I’m sure there are more applicable message boards out there for this topic since it’s not directly about the CFA exam, but this one has some intelligent followers so figure why not… I’ve never understood the infatuation with dividends, it seems most people believe them to be free money without considering the corresponding decrease in company value. If a stock price didn’t decrease by the amount of the dividend (approximate because ceteris parabus doesn’t exist in reality), it would create the easiest arbitrage situation for traders by flipping the stock around its ex-dividend date. To illustrate my point, just yesterday on CNBC they were discussing IBM’s 3-year 1.14% debt offering and comparing it to their 2% dividend yield. Obviously 1.14% is extremely low, but comparing the yields is comparing apples to oranges. They make it sound like just by holding IBM stock you’re guaranteed a total return of 2% per year, but forget that for a dividend-paying stock it has to recoup the dividend paid out in capital gains for the stock price to stay the same. If a stock does not have any capital gains and pays out a dividend you won’t get that 2% overall return. The $100 stock will now be a $98 stock with $2 now in your pocket, for a total return of 0%. Furthermore, if the stock price goes down, your return can certainly end up much worse than the 1.14% bond return. The only real advantage to dividends (as opposed to repurchasing stock when dividend and cg tax rates are equal) I see is that it’s actual money in your pocket instead of a paper gain, but then again if you reinvest the dividends this is moot. Dividends obviously have important signaling effects and affect investor sentiment, but these are intangible and based more on the “that’s the way it’s always been” line of thinking than actual logic. Does anyone else have an opinion on this? I’m not claiming that dividends don’t matter but I just don’t understand the perceived premium high-yield stocks receive.
most of the gains in long term stock values comes from dividends, which comes from company strength… snooki got arrested…
I have not read the CNBC report, but just from what you said, 1.14% for a 3-year IBM bond is poor. You can shop around and able to find term deposit from FDIC-insured bank with better rate. The default risk and potential for capital loss may not worth the risk for the bond. Bond price can go down too with 1.14% coupon. Holding the stock won’t guarantee 2% return. It barely says dividend yield is 2% before factoring in capital gain/loss, if dividend continues. If you keep the $2 cash dividend, then the stock value may be decreasing from $100 to $98 then to $96 … But you ignored the growth of the stock price and the earning growth. If furture grow is not paid out as dividend, then it retained the value of the company, aka, the stock price. If even dividend growth is zero, but the dividend yield is increasing. If you reinvest the dividend like DRIP, then you are in better position. Although you total book value may be the same before and after dividend , e.g. $100 Vs. $98 + $2, but with DRIP you now have more units of stock despite each unit of stock now worths less ex-dividend. Your units of stock is growing. Dividend is definitely win over bond in after-tax return if you factor in the dividend tax credit, while bond coupon is taxed at highest marginal tax rate.
Exactly, it’s just playing to investor’s emotions and history, because tangibly that gain could have occurred via stock repurchases instead of dividends (reinvested dividends and repurchased shares will have the same result). Plus many very strong companies do not and have not paid dividends. However, my biggest beef is comparing an interest yield to dividend yield and claiming a dividend is “pure gravy” because the stock doesn’t even have to go up for you to collect your annual dividend yield.
whui - you’re missing my point and I’m confused by your response. I agree 1.14% is poor and am certainly not arguing for bonds. I just used it as an example. Also, IBM is virtually default free over a 3 year time horizon, as evidence by their debt pricing. I did forget to add the assumption of holding the bond til maturity, which locks in the 1.14% yield so you don’t worry about price fluctuations. I agree it won’t guarantee a 2% return and that was my point! I purposely used the assumptions that the company doesn’t grow, so I’m not sure what you’re arguing here, since the dividend yield increasing doesn’t help the investor if the company doesn’t grow. My units of stock grow when a stock splits too but I wouldn’t quit my day job just because my units increased. If everyone utilizes a DRIP, this yields the same result as a stock repurchase combined with a split. Dividend is definitely the preferred choice? Hmm…in 2008 IBM’s capital loss was -22.1%, factoring in dividends was -20.4%. If you’re telling me that’s better than +1.14% I hope you never manage my money.
Compounding dividends through automatic reinvestment inside a Roth IRA is one of the sexiest tricks I have ever seen.
MrAndMatt Wrote: ------------------------------------------------------- > Compounding dividends through automatic > reinvestment inside a Roth IRA is one of the > sexiest tricks I have ever seen. If the dividend cash were instead used to buyback shares you have the exact same result. Instead of the stock price decreasing by the dividend amount it increases because earnings are spread across less shares. The only difference is the amount of units you own hasn’t increased, but it’s proportional and a stock split is just an accounting change and doesn’t affect value. Everything is tax-free at withdrawal nomatter how the gains are allocated among dividends and capital gains.
Share repurchases suck because companies do them when times are good and their stock price is high. There are countless examples of companies buying back their shares near their 52-wk high. It’s generally a poor use of capital and extremely unpredictable. Dividends, on the other hand, are at least supposed to be steady. Over the long term about 40% of the market’s total return is attributable to dividends so they’re obviously worthwhile. Problem for the “average” investor - say in a Roth IRA - is the amount you’re going to get from a dividend payout isn’t enough to justify buying more shares of stock once you take transaction costs into account. For example, that $5,000 you have in a stock that’s yielding 3% is only $150 to reinvest…yay! Now if you want to own a dividend focused mutual funds that automatically reinvests your income, then it’s a little bit better. But obvious drawbacks are included there as well.
Sweep - I’ve seen your posts in other forums and appreciate the input, they are always among the higher quality ones and you know what you’re talking about. If I may follow up with a couple questions, moreso to play devil’s advocate than actually support the opposing view, as my arguments are more based on book-learning and theory rather than industry experience which can be dangerous… First, don’t share repurchases happen when stock prices are beat down because the company thinks they’re trading below market value? At the very least it should be sending a signal to investors that the stock is undervalued because what insider would want to buy high and sell low? The company’s purpose is to increase shareholder value, be it through primary operations or indirect financial decisions, so there’s no incentive for the company or shareholders to buy when overvalued. 40% of the market’s total return is attributable to dividends, but that’s only because it’s a customary practice that people have gotten used to and older generations associate it with stability and practice mental accounting to spend only dividend income. If companies traditionally took their net income and used 40% to buy shareholders Christmas cookies then 40% of the value of holding stocks would attributable to Christmas cookies. Not a perfect comparison, but I’m contending that dividends compose 40% only because that has been the chosen payout- a proportionally higher repurchase rate would give the same total return and allocate 0% to dividends.
The effect maybe the same, but I believe Sweep the Leg is right. Share repurchases are a matter of timing the market, which has a little to much risk for my taste. I would say that dividends require a certain amount of fiscal discipline to maintain, and can not be “faked”. Much like the earnings vs. cash flow debate. Cash flows cannot be faked. Cash comes in, cash goes out. Dividends are paid out, or not paid out. Share buybacks are tricky. There are numerous instances where companies annouce a buyback but never get around to spending the full amount of money they designated; Amazon being the most recent example. So in the debate of dividend vs. buyback, I choose dividend. Even though I know they look the same on paper, and that they theoretically provide me with the same amount of value, I would always choose a dividend over a buyback.
Matt - that I agree with, and is probably the best argument for dividends with respect to not being able to fake them. However, if everyone reinvests dividends, how is this any different? It’s not as obvious, but there is the same amount of market timing - instead of the company purchasing their own shares, upon receipt of the dividend the investors use this money to purchase shares. To be honest, I would assume the board has a better idea than I would about the company/stock price’s outlook. I guess the peace of mind of knowing that you could choose to pocket the cash rather than reinvest it is worth something, but you said yourself you like to automatically reinvest the dividends.
Sun - Look at MHP huge buybacks right before the markets started to tumble…If any board should have known to wait a year or two to do buybacks, it was them. But they you get to another problem. The problem that IBM and MSFT have. Where the $3-$4 billion dollar annual buybacks are used to support the stock price. There was a report the other day that said MSFT has given back $169 billion over the past 10 years in buybacks and dividends, yet look where the stock price is. If I was given dividends, my overall holdings would have increased in value as would my percentage of equity ownership. If MSFT just did buybacks, I would own more of the equity as a percentage of the total, but my brokerage account would actually show a loss in total value over the years. Although that is a very rudimentary hypothetical example, I think it gets the dividend vs. buyback point across. Biggest reason for automatic reinvestment: No transaction fee.
we know dividends are overhyped. over the long term, small caps outperform large caps, period. but, dividend paying large caps outperform ‘growth’ large caps on a risk-adjusted basis, and maybe even absolutely, but i’m not 100% sure about that. the fact that ‘growth’ large caps buyback shares shows you the lack of confidence they have as a long-term entity. the act of buying back shares leaves you the ability to issue shares at a higher price, post-buyback, whereas a cash dividend is money leaving the company for good. any company that pays a cash dividend should expect to live forever, and if they’re not expected to live forever, its just bad management to be paying a cash dividend. in the long-run, holding all large-cap dividend payers will likely provide you with the best return should you only own 1-30 stocks in your portfolio as dividend payers are often the ones that can withstand market downturns. ‘growth’ large caps and small caps die during these downturns, and when one owns individual stocks, one death in a group of 20 can have an extremely harmful effect on your portfolio’s return. if you have a long time horizon, have access to a diversified small cap fund/etf, and do not need income, buy a small cap fund/etf. if you need income or wish to hold only a few individual equities, buy dividend paying large caps. that said, dividend payers are usually for those who need income, thats really the main reason for buying them. this stuff is rudimentary, basic Level 1 portfolio management.
another reason why people like to buy dividend payers is the fact that people don’t like to see volatility in portfolio values. again that comes back to superior risk-adjusted returns of large cap dividend payers.
Hey, great post. And the role of dividends is a natural topic for a CFA forum. A) OP, you’re basically right that investors should focus on total return (more specifically, after-tax total return, and the different tax status of dividends vs capital gains is relevant there) rather than get too focused on the dividend alone. That said, there are some ways in which the dividend is important to think about: 1) If you look at historical stock prices, you might think that the stock is not doing anything very impressive, but the dividend yield is not usually shown in the historical stock price chart, so you can get into trouble if you are comparing dividend with non-dividend paying stocks on charts. 2) In this economy, the push for dividend paying stocks is more about the fact that companies that pay stable dividends tend to have more stable cash flows from which to make payments. If there is a lot of worry about a double dip recession, these stocks are likely to take a less serious hit (in total return terms) than non-dividend paying stocks. So some of the dividend craze is specific to this moment in history. 3) If you are an investor that needs income (say a retired person that still has monthly living expenses to meet), dividends can be useful because they allow you to reduce your net exposure to bonds. What this does is allow you to substitute some interest income with dividend income and provide an opportunity to capture the equity risk premium through capital gains on the stock, as opposed to being stuck with just the bond risk premium. For someone early along in retirement, this can be beneficial because they still have time to take advantage of the equity risk premium. This does assume a normal economy in which stock prices are expected to rise over the long term. 4) If you are an institutional investor with tax-exempt status, dividends can help your reduce rebalancing transaction costs, because you do not have to sell as many stocks to rebalance - you simply allocate the cash received from dividends. 5) Companies that pay dividends are generally mature companies that cannot generate a sufficiently high ROE from retained earnings or acquisitions. Over the long term, the returns from reinvested dividends compound into an amazing percentage over capital gains alone. However, that is only realized over the long term. At the same time, any company that is not mature and survives over the long term should eventually mature and start paying dividends. So, over the long term, you don’t just want to hang on to non-dividend-paying stocks, unless you have a specific strategy that requires not paying dividends. 6) It is true that dividend payments generate a tax liability for investors, and this is generally undesirable. This can be mitigated by using the dividends for share repurchases, but the consequence is that one doesn’t know if the stock price is overvalued, undervalued, or fairly priced when the dividends are reinvested. So those are some of the things to think about when considering dividends.
Dividends are even more important than sweeptheleg said - , a 2004 study by The Brandes Insitute, ‘Examining The Income Component of Total Returns’, found that for investment periods of 10 years or more, the income component (i.e. the dividends and interest) made up a greater proportion of the overall return than capital appreciation in all income-producing assets. The proportion of the total return attributable to income return (for US equities from 1926 to 2003) amounted to 63% for each 20 year rolling average. Find it here http://www.brandes.com/Institute/Documents/Income%20Components%20of%20Returns%20Paper%20071204.pdf
MrAndMatt- I disagree. Nothing is a perfect leading indicator, and for every example of a buyback preceding a downturn can be matched with one preceding a rally. Regardless, the signal given by an announced buyback is positive, thus stocks typically react positively to buyback announcements and negatively to share issuance. MattLikesAnalysis- I’m not disagreeing with your contentions, but you’re implying cause and effect when the relative outperformance of dividend paying stocks is a weak correlation at best. Also, dividends do not add or subtract value, it is a transfer of money from the company to the investor and for the investor an accounting transfer from equity to cash, I do not see how this lowers volatility in portfolio values. The only way I can see your point being valid is considering transaction costs, because I would argue if a company repurchases stock its stock price will benefit because it’s allocating its total return entirely to capital gains instead of dividends and thus you could create your own dividend by selling an amount equal to your desired spending yield. Rudimentary Level 1 material? I disagree. bchadwick- great post, I’d nominate that for inclusion into the CBOK because it’s obvious from some of the other responses that dividends aren’t clear-cut in their utility. I agree with each of your points, just want to add a few thoughts: Regarding 5) If you agree that buying back stock is in effect the same thing as paying a dividend and reinvesting ALL investors’ dividends for them, wouldn’t the total return be equal to reinvested dividends+cg? On 6) you list a consequence of share repurchases is the uncertain value at the time of the repurchase. Perhaps it’s too much to assume, but if everyone enrolls in a DRIP then the money gets plowed back into the stock right after the dividend date and the uncertainty about over/undervalued stock prices still remains. I guess you could argue that dividends are smaller and occur at regular intervals thus diversifying the timing risk. The main reason I started this post was the CNBC comment that put me over the top. Comparing a debt yield to dividend yield is ridiculous and I often hear comments implying dividends being guaranteed returns and that the stock price doesn’t have to do anything for you to collect your annual dividend yield. I still think dividends are overrated by the great majority of the investment community, but I will concede they play a significant role in in portfolio management and are generally correlated with mature and successful companies.
newsuper Wrote: ------------------------------------------------------- > Dividends are even more important than sweeptheleg > said - , a 2004 study by The Brandes Insitute, > ‘Examining The Income Component of Total Returns’, > found that for investment periods of 10 years or > more, the income component (i.e. the dividends and > interest) made up a greater proportion of the > overall return than capital appreciation in all > income-producing assets. > > The proportion of the total return attributable to > income return (for US equities from 1926 to 2003) > amounted to 63% for each 20 year rolling average. > > Find it here > http://www.brandes.com/Institute/Documents/Income% > 20Components%20of%20Returns%20Paper%20071204.pdf newsuper - I thought I addressed this already. It’s a trade-off. Given a total return, if dividend yield goes up, capital gains yield goes down. Dividends have provided 63% of total return because the convention has been to pay dividends. I’ll admit it’s a large assumption, but if dividends were instead used to buy back shares, the compounding aspect would still be in effect because the total return would be the same. I’m not arguing that dividends have no impact on capital appreciation, but since total return is the bottom line, buying a stock based on its dividend yield makes about as much sense as buying a stock based on its forecast capital gain yield (good luck calculating that).
Don’t dividends have a similar effect as debt? Ie, paying regular dividends keeps pressure on management to stay efficient, not empire build, increase the dividend and therefore not just build up revenues but earnings, etc. Also, assuming investors pay taxes, the theoretical price of a stock is the after tax cash flow stream to the investor – after the investors individual taxes are taken into account. If you own a dividend paying stock in a tax-preferred account, like a Roth IRA, you should see the value of the stock fall by the after tax value of the dividend, but you receive the full value of the dividend as you pay no taxes on it. This should be a way of capturing tax alpha.
sundevl21 Wrote: ------------------------------------------------------- > MattLikesAnalysis- I’m not disagreeing with your > contentions, but you’re implying cause and effect > when the relative outperformance of dividend > paying stocks is a weak correlation at best. > Also, dividends do not add or subtract value, it > is a transfer of money from the company to the > investor and for the investor an accounting > transfer from equity to cash, I do not see how > this lowers volatility in portfolio values. The > only way I can see your point being valid is > considering transaction costs, because I would > argue if a company repurchases stock its stock > price will benefit because it’s allocating its > total return entirely to capital gains instead of > dividends and thus you could create your own > dividend by selling an amount equal to your > desired spending yield. Rudimentary Level 1 > material? I disagree. You’re missing the bigger picture here. You’re focusing far too much on taxes and transaction fees and are not appropriately considering risk. Volatility IS lower for dividend-paying companies over non-dividend-paying companies as cash flow is more reliable and dividend-payers experience lower highs and higher lows, relative to non-dividend payers as the more reliable, non-allocated (to capex) cash flow acts as a buffer for bad times and the lower capex on the behalf of the company reduces upside. If returns of “growth” and “value”/dividend-paying large caps are similar over the long-run, which they have shown in the past, the “value” large caps have a much greater risk adjusted return. The reason why this is rudimentary L1 stuff is because the only connection you have to make is that risk/volatility generally goes up when a company is growing or dying (i.e. they are not paying dividends). Why is this important for an individual’s not fully diversified portfolio? Generally, dividend-payers don’t die and non-dividend payers do. If you have 5 non-dividend payers in your portfolio, the odds of absolute loss in at least one holding is greater than if you had 5 dividend-payers. Risk adjusted return is what we’re after right?