One of the most important components of Simply Safe Dividends is our Dividend Safety Score, a metric that rates the safety of a company’s dividend payment by scrubbing through its most important financial metrics.
Dividend Safety Scores are available on our site for thousands of dividend-paying stocks and can help you avoid riskier investments and build a more resilient income stream. You can read more about how our scores are calculated and view their real-time track record by clicking here.
Dividend Safety Scores range from 0 to 100, and I usually suggest that conservative investors focus on stocks that score at least 60. Scores can be interpreted as follows:
As you know, I believe in complete transparency in all that I do. From reporting detailed performance information about our dividend portfolios each month to owning up to investment mistakes I make (and there will certainly be more), I will always do my best to tell it exactly how it is.
The only way to grow as an investor is to be open and honest with ourselves, tracking and analyzing our entire decision-making process and the results we achieve (good and bad) in order to continuously improve. If we give into human temptation to ignore our investment mistakes, we do ourselves a great disservice and are more likely to make the same errors over and over.
The same is true for our Dividend Safety Scores. Instead of blindly shoving thousands of scores out there on the website and hoping they have some meaning behind them, I track their performance in order to gain powerful insights that can be used to make our scores even smarter as time goes on. I know many of you rely on Dividend Safety Scores to help guide your investment decisions, and I take your trust in our metrics very seriously.
By logging real-time dividend cut announcements and recording the Dividend Safety Score we had for a company right before its dividend reduction was reported, we can view the effectiveness of our Dividend Safety Scores, learn more about why companies cut their dividends, and discover ways to further improve the risk assessment capabilities of our scores.
We began tracking all dividend cut announcements in April 2016 (see them all here). The chart below incorporates data recorded over the last year, plotting each company’s Dividend Safety Score on the x-axis and the size of its dividend cut on the y-axis.
Two observations jump out at me. First, all but one of the companies cutting their dividends scored close to 40 or below for Dividend Safety, falling in the “Unsafe” and “Extremely Unsafe” buckets. The one exception was a micro-cap stock (Ecology and Environment – EEI) that scored an 82 for Safety.
How did this happen? The company’s fundamentals were actually very healthy, but management decided it wanted to invest more for growth, freeing up additional cash for reinvestment by reducing the dividend by 17% (read the company’s press release here).
I am not really sure there was much we could have done to flag this dividend cut ahead of time since management’s decision to reduce the dividend had little to do with the company’s actual fundamentals (e.g. payout ratios, earnings growth, balance sheet, dividend longevity etc.). However, we do treat micro-caps with greater conservatism today in recognition of their generally more dynamic capital allocation policies.
The second observation from the chart above is that companies with lower Dividend Safety Scores were more likely to cut their dividends by larger amounts. You can see that most of the biggest dividend cuts (e.g. 50%+) happened with companies scoring below 10 for Dividend Safety.
Of course, not all companies that score low for Dividend Safety are at risk of an imminent dividend cut. In fact, most years there are relatively few dividend cuts compared to the number of dividend-paying stocks in the market. For example, only a dozen S&P 500 companies reduced their dividends in 2007, and just four slashed their payouts in 2010.
However, during the financial crisis (2008-09), roughly one-third of dividend-paying companies in the S&P 500 cut their dividends. Dividends generally rise over long periods of time, but the periods marked by heavy dividend cuts are usually steep and come with little warning.
Companies that score below 40 for Dividend Safety are likely the most vulnerable to slashing their dividends during the next recession. No one knows when the next crash is coming, and many economists don’t even realize that we are in a recession until it has already set in.
I prefer to remain conservative with our Dividend Safety Scores and will continue rating potentially problematic companies low for Dividend Safety, even if their dividends could remain safe until the next recession. It’s better to be aware if a company’s dividend has high risk potential before something bad happens, in my opinion.
To that point, I have some additional data on Dividend Safety Scores to share. Companies that are at greater risk of cutting their dividends usually have a number of problems. From too much financial leverage to unsustainably high payout ratios and slumping earnings, there is no shortage of potential issues.
Many of these financial problems are not only bad for the dividend’s safety, but also for the entire company in general. Many of us are conservative investors who are looking to not only generate safe, growing dividend income, but also preserve capital.
Even if the dividend remains safe, many of these low-scoring stocks can be very toxic investments that permanently destroy our capital. A “safe” dividend does little good if the stock’s price declines by 50%.
The table below takes a few minutes to digest, but it’s very important to understand. At the start of 2016, I recorded the Dividend Safety Scores of every stock in our database and placed them into the five groups below based on their scores.
Each group contained close to 400 stocks. I then recorded the total return of each stock in 2016 and calculated the following statistics:
- Average Return: the average 2016 total return of all stocks in each group.
- Minimum Return: the lowest 2016 total return recorded by a stock in each group.
- Maximum Return: the highest 2016 total return recorded by a stock in each group.
- % of Stocks With a Negative Total Return: the percentage of stocks in each group that recorded a 2016 total return below 0% (i.e. these stocks lost money for investors in 2016).
- Standard Deviation: measures the volatility of all of the individual stock returns in each group. Lower figures indicate milder price swings and a smaller chance of sharp declines.
- Average Beta: the average beta across all stocks in the group at the start of 2016. Beta is the volatility of a stock compared to the market (<1 = less volatile; >1 = more volatile).
- # of Stocks: the number of dividend stocks in each group at the start of 2016.
The first figure that really jumped out at me was the average return of each group. Stocks with a Dividend Safety Score below 20 at the start of the year had an average total return of -2% in 2016. They lost money on average despite the S&P 500 returning over 10%!
Over half (53%) of the 431 companies in this group recorded a negative return in 2016. The odds were not in your favor trying to picking safe dividend stocks from this bucket.
On the other end of the spectrum, stocks with Dividend Safety Scores between 61-80 and 81-100 achieved an average return of 22%, and less than 25% of these stocks recorded a negative total return in 2016. In other words, sticking with stocks that scored higher for Dividend Safety gave a much better chance of earning a meaningfully positive return.
Higher Dividend Safety Scores also did a good job of weeding out the stocks most likely to blow up. The worst-performing stock with a Dividend Safety Score over 80 delivered a -42% return compared to -74%, -88%, and -99% minimum returns in the Average (41-60), Unsafe (21-40), and Extremely Unsafe (0-20) buckets, respectively.
Not surprisingly, stock price volatility was the lowest (28% standard deviation) in the Very Safe (81-100) bucket of Dividend Safety Scores and highest (41% standard deviation) in the Extremely Unsafe (0-20) bucket. Healthy companies with safe dividends tend to have fewer nasty surprises, resulting in less stock price volatility and better capital preservation.
To conclude, data from 2016 suggests that the application of Dividend Safety Scores extends beyond a pure measurement of how safe a company’s dividend is. Dividend Safety Scores acted as a good overall measure of business quality and investment risk. Very conservative income investors have little to no business reaching for stocks that score below 20 (or even 40) for Dividend Safety.
However, these insights aren’t worth much if they aren’t actionable. With the start of a new year and the market sitting near an all-time high, now is as good of a time as any to review your portfolio for Dividend Safety and overall quality (especially for those living off dividends in retirement).
As the ancient philosopher Lao Tzu said, “Do the difficult things while they are easy and do the great things while they are small. A journey of a thousand miles must begin with a single step.”
You can use our Portfolio Analyzer tool to view Dividend Safety Scores for each of your holdings and your overall portfolio by clicking here. Even if you don’t see a need to make any changes, it is important to stay on top of this information and be aware of the risks you are taking.
You can also use our Stock Analyzer tool to retrieve Dividend Safety Scores for thousands of stocks to evaluate ideas you get from our site or other sources.
Thank you for studying how well the Safety Scores performed for 2016.
If I understood you correctly, you put each stock into one of the five buckets at the beginning of the year. If so, how many stocks (and maybe which specific stocks, if the list is short) moved more than one bucket from where it started the year to where it ended the year-e.g. from bucket 1 (80 to 100 safety score) to bucket 3, 4, or 5, or vice versa?? (Given that a move in the score from 81 to 79 would shift which bucket a stock is in, it seems a one bucket move should not be very consequential, hence, I would not bother with these.)
I use the portfolio analyzer to look at my portfolio probably twice a week and notice some movement in both safety scores and dividend growth scores though the movements don’t seem very large.
I am curious as to whether the stocks with large declines from among the first two buckets may have experienced notable deterioration in the Safety Score during the year.
You are welcome. I will continue to be as transparent as possible with all of our scores and information.
Your understanding is correct. I didn’t move any stocks between buckets during the year for my analysis even if their scores might have changed categories.
A number of members have inquired about the ability to track changes in Safety Scores, and that’s something we are exploring.
Out of curiosity, I reviewed stocks from the top group (81-100) that dropped into lower buckets and saw their Safety Score decrease by at least 10 by the end of 2016.
There were 99 such stocks (out of 397 in the 81-100 bucket), and their average return in 2016 was -27%! You might be on to something there.
The big question is WHEN their Safety Score changed – was it after a lot the damage was already done to the stock, or was it more of a leading indicator? It’s hard to say, but again we will be giving these issues more attention.
Generally speaking, scores should remain fairly stable (e.g. +/- 10 points) unless there is a material change with the business.
Thanks for your comment!
One (EEI) out of dozens!
Most impressive track record!
Your candid article is most appreciated. To my knowledge your site is unique when it comes to that highest level of consequent orientation toward dividends (safety, growth)
But, you know, I just feel even more confirmed why I subscribe to your service! Invaluable!
Best wishes for 2017!
Thank you, Thorsten. We aren’t always going to be right, but I can assure you that we will continue being as candid as possible, working to get better at everything we do each and every day.
I hope you have a wonderful 2017 as well, and thanks again for all of your support!
Gjg49 I concur.
Safety scores do tell a snap shot story, but Safety Scores do move. I doubt many do move to a significant amount but it might be revealing to review how many do move in either direction and by how much.
This may establish if there is a need to fix the SS at the time of purchase to it’s current score.
Over the last four years that I’ve taken over my portfolio I’ve not noticed this level of monitoring and refinement of what is unique to the site. I’m totally sold on continuing with Simply Safe Dividend.
Thanks Brian and Happy New Year
Thank you, and Happy New Year! Your feedback has been really helpful over the last year. Monitoring and tracking Safety Scores is something we plan on doing more of.
I am approaching the issue carefully because I don’t want “false indicators” to be created that result in more noise than news (e.g. you see the Safety Score dropped by 8 points last week for a holding and decide to sell as a result). I want to have some data supporting what a “meaningful” change in Safety is before getting ahead of ourselves.
Continuous improvement is our mantra! I’m excited for what lies ahead in 2017 and beyond.
My apologies, but please clarify what you refer to when you say “there is a need to fix the SS at the time of purchase to it’s current score.”
My interest gravitates toward how much and perhaps how rapidly Safety Scores vary over time. If we can get some idea of how quickly and how much the Safety Score might change for a given stock, we might have a better feel as to how often and/or how carefully one should monitor the Safety Scores of the stocks in their portfolio. As an example, I bought Chevron several years ago when the dividend appeared very safe (since I only subscribed this past year, I don’t know what the Safety Score was at the time, but CVX produced positive free cash flow and the cash flow and earnings payout ratios were well under 100% at the time and the net debt levels were lower as well); with the drop in energy prices, CVX’s Safety Score probably declined substantially over the past two or three years. So, my question leans toward: “How quickly did CVX’s Safety Score decline and over what time frame?” And, how many other companies may have had notable declines similar to CVX’s??
For me, the Portfolio Analyzer is an outstanding tool for monitoring my portfolio as it allows me to look at virtually everything that I am interested in–especially the safety of dividends and the growth potential of the dividends. I would like to have a better understanding of how stable the critical predictive scores tend to be.
My apologies for not being more clear. Incidentally I agree with exactly what you said. When I said “Fix” what meant was to record it at the time of purchase and fix it in a column that does not change. Then a column next to it could be the current SS. In that way you could track the change and how quickly the score moves over time.
This would be analogous to “cost per share” and Current Price. I realize that cost can change as you buy in a different prices, but the SS at the time of the first purchase would not.
I got caught on CALM. Sometime after I bought it I decided to DRIP all my positions over a SS of 80, CALM was one of them. Then one day CALM had a SS of 35. Given my age and memory I might not have caught the fact that it was much higher only a month or two ago. I only knew that it had to have been above 80 because I DRIPed it.
Thanks for clarifying. Your approach makes so much sense (recording the Safety Score), that I kept track of the SS in a spreadsheet as well. Could this possibly be a case of “great minds think alike”??
Absolutely! (Emphasis on great)
Simply Safety is a great site and knowing Brian is always evaluating it’s performance is so reassuring.
I review any large (i.e. 10+ points) changes in Dividend Safety each week. Always looking for ways to improve!
Hi Brian, Yep, I agree with the others, great site and, once again, explanations that anyone with common sense can understand. OK, so I’m reading about your “fixing” a SS column. How would that work? Do you add it from your end? Or, is there a way that we can add the current SS when we add a stock to our individual portfolio, and be able to see it a year later??? Once again, great site!! I, for one, sincerely hope that you do not give up this site. It has been the best I’ve found. I’ve made some significant changes to my portfolio due to your articles, discussions, etc. All for the best, I might say. Take Care, and best wishes for a successful 2017! Rick
Thanks for your kind words, Rick. Nothing beats waking up and getting to work on Simply Safe Dividends each day 🙂
We are working on an export to Excel feature for the Portfolio Analyzer that will make it easy to “manually” record your SS whenever you wish. Until then, you can always highlight the My Holdings table using your mouse (click and hold, drag over the table from top left to bottom right), hit “Ctrl” and “c” on your keyboard, open Excel, and hit “Ctrl” and “p” on your keyboard to paste in all the data. It’s a bit tedious, but it works for now.
Went to your website (I have it bookmarked) and found this article today. It was a pleasant surprise to find it and the information given is wonderful. Keep up the good work and I look froward to your articles in 2017.
Thank you! It means a lot to know that our content is worthy of being bookmarked by you, and it’s always a pleasure to publish new articles. Thanks for reading!
What % of stocks with a dividend score of less than 40 have not cut their dividend recently? This info would provide a measure of “false positives” for comparison purposes.
Dividend safety score vs dividend cut % shows a good correlation you should be proud of!
Looking at the data and references a little closer I see:
1) 7.0% of equities with a dividend score <40 cut their dividend or 93% did not. Scores of <40 represent unsafe (heightened risk of being cut) and extremely unsafe (high risk of being cut) equities.
2) 10.2% of equities with a dividend score of <20 cut their dividend or 90% did not.
Given the high % of equities that have scores <20 or 40 cut their dividend and only 1 equity with a score >40 cut their dividend for financial issue reasons. However, if one were to stay away from unsafe and extremely unsafe equities due to fear of a dividend cut, with the current scoring system they are eliminating 41% of the covered universe. This could lead to missed opportunities. The remaining 59% of the universe offers much to pick from but seems somewhat unnecessarily restrictive.
I’d be ok with 1% of equities (19 – 20 companies out of about 1950) with a score >40 cutting dividends if “false positives” (companies with poor scores that do not cut dividends) were on the order of 50% (57 out of about 114 companies) instead of 90% ( 789 out of 846).
Any thoughts on whether scoring can/will evolve to reduce the number of false positives?
Appreciate your thoughts.
If the consequences of a dividend cut were comparable to dividend increases, we would probably not worry about dividend cuts or Safety Scores.
While I don’t have data to support this, my instinct (based on experience) is to think that cuts are asymmetric–the consequences of a 50% or 75% or 100% cut (and most cuts probably are such large percentages) impact aggregate dividend income much more than the average dividend increase. I would conjecture that the “average” dividend cut is several times larger (5 to 10 times??) than the average dividend increase. Anything that assists us in lowering the probability of a cut will probably have a disproportionately positive impact on dividend income stability and growth.
Hence, isolating the 20% of stocks that have a relatively high probability of a dividend cut is worthwhile to anyone attempting to protect their dividend stream.
I’ve given this more thought and entered comments 4-5 posts below before seeing your post.
I agree 100% the key is to identify this lower 20% of higher risk stocks.
SSD is a great tool with a great track record of placing “losers” in the appropriate category (average to greater than average risk of a dividend cut). However, we should let the data drive the extent of risk and label language used. The data says that at a safety score of 60+ the risk of a dividend cut is essentially 0. From a risk perspective we are splitting hairs trying to distinguish between a 95 and 80 dividend safety score. I expand on my thoughts below.
I believe resources are best spent on identifying stocks in the higher risk categories (scores <40, <20) and then further evaluating them. For example, is JCI really a 29 (unsafe)? Payout ratio 29%, debt/cap 35%, no dividend cut in 20 years (held at $0.52 during the great recession), projected earnings growth, etc.
Again, SSD is a great tool and I believe it can be even greater. I look forward to the enhancements Brian has mentioned.
Thanks for your comments. Good stuff as usual. One of the biggest challenges of dividend safety is the very nature of dividend cuts themselves – most years tend to be relatively safe (e.g. <5% of companies cut dividends), then some years strike with little to no warning and see big waves of dividend cuts (e.g. 1/3 of dividend-paying companies in the S&P 500 slashed their dividends during the financial crisis).
The "S&P 500 Dividends Paid" chart from above highlights the wavy nature of broad dividend reductions. Furthermore, many dividend cuts are at management's discretion. There isn't a clear-cut recipe for an impending dividend cut - management could try to keep borrowing money to pay the dividend, hoping times will get better soon than later (e.g. the oil majors), or a decision could be made to cut the dividend now to preserve the balance sheet and keep investing for growth.
The best I can do is identify companies with characteristics that make them the most likely candidates to cut their dividends in the next downturn or sooner. I think it's better to know ahead of time which firms have the most embedded risk (creating more "false positives" today).
I think we will also find out as time goes on that low-ranked stocks for Dividend Safety are also most at risk of permanently destroying your capital. Even if they don't end up cutting their dividends, the risk factors possessed by many of these businesses makes them much riskier and more volatile stocks (the -2% average return in the bottom bucket, for example).
I would also add that Dividend Safety Scores are a starting point - not a be-all end end-all solution. They get your attention that something could be dangerous about a company, and then it's ultimately up to the investor to do some investigating and make a final call. Like you said, perhaps resources are best spent picking through some of the higher risk categories to find exceptions. At the very least, my hope is investors will walk away more informed about an investment's risk profile and act accordingly based on their own risk tolerances and investment goals.
Something was accidentally deleted/modified in the sentence starting with “Given the high…”
Please ignore and replace with “It is great that only 2 equities with scores >40 cut their dividend and only 1 with a score >40 did so for financial issue reasons.”
I write as a vote of confidence for your service. As a professional money manger and speaker on retirement planning topics. Your site offers a logical and actionable approach to the investment management portion as a guide to investors. Long live dividend & intrinsic value investing.
Look forward to bringing your service to the attention of others.
Director of Wealth Management
I really appreciate your vote of confidence! Coming from you, that especially means a lot. Keep up your great work, and here’s to many more years of happy and healthy dividend growth investing.
I woke up last night and realized what was bothering me. It was not the high % of false positives. It is the language associated with the 5 safety categories. I feel it does not fit the risk of a dividend cut associated with the category as supported by data.
The data says that at a safety score of 60+ the risk of a dividend cut is essentially 0. Perhaps the 60-80 category could be labelled essentially 0% risk of a dividend cut and the 80-100 category could be labelled essentially 0% risk of a dividend cut/very safe.
A safety score of 41-60 does not represent an average risk of a dividend cut. About 60 out of 2000 stocks in the covered universe cut their dividend or 3%. Of the stocks with a safety score of 41-60 only 1 out of 405 or about 0.3% cut their dividend. They are still quite below average for risk of a dividend cut. Perhaps 1/10 the average or slight risk of a dividend reduction would be a suitable category descriptor.
A safety score of <40 represents about a 2-3x greater risk than average of a dividend reduction (7% vs 3%). A safety score of 20-40 represents about the average risk of a dividend reduction (3.2% vs 2.9%) Perhaps it could be labeled average risk of reduction or about 3% risk. The 0-20 safety score stocks represent about a 3x greater than average risk of dividend reduction (10.2% vs 2.9%). Perhaps it could be labeled 3x greater than average risk of dividend reduction or about 10% risk.
Based on the data, a portfolio of 10 stocks with safety scores of 70 has no more risk of a dividend reduction than one with 10 stocks with safety scores of 90. However, a portfolio score of 90 gives the holder a greater, false sense of security while limiting their investment options. Perhaps trying to represent portfolio safety with one number is misleading/unnecessary. Maybe the investor is better suited with a histogram and table of % holdings and % dividends from holdings in each category.
SSD is a great tool and we just need to let the data drive the extent of risk and language used. I'm certain the suggested category language/descriptors above could be improved. also, including prior years data could modify the above mentioned % figures. However, I would not expect a large change.
I'm certain I'll have more ramblings but that is all for now….
Thanks, John. I think it will be easier to modify the language as more data becomes available. For example, if we were to have had Dividend Safety Scores in 2008-09 and the scores were effective, perhaps we might feel the language was actually not aggressive enough. As it stands today, I believe the language is conservative (perhaps overly conservative). I’ll keep your feedback in mind. Thanks again for sharing it.
Thank you for all of the great and thoughtful comments! My wife gave birth this week to a boy after a 19-hour labor (very proud of her), so I’ve been juggling a lot of different balls lately.
We hope to return from the hospital this weekend, and I’ll be playing catch-up over the next few weeks as it relates to replying to comments and emails. I like to reply to everyone personally and apologize for the delay!
2017 will be an exciting year for continued improvements to the site and overall service…stay tuned for some good news next month!
Congratulations! A new little one is always exciting.
Being a 19h labor, I assume this was your first child. Be sure to let the wife know the next 5-6 will be easier.
Look forward to the 2017 site and service enhancements.
Best wishes to the family.
Thanks! There has been no shortage of excitement around here. She would love to have 5-6 more, but I tell her we will take them one at a time 🙂
Congratulations on the birth of your son. Wish you all the best of health and happiness. And, thanks for the fine newsletter!
Thank you, HIgh_Sierra! I really appreciate your good wishes and am so glad you are enjoying the newsletter. I hope you have a wonderful 2017 as well.
Congratulations Brian on the birth of your son and thank you for all the wonderful analysis…I have adapted my portfolio based on your recommendations and look forward to even more success/portfolio growth in 2017.
Could you add another column to the Portfolio Analyser that tracks the Dividend Safety to show Current SS? That way, we could see change in SS at a glance?
Thank you, car123wal! Our little guy has been a big blessing and brought us a lot of joy already. I’m really happy to hear that our information has helped you build up a stronger, safer portfolio. A number of folks have requested the ability to track changes in Safety Scores, and we will be looking into it. Thanks for the suggestion, and I hope 2017 is off to a great start for you!