Message Font: Serif | Sans-Serif
 
No. of Recommendations: 7
From the Cash Flow statement we can learn that Buffett bought 17.7bn equities during Q3. 4.9bn were added to the BAC position and the remaining 12.8bn went mostly into some other financials. The 13f should be interesting!
Print the post Back To Top
No. of Recommendations: 2
Nice catch on the BAC addition
Print the post Back To Top
No. of Recommendations: 5
<<Nice catch on the BAC addition>>

Very nice catch indeed and here I thought that I had read
the report over very thoroughly but sure enough it is
printed in microdots at the bottom of p. 11.

In the name of Zeus BAC is my second largest holding after
Berkshire ( having a little fun here Jim - I greatly respect your
posts and all that you do for this Board!)

Maxthetrade I propose that you change you name to OldEagleEye.

Thanks again for the catch … BAC now a larger position than WFC.
Print the post Back To Top
No. of Recommendations: 0
the other financial might be BK, which he has been adding in previous 13fs.
Print the post Back To Top
No. of Recommendations: 0
I have to assume he is getting close to owning 10% of BAC? So he may capped soon.
Print the post Back To Top
No. of Recommendations: 18
4.9bn were added to the BAC position and the remaining 12.8bn went mostly into some other financials. The 13f should be interesting!

I too admire your eagle eyes, but I don't see how you arrived at the 4.9bn number...?

Berkshire owned 700m shares of BAC at end 2017. (actually a bit fewer in the Q2 13f, but a 13f can be off by a bit due to quirks of pension funds, voting control, etc)

At end Q3 the position was worth $26.5bn (per footnote page 11) and the shares were trading at $27.50 at end Sept, so we know they owned 963.6 million shares.
So, year to date, 263.6 million shares were added, apparently almost all of them in Q3.
If we assume that the purchases were spread over time a bit but tilted towards the weaker price spots in the
latter half of the quarter, they probably paid an average of around $31.50 a share, give or take.
That suggests they spent about $8.1bn on BAC shares, no?

Frankly, I don't get it.
Surely Mr Buffett is much better at this than I am, so presumably I'm missing something, but it must be hiding well.
It seems like throwing in the towel in terms of trying to get outperformance from capital allocation.
I can see BAC as a "too big to fail" bond substitute, so maybe they have a place in Berkshire's balance sheet sitting between the equities and the cash.
But they just don't seem to be a good place to allocate money in equities long term.
Their return on incrementally allocated capital is poor, and I don't see why one would expect that to change.
Even with the tax cuts in effect this year, net income year to date has been only 0.85% of average assets.
A bank that can't manage 1%, a number they can only dream of based on the last decade, has no real reason to be in business.
Traditionally a really good bank would manage 1.5%.
Bank leverage is lower these days, so if anything you need a higher number than in days gone by.
It's only a short period, but ROE year to date has been 7.43%, and total shareholders' equity is down.
Sure, the future might be brighter, but book per share has been growing less than 2% faster than inflation in the last 5 or 10 years.
Berkshire was buying at a price suggesting a 2% dividend yield, so they're either expecting a
big improvement in BAC's business economics or a real total return of 4% before Berkshire's tax bill.
Half current, half compounding.
The reasoning that springs to mind is "we have too much cash, and there are only 25 companies big
enough to soak up a meaningful amount, and this one is not going to go out of business, so that will have to be good enough".

Jim
Print the post Back To Top
No. of Recommendations: 2
I too admire your eagle eyes, but I don't see how you arrived at the 4.9bn number...?


Hi Jim,
Berkshire owned 700m shares of BAC at end 2017. (actually a bit fewer in the Q2 13f, but a 13f can be off by a bit due to quirks of pension funds, voting control, etc)

At end Q3 the position was worth $26.5bn (per footnote page 11) and the shares were trading at $27.50 at end Sept, so we know they owned 963.6 million shares.
So, year to date, 263.6 million shares were added, apparently almost all of them in Q3.
If we assume that the purchases were spread over time a bit but tilted towards the weaker price spots in the
latter half of the quarter, they probably paid an average of around $31.50 a share, give or take.
That suggests they spent about $8.1bn on BAC shares, no?

The 4.9bn was a typo, I meant 5.9bn and that's obviously only an estimate. BAC closing price on Sep 28 was 29.46, so we owned roughly 900mm shares at the end of the quarter. 200mm new shares at an estimated purchase price of $29.5 gave me the 5.9bn.

I think BoA has improved very substantially under Moynihan and could look a lot more like Wells in the future. Time will tell.
Print the post Back To Top
No. of Recommendations: 2
I messed up the formating in my previous post.

I too admire your eagle eyes, but I don't see how you arrived at the 4.9bn number...?

Berkshire owned 700m shares of BAC at end 2017. (actually a bit fewer in the Q2 13f, but a 13f can be off by a bit due to quirks of pension funds, voting control, etc)

At end Q3 the position was worth $26.5bn (per footnote page 11) and the shares were trading at $27.50 at end Sept, so we know they owned 963.6 million shares.
So, year to date, 263.6 million shares were added, apparently almost all of them in Q3.
If we assume that the purchases were spread over time a bit but tilted towards the weaker price spots in the
latter half of the quarter, they probably paid an average of around $31.50 a share, give or take.
That suggests they spent about $8.1bn on BAC shares, no?



Hi Jim,
The 4.9bn was a typo, I meant 5.9bn and that's obviously only an estimate. BAC closing price on Sep 28 was 29.46, so we owned roughly 900mm shares at the end of the quarter. 200mm new shares at an estimated purchase price of $29.5 gave me the 5.9bn.

I think BoA has improved very substantially under Moynihan and could look a lot more like Wells in the future. Time will tell.
Print the post Back To Top
No. of Recommendations: 4
BAC closing price on Sep 28 was 29.46...
The 4.9bn was a typo, I meant 5.9bn and that's obviously only an estimate.


I had the wrong share price for the end of period.
I made the mistake of checking Yahoo...if you ask for monthly closes it tells you $27.50 as the non adjusted September close, which is wrong.
If you ask for daily closes it tells you $29.46 for Sept 28, the figure you used, which I presume is correct.
Thus, they owned $26.5bn/29.46= about 899m shares at end of September.
Knowing Mr Buffett's penchant for round numbers, I'll wager the number was actually 900m, meaning 200m share were purchased year to date, essentially all in Q3.
So I estimate that they bought about $6.1bn worth assuming a rough guess of $30.50 per share, reasonably close to your $5.9bn.

My reaction remains unchanged...maybe the price was adequate, but I wish I could see the case for it being a firm of the quality I'd like to see Berkshire buying.
They'll stay in business as a perpetual cash cow, but I hope for more than that.

Jim
Print the post Back To Top
No. of Recommendations: 8
I think BoA has improved very substantially under Moynihan and could look a lot more like Wells in the future. Time will tell.

Certainly possible.
But I'll stick with my view that buying BAC these days it's *either* settling for a cash cow *or* a very optimistic wager on a turnaround.

Last I checked, BofA haven't had ROA as high as Wells' in even a single year in the last 30.
Value Line's recent figures for ten year rate of change of
Earnings/share: WFC +6%/year, BAC -10%/year
Book/share: WFC 11%/year, BAC -2%/year
Total assets: WFC 10%/year, BAC -5%/year
Dividends/share: WFC 4%/year, BAC -23%/year
Average ROE: WFC 10.5%/year, BAC 3.4%/year
Note, Value Line smooths their figures, so I believe these are not using a single outlier year as a determinant.
And the most important one, return on total assets: WFC 1.09%/year, BAC 0.42%/year.


That is a whole lot of gap to close, and remember that even WFC isn't as good as it used to be.
Leverage for both has been forced down for everybody, and return on assets down (other than the tax rate change), so ROE is down multiplicatively.
It would take an optimist of great fortitude to think that BofA can manage an interesting ROE when even WFC now struggles to do so.
For "interesting" read "reliably averaging double digits through the business cycle".
Yes, being a huge laggard means there is lots of room for improvement.
But when you're this bad for this long, there are two issues:
(1) why would you expect the improvement? (cf, "definition of insanity"), and
(2) are even the best really big bank franchises like WFC still truly interesting investments if they struggle to average 10% ROE?

I presume the recent tax change has been a factor in Mr Buffett's views.
What else has changed to explain why he might be a buyer now at over $30 when he wasn't during the first eight months of last year when the price was almost mostly $22-$25?
Their long run prospects aren't that much better.

Jim
Print the post Back To Top
No. of Recommendations: 0
Like the airlines the big banks have the lowest PEs out there. Maybe it’s just as simple as that.
Print the post Back To Top
No. of Recommendations: 2

(2) are even the best really big bank franchises like WFC still truly interesting investments if they struggle to average 10% ROE?


Here's one quality that could make the answer yes under the right environment: During inflationary times, well run banks don't have to have a prayer meeting in order to raise prices. The cost of cars, boats, homes - all the things people borrow money for to buy will cost more over time due to inflation. This allows the banks to grow their loan books along with inflation more or less naturally due to the inherent traits of their business models.

Another, arguably even better, example would be card firms like V and MA that earn a percentage on each purchase made by a consumer. As the cost of goods bought increases due to inflation, their revenue increases naturally and no price increase is necessary. No prayer meeting required. Also, since they don't do any lending, there's no default risk. That can't be said for banks.

Rich
Print the post Back To Top
No. of Recommendations: 0
I'm certainly no expert on BAC but the rearview mirror may not be very helpful in this case. In the last quarter ROA was 1.23%, ROE 11% and ROTCE 15.5%. They pledged to freeze expenses at this years level for the next two years and could get a nice boost from increasing interest rates, e.g. a 100bp parallel shift will add $0.22 to EPS. So the actual numbers already aren't that bad and there's certainly room for further improvement. I can see this work out very nicely.
Print the post Back To Top
No. of Recommendations: 10
I'm certainly no expert on BAC but the rearview mirror may not be very helpful in this case. In the last quarter ROA was 1.23%, ROE 11% and ROTCE 15.5%.

True, those are good numbers for a bank these days.
I guess it depends on whether you think a single quarter is a sufficient sample.
First three quarters, all of the data with the tax cut, ROA 0.85% and ROE 7.43%. That I wouldn't buy.

The optimist: bad one-off things happened in Q1 and Q2 and/or things are improving rapidly, so Q3 better represents the new normal.
The pessimist: Q3 was a short term lucky blip.

I find it interesting that so much of the improvement in net interest margins in the last year or two
among large US banks is not from a steepening [government] yield curve, nor even (much) from rising loan rates.
In effect, they've mainly been putting the screws to depositors like a monopoly/oligopoly---despite not being one.
Policy short term rates have risen, but the rates paid out on accounts have barely budged at big banks.
They seem to be wagering that people are used to lending them money for free and will continue do to so forever even if there are alternatives.
Some on line (FDIC insured) banks are paying 1.98%, and "bricks and mortar" are still paying paying a national average of 0.09%, difference 1.90%.
Typical gap 1.6-1.8%.
The gap has doubled since 2014, so competition isn't working they way you might expect it to.
They are really seriously betting on the stickiness of deposits (customers=chumps), and that there will be no tipping point of customers wising up.
Anybody in the US with cash on deposit at a bank should check out MaxMyInterest.com or similar.

From Goldman's new Marcus division for Nov 4: they're offering 2.05% APY with $1 minimum.
Their site says Oct 30 rates for WFC = 0.01%, BAC 0.06%, Citi 0.04%, Chase 0.04%, data for $2500 balance.
So, which bank is treating their clients as "muppets" now?
Let's hear it for the vampire squid.

Jim
Print the post Back To Top
No. of Recommendations: 7
Here's one quality that could make the answer yes under the right environment: During inflationary times, well run banks don't have to have a prayer meeting in order to raise prices.
The cost of cars, boats, homes - all the things people borrow money for to buy will cost more over
time due to inflation. This allows the banks to grow their loan books along with inflation more or
less naturally due to the inherent traits of their business models.


I can see that.
But you don't really have to think about inflation so explicitly.
A simpler way of thinking of it is that, absent loss of market share, BAC's loan book will track GDP.
And the real loan book will track real GDP.
Loans might grow a bit faster than GDP if the overall debt-to-GDP ratio rises, but let's not go there...it probably wouldn't be a good thing for a bank.
That probably falls in the category of something that can't go on forever, and its last throes won't be pleasant.

Will real profits then track the real loan book?
Probably so, to a first approximation.
I suspect there will be some kind of long run trend of costs that isn't linear with loans.
Costs of maintaining market share, systemically important bank costs, stepping in the proverbial doodoo each recession or crisis?

And will profits per share track the total profits of the firm?
Maybe.
Buybacks at low prices might help, issuance at low prices during crises would hurt.

Somewhere in there is a problem, though...if they aren't paying out a really large fraction of profits,
and real growth is only tracking real GDP, the retained capital other than buybacks is being largely wasted somehow.
Or rather, the costs of maintaining the market share are maybe a lot higher than they at first seem.

In any case, it still seems to be the "perpetual bond" investment case.
If you buy at a low multiple of the cyclically adjusted earnings, you'll get something like that forever.

Jim
Print the post Back To Top
No. of Recommendations: 0
True, those are good numbers for a bank these days.
I guess it depends on whether you think a single quarter is a sufficient sample.
First three quarters, all of the data with the tax cut, ROA 0.85% and ROE 7.43%. That I wouldn't buy.


Jim, I don't know where you got those numbers but here are the numbers for the first three quarters straight from the source: ROA 1.2%, ROE 10.86%, ROTCE 15.3%

http://phx.corporate-ir.net/External.File?item=UGFyZW50SUQ9N...
Print the post Back To Top
No. of Recommendations: 2
I suspect the recent BAC purchase is is a Buffett cash substitute move over the short term and not a longer term commitment.


He did the same thing with Exxon in 2013 - in and out in less than 2 years. Munger described it thus :

You know why Warren bought Exxon? As a cash substitute. He would have never done that in the old days. We had a lot of cash and we thought Exxon was better than cash over the short-term. That's a different kind of thinking than the way Warren came up.

https://www.fool.com/investing/2017/03/09/warren-buffett-exp...

if not this I would agree it just doesn't make much sense as a long term commitment of that chunk of capital in a so-so business.

https://www.fool.com/investing/2017/03/09/warren-buffett-exp...

I think the reason he has not committed this capital to buybacks is because he can get out of the position very quickly if he needs to free capital for a deal.
Print the post Back To Top
No. of Recommendations: 3
FYI. The FDIC surcharge from the great recession expires at the end of 2018. Therefore bank FDIC insurance cost will decline to pre recession levels. This will save WFC $500 MM+ in annual expense I believe, I assume BAC will benefit nicely as well.

So WFC will benefit from rising NIM. ?
lower FDIC insurance $500 MM +
lower amortization expense from old items rolling off. $800 MM ish.

BAC should at lest benefit as well from the first 2 items above.
Print the post Back To Top
No. of Recommendations: 1
Actually WFC pays market rates on brokered CD's. Highest in the land from what I see. The whole point of a branch system is that your cost of funds, including the cost of operating the branch should be lower than the cost of brokered deposits. Certainly no higher and hopefully more stable during a crisis. The benefits of branches have been small in a low rate environment. In fact I would argue that it would of been cheaper in the short run to by brokered funds than all in cost branch funds. Hopefully this will change as rates rise.
Print the post Back To Top
No. of Recommendations: 9
First three quarters, all of the data with the tax cut, ROA 0.85% and ROE 7.43%. That I wouldn't buy.

Jim, I don't know where you got those numbers but here are the numbers for the first three quarters straight from the source: ROA 1.2%, ROE 10.86%, ROTCE 15.3%


I read their balance sheet and calculated it myself.
From third quarter 10Q page 54-57.

Total assets Dec 31 2017, 2,281,234
Total assets Sep 30 2018, 2,338,833
Average of two as a proxy for average assets during the period = 2,310,034
Net income for the nine month period attributable to common shareholders, p 54: 19,657
Divide the two and you get ROA 0.85%.

Same method for ROE:
Total shareholders' equity Dec 31 2017, 267,146 from page 57
Total shareholders' equity Sep 30 2018, 262,158
Average of two as a proxy of average equity: 264,652
Net income for the nine month period attributable to common shareholders, p 54: 19,657
Divide the two and you get ROE 7.43%.

I consider preferred dividends as a cost of business like interest, not any kind of net income return to equity...depends what meaning you're looking for.
But that makes only a small difference. 0.05% ROA.

Their reported numbers are admittedly...different.
They say "Return on average allocated capital is calculated as net income, adjusted for cost of
funds and earnings credits and certain expenses related to intangibles, divided by average allocated capital.
Other companies may define or calculate these measures differently."

Well, I'm not a company, but I do indeed calculate it differently.
I calculate ROA and ROE by dividing net income to common by total assets or common equity.

Jim
Print the post Back To Top
No. of Recommendations: 1
Actually WFC pays market rates on brokered CD's.
Their site says Oct 30 rates for WFC = 0.01%, BAC 0.06%, Citi 0.04%, Chase 0.04%, data for $2500 balance.
...
Actually WFC pays market rates on brokered CD's.


Different topic...the comment was about demand deposit accounts, not term money like CDs.

Jim
Print the post Back To Top
No. of Recommendations: 0
I think BoA has improved very substantially under Moynihan and could look a lot more like Wells in the future

BAC, will improve its efficiency for 3 to 5 years compared to historical levels, setting up nicely. I also hope it doesn't get its efficiency Well's way. I think most of the efficiency is going to come from serious cost cutting (reducing employee costs) through automation.
Print the post Back To Top
No. of Recommendations: 2
Total assets Dec 31 2017, 2,281,234
Total assets Sep 30 2018, 2,338,833
Average of two as a proxy for average assets during the period = 2,310,034
Net income for the nine month period attributable to common shareholders, p 54: 19,657
Divide the two and you get ROA 0.85%.

Same method for ROE:
Total shareholders' equity Dec 31 2017, 267,146 from page 57
Total shareholders' equity Sep 30 2018, 262,158
Average of two as a proxy of average equity: 264,652
Net income for the nine month period attributable to common shareholders, p 54: 19,657
Divide the two and you get ROE 7.43%.


I'm sure you know but you have to use some kind of annualized income number to get a meaningful result. I presume that explains most of the difference.
Print the post Back To Top
No. of Recommendations: 3
I may not of explained my point well.

While the customer receives .01% on demand deposits, the cost to the bank is much more.

A branch system is very costly to maintain. Facility cost and staffing could exceed 100 Basis points of total average daily balances from deposits. But obviously the bank can't pay less than zero during what seems to be the ridiculously low rates of recent years. So, until the all in cost of deposits is well below market rates the banks will appropriately lag. The primary purpose of a branch system is to capture low rate deposits. Similar to the concept of float. So from where I sit, BAC and WFC net interest margins are still compressed and do not reflect the long term potential for normalized rates. Even paying .01% is too High. In this environment a start up branchless bank may have a lower cost of funds. As rates normalize, the traditional banks should flip this.
Print the post Back To Top
No. of Recommendations: 12
I'm sure you know but you have to use some kind of annualized income number to get a meaningful result. I presume that explains most of the difference.

Ah! Sorry about that. Having a stupid day.
My usual spreadsheet for that tracks TTM numbers, but I did this directly from the statements.
Which, as you point out, is wrong unless it's an annual report.
The reason I never do BAC from my usual spreadsheet is that I don't track BAC closely, because the numbers have always been so reliably uninteresting.

As it happens, my huge mistake makes less difference than you might expect because Q4 last year wasn't very exciting.
Redone with trailing four quarters figures = (Q1-Q3 2018) + [(FY 2017) - (Q1-Q3 2017)].
ROA 0.95%, ROE 8.19% using trailing-four-quarters net income to common and average total shareholders' equity
ROA 0.69%, ROE 5.96% using trailing-four-quarters comprehensive income and average total shareholders' equity
I don't know what adjustments they do to get the headline numbers they report. Maybe some of them are meaningful.
These are straight from the consolidated statements, the way I'd do it for any company.
They remain, in my view, reliably uninteresting.

I think there may be some truth to the notion that Mr Buffett may be seeing more BAC shares as a cash substitute or placeholder position.
This doesn't reliably work out well as a rule of thumb, as the times that you want that money back are so often the times that the price of your cash proxy is low.
But Berkshire is very unlikely to be short of funds any time soon, so maybe it makes more sense for them than it does for most.
Much as I am not fond of them as a long term pick, I admit BAC shares are pretty likely to outperform cash over many different time frames starting from prices around $30.

Speaking of cash, Berkshire still has a lot.
Using 6 month T-notes as a proxy for their average yield, average rate in 2017 was about 1.05%.
Now it's 2.5% and still rising.
Still low, but on $100bn+ the difference adds up to something. Around $1.64bn more per year after tax, or $700 per share.
If the typical after tax dollar earned at Berkshire is worth $14.25 in market value, that's $10k more share value if sustained.
Unless you were already doing a cyclical adjustment on interest rates.

Jim
Print the post Back To Top
No. of Recommendations: 6
As it happens, my huge mistake makes less difference than you might expect because Q4 last year wasn't very exciting.
Redone with trailing four quarters figures = (Q1-Q3 2018) + [(FY 2017) - (Q1-Q3 2017)].


Not to nitpick but there were some truly non recurring items in Q4 2017 like the 2.9bn loss from the Tax Act. I think the numbers for the full year 2018 will be much more representative of what we can expect in the future. I doubt that Buffett bought this as a cash proxy. You may also have noticed that Munger holds quite a bit of BAC at DJCO where there is really no need for a cash proxy. If you look through the investor presentations you can see steady improvement over the last few years. You may be pleasantly surprised, this isn't the old Ken Lewis BAC.
Print the post Back To Top
No. of Recommendations: 2
If you look through the investor presentations you can see steady improvement over the last few years.

I certainly agree that there has been improvement.
If it continues, then yes, they may become a company worth investing in.
But improvement in a business's fundamental economics is not something that extrapolates well.
Maybe Mr Buffett and Mr Munger are in a position to forecast the last half of the needed turnaround, but I am not.
I wouldn't look at a bank that isn't already demonstrably capable averaging double digit ROE through the business cycle.
Think of all the years that people have been waiting for the imminent improvement at Deutsche.


Not to nitpick but there were some truly non recurring items in Q4 2017 like the 2.9bn loss from the Tax Act.
True.
But that doesn't actually change things much. $2.9bn is a lot, but not in comparison to net income to shareholders of $22bn.
Taking out that hit, four quarter ROA and ROE based on comprehensive income rise to 0.81% and 7.05%.
Not yet up to the bottom end of acceptable, in my books.
So, the improvement has to last.

Here's what a nicer bank ROE history looks like : )
static.seekingalpha.com/uploads/2017/6/11/47781612-149721453...
That's SVNLY, trading near five year lows.

Jim
Print the post Back To Top