Confused curve signals | FT Alphaville

Confused curve signals

A while ago we observed that negative gold leasing rates were potentially signalling something awry with the Libor rate.

That judging by gold forwards, the Libor component of the gold lease rate calculation  (Libor-GOFO = Lending rate) was coming in much lower than what might otherwise be expected.

At the time we rationalised that this was likely the result of the funding market having become extremely long-tailed — a fact which had presented those banks with genuine access to this low Libor rate with something of a unique opportunity in the gold market. Since their cash had become more valuable to the market than their gold, they could now make a return from lending cash against gold, as opposed to gold against cash.

Simply speaking, the lending curve had inverted and in so doing the cash-for-gold trade had been born.

If you happened to have access to heaps of cheap cash, that meant you could now make a return by servicing the needs of those with heaps of gold but, unlike you, no cash.

A more simple way of looking at it: the spot market was now suffering from an oversupply of gold, and those who were able to lend cash on the proviso that they absorbed gold against those loans could now generate a return. The market was, in many ways, rewarding participants for taking gold out of the market and encumbering it up in their own vaults.

It’s worth pointing out that not much has changed since then. Not only is the gold lease rate still negative , the BIS is increasingly becoming the gold absorber of last resort: supporting gold bids at levels which might otherwise not be sustainable.

Thus, if the goldbug accusation used to be that gold lending was suppressing gold prices, a cash-for-gold (gold absorption) environment should naturally imply the opposite: that gold prices are, if anything, being supported by those with the capacity to fund against gold as well as the ability to hold the inventory:


Which brings us to the subject of curves in commodity markets, more generally.

If we apply the rhetoric we have learned from the gold market — that commodity markets can sometimes reflect true funding costs more accurately than interest-rate markets — we wonder, could the current “scarcity amid plenty” conundrum in the oil market be indicative of a very similar phenomenon? Namely that for a large portion of the market, with cheap access to cash, the only funding that is now profitable is the type that ensures commodity collateral is either encumbered or taken out of the market completely?

When backwardation becomes contango

At this stage we’re going to get very hypothetical. But please do stay with us. (We recommend reading our negative carry posts which explain why it is that negative interest rates tend to incentivise the destruction of production capacity and/or the cornering of markets via hoarding.)

Let’s presume the same gold market dynamics apply: That for a few market players the cost of funding is epically low — possibly even negative — while for others the cost is far higher than anything indicated by Libor. If you’re not one of the lucky few, of course, the only way to bring down funding costs is now to secure your loan with a collateral pledge. For physical commodity producers with short-term funding needs that means giving up commodities in return for cash at a negative rate.

In this way you see how a securities/commodity lending business turns into a secured funding market.

So, rather than paying to borrow a commodity so as to sell it into the market  —  presenting the opportunity  to profit from a potential price decline — you, the funder, are being rewarded for cornering the market instead. Effectively being incentivised to take the commodity out of circulation until prices are supported enough to help you realise a profit on your holdings.

So who could benefit from such a pricing environment?

The first notable observation is that the arbitrage is only open to those with access to cheap cash. The second, of course, is that access to cheap storage must be guaranteed — we are talking about warehouses, tankers and/or the ability to influence production shut-ins.

In metals this is easily done. You operate just like the BIS, dishing out cash in exchange for metal collateral that is inventoried (ideally) in off-balance sheet stockpiles.The cheaper the warehousing, the better your margin. The less visible the inventory to the public market (dark), the greater the market impact.

Oil, of course, is a bit more problematic. Not only is storage more expensive, inventories are much harder to disguise from public view.

As a result there is only so much inventory that can be encumbered via funding agreements before curves begin to adjust against you. This is the consequence of keeping stocks in plain sight, even if they are not available for sale.

The most effective way to benefit from a negative carry is thus possibly to offer funds outright to producers at what appear to be cheap (still positive rates) to them — i.e. they still pay you for the benefit of the loan– on the proviso they reduce production and allocate said capacity shut-in as collateral for the loans they’ve taken.

Such shut-ins would have been necessary to bring balance to an oversupplied market whatever the case. But by encumbering capacity in loan agreements, a cash crunch is avoided for the industry for as long as there is capacity to be offered up in exchange for loans. Banks with negative funding costs in this way end up being paid for encouraging producers to reduce supply.

Investors mis-pricing the curve

As long as investors mis-price the forward curve on the incorrect presumption that positive carry still applies, accumulated collateral (whether  in outright commodity form or hypothetical capacity rights) can also be funded at a hugely profitable positive return.

The market remains inefficient because risk is being incorrectly evaluated on both sides of the spectrum. A huge arbitrage is open for the taking as a result. Possibly the biggest commodity and carry arbitrage of all time.

You the bank are effectively being paid to lend against collateral, at the same time that investors are paying you via the forward curve a premium to store commodity collateral on their behalf. (This is because they incorrectly assume that the risk lies on the upside rather than on the downside.)

For you the bank it’s nothing short of a win win.

You have to remember that the cost of money is conventionally reflected in forward curves via a slight contango. That’s because derivatives allow you to gain exposure to commodities without the same degree of funding exposure (at least they used to before margin and collateral costs began to rise). Since you the buyer of the futures contract receive interest on top of the commodity exposure you’ve agreed to, all things being balanced, there is an opportunity loss in terms of forgone interest for anyone who opts to buy the commodity today to hold until the relevant future expires.

In other words you won’t break-even unless you’ve sold the future to the investor for more than your cost of funding and your storage costs.

Of course, if there is more risk associated with holding the commodity than not holding the commodity — because an oversupply situation exists — you the future seller may demand a premium on top of your break-even costs to be incentivised to hold that risk. This is how you get into the realms of super-contango.

The future buyer is happy to pay a premium because he believes in the case for the upside despite the oversupply (see the chart on the left below):

Of course, if the investor’s expectations start to balance out, in a positive carry world, the arbitrage closes usually with a downside spot price correction. The whole curve shifts lower. Meanwhile, because you the physical player are no longer incentivised to store — since no-one is prepared to pay a premium to do so and you’re just losing bank interest– you become incentivised to sell your stock as soon as you get it.

In fact, in a backwardation you begin to be compensated by the futures market for running the risk of having too little stock, and thus being exposed to a price hike.

A negative carry backwardation

This for the longest time is how forward curves have worked in conventional positive carry markets. Backwardation indicates market tightness because the physical market (more aware of the real supply and demand picture) is demanding an ever greater discount on tomorrow’s crude from financial market particpants to compensate for the risk of not holding enough supply today. Financial participants provide this insurance because they have the opposite view. Contango, meanwhile, indicates the physical market is worried about over supply and is demanding an ever greater premium for delaying sales into the future, to cover the risk of holding all that oil.

Spot prices, meanwhile, are determined by the degree to which investors’ future expectations balance out the spot markets concerns.

If prices fall, both sides of the equation are on a net basis bearish. If prices go up, the both sides are bullish.

A negative carry universe, however, disrupts this balance (see the second chart above).

First and foremost, it continues to reward the physical market for stockpiling even in a backwardated market. Indeed, to incentivise stock selling, backwardation roll gains (selling spot, buying future) would have to more than compensate for the negative charges associated with holding cash.

In other words you would need a super-backwardation to incentivise the usual behavior associated with the structure. Lacking a super-backwardation, the market would simply keep paying you to prop up prices by holding supply back in stock.

As for contango in a such a universe, it shouldn’t really happen since it would imply a monumental mispricing of forward risk by investors to the immensely profitable benefit of anyone holding stock. Indeed, the greater the negative carry, the more fleeting the contango — unless, of course, investors haven’t caught on to the new reality.

One last point. While we may not be in a land of negative official rates, there is a big difference between official rates and real rates. As FT Alphaville has always noted central banks are not only accommodating markets with huge amounts of liquidity, they are also propping up rates in positive territory by means of interest on reserves and on deposits.

(Though obviously the ECB has recently changed its policy, at what we think is great risk.)

Real world rates (those charged by shadow banks and the real world)– just like in the Gofo case — have possibly been negative for a long while. Indeed if you look at the Taylor rule, it’s been implying a negative rate for some time.

Related links:
The negative carry universe – FT Alphaville
Pariah profits in an age of ‘negative carry’ – FT Alphaville
The Collateralized Lending Regime: An Under-reported Shift in Capital Structure – FT
On the perils of plunging repo rates
– FT Alphaville