Dividend Safety: How to sleep better at night
“The whole secret to winning and losing in the stock market is to lose the least amount possible when you’re not right” William O’Neill
While nothing can prepare the dividend investor for the kinds of disasters that have hit a handful of high yielding blue chip companies recently, there are some important steps that can be taken to minimize the risks we’ve been discussing. In particular, the latest research by Societe Generale suggests that ‘financial robustness’ - defined as companies having strong balance sheets but also good underlying business economics - is a better indicator of dividend sustainability than even the dividend track record itself.
The irony is that giving up on the high yield dream and searching for more moderate yielders that promise greater dividend safety can actually improve portfolio returns materially, while lowering volatility to boot - a win-win proposition. We will shortly cover some of the traditional and modern approaches to assessing dividend safety, but as the best offence is a good defence, let’s first investigate the terrain of dividend cuts and traps as a prophylactic measure.
On dividend traps, cuts and other nasties
A dividend trap is a situation where a high headline dividend yield lures unwitting investors into its snare only for them to discover that it was a temporary illusion. Broadly speaking, we can think of three perfect setups for a dividend trap which you should be aware of:
A Cow Feeding Itself Its Own Milk - where a company has bad cashflow but attempts to maintain its dividend policy it may financing the payout with debt. This is highly likely to end in tears.
A Falling Knife - where an apparently compelling yield is actually the result of a substantial freefall in the share price of a dividend paying stock. Because earnings ultimately drive dividends, a sustained drop in anticipated earnings usually foreshadows a dividend cut or, in severe cases, bankruptcy.
Fools Gold - where a company decides to pay a large one off dividend payment in one year without any intention for it to persist. This might be the result of a windfall, such as a disposal, and naive extrapolation of this payment level into the future can lead investors astray.
As we’ll repeatedly hammer home, dividends are not certain cash flows and can be illusory. While CEOs do fear the repercussions of cutting dividends, sometimes management feel they have no choice. In 2011, 438 UK quoted companies paid a dividend, 59 of those actually cut their payouts while a further 31 firms cancelled their payments altogether bringing extremely unwelcome news for the shareholders involved.
Given the typical volatility in corporate earnings and cash flows, it might seem surprising that we don’t see even more dividend cuts. Wouldn’t it be rational for firms to actively reassess how much they should pay in dividends as their prospects change? It seems not. In his classic 1956 study on dividend policy, Lintner interviewed corporate managers and found: “a reluctance to reduce regular [dividend] rates once established and a consequent conservatism in raising [dividend] rates”. As a result, dividends tend to follow a much smoother path than earnings - the variability from 1960 to 2008 of year-to-year changes in dividends was just 5.2 percent, compared to 14.7 percent for earnings. Management are hell bent on sustaining them.
You would think that paying an unsustainable level of dividends is going to be worse for investors in the long term, especially if it leads to a dilutive capital raising. However, the evidence suggests that this simply doesn’t wash with investors. Research shows that the investor response to a dividend cut tends to be brutal. Several studies show that in 80 percent or more of cases, the stock prices of firms that cut dividends drop sharply at the time of the announce- ment. Furthermore, research by Michaely, Thaler and Womack found that that stock prices then continue to drift downwards in the weeks following a dividend decrease.
The traditional approach to safety
So given that dividend cuts have a tendency to reduce not only income but also the capital value of the shares, it makes sense for investors to value safety above all else. One approach to assessing this to rely on qualitative analysis aimed at identifying companies with a sustainable competitive advantage and a robust operating model. Warren Buffett likened businesses to castles at risk of siege from competitors and the marketplace. Great companies are able to dig deep economic moats around their castles that become increasingly impregnable to competition and market pressures. These moats bring either pricing power, scale advantages or cost reductions which help sustain very high returns on capital, leading to higher cashflows and returns for investors.
Judging a company’s “economic moat” is an important but time-consuming exercise. However, there are also some helpful financial ratios and indicators that can be handy short-cuts to assessing financial robustness. Obviously excessively high yields and lack of dividend history are key warning signs that all is not well, but here are a few other key health indicators which should be monitored closely.
Dividend Cover / Dividend Payout Ratio
Perhaps due to its simplicity and universality, the dividend cover has earned a reputation among investors and analysts as the essential dividend health metric. It gives investors a quick fix on how much a company is paying out in dividends in relation to the earnings it is generating. As we’ll see in the Strategies section, it’s so important that Charles Carlson gave it a 30 percent weighting in his Big, Safe Dividend formulation.
It is easily calculated by dividing the earnings per share by the dividend per share (EPS/DPS). In the United States, they prefer to invert this ratio and express it as a percentage which they call the Dividend Payout Ratio (DPS/EPS).
The preferred level of dividend cover
The usual rule of thumb is that dividend cover of less than 1.5x may indicate a danger of a dividend cut, while more than 2x is typically viewed as healthy. 1.5x cover is ultimately an arbitrary line in the sand, but it reflects the need for some margin of safety. At a cover of 1x or less, the company is distributing all of its earnings as dividends and even dipping into reserves from previous years. That’s usually a big red flag.
Because many companies are reluctant to cut payouts even if profit levels fall, dividend cover is a useful indicator of earnings persistence and financial health, particularly when tracked over a long period of years. The issue of dividend cover only really comes to the fore when a stock looks vulnerable. Companies themselves will often refer to their preferred level of cover in their overall dividend policy, which tends to get discussed in preliminary results and an- nual reports.
The debate over dividend cover
In spite of the above, there’s a lot of debate over whether investors should prefer a low or a high dividend cover.
The traditional view is that, if a company is paying out too much in dividends, it does not have the ability to reinvest profits back into the business which could hurt the business’s growth prospects longer-term. On this view, low dividend cover is bad whereas high dividend cover is good. In support of this claim, in 2006, analysts at Credit Suisse attempted to discover the optimum balance between yield and cover. What they found was that, between 1990 and 2006, S&P 1500 stocks with high yields and high dividend cover produced annualised returns of 19.2 percent – beating every other variation of payout ratio and yield. By comparison, over the same period the S&P 500 (large cap companies) delivered a return of 11.16 percent.
But in the opposite corner, in a 2003 paper entitled Surprise! Higher Dividends = Higher Earnings Growth, US finance commentators Robert D. Arnott and Clifford S. Asness concluded that high rates of dividend cover historically precede periods of low earnings growth. Essentially they showed that management of companies with low dividend cover are forced into being more disciplined with their retained profits. They become less likely to indulge in ‘inefficient empire building and the funding of less than ideal projects’ which lead to ‘poor subsequent growth’ rates.
Regardless of which camp you are in it’s still likely to be worth checking out the dividend cover. After all, dividend cover of less than 1 means that a firm paid out more than it earned as dividends – an unsustainable approach in the long-term!
Gearing
For investors who find the toing and froing over high or low dividend cover confusing, an alternative is to focus more directly on a company’s balance sheet strength and cashflow. Clearly, if the company is highly leveraged, and is having trouble meeting its short-term liabilities, then this is going to be a big red flag for the dividend. If the company has recently acquired another company, how did it finance this? Did it make a huge cash payment from its cash reserves or borrow money from banks?
There are some rules of thumb that analysts like to use to assess balance sheet strength. While we won’t go into too much detail here, a safe level of gearing (debt to equity) on the balance sheet is generally considered to be 50 percent or less. While many are comfortable with gearing of up to 100 percent the likelihood of default obviously increases as the gearing rises. It is essential to look at gearing levels in comparison to sector and industry norms as clearly high levels of gearing are more usual in some industries than others.
Other ratios
Other common ratios to use in assessing balance sheet strength include watching the current ratio (current assets / current liabilities ) which assesses the ability of the company to service short term debts. A current ratio of less than one tends to be a worry.
A modern approach to safety
At Stockopedia we are big advocates of using more advanced quantitative indicators to assess balance sheet and financial strength. The benefits of using checklists and algorithms are that they can weigh up many more ratios and indicators more effectively than a typical traditional approach, and don’t suffer the inherent bias and overconfidence that can plague human judgement.
Our eyes were really opened to the possibilities of using these indicators in dividend strategies by the astonishing results of Societe Generale’s Quality Income Index. We go into its construction in more detail in the Strategies section, but in a nutshell by filtering high yield stocks using a quality score (specifically the F-Score described below), and a balance sheet risk score they were able to improve the returns from equities since 1990 from a market average of 5.6 percent to a remarkable 11.6 percent annualised at a significantly reduced volatility.
Assessing financial robustness - The Piotroski F-Score
The primary indicator used by SocGen to assess quality is an indicator known as the F-Score. A nine test checklist that is applied to a company’s financial statements as developed by Joseph Piotroski, now associate professor of accounting at the Stanford University Graduate School of Business. A company either passes or fails each of the nine tests adding up to create a score between zero and nine.
Each of these rules looks at one aspect of a company’s financials, with six of the nine rules looking at the change in a company’s financials. Whereas most ratios (e.g. dividend cover) look solely at a company’s current financial state, the F-Score looks more deeply into the direction in which it’s financial state is moving, and herein lies it’s secret sauce - it captures fundamental momentum, earnings quality and balance sheet strength in a single very smart number.
Piotroski found that any stocks scoring 8 or 9 points had a tendency to massively outperform companies with scores in the 0-2 range in a test by 7.5 percent annually over 20 years. These findings that have been confirmed in live tracking tests in our model portfolios on Stockopedia.
Assessing Balance Sheet Strength - The Altman Z-Score
Rather than look at gearing and interest cover ratios, SocGen used a ‘distance to default’ measure to assess balance sheet risk in their paper. This methodology is lesser known and numbers aren’t widely dispersed but a commonly used alternative used by credit analysts since the 1970s. It was developed by Professor Edward Altman and is known as the Z-Score.
There are several forms of the Z-Score for different classes of company but essentially they come down to the same thing – a company with a low Z-Score signals a high probability of financial distress over the next 12 months. It weights four or five strict balance sheet factors into a single number that can be easily interpreted. Clearly avoiding companies such as these makes sense for the keen dividend investor.
In the Strategies section, we’ll also discuss a custom algorithm known as the ‘BSD’ formula for finding high safety dividend paying stocks. While not ‘academically’ derived, it again illustrates that modern approaches to dividend safety are evolving beyond simple ratio analysis.
A Technical approach to safety
There are many technical analysts and efficient market theorists who would swear that all market information is held within a company’s share price. No matter how much regulators would wish otherwise, different investors have different information and a collapsing share price on no news is a huge red flag. The following indicators are all useful factors in any sound approach to dividend safety.
Does the share price have good relative strength? This check is used by US research firm Russell Research as part of its Russell High Dividend Yield Index Series. In order to protect against the price “freefall” dividend trap, they filter out the bottom 10 percent of stocks based on 12 month price momentum. In a similar vein, a Charles Schwab study ranked the highest yielding stocks by 6-month price momentum, divided them into five segments, and found that highest yield stocks with the highest 6-month price momentum outperformed all the other momentum segments. Yet again, these studies show that price momentum works - even with dividend strategies.
Is it a low volatility (beta) stock? Another measure you can use to judge a potential dividend candidate is its “beta”. Used by traders, beta measures how much the stock price moves up and down relative to the whole market. A “high” beta stock (more than 1) is generally more volatile and prone to wider swings in either direction compared to the broad market, while a stock with a “low” beta (less than 1) is generally more ‘boring’, less volatile and less likely to excite most equity investors. Of course when things are boring, there’s a systematic tendency to leave them underpriced, which can in turn lead to higher future returns. Like anything, it’s not a silver bullet and beta has always had its critics. Even a low beta stock could tumble in price or suffer financial troubles but, in general, it looks more likely that a high beta stock with a toppy yield could turn out to be a dividend trap.
Is the market cap big enough? The small cap market is of course a wonderful hunting ground for potential market inefficiencies but, notwithstanding the recent lessons from the banking sector, it remains true that larger-cap companies tend to be more stable. Work by Bank of England analysts Andrew Benito and Garry Young has found that the smaller scale of a business is correlated with an increased chance of a dividend cut, so it’s important to measure the potential greater capital gains against the increased risks of capital destruction.
Our recommendation for income investors seeking safety is to use a combination of the above techniques. It is worth factoring in both traditional safety measures such as good Dividend Cover combined with modern filters such as the F-Score to avoid companies at high risk of default. Focusing on large, low volatility stocks will also ensure that you aren’t exposing yourself to the perils of the highest yielding, highest risk segments of the market.