This is the preface to an ambitious ongoing attempt to dig into the opaque subject of silver supply and demand analysis. Once I have this preface out of the way, I will examine and critique the “experts” including the GFMS World Silver Survey that was recently released. There is a lot of material to deal with here, so this is going to be Part 1 of a multi-part series. The preface will consist of today’s commentary and another one in the next couple of days, to be soon followed by a review of the GFMS, CPM Group and Fortis silver surveys. This series of commentaries will all be filed under the category “Silver Supply and Demand” and the tag “How Much Silver?”
I’ll start by referring to a recent commentary titled A Presidential Bombshell from Ted Butler, who makes a claim that today there is less than 300 million ounces of silver on “U.S. soil”. Here is Mr. Butler at his finest:
Here’s a statistic that is stunning and troublesome at the same time. In 1959, there were about 5 billion ounces of silver physically held on US soil. This includes the 3.4 billion Government holdings plus privately held silver, including hundreds of millions of ounces of silver objects that would be subsequently melted in the early 1980’s. Today, I think I may be exaggerating if I say there are more than 300 million ounces held on US soil, including all the 118 million ounces in COMEX-approved warehouses and privately held silver. Before you disagree, please remember that the more than 400 million ounces in ETF-type vehicles are held outside the US. If my numbers are accurate (as I believe them to be), then the amount of physical silver held on US soil is down 94% in 50 years.
Well, I’ve come up with some numbers of my own for the total amount of silver held on “US soil” and it should be no surprise that it is almost an order of magnitude bigger than Mr. Butler’s numbers. My numbers, however, are substantially less than the 1959 figure quoted by Mr. Butler and there are also qualitative factors that make my numbers very bullish under the right circumstances. The numbers will be explained and analyzed in subsequent parts of this series.
How to account for this vast discrepancy? Here’s one way. Mr. Butler states, in his latest commentary Silver Surplus?, that although there are almost 200 million ounces in Silver Eagles alone, this silver does not count as “bullion-equivalent inventory” because it would only be made available to the market (or melted down) at a much higher price:
Some may claim that the fabricated silver adds to the above ground inventory, particularly jewelry, silverware and coins, like American Silver Eagles. I agree that such silver still exists, but because it is not available for melting into bullion form at anywhere near current prices, it shouldn’t be counted as bullion-equivalent inventory. For instance, there must be near 200 million ounces of Silver Eagles fabricated over the life of the 23 year program from the US Mint, yet I doubt even a single Eagle has ever been melted for its silver content because they can be sold at a premium to their silver content. Why melt something to get less than you could sell it for? Same with the other forms of silver fabrication. Look at current recycling patterns, gold objects are being melted in record quantities, silver objects are not. When and if that changes, we’ll consider it as inventory.
I have a problem with this line of reasoning with respect to Silver Eagles because they are simply not very unique when considered as an investment. A large portion of them are in fact held as bullion and some of the remainder is held by collectors who would sell them if the price were right. In this respect, they are simply items of bullion, albeit with a higher premium than the typical items of bullion.
As with other bullion, there is no reason to melt Silver Eagles down because they serve a valid investment purpose just as they are. Evidence of this can be seen from the fact that there is a vibrant secondary market for them. Mr. Butler himself has said, although not in so many words, that if there were 200 million ounces of Silver Eagles in existence in 1980, he doubts even a single Eagle would have been melted down. Well, that’s because they can be sold at minimum for their silver content and perhaps even a premium. Once again, that makes Silver Eagles (at least many of them) items of bullion — or more appropriately “bullion-equivalent inventory”.
Despite the above criticism, I really, really like this term “bullion-equivalent inventory”. I think its invention can be safely credited to Mr. Butler considering that a Google search of that term brings up only one page of results all linking to Mr. Butler’s writings. In fact, Mr. Butler may have uttered that term just three times — once in 2005, once in 2006 and once (so far) in 2009. It would be interesting to see, via periodic Google searches, if this term will now enter the silver and gold market lexicon.
The definition of bullion-equivalent inventory appears to be simply the amount of metal that is (1) held in bullion form and (2) held for purposes of investment or gain.
To this excellent term, I would now like to add my own (admittedly less stellar): “deferred bullion inventory”. I suggest a definition for deferred bullion inventory as follows.
Deferred Bullion Inventory: Metal that exists in a form that is expected to be converted to bullion at some future date by turnover, recycling, melting, smelting and/or refining. The conversion to bullion could be the result of normal recycling activity at stable metal prices, opportunistic selling during price spikes, forced selling due to economic hardship or any of several other reasons.
In modeling the conversion of deferred bullion inventory to bullion-equivalent inventory, there is a fixed component representing normal turnover that is largely insensitive to prices and a variable component that is sensitive to prices. The fixed component is already captured in the surveys by GFMS and others in the “recycling” category, though some would argue not very well.
You will note that Mr. Butler himself admits that there is such a thing as a deferred source of bullion: “privately held silver, including hundreds of millions of ounces of silver objects that would be subsequently melted in the early 1980’s”. In other words, if the silver or gold price is high enough, additional bullion-equivalent inventory will be supplied from deferred bullion inventory.
Now, here is the important part, so please listen up. For the past 20 years the conversion process has run in reverse because silver prices were too low. That means bullion-equivalent inventory was reduced while deferred bullion inventory grew. As a result, the market today is not the same market as 20 years ago and thus the analysis that would have worked 20 years ago will no longer work today. Without examining bullion-equivalent and deferred bullion inventories, the silver analyst is simply ignoring the forest while counting the trees.
Looked at another way, bullion-equivalent inventory is the total amount of metal that is theoretically available to the market today at a given price whereas deferred bullion inventory is the total amount of metal that will theoretically be available to the market in the future at that same price. Price projections involve the future and therefore deferred bullion inventory, as a future phenomenon, must be taken into account. This is an important consideration that has been mostly ignored by “expert” supply, demand and price forecast of the silver (and gold) markets.
By now you might suspect that Mr. Butler’s 300 million ounces of silver on “US soil” is all bullion-equivalent inventory at near-current prices whereas my 2 billion ounce estimate both assumes a higher silver price and includes deferred bullion inventory as well. Before we get to the tables and figures that illustrate the difference, however, I would like to tie the above concepts together by taking a look at that very peculiar event, the Great Silver Melt of 1980.
1980 Great Silver Melt
One of the most important bullish factors for silver — regardless of how much silver Mr. Butler, I or anybody else claims is out there — is that there will never be another episode again quite like what the silver market experienced in 1980. In the commentaries to follow I will reference some of the accounts of the Great Silver Melt but for now it should suffice to say that they do little to shed light on the supply and demand undercurrents that existed in 1980. There is lots of talk about the Hunt Brothers and long lines at coin dealers but little in the way of fundamental analysis. As a result, the general impression that many people have is that “a wall” of silver supply buried investors in early 1980. So much so that “wall of supply” is now used as a canard or common refrain to portray that the current situation is every bit as viciously bearish as January 21, 1980. But that just isn’t true. Buyers weren’t overwhelmed by a wall of supply in 1980 as much as investment demand simply ran out of steam while deferred supply continued to relentlessly come on the market for many months and years. That deferred supply eventually overtook investment demand, breaking the back of the market and foiling its built-in equilibrium mechanism.
Simply put, my contention is that deferred supply, or more precisely deferred bullion inventory, had a lot to do with how the 1980 bull market ended. Even though hundreds of millions of ounces of silver was tendered to the melting pot over a very short period of time in 1980, smelter and refinery capacity was largely fixed and therefore it took a very long time to clear out the resulting backlog. Had the melt occurred at a faster pace or been smaller, more silver demand would have been satiated early in the process and prices might have reached an equilibrium level well above $5. Instead, the refinery backlog created an overhang — or deferred bullion inventory — at moderate (and declining) prices that wasn’t as much a wall as it was a steep hill. And at some point, silver investors grew tired of climbing it.
The inability of the market to clear out the deferred bullion inventory quickly in the early 1980s resulted in far-ranging effects that persisted for many years and shackled silver to a price with little range beyond $5. Stated a different way, the deferred bullion inventory so distorted the supply-demand balance that the silver market was unable to find its equilibrium for many years. By the time it did, many investors had given up.
So, the Great Silver Melt of 1980 didn’t really create a wall of bullion supply that was responsible for ending the bull market, it created a deferred supply of bullion that arrived on the market too late to satisfy the investment demand that had generated the high price level (and of course it was that high price level the deferred supply was chasing). As a result, an equilibrium process that should have taken a few months or years dragged on for more than a decade. By the time deferred supply was no longer a factor, many silver investors were ready to throw in the towel.
If we are to understand the importance of investment demand and supply to the future of the silver market, especially as prices head much higher, we need to learn the lessons of the Great Silver Melt. Some of these lessons are:
- Deferred bullion inventory has little effect on the current silver price but can have profound implications for the long term silver price. Mismatches between flows in bullion-equivalent inventory and deferred bullion inventory may prevent the market from clearing at its equilibrium price (the price might be either too high or too low). This would presumably frustrate forecasts that fail to take a holistic view of bullion inventories.
- If the rate at which deferred bullion inventory is converted to bullion-equivalent inventory accelerates, it is important to watch the conversion price. By conversion price I mean the price dealers are offering for deferred silver inventories intended for the melting pot. A conversion price at a major discount to the spot price of silver could mean that a refinery backlog is being created and that may have negative implications for the price going forward.
- The actions of all market participants, not just investors, needs to be taken into account. Many people believe that the big lines at coin dealers during the Great Silver Melt were manned primarily by people buying bullion. Not so. Most of the people were actually trying to sell deferred bullion inventory (silver coins, sterling heirlooms, etc.). There was plenty of investment demand as well, but the frenzy itself was primarily limited to people trying to cash in on sentimental junk lying around the house.
- Another important consideration was the dealers. They aggressively sought to buy both silver junk and silver treasures because the high silver price afforded them a deeply discounted offer price such that many dealers made huge profits even considering the low prices the refiners were paying. The best way to describe it might be as an orgy. Every bit of silver that was not nailed down and that did not represent bullion-equivalent inventory (investors had already bought up all the available bullion) was committed to the refiners, which were the only entities capable of ultimately monetizing this deferred bullion inventory.
We hear about walls of silver lurking over the horizon every once in a while but that is simply hyperbole. There wasn’t a wall of silver in 1980 — the Hunt Brothers turned to futures in part because there wasn’t enough physical silver sitting around. Only after the Great Silver Melt did there appear a grown silver bullion supply (or bullion-equivalent inventory) as the refiners worked down the backlogs. As already noted, this bullion supply was more of a steep hill instead of a wall.
In any case, Mr. Butler is completely correct in claiming that deferred bullion inventory levels are much lower today compared to 1979 (at least up to $50 silver), so any hills that might lie ahead shouldn’t be as steep. As the years pass, however, deferred bullion inventories will continue to expand, although not at a very fast pace. It will probably be many years if not decades (and perhaps never) before we might get a situation like what existed in 1979. A long-term silver investor ignores this long-term reality of the market at his or her peril.
For example, most new sets of sterling silverware or hollowware that are sold in the U.S. today (about 5 million ounces worth a year) become deferred bullion inventory at some price. These sets of sterling silverware are no different from the sterling silverware that were purchased in the heyday of America’s love affair with sterling (1880-1930), subsequently became heirlooms, and were mercilessly consigned to the melting pot in 1980. Actually, I take that back. Modern sterling silverware would probably not be consigned to the melting pot in large numbers at prices below $100 silver (real, not nominal price). I’m using $100 because that is around 5x the average current price of sterling silverware on the secondary market per ounce, which is pretty much consistent on a relative basis with the late 1970s. Worldwide, annual silverware and hollowware production is about 50 million ounces. In addition, some small portion of annual silver jewelry production also becomes deferred bullion inventory. Worldwide silver jewelry production is about 150 million ounces.
Slowly but surely, deferred bullion inventory has been growing year after year since about the mid-1980s even as bullion-equivalent inventory has been shrinking. We are nowhere near the 1979 level but we move closer to it, not further away, each year. Thus, deferred bullion inventory becomes an increasingly important consideration as long as silver remains at a price low enough to encourage its widespread decorative use. At (real) prices above $100 silver, deferred bullion inventory may start being converted to bullion-equivalent inventory at a significant pace. Some people accuse me of being too bearish, but I suppose these people think $100 silver is too bearish as well.
The most bullish case for silver would be if the price were to rise to a level that not only discourages widespread decorative use but also starts to gradually draw down the deferred bullion inventory. As long as mine supply does not increase substantially, the result would probably be a price equilibrium that greatly favors the silver investor in the long run as deferred bullion inventory is converted to bullion-equivalent inventory, which in turn is held in increasingly strong hands by patient investors. I believe a silver price of $25-50 would accomplish this.
If we can consider anything as a potential wall of supply, it is the ballooning holdings of the ETFs and other metal investment vehicles. As Mr. Butler points out, these ETFs primarily hold gold and silver outside the United States, but that doesn’t mean the ETF holdings are not bullion-equivalent inventory, only that they are not U.S. bullion-equivalent inventory but rather global bullion-equivalent inventory. Fortunately, the nice thing about ETF bullion-equivalent inventory is that we can track it with precision unlike deferred bullion inventory. Furthermore, if you are a subscriber to Metal Augmentor, you will have access to the ETF basis, which is the only tool so far that can accurately predict changes to ETF holdings (arguably the largest category of bullion-equivalent inventory today) in the immediate future. In any case, ETF dishoarding would take place very quickly unlike the uphill supply created by the Great Silver Melt of 1980, so after an initial price drop from ETF liquidation investors should expect the price to reach an equilibrium level in short order.
With that, I will wrap up today’s installment of this series. I would appreciate any constructive ideas, comments and criticisms in the Comments Section so that I can refine any clearly boneheaded thinking before building on it in future installments. At the same time, please allow me the opportunity to complete my entire argument before concluding that it is all rubbish or going off on some unrelated diatribe.