Why golden futures suddenly turned sour

Background

The gold lending market is the very cornerstone of the Bullion business, being the vehicle, which supports not only the greater part of the global jewellery trade but also the widely discussed hedging programmes of producers. Its existence stems from the basic historical imbalance of activity in the forward market- jewellers are under no pressure to hedge their future requirements, having a price elastic end-consumer in the general public, while producers are generally more price sensitive.

Moreover, the contango which makes forward selling so attractive to producers, makes forward buying generally unattractive. Thus, dealers having few forward buyers on their books must hedge their forward purchases by borrowing gold and selling it in the spot market. Estimates of the size of the gold lending market today oscillate around a figure of 2,500 tons ($32 billion approx.), most of which is traded on a physical loco London basis across the accounts of LBMA members. The Bank of England, in its supportive role in the London market, provides the major conduit through which central banks can access the commercial market confidentially and efficiently.

As with every market, there have to be benefits to both sides. For the lenders, yield for an otherwise dormant asset is the obvious attraction. However, this yield must be balanced against limits of both credit and tenor. These limits are of particular concern to central bank lenders who, after all, are lending reserves which arguably are their country's last line of defence. Thus, their required counterparties are first class names and the preferred term is

comparatively short, certainly within one year. Moreover, for obvious reasons, relatively small percentages of official reserves are committed to the market irrespective of yield. This is truly a market of demand and supply, untroubled by governmental fiscal policy!

Uses For Gold Liquidity

The commercial banks put this gold liquidity to use in a panoply of products. The simplest structure is a gold loan - the bank passes on the gold on a lease basis to clients with a credit spread. Such customers could be jewellery fabricators who, being unwilling to take price exposure, borrow gold forth e period of manufacture before sales. Gold liquidity is also used to support consignment stocks, traditionally lodged with Middle East or Japanese wholesale consumers.

However, the most important activity supported by gold loans is that of dealers buying forward, usually from producers. These clients seek to lock in current prices plus a contango for the future delivery of their as yet unmined gold. Essentially the contango is calculated by deducting the gold deposit rate from the currency (usually US$) deposit rate. This activity hedges the miner against falling prices but leaves him exposed to opportunity loss should prices rise. Therefore, miners rarely lock in more than a carefully calculated percentage of their future production.

Another hedging method utilized by miners involves their taking a gold loan instead of a currency loan to develop their prospect. The gold is sold immediately for cash and the loan is repaid out of future production. The economics are essentially the same as a forward sale, but the bank has a greater credit exposure to the miner and the rate charge will reflect the increased risk.

Options and more complex derivatives are other risk management products that also rely upon gold liquidity. For example, should producers wish to sell calls or buy puts, then the options dealer will generally sell out the underlying hedge for a forward value to his own in-house forward book which will require liquidity cover as a result.

The November Squeeze

The structure of the deposit market has been evolving over the last 20 years as more producers feel the need to hedge in a stagnant market and more long term holders are attracted to the yield for their otherwise non- performing asset. However, there is a limit to the availability of central bank's gold; many of the largest holders have policies not to lend, and if their reserves are deducted from the approximately 35,000 tons of total official holding then the estimated 2,500 tons of loaned gold becomes a significant per centage. Moreover, ongoing sales programmes are slowly reducing the available material.

Late last year, a confluence of factors contributed to the extreme tightness in the gold lending market. Historically there is a "seasonality" about rates over the turn of the year. This is partly explained by the Christmas bulge in the jewellery business and the concern that some central banks take back the gold for reporting purposes over year-end. Additional factors in late I995 were significant hedging programmes from South African miners (on top of ongoing activities from US and Australian producers) and relatively low prices, encouraging tl1e dealers to place out larger consignment stocks with physical consumers. Finally, some bearishness in the Comex market had afford ed dealers the opportunity to buy gold futures (and sell spot) at initially attractive arbitrage levels.

Moreover, the wish by the borrowers to match their maturities to those of the lenders had been taken to an extreme. Over the years some hedgers, not wanting to pay the premiums charged by the bullion banks for longer-dated (particularly more than one year) transactions, have decided to fix the currency interest rate component of a forward deal and leave the gold portion on a floating basis, charged as a lease rate. Thus, the central bank's preferred term would be matched by the borrower's, though of course, this does not give a locked-in term cost structure for the latter, who sacrifices certainty for the economy.

On account of one of the anomalies created by the physical nature of the gold market, overnight gold has no deposit value and all positions must be covered on a tom / next basis, generally the cheapest point of the curve. The apparent logic of running a short position clown to the very shortest time horizon, therefore, seemed attractive to an increasing number of market participants late last year. Such a strategy is of course dangerous, as there is no time for manoeuvre should liquidity dry up. The discomfort of these squeezed shorts was clearly one of the forces which pushed the market to extremes. More recently, many of these clients seem to have repositioned their int e rest further along the curve!!

A large part of gold liquidity demand springs from the fact that gold borrowing is undertaken because it is cheaper than USS funding. Therefore, there is theoretically a natural cap - at least over the medium term - at $ LIBOR. However, this will not prevent short term shocks as seen in November, as decisions to unwind positions cannot be taken swiftly. This is also the problem with central bank liquidity - generally, those lenders are fully lent out to their conservative limits, and given that their alternative yield is zero, maturities tend to be rolled over, making supply somewhat inelastic. Decisions to increase lending activity y must be taken at very high levels and this takes time. With gold deposit rates approaching those for the OM, however, some central bankers who view gold as just another part of the reserve management equation, are doubtless making their voices heard.

Other evidence of the self-balancing nature of the market comes from the demand side. For example, higher gold lease rates have been a factor in recent decisions to repurchase short positions. Indeed, hedging to enhance yield is beginning to look uneconomic as currency rates ease and gold rates rally. A forward premium of only $ IO for a year is beginning to look rather skimpy.

The Producer Imperative

However, some producers do not have the luxury of viewing hedging as a yield enhancement. Miners with rising costs, particularly those in South Africa and Australia, are being forced to hedge, or at least to buy put options out over a number of years duration in order to protect their margins and to facilitate the raising of capital for expansion. This capital must come either by way of rights issues or by banks lending, and such funding requires the comfort of gold price protection. As the forward sales are undertaken in a high yielding local currency, the benefit is much more attractive.

However, these forward sales and the large option deltas will require fixed long term gold liquidity and will contribute to the pressure on this area of the curve.

Logically, rising costs for the mining companies who produce over 35% of the Western world's annual production should create a fundamentally constructive outlook for gold prices. However, the reality of declining grades and, in South Africa, a more stable currency and a workforce with greater expectations, is creating a harsh environment where hedging is becoming a more necessary tool for ensuring the economic viability of increasingly marginal mines.

With these scenarios in place, it is hardly surprising that the gold market generally has entered a more interesting time and the curve of forward deposit rates has assumed a more conventional shape as the demand for liquidity is now more focussed at the long end. The short end of the deposit market has eased particularly in response to speculative buying in the April Comex contract which has allowed dealers to buy spot and sell the futures at attractive yields. Moreover, as we move further into the New Year almost all of the seasonal factors mentioned earlier are turning world.

Gold Investment For Yield

Another potentially bullish factor for price is that gold rates in excess of 2.5% should encourage buyers who are prepared to invest for yield, particularly in times of declining currency rates. There has certainly been some buying from Japan where currency yields have fallen to virtually zero, and with Swiss Franc rates around 1.5 % and OM rates around 3.25%, this argument becomes more compelling for Europeans. Should such investors come to the market in any size, gold deposit rates will ease even further at the short end - one presumes that these participants are not prepared to invest in and therefore to lend gold for the longer term.

Conclusion - The Future

Therefore, the challenge for the Bullion banking community remains that of bridging the gap of tenor between the lenders, who for reasons of flexibility or conservatism are only prepared to place deposits of around three months duration, and the borrowers, in some cases dealers' in­house options books, whose preferred term is probably three years! As the market becomes more mature it is probable that some central banks will be attracted by the significantly improved yield available for longer term commitment, perhaps on a collateralised basis. Moreover, the market has been inventive in offering longer term fixed for floating, cash sett led swaps settled on the benchmarks of LIBOR and GOFO. The use of these by lenders may well increase. Bulli on banks have limit s as to how much of a mismatch of tenor they can run on their books and they will pay well to reduce their risk. Yield opportunity beckons for providers of longer erm liquidity!

If one considers the longer term horizon, projected gold production numbers, particularly from Australia, Central South East Asia and South America, look set to escalate dramatically. If the mining companies concern ed are obliged or wish to hedge for whatever reason, then it is clear that the market will need to unlock fresh areas of liquidity to cover this potential demand. It is unlikely that the demand for long-dated fixed material will ever be fully met, but at the shorter end and on a floating basis it is probable that supply will be teased out eventually by yield.

What is sure is that a return to the days of gold at 3 / 8 per cent for a three month deposit have passed, and that the gold lending market will continue to be volatile at higher levels, almost certainly giving tl1e central banks a better return than they ever expected for their barbarous relic.