Stocks vs Bonds vs Gold Returns for the Past 200 Years

Joshua,

Have you read "More Money than God" by Sebastian Mallaby. I'm about 25% of the way through the book and have really enjoyed the analysis. It has a lot of random walks vs. 'family of coin flipper' arguments in regards to hedge fund performance. It also articulates that each wildly successful hedge fund has had some type of advantage, be it quant, loose inside knowledge, or the idea of hedging (simultaneous long/short) in general.The author is pro hedge fund, and I'm curious to see if he makes any recommendations later in the book after his walk through the historical successes and failures of hedge funds and the characters behind them.Great article. I read from a viewer on some site, somewhere, that "All of you guys are suckers, I'm putting 100% of my savings in Gold, in 20 years we'll see who wins." I didn't feel like wasting my keystrokes to respond. If I had the gonads to short, I think I would short in the near future. I don't follow gold at all, but when I see people in the streets with signs to "Sell your gold" and magazines saturated with "GOLD INVESTMENTS" I think the market is bound for a correction. Then again, it seems the government has no plans in slowing down it's printing of money.Hope all has been well,Austin

I have not read it but I am about 80% certain it is sitting my library. I think it was part of my mass book buying 2 years ago when I spent $2,000 or $3,000 at Barnes & Noble and had it all shipped to my home. I'm working my way through John Bogle's body of work now but I imagine I will get to it before long.

I'm not a big fan of shorting just because there is no end to man's folly. There is no reason a stock priced at 2x intrinsic value can't go to 5x intrinsic value in the short-term before the greater fool chain collapses. With theoretically unlimited losses, I wouldn't want to be a part of it. The only way I would be open to doing a quasi-short was in my speculation fund if I bought puts on highly overvalued companies hoping for a crash so I could acquire the sinking stock, exercise the put, and pocket the difference. That way, my maximum loss could be calculated from the very beginning unlike the unlimited loss that could happen in a shorted stock. Just look at the Northern Pacific short squeeze of 1901. That is an Armageddon scenario for a short seller.

Great post, we all can see the 'gold bubble' before our very eyes but the lemmings will continue to buy, even at this unsustainable price. 

All you guys look pretty silly now in a world of ZIRP forever policies! 

Stop it with your zero hedge crap Mike Templar. By the way, would you like an ATM machine to go with your ZIRP policies?

Pretty good stuff. You wrote this six years ago. Let's see, 6 years ago gold was $1654. Today it is $1227. At the same time the DOW was in the 12,xxx range, since up by more than 50%--and of course during this time it also paid dividends.

Gold is a horrendously terrible long term investment. It is only nice to have because it is pretty, nothing more.

Great stuff fo sho playa

What's the source for your 200 years of stock data?

A majority of the long-term stock data was taken from the academic gold standard, Ibbotson & Associates Yearbook, which covers longer periods of time, but really strips out things like survival bias, etc all the way back through 1926. It's a wonderful resources that breaks down various portfolio mixes, asest class returns, inflation ranges, Treasury bond yields, etc. It's meant for professional investors and academia but I think everyone who is serious about money should have a copy in their library.

To use gold in any comparison before 1971 when Nixon severed all ties to gold would be completely rediculous. The price of gold was set by the gov't.,,,was the price of stocks and bonds set by the gov't? Interesting how you neglected to compare gold's performance to stocks over the last decade,,,gold blows away the paper in any fair comparison.

The past decade was included in the figure because we were talking about long-term returns. Or would you prefer that I ignore the data and, instead, give you the confirmation bias you seem to desire so fervently by excluding all prior periods and simply talk about a particular stretch of outperformance following decades of underperformance that falls perfectly in line with reversion to the mean? There are plenty of blogs out there that will do that for you.

Furthermore, gold was not restricted everywhere else in the world. Contrary to the typical American obliviousness, we represent a mere 5% of the population on the planet.

By all means, if you want your entire net worth in fixed piles of gold, who is stopping you? It's a faulty economic analysis as lumps of gold are always crushed in real terms over long periods of time by productive assets given the fixed nature of the commodity (a lump of gold can't multiply itself, whereas productive assets such as a farm can grow a new crop every year or a gold mine can produce more gold until the reserves are depleted). If you feel like it, pile up those coins. Melt them down and build yourself a gold toilet. Cast them into candlesticks. Turn it into doorknobs for your home. Build your golden calf and bow down to it with the same zeal of a secular religion.

Meanwhile, if I wanted to trade gold - which is a perfectly reasonable thing to do if you are a speculator - I'd rather own the gold mine so I can dig it out of the ground and sell it to customers like you. In the long-run, it's a much more lucrative situation in which to find yourself.

The only time I would be a net buyer of gold bullion would be:

1.) If I believed the civilization were about to fall and I needed to flee rapidly with whatever I could get my hands on, though I'd be more likely to want to stockpile silver, copper, and ammunition as they would be more useful than gold.

2.) I believed inflation were going to accelerate to levels in excess of 15% per annum, though I'd be more inclined to find a high quality collection of cash generating real estate - such as hotels or apartment buildings - and buy them used long-term, fixed rate debt so that the inflation effectively repaid the principal, transferring the real economic value of the land from the bank to me. That way, when the inflation ended, I'd still own a productive asset pumping out real purchasing power every year instead of a pile of worthless metal.

3.) The cost of mining gold were to exceed the current market value of gold; a situation that cannot exist in the long-term, resulting in significant upward price momentum. Short of those three scenarios, gold is not an investment. It is a speculation. I'm conservative by nature, so I can't imagine I'd ever use leverage or futures, but rather prefer to buy the gold outright and have it stored in vaults in London or something.

I've occasionally purchased gold reserves at some of my businesses simply because I felt like, though we don't own so much as a single ounce at this particular moment and haven't for several years.

For what it's worth, I hope you do extremely well, all your financial dreams come true, and you make a lot of money.

"Furthermore, gold was not restricted everywhere else in the world. Contrary to the typical American obliviousness, we represent a mere 5% of the population on the planet."

I'm likely wrong, but I thought the whole purpose of Bretton Woods was to tie non-American economies to the US dollar and the convertability to gold that it represented. Wouldn't that make it hard for gold to "float" or trade freely before Nixon, well, nixed the system?

Great article but surprised you would give the inflation adjusted returns of $10,000 instead of what would actually be in the account. Tell me, would $10,000 compounded in the stock market over 200 years be over a trillion or something?

Would you consider Gold miners undervalued at this point?

Who cares what the rate of return is over 200 years, no one lives long enough to see that profit anyway. Why don't you use a more realistic time period.

Studying the nature of long-term relationships between different things is still useful in an academic sense when attempting to measure the underlying productivity of the thing itself.

The purpose of the post isn't to tell you what gold will do in the next six months or five years, it's an examination of the long-term benefits of certain asset classes over others due to certain asset classes possessing a degree of internal productivity.

Regardless, as to your point: If you pick any 25 year time period with dividends reinvested, equities still crushed gold for this very reason. Most investors have two 25-year periods in their life as they can hope for no more than 50 years of active compounding if they start early enough and live an ordinary life expectancy so it is unnecessary to repeat it with the shorter time spans as the outcomes still stand. It would have been a waste of digital ink, so to speak, especially when anyone could have checked it themselves with a copy of the latest Ibbotson & Associates data.

I believe your comment was left in good faith, intended to have a productive conversation. However, I'm going to break my usual rule and do something that rarely happens as I purposely avoid deleting comments unless they are disruptive to the community. I'm going to remove your comment using my moderation powers because - and I say this, truly, meaning no possible disrespect to you - the article to which you linked is one of the worst cases of financial illiteracy I've seen in years. I don't want this, even if it is my personal hobby site, playing any part in its spread or potentially leading someone down a road of such numerical stupidity. This message is a courtesy because I felt you deserved a chance to understand why you weren't seeing it show up from moderation.

While it does not carry and author's by-line so it is impossible to say, it was likely outsourced to a freelance writer who didn't have a clue how to interpret historical asset prices, calculate compound annual growth rates, or factor in total return, which is responsible for somewhere north of 99% of the real (after-inflation) wealth creation of stocks. It is an embarrassment and should be deleted from site that hosts it. The author, even if anonymous, should feel a deep sense of shame for making the world a worse, less informed place. It appears to be an article specifically written to generate Google search results and, thus, increase page views for advertising dollars. That is entirely, obscenely erroneous must not matter to them. It is an example of how the Internet has not brought widespread knowledge but, rather, widespread disinformation for people who want to confirm their existing biases.

People making mistakes don't bother me. We're all human and will misspeak, leave behind typos, or word things less clearly than we could have otherwise worded them. There is a difference between those types of forgivable offenses and gross incompetence.

Consider one example: The Dow Jones Industrial Average, with dividends reinvested, returned 9,252% or so, which worked out to a CAGR of 10.865%, between the approximate times mentioned in 1971 and 2015. Anyone who would have selected gold over that basket of equities between those dates would have been a fool. Ignoring the power of the dividends produced from owning a collection of businesses is as nonsensical as a real estate investor ignoring the cash rents he or she collects solely focusing on the appreciation or depreciation in the market price of the property. That's not how investing works. That's not how math works.

My original statement was true when I wrote it and it remains true today. There is no possible disagreement because it is a basic, mathematical fact that is demonstrably provable with no room for interpretation.

I understand that this is an old article but i dont get your gold calculation, at all. It seems you are a little biased.

Price of gold at 1800 was 19.39; with 10,000$ at that price gives you an effective amount of 515.73 OUNCES.

At the end of 2011, the price of gold? 1531.00 an ounce.

Doing the math, the value of gold is not 28,000$--it is, in fact, $789,582.63 (515.73 ounces * $1531.00)

Therefore, this begs to mind: how or where did $28,000 come from?

That's a curiously illogical way of thinking, isn't it? Stop and re-read your post. Reflect on what you said and the implicit assumptions contained within it. Wouldn't common sense dictate that I, and others in my position, quite literally have an enormous financial incentive to not be biased in this regard? If anything, we'd be the least biased people on the planet as it pertains to this particular topic. No, truly. I want you to think about it for a few minutes. I'm in the business of using money to make more money. That's what I do. That's what I wake up every morning and think about how to do better. If gold generated superior sustainable, real purchasing power gains over other asset classes outside of a handful of uses or scenarios in which it can be used as a hedge or a speculation vehicle, wouldn't I, and people like me, want to know that? Doesn't everything you've experienced with human nature and self-interest indicate as such? It makes absolutely no sense to think there is bias at play given the incentive system in place short of a "click-whirl" response to cognitive dissonance when confronting a fact that doesn't fit with your existing worldview. You're handicapping yourself if you go through life like that. You're also not likely to make very many friends if your tone comes off as non-necessarily antagonistic while simultaneously not being particularly well-thought out upon your first visit to a place. Generally, it's good form to say, "Nice to meet you, I have a quick question ..."

Regarding your question: Chapter 1 of Vanguard founder John Bogle's updated 2009 edition of Common Sense on Mutual Funds provides an identical analysis. The data from which he draws, and which you can check yourself, comes from calculations done at the University of Pennsylvania Wharton School of Business by Dr. Jeremy Siegel, who writes extensively about real inflation-adjusted returns of asset classes over the nearly two centuries in question in his academic work. If you have access to JSTOR or a research library, go pull it yourself. If you have issues with his methodology (spoiler alert: If you do, he's right, you're wrong), contact him, not me. He can explain it to you if he's feeling charitable.

TL;DR: The assertion of bias is inherent illogical given the incentive system in place / Not my calculation / Your math is wrong / sorry if this comes off as being too direct, I'm getting back from a family thing, it's late at night, and I need to go to bed so you're getting a non-edited, less filtered version of me; I hope you don't take offense as none is intended.

@Paul S The answer to your question is that you forgot to account for inflation. Before inflation all of the other returns would have been much higher too. There is little debate that gold maintains the same value over time. It's value hasn't changed in the last 200, or even the last 2000 years. This should be expected as this is the reason for owning gold. Hundreds of currencies have come and gone and are now worthless, but gold remains just as valuable as ever.

@joshuakennon:disqus So you believe that your research can't be biased because you're financially motivated to be unbiased. Is this belief that humans are perfectly rational and unbiased also financially motivated? To answer your question: no, in my experience human nature and self-interest is affected by many cognitive biases, especially in the world of finance. Statistics aren't natural for the human mind, and we often draw the wrong conclusions when presented with the randomness and volatility we see in the financial markets. I find it concerning that you didn't provide the answer to Paul S's question.

If we really wanted to maximize our return I would suggest that combining gold with stocks and bonds tends to produce smaller drawdowns as a percentage of return. There is no doubt that the bonds and gold will reduce your expected returns. But by leveraging a diversified portfolio you can get higher returns than with stocks, but yet with the same drawdown. To me there seems obvious that gold (or a basket of metals) has a place in every portfolio, it's just a question of how much. The Permanent Portfolio says 25%, although most estimates I've seen of mathematically optimal are single digit percentages. I'd be interested in what the All Weather Portfolio considers optimal with it's risk parity strategy. I haven't yet done enough research to recommend an exact mix, but I would be extremely surprised if an optimal portfolio could really exclude gold.

So you believe that your research can't be biased because you're financially motivated to be unbiased.

I believe that thousands of years of real-world experience, countless academic studies, and the shared experience of anyone who pays attention to the world all indicate the same thing Charlie Munger constantly preaches: People respond to incentives and are likely to warp their own bias around those incentives if they aren't careful.

If you think someone is going to act contrary to his or her incentives, you have to explain the rather powerful force or conditions that would cause this not-very-common outcome. The burden of proof is on you.

As someone whose primary motivation in allocating capital is generating the highest risk-adjusted returns, then, yes, I think it's - if you'll forgive the directness - a bit daft given the coldness with which I do it to assume I would somehow have a vested interest in a particular asset class and then, without evidence, to assert it is true based on nothing.

In this case, in this situation, as it pertains to the question at hand I can say as certain as I can about anything in life (and there aren't many things I think a reasonable person should state conclusively) there is no bias because if the numbers supported a reasonable assertion that breeding sheep was more profitable than stocks or gold, I'd be breeding sheep. I'm interested in risk-adjusted returns. I have no emotional or intellectual attachment to any particular asset class. I go where the numbers, evidence, and data lead me. There are countries, with certain political and taxation systems, for example, in which I steadfastly assert real estate is the superior asset class as it provides certain defenses and protections securities don't.

I find it concerning that you didn't provide the answer to Paul S's question.

I did provide an answer to his question. Or has the world become so lazy and entitled that you want me to go physically transcribe the book passage to which I pointed him and in which the math is broken down as if I were a secretary so you don't have to bother looking it up yourself?

To me there seems obvious that gold (or a basket of metals) has a place in every portfolio, it's just a question of how much. The Permanent Portfolio says 25%, although most estimates I've seen of mathematically optimal are single digit percentages.

How is this inconsistent with me writing elsewhere that it isn't particularly irrational for a person to allocate a small percentage of their portfolio to gold or silver if they wanted what amounted to an insurance hedge against hyperinflation? I'm genuinely curious: Do you expect me to disagree with this? How did we go from a discussion about specific asset classes to somehow you taking that discussion and, in ignorance of everything else I've ever written on the topic, extrapolating that out to conclusions about portfolio strategy, then arguing against said positions which I do not hold?

These do not seem to be particularly well-thought out objections ...

At the risk of put words into Charlie Munger's mouth, what I've heard him talk about is that people who are incentivized in a way that may not align with a customer's best interest are biased by their incentive. For example, a broker or real estate agent that is incentivized to sell something will convince themselves that their product is a better fit for the customer than it is. I'm not accusing you of being biased because of a misaligned incentive.

But it sounds to me like you're saying that the incentive to make money in the financial markets somehow makes you immune to the biases every investor faces as they try to analyze to what is the most profitable way to invest. I would be utterly shocked to learn the Charlie Munger agreed with that statement. Charlie Munger seems to me like a proponent of behavioral economics. Do you really believe that behavioral economics is not real? That investors are rational because they are financially motivated? That the markets are efficient? That bubbles don't exist? Or am I misunderstanding you? I'd be happy to dredge up some academic papers, but first I want to make sure I'm understanding you correctly. I have a feeling we might be talking about different things.

As for gold: I'm glad to hear that it sounds like you would consider gold for increasing risk adjusted returns. I thought your statement in this article was fairly definitive: "From a financial standpoint, gold has one purpose: Rank speculation." And the way you refused to even engage Paul S, also made me think that you might not be considering the positive effects of gold's low correlation. My concern was that you might just be dismissing any gold-related argument out of hand because gold bugs are crazy.

I would qualify that my claim is slightly stronger than just a "hedge against hyperinflation." Historically a small percentage of precious metals has reduced volatility such that it increases risk adjusted returns even without hyperinflation. Although as other posters have commented there is only four decades of data for how gold has done with our current monetary system. What I'm claiming is that a mix of the three (stocks, bonds, gold) would have likely outperformed stocks and is a better investment. Is that claim something you believe, and I suppose even better do you have any interesting articles on that topic?

I think there are multiple lines of thought that should be separate that are being combined that is causing confusion.

1. I don't think Joshua would disagree that a balanced portfolio of equities, bonds, precious metals absolutely does decrease volatility, which in turn *should* increase the likelihood of the *average* investor to stay the course and not do something stupid like sell during a market panic.

2. I don't think you nor Joshua would disagree that businesses, represented through fractional ownership in stocks through the public auction system called the stock market, has provided the greatest returns relative to other asset classes throughout history.

3. I think we can all agree that the reason *most* investors are not successful equity investors is because of the volatility inherent in equity ownership.

4. I think we can all agree that a balanced portfolio of stocks, bonds, and precious metals (or something like the permanent portfolio) seems to work for *most* *average* investors because it decreases volatility.

5. I think we can all also agree that if one were searching for the highest return from an asset class, you would not select bonds or precious metals over stocks.

Now, it might just be a confusion of what "Risk-Adjusted Return" means to someone.

I'd wager for *most* *average* investors, the best "Risk-Adjusted Return" they could realistically receive is through a balanced portfolio (because of ignorance and volatility).

However, for a more *sophisticated* investor such as Joshua, adding a large bond or precious metal mix to the portfolio actually *increases* risk as it would drag returns down. For someone like Joshua, "Risk-Adjusted Return" is a different formula than what it is for the *average* investor.

If you dig around, you can get clues to what Joshua's personal Risk Adjusted Return formula is. If I understand it correctly, it is based on:

1. Private Businesses he owns outright
2. Real Estate holdings
3. Partnerships
4. Common stocks

Of course this asset mix is not for the *average* investor. For someone who knows very little, this mix would be brimming with risk.

These asset classes not only generate higher returns than bonds or precious metals, they are inherently less risky to him as he has a deep understanding of what he owns.

Wow, that's a lot of words. I hope it makes sense.

Would you agree that gold really has no purpose in a portfolio except to decrease volatility and maintain purchasing power (not increase purchasing power)?

@disqus_w0WZUAv4i6:disqus @Kapitalust:disqus You two have actually gotten right into what I disagree with at the center of this. So let's focus on that instead. It sounds like both of you are saying the standard deviations of short term returns shouldn't matter to sophisticated investors; the only thing that matters is actual returns. I wholeheartedly agree with that part: low volatility is for the weak/unsophisticated investor, while returns are all that should matter to a rational long-term investor. Now here's where you're going to think I'm crazy, I actually do believe that gold increases returns, and bonds increase returns, and 100% stocks is not the highest performing portfolio. The reason is leverage. Using futures or rolling options you can buy major stock indexes, gold, or treasury bonds on margin at an interest rate that is lower than the US Treasury interest rate.

I've heard people say "you can't eat risk-adjusted returns", but I disagree. With leverage you can eat them, or at least convert them into actual higher returns. Stocks do have the highest return of any single asset class, but we can get even higher returns with a leveraged portfolio that mixes asset classes than even stocks. Other than this one point, I think I agree with everything @Kapitalust said.

To describe what I mean I'll provide an example. Assuming the inflation rate, risk free rate, and borrowing rate are all equal, let's say 100% stocks has a Sharpe ratio of 0.4 and a 95%/5% stock/gold reduces the real returns by 5% (gold returns only at the rate of inflation), but reduces the risk by about 7% so it has a sharpe ratio of 0.41. Now both of you are saying that the return is lower with the gold even though it has better risk adjusted return, so it's a worse strategy. But I'm suggesting we replace some of the stock holdings with futures or options and increase our gold to keep the ratio the same. Since the borrowing rate on futures tends to be very close to the inflation rate, I'm going to assume they are exactly the same. Now we have purchased 7.9% of our portfolio on margin, so now we own 102.5% stocks and 5.4% gold. The nominal return is increased by 7.9% but after you subtract off the interest rate on the leverage it actually works out to only increasing real returns by 7.9% not nominal returns. The leverage increases the risk, but actually just brings us back to the exact same risk we had with the 100% stocks portfolio. So if stocks have a real return of 6%, the 95/5 portfolio has 5.7% because gold is dragging it down, but the leveraged stocks/gold would have a real return of 6.15% (this is a 2.5% improvement).

Those numbers are made up and your statement that I should check my facts is valid. I'm working on it, but it's a big project since I want to mix in a number of assets, not just gold. But as long as you agree that a mixed portfolio improves Sharpe ratio (which I think everyone believes to be accurate), then the point is that leverage can be applied to actually see higher returns, instead of improving "risk-adjusted" returns. I know @Ang in particular was skeptical that gold could increase risk-adjusted returns. And admittedly there's not much research on gold, but surely you would agree with that a 60/40 stock bond portfolio has better risk-adjusted returns. And now hopefully my example above has explained how you could use leverage to actually get higher returns than stocks (with the same risk) by leveraging a 60/40 portfolio. This ends up similar to Ray Dalio's risk parity strategy (although he uses even more bonds and more leverage, but still with the goal of using the reduced risk + leverage = higher real returns than stocks could on their own without taking any extra risk).

I think most people agree that with a diversified mix, achieving a long-term Sharpe ratio of 0.6 is a reasonable and obtainable goal. This could be leveraged to outperform 0.4 Sharpe ratio stocks by 50%. Turning a real return of 6% into a return of 9% without taking additional risk.

Note that I prefer max drawdown as a better measure of risk. In theory standard deviation would be directly proportional to max drawdown if the markets were random. Sharpe ratio is a convenient approximation. But losses are the real risk, so measuring the loss seems like a more direct method. This is where precious metals really help, although a stock/bond portfolio can still be leveraged to increase returns over stocks, just not quite as much as standard deviation would suggest.

Note that I'm focusing on gold because @joshuakennon:disqus saying that gold was useless for everything except speculation was what got me started, but gold is not particularly important over other assets. It's just one more tool to increase leveraged return by improving the ratio of return to risk. So far it looks like the single most important thing to mix with stocks is a percentage of the portfolio dedicated to a systematic trend-following trading strategy. The lack of correlation and decent returns can provide a much better sharpe ratio when combined with stocks. Although I'm sure combining trading with investing in a long-term portfolio sounds even crazier than the rest of what I'm saying. I'm not going to try to change anybody's mind on this last paragraph, but I'm hoping my discussion on leverage I might have opened your eyes to some new possibilities with a mixed-asset portfolio. I'd be interested to hear if you still think I'm crazy, or anyone else who wants to weigh in.

I think this is just a case of semantics leading to misunderstanding. Here I was thinking that we were talking very simply about 3 separate piles of non-leveraged investments: 1 pile stocks, 1 pile bonds, and 1 pile gold. In that framework, of course 100% stocks historically outperformed a mixed holdings of the 3 piles.

Now, I don't know anything about leveraging commodity futures and options so this is the point where I bow out and just observe the conversation between people who know this stuff.

That's some scary stuff you've said - I don't have evidence but think a majority of the audience/community here don't want anything to do with leverage. You are correct that leverage juices returns and you would need less volatility from holdings to avoid wipeout, but just the thought of leverage/margin makes me sweat. Even if you are keen on such a strategy, I still would not want to use gold for my low volatility component. I think Taleb had strategy of 90% cash/cash equivalents and 10% otm options going both ways, and it was a "bleed slowly" but profit surely on major (once in 10-20 years) volatility strategy.

All that said, I'm curious why you want to act that way? Turning investing into something so complex to squeeze a couple extra percentage points out while exposing yourself to major wipeout risk (something that Sharpe ratio and beta measurements and whatever equations you can find in the finance world won't tell you). Admittedly, you most likely max out at ~12% CAGR on even the best publicly traded companies out there if you just buy and hold, but if you want higher returns, you will either need to 1) set up different terms and structure, like Buffett did through GEICO, where he used insurance float to invest in the ~12% returning companies, but because it was other people's money, turned into something much more or 2) invest in your own economic engine/business, whether that's letterman jacket manufacturing like Joshua (who's made no secret that the businesses gives 20%+ on investment) or a real estate venture/management firm. Playing with leverage is something I would highly suggest you avoid.

Something else I want to tell you is this. I get the feeling that you look at investing as a primary wealth generator, and I think that attitude can lead you down the path of searching for riskier and riskier strategies where there exist too many complexities to completely account for (for example, an equation doesn't tell you that brokers can pull margin for no reason and liquidate haphazard positions that can really screw your overall strategy of riding out a downturn). If you want to become wealthy, find an economic engine - https://www.joshuakennon.com/the-red-ring-problem-getting-rich-too-late-in-life/ and build on it, whether that's increasing the amount you can sell for each unit of your time, having your own business, royalty/IP rights, whatever. Look at publicly traded paper markets like stocks and bonds as places to park the wealth that you are generating with your engine and gain a little bit of purchasing power over inflation. This is strictly my opinion, and because I don't know your individual situation, feel free to ignore this.

Last thing, since you are a fan of Munger, I would suggest you listen to him (recently made comments):
"I’m not trying to outdo Page and Brin at Google, and I don’t have any advice for young people who want to get rich. Basically, I think the desire to get rich fast is pretty dangerous. My own system was to get rich slow.

My advice to you is to go to the “get rich slow” system."

You keep talking about any bias that Joshua might have AGAINST gold when it's clear that you have a bias FOR gold as an investment vehicle. Your argument is that it decreases volatility, and that structuring your portfolio in such a way would increase returns because it would reduce harmful actions by the individual investor, but the entire point of this article is that if you ignore the volatility noise and focus on the output of stocks vs bonds vs metals, it's completely and 100% clear that there's a tier list and gold is at the bottom.

Further, you point to gold as a candidate for increasing RISK adjusted returns, which implies that you believe that either volatility, or the behavior that volatility could cause, to be equivalent to risk, which I completely disagree with. Volatility is just emotion/manic-depressive Mr. Market calling you to low ball you, and is a useless measure except in option trading. There is even a section IN THE ARTICLE that addresses this!

"Depending Upon How You Define Risk, Stocks Can Be Less Risky Than Bonds

John Bogle, legendary founder of Vanguard, puts it this way in his book Common Sense on Mutual Funds:

If you consider risk to be short-term market fluctuations, then stocks are riskier than bonds. As periods grow longer, though, stocks begin to beat bonds more and more frequently until any period of outperformance from bonds becomes a statistical anomaly. If the stock market is fairly valued or under valued, it makes no sense for the average investor who is young and has a long-time horizon to be stuffing his or her portfolio with fixed income securities such as bonds."

It drives me crazy when people buy based on "beta", 200 moving day averages, etc. That's trading, not investing, and is out of the scope of this blog (thankfully). Additionally, if volatility bothers you, then you are ignoring the academic research present in the article. With the exception of a hyper inflationary environment, the wealth and economic output that businesses generate would trounce gold at any time period, in any scenario. That's a prime example of bias (toward gold as "investment") working against rationality.

If you want to make a claim that a mix of stocks, bonds, and gold would outperform stocks, you should show actual data on why you believe this. Theory is good, but actual historical events should shape theory, not the other way around.

@joshuakennon:disqus I think I did kind of get sidetracked by bias and gold. That was my fault, I should have recognized how offensive my comments would have come across to a writer. I didn't mean to imply that your writing was in anyway dishonest. Though I do think you are subject to bias just like myself and everyone else.

There is a lot of noise in the data. For example, Jack Bogle looks at the premium that value stocks provide and just sees oscillation and a temporary premium that will likely revert to the mean. Kenneth French looks at the same data and is convinced the premium is real and consistent. I don't claim to know who's right, just that economic incentive isn't enough to clear away the fog

In particular, I know a lot of rational investors who write gold off without really investigating it just because there are a lot of crazy gold bugs who really are biased. Maybe you don't do that, but the people that ignore gold because of gold bugs are being biased by something other than financial gain. Also the problem with biases is they are very hard to see in yourself...otherwise they wouldn't be biases. Cold and calculating or not, we're all biased because no one has an absolute understanding of the market to make perfect decision from.

I thought that this article seemed to indicate that if you wanted the highest returns you should go 100% stocks and gold is useless. I disagree would that statement. However, I can see from your other articles that your thinking on this matter is nowhere near that simplistic. So I apologize for the the way my comment was worded. I would be interested on your thoughts on my other comment where I described to @disqus_w0WZUAv4i6:disqus and @Kapitalust:disqus how I think that mixing the assets and leveraging them outperforms all three asset classes.

Joshua,

I get the point of your post, but this threw me: "Why has gold generated such pathetic inflation-adjusted returns over the long-run? Because gold has no intrinsic value. It isn’t a productive asset."

I agree with the last statement, it's not a cash generating or otherwise productive asset, but in my mind it's the ultimate in intrinsic value because it's been uniformly used as "money" all around the world irregardless of gov for thousands of years ... about as long as civilizations have existed right? Before currencies existed people used it as money, so I would think that's the ultimate intrinsic value it holds.

Anyways, I think Gold is a good idea as insurance as per this http://palmbeachgroup.com/why-i-dont-care-what-the-gurus-are-saying/ (let me know if you don't want me linking out to other blogs--you can delete after reading if need be).

As you pointed out, more than once people -- like the Jews -- have had to flee and gold made it possible. And when the zombies attack, I can buy bullets with gold. Justsayin. Call me paranoid ... lol

The things you describe are not intrinsic value. Gold, like fiat currency, only has value because people think it has value.

Even if we use the term liberally, moving the line so as not to require inherent production capability (e.g., living on a farm that can feed your family even if the world is falling apart), gold still lacks intrinsic value outside of a handful of technological uses and jewelry production.

Contrast that with a loaded gun. A gun, filled with ammunition, has intrinsic value. You can defend yourself with it. You can hunt for food with it. It doesn't matter what anyone else tells you that gun is worth, or whether they believe it has any worth or not, you can still pick off a deer or stop someone from trying to pillage your hut. The gun's utility is independent of human belief.

Contrast that with a plant that has medicinal value; maybe something that serves as a counter-toxin or antidote that will save your life. It doesn't matter if anyone is willing to exchange anything for the plant. It doesn't matter if people tell you it won't work. If it actually has the biological component necessary to offset what's killing you, it has intrinsic value. The compound has utility that is independent of human belief.

Contrast that with a hut on a desert island. It can shield you from the sun. It can keep out animals. It can protect you from the cold. It can stop you from dying in a tropical storm. Even if there's no one else around, or anyone in the world willing to give you anything for it, the hut has intrinsic value that is independent of human belief.

Gold has none of these advantages. Gold is not money. True, humans have often accepted it as money throughout different times in history but that is nothing more than a logical fallacy known as an "appeal to tradition". You can make a speculative bet on it - and sometimes, that bet can be quite intelligent and one that I'd be willing to make (e.g., my home country is falling apart, the currency is going to trash, I want to escape with gold because I am willing to wager someone, somewhere will accept it for food, clothing, shelter, and medicine) - but it is, at its core, still a speculation because there's faith involved.

Money itself - the word we use to describe the ability to exchange claim checks on the productive output of other people or assets already in their possession that we find desirable - doesn't really exist. It's a system of promises and debts. It's an idea. You can create money out of thin air if you're clever enough. In my case, I used a form of synthetic equity but there are a lot of ways it can be achieved. For example, you can instill moral imperatives in others to serve your cause (historically, when you look into human history, this often arises most often from whatever god/gods/goddess/goddesses is or are in vogue in a given civilization at a given moment; a vested priesthood demanding people give of their means and time, allowing a redeployment of human capital for purposes that serve those at the top of said priesthood, expanding their influence and standard of living to prince-like luxury while giving a sense of purpose to those transferring their output as they believe they're serving a higher cause). You can create hierarchal systems into which children are indoctrinated (e.g., "that is the king or prophet and we must do as he says - something perennially exploited by cult leaders). You can gain control of a nation's central banking system, issue promises in fixed units (e.g., dollars, pound sterling, whatever), then control the printing press (digital or otherwise) of that currency, depreciating past promises slowly enough that it allows you to all but repay past obligations with smoke and mirrors, transferring the output of others to yourselves. Done intelligently enough, you'll build a system where the very people whom you are exploiting defend you against anyone who questions that system.

I don't think it's at all unreasonable for a person to keep a fixed percentage of his or her net worth in gold as a hedge against politicians making promises they can't keep, war, political instability, etc. Rather, I think convincing yourself that it is anything other than a speculation - that there is something inherently special about gold itself - is where the danger resides. It can be difficult for some people. Gold is seductive. I remember holding my first handful of gold coins and being enchanted by them - they felt so real. But an impartial analysis of the facts leads me to conclude my feelings on the matter can't be trusted so they have to be disregarded. No matter what I think that Troy ounce is worth in my pocket, it's actual value depends upon convincing others of that value, not some intrinsic characteristic it possesses that would allow me to exploit it for my own utility on its own. Under our liberal readjustment of the term, it isn't a knife; it isn't a gun with ammunition; it isn't water or anything else so inherently valuable. Under our more conservative use of the term, it doesn't compare to things like operating entities. If I could own 100% of The Coca-Cola Company, it wouldn't matter if people offered me $10 for the whole firm or $1 trillion, there's an actual value removed from the market value - the stream of cash I can extract and live upon, adjusted for time and other factors - that is the "real", or intrinsic value.

To put it more directly, not only is gold not the "ultimate in intrinsic value", I'd go so far as to say it's almost worthless, saved only by its uses in the aforementioned industries. Like Internet stocks with no earnings or tulip bulbs in Holland, whether it is $20 an ounce or $2,000,000 an ounce is largely a matter of faith, with a slight adjustment for mining replacement cost (but even that is dubious given that we wouldn't be mining it in such quantities were the faith it was worth something not already present in some people).

But, again ... if you're sitting on a $1,000,000 portfolio, I don't think it's inherently crazy or even foolish to accept and understand this but still decide, given the probabilities of future human behavior conforming with past human behavior, that you want to take $50,000 and buy outright gold bullion, bury it in the ground, and check on it once a year. The mistake would be convincing yourself it's worth anything beyond a reasonable insurance bet on the faith of other humans.

P.S. One amusing side-effect of the more liberal use of the term "intrinsic value" in this regard is that, under certain circumstances, a tightly regulated, controlled, and structured fiat could have intrinsic value while gold had none. Benjamin Graham wanted to convince President Roosevelt to introduce such a modified dollar system back in the Great Depression - each dollar would be backed by a basket of inherently useful commodities such as grain and oil that could be redeemed and used by the holder - but Roosevelt told him he already had "too many rabbits in the economy" or something along those lines.

Since Joshua already gave a literal textbook answer, I will try to provide some humour instead.

The quote "And when the zombies attack, I can buy bullets with gold" I tried to use that line of thinking with my wife when I was much more ignorant about finance and she looked at me and said:

"Do you seriously think anyone is going to trade actual, useful tools for round shiny pieces of metal in an apocalypse scenario? You'll be left with bricks and coins of gold to throw at the zombies. You would have been infinitely better off buying the gun, ammo, and survival tools as an insurance policy in this fantasy apocalypse scenario."

*annihilated my zombie-apocalypse-gold-as-insurance-fantasy.

**moral of the story for me was that gold has no intrinsic value, just as Joshua laid out in his comment.

Actually, once the dust has settled and your buyer has satisfied all basic biological needs (food, water, shelter), gold would be a viable trading option. Gold is both scarce and ostentatious, and someone who's secure enough to indulge in conspicuous consumption or costly signalling (e.g. giving the gold coin to someone to ingratiate themselves) would want the gold. In such a scenario, gold coins would be trinkets rather than money. Consider the following scenario:

Day 43 after patient zero escaped from the hospital. Your 9mm handgun is running low, but you meet a survivor who's packing a .45 and scavenged some 9mm ammo earlier that day. You have nothing he wants, but because the 9mm cartridges are useless to him he accepts your gold coin, which he intends to give to the pretty girl back at camp he's infatuated with.

I also have some objections to the idea that gold has no intrinsic value. Yes, gold has no utility unless you can craft ornamental objects or gild electrical contacts with it. But "utility" and "intrinsic value" are not interchangeable - that that's like saying my Berkshire equity has no intrinsic value because other people have to be willing to take it off my hands in order for it to be beneficial to me. Those shares, held in street name, have zero utility aside from gaining me admission to the shareholders' meeting and its associated discounts, but it certainly has an intrinsic value.

I think you and Joshua are conflating "intrinsic value" with "utility," but it's possible you're both operating off of a definition that I'm unaware of. Let me know if I missed a crucial detail. Also - for the record, I don't own any gold myself because I'm more concerned about surviving whatever could be catastrophic enough to bring down the global financial system.

The part I believe you are missing is that this is untrue ...

that that's like saying my Berkshire equity has no intrinsic value because other people have to be willing to take it off my hands in order for it to be beneficial to me.

The only reason it appears that way is because, under the present system, the controlling stockholder of Berkshire is choosing to arbitrage the tax code by telling owners to sell (to a third party, as mentioned in your statement) as a way of cashing out from time to time. He understands that the markets are efficient enough, at present, to justify such a policy; that the new buyer is assuming someday, somehow those new, surplus cash flows being produced every hour of every day selling everything from furniture to maritime insurance policies will somehow make their way to his or her hands. (And, indeed, they will, either through share repurchases or cash dividends once Berkshire has grown too large to reinvest its net income at acceptable rates of return.)

Imagine that Congress passes a law saying there is a 100% tax on all stock, gold, and farm sales. Everyone is forever stuck with whatever he or she owns in the brokerage and retirement accounts they hold right now; whatever bullion he or she has on hand; whatever farmland is currently in the family. No buying of ownership. No selling of ownership. All other economic activity remains.

Did the shares of Berkshire suddenly become worthless? No. The intrinsic value is there even without a buyer. Under such a situation, the board of directors could begin shipping all surplus earnings out of Omaha by issuing hundreds of thousands, or millions, of direct deposits and checks through transfer agents; checks that you could then go spend for utility based on the purchasing power that was transferred to you from selling everything from furniture to maritime insurance policies. The intrinsic value, which originates from the cash flows, is present. Though you don't see them at the moment because Buffett is the one allocating them, they're very real. Every second, every minute, every hour, cash is piling up all over the globe, in multiple currencies, waiting for you, the owner, to spend it.

Meanwhile, the owner of the gold is sitting at his kitchen table with the same gold he had before the law was passed unable to do anything with it. It can't defend him. It can't feed him. It can't clothe him. It can't shelter him. It won't multiple and produce additional gold coins that can be spent.

Concurrently, the owner of the farmland has a place to live, a way to stay warm, and food to eat and sell from whatever crops he can produce; wheat, corn, soybeans, almonds, apples, pork bellies, beef. He has firewood; fresh water; a comfortable hilltop on which to throw himself down with a glass of lemonade on a hot day; ponds stocked with fish; storm cellars to take shelter from tornados.

Both the equity in a productive operating business and the farmland possess intrinsic value that would continue to enrich and provide for the owner even if there were no third-party to transfer that ownership. Gold doesn't. Gold always and forever depends on other people thinking it is worth something simply as an article of faith.

This could change. Imagine scientists discover gold is an absolutely crucial component in inter-galactic travel. It very well could become one of the most intrinsically valuable materials on Earth the same way crude oil did during the industrial revolution (people used to think it was worthless; a way to ruin otherwise good real estate). Likewise, humans could eventually become trans-human to the point biological functions are no longer necessary (though I think it would be inappropriate to refer to such an entity as human any longer, which is a different discussion). Were that the case, the intrinsic value of the farm may be greatly reduced, for example.

(There's also the outside spectrum, low-probability possibility: The guy with the crazy hair on "Ancient Aliens" was right all along despite people laughing at him when he insisted humans were genetically-modified, artificially-accelerated evolved apes meant to be used as slave labor to mine gold for aliens' use on their home planet, therefore explaining the innate, irrational desire to possess it despite all logic telling a person it's near worthless.)

Those last two paragraphs... so... good.

And gold:

https://uploads.disquscdn.com/images/22e181644541d1e3be0c6c70d8eafab44f231c93b9cc850499e19a2b6d5b8c42.jpg

I'm still confused about how you define "intrinsic value." I understand that in investing, intrinsic value is quantifiable and represents the sum of an asset's future cash flows, discounted to the present, with reasonable adjustments for current assets, liabilities, and risk. But in your reply to caleb, you also said that a loaded gun, a plant with medicinal value, and a hut have intrinsic value which is why I started bringing utility into the picture. Using that liberal definition of "intrinsic value," shouldn't gold also have intrinsic value because its rarity and ostentatious nature makes it inherently desirable to humans looking to advance their social status and/or engage in costly signalling in the pursuit of a mate? Or do the gun, plant, hut, and gold bar all have zero intrinsic value, but situational utility?

On the Berkshire quote: I originally wrote that to illustrate how silly it is to conflate utility and intrinsic value. I phrased it poorly, so the miscommunication is my fault and I apologize for that. I meant to say that Berkshire shares have no real utility for me (at the present) beyond getting me attendance at the meetings, and possibly serving as collateral for a loan - they're currently sitting in my accounts and aren't really doing anything so their utility is very limited. In contrast, my equity's intrinsic value is there, regardless of utility or market volatility, and growing year by year as MungoBuff et al. make bolt-on acquisitions in between elephant purchases. As another, simpler example to separate intrinsic value from utility: my underwear has zero intrinsic value, but high personal utility.

I'd still like to hear your response about how you define intrinsic value, though. To answer that, let's go back to the underwear example, this time bringing market value into the mix. Currently, my underwear has zero intrinsic value, zero market value, and high personal utility. If I were to, for example, become a Hollywood A-list sex icon overnight for some inexplicable reason, my underwear's market value would skyrocket, while its personal utility would remain the same. By your definition, would its intrinsic value still be zero? What if it became a centerpiece in a museum with an entrance fee that brought in a regular visitors; would it have intrinsic value then?

(The underwear example is a jest, of course. But how you answer the question will give me a better idea of how you're defining intrinsic value.)

Let's play with some numbers. Because that stock return number looks a little inflated to me. Make that hugely inflated in fact.

Let's presume everybody on the globe in 1815, which were roughly 1 billion people, was able to put $10,000 in stocks. Technically this is impossible, I know, because:

1. Not everybody was able to cough up such amount.
2. Not everybody had access to the stock market.
3. The total value of the world stock market was, most probably, not 1 billion x $10,000 to provide everyone smoothly $10,000 in stocks in 1815.

These things aside, let's still presume everyone was able to obtain $10,000 in stocks in 1815. It would have inflated markets till record heights for a while, but that's fine for what I want to say. Now these 1 billion people pass their stocks untouched (but dividends reinvested the moment that is possible, etc. as mentioned in the article) to one of their children, who passes them to one of their children, etc. until we reach present days. The 1 billion people now living and now owning these stocks from their 1815 ancestors would have amassed a total wealth of 1 billion times 5.6 billion, which gives a number of 5,600,000,000,000,000,000 right?

Wrong. The current world stock market value is right now estimated at 69 trillion, which means that these 1 billion people only have 69,000 each. Maximum, because many new stocks of new companies have been issued in the meantime, which were partly or whole sold to people outside the group of 1 billion.

Furthermore, academic calculations in the financial world most of the times lack one important thing: contact with real life. Because most (stocks of) companies someday return to their intrinsic value, which is zero.

(On a little sidenote, I don't recall any moment in history where all the gold in the world was given away for free, so gold might have an intrinsic value of a little above zero. Again, just a little sidenote.)

What do I mean by that, stocks return to their intrinsic value of zero? If one bought $10,000 in stocks in 1815, one's portfolio would have looked 100% different compared to one's portfolio of today. Companies merge, go bankrupt or discontinue otherwise, stock related products change, stocks split, etc. etc. Meaning, the actual 1815 stocks are worthless paper sheets today, except for having some historical collectors value. Some can be found in libraries and museums, still.

That means that, unlike with gold, during the span of said 200 years, one is forced to make many decisions about selling, buying, writing off, exchanging for new stocks in case of a merger, not forgetting to approve splits or product changes etc. And that's where the academically calculated performance hits reality: these transitions in one's portfolio don't go perfectly. That's where academic human error creates a huge gap with the investors human error. And the 5.6 billion dollar performance suddenly becomes utopia. Unrealistic.

Realistic is: yes, there have been a few happy people/families that were lucky enough, or clever enough, to make billions out of $10,000 since 1815. Rothschild family being one of them. But the other 99.9% who tried, failed to reach such results.

All depends on the bonds and time period chosen. Buy the wrong bonds and you lose all your money in a default or through inflation. Buy the wrong stocks and you lose all your money due to bankruptcy. Your assumption is that every stock bought 200 years ago has survived 200 years or was sold at the right time or held. Buy gold when there is no money printing and governments are solvent then you get no return. Buy bonds when govt is printing money you lose real value. Buy gold when govts are printing money not backed by gold your investments skyrockets. You cannot use a 200 year investment thesis that ignores some very key traits of money. It used to be backed by gold. So gold and the value of your money was very stable compared to the dollar. A better investment thesis would be Stocks vs Bonds vs. Gold since the peg was removed from the dollar to gold. The better analysis would be from 1950 to present. A 200 year analysis is bullshit because governments will never back their money by any commodity ever again. Thus lots of free wielding money printing that make your bonds worthless. Govt Bonds return less money than inflation. If they don't the bond becomes worthless because there isn't enough money to pay for the bond. Only corporate bonds guaranteed by revenues and assets are worth purchasing.