Wondering how the blow out in interest rates is impacting commercial banks, which just happen to have hundreds of billions in duration exposure in the form of various Treasury and MBS securities, not to mention loans, structured products and of course, trillions in IR swap, derivatives and futures? Wonder no more: the Fed’s weekly H.8 statement, and specifically the “Net unrealized gains (losses) on available-for-sale securities” of commercial banks, gives a glimpse into the pounding that banks are currently experiencing. In short: it has been a bit of a bloodbath.
After hitting a recent high of $34 billion in gains three months ago when interest rates were still near 2016 lows, the reported amount of net unrealized gains has tumbled, and from a gain it has turned into a loss of $14 billion as of the week ended December 14. On a 4-week rolling basis, the change amounts to $37 billion in losses, the biggest monthly drop since the 2013 Taper Tantrum.
This may not be the end of it: as the next chart below shows, commercial banks are holding just shy of an all time high of $747 billion in Treasuries and other non-MBS securities, a number which rises to $2.43 trillion if one includes all Treasury and agency securities on commercial bank balance sheets.
Should rates keep rising, the “unrealized” losses will keep building.
Where on the bank income statement do these losses appear? As we explained the last time this was an issue, in the aftermath of the 2013 Taper Tantrum, it comes down to the the Available For Sale (AFS) line, which runs through the Accumulated Other Comprehensive Income line.
It means that the November and December spike in rates will hammer those banks which hold their bond portfolios as AFS, and thus are subject to Mark to Market and ultimately flow through the P&L.
It also means that the shorthand to get a sense of how substantial the MTM losses from bond holdings will be is to look at the massacre that is going on in the AFS line and extrapolate it to all other levered commercial bank (and hedge fund) rate exposure. Expect math PhD-programmed algos that determine the marginal momentum of the S&P to figure this out some time over the next 2-3 weeks once banks begin reporting results which “unexpectedly” are well below expectations, especially since instead of steepening, the Net Interest Margin line has remained very much unchanged, and if anything, has modestly flattened, failing to offset the losses from bank holdings of rate-sensitive securities.
