposite might occur. The company goes through a slump, and the previous 10% dividend hikes get cut to just 1% (to keep its streak alive).

But, generally speaking, if you want an idea of which direction dividend growth is moving and how much growth you can anticipate, take a look at the last 1-, 3-, 5-, and 10-year averages for a ballpark figure.

It's a good measuring stick for how the company is performing. If over the past 1, 3, 5, and 10 years, a company has averaged at least 10% dividend growth, and then this year it climbs only 2%, you may want to take a hard look at it to assess whether it is likely to provide you with the growth in income that you desire.

If the following year it also hikes the dividend only by 2%, it might be time to pull the plug and find an alternative that offers much higher growth.

Special Dividends

A special dividend is exactly what it sounds like. It's a dividend that's, well, special. Any questions?

A special dividend is usually a one-time payment, often much more than the regular dividend.

Look at the dividend chart (Figure 7.3) and data (Figure 7.4) on American Eagle Outfitters (NYSE: AEO). You can see that in 2009 through March 2010, American Eagle paid shareholders $0.10 per share quarterly. It raised the dividend to $0.11 in June 2010.

Then in December the dividend spiked to $0.61 but immediately went back down to $0.11.

On December 2, American Eagle declared a special $0.50 per share dividend on top of its regular $0.11 quarterly dividend. So shareholders received $0.61 per share that quarter.

A company may declare a special dividend for a number of reasons. One of the most common is because shareholders demand it. We saw that in 2004, when Microsoft, sitting on billions of dollars in cash, paid shareholders a special dividend of $3 per share. The payout barely put a dent in the company's cash stash but somewhat appeased investors, who were unhappy that the company was hoarding cash and not putting it to use acquiring other companies or for other growth initiatives.

Investors who demand special dividends do so because they feel that the company is holding their cash. If management isn't going to do something with it, the company might as well give it back.

As you can imagine, management rarely agrees with this opinion, but sometimes when the clamoring gets too loud, it throws investors a bone with a special dividend.

The reason I bring this up is because you don't want to include a special dividend in any annual dividend growth calculations. These are special one-time items. Unless the company specifies that it plans to give special dividends every year or so, you should not assume that you will receive another special dividend anytime soon.

Since distributing a special dividend is an abnormal event, including one in your dividend growth calculation would not give you an accurate picture of the company's dividend growth policy.

If you happen to own a stock that declares a special dividend, consider it gravy, a nice little extra bonus. But don't bank on one again. Instead, be sure you're invested in a company because it has an attractive yield and dividend growth rate based on its regular quarterly dividend.

Also, if you're calculating the payout ratio, be sure to remove the special dividend from your equation.

For example, if a company's regular annual dividend is $1 per share, it declared a special dividend of $0.50 per share during the year, and there are 100 million shares outstanding, the dividends paid should equal $150 million ($1.50 × 100 million).

When determining whether the payout ratio is sustainable, remove the $50 million, and base your calculation on the regular dividend, which totaled $100 million.

One last thing, though: Do look at the total dividends paid, including the special dividend, to make sure they don't exceed the company's cash flow.

If a company has 100 million shares, has a regular dividend of $1 per share, and declares a special dividend of $3, you should be concerned if the company's cash flow totals only $200 million.

The $100 million in regular dividends would have been fine from a payout ratio standpoint, as it equals only 50%. But with the special dividend of $300 million ($3 per share × 100 million shares), the total dividend paid is $400 million—$200 million more than the company's cash flow.

You want to make sure the company has a war chest of cash to pay that special dividend and that it's not borrowing money or selling shares to pay it.

Occasionally, a powerful hedge fund or investor will force a company to borrow money to pay a hefty dividend. If the dividend is not sustainable, the company is not one you want to be invested in for the long term.

Make sure you know where the cash is coming from to pay that special dividend.

Summary

• Diversify your holdings within a dividend portfolio.
• Don't invest for income according to how much money you need or when the dividend is paid; invest in quality companies with strong dividend performance.
• When looking at payout ratios, use cash flow.
• Know your stocks' dividend growth rates.
• I make a great baked ziti (I really do).