Income & Dividend Investing

Re: mathguy from Ken

mathguy2 — Fri, 30 Nov 2007 15:19:01 GMT

"I'd be interested in your comments re cap and risk."

Ken,

I don't pretend to have any special insights into this, but here is the data I previously posted along with the Dec 2006 average market cap (in millions):

	1928 - 2006			Dec 2006
Div Yield	Total Ret	Total Ret	Dividend	Average
Decile	Incl Divs	Excl Divs	Return	Market Cap
No Divs	8.24%	7.65%	0.61%	1,455
Low 10	8.68%	6.72%	1.94%	6,201
Dec 2	10.11%	7.13%	2.97%	6,322
Dec 3	9.63%	6.14%	3.45%	6,238
Dec 4	10.83%	6.79%	4.02%	6,448
Dec 5	9.23%	4.81%	4.41%	9,447
Dec 6	10.33%	5.40%	4.96%	8,118
Dec 7	11.49%	5.98%	5.54%	6,486
Dec 8	12.19%	6.12%	6.11%	7,148
Dec 9	11.36%	4.52%	6.93%	10,734
High 10	11.20%	3.28%	8.03%	3,716
Total Market	10.03%	5.86%	4.19%	3,736
High 30	11.75%	4.95%	6.87%	7,375

As you would expect, the no dividend group is predominantly small cap. Decile 9 is significantly larger than the other dividend payers and Decile 10 is significantly smaller. The other above-average dividend payers (Deciles 5-8) are somewhat larger than the lower yielding dividend payers.

Re: El Lobo

ElLobo — Fri, 30 Nov 2007 07:12:48 GMT

Helmut,

Good luck in your upcoming retirement. Yield focused investing isn't for you.

Warmest regards,

Chuck

Re: El Lobo

ElLobo — Fri, 30 Nov 2007 07:03:10 GMT

JWR,

"A total return strategy worked great during the 1950s. It was lousy during the 1960s"

Will the 2010's be like the 1950s, or the 1960s?

(My guess is more like the 60s, because of valuations).

A yield based strategy works well in any market.

Re: El Lobo

JWR1945a — Fri, 30 Nov 2007 02:14:11 GMT

A total return strategy can work well when valuations are low at the beginning of a bull market.

A total return strategy worked great during the 1950s. It was lousy during the 1960s.

Have fun.

John Walter Russell

Re: El Lobo

JWR1945a — Fri, 30 Nov 2007 02:09:45 GMT

Latest buzzword:

If you don't understand, then it is your confirmation bias that does not allow you to even acknowledge, much less dispute any of the facts I have presented over the course of our debates showing that a total return strategy can be just as successful as any of the high yield SWR theories you have present thus far.

Any truth supported by mathematics and history is called "confirmation bias."

Have fun.

John Walter Russell

Re: El Lobo

helmut — Thu, 29 Nov 2007 23:38:22 GMT

El Lobo

No, I call it facts, not theory. My savings account example is the 'withdraw less than the yield' concept applied to savings accounts. Why do you consider this a 'theory'?

Your theory lies within your implication was that the risk on the interest and principal of a federally insured bank savings account is the same as a stock such as NAT or RAS as long as you take a distribution that is less than the yield. That type of assertion is frivolous to me.

The SWR depends on the amount withdrawn from a portfolio. Specifically, a withdrawal rate is considered 'safe' if it lasts as long as you live, or as long as you need it to last.

Yields do not guarantee SWR as you define them, earnings (profit) do. Without earnings you have nothing. The ability to earn profit along with the intrinsic value of a company determines the true value of a company. If you have a major long term drop in share value (reversion to the means) it means you paid too much for speculation.

The Russell 3000 only has 52 stocks that have a yield of 1% or more with a consecutive 20 year dividend growth record. Of those 52 stocks only two have a current yield of 6% or more. Dividend growth rates are even bleaker when the yield goes over 7%. Of all the Russell 3000 stocks yielding 6% or more only 14 have consecutive dividend growth for 10 years. Six have consecutive dividend growth for 14 years. Of all the stocks yielding 6% or more only five had consecutive earnings growth of 1% or more for as long as four years.

Bottom line earnings growth leads to dividend growth. Without dividend growth higher yielding stocks are more likely to encounter not only share value volatility, but dividend volatility as well. Now if you want to say you don't need dividend growth if you re-invest part of the dividend to grow your dividend, you still will have dividend volatility which will create the risk and uncertainty that you are trying to avoid in the first place.

This debate has not touched on the risk of bankruptcy. If you want to argue that NAT is a safer stock than XOM solely because it has a higher yield, you will have to take the risk of that being called a frivolous statement by me as well.

Whenever you withdraw more than the yield of the portfolio, you 'touch principle', you 'spend capital', you gradually liquidate your portfolio. You can do this safely only if you die before you liquidate.

If you want to parse words to win the debate so be it, but as long as your portfolio increases in the long term and you take less in the form of a distribution than the increase you will not liquidate your portfolio. If your distribution is more than the increase you will liquidate your portfolio. The size of the yield will not prevent that.

As far as separating theory from fact, just about every one of your post regarding SWR is theory with very little in the way of facts to support them. Even the Bernstein study that you quote so often to discredit income certain withdrawal strategies does nothing to support any of your high yield SWR theories.

As far as the investment strategy outlined in the beginning of this thread, I see no substance to back it up. That does not mean it will not work, but it has much less creditability than even the simple Dogs of the Dow dividend strategy which has averaged an annual total return of 17.7% since 1973.

I don't understand. Are you saying that facts support your theory, specifically that the return you get from the market (share price appreciation) is more predictable, more certain, than the return you get from the company (it's dividend?)

If you don't understand, then it is your conformation bias that does not allow you to even acknowledge, much less dispute any of the facts I have presented over the course of our debates showing that a total return strategy can be just as successful as any of the high yield SWR theories you have present thus far.

helmut

Re: EL from Ken250

JWR1945a — Thu, 29 Nov 2007 18:28:56 GMT

Bogle defines the investment return as the initial dividend yield plus the earnings growth rate.

Total return = investment return + speculative return

Have fun.

John Walter Russell

EL, re Earnings

ken250 — Thu, 29 Nov 2007 16:45:49 GMT

Don't be getting cute....

EL from Ken250

ken250 — Thu, 29 Nov 2007 15:58:57 GMT

Bogle didn't say there are 2 components to TR, he has always said there are 3:

1) Growth

2) Dividends

3) Speculation.

In his recent book he has warned investors not to expect speculation to bolster returns, the speculation excesses of the last bull need to be undone.

mathguy from Ken

ken250 — Thu, 29 Nov 2007 15:46:16 GMT

I'm glad you were able to decipher what I was trying to get across to EL.

I notice the tendency in the data for higher yield to have higher TR.

O'Shaughnessey and Damodoran showed that high yield alone doesn't beat the market.

Also, O'Shaughnessey looked at this on a capitalization and risk-adjusted basis and found the only "HY premium" is for HY US megacaps.

I'd be interested in your comments re cap and risk.

Good Luck, Ken.

Re: Well said

meyerr — Thu, 29 Nov 2007 12:02:11 GMT

mathguy2:
"You are making a mistake failing to distinguish NAV growth due to speculation from NAV growth due to underlying earnings. Both are reflected in the share price, but one is temporary and the other is pseudo-permanent. I guess this is what drives you to the other extreme, ie high yield, which I view as risky because you're investing in companies with little growth potential for the sake of yield."
Here are the 79-year returns by dividend yield decile.

Div Yield Total Ret Total Ret Dividend Div Growth
Decile Incl Divs Excl Divs Return 1928-2006
No Divs 8.24% 7.65% 0.61%
Low 10 8.68% 6.72% 1.94%
Dec 2 10.11% 7.13% 2.97%
Dec 3 9.63% 6.14% 3.45%
Dec 4 10.83% 6.79% 4.02%
Dec 5 9.23% 4.81% 4.41%
Dec 6 10.33% 5.40% 4.96%
Dec 7 11.49% 5.98% 5.54%
Dec 8 12.19% 6.12% 6.11%
Dec 9 11.36% 4.52% 6.93%
High 10 11.20% 3.28% 8.03% 3.69%
Total Market 10.03% 5.86% 4.19% 5.39%
High 30 11.75% 4.95% 6.87% 5.46%
Stocks which have paid above average dividends have had higher total returns than other stocks. You often hear that people overpay for growth and avoid risky, or slower growing, value stocks. These returns are consistent with that thinking.
This data shows that as you get into the two highest yielding deciles, the price growth is lower and the dividend growth is lower, than the other above-average dividend payers. This makes sense because more earnings are being paid out as dividends, not retained for investment in future earnings growth.
The difference in growth rates is small enough where it might not be noticeable in the early years of retirement. But as time goes on the highest yielding stocks' dividends often fail to keep pace with inflation. In order to maintain your spending level, you are then forced to sell shares whose prices have appreciated at a slower rate.

It looks like the numbers confirm a dividend based strategy. El Lobo has already talked about what might impact the data in the highest deciles and I'm discriminating based on just dividend based investing rather than high yield investing. When you start looking at the middle to third deciles both the total return and dividends are higher than a market portfolio. But what really intrigues me and I don't know if the data is available would be the difference between the U.S. market data and some of the European markets where dividend investing is much more the norm and has received more favorable tax treatment. In this country, dividends go in and out of fashion, and again, I don't know if the data is available, but wonder if the data were correlated with when dividends were in vogue or had a more tax tax treatment if the results would be different. Are some important aspects of all of this lost in the aggregation of data? Does the time period selected impact the results?

Roberta

Re: Re:ElLobo revisited.........

egyhazy — Thu, 29 Nov 2007 06:23:27 GMT

There is an unbelievable amount of garbage in this thread and on this forum in general. Anyone treads into this cesspool needs to do their due diligence - PLEASE.

The fact that stock price should fall given a dividend, all other things being equal, is because retained earnings are reduced. Current assets fall (cash), working capital suffers, various debt/equity metrics (which determine cost of capital) are worsened, and growth potential is stunted.

This is not to say that dividends are bad - they are a useful, but somewhat tax inefficient, way to distribute earnings to shareholders. For a mature and profitable firm dividends should be encouraged.

Re: Well said

ElLobo — Thu, 29 Nov 2007 06:14:12 GMT

Mathguy2,

"This data shows that as you get into the two highest yielding deciles, the price growth is lower and the dividend growth is lower, than the other above-average dividend payers. This makes sense because more earnings are being paid out as dividends, not retained for investment in future earnings growth."

Total returns, including dividends, were also lower. Rather than making sense, this can't be the reason, unless you also looked at the dividend payout ratios. That is the measure of paid, versus retained, earnings.

A more plausible and logically consistent explanation is that these high dividend payers over the years are the ones in the final death throws of their existance, thowing dividend money out to attract investors. The buggy whip boys, as it were. It's impossible for all companies to continue to grow, indefinately. Bernstein even had a paper on this subject.

Regardless, the data is what it is. What would be interesting is to see if statistical analysis of this data shows any causal relationship between the growth portion of TR and the yield portion. That is, is the amount of growth return you receive expected to be about the same, regardless of the yield portion you expect? Or do you give up growth return if you focus on yield return. A quick examination of the data does't seem to show any trend.

Re: Re:ElLobo revisited.........

ElLobo — Thu, 29 Nov 2007 06:02:11 GMT

Mathguy2,

Although I understand the Dividend Discount Model, and have played with it quite a bit, do you really think anyone would use it to place buy or sell orders for stocks?

"But you have to think that most buyers would include the knowledge of the lost dividend in the formulation of their bid."

Whenever I buy or sell, I rarely use limit orders. I buy or sell at market. The few extra dollars I might receive by using limit orders are well spent in knowing that I'm buying, or selling, close to my target prices, as explained in this thread.

Re: EL, I see the light!

ElLobo — Thu, 29 Nov 2007 05:39:26 GMT

Ken,

"I guess this sums up your rationale for investing such that you get as much of your Total Return from income as possible."

I get as much of my total return from income/yield as I need to fund my withdrawals. Anything I don't need for that purpose stays in the portfolio, in the asset that generated the yield, or another. Once that absolute need is met, then any return I receive (positive or negative) from share price changes is what it is. I simply capitalize on the volatility, as I explained in the original posting of this thread.

That is, I don't RELY on realizing capital (selling shares) to meet those withdrawal needs. If I did, the unrealized gains, each year, would have to be greater than the withdrawal in order to maintain portfolio value. That implies share prices always increasing. That's an assumption TR, or growth, focused investors have to make in order to have a safe withdrawal rate.

Go back to the simple equation I gave in this thread.

"There is no such thing as company generated return."

Yes there is. I receive a check from the accounting department of each company I own, whenever they make a dividend payment. They determine the size of the check.

The return I receive from the market, that is, realizing capital (the proceeds from the sale of shares) comes in the form of a check from a greater, or lesser, fool than I.

"You consistently imply that growth (NAV appreciation) is due to speculation."

No, I have always said that share prices don't always rise, which is growth. Sometimes they decline, which is a loss. The point is that share price behavior is volatile, implying that growth returns are volatile (actually, that's a definition, not an implication). To me, that's risk, the uncertainty of share price related returns. A risk you don't see in the volatility of the dollar amount of the yield.

What I have also always said is that there is NOT a direct, well defined, functional relationship between earnings and share prices. That is, given earnings are such and such, share prices should be so and so.

As part of the 'Active Value Investing' thread a while back, I was very clear that people, such as Jack Bogle, define total return as having two components, what he calls the investment return and what he calls the speculative return. The investment return is the yield, the speculative return is the share price behavior portion.

He, as well as I, as well as most everyone else understand that investment return is a function of earnings, that is, as long as you have earnings, then the company can pay a dividend. But you need earnings GROWTH to get dividend growth, as well as share price growth. Without earnings growth, any share price movement is speculative.

That is, in your vocabulary, an expansion or contraction of the P/E ratio is speculation. And it's also clear that, if you focus on share price growth for the major part of the return you expect, earnings also have to grow!

"For us forum posters speculation does nothing for our portfolio value over the long term."

Well, do you think you are NOT speculating if you invest in a company primarily for the growth component of total return? You expect EPS to continue to grow, year after year, otherwise share prices won't grow! You also expect the market, efficient or otherwise, to corectly price the shares with regard to that EPS growth. In fact, you are speculating on a whole bunch of things to occur, all favorably, the bottom line being share prices rise. Then the unexpected happens, and share prices tank.

Given all of the above speculation, you then estimate, or expect, a certain growth return. That, typically, will be somewhere in the neighborhood of 6-10%/year.

For a yield focused investor, we speculate that the company will pay it's regular dividend quarter after quarter, year after year. Our greatest risk is that it cuts, or eliminates, its dividend. Given this speculation, we expect the yield return to be the yield paid by the company last year! That, typically, will be the percentage yield of the company, fund, or whatever asset that we happen to choose, whether it be 6%, 10%, 1.75%, or zero %.

"Reinvestment is required to keep pace with inflation, build new facilities, purchase the latest widgets that improve efficiency, conduct R&D, etc. Do you really believe a company that is not doing these things with E/FCF is less risky because it's paying out everything to shareholders? It's a dead-end investment, in fact it's really short term speculation in the sense that you're gambling to get back more in dividends than what you invested...knowing full well there's no future."

Now your reaching, and getting into 'quality of dividend' issues. Specifically, you're talking about the dividend payout ratio, that is, the fraction of earnings paid out as dividends. Retained earnings ratio is 1 minus the payout ratio. That is, if a company earns $3/share/year, and pays a $1.50 dividend per year, it retains $1.50, for all of the reasons you listed, even those already subtracted from profits to calculate earnings. Both the dividend payout ratio and the retained earnings ratios are 0.5.

Now, tell me why it's better for a company to retain the whole $3, rather than just $1.50? Do they need the extra buck and a half for overpriced acquisitions, or to pay for improperly priced stock options given in lieu of salaries? 'Howzabout a buck or so, per share, for super-widget, that never makes it past the first line manager of the R&D department? 'Maybehaps a few billion to cover subprime loan losses!

Then, given that the company DOES pay the $1.50, does it make any sense that this stock is more risky, at $15/share (a 10% yield) than at $75/share (a 2% yield?) It's the same $1.50 being paid, by the same company with the same fundamentals, management, same everything!

"You are making a mistake failing to distinguish NAV growth due to speculation from NAV growth due to underlying earnings."

You are making the mistake in thinking that share price growth depends upon earnings, not earnings growth. That is, the mistake is thinking that share price growth without earnings growth is NOT speculation. All growth focused investors make this mistake.

"drives you to the other extreme, ie high yield, which I view as risky because you're investing in companies with little growth potential for the sake of yield."

So, you view high yield (say, 10%) as risky, but not 2% (in my above example), even though the ONLY difference between the two was share price?

Think about it.