You are here

More Proof - by James Turk

Section: Essays

by James Turk, April 21st, 2003

© 2003 by The Freemarket Gold & Money Report.

One of the statistics complied by the International Monetary Fund is the quantity of gold owned by the world's central banks. That weight is reported to be 32,291 tonnes of gold. Most people accept this number at face value and without questioning its accuracy. However, central banks actually own less gold.

In reality central banks own 32,291 tonnes of gold AND gold receivables. This distinction is important. From both a legal and an accounting point of view, gold in the vault is clearly very different from gold owed to you. The reason is that gold in the vault is much less risky than someone's promise to pay you gold.

This distinction between these two unlike assets is one of the most basic principles of accounting, namely, that cash is different from a receivable. For this reason, cash and accounts receivable appear as two different line items on balance sheets prepared according to generally accepted accounting principles. But some central banks do not report their gold assets using these sound and well-established accounting standards.

For example, the Bundesbank discloses in its 2002 annual report that it has €36,208 million of "Gold and gold receivables". It further sustains the fiction that these two different assets are one asset by stating in the footnotes to its financial statements: "At the end of 2002 the Bank's holdings of fine gold amounted to 111 million ounces." The Bundesbank does not, however, state anywhere in its annual report what portion of its gold is stored in vaults and what portion has been removed from the vault and placed at risk by being loaned.

Another central bank with a large gold asset is the Banca d'Italia. According to its 2001 annual report, which is the latest report available: "Monetary gold reserves were 48.1 trillion lire (EUR 24.8 billion, or $21.9 billion)." One would think from this statement that this "gold reserve" is sitting safely in secure vaults, as a reserve. But this central bank too has been withdrawing gold from the vault and placing it at risk. Its balance sheet also records "Gold and gold receivables", and like the Bundesbank, it fails to disclose how much of its gold has been loaned.

In contrast to these reports by the German and Italian central banks, the annual report of the Banque de France shows that none of its gold has been loaned. There is no gold receivable reported by it, so none of its gold has been placed at risk by being loaned.

There is also a third category of reporting. The Swiss National Bank, for example, uses generally accepted accounting principles to prepare its financial statements. Not only does it disclose that 254.7 tonnes of its 1,661.9 tonnes have been loaned, it provides information to assess the level of risk. For example, 158.7 tonnes were loaned on an unsecured basis.

Another central bank that discloses its gold lending is Banco de Portugal. According to its latest annual report, it has removed from the vault and placed at risk 434.1 tonnes of its 606.7 tonnes, or 71.6%, which is relatively much greater than the percentage of gold placed at risk by the Swiss National Bank, which is 15.3%.

Accordingly, there is no question that some central bank gold has been removed from vaults and loaned into the market. But because the level of reporting by the central banks is inadequate, it has been impossible to precisely determine the exact weight of gold removed from central bank vaults. This unknown weight of gold has become one of the most contentious issues within the gold industry. And the debate that has arisen as a result is well warranted.

If gold is removed from a vault and sold into the market, this dishoarding obviously will have an impact on gold's rate of exchange to the dollar and other currencies. This result from dishoarding is a basic principle of economics, but with a twist. An adaptation is necessary in a post-Gold Standard world to account for the fact that national currencies are no longer directly tied to gold.

Economic models prove that the extension of credit debases a currency, which is a principle that is true for any money, whether dollars, euros or gold. However, because goods and services are today priced in terms of national currencies - all of which are fiat and are only exchangeable for but not redeemable into gold - the impact of credit extensions in gold is different than the impact of credit extensions in national currencies.

When credit is pumped up using a national currency, it's a process that usually results in inflation; the prices of goods and services rise. The new extensions of credit increase the supply of the national currency, and if this growth in supply is greater than the demand for the currency (which has always been the case since the abandonment of the last remnants of the Gold Standard in 1971), the currency loses purchasing power. In other words, it is debased, and that debasement is reflected by rising prices. Each unit of currency purchases less and less. However, goods and services are no longer priced in terms of gold, so gold credit extensions have a different result on gold's purchasing power.

If gold credit extensions are greater than the demand for gold, it is debased, and like national currencies, it's purchasing power declines. But because goods and services are priced in national currencies, gold's debasement is manifested by a decrease in its exchange rate, or to put it in the terms commonly used, the ‘gold price' falls. In other words, gold when debased in this way purchases less national-currency-denominated goods and services. Thus, it is clear from this analysis that it is important to know how much central bank gold has been loaned, so that these credit extensions can be analyzed to assess their impact on gold's rate of exchange - the so-called ‘gold price' - compared to the many national currencies.

In recent years several efforts have been made to overcome the inadequate reporting of central banks in order to determine the weight of gold dishoarded from their vaults. Many people continue to accept the results prepared by Gold Fields Mineral Services, which have generally stated that around 5,000 tonnes have been removed from central bank vaults. However, I dismiss this number because GFMS surveys do not capture the weight of gold borrowed by commercial banks to fund their national currency assets, and my assessment is that this weight of gold represents the largest portion of gold loaned out by central banks.

Consequently, I have relied upon the work completed by Frank Veneroso and Reg Howe. Both of them have used a different methodology to reach basically the same conclusion, namely, that some 15,000 tonnes of gold have been removed from central bank vaults through lending and other forms of credit extension, such as swaps.

Frank Veneroso determined this number from a supply-and-demand perspective using various historical analyses, levels of economic activity and other statistics. His most recent report, Gold Derivatives, Gold Lending,Official Management Of The Gold Price And The Current State of the Gold Market, has been posted by GATA at

Reg Howe has concluded that the weight was 15,000 tonnes by analyzing the derivative activity of banks that is reported by the Bank for International Settlements. See the article posted on his website, Gold Derivatives: Moving towards Checkmate

Both of these reports are exceptional. The analysis of each is clear and thorough, and it is noteworthy that their conclusion from two entirely different approaches reinforces each report's accuracy. But for a while I have been wondering whether there might be a third way to reach this same 15,000 tonnes conclusion.

From time to time, I have attempted to analyze the flow of gold leaving central bank vaults, but have been thwarted by insufficient information. Some banks like the Federal Reserve Bank of New York report these flows, but the Bank of England and Swiss National Bank - both of which are among the dominant players - do not. Recently, I began to consider yet another alternative method of compiling central bank gold data.

I was put on this new path to discover a ‘third way' after stumbling upon an old report published nearly ten years ago by the IMF, The Structure and Operation of the World Gold Market. One table in particular, which I present below, caught my attention. By analyzing the UK's import and export statistics, the table showed the flow of monetary gold (i.e., essentially large bars of refined gold) through that country from 1960 to 1990.

moreproof1.gif (7536 bytes)

I was intrigued. The Bank of England is very secretive about its gold activity and does not disclose the quantity of gold held in its vaults for the custody of others, mainly central banks. For this reason, the table begins in 1960 with the 170.2 tonnes of net monetary gold imported that year. The exact weight of gold before then is unknown. Each subsequent year shows the net additions or subtractions to this cumulative stock of gold measured from 1960. From this table we can see the pattern of central bank activity in the gold market.

For example, note the huge dishoarding of gold from 1965 to 1968. We know from the dishoarding reported back then by the US Treasury that there were huge gold movements in the years immediately prior to the breakdown of the Bretton Woods system in March 1968. This table confirms that activity.

So I wondered, would the gold flows from 1990 to the present similarly reflect central bank activity? Would there be some evidence of the gold mobilization in the late 1990's that would confirm my analysis (see that central banks were actively intervening in the gold market during this period? In fact, I wondered whether HM Customs even continued to collect and disclose this data because it would provide compelling evidence of central bank gold activity.

So recently I spent one Saturday afternoon in London's Westminster Reference Library digging through stacks of data that probably hasn't seen the light of day for years. And I wasn't disappointed. Not only did Her Majesty's civil servants continue to collect the customs data with utmost efficiency, it was all disclosed. I even went back to 1989's data to make sure that I was compiling the gold flows in a way consistent with those earlier years. I was stunned, but not surprised, by the result.

Here are the results, along with data showing the weight of gold dishoarded in each of these years from the Federal Reserve Bank of New York

moreproof2.gif (6605 bytes)

I have several key observations/conclusions about this data:

(1) Note the 2,472.9 tonnes of gold exported from the UK in 1997. Not only is it a staggering amount (an amount equal to annual production), but it was in April of that year when I began to suspect that the gold price was being manipulated by central banks, see Why would this weight of gold, and indeed, the weight of gold cumulatively presented in the table above be mobilized unless to intervene in the gold market with the result of suppressing its price?

(2) It was not possible to determine from HM Customs data to where most of the gold was being exported, but if the data from the US Geological Survey for US exports is any guide, then most of the gold was shipped from the UK to Switzerland. As the IMF report states, even back in 1993 Switzerland "has retained its dominance in physical gold trading by providing specialized banking and ancillary gold services in an essentially unregulated and confidential environment." In other words, Switzerland is the center of the world's gold lending activity, and its transactions in physical metal dwarf the other gold-trading centers. While London remains the center for pricing gold, Zurich is by far the dominant location for transactions involving physical metal.

(3) Taken together the reports by HM Customs and the FRBNY show that 7,287.2 tonnes of gold were dishoarded from the UK and the FRBNY from 1991 through the first nine months of 2002. Now I recognize that this amount is not 15,000 tonnes, so some subjective analysis is required here. Specifically, the UK and US are secondary players in the physical gold market, as are the other centers such as Frankfurt, Hong Kong, Perth, etc. But in the absence of customs data from these other centers and given the 7,287.2 tonnes of gold mobilized from just the UK and US, it does not take a leap of blind faith to recognize that at least an equal amount would have been mobilized from Switzerland and the other centers, which would bring the total amount of mobilized gold to approximately 15,000 tonnes, again the same weight of gold determined by Frank Veneroso and Reg Howe.

(4) I would argue that in order to maintain the gold price suppression scheme, gold was exported from the UK and US to Switzerland for two reasons. First, physical metal had to be sent to Zurich to meet the ongoing supply/demand deficit reported by Frank Veneroso, as the available gold there was being used up in filling these deficits. Second, Switzerland is the ideal staging area for any price manipulation scheme. The huge weight of metal that has been used to suppress the gold price - some 15,000 tonnes - can be more easily hidden in a market where the largest volume of physical gold trading occurs. What's more, given the "essentially unregulated and confidential environment" in Switzerland described by the 1993 IMF report, central bank intervention in the gold market can easily be hidden, particularly if the interventions are camouflaged by acting through the secretive Bank for International Settlements. Thus, the gold market in Switzerland provides price manipulators with the ability to intervene in the gold market undetected.

(5) The marked decline in dishoarding for the first nine months of 2002 is noteworthy. Does it mean that the central banks are not going to unwisely dishoard more physical metal in their foolhardy attempt to maintain these low gold prices? Having mobilized and used much of their gold, are they now running out of ‘bullets'? Central bankers and their cohorts in the IMF have one aim - to maintain the illusion that the US dollar is worthy of being the world's reserve currency when in fact it is not. But central bankers will do everything within their power to perpetuate this illusion, including dishoarding their most precious asset, their gold. In the late 1960's they dishoarded some 10,000 tonnes of gold in a feeble attempt to pretend that gold was only worth $35 per ounce, when in fact everyone knew that gold was worth more. Today they have dishoarded probably 15,000 tonnes for the same inane purpose, different only by the near ten-fold increase in the dollar price of gold compared to back then, which is due largely to inflation and the resulting fact that it takes $10 today to purchase what $1 purchased in 1971. Central bankers know that a low gold price makes the US dollar look stronger than the underlying fundamentals warrant. My point is that they will eventually stop feeding physical metal into the market because they cannot risk using up all of their metal. If they did dishoard it all, a crisis of confidence would cause people to flee the fiat currencies they produce, which would then become worthless. Gold is power, and central bankers will not risk using all of their gold in a futile attempt to keep the gold price from moving higher. So the 2002 gold flows to me indicate that the tide has turned against the central bankers, which is a result also confirmed by the rising gold price last year.

In conclusion, I have established yet a third methodology to make obvious the weight of gold that has been dishoarded from central bank vaults. By using HM Customs reports to determine the weight of gold passing through the UK's borders and adding to that total the dishoarding from the Federal Reserve Bank of New York, along with similar activity assumed for Switzerland and the other lesser gold centers, we can deduce that some 15,000 tonnes of gold have been removed from central bank vaults.

Consequently, "gold receivables" equal 46% of the 32,291 tonnes of gold reported by central banks. So only 17,291 tonnes of physical metal is left in central bank vaults.

In 1945, 68% of the world's gold was in central bank vaults, and the total quantity of money, i.e., national currency in circulation, was about $300 billion. Today the total quantity of national currency is about $30 trillion, a one hundred-fold increase in 57 years compounded at an 8.4% annual growth rate. And central banks hold in their vaults some 17,291 tonnes of gold, which is just 40% of the weight they held in 1945 and only 11.9% of today's aboveground gold stock.

Clearly, these numbers show that today's monetary system is out-of-whack, that the money-substitutes produced by central banks have become too excessive and therefore overvalued against money itself, i.e., gold. The central banks and the fiat national currency they produce are therefore vulnerable. After decades of abusive policies undermining the purchasing power of those national currencies, the central banks are running out of gold, their most valuable and powerful asset.

Consequently, history will repeat. Gold will again soar just as it did after the last manipulation scheme failed in 1971.