There have been some intriguing developments in the precious metal markets since August 2007, not least:
- 346 tonnes of BIS gold swaps.
- The fact that rather than traditional swaps, these BIS deals were actually collateralised repo loans, using gold.
- The 250bn SDR special allocation in September 2009.
- 403 tonnes of IMF gold sales.
We propose that the first place to look to try and figure out the driver for all of this activity might actually be the US Treasury repo market itself. Why? At the very least, to determine if there is a chance these transactions could be related to gold and silver having become the optimum benchmark in collateralised lending.
We come to this notion because an article in Risk Magazine’s July edition recently referred to a very interesting development in funding post-Lehman. Namely, the emergence of a two-tiered system (yes, again) in which those who have the means to pledge collateral to secure interest-rate swaps receive rates below Libor, while those who can’t receive rates (which can vary greatly according to the institution) much above Libor.
This is because, for collateralized deals, the pricing convention is increasingly turning to the overnight-indexed rate, as opposed to Libor.
The reason for this convention shift being, since collateralized swaps are marked to market and with variation margin posted daily, they are in a sense almost risk free. Of course, that’s only providing the collateral being pledged is good enough.
Risk explained it thus:
Trades between banks are subject to credit support annex (CSA) agreements and so are backed by collateral. It is now generally accepted the future cashflows on these trades should be discounted using an OIS curve – creating a mismatch in dealer books between collateralised and uncollateralised transactions.
Clients might prefer to use high-quality banks when receiving fixed and poorer credit banks with higher costs of funding when paying fixed. “It may be to the advantage of clients to be able to say ‘for that kind of swap, I am going to use Bank X or Bank Y’. For dealers, it is not so good because there is a discrepancy in pricing methodology,” says Banks of BNP Paribas.
Meanwhile, as ISDA’s margin service report in March noted:
Collateralization rates are almost uniformly higher among the largest 15 OTC derivatives dealers than for the rest of the sample. Large dealers report that 78 percent of their overall trade volume is subject to collateral agreements, compared to 68 percent of Medium and Small firms, with percentages ranging between 97 percent of their credit derivatives trades on the high end and 62 percent of energy and other commodity derivatives on the low end.
The rise of collateralisation was also highlighted, less directly, in the following report in the Journal of Economic Finances:
Findings – Recent evidence indicates that the “safe harbor” provisions, in fact, destabilized the financial system by encouraging collateralized interbank lending, discouraging careful analysis of the credit risk of counterparties and increasing the risk that creditors will run on a financial firm.
So while the interbank has begun to function again, the lending has come at the cost of a preference for ‘interbank collateralised’ deals. Hence, conditions for those who can provide sound collateral are much better than for those who can’t.
The ECB looked into the development of this so-called collateralised lending regime back in November. Here’s a chart from the report reflecting the breakout between the two rates in the US:
The ECB also went on to suggest in its research paper that interbank lending troubles may have unwittingly spilled over into the secured market, with volatilities in the repo market running high due to general scarcity of good quality collateral.
As they concluded:
Despite the presence of collateral, the disruptions in the unsecured interbank market during the 2007-2009 financial crisis have also affected secured markets. This paper presents a model of secured and unsecured interbank lending in the presence of credit risk. Credit risk premia in the unsecured market will affect the price of riskless bonds when they are used to manage banks’ liquidity shocks.
So how does this relate to gold?
Well, for one — you’ll see from the following charts that the gold lease rate (Libor minus gold forward) has traditionally echoed the secured/unsecured spread, because of the fact that gold is simply the ultimate collateral.
Here, for example, is the secured/unsecured US rate differential first (essentially an update of the chart above) compared with the 3 month-gold lease rate – click to enlarge:
And here is the same comparison in 1-month contracts:
So in that sense, the gold forward (GOFO) rate itself can be treated as the interest an institution pays for a loan denominated in dollars using gold as collateral.
GOFO thus becomes a good proxy for the overall price of collateralised borrowing, especially if you add to it the rate the cost of gold storage borne by the lender of funds.
Of course, since it is a collateralised loan, it also makes sense the rate should be below Libor — to reflect the risk-free collateralised nature of the loan.
So what does it mean if the gold forward rate is above the Libor rate? As far as we can see, simply the fact that some banks are willing to pay over the Libor rate to access funds collateralised with gold.
That’s either because Libor is not indicative of the real cost of borrowing, or because the number of institutions funding themselves via collateralised interbank lending is now much greater than those funding themselves with unsecured interbank lending.
And since, by definition, borrowing collateralised by gold should not be more expensive than the cost of storage over US Treasury collateralised borrowing — especially now that treasuries, like gold, are pretty much zero yielding– there’s an effective arbitrage for anyone who can store gold cheaply.
Given the above, it makes perfect commercial sense for the BIS to want to borrow funds at the higher gold secured rate (GOFO) rather than invest them unsecured at Libor.
But there is one other fact to bear in mind. If GOFO is essentially a proxy for collateralised interbank borrowing , what does it mean when the traditional relationship between the secured/unsecured rate and the gold lease rate begins to break down above and beyond the cost of gold storage?
We don’t know, but here’s a closer look at the more recent past in the two spreads:
And the 3-month gold lease rate:
Does the trend suggest that the rate for collateralised repo loans was threatening to go negative too?
The story of the gold curve, so far – FT Alphaville
The gold contango trade, charts du jour – FT Alphaville
Moving down the corridor… – FT Alphaville
The Collateralized Lending Regime: An Under-reported Shift in Capital Structure – FT Long Room