If you pull up the annual report for a dividend-paying company, there’s a good chance that you will find earnings per share and dividends per share placed right next to each other. This often happens in the “selected financial data” table, where a company pulls out what it believes are key statistics for investors to see. However, much of this data is taken out of the context of the financial statements to which they belong. Making matters worse, some companies actually include dividends paid per share on the income statement, too. 

Pairing dividends and earnings is not necessarily a bad thing. Clearly, it would be ideal if a company had more in earnings than it paid out in dividends. This is, basically, the concept behind the dividend payout ratio calculation (dividends divided by earnings, on a per share basis or as bulk numbers). In fact, it is ideal if a company has a reasonable or low dividend payout ratio. However, it is mixing statistics from two different parts of the financial statements that may end up giving the wrong signal to investors.

Earnings take into account a great number of factors, including the obvious costs of doing business. To take a simple example, Company X sells $100 of widgets that cost $90 to make. Therefore, Company X had $10 in profits. Simple enough, but accounting is far more arcane than that.

Presumably, Company X had to have a manufacturing facility to build its widgets. That facility cost the company money that it paid when it bought it. However, accounting standards require that cost to be spread out over the so-called “useful life” of the facility. The “useful life” of an object can range from as little as a few years for some items to as long as 20 to 30 years for larger items, like a building. This is called depreciation. So, if Company X’s factory depreciated by $2, then earnings would actually show as $8 ($100 in sales minus $90 of manufacturing costs minus $2 in depreciation expense).

However, the depreciation expense didn’t actually cost Company X any real money—it was simply an accounting item. Such items are often referred to as non-cash items. So, if Company X paid out all of its earnings, it would pay out $8. However, since the $2 didn’t actually get spent in the period, that would leave $2 floating around. Company X could actually distribute the full $10, even though that would appear to be more than it earned.

This is a very simple example of a very complex topic. However, it demonstrates the idea behind cash flow. Cash is the total of net income plus non-cash charges (depreciation, amortization, and depletion) minus preferred dividends (if any). A back of the envelope way to calculate this figure out is to take net income and simply add back depreciation and amortization. That simple math doesn’t get you a precise result, but it is usually close enough. Either way, this is the money from which dividends are paid.

This is important because it helps to explain what would appear to be impossible anomalies if dividends were paid only from earnings. For example, there are some industries that have material depreciation expenses associated with them. Real estate related business (REITs) and asset heavy businesses (pipelines) are two examples. Companies in these industries routinely pay out more in dividends than they earn. This is why many such industries are gauged by metrics other than earnings, such as Fund From Operations (FFO) for REITs. These measures often lack industry standards, however, so investors need to pay keen attention to what is and is not included in the statistic.

So, when examining certain industries, investors must make sure to understand dividends in context. It would be very easy to dismiss a large number of very healthy REITs and pipeline stocks because their dividends appear to vastly outstrip their earnings. However, this may not really be the case and an income investor may be missing important opportunities if he or she “throws the baby out with the bathwater.”

Another example of an anomaly is when a company’s earnings fall materially, causing its dividend to be larger than its earnings. If one went under the logic that dividends were paid out of earnings, then this would mean that the company had to reach in to its cash balances to pay the dividend or, worse, take on debt (or issue securities). However, if cash flow is sufficient to fund the dividend, it’s highly likely that neither of those two things happened. Clearly, some companies do choose to use cash or take on debt to maintain dividends, but most do not do this and it would be unfortunate to disregard good companies because of temporary earnings shortfalls that don’t really affect dividends.

This isn’t to suggest that dividend payout ratios are unimportant—this statistic is an important tool for investors to use. However, understanding the tool, and its limitations, is important too. This is why Value Line provides both earnings per share and cash flow per share in its equity reports. By doing so, dividend-focused investors can take a more educated look at what is really going on.

For example, Merck (MRKFree Value Line Research Report on Merck), a so-called “Dog of the Dow”, started paying annual dividends of $1.52 in 2005. It maintained this level of distribution through 2011. However, in 2007, the company only earned $1.49 per share. All of this information can be found in the historical portion of the statistical array on Merck’s report. Looking up further in the data, the company’s cash flow per share was $2.42, more than enough to cover the dividend that year even though the dividend payout ratio (listed as All Dividends to Net Profits in the statistical array) was over 100%. That metric, meanwhile, dropped to 42% the following year.

Clearly, some companies that see their payout ratio spike toward, or beyond, 100% have material problems. However, for others such a spike may happen for a good reason. It pays to take the time to find out what’s behind the spike, particularly if cash flow is more than ample enough to cover the dividend.

At the time of this article's writing, the author did not have positions in any of the companies mentioned.