JT
The Golden Question
Is gold the next bubble? Hedge fund manager George Soros said so, and even legendary investor Warren Buffett thinks gold is over-valued. But are they right?
First, Mr. Soros recently clarified (and hedged) his comment that gold is the “ultimate bubble” by saying that it may go higher:
“Gold is the only actual bull market currently. It just made a new high yesterday. In the present circumstances that may continue,” Mr. Soros said in an interview at a Thomson Reuters Newsmaker event.
“It will be very interesting to see if there is a decline in the next few weeks because…everything that makes a new high almost immediately afterwards reverses and disappoints,” Mr. Soros said.
“I called gold the ultimate bubble which means it may go higher but it’s certainly not safe and it’s not going to last forever.”
Next, Mr. Buffett:
"You could take all the gold that's ever been mined, and it would fill a cube 67 feet in each direction. For what that's worth at current gold prices, you could buy all -- not some -- all of the farmland in the United States. Plus, you could buy 10 Exxon Mobils, plus have $1 trillion of walking-around money. Or you could have a big cube of metal. Which would you take? Which is going to produce more value?"
How does one argue with Mr. Buffett? First of all, there is this excellent rebuttal in Forbes:
http://www.forbes.com/2010/11/29/warren-buffett-dollar-gold-opinions-contributors-drew-mason.html
There is one point of his I would like to quote and emphasize:
“...History tells us the dollar will not survive, and gold will preserve wealth, yet Americans keep virtually all of their net worth in dollars...”
In my opinion, Mr. Buffett is comparing apples to oranges. Gold should be compared to currencies, not to income producing assets. Dennis Gartman, of the Gartman Letter, says that Gold has become a third reserve currency for those who have lost faith in the Dollar and the Euro.
I am more inclined towards this point of view. To simplify the argument, consider gold vs only one global currency (say, the US Dollar). Assume that all the dollars in circulation, say X dollars, could be used to buy the 67 cubic feet of gold that Buffett says exists. Then it must follow that if the amount of dollars in circulation, X, increases faster (ie. because of central bank printing of money) than the increase in the amount of gold in existence (ie, the amount that can be mined out of the ground), then the price of gold, in dollars, MUST INCREASE.
So is the amount of money in circulation really increasing faster than the rate that gold can be mined?
To answer this question, we need to go back to macroeconomics 101. If economics is not your thing, skip down to the graph. One look is all you need.
So some basics on monetary easing courtesy of Wikipedia:
“When a central bank is "easing", it triggers an increase in money supply by purchasing government securities on the open market thus increasing available funds for private banks to loan through fractional-reserve banking (the issue of new money through loans) and thus the amount of bank reserves and the monetary base rise. By purchasing government bonds (especially Treasury Bills), this bids up their prices, so that interest rates fall at the same time that the monetary base increases.
With "easy money," the central bank creates new bank reserves (in the US known as "federal funds"), which allow the banks lend more. These loans get spent, and the proceeds get deposited at other banks. Whatever is not required to be held as reserves is then lent out again, and through the "multiplying" effect of the fractional-reserve system, loans and bank deposits go up by many times the initial injection of reserves.”
According to US Federal Reserve Statistics, the US monetary base (M0) as of Oct 2010 was US$1.969 trillion. This might not mean much until you realize that just two years earlier (Oct 2008) it was US$1.182 tn, representing an increase of 67%! The reason for this increase is because the Fed had to inject excess liquidity into the system during the financial crisis in order to prevent a total collapse. So here it is visually, reproduced from the Economic Research webpage of the St. Louis Fed:
So what happened to M2 (the broadest measure of money supply) during this time? It didn’t increase nearly as much - from US$8.045 tn in Oct 2008 to US$ 8.774 tn in Oct 2010 - an increase of only 9%. See below:
This could be what Fed chairman Bernanke meant when he said on 60 minutes (Dec 5) that “money supply hasn’t increased in any significant way” - the monetary base has increased, but because banks are not lending, M2 hasn’t increased correspondingly. What I believe is happening is that this excess is flowing outside of the US to emerging markets in search of returns, and that is why the US is not yet seeing runaway inflation.
QE2
But the Fed didn’t stop there. You may have heard of QE2, or Quantitative Easing 2 - this is the Fed’s plan (announced in early Nov 2010) to purchase an additional US$600 bn worth of government bonds over the next 8 months. When and if it finishes, it could increase the monetary base by another 30% to US$2.6 tn.
So what are the risks to all of this monetary easing? Let’s see what Wikipedia says:
“Quantitative easing can trigger higher inflation than desired or even hyperinflation if it is improperly used, and too much money is created. It can fail if banks are still reluctant to lend money to small business and households in order to spur demand. Quantitative easing can effectively ease the process of deleveraging as it puts pressure on yields. But in the context of a global market, home printed money can flood abroad and spark asset bubbles in developing economies” [emphasis my own].
Therefore, all this monetary easing by the Fed will either cause runaway inflation (remember that the multiplier effect and a reserve requirement of 10% means that a $2.6 tn monetary base could result in M2 shooting up to $26 tn, or about 3X what it is now, in a worst case scenario) or it will result in ASSET BUBBLES - and not just in developing countries. One of the goals of QE2 is to target (raise) stock prices, but I believe that the biggest beneficiary of all this easy money are commodities given their scarcity and the relatively small size of their markets in comparison to stock, bond, and currency markets.
GETTING BACK TO GOLD - SUPPLY AND DEMAND
The first thing you should know (before I make my arguments) is that the gold market is already in shortage. Figures below from the World Gold Council.
Global production (mining) or supply of gold in 2009 was 2570 metric tons (production is very inelastic, meaning it reacts very slowly to increasing prices because new mines can take up to a decade before they start producing).
Meanwhile, global demand for gold was 3474 tons with the shortage of close to 900 tons made up of gold sales by central banks and recycling of existing gold.
Demand for gold is normally broken down into 3 main categories:
1) for jewelry. In 2009, demand was 1758 tons (51% of the 3474 tons)
2) for industrial use, demand was 373 tons (11%)
3) for investment, demand was 1342 tons (38%)
But thanks to Mr. Gartman, we can add a fourth type of demand for gold - as a reserve currency.
Let’s consider China, which has a whopping US$ 2.6 tn worth of forex reserves, but a measly 1054 tons of gold (as of Sept 2010, valued at only US$ 47bn using a price of $1400/troy ounce), meaning that its holdings of gold make up ONLY 1.8% of its forex reserves. In comparison, the world’s largest holder of gold reserves, the US, holds 8133 tons which comprise 72% of its forex reserves.
The below matrix illustrates how much gold (in tons) China will need to buy in order to diversify its forex reserves. The rows show the target ratio of gold/forex reserves, while the columns assume price increases of flat, a 50% increase to $2100, and a 100% increase to $2800.
% of Reserves Current Price (1400) 50% Gain (2100) 100% Gain (2800)
5% 2,888 1,925 1,444
10% 5,776 3,851 2,888
15% 8,665 5,776 4,332
20% 11,553 7,702 5,776
25% 14,441 9,627 7,221
30% 17,329 11,553 8,665
So, lets make a conservative assumption that China would like to raise its holdings of gold up very slightly from 1.8% of its reserves to 5%, and that its actions in the open market push the price up 50% to $2100/troy ounce - in this scenario, China would need to hold 1925 tons, so it would need to buy an additional 871 tons to add to its existing 1054 tons.
Remember that global production in 2009 was only 2570 tons, so China would need to buy 34% of annual global production just to get to 5%! All this buying would almost double the current annual global shortage. Furthermore, a 5% target is extremely low especially if you consider that US gold reserves are at 72% of its forex reserves.
Where will all this gold come from? 871 tons is more than 10% of US reserves. Do you think central banks around the world will be very eager to sell 871 tons of gold to China? Germany, the second largest holder of gold reserves in the world, only has 3407 tons, while the IMF (at number 3) only holds 2907 tons...
REASON NUMBER 2
The second argument for buying gold is made by Mr. Jim Cramer, host of CNBC’s Mad Money. He argues that currently gold represents only about 1% of the average investor’s portfolio. I would say that figure is fairly accurate. It would be higher for those with a lot of jewelry, and lower for those with no jewelry, but just looking at an average investor’s asset allocation of bonds, equity, and alternative investments, I think 1% might even be a little high. Cramer argues that you should buy and hold gold until it becomes around 5% of the average portfolio, so lets see how much gold needs to be bought collectively by fund managers to bring it up to 5%.
First of all, total assets under management globally by the asset management industry in 2009 was US$52 TRILLION. 5% of this comes out to US$2.6 tn. You see where this is going?
US$2.6 tn, at current prices, equals 57,765 tons of gold. Lets put this figure in perspective:
- At current annual global output of 2570 tons/year, 57,765 tons of gold equals 22 years of production!
- Total holdings of gold bullion by all the gold ETFs out there only amount to 2062 tons, so global ETFs would need to increase their holdings by 28X!
- Total gold reserves held by all the countries in the world is only 30,500 tons, which is only slightly more than half of the 57,765 tons the asset management industry needs to accumulate...
SO, IS GOLD REALLY A BUBBLE?
Not yet. We are likely in the middle of a bubble, but it probably isn’t done yet. You can check out this blog which was written this summer:
http://www.creditwritedowns.com/2010/06/time-gold-bubble.html
In particular, I would like to highlight the graph which I have taken the liberty to reproduce below:
The data for the current run in gold used in the graph ended in May of this year, but even if we extrapolate the yellow line from +300% (where it ends) to +500% to account for current prices, it is still nowhere near the +1000% run in the Nasdaq bubble or the the +1800% run in gold in the 1970’s.
So where are we in this current gold run-up? Take a look at another graph, from a 2004 blog entry, that breaks down the 1970’s bull run into 3 stages and superimposes the price action of gold as of 2004:
http://www.zealllc.com/2004/au3stage.htm
If you look carefully, you will realize this graph (again, it was made in 2004!) is scary in its accuracy because it implies gold will trade at $2000 by the end of 2010 (which would translate to the end of 1979 in the graph). In fact, the graph is only off by less than a year as $1400 on the X axis is the middle of what should be 2010!
FINALLY...BACK TO THE GOLD STANDARD? NO, BUT...
President Nixon took America went off the gold standard in 1971. Before then, each dollar was backed by gold. Now, each dollar is backed by the US government. But how much would gold be worth today if the US were to go back to the gold standard?
First of all, to be absolutely clear, I do NOT think the US should go back to the gold standard due to the deflationary implications of such a move. However, it would be very instructive to determine much gold should be worth (or how little the dollar is worth) should the US decide to go back.
Let’s see.
- US gold reserves are at 8133 tons.
- There are 32,150 troy ounces in a ton, therefore the US holds 261,475,950 troy ounces of gold.
- The US monetary base is about US$2 tn
- Therefore, if each US dollar is backed by US gold reserves, then one troy ounce of gold should be worth US$7649 (2 tn dollars divided by 261 million ounces) which is almost 5.5X what it is worth now....
So, given all these arguments, wouldn’t it be prudent to buy a little bit of gold just in case?
