Skip to main content
Message Font: Serif | Sans-Serif
 
No. of Recommendations: 5
" Berkshire has repurchased more shares since yearend and is likely to further reduce its share count in the future. Apple has publicly stated an intention to repurchase its shares as well. As these reductions occur, Berkshire shareholders will not only own a greater interest in our insurance group and in BNSF and BHE, but will also find their indirect ownership of Apple increasing as well."
Print the post Back To Top
No. of Recommendations: 14
I make it another $4.4b since year end,

Share count has shrunk 5.72% in the 12 months to Feb 16.

Hard to complain.
Print the post Back To Top
No. of Recommendations: 1
Man of his word! Loving it!!!
Print the post Back To Top
No. of Recommendations: 1
<<Last year we demonstrated our enthusiasm for Berkshire’s spread of properties by repurchasing the
equivalent of 80,998 “A” shares, spending $24.7 billion in the process.>>

The average purchase price is $304,945 per A share, or $203 per B share. The current price is 20% higher.
Print the post Back To Top
No. of Recommendations: 2
Just to be clear.... this is the first time that WEB even came close to ever saying buybacks will be sustained

Berkshire has repurchased more shares since yearend and is likely to further reduce its share count in the future

Assuming this is true, I will be a buyer on dips.
Print the post Back To Top
No. of Recommendations: 30
The average purchase price is $304,945 per A share, or $203 per B share. The current price is 20% higher.

For a longer term perspective, buybacks restarted August 7 2018.
Since then, to February 16 this year, they've bought 113991.3 A-share equivalents.
The net share count reduction has been -6.905%
The total cost has been $35.4637 billion.
Average price paid has been $311,108 per share.
The weighted average P/B paid has been 1.214.
(that's using the average of the book value at start and end of each quarter as a proxy for the book value on any purchase date in that quarter)

The most notable thing is that there have really been two stretches of buybacks since 2018.
A relatively steady rate of ~1330 A shares per month August 2018 through mid March 2020.
Then a brief pause of nothing in April, followed by a new level of commitment at seven times the rate:
A relatively steady rate of ~9260 A shares per month May 2020 through February 2021.
That new rate, if sustained, would drop the share count by 7.25%/year.

Across those two stretches, the weighted average purchase date was 2020-06-18.
(weighted by dollars of outlay)
So there have been "only" $21bn dollar-years of commitment to this endeavour.
The shares they've purchased might be worth (say) $11bn more than they cost based on value not price, which means it has been a pretty good return.
The IRR is over 50%, looked at that way.
This happy result is because they didn't step up the rate till the valuation levels were more compelling.

Jim
Print the post Back To Top
No. of Recommendations: 1
This happy result is because they didn't step up the rate till the valuation levels were more compelling

I am not sure that is the case.... but more importantly this is pretty flawed approach in my view. WEB should believe that Berkshire will be creating value in the future and today's price is going to be cheaper than his cost of the purchase, i.e., let us say Berkshire borrowed $25 B @ 2% (far higher than what they will actually pay), and invested in buybacks now, and pay down the debt in future, the value accretion would be far higher. Some may argue that is because of the leverage, which is true. On the other hand, Berky carried at least 4x to 5x cash and they could have used that cash and no need to leverage and let the future cash fill the coffers.

I hope, WEB drops his value anchor and moves to sustained buybacks, and if required sell his overvalued securities.
Print the post Back To Top
No. of Recommendations: 3
Chris Bloomstran just posted some of his comments on Twitter including:

“Depending on the pace of repos & the share price, ongoing $3B/monthly buys would shrink shares out by >6.5% in 2021. The monthly clip of buys has been steady for 9 ½ months. Worth watching. My hope is the stock stays cheap. Not sure many people get this. Most want stocks up.”
Print the post Back To Top
No. of Recommendations: 3
Chris made interesting comments before the annual report came out. In his report he wrote:

The pace of share repurchases picked up as the third quarter progressed, as did the stock price. By reason, my year-end work on Berkshire bakes in an additional $10 billion repurchased at an average price of $335,000 per share, 15% above estimated year-end book value. The crowd will applaud a large purchase, frown on a small buy and boo a no buy. Meriting attention is not only the size of the purchases but the price at which they are made. I have no doubt should the price of Berkshire’s shares again climb closer to our appraisal of value that the repurchase “program” will slow and stop between here and there. Price matters, and it’s such a unique, quaint really, concept at Berkshire compared to the broad swath of share repurchasers.
Print the post Back To Top
No. of Recommendations: 7
I am not sure that is the case.... but more importantly this is pretty flawed approach in my view. ...let us say Berkshire borrowed $25 B...and invested in buybacks now...I hope, WEB drops his value anchor and moves to sustained buybacks, and if required sell his overvalued securities.



Buffett's retort:

In 1958, Phil Fisher wrote a superb book on investing. In it, he analogized running a public company to managing a restaurant. If you are seeking diners, he said, you can attract a clientele and prosper featuring either hamburgers served with a Coke or a French cuisine accompanied by exotic wines. But you must not, Fisher warned, capriciously switch from one to the other: Your message to potential customers must be consistent with what they will find upon entering your premises.
At Berkshire, we have been serving hamburgers and Coke for 56 years. We cherish the clientele this fare has attracted...
...The tens of millions of other investors (Kingran) and speculators in the United States and elsewhere have a wide variety of equity choices to fit their tastes.
Print the post Back To Top
No. of Recommendations: 2
We cherish the clientele this fare has attracted..... ...The tens of millions of other investors (Kingran)

You are not the first person who indulges in selective reading....

Berkshire’s investment in Apple vividly illustrates the power of repurchases. We began buying Apple stock late in 2016 ... Despite that sale – voila! – Berkshire now owns 5.4% of Apple. That increase was costless to us, coming about because Apple has continuously repurchased its shares, ...
But that’s far from all of the good news. Because we also repurchased Berkshire shares during the 21/2 years, you now indirectly own a full 10% more of Apple’s assets and future earnings than you did in July 2018. This agreeable dynamic continues. Berkshire has repurchased more shares since yearend and is likely to further reduce its share count in the future. Apple has publicly stated an intention to repurchase its shares as well. As these reductions occur, ... The math of repurchases grinds away slowly, but can be powerful over time. The process offers a simple way for investors to own an ever-expanding portion of exceptional businesses.
And as a sultry Mae West assured us, and "Kingran violently agrees": “Too much of a good thing can be . . . wonderful.”
- Warren E Buffett, Chairman Berkshire Hathaway Inc, Annual Letter 2020
Print the post Back To Top
No. of Recommendations: 24
For a longer term perspective, buybacks restarted August 7 2018...
Then a brief pause of nothing in April, followed by a new level of commitment at seven times the rate...



FWIW, I looked at the rate of purchases per calendar day in all the known date intervals.
For example, we know the number of A and B shares at Sept 30, Oct 16, Dec 31, and Feb 16. Plus some purchase date ranges have been reported explicitly.
For intervals with few days, I combined them with the interval before or after which was most closely contiguous.

The main insight: there is a very strong correlation between buyback activity and valuation, as you might expect.

Price to book value is a flawed and deteriorating yardstick of share value, but it's good enough for rock and roll.
If you fit a line through the 20 observations of date ranges, you get this result.
P/B value versus number of A-share equivalents purchased per calendar day.
1.08    300.9
1.10 283.9
1.12 266.8
1.14 249.8
1.16 232.7
1.18 215.7
1.20 198.7
1.22 181.6
1.24 164.6
1.26 147.5
1.28 130.5
1.30 113.4
1.32 96.4
1.34 79.4
1.36 62.3
1.38 45.3
1.40 28.2
1.42 11.2
1.433 0


This fit suggests buybacks would halt in the vicinity of 1.433 times book.
From the actual data we have, the highest *known* purchase was at 1.438 on 2019-04-10, surprisingly close.

There are only a few outlier observations, though still not too far from the trend.
There were oddly few repurchases in May 2020 based on P/B valuation, and oddly many repurchases in March 2019 and September 2020.
But overall the trend is clear.


The valuation meaning of any given P/B ratio will be drifting, but very slowly.
Oddly enough, my valuation estimate suggest that the P/B ratio representing fair value has actually fallen a bit in the last several years.
Buybacks drive the fair P/B up, so this result is because other changes in the mix of assets have been bigger.

Since I had all the figures, I also checked how closely the prices paid match the average market prices.
It looks like Berkshire has averaged an execution price around .475% below VWAP, which pretty good given the volumes involved.
They buy more shares not just in months with cheaper valuations, but also more during the cheaper days and perhaps hours.
My guess has been that they use a moving "VWAP below limit" order.

Jim
Print the post Back To Top
No. of Recommendations: 1
"If you fit a line through the 20 observations of date ranges, you get this result."

Nice analysis. It's nice to see that Berkshire repurchases more shares when P/B is low.

I believe that Berkshire's repurchase price offers us one of the best estimates of IV. Take the highest P/B that Buffett has paid and estimate the highest P/IV that he would pay, and you get IV/BV.
The highest P/B Buffett paid was about 1.4 in Q4 2018. Buffett has said that he would not repurchase at a 5% discount to IV, but that he would repurchase "significant amounts" at a 25% to 30% discount to IV. If we use 1.4 as the highest P/B paid and assume that that represented a 5% discount to IV, then IV/BV equaled 1.47. If we assume that 1.4x BV represented a 10% discount to IV, then IV/BV equaled 1.56.
Print the post Back To Top
No. of Recommendations: 0
"Take the highest P/B that Buffett has paid and estimate the highest P/IV that he would pay, and you get IV/BV."

Actually all of the repurchases are consistent with an IV/BV of around 1.52. We don't have to start with what we think IV/BV is. We just have to guess what discount to IV each purchase represents, knowing that Buffett would not repurchase shares at a 5% discount to IV and that he would repurchase significant amounts at a 25% to 30% discount. Using Jim's table make your own guesses of what the discount to IV is for each repurchase. I think you'll come up with an IV/BV of about 1.52, plus or minus a small amount.
Print the post Back To Top
No. of Recommendations: 0
"Using Jim's table make your own guesses of what the discount to IV is for each repurchase. I think you'll come up with an IV/BV of about 1.52, plus or minus a small amount."

Call it the stock repurchase model :) As with all valuation models, each person will come up with a mean IV estimate and a range, but the range will not be large in comparison to other models.
Print the post Back To Top