Warren Buffett Re-Hedged His Bet

Also pay parity, VIX inquiries, sovereign immunity and the crypto.

Oh Warren.

Warren Buffett published his annual letter to Berkshire Hathaway Inc. shareholders on Saturday, and I guess we have to talk about it, even though it was kind of boring. "It was his shortest missive in more than two decades," points out my Bloomberg Gadfly colleague Tara Lachapelle, and it contained only a single off-color reference to the sex lives of teenagers.

Key Takeaways From Warren Buffett's Shareholder Letter

My favorite bit of the letter is a little story from five years ago. It's about Buffett's December 2007 bet with Protégé Partners, which bet that five funds of hedge funds that it selected would outperform the S&P 500 index over the next 10 years. Buffett took the other side and won in convincing, definitive, embarrassing fashion. (Protégé's Ted Seides explained himself here at Bloomberg View last year.) The letter includes the performance figures for the five funds-of-funds, all five of which underperformed the S&P, and one of which managed to return just 2.8 percent total over nine years (0.3 percent per year) before shutting down in the tenth year of the bet. (The S&P neatly returned an average of 8.5 percent a year over those 10 years, which as Buffett cannot resist pointing out is a very normal performance despite the abnormal experience of living through those years: "If a poll of investment 'experts' had been asked late in 2007 for a forecast of long-term common-stock returns, their guesses would have likely averaged close to the 8.5% actually delivered by the S&P 500.")

The way the bet worked is that each side would put up $500,000, and the winner got to distribute the $1 million to the charity of its choice. But it was a 10-year bet, and you don't want to leave that uncollateralized: Who knows if Protégé or Buffett would be good for the money in 10 years? On the other hand you don't want to collateralize it with just cash in envelopes, because that is inefficient. There is a traditional approach to collateralizing long-dated derivatives transactions, which is that you buy Treasury bonds, or even better Treasury strips, zero-coupon Treasury securities that mature at exactly the right time in exactly the right amounts. And that's what they did: Buffett and Protégé each paid $318,250 to buy strips maturing at $500,000 in 10 years.

And then after five years Buffett got antsy:

By November 2012, our bonds – now with about five years to go before they matured – were selling for 95.7% of their face value. At that price, their annual yield to maturity was less than 1%. Or, to be precise, .88%.

Given that pathetic return, our bonds had become a dumb – a really dumb – investment compared to American equities. 

What? Wait. That's not -- that's not what collateral is. The point of collateralizing a derivatives transaction is not to beat the stock market with your collateral. (By the way I'm not sure that the point of investing in hedge funds is to beat the stock market, either, but I didn't make this bet, Protégé did.) The point is to know with certainty that the collateral will mature in the amount that you need. You can't just go replace it with an index fund; that would create a risk that you might not have enough money. But Buffett didn't just replace the strips with an index fund:

Presented late in 2012 with the extraordinary valuation mismatch between bonds and equities, Protégé and I agreed to sell the bonds we had bought five years earlier and use the proceeds to buy 11,200 Berkshire “B” shares.

Look: It's Warren Buffett. It's a bet made largely for entertainment purposes, in minuscule size relative to his net worth. It worked out great ("Girls Inc. of Omaha found itself receiving $2,222,279 last month rather than the $1 million it had originally hoped for"). Buffett personally guaranteed the $1 million if there were any shortfalls. It was not a huge risk. But if you have an actual derivatives trade with a public company's chief executive officer, and he comes to you five years into a 10-year trade asking to swap the U.S. Treasuries collateral for his own company's stock, the traditional approach is to run away screaming. Only Warren Buffett could get away with that sort of thing. For a man who famously called derivatives "financial weapons of mass destruction," the guy really loves financial engineering.

One other thing from the letter: Warren Buffett is worried about GAAP accounting. Specifically he is worried that Berkshire Hathaway has adoptednew provision of U.S. generally accepted accounting principles requiring him to mark to market unrealized gains on his marketable stocks through the income statement. Since a big part of Berkshire Hathaway is, you know, railroads and insurance and truck stops and stuff, and another big part of Berkshire Hathaway is the Warren Buffett Stock-Picking Show, this is confusing: A railroad-and-truck-stop company would be measured based on its operating performance; a pure stock-picking fund would be measured on the mark-to-market performance of its stocks; but combining the two might obscure more than it clarifies. And he wants to make sure that investors aren't paying too much attention to the GAAP numbers, which he regards as misleading:

We will take pains every quarter to explain the adjustments you need in order to make sense of our numbers. But televised commentary on earnings releases is often instantaneous with their receipt, and newspaper headlines almost always focus on the year-over-year change in GAAP net income. Consequently, media reports sometimes highlight figures that unnecessarily frighten or encourage many readers or viewers.

Next time you read someone worrying about non-GAAP accounting and warning you that a company's suggested adjustments represent fantasy numbers, just remember that folksy ol' Warren Buffett thinks that his own GAAP numbers make no sense and hopes you will adjust them.

People are worried about non-GAAP pay parity.

Good news!

JPMorgan Chase & Co. said Friday that its female employees earn 99 percent of what male employees make globally, making it the fifth large U.S. bank to disclose an adjusted gender pay gap of around one percent.

So the average woman at JPMorgan makes just 1 percent less than the average man at JPMorgan? Hahaha no, quite the opposite, quite the opposite:

The new disclosures anticipate the looming deadline for all companies employing more than 250 people in the U.K. to publish their unadjusted gender pay gap numbers for their British employees -- a requirement that will include the big U.S. banks.

While the recent spate of voluntary disclosures have highlighted very small adjusted discrepancies, JPMorgan seemed to be warning employees that its U.K. filing may paint a different picture.

“The bare numbers, excluding these types of factors, will show a gap between the pay of men and women,” the firm wrote in its memo. “But we have found that employees are paid appropriately when taking into consideration their business area, their experience and the work they do.”

No, what JPMorgan's numbers show is that if you use JPMorgan's methodology for determining what pay is appropriate for a given business area, type of work and amount of experience, then men are paid 100 percent of what is appropriate and women are paid 99 percent of what is appropriate. "As a growing number of financial firms reveal whether men and women are compensated equally, they have clustered around 99 percent parity, after adjusting for factors such as job role, seniority and locale." It is a suspicious number. Really if you are doing the adjusting, you should get to 100 percent. If JPMorgan decided, based on her role and seniority, to pay a woman $500,000, and also decided, based on his role and seniority, to pay a man $800,000, then surely it also decided that she is getting paid 100 percent of what is justified by her role and seniority and that he is getting paid 100 percent of what is justified by his role and seniority. Therefore she gets paid 100 percent as much as him, as adjusted. No? Obviously if some third-party validator was doing the adjusting, you might expect its adjustments for role and seniority not to coincide with how JPMorgan pays for role and seniority. But JPMorgan is doing the paying, and it is also doing the adjusting. Why did it conclude that its own pay decisions were wrong? By 1 percent? Why did every other bank conclude the same thing?

One suspects that the answer is that if the banks all said "our women are getting paid 100 percent as much as men" then everyone would think they were lying. Especially when they also have to announce the much larger unadjusted pay gaps: so far 33 percent for Lloyds Banking Group Plc, 37 percent for Royal Bank of Scotland Group Plc and about 50 percent at Barclays. But 99 percent is as close as you can get to 100 percent while still quietly confessing that you pay women less. This way JPMorgan's head of human resources can say "We know we can always do more, and we will," and sound like she's being contrite about that 1 percent gap. (And not the much larger actual gap that takes into account the smaller number of women in senior positions.) If the adjusted number was 100 percent then how would they know they had to do more?

Really some bank should announce that the number is like 103 percent -- that its women, adjusting for seniority and role and so forth, are slightly overpaid relative to its men. The fact that they don't do this -- that the numbers are always slightly under, never slightly over, 100 percent -- is a good indication of how embarrassed they are about the whole adjusting process.

Incidentally I do not mean to pick on JPMorgan here; in many respects they are better than average. Here is a story about how Marianne Lake, JPMorgan's chief financial officer, "is one of the most senior women on Wall Street," is "on the bank's short list of possible successors" to Jamie Dimon, and has become "an even more likely contender" after a recent executive reshuffling. 

VIX, etc.

Well this is unsurprising:

U.S. regulators are scrutinizing this month’s implosion of investments that track stock-market turmoil, including whether wrongdoing contributed to steep losses for VIX exchange-traded products offered by Credit Suisse Group AG and other firms, several people familiar with the matter said.

The Securities and Exchange Commission and the Commodity Futures Trading Commission have been conducting a broad review of trading since Feb. 5, when volatility spiked and investors lost billions of dollars, the people said.

How could they not be? I have gotten like a dozen complaints about wrongdoing in VIX-related trading on Feb. 5, and I'm just a guy who writes on the internet. Surely the SEC has gotten hundreds. I cannot say that any of the complaints have received struck me as exactly a smoking gun, but then I don't have subpoena power. Perhaps the SEC or CFTC will dig up wrongdoing. I'd be a little surprised if Credit Suisse's XIV product could have lost 96 percent of its value in a day of wild trading without anyone getting in trouble. Just seems like the sort of thing that would ... cause ... trouble? I don't have a legal theory for that -- I think that very natural, fully disclosed and perfectly legal hedging of those products could have caused the wipeout -- but it does seem like blowing up a retail exchange-traded product in a day ought to land someone in hot water somehow.

Meanwhile, here is a good post from Craig Pirrong on volatility-related products, particularly the inverse volatility products that blew up:

These markets exist for a reason–to transfer risk.  Moreover, they behaved exactly as expected, and those who participated got–and paid–in the expected way.  Insurance sellers (those short volatility futures) collected premiums to compensate for the risk incurred.  Most of the time the risk was not realized, because of its “spikey” nature, and those sellers realized positive returns.  When the spike happened, they paid out.  There is never a free lunch.  Yes, the insurance sellers dined out on somebody else most of the time, but when they had to pick up the tab, it was a big one.

I find that people are often skeptical of "insurance" rationales for financial products, but really insuring against volatility makes a lot of sense. In a sense all insurance is insurance against volatility; it is meant to smooth cash flows in more volatile states of the world. You might as well just insure it directly. "Absent these markets," Pirrong writes, "the short volatility exposures wouldn’t go away: those with such natural exposures would continue to bear it, and would periodically incur large losses."

Sovereign-immunity rental.

Financial engineers are always looking around for new products, which means new sets of abstract entitlements to slice up and sell. One abstract entitlement that enjoyed a certain vogue in recent years is American Indian tribal sovereign immunity. The idea is that tribes are exempt from suits under certain U.S. state and federal laws, and so if you want to be exempt from those suits you should find a way to rent their immunity.

This became a popular approach to payday lending: An online payday lender could enter an agreement with an American Indian tribe in which the tribe became the notional lender, exempting the arrangement from state consumer-protection laws. (The online lender would do most of the actual work and keep most of the actual profits; the tribe would basically get a licensing fee.)

And it went mainstream last year when pharmaceutical company Allergan Plc transferred some of its drug patents to the Saint Regis Mohawk Tribe and then licensed them back from the tribe. Basically Allergan would continue to make the drugs exactly as it had before, but it would pay the tribe a fee. In exchange, the tribe would "own" the patents and assert sovereign immunity as a defense against certain patent invalidity claims. 

But the brief golden age of sovereign-immunity rental seems to be over. Payday lending tycoon Scott Tucker ran his business through American Indian tribal fronts -- "the tribes got 1 percent of the revenue" -- but it didn't work: He was sentenced to 16 years in prison for fraud last month.

And now Allergan's trade has been busted too, as the Patent Trial and Appeal Board refused to throw out a patent challenge based on sovereign immunity. Here is the board's decision, which is pretty boring and straightforward. "Based on the terms of the License between Allergan and the Tribe, we determine that the License transferred 'all substantial rights' in the challenged patents back to Allergan," ruled the board, meaning that Allergan remained functionally the holder of the patents and so could not assert sovereign immunity. A bare rental of sovereign immunity is too obvious a trick to actually work.

Crypto crypto crypto.

Here is Nellie Bowles on women in cryptocurrency:

Now, some early female investors and entrepreneurs are beginning to sound the alarm and push back.

“Women, consider crypto,” Alexia Bonatsos, a venture capitalist, wrote on Twitter. “Otherwise the men are going to get all the wealth, again.”

I have a half-baked, three-quarters-joking theory of cryptocurrency, which is that it is a magical incarnation of a sort of male internet grievance. People -- mostly men -- sit around on Reddit complaining that they are underappreciated geniuses and that it is unfair that they have not been rewarded with vast wealth. They feel dispossessed and betrayed: They expected the modern world to reward computer literacy, but then they grew up to realize that the modern world, much like the old world, rewards mostly people skills and creativity and emotional intelligence. And then Bitcoin came along, and paranoid computer-literate people who spent a lot of time on the internet were the early adopters, and it became the world's first economic system that allocates wealth basically for hanging around on Reddit. What Bonatsos describes is not an accident; cryptocurrency seems almost custom-designed as a way for the men to get all the wealth, again. 

I know you are going to email me to complain about this theory, but what I want to propose here is: What if you didn't?

Elsewhere, here is the great Elaine Ou on "Bitcoin as a Display of Wealth":

One of the greatest tragedies of modern money is the decoupling between a store and display of wealth. Or, more accurately, the societal decoupling between wealth and status. For most of human history, the functions of display and storage were condensed. Even before people wore clothing, they wore piercings and tattoos. A full-body ink job isn’t transferable, but it’s a sort of proof of work.

And here is Tony Yates on "The Petro and the Assignat."

Things happen.

People are worried about margin lending. Dropbox Files for I.P.O., and Other ‘Unicorns’ Are Watching. Mifid II and the return of the ‘star’ analysts. Private equity buyouts running at fastest rate since crisis. Dollar-Rate Breakdown Exposes Foreign-Exchange Mystery. Deutsche Bank lines up €2bn float of asset management unit. New Tax Law Could Spur Swap Meet for Used Business Equipment. Weinstein Co. to File for Bankruptcy After Deal Collapses. Options of Last Resort. Free Candy Creates Wild Frenzy at Wall Street Event. Former Freemason, 51, found drunk and naked inside a huge pipe organ with a toy gun and remote-controlled police car says he got lost while trying to hand out cheeseburgers to the homeless. Martin Shkreli got buff in prison. 

If you'd like to get Money Stuff in handy email form, right in your inbox, please subscribe at this link. Thanks! 

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

    To contact the author of this story:
    Matt Levine at

    To contact the editor responsible for this story:
    Brooke Sample at

    Before it's here, it's on the Bloomberg Terminal.