Anyone Else Only Buying Dividend Stocks?

[beltim]

I know it's been brought up already, but won't the market adjust and correct for that?  Or do the dividend investors know something the rest of the masses don't know?

Yes, it would, and it has been proven time and time again that it does.

No, they don't.

There is no free lunch here.

People keep saying that without any proof.  The strongest proof was your post earlier, which said that when people increase their bond allocation, their risk of the portfolio running out was greater... no shit, Sherlock.

In contrast, there's considerable data that shows that dividend-paying stocks have higher total returns.  For example: https://public.dreyfus.com/documents/manual/perspectives/dry-fsdwp.pdf

In principal, companies could use excess cash flow to buy back stocks and it would be even better than paying dividends.  However, CEOs are even worse investors than the average individual, and so are more likely to buy back stock when times are good (and the stock is high) than when times are bad, and the stock price is low (links in my post earlier).  The rare CEO who does buy when their company is undervalued - like Warren Buffett - creates tremendous shareholder value.  Even CEOs who consistently buy back stock would be better than a dividend.  But the average dividend policy in the US is to at least maintain the dividend, and grow it when possible.  This leads to better practical outcomes.

Is it your assertion that a brochure from a mutual fund company selling high ER funds, and claiming to be able to beat the market, is evidence that dividends are a free lunch?

No, and if that's what you took away from my post, you need to re-read it. If you don't like the source I gave, there's lot of others that show dividend-paying stocks have better total returns.  Here's a whole series of studies: http://www.mhinvest.com/studies.html

And it's not like this is new knowledge - people have been writing about it for at least a decade (http://www.mhinvest.com/files/pdf/NF_div_study_neddavis_2004.pdf), and probably much longer.

[hodedofome]

Enhancing the Performance of Yield-Based Strategies http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2051101

I've already shared my thoughts on efficient markets and rational investors in other topics, so I won't be responding to anything in that regard for this one. Will the market arbitrage away every efficiency imaginable? According to finance professors' theories - yes. According to actual market participants with real world trading experience - not so much.

[waltworks]

Any "studies" from anyone who *isn't* in the business of active money management? Seriously, I'd like to see them.

-W

[hodedofome]

Any "studies" from anyone who *isn't* in the business of active money management? Seriously, I'd like to see them.

-W

You would rather have 'research' from amateurs? I don't get it, the people that have the incentive to do quality research are the guys who are running money.

[beltim]

Any "studies" from anyone who *isn't* in the business of active money management? Seriously, I'd like to see them.

-W

Depends on how broadly you define the business of active money management.  The Ned Davis research above doesn't manage money.  AAII also has similar data.  Both, of course, sell research, predominantly to people interested in at least some active investing.

The larger issue, if you want someone unbiased and consider those sources biased, is how you could possibly get someone who wasn't involved in active money management.  Consider: someone does research and determines a profitable method of investing, better than index investing.  They would have to be a blithering idiot not to immediately become involved in one way or another - whether licensing their idea, becoming a consultant, working for a hedge fund, or something.

[waltworks]

Academics (folks like Schiller) do this kind of research all the time. Is there any academic research out there?

-W

[Dodge]

I know it's been brought up already, but won't the market adjust and correct for that?  Or do the dividend investors know something the rest of the masses don't know?

Yes, it would, and it has been proven time and time again that it does.

No, they don't.

There is no free lunch here.

People keep saying that without any proof.  The strongest proof was your post earlier, which said that when people increase their bond allocation, their risk of the portfolio running out was greater... no shit, Sherlock.

In contrast, there's considerable data that shows that dividend-paying stocks have higher total returns.  For example: https://public.dreyfus.com/documents/manual/perspectives/dry-fsdwp.pdf

In principal, companies could use excess cash flow to buy back stocks and it would be even better than paying dividends.  However, CEOs are even worse investors than the average individual, and so are more likely to buy back stock when times are good (and the stock is high) than when times are bad, and the stock price is low (links in my post earlier).  The rare CEO who does buy when their company is undervalued - like Warren Buffett - creates tremendous shareholder value.  Even CEOs who consistently buy back stock would be better than a dividend.  But the average dividend policy in the US is to at least maintain the dividend, and grow it when possible.  This leads to better practical outcomes.

Is it your assertion that a brochure from a mutual fund company selling high ER funds, and claiming to be able to beat the market, is evidence that dividends are a free lunch?

No, and if that's what you took away from my post, you need to re-read it. If you don't like the source I gave, there's lot of others that show dividend-paying stocks have better total returns.  Here's a whole series of studies: http://www.mhinvest.com/studies.html

And it's not like this is new knowledge - people have been writing about it for at least a decade (http://www.mhinvest.com/files/pdf/NF_div_study_neddavis_2004.pdf), and probably much longer.

Don't have time to go through all of the links now (will later), but it smells like Survivorship Bias.

http://www.bogleheads.org/wiki/Survivorship_bias

The charts simply show, "This group of stocks which have exhibited increasing returns every year for the past 25 years, have higher returns than the market as a whole."

That sounds like a reasonable statement to make.  How is that information actionable?  Shall I then invest money in the stocks which have performed well over the past 25 years, hoping they will continue to perform well in the next 25 years?  Alarm bells should start ringing on that one.

We know that of the original Dividend Aristocrats, only 7 still remain in the index.  We also know that only 30% of companies currently in the index, are still there after 10 years, with the average length for any one company being 6.5 years.  This might be why the returns look so good in hindsight, yet are difficult to achieve in real-time.  I highlighted this in the second reply to this thread:

http://forum.mrmoneymustache.com/investor-alley/anyone-else-only-buying-dividend-stocks/msg366691/#msg366691

During the 2008 great recession, the Dividend Aristocrats fund (SDY), a fund which is comprised of the 50 highest dividend yielding constituents of the stocks of the S&P Composite 1500 Index, that have increased dividends every year for at least 25 consecutive years, had lower returns than 100% bonds, and higher risk (volatility) than 100% stocks.  Let's see what that looks like graphed:



The market index VTSAX returned: 40.48%
The Dividend Aristocrats fund (SDY) returned: 18.38%
The Preferred Stock Index Fund (PFF), a high yield fund which tracks 220 preferred stocks from 44 U.S. companies and yields a yearly dividend in the 7% range: -20.90%

Maybe the real question is, how can you invest in Dividend Aristocrats, before they become Dividend Aristocrats?

[beltim]

Academics (folks like Schiller) do this kind of research all the time. Is there any academic research out there?

-W

And Robert Shiller is a founder and the chief economist of the investment management firm MacroMarkets LLC.

[beltim]

Maybe the real question is, how can you invest in Dividend Aristocrats, before they become Dividend Aristocrats?

So your response to 40 years of data, is to use a subset of the stocks I was talking about, narrow the range to the last 7 years, and say that this time it's different?

[waltworks]

Let's see a graph of a longer time period, then. How has the DA fund performed relative to a straight S&P one?

-W

[beltim]

Let's see a graph of a longer time period, then. How has the DA fund performed relative to a straight S&P one?

-W

The index that SDY is based on only goes back about 10 years.  There's 40 years of data in my link above.

[waltworks]

I thought it went back to 1989. See here: http://seekingalpha.com/article/578321-dividend-aristocrat-investing

Is that not correct?

-W

Let's see a graph of a longer time period, then. How has the DA fund performed relative to a straight S&P one?

-W

The index that SDY is based on only goes back about 10 years.  There's 40 years of data in my link above.

[beltim]

I thought it went back to 1989. See here: http://seekingalpha.com/article/578321-dividend-aristocrat-investing

Is that not correct?

-W

Let's see a graph of a longer time period, then. How has the DA fund performed relative to a straight S&P one?

-W

The index that SDY is based on only goes back about 10 years.  There's 40 years of data in my link above.

Ah, oops.  That's what I get for using Dodge's data.  Dodge used SDY, which is the "high-yield" DA index.  That link you provided has the longest back-series of the Dividend Aristocrats that I'm aware of. 

[Dodge]

Let's see a graph of a longer time period, then. How has the DA fund performed relative to a straight S&P one?

-W

You're asking the wrong question.  Since the very nature of these lists are based on past performance, by definition, any chart not based on live year-by-year data will display the Survivorship Bias.  By that I mean, don't show me a chart of today's list of the best Dividend Funds over the last 25 years, and overlay it on a chart against the market as a whole.  Show me how the list from 1950 is doing.  Then 1960...etc.  If the Dividend Aristocrats were first created in 1989, show me a chart of how well the companies on the 1989 list did in 1990.  Then show how well the list on 1990 did in 1991...etc.

That's not what we're seeing, and this could be why all the dividend funds I can find, have such poor live-data returns.  Here's the oldest one I can find:

"The iShares Dow Jones US Select Dividend ETF (DVY) is the oldest dividend-focused ETF and is the only one to follow a pure high-yield strategy."



VTSAX - 98.74%
DVY - 48.4%

The fact that we can't find any older funds is a big red flag (again, alarm bells should be ringing).  It's possible they were liquidated for having even worse returns.  This is why it's important to find studies independent of the companies which profit from the sale of these type of funds.

[hodedofome]

You think academics are unbiased? All research is biased, which is why you have to do your own.

[skyrefuge]

But if you look specifically at aristocrats, you'll see that even during tough times like 2007-2008, they continued to maintain or increase their dividend payments.

LOLWUT?

During those tough times, 6 of the 52 companies in the list cut their dividends by over 90%. Five years later, their payouts are still a small fraction of what they used to be, and their share prices remain decimated (but who cares about share prices when income is all that matters, amirite!?)

Yearly dividend payouts before, after, and now:

Bank of America: $2.46 to $0.04 (and still at $0.04)
Fifth Third Bancorp: $1.72 to $0.04 (now up to $0.48)
Comerica: $2.52 to $0.20 (now up to $0.75)
Keycorp: $1.46 to $0.04 (now up to $0.22)
Progressive: $3.20 to $0.00 (now up to ~$0.50)
Regions Financial: $1.50 to $0.04 (now up to $0.12)

I generally think that a dividend-focused approach is sub-optimal and indicates an incomplete understanding about how stocks work, but is only marginally harmful. But, yikes, seeing a dividend-focused investor with an ignorance of the basic facts of recent history is downright scary!

[monkeymind]

This graph is particularly interesting:

I’m not investment savvy enough to have a dog in this fight.  I’m just trying to understand…

In order to explore it, I thought I could extrapolate from the Vanguard “potential impact of a dividend vs. a stock sale” information.  I created the attached excel spreadsheet to try to follow the idea presented over time (bear with me, I have fair but not excellent excel skills).

First, I maintained the givens from the Vanguard table ($3/share dividend or sale of 100 shares; share value doubling over time) and extended those over 3 additional time periods.  Just kind of putting those parameters on repeat... 
I assumed that it’s because the position was shrinking at 100 shares at a clip for the total return investor, the results showed that in the second period, results were even and, by the third period, dividend investor pulled ahead.  “Oh”, I thought, “this looks like (all other things being equal), over the long run, the dividend option seems to be the better choice.”

Of course, looking at the table, I then realized that our dividend investor was getting only $3000 in cash flow every year (less tax) while the total returns investor was receiving ever increasing $$ in cash flow (less tax on capital gain).  That’s not really equivalent, is it?  Surely, we need to compare apples to apples.   

So, I tried again.  This time, I figured it’s appropriate that each investor should receive equal amounts of money per year ($3000).  With this scenario, the total returns investor would need to sell ever smaller numbers of shares in order to receive that $3000 per year.   Maybe this would show that total return investing is, in fact, superior.
   
But it didn't.  While this time, the total returns investor remained ahead for the second iteration, he was again behind by the end of third period.

Please, all of you who know your shit, tell me what I’m missing or what mistake I made to make it look like, theoretically, the dividend investment would win no matter how you look at it.   

Of course, this is a theoretical and simplified test-tube version of dividend investing and doesn't involve real-world market returns.  That alone could make all the difference.  And, according to the recent graphs that have been posted, this would seem to be the case.  Maybe theoretical tables such as this one support the opinions of the dividend crowd but the real world market graphs favor the total return group?

[Chris86]

But if you look specifically at aristocrats, you'll see that even during tough times like 2007-2008, they continued to maintain or increase their dividend payments.

LOLWUT?

During those tough times, 6 of the 52 companies in the list cut their dividends by over 90%. Five years later, their payouts are still a small fraction of what they used to be, and their share prices remain decimated (but who cares about share prices when income is all that matters, amirite!?)

Yearly dividend payouts before, after, and now:

Bank of America: $2.46 to $0.04 (and still at $0.04)
Fifth Third Bancorp: $1.72 to $0.04 (now up to $0.48)
Comerica: $2.52 to $0.20 (now up to $0.75)
Keycorp: $1.46 to $0.04 (now up to $0.22)
Progressive: $3.20 to $0.00 (now up to ~$0.50)
Regions Financial: $1.50 to $0.04 (now up to $0.12)

I generally think that a dividend-focused approach is sub-optimal and indicates an incomplete understanding about how stocks work, but is only marginally harmful. But, yikes, seeing a dividend-focused investor with an ignorance of the basic facts of recent history is downright scary!

You nailed it! 11% of those listed failed to maintain their dividend. What about the other 89%? As a comparison, in 2008 you could have held a fund that mirrored the Dow/Nasdaq which would have resulted in dropping 18% of your portfolio total value...but those are just basic facts of recent history, are they not?

[Chris86]

This is interesting. Unless you have infinity shares, you're eventually going to run out when you sell them.

That is not actually true. Build a spreadsheet where your share price increases 5% per year and you sell 4% of the value of your initial portfolio every year. You will never run out.

It depletes your shares but at a slower and slower rate such that you never run out of shares.

It is a similar concept as that of "escape velocity" where an object is slowed down by gravity but at a decreasing rate so that it can outrun the gravity of a planet.

I would love to build a spreadsheet where my share price increases 5% every year. Perhaps you can give an example? That might be a conservative average to use over a long period of time, but that's not indicative of year to year - which is how often you will be selling, at a minimum, to get income. A 10% drop isn't remedied by a followed increase of 10%. When you eventually have to sell on the dips it will only act to accelerate your depletion.

[skyrefuge]

Please, all of you who know your shit, tell me what I’m missing or what mistake I made to make it look like, theoretically, the dividend investment would win no matter how you look at it.

You're losing track of the money that you save on taxes in the total-return option. In that first year, the total-return option ends up with $225 more in total ending value, but then that vanishes from your subsequent calculations. You need to either "reinvest" that $225, or, in a more-accurate real-world scenario, sell a number of shares so that the after-tax income equals the dividend after-tax income.

The Vanguard example is set up to show the effects of taxes. If taxes were not a factor, the outcomes would be exactly equal for both cases. The only thing that makes the total-return option better (and makes it better over the long-term as well) is the tax savings.

Good job on working through it yourself...I was pleasantly surprised to see that you remembered to subtract the dividends from the share price!

The yearly price-doubling is rather unrealistic though, and its effect will be to diminish the advantage of the total-return option as the years go on, since the captial-gain amount that gets taxed quickly nears the dividend-income amount.

[waltworks]

So wait, when we look at the SA comparison of S&P vs Dividend Aristocrats, I was under the impression that it's doing the following with the DA fund: in 1989, you have 52 (or whatever) companies. You buy $1000 worth of the fund. The next year, a company drops off the list maybe so you sell that stock, and you buy some shares of a new company that just qualified (if any), and you reinvest any dividends you got. You do that every year, and that's the return number we're looking at.

That sort of sounds like a terrible strategy (sell low?) but it would at least be a consistent way to compare vs. the S&P.

If the chart is actually a chart of how the ~50 companies currently on the list performed over the last 25 years then yes, it's obviously ridiculous. I could make any number of Walt's Awesome Company Funds using this sort of "strategy" and just pick the best stocks from the last 25 years. Why limit yourself to dividend payers? I'm sure some set of companies that started from nothing in 1989 and now are worth millions/billions could produce 100%+ annual returns on paper.

So what are we looking at on the Seeking Alpha chart? Can anyone tell me?

-Walt

[skyrefuge]

You nailed it! 11% of those listed failed to maintain their dividend. What about the other 89%? As a comparison, in 2008 you could have held a fund that mirrored the Dow/Nasdaq which would have resulted in dropping 18% of your portfolio total value...but those are just basic facts of recent history, are they not?

I wasn't making a comparison between different approaches. I was simply pointing out the inaccuracy of your statement, in the hopes that naive readers would not take it as truth. I'll repeat what you wrote:

"if you look specifically at aristocrats, you'll see that even during tough times like 2007-2008, they continued to maintain or increase their dividend payments."

Do you maintain that is a factual statement?

[Eric]

This is interesting. Unless you have infinity shares, you're eventually going to run out when you sell them.

That is not actually true. Build a spreadsheet where your share price increases 5% per year and you sell 4% of the value of your initial portfolio every year. You will never run out.

It depletes your shares but at a slower and slower rate such that you never run out of shares.

It is a similar concept as that of "escape velocity" where an object is slowed down by gravity but at a decreasing rate so that it can outrun the gravity of a planet.

I would love to build a spreadsheet where my share price increases 5% every year. Perhaps you can give an example? That might be a conservative average to use over a long period of time, but that's not indicative of year to year - which is how often you will be selling, at a minimum, to get income. A 10% drop isn't remedied by a followed increase of 10%. When you eventually have to sell on the dips it will only act to accelerate your depletion.

And?  I think I'm lost on your point.  It matters not if you sell shares every year, if overall, the value of your shares increases.  Which it should over a long time period.  If this was not the case, there would be no such thing as a SWR higher than your dividend payments.

It's similar to saying which weighs more: a pound of feathers or a pound of gold?

[Chris86]

If the growth in the price of the stock is faster than your withdrawal rate, you can sell shares below that rate forever and you're fine. It's easy to prove that to yourself with a spreadsheet.

I guess you could come up with some scenario where you end up with just one share worth a trillion dollars and then have to sell it or go hungry but in real life that's not going to happen.

The point is that if your profits are generated via cash payments (dividends) your tax hit is mandatory/bigger (in the US) and you make less money on your investment, assuming you want a steady income stream from your stock holdings.

Great article from Vanguard, as usual.

-W

@Dodge, but using that example, you have less shares each year? I mean, if the share grew from $15 to $30, shouldn't you want to only about half or $7 for income use while reinvesting the other half? I'm failing to see how selling shares in this way is smart since other time you'll keep selling and end up with zero shares? I know I'm incorrect but I don't see how selling shares this way can be sustainable? I just need to wrap my head around selling shares doesn't somehow deplete your shares? If it goes up $X, and you sell Y shares to get Z money, when in this process do you get around to buying more shares? Because I'd feel a lot more comfortable owning 10000 shares worth $10 than 1 share of $100000

This is interesting. Unless you have infinity shares, you're eventually going to run out when you sell them. "If the growth price of the stock..." Well, honestly that's just another variable you're never going to be able to accurately determine.

The article seems biased. The first thing that came to mind when reading this was the fact that companies that don't pay dividends are usually considered more risky (which in turn investment firms, IE Vanguard, asks for a higher fee in equities favored in the article, as opposed to something large cap / dividend payers).

No one seemed to mention that selling (and likely buying) also comes with transaction fees, where dividends do not.
Say in that same example, you are getting $100. Whether from selling existing holdings or dividend income.

Dividend would pay 15% so you get $85. Whats the standard fee on a trade? $7-8? Plus capital gains?

Are you recommending purchasing individual stocks?  That's the only way you'd pay a fee on a sale, or worry about having "infinity shares".  Vanguard funds have 0 transaction fee, and you can buy/sell in fractional shares.

Just going to leave this here... If you think the only fees that you're paying for your mutual fund are in the expense ratio, you're wrong.

http://online.wsj.com/news/articles/SB10001424052748703382904575059690954870722

[vivophoenix]

if you were to reinvest your dividends long term, would this pretty much be the equivalent of having held normal stock?

 esp, from my understanding, during times of down turn; they merely trim the dividend. so you dont lose value, you just lose the opportunity for growth that other stocks provide when the market goes back up.


 it seems like if i understand what im reading, the dividend yielding stocks  don't drop price during low markets.

so  it would be more advantageous to purchase other stocks while the market is low.

[Chris86]

This is interesting. Unless you have infinity shares, you're eventually going to run out when you sell them.

That is not actually true. Build a spreadsheet where your share price increases 5% per year and you sell 4% of the value of your initial portfolio every year. You will never run out.

It depletes your shares but at a slower and slower rate such that you never run out of shares.

It is a similar concept as that of "escape velocity" where an object is slowed down by gravity but at a decreasing rate so that it can outrun the gravity of a planet.

I would love to build a spreadsheet where my share price increases 5% every year. Perhaps you can give an example? That might be a conservative average to use over a long period of time, but that's not indicative of year to year - which is how often you will be selling, at a minimum, to get income. A 10% drop isn't remedied by a followed increase of 10%. When you eventually have to sell on the dips it will only act to accelerate your depletion.

And?  I think I'm lost on your point.  It matters not if you sell shares every year, if overall, the value of your shares increases.  Which it should over a long time period.  If this was not the case, there would be no such thing as a SWR higher than your dividend payments.

It's similar to saying which weighs more: a pound of feathers or a pound of gold?

Here, I'll clarify with an example. Sorry for the confusion: If you have an asset worth $100 and you need $5 to live, you're going to sell 5% for that year. However, if you have an asset worth $100 and then it drops by 2%, still needing that $5 to live, you now need to sell 5.1%. Your biggest assumption (that the value of your shares is going to increase YoY) is the most riskiest.

[skyrefuge]

Your biggest assumption (that the value of your shares is going to increase YoY) is the most riskiest.

I don't think anyone was really making that assumption. We all realize that the stock market is volatile and there will be down years mixed with up years. And we have things like the Trinity Study (and its relatives) to quantify for us, exactly how detrimental are the effects of volatility on Safe Withdrawal Rates? It showed that historically, if we sold 4% or less of our portfolio per year, we would not run out of money in a 30-year period. That 4% number is substantially lower than the long-term return of the US stock market, and one major reason for that is exactly due to the volatility that you fear. It's already factored in.

Currently the yield of SDY is 2.28%. That's a much lower payout rate than the 4% withdrawal rate the Trinity Study suggests is safe. That implies one of two things.

1) You will end up selling shares for income just like the rest of us (and do just fine too!)
2) You will be accumulating a much larger portfolio than necessary (and thus, working longer) in order to fund your retirement with dividend income alone.

[Dodge]

This graph is particularly interesting:

I’m not investment savvy enough to have a dog in this fight.  I’m just trying to understand…

In order to explore it, I thought I could extrapolate from the Vanguard “potential impact of a dividend vs. a stock sale” information.  I created the attached excel spreadsheet to try to follow the idea presented over time (bear with me, I have fair but not excellent excel skills).

First, I maintained the givens from the Vanguard table ($3/share dividend or sale of 100 shares; share value doubling over time) and extended those over 3 additional time periods.  Just kind of putting those parameters on repeat... 
I assumed that it’s because the position was shrinking at 100 shares at a clip for the total return investor, the results showed that in the second period, results were even and, by the third period, dividend investor pulled ahead.  “Oh”, I thought, “this looks like (all other things being equal), over the long run, the dividend option seems to be the better choice.”

Of course, looking at the table, I then realized that our dividend investor was getting only $3000 in cash flow every year (less tax) while the total returns investor was receiving ever increasing $$ in cash flow (less tax on capital gain).  That’s not really equivalent, is it?  Surely, we need to compare apples to apples.   

So, I tried again.  This time, I figured it’s appropriate that each investor should receive equal amounts of money per year ($3000).  With this scenario, the total returns investor would need to sell ever smaller numbers of shares in order to receive that $3000 per year.   Maybe this would show that total return investing is, in fact, superior.
   
But it didn't.  While this time, the total returns investor remained ahead for the second iteration, he was again behind by the end of third period.

Please, all of you who know your shit, tell me what I’m missing or what mistake I made to make it look like, theoretically, the dividend investment would win no matter how you look at it.   

Of course, this is a theoretical and simplified test-tube version of dividend investing and doesn't involve real-world market returns.  That alone could make all the difference.  And, according to the recent graphs that have been posted, this would seem to be the case.  Maybe theoretical tables such as this one support the opinions of the dividend crowd but the real world market graphs favor the total return group?

After period 2, you stopped reducing the stock price by the amount of the dividend.  From cell C16 to E15, you did it.  But from cell E16 to G15 you did not.

[Dodge]

If the growth in the price of the stock is faster than your withdrawal rate, you can sell shares below that rate forever and you're fine. It's easy to prove that to yourself with a spreadsheet.

I guess you could come up with some scenario where you end up with just one share worth a trillion dollars and then have to sell it or go hungry but in real life that's not going to happen.

The point is that if your profits are generated via cash payments (dividends) your tax hit is mandatory/bigger (in the US) and you make less money on your investment, assuming you want a steady income stream from your stock holdings.

Great article from Vanguard, as usual.

-W

@Dodge, but using that example, you have less shares each year? I mean, if the share grew from $15 to $30, shouldn't you want to only about half or $7 for income use while reinvesting the other half? I'm failing to see how selling shares in this way is smart since other time you'll keep selling and end up with zero shares? I know I'm incorrect but I don't see how selling shares this way can be sustainable? I just need to wrap my head around selling shares doesn't somehow deplete your shares? If it goes up $X, and you sell Y shares to get Z money, when in this process do you get around to buying more shares? Because I'd feel a lot more comfortable owning 10000 shares worth $10 than 1 share of $100000

This is interesting. Unless you have infinity shares, you're eventually going to run out when you sell them. "If the growth price of the stock..." Well, honestly that's just another variable you're never going to be able to accurately determine.

The article seems biased. The first thing that came to mind when reading this was the fact that companies that don't pay dividends are usually considered more risky (which in turn investment firms, IE Vanguard, asks for a higher fee in equities favored in the article, as opposed to something large cap / dividend payers).

No one seemed to mention that selling (and likely buying) also comes with transaction fees, where dividends do not.
Say in that same example, you are getting $100. Whether from selling existing holdings or dividend income.

Dividend would pay 15% so you get $85. Whats the standard fee on a trade? $7-8? Plus capital gains?

Are you recommending purchasing individual stocks?  That's the only way you'd pay a fee on a sale, or worry about having "infinity shares".  Vanguard funds have 0 transaction fee, and you can buy/sell in fractional shares.

Just going to leave this here... If you think the only fees that you're paying for your mutual fund are in the expense ratio, you're wrong.

http://online.wsj.com/news/articles/SB10001424052748703382904575059690954870722

This information strengthens the case for indexing, as an index would incur the smallest amount of these fees.

[tomsang]

This reminds me of those investors that get all excited when their stock splits.  They don't get that it is still the same stock just divided into smaller parts. 

There is nothing magical or special about companies that have a dividend.  If anything the company is sending a message that they have no better uses for their capital than to give it back to the investors to figure out what to do with the excess capital.

If there was an efficiency in the market, somebody with a supercomputer would have figured that out and capitalized on it.  There is no free money in the stock market.

If it makes you sleep better at night, then so be it.  If you are receiving dividends that are higher than the market in total then you are taking on greater risk than the market in total.  Again, no free money.  Don't allow your guard to drop because of a dividend.  Dividends are not safer, they don't return a higher risk weighted yield, they don't stop aging, they don't do anything special.     

[Dodge]

So wait, when we look at the SA comparison of S&P vs Dividend Aristocrats, I was under the impression that it's doing the following with the DA fund: in 1989, you have 52 (or whatever) companies. You buy $1000 worth of the fund. The next year, a company drops off the list maybe so you sell that stock, and you buy some shares of a new company that just qualified (if any), and you reinvest any dividends you got. You do that every year, and that's the return number we're looking at.

That sort of sounds like a terrible strategy (sell low?) but it would at least be a consistent way to compare vs. the S&P.

If the chart is actually a chart of how the ~50 companies currently on the list performed over the last 25 years then yes, it's obviously ridiculous. I could make any number of Walt's Awesome Company Funds using this sort of "strategy" and just pick the best stocks from the last 25 years. Why limit yourself to dividend payers? I'm sure some set of companies that started from nothing in 1989 and now are worth millions/billions could produce 100%+ annual returns on paper.

So what are we looking at on the Seeking Alpha chart? Can anyone tell me?

-Walt

The Seeking Alpha chart seems genuine.  A few percentage points off from what MorningStar reports.  After doing some more research, I found why the Morningstar charts showed such poor returns.  The charts I've been posting which usually include dividends reinvested (growth charts) have actually been "price only" charts.  If you start with an ETF you get a price chart. If you start with a mutual fund it gives the total return (dividends reinvested), which you can then add an ETF to, without it changing back to a price chart.

Here's the real SDY vs VTSAX chart:



VTSAX still wins, but it's much closer.

[waltworks]

Well, just glancing at it, it looks like Seeking Alpha has the DA outperforming the S&P (and hence VTSAX since they track pretty closely) by a decent amount. So something is wrong somewhere.

-W

[beltim]

Well, just glancing at it, it looks like Seeking Alpha has the DA outperforming the S&P (and hence VTSAX since they track pretty closely) by a decent amount. So something is wrong somewhere.

-W

They're looking at different stocks and different time periods.

[waltworks]

Sure, but I thought SDY was pretty much identical. And in the time period of the Morningstar data SA says that Dividend Aristocrats was something like 8 percent returns, as opposed to something like 4.5% for S&P. The two data sets don't track at ALL, even in the same time period. What gives?

-W

Well, just glancing at it, it looks like Seeking Alpha has the DA outperforming the S&P (and hence VTSAX since they track pretty closely) by a decent amount. So something is wrong somewhere.

-W

They're looking at different stocks and different time periods.

[beltim]

Sure, but I thought SDY was pretty much identical. And in the time period of the Morningstar data SA says that Dividend Aristocrats was something like 8 percent returns, as opposed to something like 4.5% for S&P. The two data sets don't track at ALL, even in the same time period. What gives?

-W

Well, just glancing at it, it looks like Seeking Alpha has the DA outperforming the S&P (and hence VTSAX since they track pretty closely) by a decent amount. So something is wrong somewhere.

-W

They're looking at different stocks and different time periods.

No, SDY is some high yield subset of the dividend aristocrats that apparently doesn't track very well.

[waltworks]

https://www.spdrs.com/product/fund.seam?ticker=sdy

If it doesn't track well, they are doing a pretty crap job with it. Again, something is not making sense here.

-W

[hodedofome]

Another shareholder yield article I just read tonight http://jimoshaughnessy.tumblr.com/post/94453059749/the-power-of-shareholder-yield

[beltim]

https://www.spdrs.com/product/fund.seam?ticker=sdy

If it doesn't track well, they are doing a pretty crap job with it. Again, something is not making sense here.

-W

Right, there's a difference between the "high yield dividend aristocrats" and "dividend aristocrats" indices

[divinvestor]

Hopefully this will settle the debate regarding the question of dividend-paying stocks being superior to non-dividend paying stocks. Here's an excerpt on pg. 126 from "The Future For Investors" by Jeremy Siegel. Keep in mind this book was written in 2005:

"From 1871 through 2003, 97 percent of the total after-inflation accumulation from stocks comes from reinvesting dividends. Only 3 percent comes from capital gains... The sum of $1,000 invested in stocks in 1871 would have accumulated almost $8 million after inflation by the end of 2003. Without reinvesting dividends, the accumulation would be less than $250,000."

[Dodge]

https://www.spdrs.com/product/fund.seam?ticker=sdy

If it doesn't track well, they are doing a pretty crap job with it. Again, something is not making sense here.

-W

Whenever someone claims to be able to beat the market, then publishes the method, you are wise to be skeptical.  The Dividend Aristocrats:

http://us.spindices.com/indices/strategy/sp-500-dividend-aristocrats

Hold about 53 stocks which "measure the performance S&P 500 companies that have increased dividends every year for the last 25 consecutive years. The Index treats each constituent as a distinct investment opportunity without regard to its size by equally weighting each company."

There don't seem to be any ETFs that track this right now, besides the newly created NOBL.  Hopefully, over the next few decades or so (if the fund isn't closed by then), this ETF will show a live example of the returns you can expect by following the Dividend Aristocrats.

The High Yield Dividend Aristocrats:

http://us.spindices.com/indices/strategy/sp-high-yield-dividend-aristocrats-index

Hold about 96 stocks which are "designed to measure the performance of companies within the S&P Composite 1500® that have followed a managed-dividends policy of consistently increasing dividends every year for at least 20 years."  This is the higher yielding and more diversified cousin of the Dividend Aristocrats.  This is what the SDY tracks.  All information presented, shows the SDY underperforms a total stock market index (VTSAX), and has around 7x higher fees.

For the new traders reading this, some questions you should consider, in addition to all the alarm bells which should already be going off while reading this thread:

1.  If the Dividend Aristocrats represent a free lunch, and are indeed beating the market, where's the live evidence?  Certainly a fund manager somewhere would be willing to charge a 2-3% fee to offer their clients this free lunch?  Is it possible these funds all underperformed, and were liquidated? (Survivorship Bias)  Does it surprise you that there are literally 0 funds/ETFs tracking the Dividend Aristocrats (or any other dividend strategy that I can find), which beats the simple, cheap market index VTSAX?

2.  If this really is a free lunch, why hasn't the market already bought up all the stock, bidding up the price until the companies (still good companies) stock is so expensive, that any dividend gain is lost?  Is it your belief that while the market is aware of this free money opportunity (it is published, after all), the entire market chooses to not take advantage?

3.  This is investing money solely in the stocks which have performed well over the past 25 years, hoping they will continue to perform well in the next 25 years.  Does that sound like a good idea to you?  If so, you're might be much newer to investing than you think, and would do well do to some additional reading

4.  Does it surprise you that the only supporting evidence shown thus far in support of Dividend Aristocrats, were sourced from companies which profit from their higher fees, and have mangers which they claim can use dividends to "beat the market"?

Remember, all the evidence shows that over the next 50 years, the chances are 99%+ that your index portfolio will beat this, and with significantly less work and risk.  Don't be tempted, your life savings, and your ability to retire (or stay retired) is not worth it.

"Of the 355 equity funds in 1970, fully 233 of those funds have gone out of business. Only 24 oupaced the market by more than 1% a year. These are terrible odds." Jack Bogle (2007)

[Dodge]

Hopefully this will settle the debate regarding the question of dividend-paying stocks being superior to non-dividend paying stocks. Here's an excerpt on pg. 126 from "The Future For Investors" by Jeremy Siegel. Keep in mind this book was written in 2005:

"From 1871 through 2003, 97 percent of the total after-inflation accumulation from stocks comes from reinvesting dividends. Only 3 percent comes from capital gains... The sum of $1,000 invested in stocks in 1871 would have accumulated almost $8 million after inflation by the end of 2003. Without reinvesting dividends, the accumulation would be less than $250,000."

This is not relevant to the discussion.  That's an argument for reinvesting dividends vs. not reinvesting dividends.

[divinvestor]

Actually I think it's extremely relevant because it shows the importance of dividends in terms of total return over a long period of time. And from what I've read, you have yet to provide any facts and figures to illustrate stocks that don't pay dividends are superior to dividend stocks.

[Dodge]

Actually I think it's extremely relevant because it shows the importance of dividends in terms of total return over a long period of time. And from what I've read, you have yet to provide any facts and figures to illustrate stocks that don't pay dividends are superior to dividend stocks.

No one has made that claim.  I recommend reading back through the thread.

[pom]

I agree with Dodge that you have to be careful with survivorship bias in many financial studies.

If you can predict which companies will  increase dividends in the next 25 years and invest in them now, then you will make a killing. Unfortunately it is not possible and it is unlikely to be the same companies that increased dividends in the last 25 years.

Admitedly there will probably be some correlation. Lets say for the sake of argument that 75% of the companies are the same; I will bet that the 25% fallen angels will have such poor return that overall you will gain little if anything at all.

You have to be very careful when backtesting not to include any future information in the choices that you make at time zero. It sounds so logical but in reality it is not always easy to do.

[Dodge]

I agree with Dodge that you have to be careful with survivorship bias in many financial studies.

If you can predict which companies will  increase dividends in the next 25 years and invest in them now, then you will make a killing. Unfortunately it is not possible and it is unlikely to be the same companies that increased dividends in the last 25 years.

Admitedly there will probably be some correlation. Lets say for the sake of argument that 75% of the companies are the same; I will bet that the 25% fallen angels will have such poor return that overall you will gain little if anything at all.

You have to be very careful when backtesting not to include any future information in the choices that you make at time zero. It sounds so logical but in reality it is not always easy to do.

Indeed. In reality it's much worse than 75%. Of the original Dividend Aristocrats, only 7 still remain in the index.  Looking back, only 30% of the companies in the index at any one time, are still there after 10 years.

[beltim]

I agree with Dodge that you have to be careful with survivorship bias in many financial studies.

If you can predict which companies will  increase dividends in the next 25 years and invest in them now, then you will make a killing. Unfortunately it is not possible and it is unlikely to be the same companies that increased dividends in the last 25 years.

Admitedly there will probably be some correlation. Lets say for the sake of argument that 75% of the companies are the same; I will bet that the 25% fallen angels will have such poor return that overall you will gain little if anything at all.

You have to be very careful when backtesting not to include any future information in the choices that you make at time zero. It sounds so logical but in reality it is not always easy to do.

Indeed. In reality it's much worse than 75%. Of the original Dividend Aristocrats, only 7 still remain in the index.  Looking back, only 30% of the companies in the index at any one time, are still there after 10 years.

You keep arguing against a point that no one is making. Earlier I showed hard data that over long periods of time, stocks that pay dividends outperform stocks that don't.  And I showed data on why companies paying dividends do better than companies that by back stock (management buys back stock at inopportune times and prices).  No one has refuted that point, except by claiming bias on the part of the study authors, which I countered by providing other studies showing the same thing.  Your continued claims about some subset of dividend paying stocks are irrelevant to the larger point that stocks that pay dividends have higher returns.

[Roland of Gilead]

If dividend stocks consistently had gains above non-dividend stocks after accounting for dividend payments and growth, there would be a fantastic opportunity to be long dividend and short non-dividend, reaping all of the gains with none of the risk.

[tomsang]

Interesting article that is on point.

http://money.usnews.com/money/blogs/the-smarter-mutual-fund-investor/2014/02/04/7-myths-about-dividend-paying-stocks

[beltim]

If dividend stocks consistently had gains above non-dividend stocks after accounting for dividend payments and growth, there would be a fantastic opportunity to be long dividend and short non-dividend, reaping all of the gains with none of the risk.

No. Such a strategy would have very high risks due to the possibility of margin calls.  Even if you know the starting and ending points exactly, you can still lose.

[Dodge]

I agree with Dodge that you have to be careful with survivorship bias in many financial studies.

If you can predict which companies will  increase dividends in the next 25 years and invest in them now, then you will make a killing. Unfortunately it is not possible and it is unlikely to be the same companies that increased dividends in the last 25 years.

Admitedly there will probably be some correlation. Lets say for the sake of argument that 75% of the companies are the same; I will bet that the 25% fallen angels will have such poor return that overall you will gain little if anything at all.

You have to be very careful when backtesting not to include any future information in the choices that you make at time zero. It sounds so logical but in reality it is not always easy to do.

Indeed. In reality it's much worse than 75%. Of the original Dividend Aristocrats, only 7 still remain in the index.  Looking back, only 30% of the companies in the index at any one time, are still there after 10 years.

You keep arguing against a point that no one is making. Earlier I showed hard data that over long periods of time, stocks that pay dividends outperform stocks that don't.  And I showed data on why companies paying dividends do better than companies that by back stock (management buys back stock at inopportune times and prices).  No one has refuted that point, except by claiming bias on the part of the study authors, which I countered by providing other studies showing the same thing.  Your continued claims about some subset of dividend paying stocks are irrelevant to the larger point that stocks that pay dividends have higher returns.

This entire thread seems predicated on "Only Buying Dividend Stocks".

Is it your assertion that no one in this thread has promoted purchasing "carefully screened individual high-yielding dividend stocks", or the importance of finding stocks which can "keep increasing dividends in the future"?

If not, is it your assertion that information on Survivorship Bias, and how often stocks which continuously increased their dividends for 25 years or more, suddenly stopped, is not relevant information for new investors possibly considering such a strategy?

If not, then sure.  I have not explicitly addressed your point on buybacks vs dividends.  Are you suggesting investors on this forum should only purchase stocks which pay dividends, and no stocks which participate in buybacks?  Is it your assertion that these are the only two options, and that these options are mutually exclusive?

Regarding the "data that over long periods of time, stocks that pay dividends outperform stocks that don't", it smells like Survivorship Bias.

http://www.bogleheads.org/wiki/Survivorship_bias

It simply shows, "This group of stocks which have exhibited returns every year over X years, have higher returns than the market as a whole."

That sounds like a reasonable statement to make.  How is that information actionable?  Shall I then invest money in the stocks which have exhibited returns every year over X years, hoping they will continue to perform well in the next X years?  Alarm bells should start ringing on that one.

Maybe the real question is, how can you invest in stocks which have exhibited returns every year over X years, before they join that group?