A blog on financial markets and their regulation
Global banking glut, original sin, shadow banking and the cross currency basis
October 26, 2016Posted by on
This is a wonkish post that links together four concepts that are somewhat slippery even in isolation. So let me begin with a quick primer on each of them:
- Global banking glut refers to the idea that there is an excess lending capacity on the balance sheets primarily of European banks. Not finding enough outlets in their home markets, this money chases assets elsewhere in Europe and then in the United States. (More details can be found in Hyun Song Shin’s article and paper). I would extend this notion to other institutions – for example Japanese insurance companies chasing US assets.
Original sin is the idea that most lenders are willing to lend only in their own currencies and not in the borrower’s currency. Large advanced countries like the US are not subject to this constraint. By holding their foreign exchange reserves in US dollars (invested in US treasury bonds), central banks around the world lend to the US government in the borrower’s currency. But a weakened form of this constraint still exists. Banks will lend in a foreign currency only to the extent to which they themselves can borrow in that currency or can otherwise hedge the exchange rate risk. A European bank will have dollar liabilities roughly equal to its dollar assets net of hedges so that it does not bear any exchange rate risk.
Shadow banking refers to non bank vehicles for maturity transformation and credit intermediation. The vehicles most relevant to this post are money market mutual funds (MMMFs) in the United States which invested in short term instruments exposed to some (though small) degree of credit risk, but whose units were regarded as completely safe, cash equivalent instruments. Because of their ability to issue and redeem units at par, MMMFs could hide fluctuations in the value of their investments from their investors.
In the good old days before the crisis, a bank that could borrow euros at the inter bank euro lending rate (EURIBOR), was able to swap these into dollars to get funding at the dollar inter bank rate (LIBOR). Not any longer. A large cross currency basis has emerged making dollar funding through this route significantly more expensive. The BIS paper by Borio and others has details about this phenomenon. I must add though that while Borio and other economists regard the cross currency basis as a market inefficiency or failure of arbitrage, the post-crisis finance literature, no longer regards the cross currency basis as a market imperfection. Since EURIBOR and LIBOR are no longer seen as risk free, the cross currency basis is just another input to calibrate a multi-curve discounting model (See for example, Masaaki Fujii).
Now I turn to the linkages between these diverse phenomena.
For much of the last decade, the supply of credit from the banking glut in Europe was matched by the demand for dollar credit emanating from US and emerging market companies. Some US companies were levering up to fund stock buybacks; some were funding their investment (or losses) in oil fracking and other businesses. Emerging market companies sought to borrow in dollars because they could not borrow in their home currencies (original sin).
Though the banking glut was in euros and the credit demand was in dollars, the US shadow banking system (particularly, the MMMFs) stepped in to solve the currency mismatch. US MMMFs lent to the European banks in dollars and these banks then lent the funds to dollar borrowers. In this solution, the funding was in some sense coming from the US itself, but the credit risk appetite and the capital required to support this risk came from the European banks. With the implementation this year of the post crisis reforms of the US MMMF industry (abolition of stable value accounting for MMMFs), this route to matching euro banking glut and dollar credit demand is coming to an end.
But there was a second solution to the currency mismatch and that was through the derivative market, especially, the cross currency swap. The European banks had abundant access to euros, and they swapped this into dollars to fund credit in dollars. In the good old days before the crisis, a large European bank borrowed euros at EURIBOR, and swapped these into dollars to get funding at dollar LIBOR. The large and rising cross currency basis has made this solution less attractive.
In the long run, this will probably lead to a repricing of credit risk with dollar credit becoming more expensive and euro credit cheaper. The latter process is being accelerated by the ECB’s corporate bond buying programme. Borrowers accustomed to borrowing in dollars will at some stage have to accept the currency risk of euro denominated borrowing. The large reverse yankee bond issuance (US companies borrowing at zero or near zero rates in euros) is the early stage of this process. So far, however, most reverse yankee issuances have been swapped into dollars. Rising cross currency basis will force at least some of them to leave the borrowing unhedged thereby taking on euro exchange rate risk, and the US corporate sector will for the first time get a taste of what original sin looks like. For many emerging market companies, who almost instinctively borrow in US dollars, this is an opportunity to rethink their liability management strategy. The other rather remote (but frightening) scenario is that an implosion of the European banking system eliminates the banking glut in that continent.