What is the best way to measure dividends?

by Richard Allum on 15 April 2010

Dividends, it is generally agreed, are a good thing. Indeed, the majority view is that the more the better. Long term research by Barclays, Credit Suisse, Société Générale and Professor Jeremy Siegel all provide convincing evidence that the bulk of equity returns over the long term, five years and more, actually comes from dividends, growth in dividends and from reinvesting dividends.

On the basis of these data then it should be a straightforward process to construct a market beating portfolio. Simply buy a collection of the highest dividend paying shares and away you go. Indeed, for many investors over the last decade that has been a winning strategy. A number of professional money managers made great careers running high yield funds; until 2008 and 2009 that is. The last two years have been very painful for income funds with catastrophic falls in 2008 for many and lacklustre returns in 2009 when market recovered very sharply.

To the detached observer these events simply proved the validity of stock market aphorisms like “There’s no such thing as a free lunch” “You can’t beat the market” or “Higher returns only come with higher risk”.

To understand better why income and yield funds have been so disappointing over the last two years it might help to do a little analysis on what exactly a high yield share is. Yield is the product of two figures divided together. The top line is the dividend a company pays, usually expressed in pence per share. That number is then divided by a second figure, the share price, to calculate the yield. Although they are related there is no direct correlation between the two except through the yield calculation. It might therefore make sense to look at each figure separately to see how they originate and what they are telling us.

A dividend is the amount of money a company feels it can afford to pay out to its shareholders after it has paid all its creditors and invested enough to sustain and grow the business. Although it represents a cash payment of several tens or hundreds of millions of pounds it is invariably referred to in pence per share. The figure is set by the board in relation to the reported earnings per share and by what the company has paid in the past. Determination of the dividend may also be impacted by the board’s view of the future. A particularly worrisome outlook may persuade the directors to leave the dividend unchanged, cut it or reduce it.

Alternatively, the board may be so confident in the future of the company it might want to send a signal to the market that things are great, and are going to get better. An earnings per share figure is, more or less, simply a record of what the company has done and does not send a message about future prospects. Even though two of the options open to the board are reducing or not paying a dividend in practice boards are reluctant to do this and will often pay the same dividend as the previous year even when its financial circumstances might indicate otherwise.

Much is made of the dividend cover. This is the difference between the earnings per share figure and the dividends per share. Typically boards and investors like a figure of around two or more as this provides some scope for maintaining the dividend even if the company has a bad year. In practice these days the many distortions to profits from exceptional and or non-recurring items often makes the earnings figure a poor guide to what the company can pay out to its shareholders. A better understanding can be gained from analysis of the cash flow statement, although that too can be fraught with issues around lumpy capital expenditure and corporate activity.

The dividend that a company actually pays is thus a complex blend of what the company can afford and what the directors think is appropriate given the current business conditions.

In contrast the directors have no control over the share price. That said directors now are all acutely aware of the requirement to inform the market if they believe current consensus forecasts are more than five percent adrift from what they think will happen. In essence share prices are set by “Mr Market” as sagacious investor Warren Buffet calls it. All the available data is collected and assessed by existing and potential holders of the stock to discover the most accurate price for the shares. Except in periods of extreme market dislocation exactly half the market will think the shares too expensive while the other half will view them as too cheap. Share prices can therefore best be viewed as an opinion while a dividend is very definitely a fact. The question is what happens when we divide a fact by an opinion. Surely, it must just be another opinion.

And that is where income and yield funds run into problems. High yielding stocks are trying to balance the conflict between the fact of the last dividend payment and the opinion of the market over the size of the next dividend. In many cases, especially over the last few years, the market opinion that a dividend will be cut, or reduced, has been correct. Even when that downside has been priced in a dividend cut usually triggers a further fall. And that is painful for funds that hold the shares. In practice the effect is compounded because income and high yield funds will migrate to stocks that, on paper, offer a high yield and ignore stocks where the income is safer, but smaller. In effect what is happening is that dividends are being valued by price and these funds over-invest in shares whose income has a low value; in many cases for good reasons.

Investors buying conventional tracker funds ignore share prices when they invest. So what happens if we ignore the price of dividends? Why don’t we use some other measure to assess them? The simplest way to do that is to rank companies by the gross cash dividend they pay out. In other words you measure dividends by volume rather than price.

When we compare a portfolio constructed in this way we see a lot of similarities with a conventional portfolio ranked by market capitalisation. After all a company paying out several billion pounds in dividends is hardly likely to be lurking in the Small Cap sector. It is therefore no surprise to see that companies like Vodafone and BP that constitute the largest companies in the market are also the largest dividend payers. Moreover, assessing dividends in this way gives an excellent mechanism for determining how much of each stock a portfolio should hold. Using each company’s contribution to the total income of the market provides a logical basis for calculating portfolio weights.

Using dividend data in this way provides a neat riposte to Oscar Wilde’s accusation that some men know the price of everything and the value of nothing. It is more important to know the value of dividends than their price.

Article written by Rob Davies, Managing Director of Fundamental Tracker Investment Management Ltd and reproduced with kind permission of IFA Life

Leave a Comment

Previous post:

Next post: