If you want to build a high yield, low risk portfolio, looking for companies with a record of reliable dividends and profits is a good start, but it’s not enough.
A truly low risk portfolio must defend again inflation, which means it must pay a dividend which can grow fast enough to match – and preferably beat – inflation. In short, an investor looking for high yields and low risk must also look for reliable, profitable dividend growth.
But first, a note of caution.
Although I use the phrase ‘reliable’ growth, it’s a relative term. Growth can never truly be relied upon, but for some companies it’s a more likely outcome than it is for others. It’s the companies that are most likely to grow at or above the rate of inflation that I’m after, and I want them to do it year after year after year.
Finding progressive dividends
There are various way to approach this search for (relatively) reliable growth. It can be a good idea to check that the company has a progressive dividend policy, i.e. an explicitly stated intention to grow the dividend every year, or at the very least to never cut it.
However, the quickest way to find reliable growth is to search for it in the company’s financial history. I know some people say “don’t invest by looking in the rear view mirror”, but I don’t believe that. I think the best place to look for companies that can grow in the future is to look for companies that have already proven that they can grow in the past.
For me the clearest sign of a progressive dividend is that it has already been increasing every year. Exactly how long you define “every year” is debatable, but I like to look at the last 10 years.
So the easiest way to spot a progressive dividend is to count how many times the annual dividend (per share) went up in the last decade. The more times it went up, the more progressive the dividend is.
What to do about dividend cuts
A lot of investors panic when a company cuts its dividend. I’m not talking about a complete suspension and no dividend payment at all; I just mean that the dividend went down from one year to the next.
Is that reason enough to sell?
I don’t think it is. Although I don’t like dividend cuts, I am not entirely put off by them. There may be some circumstances in which a dividend cut is beneficial to shareholders; perhaps the cash could be diverted to some new internal project which promises excellent rates of return.
Even if the dividend cut isn’t beneficial to investors, it’s often a mistake to sell after the cut. The share price will often drop by 10-20% on the day of the cut, and if you sell you’re locking in that loss.
What will happen if you hold on to the shares? Of course the future is always uncertain, but I’ve seen shares rebound often enough to know that I won’t sell on a cut. That’s especially true given that, as a defensive value investor, I’m focused on owning very high quality companies, and with high quality companies a dividend cut is more likely to be temporary; there’s a good chance it will climb back up again in the years ahead.
As a defensive value investor I’m also a contrarian; I want to buy on bad news and sell on good news. In other words I want to buy low and sell high, and selling low on bad news from a dividend cut just doesn’t fit with a contrarian value approach.
As a long-term investor the idea of buying and selling shares because of dividend movements sounds too much like trading rather than investing. I think it’s better to make a decision and stick with it, taking a somewhat stoic attitude to short-term ups and downs in both the company’s fortunes and the share price.
It all starts with revenue
So far I’ve concentrated on finding reliable growth by looking for progressive dividends, but dividends do not exist in a vacuum. In some ways they’re the final output from a company.
Exactly what comes before dividends depends on how you frame the concept of cash flows through a company, but from my perspective shareholder returns begin with revenue.
Revenue is the first step where customers actually pay for a company’s products or services. Without revenue a company has nothing but dreams and plans.
Fortunately my approach here is simplicity itself, in that I just count how many times revenues went up, just as I did for dividends.
There is a difference though. With dividends I counted increases in dividends per share, but with revenue I just look at total revenue for the whole company and don’t worry about it on a per share basis.
I look at total revenue primarily because total revenue figures are easier to get hold of than revenue per share. Most data providers in the UK give revenue data and not revenue per share, and it would be another tedious step to calculate it on a per share basis. If this was an important distinction then I would gladly go the extra mile, but in all honesty I doubt that it makes much difference in the vast majority of cases, especially as I am considering dividends and earnings on a per share basis already.
Having said that, it’s always a good idea to look at the total number of shares in issue for each of the last 10 years so that you can spot large rights issues (where the company issues lots of news shares). Rights issues are often used to pay down debt when a company has borrowed too much, which is evidence of poor management. They are also used to fund acquisitions, which often turn out to be poor value for money.
For me a large rights issue isn’t a show-stopper, but they do require further examination.
Earnings are an important intermediate step
From revenues the accountants will strip out expenses of various sorts, eventually leaving us with a figure for earnings, which is easy to find on a per share basis.
Earnings can be a tricky subject, with various different definitions of exactly what ‘earnings’ are, from basic, to reported, adjusted and normalised. Where possible I prefer adjusted earnings, which strips out large one-off income or expense items; those which are not part of the company’s normal trading business.
The story here is exactly the same as it was for revenues and dividends. I’m looking for adjusted earnings that grow and the more frequently and consistently they grow the better. The same trick of counting how many times they went up in the last decade can be applied as before.
So now we have a variety of ways to measure the reliability of a company’s growth, taking into account its revenues, its earnings and its dividends.
Bringing all these steps together, the idea is to look back at the company’s finances over the last 10 years and:
- Count how many times it made a profit
- Count how many times it paid a dividend (and ignore companies that didn’t pay a dividend in every year)
- Count how many times revenues went up
- Count how many times earnings per share went up
- Count how many times dividends per share went up
Companies that score full marks – or even just close to full marks – will have the best track record of producing consistent, profitable, dividend growth, and they’re an excellent place to look for companies that can produce that kind of growth in the future.