you are viewing a single comment's thread.view the rest of the posts
woodstein2002 writes (msg# 171348):
"The day BRKa starts paying dividends is the day I sell."
Great! You'll probably be selling to me! I hope all you "dividend haters" drive the price into the ground.
Let's see, what WOULD Berkshire be worth today if it DID pay a dividend?
Well, last year Warren said he had 100 million$ pretax a week flowing in from just the operating companies. That would be about 65 mil$/week after taxes (-35%) or 3,380 mil$ per year (*52 weeks) or $2,202 per "A" (/1.534934 million "A" equivalents on 3/31/2003). So last year's dividend could have been ~$2,200 per "A" if WEB had just written checks instead of retaining that cash for reinvestment purposes.
Berkshire is a AAA credit, so let's discount the dividend by the yield on a AAA corporate bond. What's a AAA corporate bond paying?
30-Year AAA Corporate Bond yield = 5.44% on 5/16/2003.
And if WEB isn't retaining any of that free cash flow for reinvestment then real growth is probably going to be zip. But wait, there is always inflationary growth. Consider a small town barbershop. The barber has no real growth either. If he gave 4,000 haircuts last year he probably serviced ~4,000 customers/year a decade ago and ten years from now he'll probably still be averaging ~4,000 haircuts/year. But he can raise his price in line with inflation.
Let's use the compound average growth rate in the CPI-U over the last five years to estimate the nominal growth rate.
5-Yr CAGR (CPI-U) = 2.49% (from 162.5 on 4/30/1998 to 183.8 on 4/30/2003).
Now I need my trusty 20 year old hp12c calculator to calculate the present value of 100 years of dividends growing at 2.49% per year and discounted by 5.44% per year.
[f] [CLEAR FIN] ... clears the financial storage registers.
100 [n] ... 100 years of dividends.
1.0249 [ENTER] 1.0544 [DELTA%] [i] ... stores 2.8783, the growth adjusted interest rate.
2200 [PMT] ... last year's dividend.
[PV] ===> -71,957 ... calculates the present value.
So the free cash flow from the operating companies ALONE are worth ~72,000 $ per "A".
Yesterday Berkshire closed at $73,550 (3/16/2003).
"Clearly dividends determine value" - John Burr Williams (see msg# 170913).
That would really make my day, a sell off on the day dividends commence.
Yes, woodstein2002, I know what you really meant. Sorry to use you as a "foil" to make a point. As long as retained earnings grow at a rate greater than the discount rate reinvestment makes sense. But still, I would like to see Berkshire pay a dividend, even if it is a tiny one (with a correspondingly large growth rate). Come on, Charlie's WESCO does!
PS: hp has a nice write-up on calculating the present and future values of an annuity that increases at a constant rate at equal intervals here:
re:"As long as retained earnings grow at a rate greater than the discount rate reinvestment makes sense. But still, I would like to see Berkshire pay a dividend, even if it is a tiny one (with a correspondingly large growth rate). Come on, Charlie's WESCO does!"
Amen! But I would add...so long as $1 of retained earnings provides at least $1 in MARKET value (as mentioned in principle #9 of the Brk owners bible) revinvestment makes sense. This hasn't been the case lately. Therefore a dividend must be imminent. Personally I think Buffett's own ego precludes any dividend payment. It would be an admission of failure and therefore I think we will see disposition of the yardstick before we see disposition of Buffett.
Buffett's ego. Did you ever wonder why he never split the stock?? Kind of like a little kid screaming: " I want to be the most expensive stock on the board".
A real problem there, in my opinion.
Hurting the shareholders, too.
why use a 100 years. just make it a perpetually.
The math is easier and the difference is small because the PV of the years past a hundred are really not worth a lot.
By the way that was a good post, and i am just trying to add a different look.
"why use a 100 years. just make it a perpetually."
You're right, of course.
I just wanted to do one post without any algebra, using a tool accepted within the financial community.
If I had "simulated" infinity by keying in 1,000 years for "n" surely someone would have complained. You know the drill. "How many companies last a hundred years? Two hundred years? And you're discounting a thousand years worth of dividends to justify the price?"
Enjoyed your post, thanks. Can your 20 yr old HP calc handle growth rates too? I wonder what the NPV is assuming various growth rates. For a boring co., BRK certainly knows how to grow.
Thanks for the kind remarks.
I did do a two stage discounted dividend valuation using that calculator in my very first post, # 147374. I'm still not ashamed of that valuation. I probably turned a lot of people off by doing everything in the real (inflation adjusted) world and by using too much algebra. I was also afraid that I would have endless arguments over how long the growth period would last, thus my "goofy" 4.75 years based on the SQUARZ. Oh, well, live and learn. Anything more complicated than that requires a spreadsheet which is actually more comforting because it is easier to spot mistakes. Still, I do love that old machine, I really miss the company (and the quality products it used to make) that Bill (Hewlett) & Dave (Packard) ran.