In the investment world, it seems that every few months or years there is a new phrase that enters the lexicon. During the tech bubble, “eyeballs” and “clicks” quickly turned into worries about “cash burn”. The mortgage crisis introduced non fixed-income personnel to the phrase “money good” which often wasn’t so due to “liar loans”. A bull market in commodities led to discussions of “peak oil”, and ending the last recession seemed to hinge on whether we’d ever see any “green shoots”.

Of late, there has been much consternation about the US “monetizing its debt” and “devaluing its currency”, both of which tend to lead into the topic of gold. Yet gold, unlike some of the other topics that have come and gone, is in the midst of a 10 plus year bull market, leading some to wonder if we are in or entering a “gold bubble.” At State Street Global Advisors (SSgA), the Investment Solutions Group (ISG) team has placed a significant amount of resources into understanding gold, and this paper will discuss some of our beliefs. We will do so by focusing on three questions we’ve heard most often from our clients:

1. How do you “value” gold?

2. Are we in a gold bubble?

3. How much gold should I own?

HOW DO YOU “VALUE” GOLD?

In the classic fashion of a Warren Buffett type “thought experiment”, let’s say you were locked in a room by a benevolent captor. Once a year, your captor asks you to guess the price of gold. If you come within $100 of the actual price, he will let you go. If not, you stay another year. He will provide you any economic information you like—that is interest rates, inflation, exchange rates, money supply, debt/GDP, you name it. He will also provide a statistician who can develop statistical models that will incorporate this data in any way you see fit.

But here’s the catch. You are never given any price information for any security or asset class. You never know what the price of oil, gold, the Dow Jones or any other asset is. Let’s say you began your “lock up” in the year 2000. How many years would you have spent in the room? Would you have guessed $1,600/oz. in 2011? What will you guess in 2012? Would you have ever come within even $500/oz. of the actual price?

The point here, of course, is whether having any data or statistics actually helps you “value” gold. How would you even remotely connect the price of gold to any live piece of economic data? Our belief is that for the most part, you can’t. When an analyst or expert calls out their next gold target on CNBC, and says “$2,000 by year end”, it

is likely they have an embedded bias which is the current price and trading history. Well you say, they are also just guessing when they say their Dow target is 13,000 by year end. True enough, except that stock market guesses can be grounded in a valuation approach. Take a guess as to earnings, and a guess as to the multiple investors will pay for those earnings, and you have a stock market guess. Can’t do that with gold. An expert who says $2,000/oz. is his target likely says that because he knows the price is currently $1,650/oz. The next time someone gives you a gold target, give him the “benevolent captor” routine and see what their answer is.

So then what? We have an asset class that provides no earnings, no cash flow, no dividends or interest and no prospects for any, ever (note that from a classical modeling perspective, this presents quite a challenge to the statistician trapped in the room with you). In fact, there is a cost to holding gold, and that is storage and security. So not only does this asset not produce any cash flow, but effectively you must pay to own it and secure it. So, how do you value gold?

Across the industry, there is any number of proposed valuation constructs for gold. Here are three US focused constructs that have received some attention in the marketplace:

1. The inflation adjusted price of gold.

2. The value of the total gold stock relative to the US monetary base.

3. The value of the total gold stock relative to US equity market cap.

THE INFLATION ADJUSTED ARGUMENT

There is one number that arises quite consistently when investors discuss a gold target price and that is around $2,000 an oz. About $2,000/oz. is approximately the inflation adjusted high in gold that was set if you take the nominal high from January 1980 of about $850/oz. and inflate it to today.

So the logic here is that since 1979, gold has not kept up with inflation. Analysts, therefore, posit that $2,000/oz. should become a reasonable target. Now this begs the question—did the 1980 high in gold keep pace with prior inflation—i.e., is that number perhaps “frontloaded?” And the answer is yes. Inflation (ex food and energy) was up about 90% during the 1970s, yet gold was up around 14 fold. So on that point, it may be fair to say that those who focus on this reference point should understand what it had already priced in. Therefore, the key question is:

Is there any reason gold should keep pace with inflation?

The case for gold keeping pace with inflation isn’t based so much on why gold should have more value on its own merits in a highly inflationary world, as it is that paper assets will lose precipitously more value in a highly inflationary world. High inflation will make paper dollars less valuable every day while “hard assets” such as metals, land, energy and agriculture, may not see this effect precisely because they are hard and can’t be created, conjured or otherwise diluted. What’s interesting is that gold’s use as an inflation hedge is one sided. It’s not because gold has more value from inflation. It’s because it won’t lose value. Ironically investors then bid up an asset because it will “lose less”, and we end up with an asset whose price rises purely because it’s defensive.

But we still haven’t answered the question, should gold keep pace with inflation? Let’s start with gold’s utility. As illustrated in Figure 1, the combined weights for jewelry and industrial demand have averaged over 70% of demand for the past 5 years, a significant factor.

FIGURE 1: GOLD’S TRAILING 5-YEAR AVERAGE DEMAND

Source: GFMS/World Gold Council, Third Quarter 2011.

One could make the case that both jewelry and industrial use should reflect inflation increases. For example, consumers of jewelry, who have seen their income rise at the rate of inflation for say the past 30 years, should be willing to purchase jewelry at some inflated rate over that time. After all, most if not all of their other expenses have been rising with inflation—why shouldn’t gold? Similarly, industrial users should react the same way. They have cost and revenues rising with inflation, why should gold be an exception that doesn’t keep pace?

Producers too, face inflation, be it labor costs or the fact that gold exploration has, by most accounts, become more difficult since 2000 (see Figure 2). When one considers this chart, in light of an ever rising price—up nearly five times these last ten years, then to have mine production virtually unchanged is quite a statement on the difficulties involved and therefore, the likely higher costs involved as well. One would have expected that world production would have grown significantly given the nearly five-fold increase in the price of gold during the past decade.

Why would producers incur inflationary costs if they didn’t believe they could recoup them and more through higher prices?

Now one could say, that’s a fair argument if the asset you are describing has utility. Oil for example, is a key to modern commerce, and there are no replacements ready on a moment’s notice. Conversely, if all the gold in the world disappeared tomorrow, the jewelers would find a substitute, the industrial manufactures would switch their metals and the capital markets would likely focus on silver or platinum as the metal du jour. But gold’s “utility” is where we need to have, as they say in the movies, “willing suspension of disbelief.” Because while we can debate whether it actually has utility or should be valuable in the human psyche, the bottom line is, it is. And therefore, the production and demand issues around inflation must not be dismissed as “yes, but it’s not needed”. Clearly it’s not actually needed, but that doesn’t matter.

Once we accept this critical fact, our conclusion becomes clear: gold’s price should minimally reflect long-term inflation trends. It’s worth noting that the record shows this has been the case in the last half of the 20th century. Going back to 1957, we see US inflation averaging around 4% and Gold returns around 7%. However, much of these excess returns occurred after President Richard Nixon closed the gold window and took the US off the $35 an ounce exchange gold standard in 1971.

There is also, of course, the 10,000 pound gorilla in the room concerning the question as to whether gold should keep pace with inflation. The gorilla is gold’s value as a medium of exchange. Recent history, especially for US investors, has clearly illustrated what even a developed nation will do to its currency. While papers and books are written on this topic, suffice is to say that the US is literally printing money today to buy US treasuries and force “liquidity” into the system. As this printing occurs, existing dollars become worth less, and inflation becomes more likely. Gold, then, sees its (relatively) fixed supply increase in value. No wonder the calls for a gold standard have returned. While unlikely, the point is well taken. In the absence of any monetary or other authority, gold is sensible as a form of money because it cannot be created. This concept further supports the idea that gold should at least keep pace with inflation.

So, that said, does the inflation adjusted high of 1980 matter as a valuation tool? On its own merits, we don’t think it does, for as we’ve noted previously, prior gold highs had frontloaded inflation and over the past 60 years or so, gold has beaten US inflation. But, that is not the end of the debate. Because with gold, as we will discuss later, observing other investor’s attitudes is critical, and if one of those observables is a belief that gold should return to its 1980 inflation adjusted high, that can be, as George Soros calls it, a “reflexive argument” which could come true.

GOLD STOCK RELATIVE TO THE MONETARY BASE

This valuation method attempts to analyze the US monetary base relative to the value of gold held by the US Government. The rationale is that at extremes, gold will sell at some multiple of the monetary base. Using past extremes, this multiple is said to have peaked at over one times the monetary base. Where would that put gold today? According to the World Gold Council, the total amount of gold mined in the history of civilization is 170,000 tons. At the recent highs of around $1,650/oz., this means the value of all the gold in the world is around $8.8 trillion. Of this, the US government is reported to own around 261 million ounces, worth today about $432 billion. The US monetary base is around $2.7 trillion. So, at a multiple of one times, this would mean an increase of $2.44 trillion to the value of gold held by the US. The math on that works out to be a price of over $10,000 per ounce, a huge increase from today’s levels.

However, there are two immediate challenges that arise with this methodology. First, one has to ask why this matters in isolation; why is it only relevant what the US holdings of gold are relative to the US monetary base? Second, the historic level gold has traded at in the past has little relevance to today. It’s akin to some of our value investor friends who will only invest when they see the market get to a 6 price/ earnings ratio again. Maybe it will happen and maybe it won’t, but in the interim, you could be sitting for decades with no return while the market ignores your reference point bias. It’s worth noting this works in both directions. It may be that in the future, gold exceeds its prior levels relative to the monetary base. Our point is that being anchored on a reference point either way is likely a weakness in the investment process. Like any good model, the question should always be asked: and why will this occur again in the future? On this we don’t have a strong answer other than “because it did in the past,” which is really no answer at all.

GOLD STOCK RELATIVE TO EQUITY MARKET CAP

We have seen some industry work in the marketplace (verified by State Street Global Advisors) which analyzes the value of gold relative to the value of the US equities, via the Wilshire 5000. This work illustrates that gold has previously peaked at 160% of the value of the Wilshire, while today it trades around 60% of it. Again, while this is interesting and provides a bull case for a higher target price, the question is whether this prior peak should have any significance. This analysis may also provide guidance on where the next peak may be, but not anything on the current worth or when we could get back to that peak. Is there an economic reason why gold should move toward a previously attained ratio versus equities? We cannot think of any. Therefore, we cannot validate a reason why “this should occur again in the future.”

Again, we ask, how do you value gold? Our answer, to be direct, is: we don’t. Not to be trite about it, but it is worth what others are willing to pay for it. The value of gold is a lot like the value of fine art or your house. We, therefore, feel that our efforts are much better spent on understanding the mood of the “others”, and we do this by monitoring what we call “observables”. Each month our team produces an internal research note which consists of our assessment of the key observables that we think drive the market’s mood on gold, using both fundamental and technical factors. While there are approximately three dozen factors in all, a sampling of those factors can be seen in Figure 3. We believe this to be a better approach to gold investing versus constructing questionable valuation models or fixating on potentially spurious prior relationships at their highs. Thus far, it has been a successful approach.

FIGURE 3: SAMPLING OF GOLD “OBSERVABLES”

FUNDAMENTAL ANALYSIS

Source: SSgA. For illustrative purposes only.

ARE WE IN A GOLD BUBBLE?

Related to this question would be: “If not now, how will we know when?” Supreme Court Justice Potter Stewart famously once said of hard-core pornography: “I know it when I see it”. Is that possible in a bubble or only with the benefit of hindsight? It’s reputed that only a small number of professional investors actively shorted the mortgage market going into the crisis. Now, of course, it’s clear that we had a housing bubble. What does a bubble feel like? Well, having the displeasure of living through both the tech bubble and housing bubble, we have some recent insight on this. Here are three of the observables we measure with respect to “bubble territory”:

These are subjective and behavioral observations to be sure. But when thinking back on other bubbles, they would probably have helped frame the debate. You could have looked at tech stocks’ spiraling price/earnings ratios or annual housing appreciation, but in retrospect, you didn’t need to, as there was a demonstrable behavioral change. In gold’s case, there is no template for a P/E or appreciation versus a “norm”. Therefore, the “observables” become critical. Taking these one by one:

1. WE DO NOT YET BELIEVE GOLD IS WIDELY OWNED.

It is estimated by the World Gold Council in its report, The Strategic Case for Gold: January 2011, that US pension plans’ gold investments comprise less than 0.30% of their entire holdings. We would doubt individual investors are much different. How do we know this? Anecdotally. As the quote, “The plural of anecdote is not data”, is something of a creed in our group, we don’t say this lightly. Having run a number of advisor and retail client meetings in the past year, we feel that our experiences from these meetings are quite useful. At each meeting, we stop and ask how many of the crowd personally owns an investment in gold? The highest percentage of people saying they

owned gold was 20%. This continues to be a positive observation. When the gold cycle ends, it will end with everyone owning gold. We are not there yet. It’s true that ETFs have seen tremendous asset growth, but it appears that the ownership has been very concentrated in, for instance, hedge funds such as John Paulson and George Soros’. To us, “Main Street” does not yet seem fully engaged.

2. WE STILL BELIEVE GOLD IS “DIFFICULT TO OWN”.

We frequently hear from clients that “it’s already had a great 10 year move, it’s speculative, and how much upside can be left?” Similarly, while gold has received a significant increase in headlines, there is no shortage of skeptics—one need only google “Warren Buffett on Gold” to see the oracle’s consistently negative view. Buffett, the greatest investor of our time, is, therefore, the most famous critic of gold. He points out that it has no earnings, yield or way to return cash to the investor. We agree. However, that doesn’t mean one can’t make money in gold. While he won’t own gold, he also never owned Apple (up around 1,500% since January of 2000) or Google (up 530% since August of 2004) or shorted subprime mortgages (nor have we).¹ So while Berkshire Hathaway has gone up a very respectable 105% since January of 2000,2 Gold has increased nearly five-fold during the same period. It strikes us that Buffett believes that only asset classes and investments that fit his specific investment beliefs can be sensible investments. We, however, are agnostic about how to beat the market. We just want to do it.

3. NO PROLIFERATION OF ASSETS.

The ETF space is dominated by just two gold ETFs, and there currently are just a small number of gold mutual funds. It is true that “gold ATM” machines are starting to gain some attention, and the increase in “We buy Gold” advertisements cannot be denied. However, from an investable perspective, we have not seen anywhere near the proliferation of products in comparison with the explosion of tech related products which arrived on the market in the late 1990s and early 2000s.

Lastly, in conjunction with Ned Davis Research, we took a look at some other “bubbles” to see where gold currently stands in terms of duration and magnitude. As can be seen in Figure 4, while gold has had a strong 10-year run, it still pales in comparison to other “bubbles”, with the prior gold run reaching a 1,836% return, the Nasdaq 950%, and Uranium 1,171%. Of course gold isn’t required to follow any of these patterns, but it’s important to know that “bubbles” can reach four digit returns over a ten year period, and gold has only achieved just over half of that move.

THE END OF GOLD

Our view, therefore, is that we are not yet in a “bubble” and it’s worth noting that once you are in a bubble, for a time, there is still money to be made as illustrated in Figure 4. Some day, however, we could have a true gold bubble on our hands. Note the use of the word “could”. Some asset classes see bull markets and subsequent crashes, but their bull market phase may never have truly qualified as a “bubble”. Look at the housing or banking stocks climb prior to their precipitous drop in 2007/08.

But if we did enter a gold “bubble”, how bad could the aftermath be? We think it could be monumentally bad. Consider that gold is both a single security and almost an entire asset class on its own (no offense to silver, platinum and palladium, but gold is their big brother). When the tech bubble happened, you could have moved to financials and kept your equity weight in line, or even within tech, you could have underweighted web companies and overweighted hardware, semiconductors or software. The point is that there were numerous ways in which an investor (with good hindsight of course) could have fled the troubled securities yet still had similar equity exposure. In a way, this prevented the web bubble from being even worse.

Gold has none of this in its favor. If investors decide to flee gold as an investment, then gold the asset class will be crushed. There are no offshoots that investors can transition to which will cushion the asset class fall. Furthermore, without any valuation support to cling to, investors may have a hard time rationalizing holding on if a freefall begins. The end of gold could be quicker and more severe than the technology or real estate bubbles. Note from the right hand side of Figure 4 how hard that fall can be.

SO, HOW MUCH GOLD SHOULD I HOLD IN MY PORTFOLIO?

Before it’s all said and done, we are likely to see some significant advances in products and portfolio protection involving gold. One can envision a day in which your brokerage account may be translated into gold instead of US dollars. Notable hedge fund managers such as John Paulson have reportedly done exactly this by offering share classes of their hedge funds denominated in gold. It’s been reported that Paulson’s personal holdings are almost entirely in this share class. That’s how much he dislikes holding dollars.

So let’s try to answer the question: how much gold should you own? If a portfolio tool existed that allowed you to convert some or all of your portfolio into gold’s return instead of dollar returns, and this was a costless transaction, how much of your portfolio would you hedge to gold? Ten, twenty, fifty or one hundred percent? Whatever your answer is to this question is probably as good a start as any in terms of setting your actual gold allocation today. In looking at the industry for help, precise data on actual gold allocations is difficult to come by. We have seen estimates, such as the World Gold Council’s The Strategic Case for Gold, January 2010, that gold comprises just 1% of the world’s total global assets, and our experience with both Institutional and advisory based clients would support this. If anything, 1% might be on the high side of the typical portfolio’s actual ownership. From a practical standpoint, when working with an institutional client we must also consider investment policy constraints and tracking error goals. In our client portfolios which are tactically able to hold gold, we generally hold a position in the low- to mid-single digits.

A related question is what should the funding source for gold be? Fixed income or equities? We have viewed gold as a “risky asset”. Indeed, its volatility indicates as much, with gold’s volatility nearly the same as the S&P 500^®’s at around 20 for the past three years. Additionally, with no “principal value” to return, it looks more like a non-dividend paying equity than a fixed income instrument. However, in a true inflationary spiral, one would probably like gold to offset some or all of their fixed income portfolio. Given this split personality, we believe in dynamically funding our gold positions from both the equity and fixed income portfolios, with our other tactical views dictating which and how much.

On the topic of tactical asset allocation, we believe gold can have a very useful role in a portfolio both within a strategic allocation to commodities and also as a separate allocation which can be utilized as both a tactical inflation/deflation play or an “event risk” play. We are often asked, how can gold be both an inflationary and a deflationary hedge? The answer is, it can’t in the long run. What’s happening today, though, is not the long run. What’s happening today is that the rise in the gold price is reflecting the fact that today’s deflation will spur tomorrow’s inflation, due to the mechanism called the US Government. As the Government will seemingly do everything in its power to fend off deflation in the short term, it is, therefore, likely to be creating more inflation and currency debasement in the longer term.

From an event risk perspective, while moving to cash is always an option, gold can actually prosper whereas cash will be unmoved. Saving capital by moving out of a risky asset to a risk-free one can be a powerful way to preserve capital, but making money by moving out of a risky asset into an asset that provides a positive return is a result that builds capital.

CONCLUSION

Gold is clearly a unique asset class. It has caused brilliant investors like Buffett and Soros to be on exactly opposite sides of the argument. It is likely, therefore, that it requires a unique investment approach. Traditional valuation approaches seem ill equipped to be effective because they lack any true linkage to a cash flow. We are not bothered by the fact that gold’s investment characteristics don’t fit a predefined box. Rather than a dogmatic approach that must rest on valuation, our approach is dynamic and grounded in “observables” that we think drive the market. We do not believe gold is yet in a “bubble”, but do recognize the potential for that to happen and the potential downside if it does. We think a rational approach to “how much should I own” is to think about how much of your currency exposure you would ideally like to hedge to gold in a costless transaction. We also think gold can have a special role in tactical portfolios as both an inflation hedging vehicle and a flight to safety vehicle. Therefore, as of this writing, the signs that we see are still long term bullish for gold, and we will continue to tactically own it when we believe it’s beneficial to do so.