Skip to main content

Libor Manipulating Banks Used Baltimore's Tax Dollars To Help Pay Off Your Mortgage, Or Something

  • Author:
  • Updated:

There is a line forming to the left for people to beat up on Libor-manipulating banks, and it's a long line so your beating time is limited and you have to make the most of it if you want anyone to care. Today's the day for U.S. municipal borrowers. How'd they do?

The municipalities are important because they are the unusual case of a large class of politically sympathetic customers who would have been systematically disadvantaged by low Libor rates, as opposed to you and that mortgage that you won't shut up about, on which Liborgate probably saved you money. Stephen Gandel nicely sums up the situation here: the problem was that, in astonishing droves, U.S. cities and counties borrowed at variable rates, paying their own idiosyncratic floating SIFMA rate, but they then swapped to fixed, receiving a floating rate based on Libor. This led to badness, as muni credit blew up and SIFMA spiked, while bank credit blew up and Libor mysteriously didn't, because of the manipulating. So cities who had expected to pay a fixed 5% or whatever a year ended up paying 5% plus the suddenly widening gap between SIFMA and Libor.

Here is a graph I made you, comparing the SIFMA rate that munis paid to their bondholders versus a proxy for the Libor-based rate that they received from their banks*:

So that sort of looks okay outside of 2008, which looks sort of ... not okay. Here is perhaps a more suggestive thing:

  • From 1997-2004, the Libor-based rate averaged 6 basis points higher than the SIFMA rate
  • From 2005-2009, the Libor-based rate averaged 7 basis points lower than the SIFMA rate
  • From 2010-2012, the Libor-based rate averaged 4 basis points higher than the SIFMA rate

If you date your Libor manipulations to 2005-2009, and you are a municipality, you have I guess a plausible case for saying "those swaps cost us 11-13 basis points a year." The municipalities themselves calculate 20 basis points but that is because they are muppets.**

I am not a fan of most municipal carping about interest rate swaps because that carping is in the first instance of the form "rates went down and now we are paying a high fixed rate and receiving a low floating rate" and y'know what THAT HAPPENS and you wouldn't be complaining if rates had gone up and you should live in a world of expectations bro. That said you can't take that argument too far, because you do kind of get the sense that the munis were at least bulldozed a bit, no?

Like, maybe there's a good reason for decomposing a trade (borrow 30 years fixed) into a public component (borrow 30 years floating) and a derivative component (swap to fixed)? Obviously corporate issuers do related but opposite decompositions all the time (issue 10-year fixed, swap to floating) for market-receptivity reasons (taxable investors want fixed-rate bonds more than they want floating-rate bonds); I assume without knowledge that there's a plausible market-receptivity argument the other way in the municipal space. Certainly I wouldn't want to lend the City of Baltimore money at a 30-year fixed rate. Still it's always suspect when banks are earning most of their money not on disclosed underwriting fees but on swap P&L that, just guessing here, the City of Baltimore is not well equipped to calculate or second-guess.

More worryingly, it is hard to wrap your head around paying a floating rate that's based on your own credit while receiving a floating rate that isn't. Variable rate muni bonds seem to float based either on the SIFMA rate or on their own weekly auctions; in either case, you pay up when muni credit worsens, which is exactly when you don't want to, since presumably your credit is worsening for the time-honored reason of you don't have any more money. On the other hand, the floating payments that come in don't go up just because you're running out of money: they only go up because [actual risk-free rates are rising and/or] banks are running out of money, and banks have ways to avoid running out of money, or failing that to pretend that they still have money, that municipalities lack.

If banks manipulated Libor they were stealing from these municipalities, which is bad, but it's always worth keeping in mind that the total gross amount of Libor exposure dwarfs any bank's (or all the banks' together) net exposure: they have lots of Libor coming in, and lots going out, and they just clip a wee bit of it in the middle. So they were distributing the proceeds of that theft pretty widely. One fun way to think about Liborgate is to imagine the financial system, in theory a machine to intelligently use price mechanisms to intermediate the movement of money from savers to productive uses, as actually a machine to haphazardly use an imaginary-price mechanism to intermediate the movement of money from arbitrary Libor losers to arbitrary Libor winners. Barclays seems to have engineered the world so that Nassau County*** sales taxes were used to subsidize adjustable-rate mortgage borrowers in California. Was that a good idea? Beats me. Was it a conscious and defensible decision about the allocation of resources? I'm guessing no. That, and not "ooh bankers lie a lot," may be the real problem with Libor manipulation.

Wall Street's latest sucker: Your hometown [Term Sheet]
Rate Scandal Stirs Scramble for Damages [DealBook]

* SIFMA swap rate data from SIFMA. 3-month LIBOR from Bloomberg (USD0003M Index). Muni swaps tend to be based on 0.67 x Libor, rather than 1 x Libor, because taxes blah blah blah. 3mL versus 1-week SIFMA because I'm lazy but look how nice they line up no?

** Nah, there's lots of econometrics in the complaint, whatever. Twenty, or even 11, bps seems like a lot of manipulation to me - the Barclays bros were talking about half basis points and as John Carney points out they sometimes couldn't even get that - but maybe?

*** The county of my childhood, it goes almost without saying.