A Basis For Cross-border Re-leveraging

Many years ago, when certain members of TMM were fresh-faced graduates learning about curve building and how best to get the desk breakfasts, they were told that "the basis never moves". Of course, things that "never move" or "never happen" tend to follow Murphy's Law... and indeed, in August 2007, basis spreads everywhere started to blow out, resulting eventually in even single stock guys having a Libor-OIS ticker on their screen. TMM, like many, has become accustomed to esoteric portions of the financial system having a large impact upon prices, and in recent days there has been some excitement about the moves in Swedish and Sterling Basis Swaps. At the risk of boring our readership to death, TMM will attempt to an explanation at the weird goings on of the basis swap market... as a warning, this is a bit wonkish.

For the uninitiated, a basis swap is an exchange of two Floating Rate Notes (FRNs) paying different floating rates, and essentially come in two forms: (i) both floating rates are in the same currency, so there is no principal exchange (known as a tenor basis swap - Libor-OIS spreads are an example of these, but also 3m Libor vs. 6m Libor), or (ii) the floating rates are in different currencies (known as a cross-currency basis swap , see chart below- the EUR/USD basis is an example of these). Bond Maths 101 tells us that an FRN that pays the risk free rate is worth its face value because regardless of where risk free rates move, you can borrow at the risk free rate, but the bond pays you that very same rate. Obviously, that is not really a particularly good approximation of the real world, but bear with us for a minute... If there is an exchange of two FRNs in different currencies both paying the risk free rates in those currencies, then the exchange is "fair", however, if one of the Notes pays a rate with credit risk embedded, the exchange is not fair, and a Basis Spread needs to be subtracted from that note's floating coupons in order to equalise the credit risk (or, by market convention, is added to the lower risk note's coupons in the case of tenor basis swaps and to the foreign currency leg for cross-currency basis swaps. Intuitively, 6m Libor has more embedded credit risk (and liquidity risk - more on that later) than 3m Libor, so a swap of 3m Libor vs. 6m Libor would need a spread added to the 3m Libor leg to make the swap fair.

Taking the argument a bit further, one can see the term structure of the tenor basis swaps in each currency relative to risk free rates (for the sake of simplicity, we assume these to be OIS/Fed Funds/SONIA/EONIA etc, though there are plenty of caveats here) effectively gives a profile of the riskiness embedded not just in each Libor rate, but also over term. If the structure is steep (i.e. 12m Libor vs. OIS is quite a bit wider than 1m Libor vs. OIS), then there is either a perceived increased credit risk, or alternatively, liquidity constraints or liquidity-based demand for longer-term funding. Thus, the relative steepness of the two term structures should have at least some determination upon the cross-currency basis - i.e. if the steepening is due to credit concerns then the cross currency basis would move more *negative* (this is the cause behind the well-known "Japan Premium"), but if it is due to term liquidity preference rather than credit constraints, this would move the cross-currency basis *positive* as it would become more attractive to issue debt overseas and swap it back into domestic currency.

So that's the wonkish stuff out of the way. In the real world, these effects are usually dwarfed by longer-dated issuance being swapped between currencies for yield pickup by corporates and supra-nationals in the case of the medium/long-end, while in the short term by the immediate short-term funding needs of the banking system. The poster child for this was the EUR/USD 3m Basis (see chart below)in the immediate aftermath of the Lehman bankruptcy as foreign banks struggled to fund their USD assets via the FX Swap/Basis Swap markets. The EUR/USD basis moving negative has thus evoked memories of USD-funding shortages, most recently in late-April/May as concerns about the solvency of the European banking system in the presence of possibly insolvent sovereigns came to the fore.

So in recent days, when the SEKUSD (1y - white line)and GBPUSD (1y - brown line) basis swaps had large moves positive, it raised fears of a liquidity crisis in those currencies...

But TMM think there is another explanation, related to the textbook gumpf above. For the sake of tractability, TMM have rebased the tenor basis structure relative to OIS (which we take as the risk free rate, this isn't strictly accurate due to compounding, but it shouldn't affect the overall picture) in the UK (chart below, 1yr basis swaps: 12m Libor vs. SONIA - white line, 6m Libor vs. SONIA - brown line, 3m Libor vs. SONIA - yellow line and 1m Libor vs. SONIA - green line)...

...and in the US (see second chart below: 12m Libor vs. FFUND - white line, 6m Libor vs. FFUND -orange line, 3m Libor vs. FFUND - yellow line and 1m Libor vs. FFUND - green line):

It is pretty easy to see that in the US the term structure has not really changed all that much from early-2010, but in the UK it has steepened as 6m and 12m have either stayed wide or moved wider. This is interesting, because the BoE's Special Liquidity Scheme is to roll off shortly, and UK banks have been attempting to replace that funding longer-term, and this may account for the widening of these bases. This looks to TMM like the liquidity-driven term structure steepening argument from above. It is perhaps no coincidence that recently there has been a pickup in cross-currency funded issuance of RMBS in the UK given that GBPUSD basis swaps were still negative despite the moves in the basis term structure. Indeed, as market makers had kept themselves long of Dollar-funding in case of another Dollar funding squeeze, it is also perhaps not surprising that the move has been violent as FX Forward books were forced to stop out of their position. These moves would indeed drive the cross-currency basis more positive.

In addition, TMM find it hard to believe that it will not have gone unnoticed that FOMC LLC is about to print a shed load of USDs, making funding in USD a much easier proposition. With printing presses around the world not having the same productivity rate as that of the Federal Reserve, TMM thinks it makes sense for the international banking system to begin *re-leveraging* on a cross-border basis. The evidence from the UK and Sweden (related to some covered bond shenanigans) suggests that these guys are paying attention doing this. As far as credit growth goes in countries not undergoing the dreaded balance sheet recession (as regular readers know, TMM does not think the UK falls into the same category as the US, and today's GDP numbers certainly back that view up), a re-leveraging in cross-border banking is bullish. On that basis (pun intended), TMM would expect this to spread to other currencies as funding is raised in the US to buy/roll-over foreign assets. An unexpected side effect of QE2 perhaps, to add to the expected one of asset bubbles likely in Emerging Markets....

TMM wonder which esoteric portion of the market will be the next to show signs of getting tipsy on Old Ben's Bourbon.

And finally, TMM couldn't help but chuckle at the Australian Green Party's insertion of a number of root vegetables into the behinds of the merger arb guys...