Stingy Investor

[George$]

Thanks both Dan and Norm -I think I now see where I was misled.

In the Shiller 10-year CAPE average of P/E, he uses today's price for P - but for the E he uses the 10-year average and so E needs to be inflation adjusted.

In saying 10-year average of P/E, I was thinking that it averaged the P/E over the 10-years - and so after having done that, it made no sense to inflation adjust the E.

Thanks.

[George$]

In today's Saturday Globe & Mail-

Dividend stocks aren't fail-safe (but buy them anyway) - NORMAN ROTHERY

Another good on you Norm - and questions and possible suggestions for future articles.

Given that when you buy a share in any company - your reward comes via two possible streams. One is the money you receive back via a dividend stream and the second possible return comes via the market place if you sell the same share to the market.

Question one. In your 1927-2011 data from French showing average annual returns does it include both return streams - does it include both dividend returns and market price return?

Question two. You write

You can see that the experience of dividend stocks is quite similar to that of the market in down times. They didn't save investors during the grand daddy of crashes in 1929. Nor did they help investors much when they were licking their wounds in 2009. Dividends are not sure-fire bear repellants.

Thus the question is - did the dividend payout streams during the crashes also see a dividend down turn?

Suggestion. What about a future article showing the ratio of direct dividend returns from the company vs capital returns in the marketplace?

And more suggestions available should you ask.

Comment - about this data mining - while lookind at the past does not guarantee the same future result, looking at the past and history is helpful. Howard Marks' recent letter has much wisdom - including the following great quote -

"The farther back you can look, the farther forward you are likely to see ." - Winston Churchill

[StuBee]

Now, I am not an expert...
However, I believe that during the "Great Depression" of the last century, whereas the market tanked by close to 90% the income stream (dividends) "only" dropped by 30 to 50%. Mind you, it took close to 15 years for full recovery...
Of course, I will happily agree with Norm's superior knowledge in this area.

[George$]

Now, I am not an expert...
Of course, I will happily agree with Norm's superior knowledge in this area.

Even if true, not sure we should compliment or agree too much with Norm - he could change - how then will he keep his patient, modest and humble composure with the rest of us?

[NormR]

Question one. In your 1927-2011 data from French showing average annual returns does it include both return streams - does it include both dividend returns and market price return?

I think I might be able to boost strap this sort of info out.

Question two. You write

You can see that the experience of dividend stocks is quite similar to that of the market in down times. They didn't save investors during the grand daddy of crashes in 1929. Nor did they help investors much when they were licking their wounds in 2009. Dividends are not sure-fire bear repellants.

Thus the question is - did the dividend payout streams during the crashes also see a dividend down turn?

Suggestion. What about a future article showing the ratio of direct dividend returns from the company vs capital returns in the marketplace?

Might this page help?

And more suggestions available should you ask.

I'll take any and all suggestions, I'll eventually run out of ideas for the column.

[NormR]

Now, I am not an expert...
Of course, I will happily agree with Norm's superior knowledge in this area.

Even if true, not sure we should compliment or agree too much with Norm - he could change - how then will he keep his patient, modest and humble composure with the rest of us?

I look forward to getting a swelled head.

[StuBee]

Question two. You write

You can see that the experience of dividend stocks is quite similar to that of the market in down times. They didn't save investors during the grand daddy of crashes in 1929. Nor did they help investors much when they were licking their wounds in 2009. Dividends are not sure-fire bear repellants.

Thus the question is - did the dividend payout streams during the crashes also see a dividend down turn?

Suggestion. What about a future article showing the ratio of direct dividend returns from the company vs capital returns in the marketplace?

Might this page help?

Practically speaking, my take on the first "X10" graph of your link is that 1929-1930 peak to trough was roughly 50% in nominal dividends and recovery took around 18 to 20 years (nominally)

[George$]

I hope readers can help me with a clearer understanding of annual returns - both market returns for indexes (for equities and bonds) themsetves - versus what the dividends or yields can return in addition.

My impression is that markets are easy to follow via indexes - but dividends or yields (in addition to market returns) are not.

Norm - I'm posting here because I believe you may have considers this in detail.
(i) Am I correct that your posted periodic table at your webste - does not include annual corresponding dividends or bond yields?
(ii) You post a Stingy Investor Asset Mixer- using Norm's tables at Libra - which are very helpful because they are in Canadian$ - and post both nominal and real returns. But am I correct in thinking that they ignore the addition spin off that some equities and some bonds may spin off via dividends and yields?
(iii) In your own track record, you often qualify the realized returns by saying that dividends are not included.

My questions:
(i) Does anyone include total annual index returns - including dividends, etc.
(ii) I assume annual dividend returns are very time dependent - for example when P/E are low the dividends relative to P will be higher? Has anyone studies how dividends (or earnings?) have varied over times?

Any comments and suggestions are welcome. And appreciated?

[Shakespeare]

I believe Norbert's data are "Total Return" - that is, they assume dividends or coupons are reinvested.

[NormR]

Shakes is right, the figures are total returns and include reinvested dividends.

For question (ii), see this

[George$]

I believe Norbert's data are "Total Return" - that is, they assume dividends or coupons are reinvested.

Thanks Shakes. I hope Norbert can confirm this - or if not explain it further. I'm wondering who does this "total" and provides it where for Norbert.

I realize that some years the dividend spin offs may be minimal - when market P/E is high. But in years past when P/E was low - dividend %s might have been x2 to x4 (?) higher.

I'm still trying to connect the dots - and as I get older even rember what I once knew. Woe me.

[NormR]

I believe Norbert's data are "Total Return" - that is, they assume dividends or coupons are reinvested.

Thanks Shakes. I hope Norbert can confirm this - or if not explain it further. I'm wondering who does this "total" and provides it where for Norbert.

The index providers (S&P, etc) publish both regular and total return indexes.

[George$]

Shakes is right, the figures are total returns and include reinvested dividends.

Thanks Norm. Am I confused - do most index returns show "total return"? - or is it one or the other and often mixed one vs the other?

Norm, I assume that in my (iii) item for you would require individual tracking for individual equities - and so you probably do not do this?

[George$]

I believe Norbert's data are "Total Return" - that is, they assume dividends or coupons are reinvested.

Thanks Shakes. I hope Norbert can confirm this - or if not explain it further. I'm wondering who does this "total" and provides it where for Norbert.

The index providers (S&P, etc) publish both regular and total return indexes.

Again thanks. I will go to S&P to check this out.
Does MSCI do the same for their indexes?
What crosses my mind is that MSCi is more favoured by Vanguard than S&P days - they probably charge less for using it?

[StuBee]

I believe Norbert's data are "Total Return" - that is, they assume dividends or coupons are reinvested.

I have always thought the same thing...We have all heard some pundits say: "The majority of the return over the last century is largely attributed to dividends". Whether or not this is true is not the point. However, if they go about saying that, obviously the dividend component is a part of the returns that they are referring to.

However, from a purely technical aspect, I wonder how this is done? Are the dividends reinvested in the originating corporation or in the market in general? I presume that they are reinvested in the originating companies.

This means that the Total Return of a given index is what you would get if you "dripped" the same index. Mathematically, "dripping" is a very powerful tool. Real life cannot possibly approximate this because eventually in real life there is the "withdrawal phase". Conceptually, this period is the opposite of a "drip" (instead of reinvesting the income, you are removing the income).

Edited to add: "oops, a bit of cross-posting in my post..."

[like_to_retire]

Conceptually, this period is the opposite of a "drip" (instead of reinvesting the income, you are removing the income).

I don't think it's the opposite. The return is still the appreciation of the stock price plus the dividend cash that you presumably spend on hamburgers. The difference is that you're not getting the compounding return from the re-investment of the dividend itself (whether injected back into the stock through a drip or placed in another investment).

ltr

[Taggart]

This article from The Economist explains much of what you may be looking for:

"The first step is to define the equity risk premium more exactly. Mssrs Dimson, Marsh and Staunton break it down into the following components: the dividend yield, plus the real dividend growth rate, plus or minus any change in the price/dividend ratio (the inverse of the dividend yield), minus the real risk-free interest rate.

In the period 1900-2011, the average world dividend yield was 4.1%; real dividend growth was just 0.8%; and the rerating of the market added 0.4%. That comes to a real equity return of 5.4% (the calculation is geometric, not arithmetic). Stripping out the risk-free interest rate, the ERP was 4.4% versus short-term government debt and 3.5% versus longer-term government bonds (see chart 2).

The dividend yield comprised the vast bulk of the return. This was true across all the countries studied by the authors. Had investors consistently bought the highest-yielding quintile of equity markets over the past 112 years they would have earned an average nominal annual return of 13.3% compared with a return of just 5.4% for those buying the lowest-yielding quintile. High-dividend markets have also performed best so far this century."

-------------------------------------------

Robert Arnott found much the same a few years earlier in his "Dividends and the Three Dwarfs".

[StuBee]

Conceptually, this period is the opposite of a "drip" (instead of reinvesting the income, you are removing the income).

I don't think it's the opposite. The return is still the appreciation of the stock price plus the dividend cash that you presumably spend on hamburgers. The difference is that you're not getting the compounding return from the re-investment of the dividend itself (whether injected back into the stock through a drip or placed in another investment).

ltr

Sorry, I agree. Opposite is too strong a word. However, with the evidence that Taggart has stated, removing the dividend as a source of continuing growth removes an historically powerful contributor to the "Total Return".

[George$]

Thanks so much for all this input. Much appreciated. Will read the references.

A related question. I agree that annual dividends for public companies can be computed. Each share for all shares in a public equity receives the same.

But how is yield included for bonds? Can it also be included?

I also realize that gold increases are simple - only the market determines all. There is no dividend etc.

[Shakespeare]

I don't know whether the index total returns reinvest in individual stocks, George. They could just sum the day's dividends and reinvest in the index at its end-of-day value.

Added: see Total Return Index Definition | Investopedia

Further added: https://www.sp-indexdata.com/idpfiles/e ... th_Web.pdf suggests that is what is done (p. 35).

[StuBee]

I don't know whether the index total returns reinvest in individual stocks, George. They could just sum the day's dividends and reinvest in the index at its end-of-day value.

Added: see Total Return Index Definition | Investopedia

Further added: https://www.sp-indexdata.com/idpfiles/e ... th_Web.pdf suggests that is what is done (p. 35).

Well, that is quite surprising. It appears that a higher yielding subindex (for instance utilities or telecommunication) is "juicing the yield" of a relatively low yielding subindex (for instance materials) and thereby "skewing" the "Total return" calculation for the broader index!

Perhaps this would be the appropriate time to quote Solomon: "Meaningless, meaningless!" says the Teacher "Utterly meaningless! Everything is meaningless."

[Shakespeare]

It appears that a higher yielding subindex (for instance utilities or telecommunication) is "juicing the yield" of a relatively low yielding subindex (for instance materials) and thereby "skewing" the "Total return" calculation for the broader index!

Huh? How does summing the (weighted) dividends and reinvesting in the total change the subindexes? Surely the subindexes (and the 500 itself is a subindex) are calculated the same way individually?

[Taggart]

But how is yield included for bonds? Can it also be included?

I also realize that gold increases are simple - only the market determines all. There is no dividend etc.

Most certainly re-investing the yield on bonds can be included. At least that's what Elroy Dimson found regarding U.S. bonds that were included along with other assets in a paper he wrote a few years back for The Brandes Institute.

Examining the Income Component of Total Returns

[Norbert Schlenker]

The index returns collected at Libra are all total returns, i.e. with dividends reinvested. (The reinvestment occurs into the index as a whole, not stock by stock.) For bond indices, the same thing happens as coupons are collected on individual index components. What you generally hear on the radio or see in the paper (the TSX was up 42 points today) is price change only, not total return. Ferreting out where the total return indices are published can sometimes be a bit of detective work. Index providers like to hide them because they sell the time series for real money.

George mentioned MSCI indices and I should caution users of the MSCI performance tables to be extremely careful when using their site. For each of its indices, MSCI publishes six separate time series and you need to be careful to choose exactly what you want. You must choose between local currency/USD returns, and also one of price/gross/net. There are vast differences among these time series, not on any individual day usually, but over time. The returns in the Libra spreadsheet for MSCI indices (EAFE and emerging markets) are taken from the USD gross time series and adjusted for changes in the USD/CAD exchange rate.

Stubee's and like_to_retire's back and forth about what happens when dividends are not reinvested is also apropos. If you're an investor who is spending all of your portfolio dividends then, even in countries with a long history of good real returns on equities, chances are your compounded returns over time are derisory.

<ot> The Financial Post usually publishes the PC-Bond indices (which are total return by the way) for each market day here. Today there is nothing for yesterday. There should be an equivalent table in the printed Financial Post. If a forum member who subscribes to the print edition could PM me what's been printed today, I would be grateful. (The primary source is online here but publishes only a single decimal place for each index.) </ot>

[StuBee]

It appears that a higher yielding subindex (for instance utilities or telecommunication) is "juicing the yield" of a relatively low yielding subindex (for instance materials) and thereby "skewing" the "Total return" calculation for the broader index!

Huh? How does summing the (weighted) dividends and reinvesting in the total change the subindexes? Surely the subindexes (and the 500 itself is a subindex) are calculated the same way individually?

Because dividends from a part of the index (for instance BCE and all others that pay a dividend) are being used to purchase shares of the entire index (including companies that pay no dividends at all). Perhaps I should not have used the word "subindex". I was just trying to succinctly express my argument.