The Crisis Next Time and the Mergers This Week

Also: Tax shelters, convertible bond accounting, knockout options, and nostalgic floor traders.

The next crisis.

Here's Jamie Dimon's letter to investors, part of JPMorgan's annual report package. Of particular interest this year is Dimon's rather unnerving narration of the next financial crisis, on pages 30 to 34. You can be skeptical; Dimon is talking an anti-regulatory book, and it seems sort of unlikely that the next crisis will play out like the preview in a big bank's annual report. Still it's a good discussion of some of the possibly unintended consequences of the modern banking system. Here's a summary. A couple of points:

  • In a crisis, people dump risky assets and buy safe assets, but the demand for high-quality liquid assets in the next crisis will be tough to handle. Banks will be reluctant to reduce their (newly required) liquidity reserves, and will also be reluctant to accept new deposits because of increased capital requirements. And a lot of Treasuries are tied up in foreign exchange reserves, at the Fed, or in banks' liquidity reserves, leading to a shortage of safe assets for investment and repo.
  • Banks will also be more reluctant to lend, or buy risky assets, because procyclical capital rules (risk-weighted assets go up when risk goes up, and "new regulatory capital rules require losses on investment securities to reduce regulatory capital") "would force banks to hoard capital."
  • Non-bank lenders, who make up an increasing share of the lending market now, "will not continue rolling over loans or extending new credit except at exorbitant prices that take advantage of the crisis situation," because they have no sense of responsibility, unlike the big banks, which in the last crisis "understood their vital role in actively lending to clients."
  • Markets generally will be more volatile as regulation drives banks away from market-making. (There is a mention of last October's "once in every 3 billion years or so" 40-basis point move in Treasury yields that some might find hard to stomach.)

The general upshot is that banks will be safer in the next crisis, but will provide less support to the system they're supposed to serve. It is hard to know, in this generic outline, whether that's good or bad. The problem with a crisis that blows up banks is that asset prices plunge and banks stop lending; saving the banks at the cost of plunging asset prices and reduced lending seems like a Pyrrhic victory. On the other hand the hypothesis here is that there's a crisis, and the question is how best to bear it; it's quite possible that leaving more risk on the broad market and concentrating less of it in banks will lead to better outcomes.

Also good: On page 17, Dimon explains why the Fed's stress test underestimates JPMorgan's ability to respond to a crisis. In an actual crisis, JPMorgan "would be far more aggressive on cutting expenses, particularly compensation," "would quickly cut our dividend and stock buyback program," "would not let our balance sheet grow quickly," and would have lower trading losses than are provided for in the Fed's stress test. Obviously Dimon is biased to think that his bank would be smart and skillful in a crisis, while the Fed must take a more cynical view, but I am generally sympathetic to Dimon's take. The stress tests assume a certain level of stupidity -- Dimon says, "To make sure the test is severe enough, the Fed essentially built into every bank’s results some of the insufficient and poor decisions that some banks made during the crisis" -- and there's a certain amount of planning you can do to avoid that.

The next pharma merger.

I say sometimes that pharmaceutical companies have all the fun in mergers and acquisitions, and it continues to be true. The latest is Mylan's big offer to buy Perrigo for $28.9 billion, which I guess you would characterize as a hostile offer? Here is Mylan's not-un-bear-huggy letter and press release -- the letter to Perrigo is dated April 6, and it was released publicly two days later -- and here is Perrigo's noncommital we'll-take-a-look response. One question is, with all the previous deal activity in the sector, are all the potential white knights exhausted? Or is it more like, the bigger you get, the more capacity you have to swallow yet more companies?

Here is a breakdown of the banks involved in the Mylan bid, as well as Royal Dutch Shell's bigger deal yesterday to buy BG Group. Goldman Sachs, where I used to work, was on both deals, so that's nice for them, but yesterday was even nicer to Robey Warshaw LLP, "a two-year-old investment bank with less than ten employees," which advised BG in the Shell deal and is now rich. The New York Times estimates Robey's payday at $50.3 million; Bloomberg News's estimate is a bit less (a "smaller chunk" than Goldman of "as much as $90 million in fees").

Speaking of that Shell deal, BG's chief executive officer Helge Lund just started work in February and "could pocket as much as $43 million for a year’s work after Royal Dutch Shell Plc agreed to buy the company," since he has change-of-control protections in his contract and probably won't have a job at the combined company. People -- including here at Bloomberg View -- always find these change-of-control provisions unfair, and I suppose in some broad sense they are, but I don't really get the objection. He was promised a lot of money for working at a hard job that he was expected to be good at. Then a good merger offer came along that would kick him out of that job. Should he have turned it down to keep his high-paying job? Or should he have taken the deal -- which made over eight billion pounds for BG shareholders in one day -- and been paid out for his troubles? Surely the shareholders are happy to give him like half of one percent of the value he added to their shares? I mean, Robey Warshaw might be making more money off this deal than he is, and they don't have to stick around and run the company for the next few months.

A downgrade.

Here is the story of an analyst at Goldman Sachs downgrading WebMD, then realizing that he'd double-counted its convertible bonds as both debt and equity in his enterprise-value calculations, then taking back his downgrade and raising his price target to well above the current price. Oops! Here is the correction. I used to work at Goldman (disclosure!), but I also used to market convertible bonds to companies; convertibles have somewhat complicated accounting, and potential issuers would often ask if analysts and investors would be able to figure out the economic substance despite the weird accounting. My answer was, like, I hope so? But sometimes not? Really accounting for any security more complicated than common stock or straight debt can be a little confusing, and a troubling underpinning of the efficient markets hypothesis is its assumption that investors and analysts can answer basic questions like, "If this company is worth $X billion, how much is one share worth?"

A tax shelter.

The classic form of a tax shelter is that you buy losses to offset your income. One way to get losses is to lose money -- like, buy a stock, watch it go down, etc. -- but this is inefficient. A dollar of tax losses is worth around 35 cents (or whatever your all-in marginal tax rate is), so losing a dollar for a dollar of tax losses is a terrible trade. But if someone else has some losses that she can't use, she can sell you those losses for less than 35 cents on the dollar, and everyone is better off. Except the Internal Revenue Service obviously.

Here is a beautiful pure tax shelter: If you have gambling/lottery winnings, you can offset gambling/lottery losses against them. But if you just have losses -- if you buy a lot of lottery tickets that don't pay out -- then you can't deduct them from your income. So lottery losses are only useful to people with lottery winnings. So if you have losing lottery tickets, you should sell them to lottery winners on Craigslist. So people do. Sometimes they get caught -- this is not, like, allowed -- but here let us just admire their understanding of the tax code. I'm particularly fond of them because my own juvenilia includes some writing about lottery taxation.

Knockout options.

Here's a proposal from executive compensation lawyer Jeremy Goldstein (a former colleague of mine) that executives should be paid in the form of knockout stock options:

The knockout option provides that the option has its normal vesting features and term, except that the option will automatically be forfeited or expire if the price of the stock subject to the option decreases below a certain threshold. For example, a ten-year option to purchase 10,000 shares at $100 per share would automatically be forfeited or expire, whether or not other vesting conditions are met, if the price of the underlying shares falls below $50 per share.

Goldstein argues that the advantages of the knockout option are reduced compensation expense (a knockout is worth less than a regular option on a model basis, but provides the same performance incentive) and avoiding a "heads I win; tails I don't lose" situation for the executive. I'd add that some people worry that stock options can encourage excessive risk-taking, since an option is worth more as volatility increases. (Other people think this is a feature, not a bug, since executives are otherwise too inclined to avoid risk.) Knock-out options, though, are less sensitive to volatility: As volatility increases, the increasing value of the option is offset by the increasing risk of the knock-out. (Taleb says, "as volatility increases, the barrier gets flat vega while the vanilla retains its effect.") So a knock-out option is a nice way to encourage executives to try to improve performance while being sensitive to risk.

The pits.

Here is a fond reminiscence from a former Chicago Mercantile Exchange floor trader who is sad that the Merc is closing its futures pits, which, he says, "gave my brother and me, and so many others, a place in a local rough-and-tumble that helped Chicago reshape world markets." Here's how he describes Chicago Board Options Exchange floor traders:

They rendered fortunes from a witless investing public that lacked an understanding of options’ mechanics and the technology needed to price them.

And here's a story from his days in the Merc pits:

One day in January 1989, I showed up to work in the cattle futures options pits to find them eerily empty. The night before, several traders had been arrested in an FBI sting that aimed to uncover illegal collusion and market manipulation. Apparently, I had missed a widespread scheme. 

Remember, when people moan about how rigged modern electronic markets are, that they are much better than the charming humans that they replaced.

Things happen.

"Bonds with negative yields have become one of the world’s fastest growing asset classes, accounting for around a quarter of Europe’s government debt market": Switzerland did a 10-year below zero, and Mexico did a 100-year bond (in euros) at 4.2 percent. Goldman may be doing well on bond trading. "One could fairly worry that this proxy fight represents the jump-the-shark moment for activism." Argentina is still really mad at Citi. Dick Bove is excited about some Fannie Mae news; John Carney is not. SQZZ, the short-squeeze ETF. "Yes, maybe our purpose is to help participants in underfunded pensions receive the benefits they've been promised," says departing hedge fund employee who doesn't really believe that. The rise of the diva nanny. Salman Rushdie didn't like "Lucky Jim."

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

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

    To contact the author on this story:
    Matt Levine at

    To contact the editor on this story:
    Zara Kessler at

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