Smith's key point is that the 2005 bankruptcy law gives owners of derivatives ("derivatives counter-parties") a ticket the the head of the queue in a bankruptcy. Her argument is that these provisions mean that the counter-parties will grab the assets of the the bank, leaving the FDIC with the debts and the cost of guaranteeing deposits. But that's something the Dodd-Frank bill fixes. Here's Smith:
This move reflects either criminal incompetence or abject corruption by the Fed. Even though I’ve expressed my doubts as to whether Dodd Frank resolutions will work, dumping derivatives into depositaries pretty much guarantees a Dodd Frank resolution will fail. Remember the effect of the 2005 bankruptcy law revisions: derivatives counterparties are first in line, they get to grab assets first and leave everyone else to scramble for crumbs. So this move amounts to a direct transfer from derivatives counterparties of Merrill to the taxpayer, via the FDIC, which would have to make depositors whole after derivatives counterparties grabbed collateral. It’s well nigh impossible to have an orderly wind down in this scenario. You have a derivatives counterparty land grab and an abrupt insolvency. Lehman failed over a weekend after JP Morgan grabbed collateral.The Dodd-Frank resolution provisions specifically permit the FDIC to (a) block "derivatives counterparties" from collecting collateral and (b) to move derivatives and other debts to a "bridge bank" which they create at the moment. This process is described in section 210 of the Dodd-Frank act - an act that specifically states that its provisions overrule any counter provisions of bankruptcy law. So the law anticipates that the FDIC will move derivative transactions to a new company where derivatives will "net out". There is extensive discussion of this provision available online - for example here and here. Clearly, it is possible that the FDIC will not do its job properly, that the law won't work out, or that there are problems in the resolution authority - but more clearly, it's just plain wrong to cite the 2005 Bankruptcy act as the controlling law.
It's also worth noting that most banks already have their derivatives in the bank and not in a trading company subsidiary and that even for BoA, most the of the derivatives exposure is already in the bank, not in the holding company or the subsidiary. Here's the original Bloomberg article with bold added to make the point clear.
Bank of America’s holding company — the parent of both the retail bank and the Merrill Lynch securities unit — held almost $75 trillion of derivatives at the end of June, according to data compiled by the OCC. About $53 trillion, or 71 percent, were within Bank of America NA, according to the data, which represent the notional values of the trades. That compares with JPMorgan’s deposit-taking entity, JPMorgan Chase Bank NA, which contained 99 percent of the New York-based firm’s $79 trillion of notional derivatives, the OCC data show.What this means is that BOA already had most of its derivatives in the depository bank and it seems to be standard practice to do so. In response to a reader pointing this out, Smith writes:
You can argue that this is just normal business, the other big banks have their derivatives operations largely in the depositary. But BofA has owned Merrill for over a year and a half, and didn’t undertake this move until it was downgradedAnd that's correct, but why was BofA downgraded? It was downgraded by Moody's because as the government agencies issue rules under Dodd-Frank, it has become more clear that the government will not bail out bondholders of the "too-big-to-fail" (TBTF) banks anymore. Here's what Moody's says:
Moody's decision to assign a negative rating outlook reflects the possibility it may further reduce its systemic support assumptions in the future as a consequence of the process set in motion by the enactment of the Dodd-Frank Act. Under the rules recently finalized by the FDIC, the orderly liquidation authority included in Dodd-Frank demonstrates a clear intent to impose losses on bondholders in the event that a systemically important bank such as BAC was nearing failure. If fully implemented, the provisions of Dodd-Frank could further lower systemic risk by reducing interconnectedness among large institutions and could further strengthen regulators' abilities to resolve such firms.So, what's happening is the direct opposite of what Smith says: Dodd-Frank is removing the implicit government guarantee that subsidized the TBTF banks and Dodd-Frank protects the public against costs of derivatives - that we were previously on the hook for. There is a good discussion in the comments at Salmons and also see this.
Finally, if that $75 trillion money at stake seems ridiculously large - that's because it's not really the amount of money at stake. "Notional" values of trades do not tell you anything about the actual risk. The notional value is just the sum of the size of the underlying asset. For example, common interest rate swaps have a notional amount that reflects the size of a loan, but the risk is only the difference between the interest rate of the loan and the interest rate on an exchange like the London interbank rate. Furthermore when a bank sells a derivative contract with a notional value of $1M and then buys 4 contracts that are notional value $250,000 each that bet the other way, so it hedges its risk, the notional value is $2M even though worst case is at most $1M. There is a good explanation here (h/t Karoli). They have a great example which I'll just quote
Transaction Scenario 1: A derivatives transaction example that illustrates the relationship between notional amounts and potential risk. An American firm holds $2 million in cash. This company is planning to buy equipment from a German manufacturer six months from today that, given today's exchange rate, is valued at $2 million US Dollars (USD). Given its plans to purchase this equipment, its cash position in USD actually represents a foreign exchange (FX) exposure to the exchange rate between USD and German marks (DM). The firm calls a derivatives dealer and enters into a FX contract that obligates the firm to sell DM six months from today at today's exchange rate. In so doing, the American firm locks in the USD price of the equipment, thereby hedging its foreign exchange risk. If DM appreciate, the equipment becomes more expensive in terms of USD; but the firm profits in its derivatives contract by roughly the same amount. After entering the transaction with the American firm, the dealer bank enters into a contract with a hedge fund to sell DM six months from today at today's exchange rate, thereby completely offsetting its exposure to the exchange rate between USD and DM. In actuality, the exchange rates used on both sides of the transaction might be slightly different, allowing the dealer bank to make a spread on the transaction to compensate it for making a market in DM forward agreements. The hedge fund enters the transaction because its managers believe that DM will weaken against USD in the next six months and the fund is willing to speculate in order to profit from that movement. In terms of the gross notional amount of derivatives outstanding in the banking system, the transactions in this example contribute $4 million: the sum of both the bank's FX contracts. In terms of market risk, these transactions contribute nothing. There is, however, credit risk associated with the dealer bank's position. While the transaction is open, the exchange rate that the market expects between DM and USD may change. If DM depreciate (a change unfavorable to the American firm's FX derivatives position), then the contract between the dealer and the American firm increases in value to the dealer. This increase in value and potential future increases in value during the contract's life represent credit exposure of the dealer to the American Firm.PS If you really want an introduction to this try this via Alea at Salmon.