No I fully understand. The prior poster does not with an IV of $313/B
NO WAY those 2, cheapskates, are buying-in stock unless they think it was very undervalued at $322/b.
WEB is extremely price sensitive and disciplined
3 Likes
Buybacks are different from buying stock. Buyback is his way of returning excess capital to his shareholders and he is going to do that as long as the price is within the reasonable range. Folks reading between the lines on buyback price to drive what is the right price for Berkshire doesn’t understand how buybacks work.
Look around the companies that do buybacks, they do it when they have cash, when the business is booming and stop when the business climate is not good, AKA when their stock price is low.
Berkshire has plenty of cash, and the stock price is within the reasonable range, so Berkshire continued to buyback using operating earnings.
If you are expecting Berkshire to accelerate buyback now that the stock is 24% lower from ATH, you are going to be disappointed.
3 Likes
NO WAY those 2, cheapskates, are buying-in stock unless they think it was very undervalued at $322/b.
WEB is extremely price sensitive and disciplined
My thoughts:
- Munger isn’t involved in the investment decisions. It’s all Buffet.
- Buffet is price sensitive so he scaled back repurchases early this year, but…
- He’s not been waiting for the fat pitch to make some repurchases, he’s been using up excess cash flow.
Repurchases:
2021 Q1 $6.5bn $239avg
2021 Q2 $6.0bn $279avg
2021 Q3 $7.6bn $280avg
2021 Q4 $6.7bn $286avg
2022 Q1 $3.1bn $316avg
See a trend?
At $322 I expect he thought it was an OK use of excess cash.
Be interesting to see if he’s buying a lot now.
1 Like
WEB is extremely price sensitive and disciplined…
One note–
The current valuation level is far from the only consideration when the decision is made how many shares to repurchase.
Recent size of and changes to the cash pile, other opportunities (or their lack), and anticipation of possible imminent opportunities, seem to matter a lot.
The boss has done repurchases at valuation multiples that are about materially above average for the last 15 years,
yet also sometimes been entirely inactive on the buyback front at much better valuation levels.
So, yes, price sensitive and disciplined no doubt, but it’s not all about the attractiveness of the price.
Jim
4 Likes
Buybacks for BRK are generally not a great option as long as other cash deployment opportunities are available. As pessimism increases producing lower asset prices BRK benefits as it is always seeking investment opportunities for its persistent cash flow. Adding other cash flow long duration opportunities rather than buying back stock is always Buffetts preferred option. I hope buybacks are minimal now but other opportunities are deemed better by Buffett. perhaps more oil stocks… We know he keeps on adding more OXY under 56 or so. Maybe XOM has been added to the portfolio…… CVX is approaching 10% so he likely wont exceed that due to filing requirements.
1 Like
Did a quick and dirty on the top 5 investees:
Eps (from Yahoo) x shares owned = $13.4bn
That’s 72% of the portfolio, so assume total is 13.4/0.72 = $18.6bn
(Bloomstran said $16.7bn as of 3rd quarter 2021)
To make a quick and dirty IV estimate:
Non-insurance businesses trailing 4 quarters: $21.2bn
Partially owned: $1bn
Look through: $18.6bn
Total: $40.8bn
Multiple of 15: $612bn
Cash/bonds available to invest: $80bn
IV per B share: $313
I like this approach, but shouldn’t there be a line for underwriting earnings? Bloomstran uses a percentage of premiums earned (5% pre-tax), which adds about $3.5b pre-tax.
Maybe also a small deduction for the taxes on the dividend part of the equity look-through earnings?
dtb
1 Like
<<Did a quick and dirty on the top 5 investees:
Eps (from Yahoo) x shares owned = $13.4bn
That’s 72% of the portfolio, so assume total is 13.4/0.72 = $18.6bn
(Bloomstran said $16.7bn as of 3rd quarter 2021)
To make a quick and dirty IV estimate:
Non-insurance businesses trailing 4 quarters: $21.2bn
Partially owned: $1bn
Look through: $18.6bn
Total: $40.8bn
Multiple of 15: $612bn
Cash/bonds available to invest: $80bn
IV per B share: $313>>
I like this approach, but shouldn’t there be a line for underwriting earnings? Bloomstran uses a percentage of premiums earned (5% pre-tax), which adds about $3.5b pre-tax.
Maybe also a small deduction for the taxes on the dividend part of the equity look-through earnings?
I usually leave out underwriting earnings because WEB did in his 5-groves tutorial.
It doesn’t make much difference, unless you engage in counting the eggs as chickens like Bloomstran seems to.
Taxes on the dividends: IIRC dividends are about $6Bn, so about $1.3Bn deduction for taxes?
Not worth worrying about on a $690Bn valuation.
Where I really should put in more work is the look-through earnings of the investees.
Plus, I think Jim said he puts a 21 multiple on those (cyclically adjusted) earnings …that was a surprise.
It’s good to have multiple models that are fairly easy to keep updated. I’ve added this one to my spreadsheet.
Conservative IV per B share, all based on 3/31/22 numbers:
5-groves: $340
Look through: $313
Book + Float: $297
2021 Q1 average repurchase: $239
2021 Q2 average repurchase: $279
2021 Q3 average repurchase: $280
2021 Q4 average repurchase: $286
2022 Q1 average repurchase: $316
Current quote: $274
Y’all can draw your own conclusions.
5 Likes
“ Y’all can draw your own conclusions.”
Love this. Everyone should think for themselves
I usually leave out underwriting earnings because WEB did in his 5-groves tutorial.
He did mention it in another letter, if I remember correctly.
He seems to have picked a year that the underwriting profit was pretty typical.
I have an overly fancy method of estimating the cyclically adjusted underwriting profit.
I use the average of:
- Small percentage of float (1.3%)
- Different small percentage of premiums earned (2.7%), excluding one-time big figures from retroactive deals.
Both methods are intended to be a proxy for the overall size of the insurance operation.
At Q1 the annual rate came out to $1.92 bn pre-tax, or $1.52bn after tax.
On average over time, this formula is just a little less than the actual underwriting profit.
When I estimate earning power I just replace the actual underwriting gain/loss with this number.
This one adjustment, plus skipping unrealized stock gains/losses, sure makes the financial results a heck of a lot steadier.
Jim
7 Likes
Love this. Everyone should think for themselves
Hmmm, I can think of some possible exceptions I’ve met… : )
Jim
9 Likes
I usually leave out underwriting earnings because WEB did in his 5-groves tutorial.
Ok. But underwriting earnings have averaged about $1.5-2b for the last 20 years. True, that’s only about 5% of the approx. $30b in net earnings for the company, and it’s true that it is very lumpy, but I think we can safely conclude that it is non-zero and it’s not an insignificant amount.
Taxes on the dividends: IIRC dividends are about $6Bn, so about $1.3Bn deduction for taxes?
Not worth worrying about on a $690Bn valuation.
Berkshire has a market cap of about $611b, not $690b, but more importantly, the $1.3b should be compared to total earnings, not market cap. If total earnings are about $30b (ignoring unrealized gains on the equity portfolio), then this item, and the previous one, are each about 5% of that figure.
Where I really should put in more work is the look-through earnings of the investees.
Plus, I think Jim said he puts a 21 multiple on those (cyclically adjusted) earnings …that was a surprise.
No, I think he says that in his estimate of book value, he caps the individual holdings at 21x earnings, not that he multiplies all the earnings by 21. That would cap Apple at 21x TTM earnings of $102b, so $2.14b, rather than the current $2.25b market cap. This made a bigger difference when Apple shares were at $180 at the end of the year and Apple was at 30 times earnings (resulting in a $50b reduction in Jim’s assessment of the Apple stake, IIRC) than it does now with $144 Apple shares at less than 23 times earnings.
dtb
7 Likes
The formula is book+Float+ 1/2 deferred taxes
Not just bv + float
Plus, I think Jim said he puts a 21 multiple on those (cyclically adjusted) earnings …that was a surprise.
…
No, I think he says that in his estimate of book value, he caps the individual holdings at 21x earnings, not that he multiplies all the earnings by 21.
Correct.
I just apply a cap to multiples to try to avoid thinking that bubble-level market values are true value.
Some very few firms are worth more than that, for sure. Though you almost never know which ones in advance.
I’d rather be pleasantly surprised when very high valuations work out well than unpleasantly surprised when it usually doesn’t.
21x is wildly generous in our family. I use 15 for operating subs.
Jim
2 Likes
The formula is book+Float+ 1/2 deferred taxes
Not just bv + float
That comes to $682Bn or $309/B share.
Why 1/2 deferred taxes?
Going back to Buffett’s tutorial again: he took off the whole amount of deferred taxes.
1 Like
Why 1/2 deferred taxes?
The reasoning is that deferred taxes are just another kind of float.
The assumption is that there will always be some deferred taxes on average over time, and that money can be invested in the mean time.
In both cases the liability is only partly “real” because an infinite duration loan at zero interest cost is nigh indistinguishable from capital:
you get to invest it forever and get the returns on it forever.
It’s only relevant as a liability if the company were about to be liquidated, which isn’t going to happen.
I’m not defending any specific number or ratio, but the reasoning is identical for float and deferred taxes.
It quacks like a duck, so it’s (nearly as good as) a duck. Or rather, a goose.
You don’t get to own the goose, but you get to keep all the golden eggs it lays in future.
Jim
5 Likes
Berkshire has a market cap of about $611b, not $690b…
Yeah, I was talking about a valuation of $690Bn.
…but more importantly, the $1.3b should be compared to total earnings, not market cap. If total earnings are about $30b (ignoring unrealized gains on the equity portfolio), then this item, and the previous one, are each about 5% of that figure.
Fair enough.
Adding in underwriting earnings (use Jim’s $1.5Bn) and deducting taxes on dividends at 13%:
Quick and dirty IV estimate:
Non-insurance businesses trailing 4 quarters: $21.2bn
Partially owned: $1bn
Look through including dividends paid out to Berkshire: $18.6bn
Tax on dividends @13%: -$0.65Bn
Underwriting: $1.5Bn
Total: $41.65bn
Multiple of 15: $625bn
Cash/bonds available to invest: $80bn
IV per B share: $319 (vs $313 previously)
No, I think he says that in his estimate of book value, he caps the individual holdings at 21x earnings, not that he multiplies all the earnings by 21.
Yes, thanks. Jim cleared that up too. My misunderstanding.
I’m not defending any specific number or ratio, but the reasoning is identical for float and deferred taxes.
I agree that in both cases you’re using money that ultimately belongs to someone else for some time period.
But I see an important (to be) difference between float and deferred taxes. The deferred taxes are already invested - i.e. in the stock that generates them. So what you get are the dividends from that investment plus any capital gains after taxes on the deferred taxes you have working for you during the holding period. Only the dividend income is available to invest elsewhere immediately. And dividend income is already included in the “normal” earnings of BRK.
With float, you have the option to put all of it to work immediately in whatever you choose - including just holding it as cash equivalents if other opportunities aren’t attractive. Then you wait.
I also view deferred taxes from unsold equities and bonds different than deferred taxes from operating businesses. That latter are a normal part of any capital investment - and already built into the cash flow from those businesses.
One should remember also that when Gottesman stated his formula, the deferred taxes were almost totally from the equity investments. In the Wesco days, Munger in his annual letter would estimate the value of the deferred taxes on Wesco’s stock holdings at 35%. I doubt Gottesman would use 50% today. I used 25% for a while based on the below, but finally stopped.
It’s not hard to set up a spreadsheet for equity deferred taxes. Put in the dividend yield, growth rate of the stock price, holding time, discount rate, etc. and calculate a value. Do it for each major holding or just guesstimate the total portfolio. But it takes a long time before it starts becoming significant. I’ve finally quit bothering with it - it became a small factor compared with the other estimates and assumptions about Berkshire.
It’s real but not very significant.
5 Likes
But I see an important (to be) difference between float and deferred taxes.
(to ME)
I swear I sometimes think that the TMF software makes changes after I hit enter.
Quick and dirty IV estimate:
Non-insurance businesses trailing 4 quarters: $21.2bn
Partially owned: $1bn
Look through including dividends paid out to Berkshire: $18.6bn
Tax on dividends @13%: -$0.65Bn
Underwriting: $1.5Bn
Total: $41.65bn
Multiple of 15: $625bn
Cash/bonds available to invest: $80bn
IV per B share: $319 (vs $313 previously)
I like this.
When I occasionally do it, at the end of the reasoning, I take the total ($41.65b) and divide the market cap by that. In other words, instead of assigning an arbitrary multiple, I calculate what multiple it is trading at, and compare it to alternatives.
So at today’s market cap of $623b, we would get at P/E of almost exactly 15. So the question is, do you want to own a company that has steady earnings along with $140b of dry powder loaded in the elephant guns of a great capital allocator, but despite that low-yield cash manages to obtain a 100/15=6.7% earnings yield? Or do you want something like the S&P that is trading at 20x record earnings, i.e. an earnings yield of 5.0%?
Easy call for me.
dtb
4 Likes
When I occasionally do it, at the end of the reasoning, I take the total ($41.65b) and divide the market cap by that. In other words, instead of assigning an arbitrary multiple, I calculate what multiple it is trading at, and compare it to alternatives.
So at today’s market cap of $623b, we would get at P/E of almost exactly 15. So the question is, do you want to own a company that has steady earnings along with $140b of dry powder loaded in the elephant guns of a great capital allocator, but despite that low-yield cash manages to obtain a 100/15=6.7% earnings yield? Or do you want something like the S&P that is trading at 20x record earnings, i.e. an earnings yield of 5.0%?
I like that better.
You could argue that cash available for investment should be deducted.
Say that’s $80bn, giving a multiple of (623-80)/40.65 = 13.4 or a 7.5% earnings yield.
Yeah, that’s quite attractive.
3 Likes