How do we fix LIBOR?
It’s kind of hard to believe, but it’s true: many of the problems that led up to the financial crisis are still with us and simply haven’t been addressed.
- There are too few banks, and they are too interconnected. This is still a huge problem, and it’s called “Too Big To Fail.”
- In fact they’re so big they can engage in criminal activity and not fear prosecution. Still true, and it’s got a name too, “Too Big To Jail.”
- Also, the credit rating agencies, who get paid by debt issuers for AAA ratings on crappy debt? They’re still alive, there are still only three of them, and they still rate debt.
- Also, remember the LIBOR rate manipulations? Still being run by asking individual traders what they’ve been paying recently, and the answers are still being taken on faith. Oh, and they’ve been asked “not be located in close proximity to traders who primarily deal in derivative products” based on Libor. That’ll work, because nobody know how to text!
I’d like to perform a thought experiment for this last one, because it seems like a solvable problem, although I will confess up front that I don’t have a solution off the top of my head. That’s where you readers come in.
Just a little background. LIBOR is (supposedly) the very short-term interest rate that banks pay each other for loans. A huge pile – hundreds of trillions of dollars worth – of derivatives in the form of futures, swaps, and loans are tied to the LIBOR, and most of them seems to reference the three-month rate. Here’s a NY Times graphic explaining how it works and explaining how the fraud played out at Barclay’s.
So what are the attributes of the benchmark that make LIBOR so important and so widely used? And how would we start using something else instead?
Let’s answer that second question first. If we could find another 3-month benchmark rate with good properties, we could start writing contracts based on the new rate from now on, while continuing to compute LIBOR until the existing contracts have played out and LIBOR would be grandfathered out of the market.
Now on to the first question, a list critical attributes of this rate.
- It’s supposed to reflect very short-term kinds of risk, which means you don’t want to base it on, say, long term treasuries.
- It should be public data, so we don’t have manipulation behind the scenes
- But it shouldn’t be based on a market that is so small that it is worth losing money on that small market to manipulate the new LIBOR rates. Remember, the derivatives market that uses that rate is enormous, so if we base the rate on a smallish if transparent market, that would just invite market manipulation for that small market.
- The point of LIBOR-based derivatives is that it’s a floating rate, which means that as credit gets tighter or looser so does the rate references in a given derivative. But of course LIBOR is a very bank-specific kind of credit. So it’s not just “as credit gets tighter or looser” but rather “as interbank credit gets tighter or looser” (and here I’m ignoring the manipulation, since when bank credit actually got tighter, it wasn’t actually reflected in LIBOR rates).
- So I guess my question is, do we actually want bank-specific credit rates? Isn’t it good enough to have a market-wide concept of credit tightness? For most people who own one side of those LIBOR-based derivatives, I’m sure their own access to credit matters more to them than some London bank’s access, although I’m also sure there’s a relationship between the two.
I’m asking you readers to put up suggestions or explain how we can do without LIBOR altogether.