The core claim
Central bank gold buying and selling has very little to do with where the gold price trends. It may nudge the price in the very short term, but it does not drive the long-term direction of gold measured against the US dollar. This runs against the popular belief that official buying is central to gold's moves.
The reason sits in motive. A private person buys gold to protect his purchasing power because he wants goods. In the deepest, apex stage of hyperinflation he isn't really buying gold at all, he's dumping dollars, so how much gold he gets does not matter to him. A central bank buys gold to pump liquidity into the economy. It must print dollars to buy the gold, just as it prints to buy anything. When it sells gold, it is trying to drain liquidity, so the number of dollars it collects does not matter to it either. It follows made-up equations from Keynesian econometrics.
Why central banks are not real economic actors
A central bank makes no personal decision about preserving its own purchasing power. It is manipulating its own liability notes, the things it calls currency or credit. Human action cannot be squeezed into constants or plugged into an equation, which is the Austrian school point, and it is correct. So central banks as a group are not economic actors. They have no stake in preserving purchasing power as individuals or as institutions. They are trying to rescue credit by steering its price against gold using formulas that aren't real.
The balance-sheet mechanics produce a result most analysts miss. When a central bank sells gold and pushes the gold price down, it drains the gold that backs its outstanding credit, so over time selling gold weakens the credit and strengthens gold. When it buys gold, it must inject more credit into the banking system to pay for that gold, which weakens credit against gold and lifts the price. Either way, the thing to watch is individuals judging their purchasing power and deciding whether they can still trust central bank credit notes.
The 1973 gold bottom
Gold topped near $126 in May 1973, then fell for roughly five and a half to six months to a low of $90.50. The New York Times put the low around $86, slightly under that. The bottom came on November 15, 1973. From there gold ran up about 100%, reaching roughly $175 to $180 just four months later, with the bulk of the rally in January and February 1974.
Two things happened around that bottom, and neither fits the standard story.
The first was a gold revaluation by the Federal Reserve in October 1973, a few weeks before the bottom. Many people think the gold standard ended when Nixon closed the gold window on August 15, 1971, but gold was officially revalued twice after that, up to $38 and then to $42.22 in 1973. That $38-to-$42.22 move was the last revaluation.
The evidence comes from a Federal Reserve Bank of St. Louis January 1974 review, page four, an article titled "The Monetary Economics of Gold" by Albert E. Burger, written about 52 years ago.
How the revaluation moved through the books
In the week ended October 24, 1973, following congressional approval, the official dollar price of gold rose by a little over 11%. The Treasury gold stock, valued at about $10.4 billion the prior week, was now worth about $11.6 billion. The official value rose by about $1.2 billion.
The Treasury then had a choice, as it had in May 1972: neutralize the effect on the monetary system, or monetize the $1.2 billion and spend it on goods and services, which would raise bank reserves by that amount.
On October 24, the gold stock value rose $1.2 billion and Treasury cash holdings rose $1.2 billion, with no net effect on the monetary base, because the money had not been spent. It was just an accounting entry, so it had not bid up prices and was not inflationary.
On October 25, the Treasury issued gold certificates to the Federal Reserve banks, and Treasury deposits at the Fed rose $1.2 billion. Another accounting entry. The gold value was now monetized and available for the Treasury to spend, but still unspent, so the monetary base was still unaffected and it was still not inflationary.
Then, over the three weeks from October 24 through November 14, the day before the bottom, the Treasury made payments. Treasury deposits at the Fed fell $1.2 billion while deposits of the public, the monetary base, and bank reserves rose $1.2 billion. The money was spent into the economy.
What happened on November 14
The New York Times ran an article on November 15, 1973, the day of the bottom: "Price of gold drops in Europe." From Paris, November 14: gold dropped sharply in Europe's financial centers after seven nations disclosed they were ending an accord that had barred monetary authorities from bullion transactions on the free market. The decision, at least in theory, cleared the way for potentially massive selling of the metal. Central banks had controlled half the gold mined since the 17th century. The news drove a $7-an-ounce fall in London and Zurich, the two main open-market centers. Gold mining shares fell on European and South African exchanges. Bullion closed near $90 an ounce after touching $86 in hectic trading. The prior July, during a monetary crisis, gold had hit a record above $125 an ounce. Most read the move as meaning a reduced monetary role for gold.
The participants were the United States, West Germany, Britain, Italy, Belgium, the Netherlands, and Switzerland. They declared their right to sell gold in the markets if they chose. The question of gold purchases stayed open, with the US position declared and the others less certain.
So on November 14, 1973, the major central banks announced they intended to sell gold at free-market prices, something they had not been allowed to do until that decision passed. Gold bottomed the next day and took off.
Decades of data
Central banks then sold gold for the next six to seven years. A gold.org chart tracking central bank buying and selling since the 1950s tells the story.
From 1950 to 1966, central banks bought gold hand over fist, and it didn't affect the price because the world was still on a gold standard. From 1966 to 1968, central banks sold gold to keep the London gold pool alive at the $35 statutory rate, and that selling did affect the price by holding gold in line. So official selling isn't powerless. It just doesn't matter much in the larger scheme.
In 1973 there was no gold action among the big central banks. Then in 1974, two months after the November 1973 rule change, the selling started and ran through 1974, 1975, 1976, 1977, 1978, and 1979. That stretch was the biggest gold bull market in history by percentage, and it lined up with massive central bank selling. Because gold bottomed at the exact moment central banks announced they would sell, that is when the price began to skyrocket.
The pattern repeats. The gold bull market that began in 2000 to 2001 started in the face of huge selling. Central bank buying only started in 2011, right as the gold bear market resumed and ran from 2011 to 2015.
There is almost an inverse relationship between central bank buying and selling and the gold price, though not strictly inverse and not a clean contrarian signal. The takeaway holds: the strongest bull markets often begin when official action looks the most bearish. Confidence in paper assets can outweigh official intervention until that confidence finally breaks, and that break is what matters.
