The beginning of a new year seems to be a popular time to check in on Warren Buffett for some reason, perhaps because a lot of people's New Year's resolution is "invest more like Warren Buffett"? Sure, why not. So how's he doing?
Warren Buffett’s bet on Bank of America Corp. and a more-generous buyback plan helped his Berkshire Hathaway Inc. beat the Standard & Poor’s 500 Index in a year when he didn’t make a major acquisition.
Class A shares advanced 17 percent last year, beating the 13 percent gain in the S&P 500. … Book value may have climbed to $113,579 a share on Dec. 31, according to an estimate from Meyer Shields, an analyst at Stifel Nicolaus & Co. That would give Buffett’s firm a 7.8 percent annual growth rate for the five years ended 2012, compared with 1.7 percent for the S&P 500, including dividends.
Yay. It's worth noting that his buyback was controversial when it was announced like three weeks ago, but the stock is up by 6.5% since then, yielding Buffett a three-week $80 million profit.
Buffett has also made untold gazillions of dollars on his investment in Bank of America preferred and warrants in 2011. I feel like this passage, from today's "Heard on the Street" column dithering about whether BofA should buy back the preferred stock, provides some clue as to why:
Granted, retiring Mr. Buffett's preferred shares might cause BofA to cut back on plans to ask for capital returns. That could cause unease for common shareholders, especially since Mr. Buffett's preferred stock is also equity.
Yet the particular structure of the preferred shares means it doesn't qualify toward the bank's Tier 1 ratio under the new Basel rules. So it is akin to debt for capital purposes.
On that basis, 6% is expensive. The average yield on BofA's long-term debt as of the third quarter was 3.07%. And the bank, awash with deposits, has been aggressively reducing long-term debt as it seeks to bolster its net interest margin.
So it wouldn't be tough for BofA to replace Mr. Buffett's preferred stock with cheaper, long-term debt. The catch: Any additional debt issuance chips away at the bank's net interest margin, while the preferred payment doesn't because it is deducted from net income.
A thing that exists on the internet is this video of Bill Ackman explaining investing by talking about how he's gonna open a lemonade stand1, and a fun exercise is to try to explain what Bank of America is thinking about in its ponderous hive-head using only terms that would be understandable to someone who knew only the financial knowledge in that video. Like: is Bank of America looking to buy low and sell high? Or is it looking to bolster its net interest margin, or its tier 1 capital, which are not actual things?
If you were, I dunno, Warren Buffett, you might ask a question more like: is buying this preferred a good economic deal, compared to other things that Bank of America could do with its money? That seems like sort of a knife-edge question actually,2 but you definitely get the sense from the Journal that it's, like, twelfth on BofA's list of questions. This is not particularly because Buffett is smart and BofA is dumb. It's because Bank of America is, y'know, a bank, of America, and sits in the center of a tangle of regulatory and accounting knots and tries very carefully to unravel one or two of them at a time. Warren Buffett is a regulatory regime unto himself.3
Coming and going, Bank of America's motives in trading its preferred with Buffett were and will be pleasingly - to Buffett's eye - uneconomic. The initial trade, remember, was conceived in a bathtub and, economically speaking, looked pretty gifty to Buffett, though in exchange BofA got the whole "we're-probably-not-vanishing" halo effect he's selling these days. And the exit, it seems, will be conceived in a flurry of shading accounting and regulatory capital concepts in the right direction. And so each way Buffett will be overpaid, because his counterparty is not economically motivated. It's just a bank. Why would a bank be economically motivated?
1.... and then charge little kids fifty bucks to come to work selling lemonade for him, but they can totally make a million dollars selling the lemonade, even though not that many will, but some will anyway, and even if they don't they'll get a distributor discount on the lemonade mix, so they can always just drink it, and, wait, no, fuck, totally different Bill Ackman video.
2.BofA would have to pay 105% of par to buy back the 6% preferred, giving a repurchase yield of about 5.7%. Is 5.7% cheap for BofA preferred stock? Well, BofA has publicly traded preferreds; I eyeball BofA's 6.625% Series I (non-cumulative, exchange-listed, perpetual) preferreds at 6.3%, and the callable Series Ds at just about par. And BofA has recently been calling prefs with coupons as low as 5.875%. so best guess is that the "right" yield for Buffett's cumulative pref is around 5.75% to 6%, making a 5.7% repurchase yield a little expensive from BofA's perspective but not crazy.
You can compare further afield. Even the longest-dated BofA straight debt trades with a four handle, and is tax deductible, unlike this thing, meaning that the after-tax cost of 30-year debt is like half the after-tax cost of the Buffett preferred. Or you can compare to buying back stock. BAC is like 12% below tangible book value; if it gets back to TBV in the next year then buying common stock was in some sense twice as good an economic proposition as buying Buffett's preferred.
3.And you can conceive of him, verrrrry loosely, as making some of his money by renting that regime to banks. That may be a moral of the BofA story, and perhaps of this weird Deutsche Bank story.