Intuition behind the tenor basis spread in basis swaps

Question

Reading myself into basis swaps, I was wondering a couple of things. Say, one enters into a 1y - basis swap where party A agrees to pay 1M-LIBOR each month and party B agrees to pay 3M-LIBOR every quarter. As far as I understand, in an ideal environment where there is no credit risk, receiving monthly 1M-LIBOR vs receiving quarterly 3M-LIBOR are essentially the same thing? In this setting it is possible to create 3M forward from 3M and 6M LIBOR spot rates. In a world where credit risk is taken into account, this is not essentially the same thing, because 1M LIBOR and 3M Libor carry different risk.

How exactly is this difference in credit risk incorporated in the tenor spread? I would expect that, from point of view of party A that pays the 1M-LIBOR, this party would want party B to pay the 3M LIBOR + a spread to reflect that party A has more to lose. Party B is actually more sure that party A will be able to pay the monthly payments and party A wants to account for this. Is this the wrong idea?

What essentially drives this tenor basis? Furthermore, what is the relation between the tenor basis of a 1M-3M basis swap with swap maturity 1y vs 2 years vs 3 years? I just lack the feeling of how this tenor spread moves (although I red it is determined by supply and demand).

Answer 1

I don't endorse the belief that this is a credit concern. For banks operating a continual lending business they are less concerned with an isolated case of lending for 1M rather than 3M cause it is less risky. When then money from the 1M loan is repaid the money is relent so the existence of credit risk is always pertinent.

However what you have is an option. If you, for example intended to lend for 3M but in 1M instalments you would have an option every month to pull the funding, if you lent for 3M you would not have this option. From the borrowers point of view if they borrow for three months their funding is locked in and if they want money for 3M but roll it on a 1M basis they are short the option of the lender pulling the funding.

The chapter on term structure of interest rates talks a lot about these issues if you can get hold of a copy of Darbyshire:Pricing and Trading Interest Rate Derivatives. He essentially replicates my argument above and gives some maths and charts. Its pretty neat how he compares it to the different currencies basis markets and derives conclusions. Essentially it boils down to the available of money and the value of the option. When central banks pump money into the system, like in EUR, the value of these options disappears since the likelihood of say, funding and liquidity scarcity from the point of view of the borrower dramatically declines, and hence the basis collapses like it does today.

When you look at things like forward basis, this availability of money and central bank excess or drainage is important. There have been historical blips where TLTRO maturities in EUR are quite important since they correspond to particular money draining events. I can't remember the terminal date of the next one but it might be 18mths or so. Although 2016/17s LIBOR/OIS spike in USD was to do with the US Money Market reforms. Essentially the reforms meant bank funding was likely to be more threatened hence the availability of money became more scarce. These sorts of things can be speculated on a forward basis. When you go longer than 5Y its all about supply and demand and positioning of big dealers.

edit: I've since also contributed this related answer: Using option pricing methods to model real asset liquidity