Two bits of news are out today at the high and low ends of what you could loosely call big financial institutions. At the low end, Jefferies, which I'll stick with calling wee given its $40 billion balance sheet, is blasting out minute-by-minute, issuer-by-issuer, maturity-by-maturity, CUSIP-by-CUSIP accounts of its holdings of European sovereign bonds, its trading activity in those bonds, and the lunch orders of the traders trading those bonds. First it had no exposure - $2.4bn gross, $9mm net short notional, $37k of DV01, almost all cash with some futures - and then it had even less exposure, cutting gross exposure in half although apparently increasing net. The aggressive PR campaign seems to be working, with the stock basically where it was before anyone spent any time thinking about Jefferies, and up today in a down tape.
At the high end, Berkshire Hathaway, which is not entirely unlike a thinking man's AIG and has a $385 billion balance sheet, disclosed Friday that it lost two billion dollars last quarter in mark-to-market on its $34 billion notional of short S&P index puts. Also Berkshire is ramping up single-name equity investments without telling anyone what they are.
One more thing about BRK/A that you may or may not find related is that it may or may not be a "non-bank G-SIFI," that is, a financial institution that is not an FDIC insured bank but is nonetheless "too big to fail" because of its size and interrelationships:
Berkshire had $29.7 billion in credit-default swaps linked to its debt as of Oct. 28, putting it just under a $30 billion threshold proposed last month by the Financial Stability Oversight Council. AIG, now majority-owned by the U.S., had $45.3 billion in swaps written against it, and MetLife, the New York-based insurer, had $32.9 billion.
The council, responsible for deciding which non-bank financial firms are systemically important and require Fed oversight, plans to evaluate those that have $50 billion or more in assets and meet any one of five other criteria, including the credit-default swap threshold. Berkshire, AIG and MetLife all meet the asset minimum.
So, if you wanted to see Berkshire regulated more heavily, you could go write $300mm of CDS on them. Warren Buffett may or may not buy you a Blizzard / invest $5 billion in your struggling bank as a thank-you for doing so: on the one hand being officially too big to fail comes with pain, heartache, lawyers, Fed meddling, maybe prop trading limits, and maybe higher capital requirements. On the plus side, you get to not fail.
Jefferies and Berkshire look like two extremes in how to think about contagion risk in the financial system. On the one hand, you can be absurdly transparent to the market, listing all your holdings and hedges and counterparties and the risk measures for each of them, so that anyone who cares can go satisfy themselves that the risks are pretty tolerable - and, if they disagree, they can either call you up and say "hey, you've got a little too much long exposure to the Portugal 3-year than I like, wanna look into that?" or just short your stock. On the other hand, you can give nothing away but assure everyone that (1) lots of important regulators are making sure you're not running unreasonable risks, (2) you have a long track record of not running unreasonable risks, and (3) you're Warren Buffett for chrissakes, come on, relax.
There is a real argument that the Jefferies approach works for wee broker-dealers with little prop trading, where the competitive need to keep positions secret is less (since "These are not investments for us ... This inventory turns over an average of several times a week”) and where a cynic might worry that neither personnel nor regulators would be quite up to keeping an eye on the risks, while the Berkshire approach is the right one for giant basically-prop-trading-and-insurance businesses, where the whole secret sauce comes from concentrated stock-picking, the name brand is sufficient to attract and motivate smart and focused employees and regulators, and, y'know, you have Warren Buffett.
But it does seem a little backwards, doesn't it? We know the size, maturities, CUSIPs and DV01s of Jefferies's $1bn or so of gross European sovereign exposure. With bigger, more systemically important banks - Barclays, say - not so much. And here's what can be known about the exposure of the US banking system to European sovereign debt, which "lies somewhere between $46.4bn and $767.5bn." Great!
It is a somewhat hard life for clients and counterparties of the MF Globals and Jefferieseses of the world to have to monitor their dealers' exposure to risky assets on a CUSIP-by-CUSIP basis, and to have to make informed guesses about what if any derivatives lurk on their books. It is much easier to trust in the warm embrace of too-big-to-failness and regulatory oversight, coupled with a comforting helping of cheeseburgers and cherry Coke and Graham & Dodd. But if you were really worried about systemic risks magnified by interconnections among the financial system, by incentives that push customers to the biggest banks, and by regulators who are too close to the big companies they regulate to see the bigger picture, wouldn't the Jefferies approach look pretty neat? Wouldn't you be sad that it's only been adopted by Jefferies?
Buffett Broadens Portfolio by Investing $23.9B [Bloomberg]
Will Jefferies’ Open Kimono Tame the Market? [Deal Journal]