ALL DIE HARD HEDGERS BETTER READ THIS ARTICLE CLOSELY!
February 14, 2000. The New Dimension: Running for Cover
REGINALD HOWE
[Note: In preparing this commentary, I have received much helpful advice and assistance from Patrice Poyet (poyet@cstb.fr), qui s'int‚resse aux
march‚s en France et … l'ext‚rieur, and Sunil Madhok (skmoi@emirates.net.ae), whose writings on gold in India have appeared at Gold Eagle.
Merci beaucoup … chacun d'eux. Any mistakes, of course, are mine.]
Last week Barrick made its much anticipated announcement on hedging. According to its press release, Barrick during the last quarter of 1999:
(1) reduced its exposure on call options written to 2.7 million ounces (versus 4 million as reported at its website at the end of the third quarter);
(2) stretched out the delivery schedule on its its spot-deferred contracts, which now cover a total of 13.6 million ounces (versus 14 million as
reported at its website at the end of the third quarter); and (3) engaged in "an important new dimension" by purchasing call options on 6.8 million
ounces. The release further states that the new purchased call options "cover 100% of production from March 1, 2000 through 2001," at strike
prices of $319/oz. in 2000 and $335/oz. in 2001. Thus Barrick's hedging program, according to the release, "has been reduced from 18.8 million
ounces at the end of the third quarter to a net 9.8 million ounces at year-end 1999."
While the numbers do not fully jibe with those at its website for the prior quarter, the net reduction in Barrick's hedge book of some 9 million
ounces consists of 400,000 ounces delivered under forward contracts, a reduction of 1.3 million ounces in written calls, and the purchase of new
calls for 6.8 million ounces. In discussing this information with analysts, Barrick has apparently revealed three further facts of significance: (1)
the new purchased calls are for cash settlement only (CSO); (2) they were sold to Barrick by one or more bullion banks, whose name(s) are
confidential information; and (3) the total cost of the purchased calls was $68 million. An article in the New York Times last Sunday
(www.nytimes.com/library/financial/investing/021300invest-gold.html) states that the premium (average?) for the 2000 calls was $6/oz. and for
the 2001 calls $12/oz.
In fairness to Barrick, the existence of a CSO provision appears ambiguous based on its press releases. The February 7 release states that the
purchased calls give Barrick "the right, but not the obligation, to purchase gold." Its February 8 release, however, made after the conference call
in which cash settlement was discussed, states that "every dollar above [the strike price] will now be added to Barrick's floor price of US$360 per
ounce [in 2000]."
Barrick's reference to the "net" position of its hedging program is somewhat misleading. Indeed, it fooled at least one gold analyst into asserting
that since Barrick was able to close nearly half of its hedged position in the fourth quarter without pushing up the gold price, the gold market is
not as short as Frank Veneroso and others claim, and Barrick is not "trapped" by its hedge book. However, buying calls, and particularly CSO
calls, is not the same thing as closing forward contracts. Had Barrick bought physical gold in the amount of its calls (6.8 million ounces or 212
tonnes), it almost certainly would have caused the gold price to rise substantially. What is more, the very fact that it bought paper gold -- not to
mention paper gold that by its very terms is not convertible into physical gold -- suggests that Barrick was in fact unable to cover its forwards in
the physical markets without driving up the gold price.
Reaction in the gold market to Barrick's announcement appeared negative. Some have noted the irony of the positive market reaction to Placer
Dome's earlier announcement of a mere suspension of further forward sales coupled with prospective buy-backs, and the negative reaction to
Barrick's announcement of a substantial "net" reduction in its hedge book. But there is a quite rational explanation: while I was asking who sold
Barrick the calls, traders were asking what Barrick would do with the calls. Specifically, would Barrick delta hedge against the calls by going
short at or about the strike prices, thus making them serious resistance levels?
Writers (sellers) of naked options (puts and calls) typically limit their risk by delta hedging against their exposure. Delta hedging is described by
mathematical formulas and requires quick, reliable access to a liquid market. To oversimplify, let's assume I write a gold call with a strike price of
$300 when gold is trading spot at $250. As the gold price rises, I will buy gold in increasing increments so that when the spot price reaches the
strike price I am 50% covered. If the price keeps rising, I will continue to buy in decreasing increments until I am fully covered at an average
price equal to the strike price. Of course, in the real world the gold price will fluctuate, but the formulas tell me for each price exactly how much
gold I should have as cover, and I keep adjusting my cover accordingly. But note, the purpose of delta hedging by an option writer is to prevent
loss and keep the premium received as profit. For more on delta hedging and derivatives generally, see www.finpipe.com/derivatives2.htm.
Now let's suppose that I am the purchaser of gold calls for 1 million ounces at $320. My risk is limited to the premiums that I paid for the calls,
and I may choose simply to hold them as a bet (or hedge) on a rising gold price. But if I am an active trader with quick, reliable access to the
physical or futures markets, I can also use my calls to backstop a trading strategy designed to profit from shorting gold in a delta hedge whenever
the spot price is at or near my strike price. That is, if gold is at my strike price, I can sell 500,000 ounces. If my sale, especially when combined
with others doing the same thing, knocks the price down, I try to cover quietly at lower prices, book my profits, and wait for another chance to
do the same thing again. That's a successful short sale. What is more, as long as I am successful, I can keep repeating the process until the
expiry of my options. Thus, the purpose of delta hedging by an option purchaser is to earn a profit.
But suppose I don't succeed, and the gold price does not return below my strike price. There are two possibilities. First, increased volatility may
cause the value (price) of my calls to rise by more than the increase in the underlying gold price, so that profits from the sales of my calls would
more than offset the losses on my short positions. If so, despite my unsuccessful short sales, I can close out my positions at a net profit.
Alternatively, in the absence of any possibility for profit, I can cover my shorts with my calls at no net additional cost to myself since I have delta
hedged. Of course, in either of the two unsuccessful short situations, I would also like to recover the premium paid for my calls if I have not
already done so through successful short sales. But in any event, my losses should never exceed the premium amount that I was prepared to risk
in the first place.
What seems to have rattled traders, then, is the possibility that Barrick has been recruited to the short side with at least 100 tonnes of ammunition
for use in a delta hedge, and that the $319 and $335 levels will be strongly defended by Barrick and others, including its bullion banker sellers.
And my questions regarding who sold Barrick the calls, who might be backstopping them, and what is really going on here become even more
intriguing.
If the calls are CSO, several inferences about them would follow. It is very unlikely that they would be covered calls written for income by a
central bank or other big holder of gold. Rather, they would almost certainly have been written either as a short-side speculation or, and in my
view far more likely, to support short-side speculation by others. In the latter event, they are almost certainly backstopped by someone with very
deep pockets who has an interest in capping the gold price, and who therefore is unlikely to delta hedge them even in the futures (paper) markets
since doing so would partially negate the short sales of others. That can only be someone prepared to accept a $680 million loss for every $100
over the strike price. Whomever it is, Barrick apparently believes that the financial resources ultimately backing the calls are adequate to the task.
Another possibility is that they were written as CSO calls precisely because there is not enough liquidity in the physical markets to delta hedge
them. But if so, it is hard to imagine any reason for writing them other than to facilitate capping the gold price, or at least to keep Barrick from
trying to cover in the physical markets and thereby drive up the price.
Unfortunately Barrick has not provided sufficient details on its purchased calls to calculate with any precision their implied volatility under Black
Scholes. Having such a figure, it would be possible to compare the implied volatility of Barrick's calls with the implied volatility of similar COMEX
or other market traded calls to see whether Barrick paid the level of premium associated with a typical arm's length transaction. But even absent
that evidence, there is much to suggest that its purchased calls may be subject to other special conditions.
Unlike most purchasers of calls, Barrick can by its own actions affect the price of the underlying asset. Whoever wrote the calls appears
vulnerable to an effort by Barrick to cover its forward obligations, thereby also driving up the gold price. Done cleverly, especially in combination
with delta hedging its purchased calls, Barrick might cover some significant chunk of its written calls and forward contracts without driving gold
over $360 -- its claimed 2000 floor price -- at a profit or small loss. Then, more than making up the losses on its remaining forward obligations
with the profits on its purchased calls, it could glide smoothly into a gold bull market, leaving its bullion bankers holding a bag of shorts.
Accordingly, it is reasonable to assume that either: (1) the writers of the calls hold enough of Barrick's written call options and forward contracts
to ensure that Barrick cannot act in this fashion; or (2) the calls contain restrictions on Barrick's ability to cover further, or require that Barrick
maintain some minimum ratio between its purchased calls and its delivery obligations under written calls and forward contracts such that its
incentives continue to rest on the side of at most a slow, controlled increase in the gold price. The description of the calls in Barrick's February 8
press release is consistent with alternative (1) above.
Yet another possibility is that Barrick does not really control the purchased calls, but is only entitled at maturity (whether they are American or
European style options is unknown) to the appropriate cash payments. It may also be subject to further conditions, e.g., refraining from certain
actions that might drive up the gold price. In this event, the bullion bank or banks that sold the calls may be able to use them in their own
discretion to delta hedge from the short side, whether for Barrick's account or their own.
Indeed, it is not impossible to imagine a deal where the bullion banks get the profits from successful short sales, Barrick gets the profits on
exercise when and if the short sales are unsuccessful, and whoever (Exchange Stabilization Fund?) is ultimately backing the calls has a double
obligation: (1) to pay Barrick in full on exercise; and (2) also to pay the intermediary bullion banks on exercise, but subject to partial credit for any
profits they may have made on successful short sales.
In any event, the fact that Barrick bought options on paper gold -- virtual gold -- from someone who is either crazy or possessed of very deep
pockets and a strong desire to cap gold suggests: (1) that the physical gold markets are so tight that Barrick could not cover in physical metal; and
(2) that its so-called "new dimension" -- purchased calls -- is nothing more or less than Barrick running for cover. Market action at its strike
prices -- $319 and $335 -- could be ferocious. With a major breach running toward $360 -- Barrick's 2000 floor price and the level where many
think the gold banking system might implode --, Katy bar the door.
God made gold the king of money. When He made the king of beasts, He painted him gold. The lion's would-be victims know to run for cover,
and when none is at hand, to keep running, and to dodge, feint, bob and weave in hopes of shaking the golden beast. Barrick's instincts are no
less sound than those of the zebra or the gazelle. And watching Barrick run from a gold panic partly of its own making promises to be good sport,
unless perhaps you are a Barrick shareholder. In that case, Barrick's hedging program is exactly what Randall Oliphant, its chief executive, says:
"It is not a theoretical concept, it is about real money." But is real money something that Barrick and other heavily hedged mining companies
know anything about? |
