Gold Leasing by Central Banks: Reaching the Limits
Gold lease rates are soaring. What is going on? This commentary gives my best guess, which is in many respects a refinement of certain views expressed in my last essay, War against Gold: Central Banks Fight for Japan.
Were Machiavelli alive today, he might title his seminal work The Central Banker instead of The Prince. Figuring out what these princes of money are doing, not to mention why they are doing it, can be difficult. But it is not reading tea leaves. And it has a long tradition.
The original purpose of central banks was to make the classical gold standard function more smoothly than it did under free banking, to protect bank depositors (i.e., the public), and to prevent or at least ameliorate serious banking panics. Central banks, therefore, focused on three key problem areas: (1) the adequacy of gold reserves in a fractional reserve banking system; (2) the quality of bank assets derived from investing customer deposits, particularly the mismatching of maturities (i.e., borrowing short and lending long); and (3) international settlements, particularly loss of national gold resulting from imbalances in international accounts. But while central banks tried to prevent serious problems from developing, they almost never gave public warnings of imminent crisis. Quite to the contrary, banking panics and devaluations almost always took place in the wake of repeated official pronouncements that whatever the perceived problem, it was under control. For today, the lessons are two: (1) central banks are no strangers to assessing the risks in gold banking; and (2) central bankers are masters at dissembling in the face of potential crisis.
Fast forward to late 1995. My contention that the central banks decided to mobilize their gold as necessary in support of an effort to prevent a complete financial and banking collapse in Japan is really no more than an educated guess -- a deduction made after the fact on the evidence available. It rests on my view that nations, no matter what officials may say, do not part with gold absent very good reason, usually touching issues of national survival or monetary sovereignty. The claim that a government is selling gold merely to adjust the composition or yield of its foreign exchange reserves strikes me as deeply suspect. The Dutch and Belgian sales are far better understood as measures taken in preparation for the Euro. My guess is that Canadian sales were not unrelated to the Quebec issue, but that is a subject for another time. The Bank of England's current sales are quite simply inexplicable on this ground. And in time, if the proposed Swiss sales ever do occur, we will probably learn that they involved some sort of calculation or quid pro quo having to do with the protection of Swiss banking or Switzerland's uniquely independent status within an integrated Europe. In any event, the Swiss, who actually can make a reasonable claim to having excess gold, are unlikely to hold a fire sale like the British.
By 1997, with the amount of gold reserves on lease having risen sharply since 1995, two events suggest the first signs of central bank alarm. First, the London Bullion Market Association (LBMA) disclosed for the first time ever the volume of its gold trading activities and promised to release average daily clearance figures every month in the interest of greater market transparency. Second, the Federal Reserve commissioned and made public an internal study on government gold policies.
The LBMA disclosure was announced under the headline "Gold global market revealed" in The Financial Times on January 30, 1997. At the time the daily volume of gold trading by the 14 market-making members of the LBMA was about 30 million ounces or 930 metric tons. Some traders said that the number was misleadingly low because matched orders were not included. More recently, the LBMA has reported average daily clearing numbers closer to 40 million ounces (1240 tons). For purposes of comparison, 50,000 contracts or 5 million ounces, is a busy day on the COMEX, and 100,000 contracts or 100 tons a very busy day on the TOCOM. In short, the over-the-counter London gold market dwarfs the public gold markets in New York or Tokyo. More importantly, it is where most transactions that involve gold leased from central banks are conducted. Not surprisingly, therefore, the LBMA's disclosure received prominent attention in the BIS's 67th annual report released in June 1997. There, as part of an extended discussion of gold leasing (pp. 95-96; see my earlier essay), the BIS noted that the amount of daily gold turnover in London "rivals that of London trading of sterling against the Deutsche mark." It added: "According to a Bank of England survey, most of the trading was spot -- both physical and book entry -- with a significant forward market and an active option market." Exactly why in January 1997 the LBMA broke with a tradition of secrecy going back hundreds of years has never been fully explained. My opinion: the central bankers were demanding a better picture of what was happening with their leased gold.
The Fed study can be read at www.federalreserve.gov/pubs/ifdp/1997/582/ifdp582.pdf. This study contains the usual disclaimer that it presents the views of its authors, not the Fed or other staff, and could be dissected endlessly for obvious errors and omissions. It purports to analyze the costs and benefits of various courses of action, including immediate sale of all government gold and no sale of government gold. A startling feature of the presentation is the treatment of gold to be mined as part of the available gold stock. The final paragraph suggests a compromise policy (p. 26 text; p. 45 pdf):
Of course, any benefits of government ownership of gold are lost at once under a policy that involves selling all government gold immediately. However, any such benefits are lost much later under a policy that involves leasing out all government gold immediately and selling it gradually after some date in the future. It is clear that if governments lent out all their gold but wanted to keep open the possibility of using it in a crisis, they would have to structure their loan contracts so that they could get their gold back immediately in a crisis.
This study appears more an effort to rationalize a policy after the fact than to develop a new, well-thought out one. It is the sort of memo a bureaucrat might wave in explanation of a failed policy, but it could hardly persuade a smart central banker to adopt the policy in the first place. Why? The authors fail to appreciate what central bankers know too well: leased gold does not stay in the possession of the lessor.
The term "lease" is a misnomer, confusing the basic difference between banking on the one hand and bailments and leases on the other. A bank deposit of currency or gold creates a liability for repayment in currency or gold, but the money actually deposited passes to the bank for use in its business as it sees fit. The depositor necessarily becomes a creditor of the bank. A bailment creates an obligation to return the item bailed and gives no right to use it. A lease creates an obligation to surrender the item leased at the end of the lease term, during which period the lessor has the right to use but not to convert or sell the leased property. A lessor does not become a creditor of the lessee except as he may agree to accept rent in arrears. A gold loan by a central bank is a deposit in a bullion bank, not a lease in the ordinary sense of that word. The gold "leased" is effectively put out for immediate sale into the physical market, where ultimately the gold for repayment will have to be purchased. The basic point: gold leasing is not leasing at all; it is banking. The great irony of gold banking as practiced today is that the central banks are the principal depositors. Like private depositors before the era of central banking, they must protect themselves.
By 1998, with net gold derivatives rising in step with the increased leasing of gold by central banks, concern about the risks involved were rising too. Certainly they were on Fed Chairman Alan Greenspan's mind. On July 28, 1998, testifying before the House Banking Committee looking into the regulation of over-the-counter derivatives, he distinguished financial derivatives from agricultural derivatives, saying that it would be impossible to corner a market in financial futures where the underlying asset (e.g., a paper currency) is of unlimited supply. The same point, he continued, also applied to certain commodity derivatives where the supply was also very large, such as oil. And he further volunteered: "Nor can private counterparties restrict supplies of gold, another commodity whose derivatives are often traded over-the-counter, where central banks stand ready to lease gold in increasing quantities should the price rise." Needless to say, this statement provoked considerable comment in the gold community, much of it having to do with a conspiracy by central banks to control the gold price. The real question, however, is to what extent and at what risk central banks "stand ready to lease gold," whether into rising prices or otherwise.
The amount of gold available for this purpose is large, but far from infinite. In the July issue of its quarterly Gold in the Official Sector, the World Gold Council (www.gold.org) puts total official gold reserves at the end of 1998 at about 33,500 metric tons, amounting to just over 15% of total world foreign exchange reserves. Of this amount, the five largest holders (U.S., Germany, France, Italy and Switzerland) accounted for just over 20,000 tons. The U.S., which holds a little over 8100 tons, claims not to lease gold. In 1999, the numbers for Germany, France and Italy will decline in aggregate by over 500 tons due to transfers by them to the new European Central Bank (ECB). Because accounting for leased gold varies by country, it is virtually impossible to tell from official statistics how much government gold is out on lease. What is known is that some governments withdraw gold from the lease market near year end to improve the risk profile of their balance sheets. What is also clear is that at a daily rate of 1000 tons, the LBMA's annual turnover is around 250,000 tons, or almost 8 times total official reserves.
After excluding from world reserves the U.S. and the international financial institutions (BIS, IMF and ECB), there are some 20,000 tons theoretically available for lease. Reasonable current estimates of the net short position in gold derivatives run from under 5000 to over 10,000 tons, implying that about this same amount is out on lease from the central banks. Why? Because, subject to the caveat discussed in the next paragraph, it is the destiny of such leased gold to be sold at spot into the physical market, creating a short position of equal amount. This position may be hedged many times over, but it remains a net short position until the gold is repurchased and returned to the central bank that deposited it.
An ancient rule of thumb in gold banking is that under ordinary circumstances gold placed at sight (i.e., in demand deposits) should be backed by a reserve of 40% in physical gold. Indeed, as recently as 1946 the legislation governing the Federal Reserve System required a gold cover of 35% against deposit liabilities of Federal Reserve banks and 40% against Federal reserve notes in circulation. Although I have seen nothing to this effect, it is possible that the bullion banks engaged in gold banking with deposits from the central banks are holding some of this leased gold in reserve. If so, the amount of central bank gold out on lease would exceed the net short position in gold derivatives by an amount equal to the total leased gold held in reserve by the bullion banks. Of course, in this event the bullion banks' rate of return would also be less. On the other hand, if the bullion banks are not retaining significant physical reserves (as I rather suspect), the risk profile of the central bank gold out on lease is increased, and at a time when general financial conditions are far from ordinary. Adding to the risk, the central banks by their own leasing have driven gold prices to bargain levels that have stimulated unprecedented physical demand, accelerating the withdrawal of physical gold from the reaches of western bullion bankers into India, other parts of Asia and the Middle East. Finally, now pressing in on the central banks is a real wild card -- Y2K, a subject on which the BIS has issued several far from reassuring reports (www.bis.org/ongoing/index.htm).
By any historical or common sense measure, and particularly under current circumstances, a net short gold derivatives position of anything like 10,000 metric tons is pushing the limits of any reasonable assessment of central bank tolerance for risk. To me at least, the message of today's high gold lease rates is that central banks no longer "stand ready to lease gold in increasing quantities" and are instead focusing on the problem of recovering their gold in preparation for Y2K and whatever other crises may lie directly ahead. Expect them to issue reassuring statements; don't expect them them to part with a lot more gold. No central banker wants to be remembered in history for losing his nation's gold, or for giving it away.
Reg Howe
row@ix.netcom.com
goldensextant.com
18 September 1999
gold-eagle.com |
