On July 22, I summarized Seagate’s (NASDAQ:STX) earnings commentary partially blaming a “bubble” in rare earth prices for causing a margin squeeze for the company. Seagate management claimed that the soaring prices of rare earth elements (REEs) are very similar to a brief episode of soaring prices of ruthenium back in 2006. The price bubble for ruthenium was quite dramatic. In about a year, ruthenium prices increased about 8x. The Minor Metals Trade Association (MMTA) provides a graph from Johnson Matthey showing the 2006-07 run-up in price:
[Click to enlarge]
Before the 2006-07 run-up, ruthenium's price had been relatively stable. Seagate’s implication was that ruthenium’s price run was a bubble and so to the price of rare earths must soon collapse.
Ruthenium is a rare metal (not a REE) found in platinum ores. Seagate’s early use of ruthenium as a thin barrier between two magnetic layers allowed the company to achieve greater data densities (see Performance Media: Tweaking Magnetic Capabilities, August, 2001). Seagate holds a patent using ruthenium for the read head of a disk drive (see Giant magnetoresistive sensor with high-resistivity magnetic layers, filed Jan. 2002).
I decided to dig a little deeper into what happened in 2006 to ruthenium. I did not find a supply/demand situation that looks anything like what is going on with rare earths today. For this research, I sought the advice of Johnson Matthey Public Limited Company, a company quoted by the MMTA in describing ruthenium. Based in the United Kingdom, Johnson Matthey is a specialty chemicals company focused on advanced materials technology. Terry Harris, a regional sales manager in Asia, has spoken at conferences about the ruthenium market (for example, Minor Metals & Rare Earths 2010 in Xiamen, China, October 18-21, 2010). Harris was not available to answer my questions, but Peter Duncan, head of market research, provided all the details I needed.
Here is what Duncan explained to me:
Ruthenium is one of the minor platinum group metals, with an annual supply of around one million oz (as compared with platinum supply of about six million oz). Although supplies of ruthenium are usually sufficient to meet industrial demand for the metal, being a relatively small market it is always at risk of periodic price spikes when a sudden surge in demand places an unusual pressure on supplies. As a by-product of platinum mining, there is no possibility of a primary supply-side reaction to demand.
The hard disk industry … uses ruthenium as a support layer coating on hard disks. Towards the end of 2006, the hard disk industry underwent a step change technology shift from “horizontal magnetic recording” to “perpendicular magnetic recording” (or “PMR”). This resulted in a considerable increase in the amount of ruthenium deposited on each disk. Whilst the net demand for ruthenium in this application was not in itself enough to push the market into deficit, the process by which the metal is deposited, called “sputtering,” requires an enormous pipeline of metal because only about 4% of the weight of the starting material (the sputtering target) is deposited on the disk. This meant that the entire hard disk industry was suddenly looking to fund enormous pipeline builds, and the only way to do that was to buy the metal. This created a massive surge in demand at the end of 2006 and into 2007. Once the pipeline was filled, however, metal started to be released from the back end of the process. Refining capacity rose to meet the new inflows, and the net demand for ruthenium settled down at a level close to that of the net [supply]. This brought the market quite rapidly back to balance, causing the price to fall sharply.
The term “bubble” was quite appropriate for what happened to the
ruthenium price at that time. A market which had been long in adequate supply was suddenly tipped into sharp near-term deficit, until the PMR impact had run its course.
In the absence of any significant similar surges in demand, the
ruthenium price has been stable for several months now at around the
level of US$180/oz. That’s not to say that in such a small market the
price couldn’t spike again, but I would have to say that the size and
nature of the PMR technology move was rather unprecedented ….
It is certainly different from the rare earth situation, where demand
is rising at a steady and reasonable pace, but supply is of course
restricted by export quotas from China.
In other words, ruthenium had a brief price spike because one new application caused an unprecedented surge in demand for the metal. Once the materials industry was able to adjust to that demand, prices fell back to historical levels.
The current REE crunch was exacerbated by Chinese quotas which are not likely to improve significantly anytime soon. The supply adjustment that the REE industry must make is very significant given entire mines have been restarted and supply chains built and/or rebuilt. This adjustment has been unfolding over several years thanks to companies like Molycorp (MCP), Lynas (OTCPK:LYSCF), and Great Western Mineral (OTCPK:GWMGF), and it will take several years more – quite unlike the mere months it took to increase ruthenium supplies from mines already in operation with functional supply chains. While this cycle plays out, the demand for REEs will continue to increase across a broad range of applications and products.
Seagate appeared very hopeful that the quota situation will turn out to be a quirky aberration and that the supply situation, especially in the United States, will respond just as it did for ruthenium. I do not see the basis for this optimism, and suspect the company is actually rapidly moving to accommodate a prolonged period of higher prices.
Finally, note in my previous piece that I made a reference to dysprosium as a likely rare earth used in Seagate’s hard drives. REE expert Gareth Hatch declares this cannot be the case in a technically detailed piece called Seagate, Rare Earths And The Wrong End Of The Stick. Here is the “slam dunk” quote:
Virtually every man and his dog in the rare-earths sector will tell you that all Nd-Fe-B magnets need Dy as an additive to make them work at higher temperatures. And I’m here to tell you that every man and his dog would be wrong.