Zero hedged banks

FT Alphaville dug into Silicon Valley Bank’s interest rate hedging strategy last month, and concluded that it was not quite as stupid as you’d think, but still pretty dumb.

Basically, banks shouldn’t (and usually don’t) hedge the rate risk in “Hold To Maturity” bond portfolios because in practice it would function as a directional bet. But they should do on any bonds held as “Available for Sale”, and SVB actually ditched their hedges to juice short-term earnings.

Now Lihong McPhail of the CFTC and Philipp Schnabl and Bruce Tuckman from NYU Stern have looked at the US banking system as a whole, by examining regulatory data on the interest rate swap positions held by the 250 biggest US banks. So how unusual was the lack of hedging?

Well, according to their paper — just published by the National Bureau for Economic Research — the banks had a nominal $434bn worth of swaps on their books, but if you net them out the overall economic impact is basically zero.

Here’s the abstract, with FTAV’s emphasis:

We ask whether banks use interest rate swaps to hedge the interest rate risk of their assets, primarily loans and securities. To this end, we use regulatory data on individual swap positions for the largest 250 U.S. banks. We find that the average bank has a large notional amount of swaps — $434 billion, or more than 10 times assets. But after accounting for the significant extent to which swap positions offset each other, the average bank has essentially no net interest rate risk from swaps: a 100-basispoint increase in rates increases the value of its swaps by 0.1% of equity. There is variation across banks, with some bank swap positions decreasing and some increasing with rates, but aggregating swap positions at the level of the banking system reveals that most swap exposures are offsetting. Therefore, as a description of prevailing practice, we conclude that swap positions are not economically significant in hedging the interest rate risk of bank assets.

It’s somewhat surprising that there is in practice no real hedging going on, as there are other banks with chunky AfS portfolios of high-duration bonds that are jerked around by shifts in rates. We gather that some banks will even have some hedges on HTM portfolios, just in case.

There are other ways to hedge interest rate exposure of course. And you could argue that cheap, sticky deposits is a pretty good natural one. SVB’s problem was the combination of an unusually large and long-duration securities portfolio that was virtually unhedged and a fickle depositor base. Still, the systemwide findings are interesting.

By the way, if you want more on the SVB debacle, our mainFT colleagues have published a fantastic autopsy today.