Looking at Silver
Jessica Cross and Gary Mead
This time 5 years ago
June 14th 2002
Daily fix: $4.89/oz
Dollar to Sterling: 1.47
Yen to Dollar: 125.1
Euro to Dollar: 1.06
1 month lease rate: 0.72%US Mint, June 2nd 2002:
According to mint statistics the US Mint's American Eagle silver bullion coinage program passed the 100 Moz sales mark. With an average London price fix of $4.35/oz (since the programme's inception in 1983) total revenue from sales amounted to approximately $517 million.
Innovative Medical Services (IMS), April 1st 2002: IMS has made an announcement regarding its silver based Axenohl product which could become an important non-toxic and non-irritant, over the counter topical remedy to kill off bacteria that can cause acne.
Silver Institute, 18th April 2002:
Kodak are to introduce a camera called the Advantix EasyShare that will combine digital photography with traditional silver backed film. Customers will be able to receive prints from their retailer whilst maintaining a negative and also having the option of e-mailing the images as a digital picture.
Reuters, May 23rd 2002: According to the Peoples Bank of China, 21 types of silver coins along with a series of 10 commemorative precious metals coins will be issued throughout the year. More than 2 Moz of silver will be used to mint the coins, 474,000 oz more than in 2001.
Feature: Deficit? What deficit?
In 2000, an American-based precious metals investment website posed a very pertinent question concerning publicly available silver statistics: if the market is in deficit and had apparently been so for many years, why had the price not risen?[1] This is an entirely valid point − the accompanying chart covering the period 1982 to the current year, shows that the silver price from around late 1985 to 2003 languished between $4 and $6 per ounce (with the exceptional price rallies of 1987 and 1998). A market in persistent deficit yet prone to a consistently weak price flies in the face of economic theory. What was really going on?
Source: Reuters EcowinThis question we set out to answer by re-examining the silver industry from scratch, putting together an independent supply/demand imbalance going back to 1980. In line with the statistical methodology we apply to all commodities - metals, energy and agri-business - our supply/demand balances do not balance. Each sub-sector identifies what we can measure or proffer model-based estimates, the difference between any annual overall supply and demand being an indication of either a surplus or a deficit. These surpluses or deficits represent what we cannot measure (e.g. investment demand) or measure with any degree of certainty (e.g. dishoarding). Our residual therefore represents - in Donald Rumsfeld's memorable phrase - the known unknowns; the inference being that this is metal either available (surplus) or not available (deficit) for investment. Markets in annual deficit rely on above ground stocks to alleviate the shortage in the underlying commodities. Markets in annual surplus see the levels of aboveground inventories rise.
Thus a supply/demand balance in large deficit, with anecdotal evidence of strong investment interest, could be expected to result in strong upward price pressure. Conversely, the price of a commodity whose market remains in surplus, with scant investment interest, might be expected to show a tendency to price weakness. This we believe is exactly what the status quo happened to be in the silver market for the years 1985-2003.
This next chart shows our calculated annual silver surplus for the period 1982-2000. The left hand axis shows it in tonnes and the right hand axis shows the surplus as a percent of annual supply. Interestingly, it shows a declining surplus over the period.
Source: VM GroupWhat we are saying is that precious metals investors asking the question posed at the start of this argument should turn the question on its head. What they really need to be asking is - given the fact that the silver price has remained so low, how come the data shows this was a market in deficit?
In the first instance our answer is posited on some reconsideration of fundamental supply data. We believe that the recycling of silver from the photographic industry has always been more efficient than claimed. Our development of a detailed recycling model supported this and our methodology is discussed in more detail in the next section. Unlike e-waste, this industrial process tends to be concentrated geographically. Collection of the waste is much simpler and hence more efficient than with e-waste. Contrast the position of a recycler collecting photographic waste from film processors and hospitals with that of one faced with the task of collecting personal computers from millions of individual homes. Second, although there are high collection losses largely associated with low silver prices, there has always been an awareness of the potential value in photographic waste; a phenomenon which shows a high degree of price sensitivity.
Secondly, we also believe that silver jewellery offtake in many parts of the world, especially countries of the Indian subcontinent like Pakistan and throughout much of the Middle East, has been underestimated. Our analysis reveals a relationship between the volumes (as opposed to values) of gold jewellery and the silver jewellery bought in various countries and regions. In many parts of the Indian subcontinent silver is the first point of market entry with households intending to upgrade to gold, when and as soon as they can afford to. This implies that the ratio of silver to gold jewellery purchases in volume terms is high. It also implies that silver jewellery is readily sold back and recycled. Our analysis supports both these implications. These higher jewellery figures go some - but not all - of the way to mop up the higher recycling figures.
Thirdly, as a direct consequence of higher jewellery offtake in a sector which exhibits a high degree of price sensitivity, our jewellery recycling figures are higher than public domain figures.
Taking all these factors into account we conclude that the silver market during the period 1985-2003 was substantially more liquid than might have been the case. Our analysis suggests an inferred physical surplus which would be available for investment. But of course we know that for long periods the investor was conspicuously absent, and it therefore follows that the silver price showed such sluggish behaviour.
Should this come as a surprise?
Not at all. With kind assistance from Tim Green, we have computed the total amount of silver we believe has been mined over time. Going back to the eleventh century, we estimate that this figure totals 1.17 million tonnes (Mt) of which the great majority was in the last 100 years.
Source: Virtual Metals extrapolated from various historical sourcesGiven the relative value of silver and gold, we assume that losses of silver stocks are far greater than those of gold. If one accepts that at least 40% of the silver mined has over the years been lost, then above-ground stocks today can be estimated to amount to 700,000 tonnes of metal[2]. A good deal of this metal has been recycled many times over.
But what of the future?
Since 2000, market circumstances have changed. Because of the levels of aboveground stocks, the silver market has and continues to be prone to surplus, in exactly the same way gold is. We must accept this as a chronic feature of the silver market. What has made the difference since 2000 is that the investor, like the Terminator, is back. Thus, in the midst of one of the strongest resource booms on record, investment interest could have easily absorbed surplus silver and asked for more, hence the recovery of the silver price and its upward trajectory.
We believe therefore that while the investment boom in commodities remains intact, the silver price is likely to remain high, in the $10-15/oz range. This is despite the fact that we believe that the silver market remains in surplus. In addition to the initial comments we note the following market features:
• Mine supply of silver is very largely dictated by the production of copper, lead and zinc and gold, the silver being recovered as a by-product to the other metals. This has always been the case and it is unlikely to change. The implication then is that silver mine production will remain largely insensitive to movements in the price of the metal.
• Industrial recycling computes at much higher levels than public domain data. Not only do we calculate that silver recovered from photographic waste is more efficient than previously thought, there is also recycling from electronic waste which is likely to become of increasing importance especially in the next three to four years. Our photographic recycling model is considered in more detail later in this report but of importance is the fact that the volumes of recycling from this sector will decline in future years in line with declining usage of traditional photographic film. This is the one supportive side-effect of the move to digital technology.
• There are a number of new industrial end uses of silver beginning to emerge as commercially viable. These in general are predicated on silver's place as an effective biocide, and its conductivity. Our supply/demand balance to 2007 computes demand from these sectors at either zero or nominal volumes - essentially just setting the scene as a precursor of things to come. We expect many of them to grow rapidly from this zero base line and therefore any forecasting ought to take them into account. While collectively they are unlikely to take over where photography has left off in volume terms, their potential contribution to supporting the silver supply/demand profile should not be underestimated. Most important is that these end uses are unlikely to become the source of large amounts of future recycling of the metal.
• The VM Group acknowledges that it does not have a complete data series for silver coinage (now a relatively small sector), producer hedge positions (also relatively low figures) and the most recent official disposals of silver. These figures, especially the official sales, are difficult to obtain with any complete reliability; but we will update our data as and when we obtain more accurate information. This lack of data does not materially affect our conclusions, since much of the official disposals of silver (for example from the USA) occurred during the 1960s and 1970s and sales since then have been at much reduced levels. The country of most importance now remains China, which officially came off the silver standard in 1935. Other official holders of silver also appear to be Russia and India. As far as we are aware, there is no public domain data covering the levels of official silver inventories in these countries.
Four people were particularly helpful with this exercise of modelling the silver market. We thank in the first instance Rosemary Van Musschenbroek and Robin Adams formerly associated with MMRS and Martin Fewings of Xstrata. But we also extend very grateful thanks to Tim Green who scoured his attic; a veritable Aladdin's Cave of silver data.
1 Ted Butler's www.gold-eagle.com
2 Compared with our total of 154,000 tonnes of gold. See The Yellow Book, October 2006 edition.
Jessica Cross and Gary Mead
VM Group
London, June 2007
