January 25, 2000. Gold Shenanigans: Suspicion Shifts to the Treasury

Fed Chairman Alan Greenspan denies that the Federal Reserve sells, leases or trades in gold or gold derivatives. In a letter to Senator Lieberman (Dem., Conn.) responding to questions from GATA, the Fed chairman goes even further: "Most importantly, the Federal Reserve is in complete agreement with the proposition that any such transactions on our part, aimed at manipulating the price of gold or otherwise interfering in the free trade of gold, would be wholly inappropriate."

Apart from the Federal Reserve, the only other arm of the U.S. Government with broad statutory authority "to deal in gold, foreign exchange, and other instruments of credit and securities" is the Exchange Stabilization Fund. 31 U.S.C. s. 5302(b). "Subject to the approval of the President, the fund is under the exclusive control of the Secretary [of the Treasury], and may not be used in a way that direct control and custody pass from the President and the Secretary." 31 U.S.C. s. 5302(a)(2). The Exchange Stabilization Fund is also responsible for administering U.S. holdings of SDRs. 22 U.S.C. ss. 286o, 286p. The Secretary is required to provide detailed monthly financial reports of its activities to the House and Senate Banking Committees. 31 U.S.C. s. 5302(c)(1).

My suggestion that the Fed might be selling gold call options represented an effort to assign some credible motive to the surprise announcement of British gold sales on May 7, 1999. At the time gold was threatening to move above $300/oz. due in part to increasing doubts that the proposed IMF gold sales would be approved. Short positions in gold were thus in considerable peril. The manner of the British sales -- periodic public auctions versus unannounced sales through the BIS -- belied any effort to get top dollar and smacked of intentional downward manipulation of the gold price. All indications are that these sales were ordered by the British government over the objection of Bank of England officials. British Treasury officials provided some spurious reasons for the sales but no persuasive ones, leaving only one logical conclusion: the gold sales were directly ordered by the Prime Minister for unknown political or other reasons. What is more, his reasons are unlikely to have been frivolous. As leading supporters of the proposed IMF gold sales, the British clumsily put themselves in the position of front-running them, and ultimately the British sales were an important catalyst in forcing the IMF to change tack.

These developments led me to hypothesize a scenario in which U. S. officials called on Prime Minister Blair for assistance in containing the gold price while they unwound short positions put in place to cap it. Chairman Greenspan makes no mention in his letter to Senator Lieberman of any activities by the U.S. Treasury or the Exchange Stabilization Fund designed to influence the gold market. In responses to questions propounded earlier by GATA, a representative of the Treasury dodged questions relating specifically to writing or otherwise dealing in gold call options. Surely former Secretary of the Treasury Rubin was aware of these possibilities since his former firm, Goldman Sachs, is Ashanti's principal gold banker and a major purveyor of gold derivatives. Maybe the Treasury will be more forthcoming in its responses to Senator Lieberman.

In the meantime, results of the BOE's fourth auction held this morning are available at The auction was 4.3 times oversubscribed and the allotment price was US$ 289.50/oz., i.e., all bids over this price were filled and bids at this price were filled on a proportional basis. The allotment price was almost $2 above yesterday's close in New York and $1 over the London spot price ahead of the auction. The results should have been bullish for gold, and it did trade slightly higher in London immediately afterwards. But as soon as the COMEX opened in New York, gold fell back sharply to the $287/oz. level with Goldman Sachs and Chase reported as heavy sellers.

Could it be that the Treasury did not want gold rising with stocks under pressure, not to mention Al Gore facing a big test in New Hampshire? (If I were Senator McCain, I'd take a careful look at the monthly reports filed with the Senate Banking Committee by the Exchange Stabilization Fund.) More importantly, even with reported lease rates quite low, physical gold in size now appears to command a slight premium over paper -- not a good sign for the shorts.

January 24, 2000. Gnomes and Kids: They Say the Strangest Things

President Kennedy left his German audience a bit mystified when he announced: "Ich bin ein Berliner!" Did he really mean to call himself a doughnut, a rough translation of "berliner" -- a German breakfast pastry? So maybe what appears a singularly bizarre statement by Swiss National Bank Vice President Jean-Pierre Roth last week is just a translation problem. Indeed, I do not know whether he was actually speaking in English, German or one of Switzerland's other three official languages, but he is reported to have said that the SNB has only "a small window of opportunity" for its planned gold sales of 1300 tonnes and "couldn't afford to wait too long."

The central banks who are signatories to the Washington Agreement or have otherwise indicated their intention to abide thereby are limited to selling a total of 2000 tonnes of gold over the five years beginning October 1999. The full amount of the planned Swiss sales are expressly allowed. Absent demand for a popular referendum, the legal requirements to permit these sales should be in place by the end of March. The SNB will then have about four and one-half years to sell 1300 tonnes of gold into a market where annual demand now exceeds new mine production by more than this amount. What is more, by reason of the Washington Agreement, the SNB is virtually assured -- particularly during the later years of the five-year period -- that it will have little competition on the sell side from the world's largest official holders of gold.

What, then, is M. Roth talking about? Why would the SNB want to rush into its gold sales program before the Bank of England, not to mention the Dutch, finish theirs? Are the gnomes of Zurich under the same mysterious compulsion as the suits on Threadneedle Street? Indeed, have the world's central bankers gone bonkers on gold? On the one hand, through the Washington Agreement they come to gold's support; on the other, they compete with each other to see who can sell gold sooner at a lower price.

To attribute this strange behavior to a peculiar form of schizophrenia affecting central bankers around the globe is probably a mistake. Historically, coordinated central bank gold sales are a good leading indicator of an impending monetary storm. They are the proverbial finger in the dike, holding back the flood to give more time for preparation. It's an old art, selling as little gold as possible to buy time, and saving as much as possible for the monetary tempest to come.

What could be an interesting twist this time would be Swiss sales made through the BIS at negotiated prices. It's very hard to believe that the SNB would sell gold in as stupid a manner as the BOE. But negotiated sales, particularly to other official holders, at ever rising prices (possibly in euros) might be a way to achieve higher gold prices at a measured pace while leaving central banks in apparent control. It could also permit a more controlled downsizing of the paper gold and gold derivatives markets. Indeed, absent some plan such as this, there is a good chance of a sudden large spike in gold creating chaos in these markets. And for central bankers, chaos in the gold market is the most virulent plague of all.

January 22, 2000. IMF's Off-Market Gold Sales: Toward the New Order?

Speaking this week in Gabon, Michel Camdessus, the retiring managing director of the IMF, revealed that the Fund has "almost tripled the resources of gold" that it is prepared to use in support of its plan to lower the debt burden of poor countries (

Originally the IMF, with the support of Britain and the U.S. but not that of several European countries, particularly Germany, proposed to sell some of its gold reserves to finance its so-called Heavily Indebted Poor Countries (HIPC) and Enhanced Structural Adjustment Facility (ESAF) initiatives. The income from investment of the proceeds of these sales was to be applied to fund these initiatives. However, given the relatively small amount of income likely to be generated, there were suspicions in many quarters that the scheme had an important ulterior purpose: namely, to help balance the growing gap between annual gold demand -- in excess of 4000 metric tonnes -- and new mine production -- around 2500 tonnes. Gold bugs, of course, smelled manipulation aimed at capping the gold price.

The weak gold prices that followed in the wake of the Bank of England's May 1999 announcement of its own gold sales program galvanized additional strong opposition to the IMF's original plan, which also required approval of the U.S. Congress. As a result, the IMF advanced and secured approval of an alternate plan to fund its HIPC-ESAF initiative. This plan replaced the proposed gold sales with certain off-market transactions in gold (

These transactions, which involve settlement of certain obligations coming due to the IMF by member countries, operate as follows. First, the IMF sells gold to the member at the current market price, placing the profits from the sale in a special account to be invested for the benefit of the HIPC-ESAF initiative. The profits are the difference between the IMF's book carrying value of the gold (i.e., one ounce = 35 SDRs, or approximately US$ 48) and its market value. Second, the member purchasing the gold then satisfies its obligation to the IMF by returning the gold at the same market price. Accordingly, the amount of gold involved in any single transaction will be that amount which at the current market price equals the member's obligation. The IMF explains further:

The net effect of these transactions will be to leave the IMF's holdings of physical gold unchanged. No gold will be released to the market, and thus there will be no impact on the supply and demand balance in the market. The IMF's gold holdings accepted in settlement of members' obligations ... will be recorded at a higher value in the IMF's balance sheet, and acceptance of this gold (instead of currencies or SDRs) in such settlements will reduce the IMF's liquidity, by the amount of profits transferred for the benefit of the HIPC and ESAF initiatives ... , and its net income.

According to the above link, the original plan called for transactions involving in total 14 million ounces (about 435 tonnes) of gold. Taking M. Camdessus at his word, it appears that now more than 40 million ounces are committed to the plan. The IMF's total gold stock exceeds 103 million ounces (3200 tonnes), making it one of the largest official gold holders in the world. More information on its gold holdings and policies with respect thereto can be found at Subject to approval by an 85% majority vote, the IMF can sell gold at market prices and accept gold in settlement of members' obligations at agreed prices based on market prices. It cannot buy gold or engage in any other transactions in gold (e.g., loans, leases, swaps, etc.), nor can it use its gold as collateral for loans.

The IMF's second off-market gold sale, involving just over 655,000 ounces, was completed with Mexico on December 17, 1999 ( The press release states in relevant part:

The IMF retained about SDR 23 million on its own account as required by the Articles of Agreement. The remainder of the proceeds -- SDR 111 million (about US$ 152 million) -- was invested with the Bank for International Settlements to generate income for the Heavily Indebted Poor Countries (HIPC) initiative.

The term SDR refers to the IMF Special Drawing Right, which is essentially an accounting convention of the Fund. SDRs are described at The SDR is a weighted composite of the currencies of the U.S., Britain, France, Germany and Japan, which today means the dollar, euro, pound and yen. Its value is determined daily by the Fund on the basis of market rates for its components. The SDR interest rate is adjusted weekly on the basis of certain domestic short term rates in the five specified countries. SDRs can be used in transactions among the Fund, its members and 15 other "prescribed" institutional holders, presumably including the BIS. However, while the IMF can create international liquidity by allocating SDRs to its members, it cannot allocate SDRs to itself or prescribed holders.

In essence, the IMF's off-market gold sales are a mechanism for creating SDRs for the Fund itself by revaluing a certain portion of its gold from SDR 35/oz. to current market prices expressed in SDRs. These new SDRs are apparently being placed with the BIS, a prescribed holder, because IMF rules prohibit its Special Disbursement Account, which includes the HIPC-ESAF initiative, and other accounts administered by it from holding SDRs directly.

Although I have read speculation that some IMF gold may be reaching the market through its off-market gold sales, I am unable to come up with any plausible scenario as to how this might happen. On the other hand, the apparent success of the IMF's program may have favorable longer range implications for gold.

Certainly until all its gold is revalued from SDR 35/oz. to market prices, the argument for outright gold sales by the IMF will be difficult to make. Indeed, if ever there were a situation of eating one's cake and having it too, these current off-market gold sales would seem to be it. Of course, once all the Fund's gold is revalued to market prices, only further increases in the gold price itself will support creation of additional SDRs in this manner for the Fund's own use.

A good argument can be made that the EMU's ultimate vision is not a euro fixed to gold, but a euro regularly measured against gold at market prices. Under a system of this sort, the euro could serve as the domestic currency of the EMU countries without the rigidities of the classical gold standard. It could also serve as a major currency, possibly the dominant currency, for settlement of private international transactions and for international finance. But gold would resume its role as the international standard of value -- the yardstick for all currencies. It would also become the principal component of most nation's international monetary reserves and, at market prices, the standard medium for settlement among national governments (i.e., central banks) and international monetary authorities. The days of one nation's paper currency serving as the world's principal international monetary reserve would be over.

The beginnings of a system of this sort can already be glimpsed in both the EMU's practice of marking its gold reserves to market quarterly and the Washington Agreement affirming that gold will remain an important part of the Euro Area's international monetary reserves. Although starting with a proposal for gold sales much like the BOE's, the IMF ended with a system of off-market gold sales involving settlement of international obligations in gold at current market prices. In hindsight, this IMF program could well be perceived as another step, however small, toward restoring gold to a major role in the international payments system.

In any event, by dropping the fiction of gold at SDR 35/oz. and embracing market prices, the IMF has aligned itself with the EMU as an official sector institution in which the market price for gold has a significant and current balance sheet impact. In the long run, institutions operating in this manner are likely to find that it is more in their interest to have an honest, free market price for physical gold than an unrealistic price subject to manipulation in largely paper markets such as the COMEX, TOCOM and LBMA. Increasingly these large official holders of gold should come to realize that rising gold prices are not always a harbinger of rising prices or proof of bad monetary management. Instead, they can be at times an indicator of rising general productivity relative to gold, and in that context an opportunity for a fundamentally benign upward revaluation of existing gold stocks in accordance with market dictates and to the benefit of gold holders.

Put another way, if the IMF can do a lot for poor countries with a gold price near $300/oz., it could do twice as much with a gold price near $600/oz. That is the price that even today many knowledgeable gold analysts suggest may be necessary to bring physical gold demand and new mine production back into equilibrium. Why, then, should the IMF fight it? Indeed, why should any large official holder of gold, except of course the U.S. which now enjoys "the exorbitant privilege" of printing its way out of its international deficits?

January 14, 2000. Crimson and Gold: Ivy League Suit Test

It used to be said that a good man's suit could be purchased for one ounce of gold. Today a Campagna suit runs anywhere from $3400 to $12,000, with a ready-to-wear line soon to be introduced starting at a mere $3000. "The Man with the Golden Needle," Forbes, Jan. 24, 2000, p. 171. The Andover Shop in Harvard Square, long a vendor of suits to budding young executives headed for their first job interviews, today offers suits on-sale at prices ranging from $500 to $1000.

If Harvard undergrads are finding even their relatively cheaper suits historically expensive in terms of gold, what about the cost of their education?

When my father entered Harvard in 1933, the full cost of attending for one year (i.e., tuition, room, board, books, etc.) was about $1000, or almost 50 ounces of gold (1000/20.67). Four years later, the dollar cost remained about the same but the cost in gold had been reduced to less than 30 ounces (1000/35). When I was at Harvard in the 1960's, the total cost of one year was around $3000, or 86 ounces of gold (3000/35). Later events suggest that even by that date the fixed $35/oz. gold price had become fundamentally untenable, and a gold price of about $60/oz. would have been more realistic. It also would have brought the cost in gold back to 50 ounces, the same as when my father entered before the demise of the gold standard. Today I'm told that the full cost of one year at Harvard is about $30,000, or about 110 ounces of gold at $275/oz. To bring the gold cost down to 50 ounces would require a gold price of $600/oz. (with the dollar price kept constant at $30,000).

On the other hand, adjusting $1000 from 1933 and $3000 from 1965 using the CPI gives around $13,000 and $16,000, respectively, in current dollars. See Similarly adjusting the $20.67/oz. gold price from 1913, the last year before the distortions caused by World War I and subsequent events upset an equilibrium price that had prevailed for more than a century, gives about $350/oz. in current dollars. Using this price, one year at Harvard costing 50 ounces of gold would equate to $17,500. Measured on a cost-of-living basis, then, the total cost of one year at Harvard today ought to be not much over half of what it is.

Even if Harvard is the gold standard of education (a proposition that not everyone admits), its prices are fairly representative of many good schools and colleges. The full cost of one year at any of them involves an amalgam of prices for a very broad array of goods and services. But like a suit, the end product has remained pretty much unchanged in function and purpose through the years. If the price of a suit or a college education can appreciate at almost twice the rate of the CPI, there is no obvious reason why the price of gold should not rise at a similar rate. Interestingly, a price of $500 to $600 per ounce is what some analysts claim would now be necessary to bring gold demand and annual new mine production back into equilibrium. It would also bring the cost of a suit (albeit a cheap one) and a year at college back into more normal ratios with gold.

January 7, 2000. Good Rookie Year: What's Next?

After a 14% decline against the dollar in its first year, some commentators are calling the euro a flop. See S. Hanke, "Euroflop," Forbes, Dec. 27, 1999, p. 98 ( On the other hand, Wim Duisenberg, the former Dutch central banker who now heads the ECB, thinks that the euro's first year went reasonably well, and expects that this year the euro will gain ground against the dollar. See A recent IMF study supports the view that the euro remains significantly undervalued against both the pound and the dollar. See;

What is clear, however, is that in its first year the euro had great success as a vehicle for international finance. Figures from Capital Data Ltd. reported this week in The Wall Street Journal ("Euro Secures Big Role in International Finance," Jan. 4, 2000, p. A17) are quite instructive. In 1999 euro-denominated bonds were issued in the amount of $602.2 billion on international markets, compared to $572.5 billion of bonds denominated in U.S.dollars and $174.2 billion in other currencies, including sterling, Swiss francs and yen. In percentage terms, the euro took 45% of the market, the dollar 42% and other currencies 13%. In 1998, 48% of all bonds sold on international markets were denominated in dollars, and only 22% were denominated in European currency units or the currencies of the 11 EMU countries.

Neither the IMF currency valuations nor the 1999 bond issuance figures can be good news to the British or the Japanese. In Britain's case, not only is the pound being increasingly marginalized as a major currency, but also the difficulties of joining the EMU are mounting. For Japan, the bond statistics underline the yen's inability to gain acceptance as a major reserve currency on a par with the dollar or the euro.

More generally, of the world's three major central banks, only the ECB appears comfortable with both the gold and the currency markets. Japan is intervening in the currency markets to weaken the yen against the dollar. The U.S. is increasingly believed to be behind ever more obvious manipulation of the gold market. Since Japan eschews gold as a major component of its foreign exchange reserves and the U.S. continues to value its official gold at a derisory $42.22/oz., neither country has much interest in an honest price, particularly if it would embarrass their currencies.

However, the ECB now marks its gold reserves to market quarterly. What is more, it has reaffirmed through the Washington Agreement that gold will remain a principal component of its reserves and that it has at least some interest in maintaining the integrity of the gold market. An important question for the ECB in 2000 is whether to continue to tolerate politically motivated Anglo-American interference in this key market.

At the end of the day, the difference between the euro-bears and euro-bulls may not be as great as first appears. Steve Hanke concludes his Forbes article: "The only thing that can stop the euro's decline versus the greenback is if it gets lucky and Wall Street takes a dive." But in the latter event, will it be luck? Or will it be the inevitable end to an investment mania fueled by imprudent monetary and credit expansion? Just because Alan Greenspan is unable to recognize an investment bubble before it pops does not mean that Wim Duisenberg and the other European central bankers cannot. Luck often smiles on those who plan for it.

December 27, 1999. Interest Rates: The Golden Connection

The absence of an international monetary order rooted in gold makes the century now ending unique. Professor Robert H. Mundell emphasized this point in accepting the 1999 Nobel Prize in Economics a couple of weeks ago. See R. L. Bartley, "Money: The Century's Agony," The Wall Street Journal, Dec. 10, 1999, p. A18. Cf. A. Swoboda, "Robert Mundell and the Theoretical Foundation for the European Monetary Union," IMF Views and Commentaries for 1999,

Gold's propensity to retain over long periods of time a reasonably constant purchasing power is widely recognized. Less widely appreciated but just as significant is the long term stability of gold interest rates. Both together are the defining attributes of gold money, features which governments have heretofore proven incapable of replicating with their fiat money substitutes.

Relatively low and stable interest rates under the gold standard were the product of measuring economic value by a shared and real international yardstick. Money -- dollars, pounds, francs, etc. -- was a certain weight of gold, not an artifice of bankers or governments. A lawful dollar had a real cost of manufacture, related to the cost of producing gold. Seigniorage was close to zero, not virtually 100%. Money was not simply a means to facilitate exchanges; it was both a store and standard of value.

Because international balances were settled in gold, small countries could trade on relatively equal terms with larger ones. Trade deficits could be offset by capital flows, but no country was required to hold large amounts of another's paper in its reserves. Any country, small as well as large, could achieve monetary sovereignty and a sound currency simply by following the prudential rules imposed by gold. Quality of monetary policy and banking practices mattered more than economic size, permitting Switzerland, one of Europe's smaller countries, to become a banking and financial powerhouse.

Of course, the gold standard was not perfect, and some of today's monetary problems were also issues a century ago. For example, excessive credit inflation was always a potential problem under the gold standard, and many were the panics resulting from over-exuberance in this regard. So too, in the area of productivity, whether the gold supply could grow sufficiently to provide adequate increases in the monetary base remained a constant concern, particularly for expanding industrial economies.

As it turned out, gold discoveries in California, Alaska, and later South Africa were adequate to the task, enabling most major countries to maintain substantially unchanged gold parities from the early eighteenth century to the outbreak of World War I. Indeed, the gold discoveries in South Africa were large enough to cause a short but unusual period of U.S. peacetime wholesale price inflation averaging 2.5% annually from 1897-1914. See M. Friedman et al., Monetary History of the United States (Princeton Univ. Press, 1963), p. 135.

World War I so shaped the history of the twentieth century that it is hard to imagine what it would have been like without this almost inadvertent cataclysm. The classical gold standard could not accommodate at existing gold parities the wartime financing requirements of the principal belligerents. Considerable gold flowed to the United States, swelling its money supply and raising the general price level. After the war, the British made a critical error in trying to return to gold at the prewar parity, effectively forcing a severe deflation. France, which devalued after the war, fared somewhat better.

The gold standard, in a sense, fell victim after the war to its own earlier success, for a century of largely stable gold parities rendered the notion of a "good" or "necessary" devaluation anathema to many. Economists who assign major blame for the Great Depression to the effort to stay on gold are partly correct. But it was not so much the effort to stay on gold as Anglo-American policies aimed at preserving prewar parities that lay at the root of the difficulty. The enormous credit expansion associated with World War I was beyond remedy by a mere panic; it simply could not be handled other than by severe deflation or devaluation.

Although the gold standard could not prevent excessive credit expansions or even fix permanently appropriate gold exchange rates, it did effectively set interest rates within a rather narrow range. Under the classical gold standard prior to World War I, short term interest rates in both the United States and Britain tended to cycle between 2% and 5%. Very rarely and never for long did they breach these limits. S. Homer et al., A History of Interest Rates (Rutgers Univ. Press, 3d ed., 1996), pp. 207, 321, 357, 364-365. For a brief graphic history of real and nominal long term interest rates, see P. Brimelow, "What Drives Interest Rates," Forbes, Dec. 27, 1999, pp. 218-219.

Under the gold standard, business and credit expansions were typically associated with higher interest rates. Panics normally brought lower rates as fear reduced both willingness to lend and demand for credit. Prior to the stock market crash in 1929, short term rates moved over 5% as they had prior to the Panic of 1907 and during the war years. What was different in the 1930's was that short rates not only fell, but also remained stuck under 1% for several years. Central banking under the Fed, exacerbated by the monetary excesses of World War I, managed to accomplish what free banking and the Civil War never could: a severe multi-year national bust.

Today what was once simply banking is "gold" banking. Interest rates on gold are now "lease" rates. Yet their levels cycle within substantially the same range as before. Last fall's gold banking crisis demonstrated 5% gold lease rates to be as much a harbinger of trouble as 5% short term interest rates under the gold standard. Both signaled too much paper gold -- too much gold credit -- relative to available physical gold.

The question now is whether the recent gold banking panic will prove a relatively brief episode caused largely by temporary factors, or whether more fundamental distortions were at work. In the latter event, the 1929 experience suggests that gold lending and gold interest rates could remain depressed for a considerable period of time, and that a fundamental revaluation of gold may be necessary before the gold credit market can fully recover.

As the millennium turns, U.S. economists hail the "Goldilocks" economy. The Fed, originally formed to stabilize the gold value of the dollar, instead wages an undeclared hidden war on the discipline of gold. And for now, at least, relegated to the realm of quaint ideas from long ago is John Stuart Mill's admonition (Principles of Political Economy (orig. ed. 1848, 5th ed. 1877), Bk. III, Ch. XIII, s. 3):

Although no doctrine in political economy rests on more obvious grounds than the mischief of a paper currency not maintained at the same value with a metallic, either by convertibility, or by some principle of limitation equivalent to it; and although, accordingly, this doctrine has, though not till after the discussions of many years, been tolerably effectually drummed into the public mind; yet dissentients are still numerous, and projectors every now and then start up, with plans for curing all the economical evils of society by means of an unlimited issue of inconvertible paper. There is, in truth, a great charm to the idea. To be able to pay off the national debt, defray the expenses of government without taxation, and in fine, to make the fortunes of the whole community, is a brilliant prospect, when once a man is capable of believing that printing a few characters on bits of paper will do it. The philosopher's stone could not be expected to do more.

For almost 70 years, the United States -- contrary to its own Constitution and the most deeply held beliefs of its Founding Fathers -- has led the world down the path of unlimited fiat money. Its paper dollar has become the de facto international monetary standard; its debt the world's principal international reserve asset; and its trade deficits the world's main source of international liquidity. As a result, some 40% of outstanding U.S. marketable debt securities are now held by foreigners, up from 20% just five years ago. See M. M. Phillips, "Foreigners' Share of Treasurys Is Growing," The Wall Street Journal, Dec. 20, 1999, p. A2. And the U.S. trade deficit is now running at an annual rate exceeding $300 billion, a level previously quite unimaginable.

This situation would be dangerous under any circumstances. An historic U.S. stock market bubble fueled in large part by an out-of-control domestic credit expansion makes it explosive. Why? Because a simultaneous decline in the stock market and the dollar could cause interest rates to rise sharply rather than decline. The Fed cannot simultaneously support the domestic financial structure with lower rates and defend the dollar with higher ones. Its vaunted domestic powers could be checkmated by international demands, heightened by the dollar's role as the world's main reserve currency.

Under the severest strains, a system of unlimited paper money backed by a lender of last resort behaves quite differently from a system based on gold -- the money of last resort. Ultimately neither system can save imprudent lenders or borrowers from the consequences of their acts. But whereas the latter will stabilize at lower interest rates with the underlying monetary system still intact, a system based on unlimited paper will tend toward hyperinflation unless checked by very high interest rates, themselves business killers which will prolong and intensify the economic downturn.

In recent years many small countries have learned this lesson the hard way as international capital fled their currencies and financial markets. Boom has turned to bust, often quite suddenly. Few illusions are as dangerous as: "It's different this time." Except, perhaps: "It can't happen here."