Should Berkshire Hathaway be designated a systemically important non-bank financial firm and subjected to Federal Reserve oversight, as the Financial Stability Oversight Council is considering? Oh I don't know. Obviously, insurers don't want to be subjected to Federal Reserve oversight; pretty much no one ever wants to be subjected to any oversight. I'm not entirely clear on what that oversight would entail, though the Fed might "impose stricter capital, leverage and liquidity requirements and demand stress testing for crisis scenarios." I don't know if that would be good or bad or irrelevant; so far Berkshire seems to have done a decent job of avoiding crises all on its own. Better than the Fed, even.
And it would sure be a shame if a systemic-importance designation took away Berkshire's ability to, I don't know, bet a billion dollars on some random numbers picked by a monkey rolling dice. That seems like the sort of thing you can get away with as a scrappy little $280 billion AA/Aa2-rated insurance company, but that gets a little bit more awkward once you're systemically important. AIG, which has received the systemic-importance designation, hasn't bet that much money on monkeys since it closed AIG Financial Products, ZING!
Maybe a simpler question is, is Berkshire Hathaway systemically important? Arithmetically the answer seems to be yes, or yes-ish:
non-bank financial companies that have $50 billion or more in assets and meet any one of five other criteria, including having $30 billion in credit-default swaps linked to their debt, can be evaluated.
Berkshire had $458.1 billion of assets as of Sept. 30, the company said in a filing with the SEC. It had $31.4 billion in credit-default swaps linked to its debt as of Jan. 17, according to data from the Depository Trust & Clearing Corp.
Berkshire also had $5.8 billion in derivative liabilities as of Sept. 30, more than the $3.5 billion trigger set by the FSOC.
That $30 billion in CDS criterion is particularly interesting: Why is there so much CDS on Berkshire? Well, there's only like $5.8 billion of net notional (that is, most of the $31.4 billion is offsetting trades within dealers), but that's still a lot, more than the net notional outstanding on systemically important issuers like JPMorgan, Citigroup, Bank of America, Deutsche Bank, Goldman Sachs, the United Kingdom and the United States of America.
Are people really into betting against Berkshire? Meh. If you wanted to bet on a really serious meltdown of the global financial system, I guess buying Berkshire CDS at say 70 basis points running might be a cheap way to do it. But, you know, who are you buying it from? If you're expecting the sort of global meltdown that bankrupts Berkshire, your Berkshire CDS starts looking dicey. Better to buy gold or farmland or ammunition or Dogecoins.1
More likely -- well, notice how close that $5.8 billion net notional is to Berkshire's $5.8 billion in derivative liabilities. There's probably a link there. Berkshire's insurance and insurance-ish businesses are about taking financial risks away from other people -- a Nomura analyst describes it as, "You’re taking volatility away from other people and accepting it to your own balance sheet" -- and some of those people obviously want to be sure that Berkshire will pay up on that insurance. Buying Berkshire CDS is a way to insure that insurance, as it were, though it raises the same "who are you buying it from" issue as a straight bet against Berkshire. (More practically, buying CDS is a way for banks to reduce CVA charges for capital purposes on their trades with Berkshire. Capital regulation doesn't care as much about the who-are-you-buying-it-from issue.)
So the CDS notional outstanding serves as a rough proxy for how interconnected Berkshire is with the rest of the financial system. And the answer is, relatively speaking, pretty big.
But there's no sense in measuring Berkshire's interconnectedness in purely mechanical ways. It's Berkshire! It's Warren Buffett! He's the frequent savior of the financial system! His mere stamp of approval -- typically in the form of a large investment in a risky-looking institution -- is enough to calm markets and bring firms back from the brink. Once Buffett has invested in a bank, the conspiracy theory goes, regulators will find it a bit harder to let that bank fail, because how can you look Warren Buffett in the eye and take away a toy from him?2 And if Buffett has the power to bestow halos on banks, then that's a decent argument that the Fed should be monitoring his halo-distributing activities for signs of excessive risk.
Because the main systemic thing about Berkshire is, come on, it's Berkshire Hathaway. It's got Warren Buffett and Graham and Dodd and Cherry Coke and weird annual meetings in Omaha and that 29-year-old who's in charge of everything and whose name is literally Cool. It's a piece of wholesome Americana incongruously deposited in the heart of the financial system. That's irreplaceable, and a collapse of Berkshire could destroy America's already near-zero trust in its financial system. You can see why the FSOC would think it's too big to fail.
1 By the way, if there's an argument of the form "If X goes bankrupt, then the entire financial system will probably be bankrupt, so you might as well stockpile weapons in your remote cabin," then I guess that itself is an argument that X is too big to fail? I mean, not strictly -- correlation with end times is not causation -- but still.
2 Salomon Brothers is arguably a counterexample, though not a clear-cut one.