The Buyer of Last Resort Resigned in 1999

Rough diamond prices have roughly halved since their peak in early 2022, and the Bloomberg measure of natural diamond prices reached the lowest level in its history this April. Global production is at its weakest since 1987.

Anglo American bought its controlling stake in De Beers for close to thirteen billion dollars in 2011. It has written the business down three years running, to 1.56 billion in 2023, to a carrying value of 4.1 billion a year later and to 2.3 billion at the end of last year. Reporting this summer put an advanced offer for the majority stake at around a billion dollars. That is something like ninety two percent below what was paid fifteen years ago for the most famous name in the industry.

The substitute has taken the market by volume. Fifty one percent of newly appraised American engagement rings in the first half of this year had a laboratory grown centre stone, up from forty seven percent last year and six percent in 2019. That is the first time it has passed half.

The Historical Echo

For most of the twentieth century this market did not have a price. It had an administered one, and the administration was private.

The Central Selling Organisation grew out of the Diamond Corporation, formed in 1930 as a producers' cooperative. Rough stones from most of the world's mines were consolidated at a central sorting operation in London and offered to a closed list of buyers called sightholders, ten times a year, at prices that were not negotiable. The sightholder was handed a box and told the price. Refusing it risked not being invited back.

The system worked because it did two things at once. It controlled what reached the market and it had the balance sheet to absorb what did not. In the early 1990s De Beers mined about forty five percent of the world's rough and sold about eighty percent of it; the difference was bought from others specifically to keep it off the market.

Holding that much inventory is expensive, and by the end of the 1990s the stockpile had grown from about 2.5 billion dollars to five billion. In 1999 the company brought in outside consultants and concluded that the role of buyer of last resort was no longer worth its cost. It stopped stockpiling, stopped buying on the open market, and committed to selling only what it mined. The new strategy picked sightholders on their ability to market diamonds rather than their ability to finance inventory.

De Beers' share of world rough sales fell from around eighty percent in the early 1990s to sixty five percent by the end of that decade and to about forty five percent by 2007. Nobody took over the stabilising function. The industry simply stopped having one.

Where Patient Capital Is Positioning

Twenty seven years later the function is being performed again, by an unlikely party. Alrosa, the Russian state producer, mined about 29.7 million carats last year and has planned for 25 to 26 million this year, a cut of around fourteen percent its chief executive described openly as inventory management. A sanctioned producer is voluntarily withholding supply to support a price in a market it cannot freely sell into.

The supply side is shrinking without anyone's help. Argyle closed in 2020. Diavik produced its last rough this March after twenty three years and more than 150 million carats. De Beers has paused Venetia, South Africa's largest diamond mine, for about two years, and its raised production guidance for this year is an accounting consequence of consolidating a Canadian joint venture, not more rock.

The retail numbers contain the subtlety that most of the coverage misses. Laboratory grown stones have taken the majority of American engagement rings by unit, but the median laboratory grown ring sells for about 4,557 dollars against 12,961 for a natural one. Per carat, laboratory grown prices have fallen from roughly 4,622 dollars in 2019 to 2,057, while natural prices in the same retail series rose from 6,237 to 8,428. The substitute took the units and left the dollars.

For capital that thinks in decades this is a case study in what happens to a physical asset when the institution that priced it walks away. The diamond did not change. The scarcity did not change, and by production volume it has got more acute. What disappeared was a private balance sheet willing to hold inventory through a cycle, and nothing in a market of listed companies answerable quarterly was ever going to replace it.

The Capital Memo