Although I’ve written in the past about different categories of stocks and portfolio diversification, I haven’t spent much time on the question of dividend stocks. Everybody seems to be in love with dividend investing right now. Looking at the Guide to Personal Finance Blogs, there are at least 40 investing blogs that focus specifically on dividend investing and hundreds of recent posts that discuss some aspect of dividends. We all love the idea of lounging in piles of cash dividends, but is it mindful for individual investors to seek out dividends or prefer dividend stocks?
The Dividend Investing Love Story
I reviewed a bunch of articles from personal finance blogs, websites, and other publications to find out why so many people love dividend investing. How do they love thee, dividend investing? I could count many ways, but the core reasons for dividend investing generally fall into five main categories.
1. Better Performance – Dividend stocks have historically out performed the broader stock market. It’s easy to find data presentations like this one from Ned Davis Research supporting the superior track record of dividend stocks.
2. Less Volatility – Dividends help to moderate stock price fluctuations in two ways. One, the prices of dividend stocks tend to be less volatile over time than non-dividend payers or “growth” stocks. Here’s one example from Portfolio Visualizer showing the volatility (standard deviation) and maximum draw downs from 2007 to present for Vanguard’s Dividend Appreciation Exchange Traded Fund (VIG) as compared to Vanguard’s All Market Exchange Traded Fund (VTI).
|ETF||Volatility||Maximum Draw Down|
Two, dividends help moderate downward price fluctuations by providing additional value in the form of regular cash payments. The reasoning goes like this: if the market price of your dividend Exchange Traded Fund (ETF) drops by 5% in one year, but pays a 3% annual dividend, then the net loss in value is only 2%.
3. A Proxy for Financial Health – The fact that a company can regularly generate sufficient cash to pay dividends demonstrates its stability and health. While myriad accounting tricks can be used in annual reports to obscure the true financial health of a company, as someone clever once said, “Cash doesn’t lie”. Warren Buffett, perhaps the greatest all-time assessor of excellent company management, is known for buying stocks like Coca-Cola that have consistently increased dividends over the long term.
4. Reliable Income Stream – Undoubtedly, the main source of the love for dividend stocks is the regular income for retirees and “passive” income fans alike. The reliability of this income stream can be improved by focusing on companies with a consistent history of paying and even increasing dividends, which are called “dividend growers”. And the impact of any one company’s decision to cut dividends can be minimized by investing in baskets of dividend stocks using funds.
5. Lower Taxes – The U.S. government taxes most stock dividends at a lower rate than more ordinary income from cash, certificates of deposit, or bond interest payments. Depending on your income bracket, the dividend tax rate can be as low as 0% and no higher than 20%. This favorable tax treatment was unaffected by the new Tax Cuts and Jobs Act of 2017.
Breaking Up with Dividend Investing
Those five reasons may be stoking your passion for dividend stocks. But if you know anything about Mindfully Investing, you know we don’t accept things on face value. It turns out there are some pretty good cold-shower counterarguments for each of these five reasons.
1. The Better Performance Claim is Questionable – Although that Ned Davis graph above looks compelling, it’s based on some questionable statistics. (Mark Twain would’ve seen this one coming from a mile away.) Meb Faber at Cambria Investment Management picked apart this graph. He calculated the return since 1982 for a broad swath of decent performing stocks (the top 3000 companies), all dividends stocks, and dividend growers.
|Weighting||Top 3000 Stocks||All Dividend Stocks||Dividend Growers|
|Market Cap Weight||11.64%||11.89%||11.54%|
Because most stock funds are market-capitalization weighted, the top line of this table is probably the most relevant for individual investors like us. Meb asks in his post, “Where’s all that towering out performance of dividend growers?” Well, it turns out that the Ned Davis graph and Meb’s table are calculated two different ways, and the graph method is a bit outdated for reasons you can read about in Meb’s article. In their own updated calculations, Ned Davis now shows the following performance metrics since 1972 (the period covered in the graph):
- S&P 500 – 12.35%
- All dividend stocks – 12.83%
- Dividend growers – 12.89%.
In the words of Rick Perry, “Oops”! While some of these newer metrics show dividend stocks and/or dividend growers performing slightly better than the broader market, it’s certainly not the slam dunk for dividend stocks that was initially implied by the Ned Davis graph. In fact, these differences in performance levels are so small that they may be difficult to reproduce in the real world. For example, perhaps in 2007 you decided to pick Vanguard’s Dividend Appreciation ETF (VIG) to execute your dividend investing plan. Portfolio Visualizer indicates that VIG provided an annualized return since 2007 of 8.69%, while Vanguard’s All Market ETF (VTI) returned a slightly better 8.86%. Somehow that expected small performance bonus for dividend growers evaporated just because of your implementation choices and time frame.
Further, Larry Swedroe points out that for the past 20 years, models using just four factors explain about 95% of the differences in returns between diversified portfolios. (Those four factors, which are explained more here, are volatility, size, value, and momentum.) Adding a few other factors can provide a few percent more in explanatory power, but none of the widely accepted factors isolates dividends. Meb Faber supports this point by presenting the historical performance of portfolios based on the “value” factor as compared to an example dividend investing portfolio, as shown in this graph.
The line colors are a bit hard to see, but most of these lines represent variations on a “Valuation Composite” portfolio. The order of performance for these portfolios since 1995 was:
- Valuation Composite (ignores dividends completely) – 16.79%
- Valuation Composite excluding top quintile of dividend payers – 15.62%
- Valuation Composite excluding dividend payers above 1% yield – 14.68%
- Valuation Composite excluding the top half of dividend payers – 13.72%
- Top 400 Dividend Yield stocks – 11.86%
- Top 2000 Stocks Equal Weighted (broader market) – 9.99%
A lover of dividends will quickly note that the Dividend Yield portfolio (#5) out performed the broader market portfolio (#6) in this period. But that’s putting lipstick on a pig. The simplest value portfolio that completely ignores dividends got the gold medal, while the dividend portfolio was entirely out of medal contention at fifth place. Surprisingly, value portfolios that actively exclude some portion of the top dividend payers (the second through fourth portfolios) also beat a dividend-focused portfolio. Simply put, a dividend portfolio is just a poor man’s version of a value portfolio; one that’s unfocused and inefficient.
Even further, bonds haven’t performed their usual income function over the last decade because of super low interest rates. As a consequence, many investors turned to higher yielding dividend stocks to boost income. This income seeking behavior caused dividend stocks as a group to go from being unloved and undervalued to overvalued relative to the broader market, as shown in this graph presented by Meb Faber.
Historically, dividend stocks were about 25% less pricey than the rest of the market, but today they’re about 10% more pricey. As I’ve discussed before, higher present-day valuations are a pretty good predictor of lower future returns over the next 10 to 15 years. Thus, while dividend stocks may have slightly out performed the broader market in the past, the recent popularity of dividend stocks may have ended that story.
2. Volatility Does Not Equal Risk – Dividend lovers hope to moderate volatility in two ways: 1) smaller intrinsic price fluctuations and 2) counter balancing price declines with cash dividend payments.
On the first count, lets look at our table of VIG (dividend) versus VTI (all-market) volatility again but with some added performance statistics.
|Measure (2007 to 2017)||Dividend (VIG)||All-Market (VTI)|
|Volatility (standard deviation)||12.73%||15.14%|
|Maximum Draw Down||-26.69%||-36.98%|
|Best Subsequent Year||28.87%||33.45%|
As noted before, VIG had about 2.5% less volatility (standard deviation), and the difference in maximum draw down between the two funds was about 10%. While that sounds appealing, you have to ask yourself whether you’re going to feel completely different when your portfolio declines by 27% instead of 37%. I’d wager your going to feel pretty bad either way. Further, what different actions would you take when faced with these two outcomes? If you had sold either fund at the depths of the 2008 market crash, you would have missed out on huge gains in the next few years. And both funds provided a nearly identical annualized return over this period.
As I’ve discussed before, temporary declines in prices are not the true risk that we need to avoid. The true risk is the permanent loss that’s created by panic selling during a market crash. Standard deviation or other measures of routine volatility are a dismal gauge of the risk that matters most to real-life investors. The true risk is our own counterproductive behaviors, not 2 or 3% differences in routine price fluctuations. And successful investing requires that we control those bad behaviors regardless of whether we buy dividend stocks or non-dividend stocks.
On the second count, the argument for how dividend payments moderate price declines is fundamentally flawed. When a dividend is paid, the stock’s value is offset by an equal fall in the stock’s price. Following the previous example, if your dividend ETF declined by 5% one year and paid a 3% dividend in the same time span, 3% of that decline was caused solely by payment of the dividends. You can tell yourself a comforting story that the dividend payments caused a net 2% decline, but that’s just confusing cause with effect. In fact, an exactly equivalent stock without dividend payments would have also declined by only 2% under the same circumstances.
3. Cash Can Lie – While cash dividends may be related to financial health of a company, there are more direct measures of health. We’ve already talked about value-based funds, which try to select undervalued companies using measures of financial health such as: price earnings ratio, price book-value ratio, free cash flow, debt-equity ratio, and price earnings to growth ratio. And by emphasizing these more direct measures of health, we’ve seen that value portfolios can often perform better than dividend-focused portfolios.
More importantly, dividend-paying companies are under considerable pressure to maintain and even raise their dividends over time, exactly because dividends are widely believed to indicate financial health. Company managers understandably loath the stock price drops and increased public scrutiny that follow dividend cuts. But avoiding such cuts may simply exacerbate existing financial stresses until something snaps.
Some companies go to great lengths to continue dividend payments, almost regardless of their actual financial health. Just after the 2008 crash, many ailing companies issued low-interest bonds and then used the borrowed money to pay dividends to stock holders. I’m no Fortune 500 CEO, but issuing dividends with borrowed money seems like a perfect example of placing perception ahead of responsible management. If imprudent managers of non-dividend companies can game their annual reports to goose earnings, it’s equally possible that imprudent managers of dividend companies can game their finances to issue cash dividends.
Also, trotting out Warren Buffett, America’s favorite investing uncle, in support of dividend investing blatantly ignores the history of his management of Berkshire Hathaway, which in its entire company history only issued a dividend once in 1967 for 10 cents per share. Uncle Warren quipped, “I must have been in the bathroom when the decision was made.” Buffett has been clear that a good manager should be able to return more shareholder value by using cash to reinvest in the business, reduce debt, buy back company stock, or buy new businesses. His advice points investors toward companies that have a demonstrated ability to use cash productively, not toward companies that consistently pay dividends regardless of their financial situations.
4. Just Another Income Stream – Above all, lovers of dividends covet those regular cash payments. You buy the stock, and with no further effort, beautiful fresh cash shows up in your account like clockwork. I’ve already mentioned that when dividends are issued, the price of the stock declines by the same amount. So, what does that mean for dividend stock owners versus non-dividend stock owners when they need to generate income?
- Dividend Owner’s Income – Imagine you own 100 shares of The Heart Company, which is priced at exactly $1 per share ($100 total value) on the day it issues a one cent per share dividend at the close of trading.
- The next day at the opening of trading that same 100 shares of stock is priced at $0.99 per share to account for the one cent dividend paid the previous afternoon.
- You receive $1 in dividends in your account. Your account value is now $99 in the form of 100 shares of stock and $1 dollar in cash from the dividend.
- You have exactly the same total value of $100 as you had on the previous day, but in a slightly different form.
- Non-Dividend Owner’s Income – Now suppose you owned 100 shares (at $1 per share) of The Brain Company, which is identical to The Heart Company, except The Brain Company doesn’t issue dividends.
- What if you need $1 of income to complete a deposit on a new Tesla Model 3?
- You sell one share of The Brain Company to create “homemade” income of $1.
- The next day your account has $1 in cash and $99 worth of stock, for a total value of $100.
The resulting value for the Dividend Owner and the Non-Dividend owner is the exact same $99 in stock and $1 in cash. (I’ll tackle the issue of taxes and transaction costs in a moment.) There’s nothing inherently better about $1 received as a dividend and $1 received by selling one share of stock; it’s still a dollar.
However, a dividend company can cut or eliminate its dividend with little warning. In contrast, you can generate homemade income at anytime from any stock. Dividend investors hope to avoid dividend cuts by buying companies with a history of consistently paying and increasing dividends. Unfortunately, history is littered with good companies that paid dividends for decades, until they simply had to stop or went bankrupt. Jason Lina pointed out in a recent Forbes article that Eastman Kodak paid an uninterrupted dividend for over a century before going bankrupt. Companies like Bethlehem Steel, General Motors, US Steel, and Goodyear Tire paid dividends nearly as long before suffering the same fate.
More recently, dividend darling General Electric, which hadn’t cut its dividend since the Great Depression, made two cuts in the last decade, in 2008 and 2017. Does this mean that GE is headed for bankruptcy? I have no idea. But I think GE would be in better shape now if it hadn’t aggressively increased dividends again after the 2008 cut, as shown in the middle panel of this chart from Macrotrends.
Even after cutting the dividend in half in late 2017, GE will continue to pay out nearly 85% of its free cash flow as dividends, which still sounds pretty unhealthy to me. Dividend investors often claim that dividend income is safer and more reliable than homemade income, but in actual fact, nearly the opposite is true.
Some dividend lovers further guard against cuts by selecting dividend funds, which are baskets containing many dividend stocks. But even funds are not entirely immune to dividend cuts in dire times. Corey Hoffstein points out that the S&P the 500 Dividend Aristocrat Index, which has grown dividends at a more than 6% annual rate over the last nine years, saw annual dividends collectively decline by 8.5% from 2008 to 2009.
5. Higher Costs (Lower Net Income) – It’s true that dividends are taxed lower than bond income and other forms of interest. But that’s a misleading comparison. Given we can produce homemade income by selling stock, the more appropriate comparison is between dividends and capital gains. So, let’s compare the The Heart Company to the The Brain Company again, and look at how dividends versus homemade income compare after taxes and trading costs. We can make the example a little more realistic by bulking up the income needed to $1,000.
|Costs on $1,000 of Income||Heart Company Dividends||Brain Company Homemade Income*||Difference|
|Tax Due 0% Bracket||$0||$0||$0|
|Tax Due 15% Bracket||$150||$75||$75|
|Tax Due 20% Bracket||$200||$100||$100|
|Trading Cost||$0||$10||– $10|
|0% Bracket||$1,000||$990||– $10|
*Assumes 50% of the income is subject to capital gains taxes.
Only in the lowest income bracket does the dividend approach net more income, but it’s only $10 more. In all other cases, the net income is greater using the homemade income approach. For homemade income, I used a cost of $10 per trade, which is typical for a low-cost online broker. I also assumed that 50% of the value of the stock being sold was subject to capital gains. Obviously, this number will vary in real life, but unless you assume the stock was bought for $0, which is impossible, the tax burden will always be higher for the dividend approach.
With homemade income you can further drive down costs by selecting which tax lot of shares to sell. You can even sell shares that have declined in value, and get a tax break on your other income. The selling-at-a-loss option often confuses dividend lovers, who focus on the fact that they don’t have to sell shares at a loss when reaping dividends. But remember, to make a fair comparison, the Heart and Brain companies must be identical including any decline in share price; the only difference is whether dividends are paid. If both companies’ stock prices decline, they both lose price value, but only the Heart Company also loses value because cash was paid out.
After the Break Up
The metaphors I’ve been using signal my opinion. A preference for dividend investing is mainly an emotional decision, not a rational or particularly mindful one. However, given that dividend portfolios can perform well over the long term, I don’t think dividend investing is totally irrational either. Further, I’ve often cautioned against arrogant attempts to highly optimize and diversify stock portfolios with the hope of devising the “best” portfolio for every potential future market and economic condition. A dividend-focused portfolio may turn out to provide as good, or even better performance in the next decade as compared to, for example, an all-market portfolio. My only complaint is the dubious and emotional assumption that any superior performance of dividend portfolios is mostly due to the magic of dividends.
If dividend blogs are any indication, many dividend investors seem to be spending the “passive income” generated from their dividend portfolios in early retirement or just to supplement regular income. Given that 40% of the S&P 500 historical returns come from dividend reinvestment, it’s important to realize that spending dividends eliminates any hope of keeping pace with the performance numbers presented in this post.
If dividend investing is such an emotional decision, why do so many investors seem to prefer it? In my next post, I’ll talk about the emotional motivations and cognitive biases that are likely behind the love for dividend investing. In the meantime, I suggest you think about how your going to end your love affair. Perhaps you should take your dividend portfolio out to lunch at a busy restaurant, where you can make the break gently.