The Popping of the Silver Bubble: Still Too Early to Buy

Silver has been on a tear for the past year.  If you haven't followed it, SLV, the underlying ETF, has traded from $19.50 an oz (or per share) as of just last August, to a high of $49.50 an oz just last Friday.  Then, over the weekend, the bubble popped.  In a mere four trading days this week, silver fell an astonishing 32%! Crushed.  Today its rebounding a little and trading around $34 an oz.  I find this subject endlessly fascinating, and apologize if this gets too long, but there is a conclusion you can skip to at the bottom.

HISTORY & THE HUNT BROTHERS
I admit to following the silver market longer than any other market.  Being a kid in Texas in the early 1980s, I still remember when the Hunt brothers tried to corner the silver market.  In 1979, the Hunt brothers, Herbert and Bunker (names of kids you would beat up in elementary school), began buying massive amounts of silver futures.  They of course only needed a certain margin level to enter into the contracts (ie they levered to the hilt).  Eventually they controlled, via futures and their own physical holdings, almost 1/3 of the entire world's deliverable supply of silver, some 200mm ounces. 

Silver traded between $3 and 4 an ounce for most of the 1970s.  However, by fall 1979, it had reached $11 an ounce.  Markets were in a frenzy.  Then, over the next few months, continued Hunt buying in conjunction with a Middle Eastern partner, drove silver to over $50 an ounce!  Sometime during this ordeal, my older brother gleefully doled out his life savings ($750, he was in high school), and loaded up on silver coins.  I think he paid $40 an ounce. 

It was very shortly after when disaster struck:  the COMEX,  the futures exchange governing silver, changed the rules on silver margin requirements, forcing the Hunt brothers to stop buying.  When they stopped, silver began to fall, and fall hard.  When the Hunts received a $100mm margin call that they could not meet, silver  went into a free fall.  In its darkest day, silver fell by 50% on March 20, 1980.  Nobody wanted anything to do with silver for a long time.

After the collapse, my brother never forgave himself, and silver continued to fall for the next 20 years.  The Hunt brothers eventually went bankruptcy in 1988, losing some $1.7BB on their silver fiasco.  Today in fact,  Bunker Hunt lives just a few houses down from my parents in Dallas.  (Its nice, but quite modest for someone who was at one time perhaps the wealthiest person on the planet).  Personally I owned some physical silver starting in the late 1980s, and really loading up in the late 1990s in the $5-7 range.  I missed all the Internet millions, but couldn't resist buying silver eagles.

As an aside, I highly recommend the book, Bryan Burrough's The Big Rich, which details the Hunts, the Murchisons, the Basses (via Sid Richardson), and the Cullen's family histories and how they came to their wealth.  Hint, starts with O, ends in L.  Great stuff.

SILVER SUPPLY & DEMAND
First of all, its not exactly clear, but a variety of estimates seem to converge on one level of total silver in the world today:  45BB ounces.  That is, the world to date has produced 45BB ounces of silver give or take.  What is more interesting however, is to examine silver supply and demand over the past 10 years.  New production has been remarkably stable throughout the 2000s, averaging a steady 900mm ounces per year of new mined silver.  That means we add about 2.5% to the world's supply of silver every year. 

The general consensus on silver vs gold is the following:  Gold has very little in the way of industrial uses, 50% of its production goes into jewelry, and the rest into world central banks and people's closets.  Silver on the other hand, actually has numerous industrial uses.  Of the 900mm or so ounces of silver mined per year, 700-800 gets used up every year in industrial applications, photographic applications, and in jewelry & silverware.  However, last year silver took a big upturn in turns of total supply.  While the world produced 920mm oz of silver in 2009, it jumped to 1,060mm ounces in 2010, up 15%.  The net increase was naturally due to the price increases that we saw.  More importantly, the net incremental demand came entirely from investment silver buyers. More on this in a sec.

Now, one problem I have with silver is that its demand is declining in 2 key categories:  Photography and Silverware.  Clearly digital photography has impacted silver demand, so much so that in the past 10 years, annual demand has fallen from 214mm oz to a mere 73mm oz in 2010.  Same with silverware, with demand falling from 105mm ounces to 50mm ounces last year.  People just don't buy silver wedding forks like they used to.  That decline, combined nearly 200mm ounces per year, is a LOT of silver relative to supply of 1060mm ounces per year.

To make up for these losses in demand, the implied net investment demand, has risen dramatically in the past 2 years.  The average annual investment demand was a mere 22mm ounces from 2000 to 2008.  Then in 2009, I assume with QE1, investment demand jumped to 120mm ounces, up sixfold!  Then in 2010 it was 178mm ounces!  I have no way of knowing what this number will look like over the next 5 years, but based on trends for supply and demand, we will need to see investment demand AVERAGE over 195mm ounces per year, just to absorb new mined supply! 

This demand can only come from new investment buyers.  I honestly don't know if investment demand goes to 300mm ounces, or back to 22mm ounces per year.  But HIGHER levels than 2010 are required now just to equalize supply & demand, and at significantly higher prices of silver.  I doubt how much much higher demand can go. 

Looked at another way, there are 425mm total ounces of physical silver via the 3 traded ETFs in the market.  That number in 2 years has to almost double just to absorb new mining supplies.  To put that in dollar terms, that implies $12BB in new buyers of silver (at 35/oz) over 2 years.  Compare that the 2000-2008 time frame.  Cumulative silver investment demand was a mere $1.7BB over that 8 year period.  That is a massive increase.

CYCLICALITY
While the industrial applications category makes up about 50% of annual mined silver, it should be noted that this is heavily cyclical.   Silver goes into hundreds of devices, including electrical components, catalytic converters in cars, etc.  So in 2009, demand was especially hit hard during the recession.  (demand fell 19%, from 493mm oz to 404mm in 2009).  So, I think its important to realize that silver prices will likely be fairly correlated to the economic cycle, and cyclical stocks.  The time to buy clearly is at the end of recessions, as likely it will get hit hard during a downturn.  Silver has historically been more volatile than gold, I suspect this will continue.  The problem with volatility is, when prices fall by 32%, people start to question whether they want to hold too much of it.

Most of all though, I think you want a silver investment thesis that doesn't depend on investment demand, but rather one that depends on increasing industrial demand, which is far more predictable.

THE GOLD-SILVER RATIO

To me, perhaps the best measure of whether or not silver is cheap is to look at the long term ratio of gold to silver prices.  The chart below shows 1971-2009 what this ratio looks like.  (prior periods were not terribly relevant given the Bretton Woods partial gold standard for the dollar).

If you drew a trend line on the chart, I think you would find that the average has been around 50.  That is, gold typically trades 50x higher than silver.  When gold is $1000/oz, silver historically has traded around $20 an ounce.  You can see the silver price-spikes in the late 1979 time frame, when the Hunt brother tried to corner the market.

It wasn't long ago at all, however, 2009 to be exact, when the ratio got to as high as 86.  That is, gold got pretty expensive, and silver was a great buy.  What this chart fails to show however, is the spike in silver in just the last 18 months.  The ratio today, even after a 32% fall, is an astonishingly low level of 41.  That is still 20% below the historical average, meaning that silver is still perhaps 20% overvalued.  History suggests buying silver when it hits a 65 or higher ratio, which today means I don't touch it until its around $22 an ounce.  (I think a long GLD, short SLV trade probably is a good trade, but be careful).

SILVER COSTS
I tried in vain to figure out what it costs to mine silver on a per ounce basis.  However, most silver mined is mined in conjunction with many other minerals, its hard to gauge.  Generally, though, I found that anywhere from $6-12 per ounce in cash costs are what a miner pays to dig more silver out of the ground.  Clearly, when Silver got to be above $15 in 2010, producers jacked up supply by quite a bit.  There is a spigot here, and it got turned on full blast last year and likely this year too. 

CONCLUSION
Don't touch silver now.  I laughed on Tuesday when a guy on Seeking Alpha wrote an article saying he was buying more (at 42), and expected it to go to 62.  You can see my comment there.  Even today watching SLV, it initially rose almost 5% this morning, but has succumbed to more pressure.  Its sits right now at $34, up less than 1%.  I call this a dead cat bounce.  (ie, even a dead cat will bounce a little).  I will probably wait to buy more (physical or SLV, not sure) at $22 or lower.  Fair value is probably the historical gold-silver ratio, which at 50 ratio implies around $29/oz.  I don't know silver falls below $30, we might get one more run back to $40, but I think the bubble has popped, and we are due to continue correcting, or at least languishing where we are.