Transcript
Paul Belanger joins the Gold Exchange Podcast to talk about his gold analysis, why the Random Walk Hypothesis and Monte Carlo analysis are wrong and how much gold is the “right amount” to have in an investment portfolio. Keith and Paul also discuss inflation and the 1970’s, the problems with data science, back tested models, and the failures of the mainstream financial consensus.
Show Notes
Thoughtful Disagreement with Ted Butler
The Case for Gold Yield in a Portfolio
The Rise and Fall of Interest Rates and Radio Shack
Investments, Speculations and Money by Paul Belanger
Unit of Account and Current Valuations by Paul Belanger
Make sure to subscribe to our YouTube Channel to check out all our Media Appearances, Podcast Episodes and more!
Transcript
Dickson: Hello everyone, and welcome to the Gold Exchange Podcast. My name is Dickson Buchanan. I’m the Vice President of Marketing for Monetary Metals and I’ll be hosting today’s episode. I’m joined by Keith Weiner, the founder and CEO. I’m very excited about today’s episode because we have the Paul Belanger as our guest today. Paul is, in my opinion, one of the premier analysts on gold and gold investing and I would say a real gift to the precious metal community. Let me unpack what I mean by that. If you spent any amount of time in the precious metal interwebs, which we all know can be a dark and gloomy place at times, it is excruciatingly difficult to find solid, unbiased, well researched analysis on precious metals. And that’s because the vast majority of what passes for “analysis” is simply pandering to people’s fear or greed as they try to sell you something. There’s the fear camp, which is epitomized by “the dollar is going to crash tomorrow, so hurry up and buy your gold and silver now before it’s too late!” And then there’s the greed camp, which is gold is on its way to 510, $60,000 an ounce, as we heard one commenter say one time.
So you better pick up the phone and dial 1800-Buy-Gold-Now. And of course, these are two sides of the same coin, right? Fear and greed are two sides of the same coin. So you get this kind of toxic cocktail where it’s almost like “how to profit with gold from the coming collapse”, right? How to get rich with gold as the world burns around you. That’s unfortunately what a lot of analysis is. But Paul is one of those rare breeds that doesn’t do any of that. Instead, he has produced hundreds of videos and most recently a book, which include timeless, data driven analysis on the benefits of owning gold and how gold can help investors achieve their financial objectives. He’s done all of this on his own as an independent thinker, researcher and investor himself, all while concluding a successful career as a chemical engineer. Paul, Keith and I are very pleased to have you. Welcome to the show.
Paul Belanger: It’s a pleasure to be invited. Thanks for having me.
Dickson: Great. So before we get started, I want to briefly mention a few ways that our audience can connect with Paul and his work. First, there’s his YouTube channel. Most of his analysis, I want to say going back to almost a decade, Paul, if that’s right. Okay, so about a decade, Paul has been producing videos on his YouTube channel which is called “belangp”. That’s B-E-L-A-N-G-P. I highly encourage you to press pause right now, find “belangp” on YouTube and hit subscribe on his channel. You’ll thank me later. In addition to YouTube, Paul has published, just recently published, I believe last month published his book, which is called Evidence Based Wealth How to Engineer Your Early Retirement. And that can be purchased on Amazon. Any other place that can be purchased, Paul, or is that the best?
Paul Belanger: No, just Amazon for now.
Dickson: Just Amazon. Okay, great. Last but not least, he is also the owner and author of Evidence Based Wealth, the website which contains his written analysis. I’m sure most of you are probably familiar with Paul watching his videos, but he does write as well. He does publish written word content on his site and I can attest that it is as good as his videos are. So be sure to check out his website. We’ll include links to all of those in the show notes once we publish this episode. Okay, so for today’s episode, I think it’s only fitting that we take a deep dive into gold on this episode. Between the two of you, Keith and Paul, we have a large amount of combined knowledge on gold in this room right now. In this virtual space, we might break your speakers or crash the internet with the amount of gold knowledge between the two of you. So I think we should take that second layer, third layer down into Gold in today’s episode. And I think what would be particularly interesting is to cover because when I think about both of you, you’ve arrived to similar conclusions on gold, but you’ve taken very different paths to get there.
So I think that would be really fruitful ground to cover for today’s episode. Obviously our audience is going to be familiar with Keith and his work. So we should focus more on you, Paul, and kind of let you explain and expound on your body of work and give Keith the opportunity to comment on that. I would love for you to kind of go through your research and findings on gold and tell us a bit about your methodology, kind of the tools that you use when you do your research. But why don’t we start by just you, Paul, telling us a little bit about yourself, your background and maybe how you were introduced to gold originally.
Paul Belanger: Sure, I’ll try to keep it brief. So I graduated with my PhD in chemical engineering from Lehigh University in the mid 90’s. And that’s when I took my first job in the industrial world. And as we all know, when we first get a job, one of the first things that they do is they introduce you to the that’s when they introduce you to the 401K and I decided that, hey, it would probably be a good idea to figure out how these things work and how to do it properly. So I did a deep dive into stock investing because that’s primarily what they offer, and I gravitated towards the value investing school of thought that’s the Ben Graham’s, Warren Buffett’s and so forth. And I thought it was really interesting stuff. It seemed like a good way to study stocks and figure out how to buy things that were better than what everyone else was buying. One thing that caught my attention in 1997 was hearing that Warren Buffett had bought a lot of silver and was shortly thereafter persuaded to divest himself of it. But I found that to be very curious, and I started studying why it was that he bought silver.
And long story short, I got into the metals. Unfortunately, I think I got into the wrong metal for the wrong reason. But it led me down a path that allowed me to understand gold eventually. So I started reading articles. As you know, the way that things usually work out is when you’re looking for things to confirm your hypothesis, you tend to find all kinds of things. And I stumbled upon some work by Ted Butler, who had even compiled all of his articles into a book. So I read it, and I was thoroughly convinced that I was right. Unfortunately, later on, I found out that Mr. Butler had missed a bunch of fundamental things. For example, there’s a lot of silver out there on the market, and it is being lent, not necessarily shorted to hammer things down, but it’s being lent by bullion banks on behalf of their customers to make a profit. But anyway, as I went further and further down the rabbit hole, I started learning a little bit more about central banking. I learned more about the world of gold. I learned about the LBMA. I learned how things work on COMEX.
I pretty quickly figured out that gold was one of those rare things that the common man could own in similar quality to what the rich families use in order to preserve their wealth during times of crisis. They can’t afford fine art or the choicest land or antique furniture in the same quality as what the rich people own, but they can own gold. So over time, over the span of the 2000’s I not only accumulated stocks, but I started accumulating gold in addition to silver. And towards the end of I guess it was the beginning of 2011, I noticed something very curious, and that was the emergence of a lot of silver oriented YouTube channels where self proclaimed experts in silver were talking about how easy it would be to get super rich buying silver. And they were talking about the rules that they follow how to buy silver lunar coins from the Perth Mint, how eventually we’d get back to a point where one silver dime would be equivalent to a day’s wages, and the price of silver was going absolutely parabolic. And that’s when I realized that it’s probably time to lighten up because there’s absolutely too much speculative activity in it.
And I managed to and this was fortunate, it was just luck on my behalf but I managed to dump most of my silver in exchange for gold. When the gold to silver ratio hit about 35, I think I got a little bit out at 32 but it was just dumb luck. But I’ll take dumb luck, I’m fine with that. And I just held on to the gold throughout the rest of the 2010 decade and added a little bit to it as time went on. But I continued to study and that’s when I learned a lot more about central banking and the relationship between gold and the value of currencies like the euro and the dollar. And also took a deeper dive into some statistical analysis around mixtures of bonds, stocks and gold. And to this day I basically follow a lot of the conclusions that I came up with on how the three assets relate to each other. So it was last year that I reached a point where I was able to retire. So I had reached financial independence. My company had merged with another competitor a couple of years before and the cultures really didn’t get along together so it became an unpleasant place to work.
So it seemed like a confluence of events kind of pushed me out the door. And that’s when I started putting out chapters in my book. I was providing them for free on my website but I had to take those down recently due to a contractual obligation with Amazon when I actually did publish my book. But the website is still there with free articles and still producing videos on belangp on YouTube. And so that brings us to the present.
Keith: I was going to say I have two things to add. One is I had my own thoughtful disagreement with Ted Butler. He had put out this article, this is years, maybe a decade or two after you counted them in which he was taking the tone and the approach of like a professor emeritus who is an old man now and has made his contributions to the world, won his Nobel Prize, taught all the influential professor emeritus in the field and all the other top universities and now in this magnanimous gesture saying hey, here’s my body of work and here’s my contribution and if anybody has any thoughts or disagreement I’d love to hear it. And so I wrote an article. Okay, in that spirit, let me pick you up on that and first let me attempt to summarize your argument that’s over manipulated using terms and framed in a way that you would have to agree with and then based on there, let me criticize the idea and by that point we licensed what’s called the Tick History Database from then Thompson Reuters now Refinitive that contains every bid and every offer on gold and silver, both spot and every futures contract going back to 1995 with sub millisecond resolution.
Terabytes of data. So we first had to build a big data platform. You can’t just show terabytes of data into a sequel database there you go. And then built our own data science platform to start analyzing it and normalizing and cleaning it up. And there’s a lot of different data issues that you went into there. Anyway, by that point, we have the data to show what happens when every futures contract matures. And I said, in my thoughtful disagreement with Ted Butler I said, whenever scientific theories compete, what scientists look for is an edge case where the two series predict the opposite behavior. If Einstein says that when we find a super duper, pulsing magnetic object in the distant universe, it’s going to bend light to the left. And suppose Lorenz said it’s going to bend. His theory would predict it would be light on the right. All of physics are sitting on the edge of their seat, waiting with bated breath for when we can finally build a telescope big enough to go find one of these objects. And then let’s say we put the Hubble Space Telescope out there. Then everyone’s like, chewing their fingernails down to stumps.
What is it? Is it left to right, left or right. And then when the answer finally comes in, it’s left. Then Lorenz’s theory is relegated to the ash heap of history. Einstein is vindicated and life goes on. We have two theories here. One is that the banks are naked short in the futures. The other is that the banks are arbitraging between spot and features, and therefore you lead to opposite predictions. First of all, how this whole thing behaves, and then particularly when each contract matures. What would you predict from each of these theories? I lay that out and I said, now, here’s the data on 130 contracts as everyone that has expired all overlaid on top of each other. You can see what happens to the silver basis and the gold basis as you approach contract expiring. And therefore, the truth is definitively that it’s not naked short manipulation. It’s arbitrage. And anyway, with all that data and what’s interesting is that article was covered by everybody. It was on zerohedge. It was on GATA, it was on the MetropolCafe. It was on every possible site for gold bugs and alternative investing in alternative finance.
I’m sure that Mr. Butler woke up that morning with 50 or 100 emails. Did you see what that guy Weiner said? What are you going to say? What are you going to say? It was every place that he would have had to have been reading, and certainly his fans and followers would be reading. Dead silence. Not a word
Paul Belanger: I personally believe that there is a certain amount of manipulation that happens on the precious metals markets, but I don’t think that it’s evidenced by pointing at JPMorgan and saying, hey, they’ve had 400 trading days without a single loss. I think that’s just a sign that JPMorgan is doing bullion banking as opposed to trading. But you can see during thinly traded hours, occasionally a big trading house will try to run stops. So a common practice of traders is to just use technical analysis to explain everything, whether it’s bond, interest movements, stock movements, precious metals movements, commodity movements. And these technical traders do trade that way, and they put in trailing stops. So one of the easiest ways, I think, for a lot of the big trading houses to make money is to program some algorithms to seek out where these stops are during thinly traded hours and to try to run them. And then the computer based trading will basically take over once those stops get hit, and it will drive the price down, and it allows the big trading companies to basically take advantage of what I would consider to be a fairly unsophisticated trader.
Keith: Yeah, they’re front running. There’s all kinds of things there. I mean, JPMorgan, one of their traders was criminally convicted. Now, it wasn’t running stops, but it was front running or spoofing, putting out fake orders that they didn’t intend to honor. I’m not sure how the heck a court would determine that versus a market maker that’s arbitrage between GLD, futures, spot, forward market, all kinds of different markets. And then maybe other ETFs have been making a market in as well. How they differentiate any one move in any one market would necessarily necessitate changing all the bids and all the offers and all the other markets. How the court would determine that versus an order that was put in without the intention to fill. But the guy was caught for that and criminally convicted, and everyone says there’s a proof of manipulation, which I say Motte and Bailey fallacy, sir. Motte and Bailey fallacy is when you have an uncontroversial, but fairly useless, usually fairly smaller, trivial claim, such as somebody or something bad. You know, the Al Capone kid stole a candy from the five and dime store and then using that to freely migrate to. And therefore, Al Capone is the godfather, and therefore the price of gold would be $60,000, but… it would have been $60,000 decades ago but for this suppression, the stark cabal that for decades keeps the price tens of $1,000’s below where it should be. Anyway. So the article that I wrote in response to Butler doesn’t address everything that addresses. Specifically, if they held this naked shark position, what would happen and expiring, which is one of the leaving claims that they make.
Dickson: I think it’s fascinating. I don’t think there’s too many investors in precious metals who can say that they got their start in precious metals from reading the likes of Benjamin Graham and Warren Buffett, right? I mean, typically when you think of Warren Buffett, you think of this vocal opposition to owning gold and silver. But it’s interesting to me that that’s actually how you got introduced to those markets. I’m curious if you can fill in the gaps for us. So you’re reading about value investing. You see Buffett make his silver trade, and then that’s how you get into the metals yourself. But when did you start incorporating this kind of data driven approach? Because you’ve produced all these videos, you’ve got just gobs of data and research that you’ve done on portfolio composition and how gold and silver can occupy a role in a diversified portfolio. Connect the dots for us a little bit there.
Paul Belanger: Going back to Buffett, one of the principles of science is that when you find something that doesn’t make sense, there’s an opportunity there to dig in and figure out whether your theories are wrong or if there’s a data point that’s wrong. And so I think it behooves anybody that has any kind of a scientific inclination, anytime they see something that contradicts what they believe is true, they really ought to pay attention to that and try to dig a little bit deeper.
Keith: I just want to just pause there and say… this.
Dickson: Highlight and underline that comment.
Keith: Yes.
Paul Belanger: As I went through time and studied this, I came across a lot of different theories for how the world works, at least the world of economics. I mean, everybody has their models. Models, they make people comfortable because they tend to assign numeric values to things. But going back to my engineering studies, what I’ve concluded is that oftentimes the models that people form will reflect their own preconceived notions. So models are only as good as the assumptions that go into them. Right. And so usually, especially, the more elegant the model, the more you’ll find that it’s going to give back to you exactly what you think is going to happen based upon the inputs.
Dickson: When this discussion comes up. “All models are wrong, some models are useful.”
Paul Belanger: That’s right.
Dickson: I was thinking the same thing. I was wondering who said that.
Paul Belanger: I don’t know if it was Feynman or not. I seem to remember it coming up in Statistics for Experimenters by Box, Hunter and Hunter.
Anyway, going back to theories, so the only way to really test the theory is to use data. And you’re never really going to confirm that the theory is true, but you can certainly invalidate the theory. I mean, there’s nothing that ruins a beautiful theory, more than one data point that disagrees with it. Right. And so I tend to focus on coming up with as much data as I possibly can. And whenever it tells me that there’s something inconsistent with the way that I view the world. I tend to go back and examine my hypotheses and try to figure out where it might be wrong. So all that said, whenever you see all the data that’s been compiled in my videos and my writings over the years, one thing that you will probably notice is that there are a lot of conclusions that I drew early on when I was publishing these videos that I don’t really hold to anymore. I just haven’t taken those down. So it’s kind of like a giant, decade long stream of consciousness that’s led from the very beginning up until where it is currently.
Dickson: That’s great. I’m wondering, actually, just for the benefit of our audience, if you could spend some time and maybe walk us through some of your greatest hits over the years. Right. Some of the big picture questions that a lot of precious metal investors have. You’ve kind of attacked those head on using a lot of data and have produced really solid answers that I think are a real benefit to the precious metal community that’s looking for the answers to those kinds of questions. So I got a little list here. This is what I consider to be your greatest hits. You can add or subtract to this list, or we can spend time discussing something else. The first one I have here is, you note the problems with using Monte Carlo analysis. So why don’t you unpack that one for us?
Paul Belanger: Okay. Well, as I said, people tend to like models. They’re comforted by models. And especially when experts put together models, it tends to lead people to believe that the models are credible and you can do something useful with them. Now, a few researchers back in the 1990s, chief among them was Bill Bengan, who was first to publish this information, took a look at stocks and bonds, and the main question he was trying to figure out was how much could a person withdraw from their portfolio each and every year without running a major risk of running out of money in retirement? He had compiled a lot of data through the 20th century, at least the data that was available on stocks and bonds. And he concluded that most of the financial analysts or the consultants were incorrect because they were saying that because the stock market had returned 7% real on average over those few decades, people could feel safe withdrawing 7%. But he knew that that wasn’t the case because the stock market is volatile, and if you sell during a bear market, you’re going to end up selling more of the stock than otherwise you would during an average year.
So he tested the theories at the time by compiling decades long periods of data and concluded that, no, the answer was closer to four. Now, of course, that was a historical look. It’s not a futuristic look. But unfortunately, if you take a look at the data that he had available to him he was trying to answer, how could you make it so the portfolio would last you for 30 years? But if you’re going back to the 1920s and he wrote the paper in the 1990s, how many independent 30 year periods do you have? You really only have two, right? That was one of the key limitations of the work that he had. But a lot of financial analysts had basically globbed onto his work and the paper that was written a few years later by researchers at Trinity College, which by the way, they never even cited Began, which was a really weird thing. So a lot of financial analysts now they’ve extended that work and they think that they can use a tool called Monte Carlo analysis, which is where you take the statistical properties of the various asset classes. So they started with stocks and bonds, and then they included other assets such as commodities, gold, real estate and things like that.
And the theory goes that you can take a particular asset mix, do a lot of simulations on the computer using statistical distributions that have been collected over time, and come out with a precise answer for how long or a particular asset mix will last with a given withdrawal policy. Makes people feel very comfortable because the computer simulations will spit out answers like, oh, you only have a 5% chance of running out of money after 30 years. You only have a 2.1% chance of running out of money after 30 years. So it makes people feel very comfortable, very confident, so they can basically say to their boss, okay, I’m leaving the workforce. Now the 4% rule tells me that I’ll be safe for the next 30 years. So I have taken a look at that because I was kind of curious about what would happen with mixes of gold in stocks, because some of my research had indicated that a gold/stock mix was actually more predictable or more consistent in performance over especially the difficult decades of the 2000 to 2010. Then the stock bond portfolio was and I used my own Monte Carlo analysis to analyze what would likely happen to various mixes for a person entering a 30 year retirement.
And I found a couple of data points that were inconsistent with what some of my other work had said the results should be. And so I dug deeper and what I found out was that one of the key assumptions behind Monte Carlo analysis is critically flawed. And that is the concept of The Random Walk. Are you familiar with what a Random Walk is? Yeah. Okay, so a Random Walk basically says that future performance is completely independent from current performance. In statistics, you refer to that as independent variables, time invariant. So it says that, all right, if you have 7% return in stocks this year, you can’t use that to predict what’s going to happen in stocks next year or the year after that. So no amount of past data will tell you what’s going to happen in the future. And to a certain extent that’s true for individual asset classes because the markets tend to be somewhat efficient. And so if you could use the past in order to predict the future, everybody would do it. And I believe that a lot of people do try and in so doing when they try they tend to remove any predictive capabilities because it tends to be priced into the market at any point in time.
But the key assumption in Monte Carlo is basically that you can take standard deviations of returns and cross correlations between assets and use that to predict what’s likely to happen in the future. So you do your simulation for this year, then you do another simulation for the following year, then you do another simulation for the year after that and none of it is dependent upon anything else. It’s a true random walk, just like a drunk walking down the street. You don’t know which direction the drunk is going to go, but maybe he’ll manage to meander down the street after successfully bouncing off a couple of buildings. But the problem with that is that the correlation between stocks and bonds actually tends to grow over time. So if you hold stocks for one year, the correlation between them, the Pearson r correlation, is about .3. So that’s for a one year holding period, and for stocks versus gold, or bonds versus gold, it’s basically zero. So there’s almost no correlation for a one year holding period. But if you look at a five year holding period, or a six year holding period, or seven year holding period, the correlation between stocks and bonds actually becomes fairly high.
And the negative correlation between stocks and gold or bonds and gold is actually very negative. So we’re talking about .7 between stocks and bonds and minus .8 between stocks and gold. And what that says is that if you have a long stretch of time when bonds or stocks are doing better than average, there’s a pretty good probability that gold will do worse than average and vice versa. It’s almost as if gold is the forgotten asset when bonds and stocks are doing well. But then when stocks and bonds take a turn for the worst, all of a sudden everybody gets religion and they start paying attention to gold again. And so we get periods of time like the decade of 1970s.
Keith: Which makes sense from a theory perspective. Gold being money. It’s the thing you hold when there’s no other investment or no other speculation that you prefer based on a risk return analysis. The gold is the thing you own when there’s nothing else you want to own. Which is exactly what you’re observing in this negative correlation. When stocks are going gangbusters, everybody wants to own stocks, they sell the gold to buy more stocks, right.
Paul Belanger: Except they do it in the rear view mirror. Until early this year, stocks had insane valuations and bond yields were very low. But you didn’t have many people coming out of the woodwork and saying, buy gold, you should stay away from stocks and bonds.
Keith: Right.
Paul Belanger: These guys what’s the old saying? I think it was Upton Sinclair said, you’d be amazed at how much people can ignore when their livelihood depends on them not knowing it.
Keith: It’s impossible to teach somebody something that his salary depends on who’s not knowing. I was thinking of was Roy Rogers. The key to making money in the stock market is to buy the stocks that go up, and if they don’t go up, don’t buy them.
Paul Belanger: Yeah.
Keith: Rearview mirror.
Paul Belanger: Anyway, back to Monte Carlo, because of that strong negative correlation between gold and the financial assets and the strong positive correlation between bonds and stocks, I think people are misguided by using Monte Carlo from doing any kind of prediction, because you will end up with decades such as the 70s or 2000 through 2010. And Monte Carlo is an invalid tool for trying to predict that kind of thing if you don’t include that kind of information in it.
Keith: That’s a problem with data science more broadly, right? If you don’t understand causality of anything, just looking at data, it’s so easy to trip yourself up. It’s so easy to make a bunch of assumptions. You know, maybe a few of them aren’t quite right, but whatever, and you just file on forward and then you end up with a conclusion that, because it comes out of a computer, seems like it’s got a lot more weight than we actually should have. By the year 2020, the Earth will be so hot, it won’t be able to sustain life. That sort of screaming headline that makes when somebody was doing efficient gravitas denounces it. But what do you really have at the end of the day? Garbage in, garbage out, right?
Paul Belanger: That’s what, personally, I think makes the Federal Reserve right now so dangerous in their policies, is that you have a lot of PhDs who have studied data in the past, and they’ve come up with some very elaborate models, but what they haven’t done is they haven’t built in the capacity of human psychology to change. And so I happen to believe in an old thesis of Ludwig von Mises, which is that for things that are economic, you’re almost best off not trying to quantify them, because it’s driven by people’s psychology more than anything else.
Keith: And the fundamental uncertainty of the entrepreneur. You think you know what all the jobs are that we need, and then the entrepreneur creates a new industry, and suddenly there’s 100,000 jobs that “we” didn’t think “we” needed six months ago and the Fed is trying to manage this. One of those anomalies that nobody ever seems to ask the question. Remember earlier when I said this, you have a theory, right? And then you go out and look at the world. And when the world contrasts to theory, you’ve got to ask them questions. One of those questions that nobody asks is is Quantity of Money the singular cause of rising prices? And so as early as at least as far back as Newt Wixil in the 1890’s. Wixil went out to try to prove that because he was a moneterist. And proved that actually there isn’t really a correlation between quantity and prices, the correlation is between interest rates and prices. Rising interest rates correlate with rising prices, falling interest rates correlate with falling prices. Later a guy named Gibson comes along and then they name it Gibson’s Paradox. Paradox only so called because their theory doesn’t get it. It predicts it wrong. This whole thing seems to be forgotten back to today where, okay, we have inflation and therefore the Fed should hike interest rates.
Two problems with that. One is the cause of rising prices today is clearly disruptions caused by trade war, green energy restrictions and the lockdown and the consequent unlocking. And now finally Ukraine war. Ukraine was one of the biggest exporters, both of fertilizer and grains. And that’s offline this year. We’ll see about next year. But for this year that’s gone offline. What does that do? Prices obviously skyrocketing, but even from a monetary perspective, if you have higher interest rates, that means higher input costs to every producer in the theory this is going to cause lower prices. Anyways, if you look at the 1970s, actually 1950s through 1970s, you see a great correlation between rising interest rates and rising prices. If rising interest rates was supposed to correct all this why did inflation just keep accelerating? And then the moment that inflation is supposedly broken, we have a falling interest rates from 1981 arguably through present. Although right now there’s a little correction. It’s one of those things where the theory doesn’t match the observed data and so therefore people just ignore the data. That’s much easier than challenging the theory.
Paul Belanger: Yes. And of course, as prices rise, the real danger is when the average everyday person starts to expect continuous price rises and ever increasing rates of price inflation and they start taking action accordingly. And that tends to feed on itself and it becomes a positive feedback loop.
Keith: That was the 1970’s in a nutshell. My parents got to the point where, and a lot of people did I don’t think there’s anything neat about my household, would go grocery shopping and then whatever is on sale. It was terrible. Like, let’s say can’t the tuna fish or paper towels or toilet paper or dishwashing detergent or whatever, we just buy enough for a couple of years, right? And I use the term they traded a bank balance for a pantry balance. That’s because in the pantry were of known and predictable value and cash in the bank was not. Then the real danger. That’s positive feedback loop number one, positive feedback number two, which feeds on that and makes it work as corporations get in on the act because they sell bonds to finance ever increasing hordes of raw materials and work in progress in between every stage of production. So if you go 17 steps from buying your raw materials to a finished product, you have increasing buffer filling up a warehouse in between each of those 17 stages. And then, of course, the buffer of finished goods. The longer you go from buying raw materials to selling finished gold, the greater your profit and nominal terms.
Paul Belanger: One of my favorite articles of yours, Keith, was actually the one where you talked about why Radio Shack did as well as they did in the 1970s because of their inventory, right?
Keith: And then they start to get into real pressure in the 1980s when the inventory becomes a problem and not a solution. Right? So then the corporations are selling bonds, which means pushing interest rates up in order to chase prices up. So you have chasing interest rates up. The more rates rise, the more they’re doing it in order to buy. The more commodities, more commodity prices rise, the more they’re doing it, which is I don’t think that we’ll see, but I don’t think that’s the dynamic we have today at all. This whole elaborate extended point to corroborate what you’re saying about confirmation bias and models being wrong and assumptions to the models being wrong. And ultimately the Austrian school, Menger talks about this, and a lot of people think he’s anti scientific and saying that economic knowledge is a priori and not empirical. And of course, modern philosophy traits this dichotomy, it’s a Kantian dichotomy of nominal versus phenomenal. But it’s either pure analytics has nothing to do with reality, or pure lived experience without any resort to reason or principles of any kind. Menger is saying, I think Menger being Aristotelian very clearly to me.
When you read his method, he is so Aristotelian. He’s saying that you have to understand economic principles from the causality of it. Suppose two guys come to market, each of them has wheat that they want to sell to the market. He’s debunking the idea that there’s a “right” price of wheat that all the previous people would have calculated based on the number of people and the number of wheat bushels being produced and saying that a wheat bushel is supposed to be one ounce of silver or whatever it is. He said clearly, if two guys come to market both having wheat, there will not be a trade. And that’s obvious, right? And then he’s going into why it’s not just because you have wheat and I have wheat, so have a nice day. But my ask price on the wheat is above your bid price. You’d potentially be a buyer of more wheat if you get it cheap enough, right? Then you could bring double the wheat to market. So anyways, then you get into bid and ask and you get into the spread being inverted.
And it’s in neither of our best economic interests. If we’re economizing actors to trade my wheat for your wheat, what’s the point of this? Both of us are going to look for advantage at the end of the trade. I end up with a bushel and a half. Then I would do it. But of course, why would you? Because then you end up with only half a bushel. He’s emphasizing you have to think this through from first principles, which becomes very, very difficult when you’re talking about the entire economy. And so then the macro guys love to write a differential equation that purportedly describes the economy. I’ll never forget a discussion in Fekete’s classroom when you’re talking about this idea of an equation. First of all, he said you have an integer number of actors. It’s a whole number. It’s not down to any fraction. So we just use difference equations, not differential equations. That would be the first problem. Second problem is people have free will. They’re not particles of an ideal gas. And then you have the entrepreneur. But even then, I don’t know if he said this or if I said this.
Suppose you have two steak dinners and two hungry people. And so all these equations would say, I think I brought this up as something as a problem with this whole equation thing in class, that the equation would say, okay, fine. You have two steaks to satisfy two people, everything’s fine, right? Well, what if you have a fat guy in a restaurant eating two steak dinners and a starving guy looking through the window? These equations just can’t tell you anything like that. And so the central planner is due to fail before we even get started. And not to mention, the models are wrong, the assumptions are wrong. The distributions are never Gaussian. Wasn’t that essentially the bottom line for why Long Term Capital Management blew up. It’s a power curve, not a Gaussian curve.
Paul Belanger: It also explains why gold coin and or silver coin basically went out of circulation during 19th century America. It’s because government tried to fix the price of one relative to the other by stamping the number on the coin.
Keith: Right. Whichever one’s undervalued comes to market and whichever one is overvalued, it’s either hoarded or gets sold in global markets where you get a better price for it. Yeah, that’s right.
Dickson: Despite the terrible historical track record of price fixing, it seems like we’re still having a go at it. I saw a headline last week that they’re talking about putting price controls in for gasoline for fuel. We’ll never learn.
Paul Belanger: Back to the 70s, right?
Keith: I for one look forward to odd and even license plate days! ;)
Dickson: I want to go back to something that you kind of glossed over, Paul, because I think it’s actually a really key point. When you were talking about the problems with Monte Carlo analysis, you said that as it turns out what your research shows is that a stock to gold portfolio, I think it’s 65/35, or I can’t remember the exact proportion, but actually when you had gold in the portfolio, or if you replaced bonds with gold, you actually have better performance. Tell us a little bit about that, how you came to that. What are your findings there on what role can gold have in a portfolio and how it helps performance?
Paul Belanger: Okay, well, better performance I think is a little bit of a problematic phrase. So what I would say is that gold tends to be able to make things more predictable, which is to say it has. Everything that we’re talking about right now is based upon past record. It doesn’t mean that it’s going to hold. But my feeling is that without taking a look at the past, how are you going to make any kind of decisions about what the future might hold? So if you take a look at the past data, and I start with 1971, because that was the seminal event that completely changed the nature of what the dollar was and what bonds are. I looked forward from 1971 all the way through 2021, and what I found was that the inclusion of even a little bit of gold to a stock portfolio would tend to make year to year performance a lot more even. And what’s more important is that it tends to make stretches of five to ten year performance much more consistent. And that’s a little bit paradoxical because everybody points at gold and says, hey, look at the volatility of gold, it’s all over the place and it doesn’t really produce a return.
So why the heck would you want to own this? And the answer really goes back to what I was saying, which is that people tend to find religion when the financial assets tend to do poorly and they tend to focus on gold. I think over the past, gold has at least allowed people to stay whole when their financial assets are falling apart. I think it behooves everybody to put at least a little bit of gold into a portfolio. And I happen to know that a lot of wealthy families always have done this throughout history because they know that there have been periods of time when financial assets just basically fall out of bed and the only thing that they can count on to preserve family wealth is find real estate, collectibles, art, antiques, and gold. Right? So what I found was that the most consistent performer over the past 50 years was a mix of about one third gold and two thirds S&P500. I also took a look at foreign stocks and the relationship wasn’t nearly as clean. I mean, some of the conclusions did still hold, but I think part of the problem was that I was taking a look at performance in US dollar terms, and capital flows tend to partially be dictated by how well the country’s stock market is doing.
When you reduce the amount of gold in that portfolio, the average performance did happen to go up. Average performance meaning average yearly real return, but the volatility got a little bit higher.
And so I think the sweet spot for most investors is probably in honestly the 10% to 20% level. And the reason why I say that is that although I found that about one third gold seems to be the most consistent, I don’t know that a lot of people have the temperament to maintain that kind of asset mix. Other problems that tend to come up are the frictions of rebalancing. When you sell stocks or sell gold, you’re obviously going to be paying a bid ask spread, and especially if you’re involved in physical metals, that bid ask spread can be pretty punishing, especially in silver, as you know. But there’s also taxes, right? So there are certain frictional costs when you do the rebalancing, which is another reason why I think that a lot of people are probably better served with a 10% to 20%, and I think they’re probably best served by not rebalancing every year. Because again, as I said, the correlation between these assets tends to become more and more negative with longer holding times. And so doing a rebalance gradually, for example, if you had a 15% target allocation of gold, you wait to the end of the year, you find out how much you’re off, and maybe you rebalance it 1/3rd of the way.
Dickson: Oh, I see. That’s what you mean by gradual rebalance. So you don’t complete the entire okay. You don’t go back to the target immediately. You kind of scale into that target over time.
Paul Belanger: A bit of a more gradual rebalancing. Another approach that I propose is that for a person who’s accumulating assets, if you have new funds that you want to deploy, maybe you just use those to buy the asset that’s under your current allocation. And similarly, when you’re on the other side of retirement and you’re drawing down your portfolio, just sell the one that’s over its allocation, and that way you can minimize the trading frictions that I was talking about.
Dickson: Right. When I hear that, it always just makes me scratch my head a little bit, because when you look at going back to your original point about 401K’s, retirement accounts, and when you think about mainstream investing, the 60/40 portfolio, even though it’s, I think, coming under a lot of heat recently, still is kind of the mainstay for that world. And it’s just like gold has kind of been relegated, just kind of pushed off the playing field. Why is that? Because the data shows something else. Right? The data shows that there is a role for gold.
Keith: I chuckled slightly when Paul, you were talking about 30%. We did a white paper on that 60/40 portfolio. And what happens if you add I call it a little slug of gold to it. I’m not yet going to reveal what I mean by little slug. We found that by putting gold in the portfolio and we did probably similar analysis to what you did. We just went back to data, all the data going back to 71 for stocks and bonds, S&P500 and the 10 year treasury specifically. And we found that by putting that little slug of gold, you got lower volatility as measured by the sharp ratio, smaller drawdowns. So if you’re an institutional investor or retiree, the drawdowns are what you’re really worried about. Like above all else, don’t have big drawdowns, right? Yes, you want big upside too, but the fact that you might have a big upside next year if you have a big drawdown this year is really bad.
Dickson: There might not be a next year.
Keith: For the amount of gold that we used, the enhancement of the return wasn’t enough to really write home about. But anyways, by little slug I mean 4%. Not that we didn’t think people could or should go for more than 4%, but rather we felt that we would not come across as a credible white paper in institutional circles. If we said more than 4%, we would start to look like a little wild haired and a little willly to be talking that’s crazy talk that’s outside the Overton window. So we just left it at 4%. And I think, did we mention, Dickson, that if you go with a greater percentage of gold you get even better improvements? We just had a little footnote or something.
Dickson: Yeah, no, you’re right, we did.
Keith: And moved on and trying to make the case for 4%, and then obviously our case was with yield on that 4% of 3%, then it’s a game changer to the portfolio over a time period of 50 years, like 1971 to present over that kind of time frame. If you’re given 3% on your 4%, then the portfolio really takes off relatives to the 60/40 standard portfolio. That was the point we were trying to make anyways. But anyway, it’s just interesting how even in the gold community, if you wrote on KITCO or ZeroHedge or a similar site and said, I recommend 30% gold in your portfolio, do you just take yourself out of serious consideration? I don’t know. I don’t know if you’ve said that publicly or what the response has been. Whenever people ask me about it, I just said, look, I can’t give financial advice. I just think everybody should have some and leave it to them as to what they should say, which is probably appropriate given my position as CEO of Monetary Metals. I’m curious if you have said anything like 20 or 30 and how people responded if you said that.
Paul Belanger: Oh yeah. In my videos I’ve shown how 35% 65% stock and gold portfolio has basically beat the pants off of a 60/40 stock bond portfolio over the past five decades and was more consistent in terms of performance. It did not have a decade of bad performance in the 70s or again in the 2000s. It was just pretty much up. I mean, of course there were swings up and down. You can’t have return without having some kind of variability. But I think most people did see that with an open mind. And fortunately for me, I’m a small guy without too much of a following, and I’m an engineer. I’ve had this thought in the past that the financial industry has grown to resemble very closely the medical industry. And when I say that, what I mean is that whenever a doctor is giving medical advice, he feels a lot of pressure to conform with what the vast majority of doctors in the medical field basically say. And if he deviates and something goes wrong, he opens himself to a malpractice suit. I think the same is true for a lot of financial professionals, especially the credentialed ones, where if they deviate from what everybody has accepted as true and something were to happen, they would either get sued or lose their job.
Keith: It’s okay if we lose 50% of your value in this bad year because our benchmark is S&P500 and S&P500 went down 51%, we outperformed our benchmark. If we did something that’s different and nonstandard and therefore we lose out a benchmark to hide behind, then what the hell’s wrong with you, you idiot? You lost 50% of your clients money. The regulators are going to shut you down, and the clients will become plain distance to you out of existence.
Paul Belanger: And if you have problems, you can always rely on your Monte Carlo simulations, right?
Keith: That’s right. That’s right. My back tested models. The thing I was going to say earlier about the back testing and the problem with data science, is that to your point about if you’re doing 30 year periods and you’re writing this in, let’s say, early 90s, so let’s say you start with 1925-1955 and then 1955-1985 you only only have two good period plus, whatever, another few years before you wrote your paper. Anything… and don’t even get me started on so called artificial intelligence. I had a few tweets back and forth with a couple of folks about AI, which is really nothing more than either training models to interpolate, hopefully nonlinear interpolation at least, or pattern recognition. Then the problem is if you have less data than you think you have, because either if you look at it, there’s not very many periods in it, or there’s a lot of correlation in your data because three variables that you think are independent are really actually two dependent and one independent. So you have less data than you think. For whatever reason, you can easily overtrain the model, and the model is going to give you really good fit.
Should 1925-1955 ever recur, you’ve got the perfect model to train you for that. So I guess if you went back in the time machine t0 1925. Man, you just got a great model for what to do. But in terms of predicting 2025 to 2055, not so much.
Paul Belanger: Yeah, and don’t forget, once in a thousand years event seems to happen about once every ten or 20 years right now. The unlikely event seems to be a lot more likely than the models say.
Keith: It’s not like every year companies seem to have non recurring charges.
Dickson: It goes without saying. But this was great. Paul, it was a treat to have you. We did not even come close to addressing all of the items on our list. So what that means is we’ll just have to have you back on for another show and we can address the other items on my list here.
Paul Belanger: That would be great.
Dickson: Yeah. Thank you so much. Hope to have you again soon.
Paul Belanger: Okay, well, thanks for the invitation, Dickson, Keith.
Keith: Let’s do that.
Additional Resources for Earning Interest on Gold
If you’d like to learn more about how to earn interest on gold with Monetary Metals, check out the following resources:
The Case for Gold Yield in Investment Portfolios
Adding gold to a diversified portfolio of assets reduces volatility and increases returns. But how much and what about the ongoing costs? What changes when gold pays a yield? This paper answers those questions using data going back to 1972.
In this paper we look at how conventional gold holdings stack up to Monetary Metals Investments, which offer a Yield on Gold, Paid in Gold®. We compare retail coins, vault storage, the popular ETF – GLD, and mining stocks against Monetary Metals’ True Gold Leases.
https://mises.org/library/carl-menger-nature-value