Dividend and Buyback Yields, S&P 500, 1982-2015

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Dividend and Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

Many investors have been discouraged by the declining dividend yields of the large-cap U.S. stock indexes over the last three decades. However, when the net effect of share buybacks (the net buyback yield) is added to the dividend yield, the overall return to shareholders looks much more encouraging (chart below). In fact, the recent years' overall returns have been a bit above the 3.2% average of the last 35 years:
  • • Dividend Yield = Trailing 12 months cash dividends / Market capitalization
    • Net Buyback Yield = (Trailing 12 month stock repurchases - stock issuances) / Market capitalization
Bottom Line: It appears that no matter what happens to stock valuations and prices going forward, if dividends and net buybacks continue at their current pace, investors can expect to earn at least 3%-4% (in real, inflation-adjusted terms) on their U.S. large-cap stock investments.

Any thoughts on this way of looking at returns?
Last edited by SimpleGift on Mon Jun 20, 2016 8:48 pm, edited 1 time in total.
User avatar
triceratop
Posts: 5838
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Dividend & Buyback Yields, S&P 500, 1982-2015

Post by triceratop »

SimpleGift wrote:Bottom Line: It appears that no matter what happens to stock valuations and prices going forward, if dividends and net buybacks continue at their current pace, investors can expect to earn at least 3%-4% (in real, inflation-adjusted terms) on their U.S. large-cap stock investments.
Is this not true only for the first year (before inflation actually starts eroding the value of your $$)? Or perhaps you're implicitly concluding that the market prices inflation into Large Cap stocks ? I'm no sure I follow the logic here, because it seems if you were to have a nominal bond with a 3-4% coupon you'd be in the exact same cash-flow situation as with Large Cap (except with equity risk). Where is the inflation protection?
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."
chx
Posts: 429
Joined: Sat Jan 16, 2016 10:52 am

Re: Dividend & Buyback Yields, S&P 500, 1982-2015

Post by chx »

It's hard to read the chart in that form. For example, in 2008, it looks like there are both negative and positive amounts. How do you get a negative amount of buybacks? It appears that for 2008, the chart does not explicitly show a total to compare with other years.
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Re: Dividend & Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

triceratop wrote:Where is the inflation protection?
Sorry not to be more explicit with the assumption that real dividend growth rates (and real net buyback rates) would be positive going forward. Obviously no future guarantees, but dividend growth rates have consistently outpaced inflation in the U.S., especially since the 1940s:
  • Image
    Note: These are growth rates of dividend dollars, not dividend yields.
    Source: WisdomTree
chx wrote:How do you get a negative amount of buybacks?
On a net basis, more stock was issued for the year than was repurchased. Here’s the equation from the OP:
  • Net Buyback Yield = (Trailing 12 month stock repurchases - stock issuances) / Market capitalization
User avatar
gordoni2
Posts: 213
Joined: Wed Aug 15, 2007 12:20 am
Contact:

Re: Dividend & Buyback Yields, S&P 500, 1982-2015

Post by gordoni2 »

I think this might be a case of miscounting. It seems to me that the cash from buybacks is not received by investors who continue to hold shares in the S&P 500 index, or if they do receive the cash then they have to content themselves with holding a reduced number of index shares going forward.

I could be wrong on this point. Damodaran likes to include buybacks in his equity risk premium estimates, but I have a hard time reconciling doing this with the S&P 500 index definition which includes a divisor adjustment in response to buybacks -- meaning the index level is unchanged in response to a buyback. The index thus represents the performance of an investor who elects to hold his or her underlying shares. If no shares are given up in response to a buyback, then I am afraid no cash is received. Would love to know what others think on this.
AlohaJoe
Posts: 6589
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Dividend & Buyback Yields, S&P 500, 1982-2015

Post by AlohaJoe »

gordoni2 wrote:I could be wrong on this point. Damodaran likes to include buybacks in his equity risk premium estimates, but I have a hard time reconciling doing this with the S&P 500 index definition which includes a divisor adjustment in response to buybacks -- meaning the index level is unchanged in response to a buyback. The index thus represents the performance of an investor who elects to hold his or her underlying shares. If no shares are given up in response to a buyback, then I am afraid no cash is received. Would love to know what others think on this.
Interesting, I've never seen the methodology before.

It does look like the S&P Index would change the Divisor when a stock buy back happens. (It explicitly says that in the chart on page 10, after all.) But they also say
An index is not exactly the same as a portfolio. For instance, when a stock is added to or deleted from an index, the index level should not jump up or drop down; while a portfolio’s value would usually change as stocks are swapped in and out.
So the index doesn't represent the performance of an investor who elects to hold the underlying shares. If I understand what they're saying, then if there were some massive share buy back then the index itself would remain unchanged but the NAV of mutual funds that track the index would go up. Which would show up as index tracking error, I guess?
User avatar
whodidntante
Posts: 11720
Joined: Thu Jan 21, 2016 11:11 pm
Location: outside the echo chamber

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by whodidntante »

When are in a period of exceptionally low rates and long-term investment is viewed with suspicion, or just "wait and see." Might as well buy back some shares.
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

gordoni2 wrote:I think this might be a case of miscounting.
My understanding is that it would be miscounting (actually double counting) if the net effect of buybacks was added to per share data of the S&P 500. For example, if earning-per-share was adjusted upward due to the effect of net buybacks, this would be double counting.

However, by adding the buyback yield to the dividend yield, as in the OP, we are not dealing with per-share data. Also, granted, the S&P 500 shareholder doesn’t receive the net buyback yield directly — but the net effect of share buybacks should ultimately be reflected in the price of the index (as the S&P 500 divisor is reduced).

This short article from S&P might shed some light: Buybacks and the S&P 500 EPS
John Hussman also discussed this issue in an article: Double Counting
TJSI
Posts: 399
Joined: Thu Jan 27, 2011 4:03 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by TJSI »

Simplegift,

That buyback yield is not a yield in any conventional definition of yield. You are just adding two dissimilar percent changes to produce a number which means nothing. A yield is a return on investment. It tells you what you are gaining from your investment.

When people talk of stockholder yield they should state which stockholders they are referring to: selling stockholders, non-selling stockholders, or perhaps an average stockholder who sells a fraction of their shares. Each would have a different return.

One simple way of computing a selling shareholder buyback yield is by using the formula Y = (selling price - buying price)/buying price * number of years held. Holding shareholders get nothing directly. Their direct buyback yield is zero. They may over time see an increase in dividends per share, earnings per share, or valuation change. Or, they may not if the company paid over the intrinsic value of the stock.

Buyback dollars are huge and only time will tell if they were worthwhile.

TJSI
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

TJSI wrote:That buyback yield is not a yield in any conventional definition of yield. You are just adding two dissimilar percent changes to produce a number which means nothing.
Granted, when companies return cash to shareholders through buybacks, the impact is more ambiguous to the shareholder than if by direct dividend payments. But it still seems like a return of capital to shareholders that should be accounted for in some way — especially with the massive amount of net buyback dollars being spent in recent years.

With dividends, it's a return of cash that all shareholders partake in, in proportion to their stockholding. In a stock buyback, only those shareholders who tender their shares back to the company get cash. The remaining shareholders get a larger proportional stake in the remaining firm. In both cases, it seems there's a return to shareholders.
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by ogd »

TJSI wrote:When people talk of stockholder yield they should state which stockholders they are referring to: selling stockholders, non-selling stockholders, or perhaps an average stockholder who sells a fraction of their shares. Each would have a different return.

One simple way of computing a selling shareholder buyback yield is by using the formula Y = (selling price - buying price)/buying price * number of years held. Holding shareholders get nothing directly.
Wait -- do you make the same demands of dividend yields vs reinvesting investors? Like "I know Chevron paid 4% a year for the past 30 years, but some investors reinvested the whole period, others reinvested for 20 then withdrew for 10, others withdrew the whole time; what are all of their yields?", or "XYZ paid 1 billion in dividends, but some investors who bought at this unfortunate time and kept reinvesting never saw a penny, and now their return is negative?"

No, because you understand that whether or not investors increase or keep their share of the company at a given price is their decision, not the company's. It's the same with buybacks. During a buyback, each investor has the option of keeping their share constant (by participating or selling) or increasing it.

You seem to think of the buyback as some sort of investment. It isn't. I don't want management to worry about whether this is a good price or not, any more than I'd want them to time their dividend payments to try to get dividend-reinvesting shareholders a good price. The less attention they pay to the price, the better. If I think this is a bad price to be holding company stock, I know what to do.

The buyback is a return of cash to the shareholders, cash that will no longer participate in company operations; it's just done in a way where the position of inaction is to reinvest and increase one's share, and a way that happens to be more tax-efficient. So the term "yield" is justified -- this year ABC returned X% of its cash to shareholders. If you think cash returns (dividends or buybacks) are an indication of the "health" or "soundness" of the company's business (I don't, personally), then you care just as much. And if you're a taxable investor, you should be grateful that the company chose the method that's in your best interest. It's no accident that we see an increase in buyback as portion of yield in the current tax regime and with the current, extremely low, cost of selling shares.
User avatar
siamond
Posts: 5888
Joined: Mon May 28, 2012 5:50 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by siamond »

I've seen this kind of strange math (combining dividends with net buybacks) for a while now, this left me cold to begin with, this continues to leave me cold. Buybacks affect prices, but so are many other factors, and there is just no way to distinguish one from the other.

I think this is an ill-fated way to try to stick to the dividends yields of the past as a 'pseudo-constant', and not acknowledge that things are changing, some distributed earnings (dividends) shifting to undistributed earnings and subsequent price changes (whether it comes from immediate buybacks, longer term effects of reinvestment, or whatever mood the market is in).

Yes this historical shift makes various past rules of thumb become less and less applicable (the CAPE 'mean' being the prime one), but that's just because those rules of thumb were flawed. Better seek better rules of thumb, that work irrespective of the way earnings are going (distributed, buybacks or reinvested).
User avatar
rmelvey
Posts: 826
Joined: Sat Sep 18, 2010 5:17 pm
Contact:

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by rmelvey »

Technically the buy back yield is part of the dividend per share growth rate. So, even if earnings never grew again, the dividend per share would grow at the rate of the buy back yield.

I found the chart encouraging. Thanks for posting! 8-)
TJSI
Posts: 399
Joined: Thu Jan 27, 2011 4:03 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by TJSI »

Ogd,

You conveniently forget that when a company does a buyback they take from stockholders something of value. The stock surrendered does have value. And in some cases it is worth more than the cash received. The value of stock removed from stockholders needs to be netted out of any calculation of the effect of buybacks.

I recommend you read what Warren Buffett wrote about buybacks. He nailed it.

TJSI
TJSI
Posts: 399
Joined: Thu Jan 27, 2011 4:03 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by TJSI »

Simplegift,

You wrote "It seems there's a return to shareholders." There is indeed a return to shareholders but the calculation of it is difficult. One set of shareholders wins while the others lose. That is why it would be great if those who write about buybacks clearly state which stockholders they are referring to.

A buyback may produce a positive return or negative return for holding shareholders. The large amount of cash spent on buybacks may be good or it might just be a bad mistake. Someone said that eventually the market is a weighing machine, so we will see.

TJSI
Bfwolf
Posts: 2084
Joined: Thu Oct 14, 2010 11:19 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by Bfwolf »

TJSI wrote:Ogd,

You conveniently forget that when a company does a buyback they take from stockholders something of value. The stock surrendered does have value. And in some cases it is worth more than the cash received. The value of stock removed from stockholders needs to be netted out of any calculation of the effect of buybacks.

I recommend you read what Warren Buffett wrote about buybacks. He nailed it.

TJSI
When a company does a buyback, it takes shares worth some amount of market capitalization from its shareholders in return for the cash. When a company does an equivalent dividend, it takes an equivalent amount of market capitalization from its shareholders in return for the cash--this is reflected in the stock price drop equal to the amount of the dividend paid.

Simplegift, I think your analysis is correct. I don't think it answers a particularly interesting question....these yield %s don't matter much to me. But to somebody who cares about them, I think the buyback yield needs to be added to the dividend yield.
garlandwhizzer
Posts: 3473
Joined: Fri Aug 06, 2010 3:42 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by garlandwhizzer »

Great post, Simplegift, as usual. I find it somewhat reassuring and at times like this reassurance is nice.

Dividends put money directly in the pockets of investors. Buybacks work by reducing share number and therefore (as long as profits are constant) having a positive effect on PE ratios and tending to elevate share price. Those benefits are only achieved on paper until shares are sold. One factor that should be mentioned with regard to buybacks is the effect of changing PE ratios on share price. PE ratios are generous by historical standards now but of course relative to the very low yields on bonds they seem to me not outrageous. In the future they may either rise higher which would tend to magnify the positives of share buybacks or decline which would tend to reduce the effects of share buybacks. If share buybacks increase profits per share by 10% but PE decrease by 15%, the net change for investors is negative. Giving money directly to investors now with dividends may be less efficient from a tax point of view but it takes the risk and uncertainty out of changes in PE ratios over time. Also the uncertainty of future corporate profits. It's a risk/reward tradeoff like everything else. Hence some investors prefer dividends over share buybacks (a real live bird in the hand) while others prefer buybacks (no taxes generated until sale of shares, possibly lower rate on LTCG when shares are sold). However you look at it, judged by the formula of share buyback plus dividends it seems to me that current stock prices, while generous, are not presently in bubble territory relative to 10 year Treasuries yielding 1.6%. If bond yields increase dramatically (which does not appear on the horizon now) PE ratios may well decline reducing the multiplier effect of share buyback's delayed benefits.

Garland Whizzer
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

TJSI wrote:There is indeed a return to shareholders but the calculation of it is difficult...(snip)...A buyback may produce a positive return or negative return for holding shareholders.
Totally agree that the direct benefit of net share buybacks to shareholders is difficult to calculate exactly — there are several moving variables. But this doesn’t mean that we can ignore the benefits entirely. For example, for buy-and-hold index investors, there’s been a significant decrease in the S&P 500 Index divisor since net share buybacks became so prominent in about 2005:
  • Index Value = Sum (Market cap all S&P 500 stocks) / Divisor
    Image
    Source: Yardeni
And this decrease in the S&P 500 Index divisor appears directly related (inversely) to the buyback yield:
  • Image
    Note: The scale for the buyback yield at right is inverted.
    Source: JPMorgan
Thus buy-and-hold index investors, due to net share buybacks, are experiencing index price appreciation and are owning a larger proportional share of the S&P 500 companies’ earnings. Though they aren't getting this return in the form of a cash dividend check, they are getting a tangible return that needs to be considered, I believe.
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by ogd »

garlandwhizzer wrote:Dividends put money directly in the pockets of investors. Buybacks work by reducing share number and therefore (as long as profits are constant) having a positive effect on PE ratios and tending to elevate share price.
But -- buybacks too put company cash in the pockets of some investors. It's actual cash that leaves the company coffers, not a paper maneuver. The others increase their share in the company. This is the same as the distinction between investors who take the dividend and those who reinvest it, except that the latter group is forced to pay taxes. And the other difference is that the position of inaction is to reinvest the "dividend" (i.e. the cash paid out). But the investor is always free to switch to the other group by participating or selling.

From this perspective, it makes complete sense to add up the two. This much cash left the company and was paid out to investors. It's not artificial and it doesn't depend on the future of the shares -- that's a separate decision that an investor should make at all times, whether or not they want to be holding stock at this price.
Da5id
Posts: 4949
Joined: Fri Feb 26, 2016 8:20 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by Da5id »

Doesn't change my asset allocation, but seems like a useful graph and interesting take. Thanks for posting it.
chx
Posts: 429
Joined: Sat Jan 16, 2016 10:52 am

Re: Dividend & Buyback Yields, S&P 500, 1982-2015

Post by chx »

Simplegift wrote:
triceratop wrote:Where is the inflation protection?
Sorry not to be more explicit with the assumption that real dividend growth rates (and real net buyback rates) would be positive going forward. Obviously no future guarantees, but dividend growth rates have consistently outpaced inflation in the U.S., especially since the 1940s:
  • Image
    Note: These are growth rates of dividend dollars, not dividend yields.
    Source: WisdomTree
chx wrote:How do you get a negative amount of buybacks?
On a net basis, more stock was issued for the year than was repurchased. Here’s the equation from the OP:
  • Net Buyback Yield = (Trailing 12 month stock repurchases - stock issuances) / Market capitalization
Thank you for the explanations.
User avatar
triceratop
Posts: 5838
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by triceratop »

Thanks Simplegift, for the followup. I concur with you and garlandwhizzer now.

Thank you for the informative post, too.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

In a PM, the question was raised whether U.S. small-cap stocks have been repurchasing their shares at the same rate as large-cap stocks. After doing a little Google searching, it appears that small-cap companies have actually been net issuers of stock over the past 15 years (chart below), with an average net buyback yield of -2.3% since 2001:
  • Image
    Note: Large caps are the 500 largest U.S. companies, while small caps are the smallest 2,000.
    Source: SpiderFinance
According to the article, nearly all of this net new share issuance came from the higher-priced, small growth companies (with an average net buyback yield of -5.5%) — and the lower-priced, small value companies issued far fewer net shares. This may help explain why small growth companies tend to underperform, if they have to overcome an average dilution rate of over 5% per year in driving their earnings-per-share growth.
Last edited by SimpleGift on Tue Jun 21, 2016 4:19 pm, edited 1 time in total.
User avatar
gordoni2
Posts: 213
Joined: Wed Aug 15, 2007 12:20 am
Contact:

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by gordoni2 »

In a simplistic model: shares fairly valued and company earning a uniform rate of return on its capital, buybacks appear to have no effect for buy and hold shareholders of a company. To see this imagine a company worth $100b that has 100b shares outstanding, earns 5%, and pays out 50% of its earnings in dividends. It performs a $50b buyback of 50b shares. Because selling shareholders have been paid off, the buy and hold shareholders now each own a larger stake in the company. Unfortunately, because cash has been paid out, that company is worth $50b less than it was before. The buy and hold shareholders now have 50b shares in a $50b company that earns 5%. In other words EPS and DPS is unchanged.

This is a simplistic model, but it captures the world to first order. Second order effects are if shares are not fairly valued, or if the rate of return on capital is non-uniform, but the effect of share buybacks is still only the difference between that and the simplistic model.

I believe the S&P 500 index works in a similar fashion.
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

gordoni2 wrote:In a simplistic model: shares fairly valued and company earning a uniform rate of return on its capital, buybacks appear to have no effect for buy and hold shareholders of a company.
Gordoni2, I appreciate you pressing your case here, as it’s caused me to drill down deeper into the impacts of net share buybacks on buy-and-hold shareholders.

Given the assumptions in your simplistic model, I believe you are correct in theory, not only for your example company, but for the S&P 500 Index as a whole. However, in the real world, when a company’s after-tax interest costs are much lower than its earnings yield (as is so widespread today), net share buybacks do increase EPS, as each dollar spent buys more earnings than the cost of funding the buyback. For the mathematical details, see page 10 of this Credit Suisse white paper.

With interest rates near historical lows today, and forward P/E multiples (earnings yields) about average, net share buybacks are boosting the earnings-per-share of S&P 500 companies. Of course, this may not hold in the future. If interest rates trend higher or companies trade on lower earnings yields, buybacks could start to become a negative for EPS ratios — as best I understand it.
TJSI
Posts: 399
Joined: Thu Jan 27, 2011 4:03 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by TJSI »

Thank you Simplegift for that Credit Swiss reference. It is by far the best discussion of buybacks I have seen.

They recognize the difference between holding and selling shareholders. And from their conclusion:

" Buybacks only benefit continuing shareholders when management executes them when the stock is undervalued."

Like Mr. Buffett they got it right.

TJSI
qwertyjazz
Posts: 1999
Joined: Tue Feb 23, 2016 4:24 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by qwertyjazz »

I buy total stock market thinking I am more diversified than S&P 500. Given how close they track I know I am probably deluding myself. But just maybe there is a small growth stock in there over the long term will payoff. But I am probably deluding myself even more because of stock dilution by the time Apple becomes APPLE of great.
Interesting
Lobster
Posts: 285
Joined: Sat Mar 26, 2016 3:11 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by Lobster »

TJSI wrote:Ogd,

You conveniently forget that when a company does a buyback they take from stockholders something of value. The stock surrendered does have value. And in some cases it is worth more than the cash received. The value of stock removed from stockholders needs to be netted out of any calculation of the effect of buybacks.

I recommend you read what Warren Buffett wrote about buybacks. He nailed it.

TJSI
The 24/7 Wall St article that hit as the first response for Warren Buffett on buybacks suggests that Buffett is a fan of buybacks because it represents an increase in intrinsic value for shareholders. This corroborates ogd's post per my understanding.
Submit to the relentless rules of humble arithmetic and avoid the tyranny of compounding costs.
TJSI
Posts: 399
Joined: Thu Jan 27, 2011 4:03 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by TJSI »

Lobster,

Mr. Buffett has commented several times on buybacks and I believe most recently on IBM and their repurchase program. He believes IBM did the right thing. He is a very firm believer in buying stocks below intrinsic value. He has never said that buybacks are good in general. In one of letters to shareholders he commented in some detail. Do a search on shareholder letters, Buffet, and buybacks and you will probably find his comments.

He has said that some buybacks are good and some foolish. Selling shareholders don't just receive cash. They exchange their shares which may in some cases is worth more than the cash received. This is a case where a buyback had a negative yield for the selling shareholder.

TJSI
Lobster
Posts: 285
Joined: Sat Mar 26, 2016 3:11 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by Lobster »

TJSI wrote:Lobster,

Mr. Buffett has commented several times on buybacks and I believe most recently on IBM and their repurchase program. He believes IBM did the right thing. He is a very firm believer in buying stocks below intrinsic value. He has never said that buybacks are good in general. In one of letters to shareholders he commented in some detail. Do a search on shareholder letters, Buffet, and buybacks and you will probably find his comments.

He has said that some buybacks are good and some foolish. Selling shareholders don't just receive cash. They exchange their shares which may in some cases is worth more than the cash received. This is a case where a buyback had a negative yield for the selling shareholder.

TJSI
TJSI,
Found the excerpt from the shareholder letter you mentioned. Absolutely agree, in fact this perspective holds for any purchase and doesn't strike me as unique to stock buybacks at all. Absolutely adheres to basic investing principles espoused by Graham and other value oriented investors.

Cheers :sharebeer
Submit to the relentless rules of humble arithmetic and avoid the tyranny of compounding costs.
User avatar
abuss368
Posts: 27002
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
Contact:

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by abuss368 »

Interesting chart.
John C. Bogle: “Simplicity is the master key to financial success."
3504PIR
Posts: 978
Joined: Mon Jul 26, 2010 2:46 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by 3504PIR »

siamond wrote:I've seen this kind of strange math (combining dividends with net buybacks) for a while now, this left me cold to begin with, this continues to leave me cold. Buybacks affect prices, but so are many other factors, and there is just no way to distinguish one from the other.

I think this is an ill-fated way to try to stick to the dividends yields of the past as a 'pseudo-constant', and not acknowledge that things are changing, some distributed earnings (dividends) shifting to undistributed earnings and subsequent price changes (whether it comes from immediate buybacks, longer term effects of reinvestment, or whatever mood the market is in).

Yes this historical shift makes various past rules of thumb become less and less applicable (the CAPE 'mean' being the prime one), but that's just because those rules of thumb were flawed. Better seek better rules of thumb, that work irrespective of the way earnings are going (distributed, buybacks or reinvested).
Geepers, has anyone in your life ever mentioned a tendency toward dramatics? Your excellent writing style aside, I fully agree with you.
3504PIR
Posts: 978
Joined: Mon Jul 26, 2010 2:46 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by 3504PIR »

Per the chart, I guess ye olde buyback theory isn't all it's cracked up to be. Why should it? It theoretically makes sense that a company buying up shares would add definitive value, but obviously market effects sometimes don't play along, according to the chart anyway. A bird in the hand...
Da5id
Posts: 4949
Joined: Fri Feb 26, 2016 8:20 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by Da5id »

3504PIR wrote:Per the chart, I guess ye olde buyback theory isn't all it's cracked up to be. Why should it? It theoretically makes sense that a company buying up shares would add definitive value, but obviously market effects sometimes don't play along, according to the chart anyway. A bird in the hand...
Per the above, I think the chart clearly is a Rorschach test.
IlliniDave
Posts: 2388
Joined: Fri May 17, 2013 7:09 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by IlliniDave »

siamond wrote:I've seen this kind of strange math (combining dividends with net buybacks) for a while now, this left me cold to begin with, this continues to leave me cold. Buybacks affect prices, but so are many other factors, and there is just no way to distinguish one from the other.

I think this is an ill-fated way to try to stick to the dividends yields of the past as a 'pseudo-constant', and not acknowledge that things are changing, some distributed earnings (dividends) shifting to undistributed earnings and subsequent price changes (whether it comes from immediate buybacks, longer term effects of reinvestment, or whatever mood the market is in).

Yes this historical shift makes various past rules of thumb become less and less applicable (the CAPE 'mean' being the prime one), but that's just because those rules of thumb were flawed. Better seek better rules of thumb, that work irrespective of the way earnings are going (distributed, buybacks or reinvested).
To me it doesn't seem any stranger than adding dividend yield to share price appreciation and computing total return. All this is doing is adding dividend yield to the portion of share price appreciation attributable to buybacks to isolate what non-reinvested profits are contributing to investment returns. I can't speak to its profundity, but it is interesting. My question would be: how reliable are the figures for return due to buyback and what is the efficiency (i.e., what percent of the buybacks are offset by new issues or just handed to executives as compensation). Maybe all that is well accounted for.
Don't do something. Just stand there!
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

IlliniDave wrote:My question would be: how reliable are the figures for return due to buyback and what is the efficiency (i.e., what percent of the buybacks are offset by new issues or just handed to executives as compensation). Maybe all that is well accounted for.
In the data on share buybacks, one has to distinguish between gross buyback yield (does not account for new issues) and net buyback yield (as used in the OP). As I understand it, new issuance data is harder to come by, so S&P only publishes gross share buybacks in their annual reporting — and as a result, many researchers use this gross data alone. However, a few (like Meb Faber and WisdomTree) go to the trouble of researching and calculating the net share buyback figures, which seem the most accurate to use in determining the final impact on shareholders.

Overall, my sense is the net share buyback data is certainly squishier than the dividend data, but not too much so.
IlliniDave
Posts: 2388
Joined: Fri May 17, 2013 7:09 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by IlliniDave »

Simplegift wrote:
IlliniDave wrote:My question would be: how reliable are the figures for return due to buyback and what is the efficiency (i.e., what percent of the buybacks are offset by new issues or just handed to executives as compensation). Maybe all that is well accounted for.
In the data on share buybacks, one has to distinguish between gross buyback yield (does not account for new issues) and net buyback yield (as used in the OP). As I understand it, new issuance data is harder to come by, so S&P only publishes gross share buybacks in their annual reporting — and as a result, many researchers use this alone. However, a few (like Meb Faber and WisdomTree) go to the trouble of researching and calculating the net share buyback figures, which seem the most accurate to use in determining the final impact on shareholders.

In all, my sense is the net share buyback data is certainly a bit squishier than the dividend data, but not too much so.
Thanks. I guess what I was thinking is along the lines of, if a company takes $1/share it could have given as a dividend, buys back stock of which they retire $.50 worth and hand out/reissue $0.50 worth, that's not so good a deal for any average shareholders. So seeing the net as in the OP is good, but it would also be interesting to see the difference between the net and gross that, from a common shareholders perspective, sort of vanishes. If the net/gross ratio is high, then buybacks could be a good deal for taxable shareholders. Sounds like all the data is out there, I guess I'd just need to aggregate it. But it's one of those things I suspect I don't want to know the answer to. No use getting cranky about stuff I'm powerless to affect.
Don't do something. Just stand there!
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

IlliniDave wrote:So seeing the net as in the OP is good, but it would also be interesting to see the difference between the net and gross that, from a common shareholders perspective, sort of vanishes. If the net/gross ratio is high, then buybacks could be a good deal for taxable shareholders.
Not exactly sure what you're getting at here, but this chart shows the gross and net buyback yields from 1987-2015:
  • Image
    Note: This data is for the 750 largest-cap U.S. listed companies.
    Source: IFG
dkturner
Posts: 1763
Joined: Sun Feb 25, 2007 7:58 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by dkturner »

Lobster wrote:
TJSI wrote:Lobster,

Mr. Buffett has commented several times on buybacks and I believe most recently on IBM and their repurchase program. He believes IBM did the right thing. He is a very firm believer in buying stocks below intrinsic value. He has never said that buybacks are good in general. In one of letters to shareholders he commented in some detail. Do a search on shareholder letters, Buffet, and buybacks and you will probably find his comments.

He has said that some buybacks are good and some foolish. Selling shareholders don't just receive cash. They exchange their shares which may in some cases is worth more than the cash received. This is a case where a buyback had a negative yield for the selling shareholder.

TJSI
TJSI,
Found the excerpt from the shareholder letter you mentioned. Absolutely agree, in fact this perspective holds for any purchase and doesn't strike me as unique to stock buybacks at all. Absolutely adheres to basic investing principles espoused by Graham and other value oriented investors.

Cheers :sharebeer
This discussion of buy backs vs. cash dividends raises an interesting question in my mind. From a strategic standpoint wouldn't it make sense for a company to buy back its shares when its stock price is low, but pay substantial cash dividends to shareholders when it's stock price is high? A company would be taking shares off the table when it can buy more of them for the same number of $ but return cash to shareholders when it's shares may be "overvalued". We frequently stop reinvesting our dividends when stock prices are trading at well above average P/E ratios and purchase bonds with the proceeds.
TJSI
Posts: 399
Joined: Thu Jan 27, 2011 4:03 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by TJSI »

dkturner,

You have raised a good point. Buybacks could be an excellent tool to damp the volatility of stock price. They could step in if prices dropped significantly below their estimate of intrinsic value. And I am sure than some companies do make a good faith estimate of intrinsic value. But there is also evidence that companies buy high and stop buying when things go bad. The exact opposite of what should happen.

TJSI
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by ogd »

TJSI wrote:,
You conveniently forget that when a company does a buyback they take from stockholders something of value. The stock surrendered does have value. And in some cases it is worth more than the cash received. The value of stock removed from stockholders needs to be netted out of any calculation of the effect of buybacks.

I recommend you read what Warren Buffett wrote about buybacks. He nailed it.
I am not 100% sure what you mean here.

I think you mean that the stock bought may be under- or over-valued, benefitting ongoing or outgoing shareholders respectively. This is the same notion that I see in other posts:
TJSI wrote: They could step in if prices dropped significantly below their estimate of intrinsic value. And I am sure than some companies do make a good faith estimate of intrinsic value. But there is also evidence that companies buy high and stop buying when things go bad.
TJSI wrote: He is a very firm believer in buying stocks below intrinsic value.
Credit Suisse wrote: Buybacks only benefit continuing shareholders when management executes them when the stock is undervalued."
I think asking management to perform or withhold buybacks based on price is fundamentally misguided; in the hands of the analysts, it's a tool for scapegoating management for the analyst's poor investment decisions. Here's why: it's the investor's decision to buy or sell stock at the current price, and it should be none of the management's business. If one sees management buying below market value ("the stock surrendered is more than cash received"), they are free to buy more with the cash received. If one sees management buying above, they are free to sell. Too many pundits blame management for "buying high" during a period and at a price where they -- the pundit -- had a buy or hold on the stock. If the stock was overpriced at the time (and not just with 20/20 hindsight) -- sell the darn stock, it's your job.

Like I said in my first post, it's as if a dividend reinvestor blamed management for paying high dividends when the price was high, making them reinvest unfavorably. Except that the politics of dividends are well established, whereas buybacks are relatively new and tend to confuse people with price considerations.

From my perspective, management should pay absolutely no attention to the current price when they buy back shares. They are not investing in those shares, they are removing cash from the company books. It is my job, as an investor, or the market's, to decide on the value of those shares. Management tends to have rose-colored glasses and think of their stock as undervalued; and on the other hand be afraid of being accused of buying high. The harm from taking price into account is not that they might be wrong on the price; it's that it distracts the management from their real work in all of this, to consider whether or not the cash is of use inside the company, e.g. if it will help them avoid expensive debt in the future or survive a debt crisis or buy a very attractive expansion property, and so forth. That's all I want them to think about. The price is my responsibility. Again, just like whether or not to reinvest dividends.

The example of Warren Buffett muddies the waters because he is an investor-manager. When you buy his company-cum-mutual-fund, you do want him to make that sort of decisions; that's to a large degree what you pay for. I'd note that he hasn't made them exceedingly well as of late, but yes he has earned the right to do this. Any other management has no business making decisions about price.

I like buybacks as a means of returning cash primarily because of taxes, but I also like that they (ideally) remove the politics surrounding dividends from the considerations about cash payouts, because the custom of regular and mildly increasing payments doesn't exist. So when they see an outstanding opportunity for that cash -- like a factory site at a liquidation price -- they should be free to deploy it without worrying they'll be tarred and feathered for lowering or skipping a dividend. But if they instead have to worry about "not buying under-valued stock" instead, that's no good and we've just replaced one type of politics with another. Again -- "under-priced expansion site we need" is their job; you have to trust them to do that sort of thing well, but if you're in the company you'd better do. "Low stock price" is not their job, it's yours.
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by ogd »

dkturner wrote: A company would be taking shares off the table when it can buy more of them for the same number of $ but return cash to shareholders when it's shares may be "overvalued". We frequently stop reinvesting our dividends when stock prices are trading at well above average P/E ratios and purchase bonds with the proceeds.
The equivalent for a buyback is to participate in the buyback (or replicate the sale in the market) when you think the price is high, and not participate (increasing your share in the company) when you think it's low. It's like withdrawing a dividend vs reinvesting it. You are on the right track -- how many shares you want to hold is your responsibility, not the management's. Or, if you're a passive investor, it's the market's responsibility to determine these things, not management's.
IlliniDave
Posts: 2388
Joined: Fri May 17, 2013 7:09 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by IlliniDave »

Simplegift wrote:
IlliniDave wrote:So seeing the net as in the OP is good, but it would also be interesting to see the difference between the net and gross that, from a common shareholders perspective, sort of vanishes. If the net/gross ratio is high, then buybacks could be a good deal for taxable shareholders.
Not exactly sure what you're getting at here, but this chart shows the gross and net buyback yields from 1987-2015:
What the chart says is most recently, in the aggregate, for every $1 dollar of gross buyback, I see $0.75 in benefit (2.1/2.8=75%). Even in a taxable account, I'd rather get the $1 as a dividend. That's all I'm saying. If it was 90% I'd prefer the buyback when looking at taxable, but the dividend in tax deferred. At 100% it's a wash in tax-favored and a nice way to reinvest in the stock without double taxation in a taxable account.
Don't do something. Just stand there!
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by ogd »

I would add that a dividend and a buyback of the same magnitude have the same effect on total return numbers -- they decrease the company's cash holding while increasing your $10k's (as in "growth of $10K") share of the company. Total return is with dividends reinvested. So it makes perfect sense to add together dividends and buybacks as a combined yield; it's only when you think "per share" that the waters are muddied with these EPS questions, because the number of shares you hold is indeed different in these two scenarios.
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by ogd »

IlliniDave wrote: What the chart says is most recently, in the aggregate, for every $1 dollar of gross buyback, I see $0.75 in benefit (2.1/2.8=75%). Even in a taxable account, I'd rather get the $1 as a dividend. That's all I'm saying. If it was 90% I'd prefer the buyback when looking at taxable, but the dividend in tax deferred. At 100% it's a wash in tax-favored and a nice way to reinvest in the stock without double taxation in a taxable account.
Why do you think share issuance would have been $0 if dividends had been used instead? It's hard to verify a hypothetical, but what if it would have been exactly $0.25? That 0.25 difference did have a positive impact on the company's business; worst case, it replaced compensation paid in another form because it's naive to think today's executives would just meekly accept lower pay if their options aren't returning anything. Best case, it allowed them to invest and have the returns they've had since.
ralph124cf
Posts: 2866
Joined: Tue Apr 01, 2014 11:41 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by ralph124cf »

I'm going to stick my neck out here, and say that there is no such thing as buyback yield.

Suppose that you have a company with $1M in the bank, no other assets, no liabilities, and only earnings from bank interest. It currently pays out all earnings as dividends, currently yielding 3%.

Further, this company has exactly one million shares outstanding, and these shares are currently priced at $1, exactly equal to shareholder equity/share.

If this company buys back 500,000 shares for $1 each, it decreases its cash in the bank to $500K, and leaves shareholder equity/share unchanged. In a rational world, the price per share should remain unchanged, as the EPS potential from cash in the bank is unchanged. The dividend yield would be expected to stay at 3%

This seems to be a very neutral action from a shareholder point of view.

Ralph
Bfwolf
Posts: 2084
Joined: Thu Oct 14, 2010 11:19 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by Bfwolf »

ralph124cf wrote:I'm going to stick my neck out here, and say that there is no such thing as buyback yield.

Suppose that you have a company with $1M in the bank, no other assets, no liabilities, and only earnings from bank interest. It currently pays out all earnings as dividends, currently yielding 3%.

Further, this company has exactly one million shares outstanding, and these shares are currently priced at $1, exactly equal to shareholder equity/share.

If this company buys back 500,000 shares for $1 each, it decreases its cash in the bank to $500K, and leaves shareholder equity/share unchanged. In a rational world, the price per share should remain unchanged, as the EPS potential from cash in the bank is unchanged. The dividend yield would be expected to stay at 3%

This seems to be a very neutral action from a shareholder point of view.

Ralph
It is a neutral action. Just like a dividend is. Neither a buyback nor a dividend creates wealth out of thin air. In your buyback example above, half of the company market cap is removed as are half the shareholders, leaving the share price and dividend yield unchanged. $500,000 has been returned as cash to shareholders. If the company had instead paid a special dividend of $0.50 per share, half of the company market cap would be removed, the total number of shares and dividend yield would remain unchanged, and the share price would halve. $500,000 has been returned as cash to shareholders. The market cap, amount of cash returned to shareholders, and expected future absolute earnings are the same either way, it's just the number of shares outstanding that is different.

If instead of a 3% dividend every year, this company did a 3% buyback every year, the results would be the same, just with a different number of outstanding shares. If the point of measuring dividend yield is to understand how much of a company's value is being returned to shareholders as cash, then buyback yield is an equally valid measurement as it accomplishes the same thing.
IlliniDave
Posts: 2388
Joined: Fri May 17, 2013 7:09 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by IlliniDave »

ogd wrote: Why do you think share issuance would have been $0 if dividends had been used instead? It's hard to verify a hypothetical, but what if it would have been exactly $0.25? That 0.25 difference did have a positive impact on the company's business; worst case, it replaced compensation paid in another form because it's naive to think today's executives would just meekly accept lower pay if their options aren't returning anything. Best case, it allowed them to invest and have the returns they've had since.
I was speaking in terms of my hypothetical preferences only. As a current shareholder, I'd always prefer no or minimal new share issuance all things being equal.

I think what we're seeing is the exact opposite of executives in US companies meekly accepting lower pay. My guess would be that their compensation is growing at a faster clip than the earnings of the companies they manage.

In the end issuances and retained earnings amount to the same thing, except that issuances happen in relative secret. It all depends on how well the company puts them to use.
Don't do something. Just stand there!
User avatar
Topic Author
SimpleGift
Posts: 4324
Joined: Tue Feb 08, 2011 3:45 pm

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by SimpleGift »

ogd wrote:Total return is with dividends reinvested. So it makes perfect sense to add together dividends and buybacks as a combined yield; it's only when you think "per share" that the waters are muddied with these EPS questions, because the number of shares you hold is indeed different in these two scenarios.
:thumbsup It's the "per share" data where many folks get confused about net share buybacks — and you've stated it very nicely.
Da5id
Posts: 4949
Joined: Fri Feb 26, 2016 8:20 am

Re: Dividend and Buyback Yields, S&P 500, 1982-2015

Post by Da5id »

I do wonder if there is a fundamental difference between buying back shares, as discussed here, and buying something else. e.g. if the company uses the cash to buy and hold S&P 500 ETF, or buy and hold shares in some other company (as Buffett does)? Seems like largely the same thing to me, in which case I am a bit confused by those who seem to argue that value has been vaporized by the buyback. Anyway, very interesting data in this thread.
Post Reply