"homemade dividends" from Total Stock Market fund

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
VaR
Posts: 572
Joined: Sat Dec 05, 2015 11:27 pm

Re: "homemade dividends" from Total Stock Market fund

Post by VaR » Mon Jul 02, 2018 2:52 am

snarlyjack wrote:
Mon Jul 02, 2018 1:21 am
The conversation started when Carol8888 (OP) was asking
about selling off the portfolio vs. using dividends.
Working on a total return strategy is not equivalent to "selling off the portfolio".
I realize most people here use the 4% rule & fully intend to
sell off their portfolios in retirement. I have no problem
with that.
I don't think most people here "use" the 4% rule. The "4% rule" is otherwise stated as the 20x expenses rule and says that you need at least 25x retirement expenses in your portfolio in order to be assured of a low probability of failure of your retirement portfolio. This rule of thumb applies regardless of what your strategy is. It even applies to any of the variety of dividend strategies.

What dividend strategy are you a proponent of? Is it the 100% high dividend equity strategy? Specifically, are you proposing 100% in the Vanguard High Dividend Yield Index Fund with current SEC yield of 2.985? Or alternatively in the active 100% equity Vanguard Equity Income Fund with current SEC yield of 2.64%?
I was trying to be opened minded and tell him their are other
ways & show him some articles from CPA's that have run the
numbers & studied it.
I don't how knowledge of accounting standards qualifies one to recommend a retirement strategy. In particular, I think that the fact that the author of the article that you quoted links sells a newsletter subscription where he provides the specifics of this dividend strategy he is promoting raises red flags to me about whether he is providing objective advice or whether he is selling his newsletter.
My personal opinion is (as I stated up stream) is it depends on
your portfolio size. If you have a huge portfolio you can
use whatever method you want. People with smaller portfolios
have more limited choices. I do not know the size of Carol's
portfolio or his needs or wants (expenses). The point I was
trying to make is everyone is different. My situation is totality
different than your situation. It's not about right or wrong it's
about your personal situation. That's why they call it personal finance.
Portfolio size isn't really the critical factor in determining acceptable retirement income strategies. For the overwhelming majority of people it's the ratio of their portfolio size to their anticipated retirement expenses in relation to their current age and joint life expectancy. When your portfolio exceeds 50x expenses, every strategy starts to have a vanishingly small probability of failure.
Every one is different. Their is no black or white. It's about you
& your situation.
Every individual is different, but the optimal strategies that we discuss here are generally applicable and can be tailored to specific situations. I assume you're not suggesting that everyone needs to know about whole life and variable annuities just to know all the possible 99.99% bad choices they could make, right?

Are you sure you're not just trying to gaslight people coming to this board for sound advice? You've certainly succeeded in derailing this thread.

OP: The conventional wisdom of this board says to invest in the 3-fund portfolio with an asset allocation that is appropriate to your risk tolerance. If you want to be conservative (which people on this board have increasingly become with shrinking dividend yields and interest rates), you would pick a withdrawal rate that is no higher than 3.5% and if you want to be even more conservative also no greater than the income distribution of your portfolio. This latter constraint currently sits as about a 2% yield. Everyone here will agree that 2% is ultra-conservative and has no chance of failure.

IlliniDave
Posts: 2283
Joined: Fri May 17, 2013 7:09 am

Re: "homemade dividends" from Total Stock Market fund

Post by IlliniDave » Mon Jul 02, 2018 6:10 am

Phineas J. Whoopee wrote:
Sun Jul 01, 2018 11:00 pm
IlliniDave wrote:
Sun Jul 01, 2018 4:48 pm
...
So it seems loosely analogous to buying a stock immediately prior to the ex-dividend date, assuming a market estimate of the value of the upcoming dividend is implicitly priced in as the time after the last ex-dividend date progresses.
Do you want to invest based on analogies or on facts?
...
PJW
You misunderstand. I won't make any investment decisions based on this conversation. I'm just thinking out loud, as it were, about how a dividend gets priced into a stock. In these type of threads there is always reliance on the mechanism of how dividends get priced out of a stock, I was curious about the flip-side.
Don't do something. Just stand there!

david1082b
Posts: 368
Joined: Fri Jun 09, 2017 12:35 pm

Re: "homemade dividends" from Total Stock Market fund

Post by david1082b » Mon Jul 02, 2018 6:11 am

snarlyjack wrote:
Mon Jul 02, 2018 12:58 am
MT,

I realize that, but...

What companies does Warren Buffett invest in?
(dividend paying stocks).

Also, the reason WB does not pay a dividend out
is because he thinks he is a better investor than most people.

Getting a dividend & paying out a dividend are 2 different strategies.
Warren Buffett is the greatest...he knows what he is doing.
Apple's dividend yield is under 2% right now, not very high. Even a 2% to 3% withdrawal strategy would require selling off some each year, which would be an OK strategy. Retirees can use a combination of dividends and sales to create a withdrawal rate that works for them. The portfolio is not sold off just because of a withdrawal strategy not 100% based on dividend income. A dividend is simply an involuntary sale in itself, since the stock value declines to match the dividend payout.

Berkshire Hathaway has a plan for stock buybacks if the price to book falls low enough. Using BRK in a portfolio would mean selling some off each year in retirement, something Buffett has said is an OK strategy. There is no need for a dividend and never will be. With stock buybacks there can be return of capital without paying out a dividend. Alphabet can also do buybacks and sometimes has done so. Again no need for dividends. Total return is the important issue for withdrawal rates.

Most of BRK's assets are in wholly-owned subsidiaries, not dividend paying stocks. The profits from the subsidiaries can be used for buybacks if necessary. Apple uses a lot of its profits for stock buybacks too. Dividends are the only strategy Apple has. Apple could stop paying its dividend and do buybacks and everything would be perfectly fine for retirees using Apple or market index funds for retirement withdrawals.

22twain
Posts: 1444
Joined: Thu May 10, 2012 5:42 pm

Re: "homemade dividends" from Total Stock Market fund

Post by 22twain » Mon Jul 02, 2018 6:28 am

david1082b wrote:
Mon Jul 02, 2018 6:11 am
Dividends are the only strategy Apple has.
I suspect you inadvertently omitted a "not" here.
My investing princiPLEs do not include absolutely preserving princiPAL.

david1082b
Posts: 368
Joined: Fri Jun 09, 2017 12:35 pm

Re: "homemade dividends" from Total Stock Market fund

Post by david1082b » Mon Jul 02, 2018 6:33 am

Let's look at the dividend yields of major BRK stock holdings using Yahoo Finance:

Apple: 1.59%
Wells Fargo: 2.84%
Kraft-Heinz: 4.00%
BofA: 1.70%
CocaCola: 3.62%
AmEx: 1.45%
Phillips 66: 2.49%
US Bancorp: 2.36%
Moody's: 1.02%

Imagine if you were creating a retirement portfolio based on BRK's stock holdings, you would either need a withdrawal rate below 3% generally or would have to sell some stock off each year to match the withdrawal rate. People with very big portfolios might not need to spend more than the dividend, but those with more modest portfolios may need to spend more than the dividend payouts, so some sales would be in order. This is not very difficult to do.

With dividend payout ratios having declines a lot since the 1980s, clearly dividends have made up a smaller % of total return since stock buybacks were legalized.

IlliniDave
Posts: 2283
Joined: Fri May 17, 2013 7:09 am

Re: "homemade dividends" from Total Stock Market fund

Post by IlliniDave » Mon Jul 02, 2018 6:48 am

Artsdoctor wrote:
Sun Jul 01, 2018 8:04 pm

It seems the thread has veered since your post here, but I'll explain it another way so hopefully there will be no confusion.

Do not think of equity fund dividends and bond fund dividends in the same way. They are not.
Yeah, I get the mechanism with stocks, I'm just thinking through the other side of the process--what happens in between dividend dates for consistent dividend payers that everyone believes is highly likely to pay a dividend of a predictable size at a predictable interval. Intuition tells me there is an implicit process that is roughly analogous to what PJW mentioned above as the prorated coupon added on top of the value of the bond that must be paid to purchase it from another holder. In a solid, profitable enterprise cash is accruing to pay the dividend (yeah I know there is sometimes financial trickery wrt dividends). The accruing cash should raise the price of the stock in an efficient market by the same amount the price of the stock drops when it's paid I guess is what I'm getting to in a roundabout way.

"Analogous" to me does not mean "identical to", and intuition is maybe something I should just keep to myself,eh? :)
Don't do something. Just stand there!

User avatar
triceratop
Moderator
Posts: 5469
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: "homemade dividends" from Total Stock Market fund

Post by triceratop » Mon Jul 02, 2018 8:59 am

snarlyjack wrote:
Mon Jul 02, 2018 1:21 am

My personal opinion is (as I stated up stream) is it depends on
your portfolio size. If you have a huge portfolio you can
use whatever method you want. People with smaller portfolios
have more limited choices. I do not know the size of Carol's
portfolio or his needs or wants (expenses). The point I was
trying to make is everyone is different. My situation is totality
different than your situation. It's not about right or wrong it's
about your personal situation. That's why they call it personal finance.
Our situations in terms of age and tax brackets are more similar than you realize.

What is not similar is that I have taken a evidence-based rational approach to the dividend conversation and have found errors in the analysis you have posted and implicitly advocated. You have not addressed this which implies to me as a reader that you are not interested in an evidence-based discussion regarding this. If I am incorrect about this you would defend the points (show me how I’m wrong) or concede some error—until one of those happens I will sign off this thread as I feel there is limited productive use to continuing this conversation.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

TN_Boy
Posts: 352
Joined: Sat Jan 17, 2009 12:51 pm

Re: "homemade dividends" from Total Stock Market fund

Post by TN_Boy » Mon Jul 02, 2018 9:09 am

snarlyjack wrote:
Mon Jul 02, 2018 12:58 am
MT,

I realize that, but...

What companies does Warren Buffett invest in?
(dividend paying stocks).

Also, the reason WB does not pay a dividend out
is because he thinks he is a better investor than most people.

Getting a dividend & paying out a dividend are 2 different strategies.
Warren Buffett is the greatest...he knows what he is doing.
No, he doesn't pay out dividends because he believes the money can be better deployed to grow/run the business (I think I linked a article discussing his reasoning the last time we had one of these "dividends are better, listen to Warren Buffet" discussions).

Consider that other companies might make the same decision about dividends. But I am convinced that nothing anybody says would convince you that dividend investing is not superior. (And hey, I like dividends myself. Nothing easier. But they are not "better").

snarlyjack
Posts: 780
Joined: Fri Aug 28, 2015 12:44 pm
Location: Montana

Re: "homemade dividends" from Total Stock Market fund

Post by snarlyjack » Mon Jul 02, 2018 9:50 am

Good Morning Bogleheads.
(I' am on my way to work...).

T.

I' am glad our situations are similar. That makes things more relatable.

I think I have been very evidence based. For example I have shown you
assorted studies. If you read Jack Bogle's book "The Little Book of
Common Sense Investing, 10th anniversary edition", Chapter 6, Jack Bogle
talks about dividends. I have also shown you articles that Jack Bogle has
written about dividends.

I have shown you reports (white papers) written by PHD. level University
Papers & Studies. I have shown you reports by CPA's, I have shown you
reports by Brokerage Firms with analysis & graphs. I have shown you videos
by expert investors, for example Kevin O' Leary. The list goes on...

Imho, your not a believer in dividends. That's ok I don't care. But I do
think you should be more open minded. Their are thousand's of dividend
investors, managing $Millions of dollars. My impression is a lot of these
investors are well educated, very smart & have a strategy that they can
relate to. That's what matters. A strategy that they can relate to without
making emotional mistakes. I've shown you investors that have made
Millions of dollars with the strategy, for example Ronald Read, The Janitor
Next Door. There are to many examples like Ronald Read with Millions of
Dollars to blow it off & say their stupid.

Since our situations are similar, I would suggest you expand your knowledge base.
I do like you & I think your good at what you do. But open up & expand...

(gotta go to work...no time to play)… sj

User avatar
Phineas J. Whoopee
Posts: 7234
Joined: Sun Dec 18, 2011 6:18 pm

Re: "homemade dividends" from Total Stock Market fund

Post by Phineas J. Whoopee » Mon Jul 02, 2018 10:12 am

snarlyjack wrote:
Mon Jul 02, 2018 1:21 am
...
I realize most people here use the 4% rule & fully intend to
sell off their portfolios in retirement. I have no problem
with that.
...
There ain't no 4% rule for people to use, most, some, or none. It's a fiction designed to sell you Ford F-150s.

One reason you and we aren't connecting is you attribute to us opinions most of us don't hold.

PJW
Last edited by Phineas J. Whoopee on Mon Jul 02, 2018 10:30 am, edited 1 time in total.

User avatar
Phineas J. Whoopee
Posts: 7234
Joined: Sun Dec 18, 2011 6:18 pm

Re: "homemade dividends" from Total Stock Market fund

Post by Phineas J. Whoopee » Mon Jul 02, 2018 10:19 am

IlliniDave wrote:
Mon Jul 02, 2018 6:10 am
Phineas J. Whoopee wrote:
Sun Jul 01, 2018 11:00 pm
IlliniDave wrote:
Sun Jul 01, 2018 4:48 pm
...
So it seems loosely analogous to buying a stock immediately prior to the ex-dividend date, assuming a market estimate of the value of the upcoming dividend is implicitly priced in as the time after the last ex-dividend date progresses.
Do you want to invest based on analogies or on facts?
...
PJW
You misunderstand. I won't make any investment decisions based on this conversation. I'm just thinking out loud, as it were, about how a dividend gets priced into a stock. In these type of threads there is always reliance on the mechanism of how dividends get priced out of a stock, I was curious about the flip-side.
That's pretty weird, since you asked specifically about bonds.

PJW

User avatar
Oicuryy
Posts: 1235
Joined: Thu Feb 22, 2007 10:29 pm

Re: "homemade dividends" from Total Stock Market fund

Post by Oicuryy » Mon Jul 02, 2018 12:18 pm

Here is what the Bogleheads wiki says about the spend-only-the-dividends withdrawal method.
If you wish to keep your principal investment amount intact, you might consider using a method where you only spend the dividend and interest income from your investments.

The advantage is clear, but like the constant percentage method, it may result in fluctuating income amounts as dividend and interest income rates for your investments change. Also, while having a large bond percentage will increase your interest income with this method, having too few growth investments exposes you to the risk of not keeping up with inflation over the long term. You might prefer this method if your expenses are small in relation to your portfolio size, and if you wish to leave a large amount to your heirs.
https://www.bogleheads.org/wiki/Withdra ... _dividends

Ron
Money is fungible | Abbreviations and Acronyms

MIretired
Posts: 712
Joined: Fri Sep 06, 2013 12:35 pm

Re: "homemade dividends" from Total Stock Market fund

Post by MIretired » Mon Jul 02, 2018 12:18 pm

Phineas J. Whoopee wrote:
Mon Jul 02, 2018 10:19 am
IlliniDave wrote:
Mon Jul 02, 2018 6:10 am
Phineas J. Whoopee wrote:
Sun Jul 01, 2018 11:00 pm
IlliniDave wrote:
Sun Jul 01, 2018 4:48 pm
...
So it seems loosely analogous to buying a stock immediately prior to the ex-dividend date, assuming a market estimate of the value of the upcoming dividend is implicitly priced in as the time after the last ex-dividend date progresses.
Do you want to invest based on analogies or on facts?
...
PJW
You misunderstand. I won't make any investment decisions based on this conversation. I'm just thinking out loud, as it were, about how a dividend gets priced into a stock. In these type of threads there is always reliance on the mechanism of how dividends get priced out of a stock, I was curious about the flip-side.
That's pretty weird, since you asked specifically about bonds.

PJW
I think that PJW is right upthread: that the capital structure, so to speak, is completely different for bonds and coupons than for stock ownership and dividends.
eg: A bond coupon is under a contract to pay that coupon; else in default.
With stocks, you are a company owner. A dividend payment is a drawdown(term used in bookkeeping), or cash distribution of part of the assets which you are an owner of, again.
In a company financial statement and income statement done quarterly, the net (reported) earnings move over to the balance sheet (assets - liabilities). The dividend is then withdrawn from the assets(balance sheet.) This is not contractual. It is just a disbursement(drawdown) of company assets to you, an owner of the company.
I didn't find a wiki reference to company bookkeeping in a search.

And, as the time for a dividend payment approaches, there is no appreciation built into the stock price for the coming divdend. It is simply a one time withdrawal of capital to whomever owns the stock at the date. The dividend is not a gain at all. It is a transfer of cash to the owners of the company.

IlliniDave
Posts: 2283
Joined: Fri May 17, 2013 7:09 am

Re: "homemade dividends" from Total Stock Market fund

Post by IlliniDave » Mon Jul 02, 2018 12:29 pm

Phineas J. Whoopee wrote:
Mon Jul 02, 2018 10:19 am

That's pretty weird, since you asked specifically about bonds.

PJW
Not really, you posted about bonds in a thread about stock dividends and I asked you a question. I shall not do that again as you seem a little prickly about it and I don't like to aggravate people.
Don't do something. Just stand there!

IlliniDave
Posts: 2283
Joined: Fri May 17, 2013 7:09 am

Re: "homemade dividends" from Total Stock Market fund

Post by IlliniDave » Mon Jul 02, 2018 12:41 pm

MIretired wrote:
Mon Jul 02, 2018 12:18 pm

I think that PJW is right upthread: that the capital structure, so to speak, is completely different for bonds and coupons than for stock ownership and dividends.
eg: A bond coupon is under a contract to pay that coupon; else in default.
With stocks, you are a company owner. A dividend payment is a drawdown(term used in bookkeeping), or cash distribution of part of the assets which you are an owner of, again.
In a company financial statement and income statement done quarterly, the net (reported) earnings move over to the balance sheet (assets - liabilities). The dividend is then withdrawn from the assets(balance sheet.) This is not contractual. It is just a disbursement(drawdown) of company assets to you, an owner of the company.
I didn't find a wiki reference to company bookkeeping in a search.

And, as the time for a dividend payment approaches, there is no appreciation built into the stock price for the coming divdend. It is simply a one time withdrawal of capital to whomever owns the stock at the date. The dividend is not a gain at all. It is a transfer of cash to the owners of the company.
I'm sure PJW is right. And the part I highlighted is what I find rather strange (not about your post, but the concept in general). If the stock does not appreciate as cash builds up, but depreciates as cash is paid out, then dividend stocks that are in situations where there is little/no room to grow but plenty of profits to be made/passed out to shareholders, should have ever-declining prices. Something seems to keep them afloat (and typically growing over time). So my thought was that the dividend decrement is offset, if not literally/directly than at least effectively (I dunno how the accountants shuffle it around), by ongoing cash flow into the coffers. Apparently that's wrong so I'll stop asking about it as it's derailing the topic too much.
Don't do something. Just stand there!

MIretired
Posts: 712
Joined: Fri Sep 06, 2013 12:35 pm

Re: "homemade dividends" from Total Stock Market fund

Post by MIretired » Mon Jul 02, 2018 12:54 pm

IlliniDave wrote:
Mon Jul 02, 2018 12:41 pm
MIretired wrote:
Mon Jul 02, 2018 12:18 pm

I think that PJW is right upthread: that the capital structure, so to speak, is completely different for bonds and coupons than for stock ownership and dividends.
eg: A bond coupon is under a contract to pay that coupon; else in default.
With stocks, you are a company owner. A dividend payment is a drawdown(term used in bookkeeping), or cash distribution of part of the assets which you are an owner of, again.
In a company financial statement and income statement done quarterly, the net (reported) earnings move over to the balance sheet (assets - liabilities). The dividend is then withdrawn from the assets(balance sheet.) This is not contractual. It is just a disbursement(drawdown) of company assets to you, an owner of the company.
I didn't find a wiki reference to company bookkeeping in a search.

And, as the time for a dividend payment approaches, there is no appreciation built into the stock price for the coming divdend. It is simply a one time withdrawal of capital to whomever owns the stock at the date. The dividend is not a gain at all. It is a transfer of cash to the owners of the company.
I'm sure PJW is right. And the part I highlighted is what I find rather strange (not about your post, but the concept in general). If the stock does not appreciate as cash builds up, but depreciates as cash is paid out, then dividend stocks that are in situations where there is little/no room to grow but plenty of profits to be made/passed out to shareholders, should have ever-declining prices. Something seems to keep them afloat (and typically growing over time). So my thought was that the dividend decrement is offset, if not literally/directly than at least effectively (I dunno how the accountants shuffle it around), by ongoing cash flow into the coffers. Apparently that's wrong so I'll stop asking about it as it's derailing the topic too much.
To me it could be kind of on topic. As where the dividends lie is somewhat relevant.
But, I wouldn't say there is cash building in the coffers as dividends, but as just earnings. Earnings are accruing, and rate how the company is doing (at least as a quarterly measurement, unless there are anouncements, etc.)
Dividend payments are NOT accruing. They are a block transfer of assets(cash/drawdown) back to the owners. 2nd thought: maybe "drawdown" isn't the best description, as it is used in sole proprietorships and bank transfers. But it's a cash-out of the company to the owners.
Also, 2nd thought: there are no earnings (accruing or otherwise) until the reporting happens. But there is an accruing (usually) cashflow. It's not called earnings until the reporting to IRS is done.
Edit: or it's the SEC that's reported to? lol
Last edited by MIretired on Mon Jul 02, 2018 1:13 pm, edited 2 times in total.

TN_Boy
Posts: 352
Joined: Sat Jan 17, 2009 12:51 pm

Re: "homemade dividends" from Total Stock Market fund

Post by TN_Boy » Mon Jul 02, 2018 12:55 pm

snarlyjack wrote:
Sun Jul 01, 2018 4:25 pm
T.

Here is his source article.
(I' am just sharing public info).

http://www.mhinvest.com/files/pdf/NF_hi ... kstrat.pdf
Okay, so I skimmed this article. Here is a question I have. He compares the performance of a set of dividend stocks with the S&P 500. Here is what he says about the dividend stocks:
We purchased a Standard and Poor’s database of monthly returns for the top 20 percent of dividend-paying companies in the S&P 500 Index (hereafter, the S&P Top 100 Dividend Payers). This database begins in 1968, is equally
weighted, and provides returns through 2007. Performing an initial comparison of the S&P Top 100 Dividend Payers to the S&P 500 Index reveals the characteristics shown in table 1.
Where table 1 shows the return of the dividend stocks as 13.52% versus 10.53% for the S&P, from 1968 to 2007.

Cool one says, "How could you not like 3% better returns per year??"

But, exactly what stocks are being used here? The top 20% in 1970 may not be the top 20% in 1971. If you use the best 20% every year to compute your returns, then you must have a process for identifying the best 20% before the year starts ....... this is not, as they say, a trivial problem. Do note that Gardner does not state this is an index anybody is maintaining (paper written back in 07 or so).

Am I missing something? Based on that paper, I cannot figure out how the performance metrics were computed. I mean, you could have an algorithm such as every year on Dec 31, rework your portfolio to include the top 100 dividend performers for that year. That's backward looking though (and ignoring any tax consequences). I couldn't tell what the author of the paper was doing. Maybe what he did was okay. But I'd need a bit more explanation.

On the other hand, at the risk of adding fuel to the fire, there IS now an index (and an ETF) that correspond pretty closely to what Gardner describes. It's the S&P® 500 High Dividend Index, and State Street has an ETF based on this index. Its the top 80, rather than 100, dividend stocks in the S&P 500. Digging around, they claim the index has returned 12.16 annualized over the last 10 years, versus 10.17 for the S&P 500. See https://us.spindices.com/indices/strate ... dend-index. That URL describes their index methodology, as well as backtested results (but only going back 10 years).

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: "homemade dividends" from Total Stock Market fund

Post by rbaldini » Mon Jul 02, 2018 12:58 pm

Scenario 1: Stock fund pays dividends; you use them for living expenses.
You own $v of stock fund. The fund pays a dividend of x% of stock's value. You do not reinvest. You own the same number of shares, but the stock price drops by an expected x% due to the dividend payout. So, you now have $v*(1-x/100) invested in stocks.

Scenario 2: Stock fund does not pay dividend; you sell at some periodic rate x% of the fund to pay for living expenses
You own $v of stock fun. You sell x% of your stock. So, you now have $v*(1-x/100) invested in stocks.

The outcome is the same, with regard to "how much do I have invested in stocks?" The only difference is that, in the latter case, you "own" a slightly smaller percentage of the companies in the fund (because you actually sold shares to someone else). I can't imagine that the difference between owning one millionth and 98% of one millionth really matters, though.

A non-trivial difference is that you probably paid less tax in the second scenario.

SGM
Posts: 2680
Joined: Wed Mar 23, 2011 4:46 am

Re: "homemade dividends" from Total Stock Market fund

Post by SGM » Mon Jul 02, 2018 1:12 pm

I don't find the term homemade dividends to be confusing, but I understand how some people could object to it. Meir Statman uses the term in Finance for Normal People when he discusses the "dividend puzzle". This is a good book to read concerning psychological errors and investor wants for utilitarian, expressive and emotional benefits regarding investing.


A nice easy to understand and fairly comprehensive book by Jane Bryant Quinn How to Make Your Money Last does not use the term homemade dividends. However, she discusses a "homemade paycheck" which can include a portfolios interest, dividends, and capital gains, and if that is not enough some of your principal too. Of course some could argue that a portfolios dividends and capital gains are principal as well. She also suggest skipping the withdrawals and paying your bills from a cash reserves if the value of your stock and bond mutual funds have both declined. She suggests a two year cash reserve. When the market starts to recover she says to begin to replenish your cash reserve. Her comments also are made with respect to a 4% withdrawal rate. Quinn has done a lot of research and includes many BH concepts in the book.

User avatar
vineviz
Posts: 999
Joined: Tue May 15, 2018 1:55 pm

Re: "homemade dividends" from Total Stock Market fund

Post by vineviz » Mon Jul 02, 2018 1:30 pm

SGM wrote:
Mon Jul 02, 2018 1:12 pm
Of course some could argue that a portfolios dividends and capital gains are principal as well.
A really wise investor would accept that, when it comes to stocks or equity funds, there is really no such thing as "principal". It's a term people use all the time, but it has no real meaning in this context.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: "homemade dividends" from Total Stock Market fund

Post by dbr » Mon Jul 02, 2018 1:44 pm

vineviz wrote:
Mon Jul 02, 2018 1:30 pm
SGM wrote:
Mon Jul 02, 2018 1:12 pm
Of course some could argue that a portfolios dividends and capital gains are principal as well.
A really wise investor would accept that, when it comes to stocks or equity funds, there is really no such thing as "principal". It's a term people use all the time, but it has no real meaning in this context.
I've tried to point out that problem from time to time as well. The "principal" is the total of whatever is in the account at the time. The math, by the definition of the word "return," is that the principal increases or decreases by the net result of contributions and return being added to the principal and withdrawals being taken from the principal. Note that adding a negative return decreases the principal, of course. If we want to get pedantic, dividends and interest are not principal, but are transactions that change the principal. After a dividend or interest payment is made that cash is an asset to be included as one element of the principal. Also note that the definition of return is explicit that return includes changes in prices of shares and dividends and interest paid out. Sometime this is called "total return," but the concept is not about asset selection or stock picking; it is about book keeping. There is no such thing as "total return" vs "dividend." The idea that there is would be a misunderstanding. A so called "dividend investor" would still want to do correct and helpful bookkeeping, though sometimes people neglect to do that.

The reason for all this semantic finagling is that the above outline of terminology and relationships gives the investor a very clear, straightforward, and mathematically correct and simple scheme to account for what goes on in or happens to a portfolio, no matter what assets are in the portfolio. It is a picture that is well aligned with how individual investors accumulate and disburse money. It is also consistent regarding such terms as gain, return, contribution, and withdrawal and with the way the industry talks about and reports investments and accounts most of us hold.

snarlyjack
Posts: 780
Joined: Fri Aug 28, 2015 12:44 pm
Location: Montana

Re: "homemade dividends" from Total Stock Market fund

Post by snarlyjack » Mon Jul 02, 2018 3:06 pm

(Lunch Time)

Phineas J. Whoopee

I' am sorry...I' am trying my best.
(It's hard to put your thoughts down on paper
when your being constantly attacked. I can go to
Seeking Alpha or other blogs & feel right at home).

However, I' am trying...thanks for being patient with me.

User avatar
Phineas J. Whoopee
Posts: 7234
Joined: Sun Dec 18, 2011 6:18 pm

Re: "homemade dividends" from Total Stock Market fund

Post by Phineas J. Whoopee » Mon Jul 02, 2018 3:07 pm

I'm afraid the word principal, as used by dbr just above, isn't right. Principal is the central sum somebody owes somebody else. With stocks nobody owes anybody anything, and in a mixed stock/bond portfolio the term isn't applicable.

I believe the appropriate term is capital. Even if there's only principal, it's still capital, a more inclusive term.

PJW

User avatar
Phineas J. Whoopee
Posts: 7234
Joined: Sun Dec 18, 2011 6:18 pm

Re: "homemade dividends" from Total Stock Market fund

Post by Phineas J. Whoopee » Mon Jul 02, 2018 3:10 pm

IlliniDave wrote:
Mon Jul 02, 2018 12:29 pm
Phineas J. Whoopee wrote:
Mon Jul 02, 2018 10:19 am

That's pretty weird, since you asked specifically about bonds.

PJW
Not really, you posted about bonds in a thread about stock dividends and I asked you a question. I shall not do that again as you seem a little prickly about it and I don't like to aggravate people.
I'm sorry if I came off as prickly. At the danger of being interpreted that way again, the thread wasn't originally about stock dividends at all. It was about portfolio withdrawals. That's what prompted my plea upthread that we stop calling portfolio withdrawals, or even sales with the proceeds kept in the portfolio, "dividends."

PJW
Last edited by Phineas J. Whoopee on Mon Jul 02, 2018 3:23 pm, edited 1 time in total.

User avatar
Phineas J. Whoopee
Posts: 7234
Joined: Sun Dec 18, 2011 6:18 pm

Re: "homemade dividends" from Total Stock Market fund

Post by Phineas J. Whoopee » Mon Jul 02, 2018 3:14 pm

TN_Boy wrote:
Mon Jul 02, 2018 12:55 pm
snarlyjack wrote:
Sun Jul 01, 2018 4:25 pm
T.

Here is his source article.
(I' am just sharing public info).

http://www.mhinvest.com/files/pdf/NF_hi ... kstrat.pdf
Okay, so I skimmed this article. Here is a question I have. He compares the performance of a set of dividend stocks with the S&P 500. Here is what he says about the dividend stocks:
We purchased a Standard and Poor’s database of monthly returns for the top 20 percent of dividend-paying companies in the S&P 500 Index (hereafter, the S&P Top 100 Dividend Payers). This database begins in 1968, is equally
weighted
, and provides returns through 2007. Performing an initial comparison of the S&P Top 100 Dividend Payers to the S&P 500 Index reveals the characteristics shown in table 1.
...
If that's what it says the article is to be dismissed out of hand. To draw such conclusions from comparing an equal-weight portfolio to a market-weight one isn't even wrong.

PJW

TN_Boy
Posts: 352
Joined: Sat Jan 17, 2009 12:51 pm

Re: "homemade dividends" from Total Stock Market fund

Post by TN_Boy » Mon Jul 02, 2018 3:43 pm

Phineas J. Whoopee wrote:
Mon Jul 02, 2018 3:14 pm
TN_Boy wrote:
Mon Jul 02, 2018 12:55 pm
snarlyjack wrote:
Sun Jul 01, 2018 4:25 pm
T.

Here is his source article.
(I' am just sharing public info).

http://www.mhinvest.com/files/pdf/NF_hi ... kstrat.pdf
Okay, so I skimmed this article. Here is a question I have. He compares the performance of a set of dividend stocks with the S&P 500. Here is what he says about the dividend stocks:
We purchased a Standard and Poor’s database of monthly returns for the top 20 percent of dividend-paying companies in the S&P 500 Index (hereafter, the S&P Top 100 Dividend Payers). This database begins in 1968, is equally
weighted
, and provides returns through 2007. Performing an initial comparison of the S&P Top 100 Dividend Payers to the S&P 500 Index reveals the characteristics shown in table 1.
...
If that's what it says the article is to be dismissed out of hand. To draw such conclusions from comparing an equal-weight portfolio to a market-weight one isn't even wrong.

PJW
Hmm, I don't know about that. You snipped the rest of my post noting that State Street is currently running an ETF using an index of 80 dividend stocks from the S&P. I did not note in my post that "The index is equal-weighted, with constituents reweighted semi-annually."

It is perfectly valid to compare the performance of an equal-weighted index with that of a market weighted index, if all I'm interested in is the performance. There is, for example, an equal weighted S&P 500 index. I am aware that equal weighted indexes introduce other confounding issues -- typically value and size factor differences from market weighted indexes. And people argue about those things.

However, though I personally don't believe in dividend focused investing, I found it interesting that a real index now exists very similar to the dataset used in the paper snarlyjack mentions. And backtested index returns (the ETF is three years old or so) are good for that index (and roughly similar to the returns in the paper, maybe 1% or so less). There are lots of potential reasons for the performance difference, but there it is.

And of course there are other dividend indexes out there, and have been for some time.

User avatar
vineviz
Posts: 999
Joined: Tue May 15, 2018 1:55 pm

Re: "homemade dividends" from Total Stock Market fund

Post by vineviz » Mon Jul 02, 2018 4:02 pm

TN_Boy wrote:
Mon Jul 02, 2018 3:43 pm
It is perfectly valid to compare the performance of an equal-weighted index with that of a market weighted index, if all I'm interested in is the performance.
Not without doing some very fancy controls it isn't. Generally when performing any sort of experimental test, you want to limit the number of variables as much as possible.

In this case, you've got at least three different changes being made at once (dividend yield, number of stocks, and weighting scheme). It's likely that other factors aren't being held constant either, which leaves it an open question as to wheether it was the dividends or the weighting scheme or some other variable that led to the difference in performance.

Indeed, without decent controls it's hard to say if the difference was even statistically significant.

Worse, we have lots of studies which demonstrate that the weighting scheme can have a BIG impact on compounded returns. Comparing an equal-weighted dividend-sorted portfolio to a market weighted unsorted portfolio is a huge red flag.

That is D minus work in any decent Finance 101 class.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

TN_Boy
Posts: 352
Joined: Sat Jan 17, 2009 12:51 pm

Re: "homemade dividends" from Total Stock Market fund

Post by TN_Boy » Mon Jul 02, 2018 4:38 pm

vineviz wrote:
Mon Jul 02, 2018 4:02 pm
TN_Boy wrote:
Mon Jul 02, 2018 3:43 pm
It is perfectly valid to compare the performance of an equal-weighted index with that of a market weighted index, if all I'm interested in is the performance.
Not without doing some very fancy controls it isn't. Generally when performing any sort of experimental test, you want to limit the number of variables as much as possible.

In this case, you've got at least three different changes being made at once (dividend yield, number of stocks, and weighting scheme). It's likely that other factors aren't being held constant either, which leaves it an open question as to wheether it was the dividends or the weighting scheme or some other variable that led to the difference in performance.

Indeed, without decent controls it's hard to say if the difference was even statistically significant.

Worse, we have lots of studies which demonstrate that the weighting scheme can have a BIG impact on compounded returns. Comparing an equal-weighted dividend-sorted portfolio to a market weighted unsorted portfolio is a huge red flag.

That is D minus work in any decent Finance 101 class.
Well, I think that is slightly missing my point. The paper snarlyjack put out had a guy essentially putting together a 100 stock "index" back in 2007. There was not a lot of detail around how he handled that data, so I was not especially impressed.

But there is a real ETF using a real equally weighted index compromised of 80 dividend stocks from the S&P 500, run by State Street. State Street compares that index to the S&P 500 over the last 10 years and shows outperformance.

Now, we are drifting far afield, but personally I think it is nothing but a religious argument to say if an index is not market weighted, it is somehow evil and cannot be compared to a cap weighted index. I *understand* that using dividend sorting (for example) typically changes the size and value factors. And that any outperformance, if it exists, might be due to reasons other than simply the dividend focus.

But that doesn't necessarily matter to the investor. Suppose dividend weighted index X consistently seems to do better than market cap weighted index Y. An academic might set forth arguments that it would be more efficient to have index Z that replicates the size and value factors we see in index X and use that instead. An the investor might say, I'll take the extra money I can get from index X right now, while the academics write papers.

It would be a fine exercise to look at the State Street Index and determine why it is a "bad thing," or that its apparent outperformance can easily be explained by other factors. In fact, that would be the real debunking of the paper snarlyjack mentions (though we don't know exactly how those stocks compare to the State Street index, they sorta feel the same based on the descriptions I can see). For example, the reason might be size factor (the index might look like a midcap index).

User avatar
vineviz
Posts: 999
Joined: Tue May 15, 2018 1:55 pm

Re: "homemade dividends" from Total Stock Market fund

Post by vineviz » Mon Jul 02, 2018 4:57 pm

TN_Boy wrote:
Mon Jul 02, 2018 4:38 pm
Now, we are drifting far afield, but personally I think it is nothing but a religious argument to say if an index is not market weighted, it is somehow evil and cannot be compared to a cap weighted index.
I never said anything about "evil".
TN_Boy wrote:
Mon Jul 02, 2018 4:38 pm
I *understand* that using dividend sorting (for example) typically changes the size and value factors. And that any outperformance, if it exists, might be due to reasons other than simply the dividend focus.
Bingo. You should stop right there. Everything after this statement is irrelevant (and wrong, but that's a story for another post).
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
jainn
Posts: 246
Joined: Tue Jun 28, 2011 6:41 pm

Re: "homemade dividends" from Total Stock Market fund

Post by jainn » Mon Jul 02, 2018 5:20 pm

I'm currently reading Against the Gods, by Peter Bernstein...this is my second time reading it...first time about 15 years ago...
Quite enjoyable..just came across the word in this thread title...it discusses homemade dividends in Chapter 17 - The Theory Police.

https://books.google.com/books?id=uTje6 ... &q&f=false

Page 290+

enjoy!
Jainn

VaR
Posts: 572
Joined: Sat Dec 05, 2015 11:27 pm

Re: "homemade dividends" from Total Stock Market fund

Post by VaR » Mon Jul 02, 2018 5:38 pm

TN_Boy wrote:
Mon Jul 02, 2018 4:38 pm
But there is a real ETF using a real equally weighted index compromised of 80 dividend stocks from the S&P 500, run by State Street. State Street compares that index to the S&P 500 over the last 10 years and shows outperformance.

Now, we are drifting far afield, but personally I think it is nothing but a religious argument to say if an index is not market weighted, it is somehow evil and cannot be compared to a cap weighted index. I *understand* that using dividend sorting (for example) typically changes the size and value factors. And that any outperformance, if it exists, might be due to reasons other than simply the dividend focus.

But that doesn't necessarily matter to the investor. Suppose dividend weighted index X consistently seems to do better than market cap weighted index Y. An academic might set forth arguments that it would be more efficient to have index Z that replicates the size and value factors we see in index X and use that instead. An the investor might say, I'll take the extra money I can get from index X right now, while the academics write papers.

It would be a fine exercise to look at the State Street Index and determine why it is a "bad thing," or that its apparent outperformance can easily be explained by other factors. In fact, that would be the real debunking of the paper snarlyjack mentions (though we don't know exactly how those stocks compare to the State Street index, they sorta feel the same based on the descriptions I can see). For example, the reason might be size factor (the index might look like a midcap index).
Why did S&P construct the index as an equal weighted index rather than a market weighted index? I'm a bit worried that they constructed it this way because it happened to backtest better when equal-weighted vs market-weighted. This would be red flag.

10-year annualized returns through June 29th, 2018
S&P 500 market weighted: 10.68%
S&P 500 equal weighted: 11.62%
S&P 500 High Dividend equal weighted: 12.16%
S&P 500 Dividend Aristocrats equal weighted: 13.28%

10-year annualized returns through May 31st, 2018
MSCI USA High Dividend Yield Index market weighted: 10.63%
MSCI USA Index market weighted: 10.7%

It looks to me like 2/3rds of the outperformance of the high dividend index was due to the fact that it was equal-weighted.

All this said, there is some evidence of the high dividend yield *factor* providing a modest performance boost. This is arguable as is the outperformance of any of the other factors like size, value, and quality.

Just looking at 10-year performance of the high dividend yield factor may conflate sector performance differences going on during this period since the S&P high dividend yield equity index has different sector weights from broad market indexes. The MSCI index excludes REITS but the S&P one does not. The top sectors in the S&P High Dividend index are Real Estate, Utilities, Consumer Discretionary, Energy, and Consumer Staples, with Real Estate and Utilities comprising 50% of the index. In contrast, the top sectors in the S&P 500 are Information Technology, Health Care, Financials, Consumer Discretionary, and Industrials, with Real Estate and Utilities comprising less than 6% of the index.

TJSI
Posts: 335
Joined: Thu Jan 27, 2011 4:03 pm

Re: "homemade dividends" from Total Stock Market fund

Post by TJSI » Mon Jul 02, 2018 6:58 pm

IlliniDave,

Your suspicions are correct. The stock price of a dividend paying stock does indeed rise before the payment. See: Dividend Capture and the EX-Dividend Price Behavior of Electric Utilities.

Unless there is an impairment to the ability of a company to pay future dividends, the intrinsic value of the company remains the same. The market price reflects that as the price will first increase then drop as the dividend is paid.

TJSI

Turbo29
Posts: 68
Joined: Tue May 01, 2018 7:12 am

Re: "homemade dividends" from Total Stock Market fund

Post by Turbo29 » Mon Jul 02, 2018 7:50 pm

jainn wrote:
Mon Jul 02, 2018 5:20 pm
I'm currently reading Against the Gods, by Peter Bernstein...this is my second time reading it...first time about 15 years ago...
Quite enjoyable..just came across the word in this thread title...it discusses homemade dividends in Chapter 17 - The Theory Police.

https://books.google.com/books?id=uTje6 ... &q&f=false

Page 290+

enjoy!
Jainn
Thanks, that was good! Also explains why stock analysts will give "Buy" or "Hold" recommendations. In reality a "Hold" is a "Buy"; deciding to hold is the same as deciding to buy the shares you already have every day. Psychologically of course it's not just as spending only dividends is not the same as selling shares when in reality it is.

TN_Boy
Posts: 352
Joined: Sat Jan 17, 2009 12:51 pm

Re: "homemade dividends" from Total Stock Market fund

Post by TN_Boy » Mon Jul 02, 2018 9:24 pm

VaR wrote:
Mon Jul 02, 2018 5:38 pm
TN_Boy wrote:
Mon Jul 02, 2018 4:38 pm
But there is a real ETF using a real equally weighted index compromised of 80 dividend stocks from the S&P 500, run by State Street. State Street compares that index to the S&P 500 over the last 10 years and shows outperformance.

Now, we are drifting far afield, but personally I think it is nothing but a religious argument to say if an index is not market weighted, it is somehow evil and cannot be compared to a cap weighted index. I *understand* that using dividend sorting (for example) typically changes the size and value factors. And that any outperformance, if it exists, might be due to reasons other than simply the dividend focus.

But that doesn't necessarily matter to the investor. Suppose dividend weighted index X consistently seems to do better than market cap weighted index Y. An academic might set forth arguments that it would be more efficient to have index Z that replicates the size and value factors we see in index X and use that instead. An the investor might say, I'll take the extra money I can get from index X right now, while the academics write papers.

It would be a fine exercise to look at the State Street Index and determine why it is a "bad thing," or that its apparent outperformance can easily be explained by other factors. In fact, that would be the real debunking of the paper snarlyjack mentions (though we don't know exactly how those stocks compare to the State Street index, they sorta feel the same based on the descriptions I can see). For example, the reason might be size factor (the index might look like a midcap index).
Why did S&P construct the index as an equal weighted index rather than a market weighted index? I'm a bit worried that they constructed it this way because it happened to backtest better when equal-weighted vs market-weighted. This would be red flag.

10-year annualized returns through June 29th, 2018
S&P 500 market weighted: 10.68%
S&P 500 equal weighted: 11.62%
S&P 500 High Dividend equal weighted: 12.16%
S&P 500 Dividend Aristocrats equal weighted: 13.28%

10-year annualized returns through May 31st, 2018
MSCI USA High Dividend Yield Index market weighted: 10.63%
MSCI USA Index market weighted: 10.7%

It looks to me like 2/3rds of the outperformance of the high dividend index was due to the fact that it was equal-weighted.

All this said, there is some evidence of the high dividend yield *factor* providing a modest performance boost. This is arguable as is the outperformance of any of the other factors like size, value, and quality.

Just looking at 10-year performance of the high dividend yield factor may conflate sector performance differences going on during this period since the S&P high dividend yield equity index has different sector weights from broad market indexes. The MSCI index excludes REITS but the S&P one does not. The top sectors in the S&P High Dividend index are Real Estate, Utilities, Consumer Discretionary, Energy, and Consumer Staples, with Real Estate and Utilities comprising 50% of the index. In contrast, the top sectors in the S&P 500 are Information Technology, Health Care, Financials, Consumer Discretionary, and Industrials, with Real Estate and Utilities comprising less than 6% of the index.
Thanks for looking up the returns of the relevant indexes; I was too lazy to do that earlier.

I agree with about everything in your post. The equal-weighting of the index is probably contributing to its "success." And I don't know why that index is constructed equal weight. Very possibly due to backtesting success. You also make a good point about the sector factor. It is interesting that Gardner's original paper, which maybe used a set of stocks similar to the State Street index (as I noted, impossible to tell exactly what he did), covered the years 1968 to 2007 and got a similar outperformance over the S&P 500.

TN_Boy
Posts: 352
Joined: Sat Jan 17, 2009 12:51 pm

Re: "homemade dividends" from Total Stock Market fund

Post by TN_Boy » Mon Jul 02, 2018 9:27 pm

vineviz wrote:
Mon Jul 02, 2018 4:57 pm
TN_Boy wrote:
Mon Jul 02, 2018 4:38 pm
Now, we are drifting far afield, but personally I think it is nothing but a religious argument to say if an index is not market weighted, it is somehow evil and cannot be compared to a cap weighted index.
I never said anything about "evil".
TN_Boy wrote:
Mon Jul 02, 2018 4:38 pm
I *understand* that using dividend sorting (for example) typically changes the size and value factors. And that any outperformance, if it exists, might be due to reasons other than simply the dividend focus.
Bingo. You should stop right there. Everything after this statement is irrelevant (and wrong, but that's a story for another post).
Please do elaborate on why everything after that statement is wrong.

CantPassAgain
Posts: 577
Joined: Fri Mar 15, 2013 8:49 pm

Re: "homemade dividends" from Total Stock Market fund

Post by CantPassAgain » Tue Jul 03, 2018 8:01 pm

Profitability and increasing earnings retained in the business (equity) would cause a stocks price to increase in a rational market. In other words, a rational player would see that out of two identical companies with the same future earnings prospects and identical balance sheets except for cash/retained earnings, the one with higher cash/retained earnings should command a higher stock price. Now, of course no two companies are absolutely identical but that's the idea.

A cash dividend reduces cash/retained earnings and thus the value of the enterprise.

The difficulty in parsing all of this comes down to changing market expectations which masks the more rational aspects of company valuation in daily noise of the market. But over time, it plays out pretty much the way it should.

Post Reply