Everybody take a deep breath.
I know all you read was the title and I can see half of you immediately queuing up your angry comments. What follows will probably cause the other half of you to get your angry comments ready as well.
So first let me set the stage at the risk of undermining everything I'm about to do:
Predicting the price of gold is sheer folly.
We all know this, right?
Gold (NYSEARCA:GLD) is the investment equivalent of your crazy ex-girlfriend. It does things that make zero sense. It moves in ways that are impossible to understand and have no basis in logic. The only pattern of behavior that can be counted on is that it'll always do what you least expect -- and nine times out of ten, this'll also be the thing that pisses you off most.
Gold may be insane, but like our crazy ex-girlfriends, there's a certain sex appeal that keeps us coming back again and again and making the same stupid mistakes over and over. None of gold's baggage seems to bother us. We put up with it and at times even enjoy it.
Perhaps we do it because when you take a step back, there is a little bit of sanity and certainty. Over the long run there is one very basic driver that profoundly affects the price of gold: inflation.
Fundamentally, this makes sense. As critics frequently and accurately point out, gold doesn't really do anything. It doesn't generate earnings. It doesn't grow or compound. It doesn't pay a dividend. It's just sits there and looks pretty. I know it has a long and colorful history, and I know that many others view it as something of a "neutral currency" . These are legitimate points and I'm not going to argue for or against those.
But as an asset, gold is just a shiny lump of stuff that we pull out of the ground and store in our safe deposit boxes or drape on our bodies.
With that, here's what we're going to do today:
- Take a look at the modern history of the gold price.
- Answer questions about whether it's expensive or cheap right now.
And here's what we're going to do with part two:
- Build a basic model that attempts to predict the future returns in gold.
- Create a practical long-term strategy for investing in gold that's hopefully a little better than holding a fixed allocation or buying and selling willy-nilly.
There's a lot to cover. This won't be your generic gold bug song, extolling the virtues of the yellow metal. Nor will it be the typical takedown on why gold stinks as an investment. I've read a lot of research and studied a lot about the gold market but I've never heard a story quite like this told in this way. So that's why I'm going to tell it. It'll be relevant if you own GLD, the iShares Gold Trust (NYSEARCA:IAU), the cash commodity, or one of the myriad linked investments.
Don't worry, there will be lots of pictures.
Prologue: Drivers & Data Sets
We all know and agree that in the broad stock market, very long-term returns are a function of a few simple factors. GDP growth & corporate earnings. Dividends. Inflation. There's an element of psychology (i.e. multiple expansion & contraction) over shorter cycles.
That's pretty much it. All the action in the stock market over the last century can be explained with those variables. I've never encountered anybody in the industry who disagrees with this.
So in a lot of ways, a long-term model for explaining the gold price is even easier. As a real asset, it's powered by inflation with shorter term supply/demand movement on the margins. And the good news is that we have a really great long-term data set on this, the Consumer Price Index.
Don't like the CPI? That's cool. But don't tell me about it. Take your beef to the really intelligent people over at the Bureau of Labor Statistics. Tell them you have more accurate data on this stuff than they do and that you know of a better way to do things. Shadow Stats Guy, if you're reading, send me your data and I'll run this with your numbers. I'm curious how it'll look.
There's a also slim possibility that you don't believe gold is or should be influenced by the rate of inflation and the way that affects the value of the Dollars in which gold is denominated. This, too, is cool. But let me tell you a quick story. It involves black magic and wizards and government conspiracies. It might just change your mind.
Chapter 1: Serendipitous Happenstance or Cold Calculation?
Our story begins way back in the year 1900. This is a less arbitrary beginning than you'd think. That wasn't just the first year of a new century and an easy entry point for calendar math. (Working with dates in Excel prior to 1900 is a pain in the ass and I have a secret theory that that's why you see so few charts go back before then. Analysts are a lazy lot.)
1900 was the year when the Gold Standard Act was passed. Unlike today's legislation, this one actually did what the title suggests. (zing!) President William McKinley signed it into law and, presto, the United States was on a gold standard. One dollar was fixed to "25 8/10 grains of gold at 90% purity." By extension, one ounce of gold was assigned a dollar value of $20.67.
That price would stand for over three decades.
The level of inflation, however, would not stand still for three decades. It would explode during World War I and then turn into a different sort of nightmare, systemic deflation, during the early years of the Great Depression. Under the specter of deflation, the idea of sustainable economic recovery seemed impossible.
By 1934, something had to be done and the wizards of the Roosevelt administration were just the people to do it.
So Congress passed the Gold Reserve Act of 1934. That was when it became illegal for the public to own gold. They were forced to sell it to the Treasury. On a side note, think about how crazy that is. Can you imagine something like that happening today? Let this be another lesson about the caution one must adopt when doing apples-to-apples comparisons of today's economic data to the data of yesteryear. Environments change. Dramatically.
But more importantly for our purposes, January of 1934 was when the nominal price of gold was reset from $20.67/oz to $35/oz. That was a dramatic devaluation of the Dollar under gold standard terms. The Treasury saw the value of their gold holdings increase by a tidy $50 billion overnight (inflation adjusted).
The country suddenly realized that these guys weren't messing around. After deflation of around -10% per year, the CPI turned around and started growing again. The Government was successful. Obviously, there were a lot of factors that helped slay the deflation dragon and save the U.S. economy -- this is what Ben Bernanke is an expert on -- but the point is that after this intervention, the price level started going up again.
That's where our first magical coincidence occurs.
Had the Gold Standard Act never been passed and had the price of gold simply been allowed to rise and fall in accordance with the CPI, the gold price would have landed on January 1, 1934 at $34.57.
Just to be clear, that blue line is a theoretical price of gold. That's what the price would be at any moment if you let it appreciate every month in accordance with the change in CPI. It's the CPI in dollar terms, indexed initially to $20.67. No black magic here.
Now, to be fair, under a gold standard, the rate of inflation and the value of the dollar relative to other goods and the price of gold are all intimately linked. These variables should all track over the long run.
And here's where I also must confess my lack of knowledge as a gold historian. I have no idea how the Treasury determined that $35 was the right price. Perhaps somebody back then ran this exact same study and concluded that, OK, here's a defensible case for adjusting the price of gold. It could have been difficult to defend setting the price any lower than what the CPI would have suggested. It may have had no choice under the gold standard. I apologize for my ignorance.
Ultimately, it doesn't matter whether it was magic or math.
What does matter is that the Powers that Be revalued gold to exactly where it should have been at that exact moment.
Chapter 2: It's My Gold And I'll Value It How I Please
I missed out on the period between Roosevelt and Nixon. While I can't give you a first hand account, my sense is that in the public space, gold was largely forgotten. It was just materiel for making jewelry. Gold was illegal to own privately as an investment. The Gummint held all the gold! They defended the $35 peg for a lot of years, buying from whomever and selling to approved central banks, intervening however and whenever to maintain the standard. The rest of us went about our lives.
But during the 1960's, this and other things in the country started to change. Everything else in the world had grown more expensive over the years. Wages. Cars. Orange juice. But not gold. Gold was still $35/oz. Gold had grown steadily more undervalued relative to assets with similar characteristics.
Things started changing culturally around gold, too. I'm talking about Ian Fleming's Goldfinger, of course. When we left the theater in 1964, we were hooked. Whoah, Ft. Knox! We all wanted our own gold! It was coming back into the public eye.
This is when the second magical coincidence occurs.
This is simply a continuation of that first chart. The fixed price of gold vs. the CPI.
Here's what happened in the years when the gold price finally started changing.
- 1968 -- The sterling price was abandoned in London and the Dollar became the final and formal basis for the gold price. At the same time, gold began to float freely in private markets, while only central banks could trade with the U.S. at the $35 peg.
- 1971 -- Nixon ends international convertability of the dollar to gold. The post-WWII Bretton Woods system breathes its last gasps.
- 1975 -- Citizens are free to own gold in any form without any restrictions.
The mid 70's were an interesting time. For the first time in nearly everyone's lives the market had to figure out for itself at which price gold ought to settle. And this time, the free market managed to carry the price almost immediately to where it should have been priced had gold simply appreciated over the years at the same rate as the CPI.
The market was free, though, so it wasn't without characteristic volatility. Gold overshot, then undershot -- magically and coincidentally oscillating perfectly around this historical mean -- before hitting the target on the nose again in 1977. Free markets may be messy, but more often than not they work.
After all this drama about The Inflation, psychology started getting out of hand. By the early 1980's the first real Gold Bubble would develop. It would be an equally entertaining future beyond that.
But 1977 wouldn't be the last time that gold and inflation would intersect.
Chapter 3: Putting the Free Market to the Test
When we think of gold, we think of this rich, thousand-year history. We think of it as being inseparable from our classic concepts of currency. But what we forget is that much of this is strictly a modern convention and a retroactive re-writing of the way that things actually were. For a large chunk of the U.S. history, gold existed outside the public eye. Nobody cared except the wonks. We had a gold standard, but the way most of us think about gold today would have seemed completely foreign to someone 50-75 years ago.
History isn't just about fact, it's about perception. It's about what happens, but also how we interpret its meaning. The way that perception and cultural narratives quietly shift over time is an occasionally-insidious factor that is perpetually under-appreciated in the present.
Since gold started trading freely on the open market -- and especially when it became possible to trade gold in the equity market through ETFs -- both our perception of it and its price behavior have changed dramatically. The long term trend is still the same, but we start to see more of that inevitable side effect of free markets: volatility.
Here's what it looks like when we reset our clock to 1973 and run it through the end of 2005.
Again, no black magic here. It's the market price of gold and the inflation rate.
If I came to you with this chart in 2005, you would have said that this all makes pretty good sense. You'd think that using the inflation rate as a basic long-term proxy about which the gold price ought to oscillate would be really clever. That's assuming I could have distracted you long enough from your condo flipping to listen for a minute or two. You would have conceded that, yeah, gold was probably too cheap in 2001 and too expensive in the early 80's, and that gold was probably fairly valued today.
Keep in mind that little blue trendline is still the exact same one that we started way back in 1900. It's as sensible a trend line as any.
Now, there are obviously a lot problems with gold in that it can overshoot this trend line dramatically. Like, by a factor of 3. Under this study, gold should have been worth a little more than $200 in 1980 but we all know what happened in the midst of that bubble. What we have forgotten to appreciate is how quickly (relatively) gold fell back to "fair value." By 1985 it was more or less there. It had another act left in it, but by the 1990's and nearly a century of drama and intrigue, it had found fair value.
Stocks overshoot and undershoot too. Equity prices have reached some equally crazy extremes in their history. Think about how rich valuations were in 1929 or 2000. Or how irrationally cheap stocks were in the early 80's. And these valuations can persist for a little while as well.
This simple model may seem to work really well over long windows of time, but never forget that all sorts of wild things can happen in the markets. It can get even worse with gold, a market that's tiny compared to the size of the equity, bond, and real estate markets.
Chapter 4: Cultural Changes
One more thing we've forgotten to appreciate is how remarkably the gold narrative changed after 2006. Or technically, after 2008. That's when a number of factors had finally come together for gold:
- Good momentum. Gold was in the middle of a cyclical bull trend, much of which was due to simple mean reversion.
- New ETFs had given a massive number of investors easy access to an asset that was historically difficult to access.
- There was a collapse in multiple asset classes and the emergent need for new destinations for capital aside from cash.
- Bailouts, "money printing", and the resulting hyper-inflation meme.
- Cultural backlash to the election of Barack Obama and formal & informal gold "marketing" from partisan opposition.
- Ron Paul
- Distrust in government institutions, particularly central banks.
- Doomsday prepping and dark fantasies about apocalyptic scenarios, financial and otherwise.
Gold wasn't just a thing anymore. It had become woven into the fabric of who we were as individuals. Or who some of us were, rather. These psychological changes plus simple price action helped re-write the way we thought about the value of gold. People who had never paid attention before started to.
So where does that put us today?
The idea of setting gold to appreciate in accordance with the inflation rate may have sounded absurd at first. And if I'd come to you with this story in 2011 when gold was approaching $2,000/oz, even the gold skeptics would have dismissed me outright.
But now that we've calmly walked through more of the history, it doesn't sound so crazy when narrated this way, does it?
If the last 5 years had never happened would we be so quick to reject such a story? I highly doubt it.
We can twist this perspective a bit to state it a different way: without the previous 100 years of history, the post-crisis run in the gold price would seem so aberrant as to be unconnected from that original driver at all.
Seriously, what do those two lines have to do with each other? But we know they are and should be intimately linked.
It's no wonder so much of the chatter about gold in recent years has to do with what could happen in the future. Future inflation. A future monetary collapse. This is incredibly revealing about our deeper psychology. It's as though we're subconsciously conceding the point that current prices have disconnected themselves from history and require specific future outcomes to justify themselves.
If there is no hyperinflation and monetary collapse, then what?
Unfortunately, we don't get to pick and choose which characteristics of gold are ancient. If we're willing to tell stories about that are thousands of years old about gold, we can't be so quick to ignore crucial aspects of its history that occurred within just the last century.
So let me now stitch the whole thing together for you log scale.
Still think the price of gold is going to $5,000/oz?
If I'd told you right off the bat that I thought an ounce of gold was worth $624, even the most bearish of you probably would have laughed me out of the room. Nobody would have heard me out.
But when you step through this history in a linear fashion without any of the historical re-interpretation of the post-2008 years, this hardly sounds crazy.
It might take a decade to get $624. In fact, it probably will take a decade to get to fair value if the past is any guide. These bigger trends take a long, long time.
Or maybe it never gets there at all and inflation simply catches up. Maybe in the year 2020 these lines finally meet again in a moonlight rendezvous at $1,000/oz. Maybe at that point the rest of us will have abandoned gold as a decrepit, uninteresting asset that only has the power to move sideways.
Y'know, sorta like how we felt in the 90's the last time we watched that movie.
Chapter 5: Caveats Aplenty
You don't need to tell me there are problems with this kind of basic study. First of all, it does nothing to account for supply and demand in the gold market. These are the factors that move the price of any asset up and down.
I'm not sure there's anyone on the planet that understands the true demand for gold and the way it intersects with supply. I mean, the supply is easy enough. But what is the aggregate demand for gold? I sure don't have a clue. Do you?
I suppose this too is implicit in the price. When gold gets too expensive, we sorta stay away. When it gets sufficiently cheap, demand eventually stirs. Between those poles are two factors: momentum and the inflation rate. One sets the basic long-term trajectory while the other drives the oscillation about it.
With an asset like gold that literally doesn't do anything except sit there, price influences demand as much or more than demand influences price. This isn't like the latest iPhone or a stick of deodorant where we can get a fairly accurate read on a level of demand that's largely independent of price. Whether or not we choose to buy or sell gold has as much to do with plain old price as anything.
Second, on the supply side, gold gets interesting. We produce six times as much gold per year today as we did in the year 1900 (with five times the global population). The amount of gold produced has also fluctuated over time and it too is influenced by the price. When gold prices are high, it makes sense to mine gold from expensive locations. This increases future supply. When prices are low, it doesn't make sense for businesses to spend more to mine an ounce of gold than they can sell it for. Supply slows.
In case you were wondering why gold tends to move in strange cycles, I think this is another contributing factor. As prices get higher and more gold gets out on the market, the demand invariably weakens relative to the supply and prices begin to fall. The converse is true when prices get too low. Gold is more vulnerable to long-term feedback loops than a lot of assets.
With an asset like crude oil, it makes sense to assume quantum leaps in its "fair value". The average cost to produce a barrel of crude oil is significantly more expensive than it was 20 years ago, much more than the corresponding increase in the inflation rate over that time. And it's with good reason, too. Like untapped gold reserves, today's crude oil supplies are in difficult, expensive places to reach. Crude can't trade at $25 for very long before the supply dries up and prices leap higher.
The kicker with gold is that we don't burn up what we harvest the way we do with oil. The oil we pumped in 1994 is long gone; we've gotta go back to the ground to get more. But all the gold that man has ever mined is, more or less, still with us. It's yet one more way why gold doesn't work the way other commodities do and it's why the fundamentals behind supply and demand matter less.
If every gold company in the world shut off their mines, there's no reason price should spike the way it would with consumable commodities. I mean, surely it would for short-term psychological reasons. But would it be justified and sustainable? I've got some gold. I'll happily trade it to you for Dollars at a price somewhere sufficiently higher than here. And I'm sure you'd trade it to someone else for a price sufficiently higher than that. And that gold won't ever go anywhere. Nothing on the supply side will ever prevent us from trading gold for Dollars and Dollars for gold. We can do this indefinitely.
If you think about it, we should have started shorting gold the moment those "we buy gold" signs became ubiquitous. Almost by definition, that was the turning point. That was when the feedback loop started bending back the other way.
Eventually, we'll reach a point where we will have dug up all the reserves. "Peak gold" may be an easy concept to understand but it's a very tricky thing to forecast. Right now, the largest gold miners all have an estimated 15-20 years worth of reserves in the ground based on current production rates.
Lastly, if you still think another 50% drop in the gold price should be dismissed outright, consider that these gold miner stocks are all trading as though that's actually where gold is headed. The equity market thinks gold is going to $600/oz.
Despite everything I just wrote, I'm not sure I agree with that. At least, I don't think we'll get to $624 overnight. I wrote about the gold miners extensively last month and even made them one of my Predictions for 2014. There's a short-term opportunity here with the gold stocks. The Gold Miners ETF (NYSEARCA:GDX) is already up 20% in 2014 and I totally lucked out with my timing and adding it at the beginning of the year to our Aggressive Portfolio. The Juniors (NYSEARCA:GDXJ) are up even more.
I realize that using inflation as the basis for a really big picture model for understanding the gold price has flaws. The real story is more complicated, but not as much as you'd think. When you get down to it, very few markets are as simple as gold. There aren't many macro inputs. Paradoxically, this could actually be one of the reasons why we have such a hard time understanding what it does. Price can be more easily whipsawed by fundamentally unimportant and misleading narratives.
Barring massive exogenous shifts in supply or demand -- which neither you nor I can confidently predict or model -- an asset like this should over the long run increase in value in basic accordance with the amount by which the dollars denominating it decrease in value. The inflation rate is as sensible a trendline as we can responsibly draw for something like gold.
While the post-2008 gold narrative may indeed be weakening alongside the price, some of these relatively recent ideas are still strong in our collective psychology. It will take a while for us to fully divorce ourselves from these and reach a point where we think about gold rationally, as a society, once again.