Central Bank Gold Agreement
Last Friday, the central banks of Europe extended their landmark agreement on gold sales. 18 national central banks, along with the European Central Bank itself, signed the third Central Bank Gold Agreement. CBGA 3, like its two predecessors, has major implications for gold that investors need to understand.
Due to their propensity to sell gold from their massive hoards, central banks have long sparked fear and suspicion among gold investors. While CBs absolutely add supply, thus weighing on gold’s price, the misinformation and intentional disinformation surrounding these institutions is often way overdone. Instead of fearing them, taking a pragmatic perspective is far more prudent and profitable for investors.
Paranoia aside, ultimately central banks are nothing more than large gold investors. They buy and sell great quantities of gold over long spans of time measured in decades. They manage portfolios of reserve assets, the most important of which is gold. CBs are much like private investors in many ways. They are heavily influenced by popular greed and fear, they strive to maximize the values of their portfolios, and they need to diversify and rebalance these portfolios from time to time.
Like any investor, CBs’ intentions and biases can best be inferred by studying their trading behavior. And there is no span of time more relevant to today than the past decade. Incidentally, this is the very period over which the first two CBGAs were in force. Walking through CBGA 1 and CBGA 2, seeing the history that shaped the newly-signed CBGA 3, will help investors better understand CBs’ likely future impact on gold.
A decade ago this month, gold looked terrible. After suffering through a 20-year secular bear, it was languishing in the $250s in August 1999. Investors understandably wanted nothing to do with it. Gold was perceived as a capital wasteland, an anachronistic relic passed by in the brave new era of the surging technology-stock bull. Provocatively, this despair-ridden sentiment was quite a problem for CBs.
Conspiracy theorists often point out that gold, history’s ultimate form of money, is the mortal nemesis of today’s fiat paper currencies. They are certainly right on this fact. But then they extrapolate it too far, inferring that CBs want to drive the gold price as close to zero as possible. What better way to eliminate the monetary competition than killing gold? The 10 years of CBGA history utterly refute this extreme thesis.
Just like any investor at the end of a multi-decade bear, in 1999 CBs wanted to lighten their gold holdings. No one is excited about any asset after 20 years of price declines. In addition, the European central banks’ reserves portfolios were dominated by gold. With 70% to 90%+ of their foreign-exchange reserves in gold, they felt the need to diversify out of such a poor-performing asset. But each time one sold gold, it would further spook private investors. Few would buy with the ever-present threat of future CB sales.
Thus the European CBs, wanting to minimize the adverse price impact of their own gold selling on their own reserve gold, collectively decided to create a formal and transparent framework for gold sales. On September 26th, 1999, 15 European central banks signed the equivalent of a treaty then known as the Washington Agreement (they met in DC during the annual IMF meeting). They issued a simple press release outlining the principles of what would later be called the Central Bank Gold Agreement.
CBGA 1 started out by stating “Gold will remain an important element of global monetary reserves.” The 15 CBs agreed not to enter the gold market as sellers, except for already-decided sales. They said these sales would not collectively exceed approximately 400 metric tons per year. And since this initial Agreement would run for 5 years, the 5-year total gold sales wouldn’t exceed 2000t. They also “agreed not to expand their gold leasings and their use of gold futures and options over this period.”
Now with worldwide gold production running 2540t in 1999, an additional 400t a year in supplies from CBs was not trivial. CBGA 1 effectively increased the annual gold supply by 16%. You’d think this was bearish, right? On the contrary, it was very bullish! Over the 3 trading days after this Agreement was announced, gold rocketed 13.7% higher. Gold has not seen a bigger 3-day run over the past decade, it was massive. Had it started at today’s levels ($950), such a run would blast gold $130 higher in 3 days!
The original CBGA was bullish because it took a great deal of uncertainty out of the gold market. Investors finally knew that European CBs would not dump too much gold too fast, cratering the gold price. This started to restore their confidence in buying gold near multi-decade lows. Even though the secular gold bull wouldn’t formally start until 18 months later in April 2001, the CBs’ proclaimed restraint laid a critical psychological foundation stone for the huge gains to come in this gold bull.
Make no mistake, CBGA 1 did not emerge out of concern or charity for the gold investors. The CBs were simply acting in their own self-interest in trying to restore confidence in gold so they could offload their “excess” gold at more favorable prices. Their gold-heavy reserves also look better, are worth more, when gold is trading higher rather than lower. CBGA 1 stabilized the gold market, enticing investors back in which gave CBs someone to sell to.
Over the 5-year term of CBGA 1, the signatory CBs sold exactly 2000t as planned. Yet despite these major supply headwinds, gold still rose 52% between September 1999 and September 2004. Incidentally, CBGA years end on September 26th since that is when CBGA 1 was signed. CBGA 1 was such a success from the European CBs’ perspective that they extended it by signing CBGA 2 on March 8th, 2004.
Once again 15 European CBs signed CBGA 2 (although Greece replaced the UK). Once again, it opened with “Gold will remain an important element of global monetary reserves.” While the signatory CBs again agreed on a new 5-year term ending September 2009, they upped their annual sales quota to 500 tonnes. They committed to not exceed 2500t of total selling over the CBGA 2 period. They also agreed not to expand their gold leasing, futures, and options use beyond their pre-CBGA 1 levels from September 1999.
Interestingly, despite this 25% increase in the annual quota of CB gold hitting the marketplace, gold did nothing after CBGA 2 was signed. It traded within a half-percent band of the $400 it was at when this news hit for the next 5 trading days. While bigger CB sales were bearish, the fact that the CBGA was extended was bullish since it radically reduced the probability of a big surprise CB sale tanking gold.
Later several other smaller countries joined the CBGA as they adopted the euro. These include Slovenia in December 2006, Cyprus and Malta in January 2008, and Slovakia in January 2009. These additions brought the total signatories to the CBGA up to the 19 CBs that signed CBGA 3 last week. But before we dive into CBGA 3 and its implications for gold, we need to dig deeper into CBGA 1 and CBGA 2 sales.
This table includes the official gold reserve holdings of the major central banks involved in the first two CBGAs. I defined “major” as any signatory CB that had more than 100 tonnes of reserve gold. This data is courtesy of the World Gold Council, which does outstanding fundamental research. And despite the CBGA years ending September 26th, the numbers in this table are as of calendar year-end (except for this year which is current to Q2).
To better visually parse the changes in reserves, I highlighted large selling. Yellow, orange, and red cells respectively represent CBs selling 5%+, 10%+, or 15%+ of their gold reserves in a single calendar year. The total percentage decline in each CB’s gold reserves between the end of 1999 and Q2’09 is shown in the right column. Amazingly the CBs stuck to both of their Agreements so their selling was very orderly.
The big sellers over the CBGA 1 years were Austria, Portugal, Switzerland, and the UK. The early Swiss and British gold sales, while within CBGA limits, were particularly irritating to gold investors at the time since gold was languishing near secular-bear lows in early 2001. But the neat thing is these sales were coordinated within CBGA boundaries. The Swiss didn’t start selling really aggressively until the lion’s share of the British sales were done. And Portugal waited until late in CBGA 1 to start selling its gold.
The net result of all this selling, 2000t worth, was remarkably orderly. The total gold holdings of the major CBGA 1 CBs shrunk by 2.7% in calendar 2000, 2.6% in 2001, 2.7% in 2002, 3.0% in 2003, and 2.7% in 2004. While such a vast amount of gold would have crushed prices if dumped all at once, the secular gold bull still had no problem getting underway since this CB selling was measured and orderly.
And this is saying a lot given the magnitude of the selling certain CBs wanted to do. In 2000 and 2001, the UK sold 24% and 27% of its total reserve gold! Incidentally the driving force behind these gold sales, Gordon Brown, is now the Prime Minister of the UK. The 415t he sold near multi-decade lows ($285 4-year average) cost the British people $9b compared to what that gold is worth today. Gordon Brown, and other guys running CBs, made the classic investor mistake of succumbing to fear and selling near secular lows.
By the time the second CBGA went into effect (September 2004), different CBs were initiating sales at far more favorable gold prices than seen in the first CBGA. The European Central Bank itself, the Netherlands, and Spain became major new sellers. Still, even with the annual limits raised from 400t in CBGA 1 to 500t in CBGA 2, the overall selling was pretty orderly. In calendar 2005 to 2008, total major CBGA CB holdings fell by 5.3%, 2.6%, 4.0%, and 2.2%. Only 2005’s 5%+ decline is notable, everything else was moderate.
As the right column shows, the total declines in CB gold reserves over this decade-long CBGA-ruled span were often massive. Switzerland sold 60% of its reserve gold, the UK 51%, and Spain 46%! Total major CBGA CB gold reserves fell by 25% over this span, by 4000t. This is really a huge number. In 1998, the total global gold reserves per the WGC ran 33,536 metric tons. By the end of 2008, they were down to 29,727t. The European CBGA selling was responsible for the fastest decline in official gold stockpiles ever witnessed.
Despite this massive headwind of new supply, gold more than held its own. Its bull market over the entire decade-long CBGA span was one of the biggest and best in all the world’s markets! How could gold soar in the face of relentless CB force-feeding into the market? Simple, global investment demand increased at a faster pace than CB selling. This next chart highlights the CBGA’s impact on gold prices.
The CBGA-year quotas are rendered in red, with actual selling in CBGA years in blue. These numbers are in tonnes. The blue percentages show how much gold was sold relative to quota in each CBGA year (ending September 26th). Against this backdrop, the calendar-year average gold price is shown in yellow along with annual percentage gains. And for a read on just one form of investment buying, GLD gold ETF demand from stock investors, the net gain in GLD’s tonnage in each calendar year is shown in green.
Across all the major and minor CBGA central banks, 3867 tonnes of gold were sold in the last decade. Despite this, the average annual gold price surged 229% since 1999! And this measure is conservative since gold didn’t truly bottom until early 2001. The most important message from this graph, and indeed this entire essay, is that growth in gold investment demand during a secular bull easily outpaces even aggressive CB selling. As long as all the CB gold isn’t dumped at once, gold can still thrive.
Every year since this secular gold bull began, the average gold price has risen considerably. These gains range from a respectable 9% to a stellar 36%. If 400t to 500t of CB gold dumped annually failed to short-circuit the first half of this secular gold bull, then why on earth does anyone fear central banks today? Sure, their sales add supply and retard gold’s rate of advance. But in the end investment demand rules the day in a secular bull.
And investment demand only grows throughout such a bull. Just as no one wanted to own gold in 1999 late in its secular bear, everyone will want to own gold late in this secular bull. Yet so far, gold remains a fringe contrarian asset with very little mainstream following. The net addition to GLD’s gold holdings each calendar year, shown in green, highlight the increasing gold investment demand as this bull matures. While just one form of gold investing, the surging interest in GLD shows how easily CB sales can be absorbed.
Remember that CBs are just large gold investors subject to the same psychology as all investors. Thus it is fascinating to see the declining rate of CBGA sales relative to quota. The last time the CBGA CBs really got close to hitting their CBGA 2 500t annual limit was way back in 2005. They only sold 79% of their allotment in 2006, 95% in 2007, and 72% in 2008. And so far in 2009, with this CBGA year ending in just 6 weeks, CBGA selling is only running 27%. It’s at just 136t out of the 500t annual limit!
Clearly for some reason the zeal for selling gold reserves among the CBGA CBs appears to be waning. While individual CBs have different reasons for slowing their gold sales, a couple overarching themes probably apply to all. When gold was in the $200s, psychology was bearish and no one including the CBs wanted to hold it. But now with it up in the $900s, central banks are much more bullish on it and thus less inclined to diversify away from it. The CB fear is gradually morphing into greed, just like in all investors!
On top of this, most of the European CBs selling gold over the last decade were diversifying into the US dollar. But since July 2001, 3 months after gold’s secular bottom, the US Dollar Index has lost 41% of its international value in a nasty secular bear. If you ran a central bank, even if you felt you had too large of allocation to gold, would you want to sell it to buy US dollars when the former is growing stronger while the latter is growing weaker? Me either. Gold looks far more relatively attractive today, thus much harder to sell.
Interestingly, the European CBs themselves just officially acknowledged they are going to slow their gold sales in the future. Last Friday, August 7th, 2009, the 19 CBs now in the CBGA officially signed CBGA 3. Once again, they led off with “Gold remains an important element of global monetary reserves.” And incredibly, they actually shrunk their annual quota back down to 400t per year between now and 2014! This alone shows they are selling less gold, but the IMF comments in the release cement this case.
For years, the IMF has wanted to sell gold. This supranational bank-like organization is actually the third largest holder of gold on the planet with 3217 tonnes. For at least a decade now, the IMF has been campaigning to sell 403t of gold. Periodically potential IMF gold sales scare gold investors, but it is important to realize liquidating IMF gold is not an easy thing. Its “shareholders” are nations, and 85% of the “shares” must vote in favor of selling gold. But the US alone, with its 17% stake, has a de-facto veto. Thankfully it is not easy to get all the world’s nations to agree on gold sales, as many benefit from strong gold prices.
Anyway, even though the IMF is not part of the CBGA, the CBGA 3 signatories said, “[we] recognize the intention of the IMF to sell 403 tonnes of gold and noted that such sales can be accommodated within the above ceilings.” In other words, this lower 400t annual quota will include any IMF gold sales. This means that the CBGA CBs themselves plan to sell even less gold than their lowered quota indicates.
Obviously this is very bullish. Not only is the risk of a big surprise sale radically reduced, but today’s much slower selling pace by European CBs is likely to persist. They aren’t so keen to sell gold after it has nearly quadrupled in its secular bull in order to buy the US dollar which is approaching getting cut in half in its secular bear. They see the Fed’s highly-inflationary monetization and Washington’s mammoth deficit spending, so they know gold is going to fare vastly better than the ailing US dollar.
Today the CBGA signatories have 54.9% of their collective forex reserves in gold. So they will probably continue selling even if it is at a much slower pace. Remember that virtually all the global CB gold sales over the last decade emerged from these European CBs. But their “market share” is dwindling. Every tonne of gold they sell reduces their holdings and future impact on the gold market. This trend towards less importance is exacerbated by the 2350t or so of new gold mined every year. CBGA CBs’ gold dominance is eroding.
Meanwhile, the Asian CBs that are likely to grow dramatically are radically overweight US dollars and need to buy lots more gold in the coming years. The top 5 European CBs in terms of gold holdings are Germany at 69.5% of its reserves in gold, Italy at 66.1%, France at 73.0%, Switzerland at 37.1%, and the Netherlands at 61.4%. In stark contrast, the top 5 Asian CBs are China at 1.8% of its reserves in gold, Japan at 2.1%, Russia at 4.0%, Taiwan at 3.8%, and India at 4.0%. Unlike the European CBs, the Asian CBs need to diversify into gold. They are woefully underweight it.
And they will. As this secular gold bull continues higher and gold becomes ever more attractive as an investment, CB gold buying from the East should easily eclipse CB gold selling in the West. Worldwide, there is little doubt that CBs will become net buyers of gold before this secular bull ends. Obviously this is extremely bullish, since CB supply up to this point is the only thing that capped gold prices at a mere quadrupling.
So contrary to Internet mythology, central banks are no threat to this gold bull. If the CBs really wanted to crush gold, they wouldn’t have agreed to 15 years of measured and orderly sales via the CBGAs. The European CBs could have all sold in 2000 and 2001, which would have slaughtered gold since sentiment was so poor then and they had such dominant market share. But today, even without the CBGA, they are just too small relative to investment demand. Gold’s bull did and will easily power higher despite them.
At Zeal, we zealously study the markets because superior knowledge leads to superior investing and speculating returns. We don’t get emotional about anything, including the shrill anti-CB rhetoric, because emotional trading always yields poor results. Instead we strive to really understand what is going on, accept the reality of the markets, and trade accordingly. If you want to really deepen your knowledge of the markets and learn of high-potential trading opportunities, greatly increasing your odds for a successful financial future, subscribe today to our acclaimed monthly newsletter.
The bottom line is the three Central Bank Gold Agreements prove the world’s biggest gold-selling CBs aren’t trying to drive gold to zero. Over the 10-year span of the first two CBGAs, within which gold nearly quadrupled at best, these CBs stuck to their word and only sold in a measured and orderly fashion. Based on this long history of CB gold-trading action, it is illogical and naive to actually fear central banks today.
This secular gold bull, driven by growing global investment demand, will continue powering higher no matter what the CBs do with their gold hoards. Every tonne of gold the CBs sell lowers their current market share and future influence in the global gold market. And outside of Europe and the US, most of the rest of the world’s CBs have too little of their reserves portfolios in gold so they’ll probably become big buyers.
Adam Hamilton, CPA
Aug 14, 2009
Thoughts, comments, or flames? Fire away at email@example.com. Due to my staggering and perpetually increasing e-mail load, I regret that I am not able to respond to comments personally. I will read all messages though and really appreciate your feedback!