aws

Such a short letter this year. Two pages less than last year which was already the shortest in 20 years before that. And much less useful content. Kind of disappointing to be honest.

The loss from the writedown was not that much different from the market value decline during the period, so it would be more or less the same if they were not reporting it on the equity method. KHC was 55.11 on 9/30 and 43.04 on 12/31. 325mm sh x 12.07 loss per share = 3.92b. And since the market value is a further 20% lower QTD it will actually be carried well above fair value, as compared with most of their equity portfolio which has rallied strongly since yearend.

The linked percentage ownership are off. It's more like 9.3% of each WFC and BAC. Like one year away from being at 10% at current buyback rates. They are probably using old or average shares outstanding figures.

Net buys were only about 1b for the quarter, and unless they got lucky and bought RHT before the IBM news then most of that is probably just a merger arb. Not exactly swinging for the fences on one of the worst quarters in recent memory. I wonder if he was expecting it to get a lot worse, or if he was spending all his funds on Berkshire stock. Berkshire did do relatively well compared to the market up until the start of the year, so maybe he was buying a lot then but has been locked out lately.

I just don't see how they are going to come up with an equitable solution in a bankruptcy. The fire victims likely won't get anywhere near full compensation. There's talk of bigger cities like San Francisco wanting to break-up PG&E and take just slices that serve their own region, but if that were allowed there would be no affordable power in the rural regions subject to the fire risk. The current system sucks, where you can be hit with multi-billion dollar litigation even in the event you did nothing wrong because of inverse condemnation (not saying PG&E did nothing wrong, but that's a risk any buyer would have to assume), and the only reason it remotely works is because at least you can spread the risk over millions of ratepayers. Not only do the cities help subsidize the overall cost, but it provides some diversification so at least you have a large enough pool of overall earnings that it's possible to come back from a single disaster. If you broke up PG&E and individual utilities controlled slices, no one with sufficient capital to actually cover the potential damages that could occur would take the risk of providing power to the rural regions. Power would be extremely expensive to those residents and the company that provides it would fold like a lawn chair if there was another fire leaving future victims much less likely to recover than they are now. So if you sell off PG&E piecemeal you will hurt a lot of people outside the major cities. No one will buy the whole mess under the current law. If you change the law or do something to make it possible to own, then maybe you should just go ahead and do that now instead of letting PG&E fail first. A protracted bankruptcy can't be the best solution for anybody.

If California lets them go bankrupt, who would dare step in with new capital post BK? In good years they get a decent but not amazing return on equity, are crimped on their ability to upgrade infrastructure and safety by not getting enough rate increases, and then eventually there is a disaster and you're wiped out whether at fault or not. Not an attractive value proposition for anyone. The stock is trading like an option on California doing something to bail them out, but even if they don't get something like the massive rate increases or debt securitization I would have to think they would be fighting responsibility for wildfire claims in the courts for years. Not sure how likely it is, but given no option but to fight back against accepting full financial responsibility for the disaster, they might be able to mount a challenge to the laws governing liability vs. act of God. They are only insolvent if massive claims like $30B+ are allowed.

I'm in at 50ish% now. I increased it a lot initially last summer in the high 180s, and now again around 195 in these days while BRK underperforms the market bounce. Now that I'm basically fully loaded I would be perfectly happy for whoever is in charge here to take the lid off the price and allow it to move up with the general market.

I would be shocked if the share count of Apple decreased. He's not one to try to flip shares for a quick profit even if they start to look overvalued, and the best he could have done is record something like a 30% gain on his recent purchases. I'd bet there is a higher chance he nearly doubled it than reduced it. We know he added tens of millions of shares at prices around 25% above where we are right now so he must have thought it was a big bargain then, and I can't see anyway he could think it has been impaired so much by a little blip like a 9% revenue miss in a single quarter. It's a huge position, but the percentages are tiny compared to what he worked with before. Don't forget he put 1/3rd of Berkshire's net worth into Coke, and for most of the 90s its value was something like half of Berkshire's book value and he never sold a share. I believe the latter he views as a mistake, but that was when he was sitting on a 10-bagger in ten years, not less than a double like he had with Apple. Apple was never more than around 6% of Berkshire's net worth at cost or more than 10% at market value. The numbers are bigger now, but I don't think that would scare him if he still has conviction in Apple. I could see him going to the 9% range - probably not up to 10 because then he'd be immediately selling along with the repurchases - which would be around 430 million shares total or 177.5 million in new purchases.

The under performance for the week continues today with Berkshire lagging the S&P by more than 1%. Berkshire down 3.54% for the week with the S&P up 1.39% as I write this, for an underperformance of 4.93%. For perspective, that's the worst week of relative performance since 2009. If this keeps up I'm guessing there will be plenty of opportunities for buybacks.

I'm aware there has been outperformance vs. the S&P this past year, mostly due to the pops from the buyback announcement and Q3 earnings. As an aside though, you can extend that time frame further back and find periods of underperformance despite that. For example the S&P total return since 12/31/14 is 30% and Berkshire over the same time is 29%. Anyway, my point isn't that it's the cheapest it has ever been. I just have been tracking the interplay between Berkshire's share price, the S&P 500, and Berkshire's portfolio (as a proxy for Berkshire book value). I expect those are the three most important things Warren would be looking at when making buyback decisions. We have some data from prior repurchases about what things looked like when he was buying shares, so I'd just like to have a handle on how those factors have moved since then.

With today's selloff Berkshire seems the most relatively undervalued that I have seen since I have been tracking it. It's only 3.2% off the lows from Christmas eve, compared to a 5.2% gain for the S&P 500. Also, even with Apple trading at new lows the overall Berkshire portfolio is up 4.6% since that time so that doesn't explain any of the difference. So relatively speaking Berkshire is about 2% cheaper than it was at the market lows late last year.

I updated my spreadsheet on portfolio losses to today's numbers. Pretax QTD market loss is now 49.1B, which is 38.8B or 15.76 per B share after tax if you use a 21% tax rate. With a total market cap loss of 64.9B QTD on Berkshire overall, that implies the operating company and cash are worth 8.51% less today than on 9/30. And since you figure they probably had decent operating profits it's more like a 10% drop. Certainly seems much more in the buyback range now than it did a week ago, although obviously all other stocks they are looking at probably dropped more than 10%, so we'll have to wait and see what they actually did. Edit: Post didn't age well as I wrote it when the market was down for the day. brk_declines.xlsx

Don't forget to subtract the massive portfolio losses from book value. I haven't calculated it exactly but it's closer to 1.40 than to 1.20 for sure. Part of it would depend on if they need to write down their Kraft, not sure of the rule on that. Market value is below Kraft book value but they account for it on the equity method.

I ran some calculations to add additional perspective to market price change in Berkshire since prior quarter end and its portfolio holdings. I used the data from the CNBC portfolio tracker and I excluded the handful of stocks under $400mm in holdings which in the aggregate are less than $1 billion total. I attached my spreadsheet with more details but I wanted to point out a few things: Berkshire's market cap has fallen much slower than the average stock it holds. In fact, the majority of the decline since prior quarter's end was the decline in portfolio value. Assuming a tax rate of 21%, Berkshire's portfolio has lost 11.69 in per B share value after tax, compared to the market price drop of 18.94 over the same period. The roughly 42% of market cap weighted to the portfolio accounted for 62% of Berkshire's decline. If you subtract out the impact of the investment portfolio then Berkshire's market cap of remaining assets dropped by 5.82%. So effectively Berkshire the operating company is 5.82% cheaper that it was on 9/30, when the average company in the S&P 500 is 13.8% cheaper. I would have to think that heavily skews in favor of NOT buying in Berkshire over other opportunities. Or maybe he's buying back so much that the price is propped up a bit and otherwise would have fallen closer to market average. As an aside, how much do you think the market is going to overreact when they see Berkshire's enormous net loss for the quarter? We're already at something like $25 billion after tax loss here assuming good operating earnings. Since it seems like everything is more volatile and short-term oriented lately I could see a big drop if we are still in a weak market then, even though it is obviously old news by that point. That might be the time the buybacks kick into high gear. brk_declines.xlsx

I calculate that as of today the top 7 holdings (aapl, bac, wfc, ko, axp, khc, and usb) have declined a total of $24.23B from the prior quarter as of today. These 7 make up at current value 141B of the 191B market cap of their portfolio according to CNBC tracker. I didn't bother to calculate the loss on the remaining $50B in securities, but let's just assume it washed with operating earnings QTD. So the loss from those 7 big positions brings the book value down roughly $10/sh pretax or something like $7.50 after tax. So it seems like after subtracting portfolio losses Berkshire is trading at around exactly where we know he was buying it back. That was before marking up for excellent Q3 earnings, but before a plethora of additional opportunities to deploy capital in the markets opened up. So if I had to guess I would think he's buying slowly just like in Q3. Maybe a few billion total since there have been more days with a price that may be significantly below IV than there were in Aug-Sep.

https://money.cnn.com/data/markets/sandp/ Although they only put 15 companies per page with no sorting except alphabetical so navigating is a bit tough

Surprised there weren't more buybacks in the drop after Q3 end. I wonder if any buyback restriction was involved, or simply that the market value loss in their equity portfolio depressed book value enough to limit buys. Anyone have a sense of what the delta was between their portfolio values when they were buying back at 207 in Q3 vs what it was when they presumably bought a little back under 200?

88 was right the first time. He's also going to be on Bloomberg at 11:30am tomorrow, pre-taped I guess.

I'll be perfectly happy if this buyback bump is a nonevent. Berkshire the stock under-performed Berkshire the company so far in 2018, but I'm not sure that was the case in 2013, 2014, 2016, or 2017. It seems like there have been many more undervalued spots in recent history than there are right now so I don't see why Buffett would be rushing to buy back shares hand over fist right now. It would be great if some more China fears, Gates Foundation sales, and disappointed speculators drop the stock back under 200 for the foreseeable future. I don't want to see a crash, but I just don't want the stock to look fully valued as it will be harder to buy new shares when money becomes available. I have a lot more shares I want to buy over the next 10+ years, and I wouldn't want to be competing with the Company for shares, especially not over $200.

Buffett did add to the position a little after it was already above 10%, but it was a small addition. In 1997 they reported a 10.7% stake in AXP, and then in 1998 added 1.08 million shares (3.24 million split adjusted). There was a sharp drop that summer and he was able to pick up some more shares for around $24/sh. It recovered quickly though and probably beyond where he wanted to add any more. Since then he hasn't bought or sold a single share.

I'm not sure if it's been mentioned before in this thread or others, but I would think you need to be careful with the 1.2x book calculations when there is so much cash laying around. Cash is clearly not worth more than 1x book, and with around $42/B share in cash that's not insignificant. If they were to buyback any material number of shares starting at 1.2 book, they would soon be buying above that level. Imagine for a simplified example that there are exactly 2.5 billion B share equivalents, that the market value is $200/sh, the book value is 166.67/sh (so trading at 1.2/book, which is 416.67 billion total), and they have well over $100B in starting cash. If they did a $100B tender at the $200/sh and it was fully subscribed then 500 million shares would be retired all at 200/sh, leaving 2 billion shares and a new market cap 400B, and a new book value of 316.67B. Assuming the market value is still $200/sh at that point they would now be trading at 1.26 times book. With that said, I've been a buyer lately in the high 180s and am glad to see it.

gfp, I thought the same. I'm speculating that this speaks to WEB's discipline in not accepting a return below whatever his hurdle rate is, even with 100B in his wallet ;) -CM After the warrants were exercised, if he still held the preferred stock (meaning he paid cash to exercise) then he would have lost any optionality in redeeming the preferred. It would have simply been a 6% perpetual fixed return redeemable at the bank's discretion and not his. It's a nice return for the time being, but I'm sure he doesn't want that capital tied up indefinitely.

He's probably referencing when they bought back over $1 billion in stock from a single shareholder right before tax rates went up. It saved the shareholder money and probably was necessary for estate planning, while it was done at a price that was cheap enough to benefit the remaining shareholders. I guess it would be any time a shareholder is required to sell anyway and the stock is trading at a significant discount to intrinsic value. Such situations are probably few and far between but they could come up again during market panics or before major reforms take effect.

redskin, What's your actual basis [Annual Report 2016] for posting this? The warrants are listed in as an other investment in note 5. They are listed at fair value, and this amount is included in the insurance segment balance sheet. Notes to Consolidated Financial Statements (Continued) (5) Other investments Other investments include preferred stock of Bank of America Corporation (“BAC”), warrants to purchase common stock of BAC and preferred stock of Restaurant Brands International, Inc. (“RBI”) and in 2015 also included preferred stock of Wm. Wrigley Jr. Company (“Wrigley”) and The Dow Chemical Company (“Dow”). Other investments are classified as available-for-sale and carried at fair value and are shown in our Consolidated Balance Sheets as follows (in millions

For Buffett using NOLs? Buffett mentions using Dempster NOLs in the partnership letters. The Berkshire NOL you can find in the early annual reports.