Watch this conversation on the Rational Reminder YouTube channel
Subscribe to the Rational Reminder podcast
Key Points From This Episode:
-
We touch on future guest Jennifer Risher’s book, We Need to Talk. [0:00:19]
-
Hear about the new Bitcoin ETFs and other cryptocurrency news. [0:03:15]
-
Introducing today’s investment topic; fixed income products. [0:07:31]
-
Approaches to building fixed income portfolios and forecasting expected returns. [0:10:16]
-
Exploring the factors that impact fixed income risks and returns. [0:15:20]
-
Using forward rates to predict your fixed income returns. [0:17:00]
-
Conflicting research on the power of forward rates to predict term premiums. [0:19:22]
-
Why forward rates do contain information about expected term premiums. [0:22:21]
-
What Barclays’ intermediate indexes say about fixed income allocation. [0:26:19]
-
The summarised formulas for expected bond returns. [0:28:33]
-
Evidence on why credit spreads have low explanatory power for default rates. [0:29:37]
-
The main takeaways on how we should view bonds and returns. [0:32:50]
-
Comparing fixed income with cap-weighted indexing. [0:34:57]
-
Why Dimensional Funds looks at credit spreads and yield curves around the world. [0:36:50]
-
Introducing today’s planning topic: expected return assumptions. [0:39:25]
-
How important expected returns models are to financial decision-making. [0:40:25]
-
Different models that are used to derive expected returns. [0:41:50]
-
Planning for short-term versus long-term predictability. [0:44:38]
-
The danger of using historical returns as the basis for your expected returns. [0:47:50]
-
Damodaran’s research on how different forecasting models perform. [0:48:18]
-
Insights into PWL Capital’s expected return models. [0:49:37]
-
We answer questions in our ‘Talking Sense’ segment. [0:53:56]
-
This week’s bad advice; incorporate Bitcoin into your retirement investments. [0:56:06]
Read The Transcript:
Ben Felix: This is the Rational Reminder Podcast, a weekly reality check on sensible investing and financial decision-making for Canadians. We are hosted by me, Benjamin Felix and Cameron Passmore, portfolio managers at PWL Capital.
Cameron Passmore: You nailed it. You nailed the new intro perfectly.
Ben Felix: Big change.
Cameron Passmore: We wanted to, due to popular demand, get an update on your BattleBots. What’s the latest development?
Ben Felix: I’m surprised, first of all. Anybody that’s involved in our Rational Reminder community would have seen the survey that Angelica sent out or posted, I guess, for people to participate in. She’s trying to do that to get a better feel for how to organize the content, largely, just based on who the audience is, so we’re collecting information for that reason but she also asked for feedback for us. One of the things that came up more times than I would’ve guessed was that people want to hear more about BattleBots. I always feel like people don’t want to hear about it, so I never want to talk about it, but anyway, this is the one I mentioned last time that we’re working on, the drum spinner. There it is there, if you’re on YouTube, you can see it.
Cameron Passmore: And then over your right shoulder is the printer we can see with that orange spool thing, is that correct?
Ben Felix: Same filming, yeah, that’s right.
Cameron Passmore: That’s pretty cool.
Ben Felix: It’s pretty cool. It’s pretty-
Cameron Passmore: But you fried the battery, so we don’t get to hear it.
Ben Felix: Yeah. Lithium polymer batteries are a bit finicky, so I can’t spin the drum up right now, but I’ll replace the batteries maybe in a couple of weeks, everybody can hear it. It sounds pretty cool.
Cameron Passmore: I know you followed a few of my recommendations for shows and I think I mentioned this one before, but the final episode came out last week. It’s called Your Honor with Bryan Cranston, of course, of Breaking Bad. Oh. The 10th episode, the last one came out, the last and final episode of the series, incredible. You got to watch it. Such a great series.
Ben Felix: We don’t subscribe to Crave. It’s 28 bucks a month or something, isn’t it?
Cameron Passmore: I think it is if you get your HBO package, I don’t think you need the HBO package but …
Ben Felix: Oh, yeah?
Cameron Passmore: Anyways, if you’re on Crave and you’re looking for another show to watch, it’s fantastic.
The other thing, two weeks ago, we mentioned a new mug that was coming out that featured past guests and their episode numbers on the mug. Unfortunately, we made a mistake and missed our very first guest and dear friend Carson, so we are getting new mugs printed and they will be available soon and we’ll be sending them out. That’s why we don’t have-
Ben Felix: They look great, though.
Cameron Passmore: They are great. They’re going to be great and that was a mistake, so we were fixing the mistake. The other thing too is we’re going to continue the offer of getting a free pair of socks with every order, so our good friend who’s in the sock business and a regular listener extended the same offer to us. The cost on the socks is low enough, we’re quite happy to throw in a pair for every order. We’ll keep that up.
Also, we haven’t done outs for a while. There’s a bunch of people who’ve been very kind to leave reviews. I know when I read the names, it makes you laugh, but there’s a lot of great people, some I can’t pronounce, but Modern Family, Chris Hazelgrave , Kalmi Boot Stick 16 , M. Bogle, Peter Pietro, Mariana 37, Jason Myerson, thanks to all of you for great and very kind reviews that you left online.
Ben Felix: None of those names were that funny. I hope somebody –
Cameron Passmore: The ones that were funny I shouldn’t say, like FP, I don’t know how to … Anyways, people know who they are and we appreciate it very much.
Ben Felix: Someone might do a review with an obviously ridiculous name just to hear you say it.
Cameron Passmore: Well, you’ve had some good ones in the past that you’ve had a good laugh at.
Ben Felix: Yeah.
Cameron Passmore: Pretty cool to see how many people linked up with me on Goodreads. I got over 20 people in the past two weeks, which is great. Love to see what other people suggest and it’s neat how Goodreads picks up on what you’re reading and what your followers are reading and makes really good suggestions. Also, over 30 members in the hashtag #rationalremindergroup on Peloton and we’re well over 2000 members in the community, right?
Ben Felix: Yeah, it’s coming up on 2,200, which is more than we ever expected when we started it.
Cameron Passmore: Yeah. And we have, is it a … We will have a poll going, right, inside the community board for CE credits?
Ben Felix: Yeah. When this episode is live, we’ll have a poll in the community regarding CE credits. Something that we want to do this this calendar year for the podcast is get it set up so that listeners can answer a quiz based on the podcast episode and get CE credits for having done that. There’s a whole process involved with doing that and some costs and there are a bunch of different ways that we can approach it so before we figure out which path we want to take, one of the things that we want to know from our listeners is how many of our Canadian licensed listeners, licensed financial professional listeners would be interested in having CE credits available for listening to the episodes and answering the quiz. We’ll put a poll up online to gauge that interest and if there’s enough, then we’ll pursue that as a project on our end.
Cameron Passmore: It’d be neat to gauge the difference between, say, [iRock 00:05:17] and the insurance side for credits. Perhaps we can include that in the survey too.
Ben Felix: Yeah. Yeah, definitely. Anything else to add?
Cameron Passmore: No, I think that’s good. We can go ahead with the episode.
Ben Felix: Welcome to episode 138 of the Rational Reminder Podcast.
Cameron Passmore: Quick books that we’ve actually both read and we’re interviewing the author very soon, it’s a book called We Need To Talk: A Memoir About Wealth by Jennifer Risher. She’ll be on the podcast in a month and a half or so, but I think it’s worth checking out the book ahead of time. It’s an interesting story about a couple that come to terms with their wealth as she and her husband basically became wealthy beyond their wildest dreams, very modest upbringing and they were both early employees at Microsoft in the nineties. And then, her husband left Microsoft to join a small startup called Amazon. Jennifer wrote the book not as an expert in money per se, but as somebody who wanted to get conversations around money going. I think the book does a very good job of that and talking about the challenges that they went through as their net worth increased and the impact that had on all kinds of people around them, including their social network, friends, work, family.
Ben Felix: It’s a very unique perspective because their experience is living in the 1%, not the 99% and you don’t often hear people in the 1% talking about, it sounds terrible, but the 1% problems and she talks about this in the book too, that everybody else looks at people that have wealth and think, “How can you have problems?” Or, “What do you possibly have to complain about?” Which makes it hard for wealthy people to relate to other people in a lot of cases. It’s a fascinating overview of their story, the challenges that they dealt with related to wealth, but it’s not just about them. She surveyed, somewhat informally from what I can tell from the book, quite a few other wealthy friends of theirs, which gave a whole other perspective because not all of the perspectives are the same.
And then, there are also quite a few references to studies related to wealthy people. The information content was really good and very niche. You don’t get to see this side of the story very often. It was interesting and I think our conversation with her is going to be good. It’s maybe tricky for someone, our job is to deal with wealthy families, we’re used to seeing these real problems, so hearing them describe from somebody who’s gone through it, it was interesting … For someone who’s not wealthy, I can see reading the book being difficult because it’s talking about problems that someone who’s not wealthy would maybe wish they had.
Cameron Passmore: Anyways, worth reading, I think. Speaking of Microsoft, I just started reading the head of Microsoft’s book, Hit Refresh by Satya Nadella. Really interesting book. Just started it, but he talks about how he wanted to change the culture as soon as he was given the position, awarded the position by the board. I’ll have more to follow up on that book in a few weeks.
Last week in Canada, first and second Bitcoin ETFs in North America were launched. It’s incredible what happened here. Again, another first for Canada, but I guess it’s due to some restrictions by the SEC in the US and they just haven’t been able to come up with an ETF structure that’s been approved in the US but Canada now has two and one of them is the Bitcoin, the Purpose Bitcoin ETF, BTCC is the symbol. And if you can believe it, they took in 165 million on the first day and the second day, over $200 million of assets. One commentator I saw today said they could be at a billion dollars by the end of week one.
Ben Felix: Yeah.
Cameron Passmore: Just an unbelievable bonanza.
Ben Felix: They’re listed in Canada, but like you said, in North America, they’re the only one for now so you can definitely see how they pick up a lot of assets quickly.
Cameron Passmore: Yeah. Purpose came out first, then the next day, Evolve Funds came out with their Bitcoin ETF, EBIT E-B-I-T, which also trades at the TSX. Couldn’t tell how much they picked up in assets, but the articles I read suggests that Purpose being out first just gobbled up a ton of ton of money. What’s interesting about them, why they’re so popular is, based on the ETF structure, they should be trading much closer to net asset value than some of the comparables in the US. This has been an often cited issue with the closed end Bitcoin units in the US. You’ve got that spread between what it trades at and what that asset value is. Pretty incredible. Speaking of Bitcoin and the rising price of Bitcoin, last week, I hit what? $50,000 per coin?
Ben Felix: It’s about that now, yeah. It’s even gone higher.
Cameron Passmore: Total market cap of Bitcoin, I think, is over a trillion dollars. We came across this article that was citing a study from Cambridge University that Bitcoin mining, the energy to do the Bitcoin mining has taken over the country of Argentina in terms of being inside the top 30 countries using the most energy. It’s between Norway and Argentina, if Bitcoin was a country for energy usage, which is just a mind-boggling statistic to me.
Ben Felix: Yeah. It is. I saw Janet Yellen today, the day that we’re recording, on the 22nd, she was out making comments about the inefficiency of the technology for monetary transactions. She was also talking about its use in illicit activities and then there were people that saw this on Twitter and there were a lot of people jumping on that saying, “Well, what is the number one currency used in illicit transactions? It’s US dollars.”
Cameron Passmore: US dollars.
Ben Felix: Paper dollars, I don’t know how legitimate that is. The thing that I find just amazing about the Bitcoin phenomenon is it’s become cultish. There are people who believe that Bitcoin is the absolute future and if you don’t own it, you’re going to be poor and that’s it. There’s no questions asked and it is stupid. I don’t like using that word, but that’s the perception that you get if you look around people talking about Bitcoin online, you’re basically an idiot for not owning it. Whatever your reason is for not owning it, you’re stupid, that’s how it comes across. It’s going to be interesting to see how it all plays out.
Cameron Passmore: But this trend of many people saying to have 1% of your portfolio in Bitcoin, that’s the common wisdom that’s often-
Ben Felix: In hindsight, that would have been great.
Cameron Passmore: Absolutely.
Ben Felix: Because everyone says, “Oh, yeah, it’s a hedge against something, whatever, so you should have 1%.” And, hey, yeah, look at that. It actually did work as a hedge. That’s cool and having just 1% worked really well because it increased so much. Great. Is there any reason to expect that in the future? I don’t know about that. I don’t know about that. I talked to, this is a casual conversation that I had with somebody who was involved, very involved in the cryptocurrency space early on. I’m not going to go into the details, but their outlook, they were there from the beginning, involved in Bitcoin in a major financial center. All the transactions and the type of people that were making transactions and their outlook from here, it was pretty bleak. Not that that means anything, their speculation is no more valuable than mine, but it was interesting to hear that perspective.
Cameron Passmore: We shall see. Our investment topic today, kind of the opposite of Bitcoin, I guess.
Ben Felix: Yeah. Some people are saying that Bitcoin should be used instead of bonds, Bitcoin or gold. They’re kind of the same thing, Bitcoin and gold, to hedge.
Cameron Passmore: The question about bonds and fixed income has come up a lot lately. Of course, why would I ever invest in bonds now and what about negative interest rates? It doesn’t make any sense to own bonds.
Ben Felix: Yeah. One thing I do want to say before we get into the topic and I picked this idea of, it’s not even a new idea, but I gave a talk to a medical school faculty recently, and they asked me, at the beginning to disclose any conflicts of interest. It’s kind of a big sense to do that, actually, somebody in the Rational Reminder community, when we had David Blanchet on, they were speculating about conflict of interest that he might’ve had and while he was talking about annuities.
It’s a nice idea to, before talking about something, disclose any conflicts of interest so I was going to do that. We’re talking about fixed income. The evidence that we’re going to talk about related to fixed income is all used in the dimensional fund advisors fixed income products that we use for our clients, so there’s my conflict of interest disclosure.
We’ve, I think, done a pretty good job. Since Rational Reminder started, my sense is that there are a lot more people in the world who have an understanding that different types of stocks can have different expected returns. Maybe it’s a bit of a stretch to say that we’ve influenced the world in a meaningful way, but at least the people that listen to the podcast, I know a lot of them, prior to participating, didn’t know anything other than cap weighted index funds and then you hear about, oh, small cap in value and all these other systematic risk factors.
Now, the same thing, or a similar thing exists in fixed income and it gets discussed even less than factor investing in stocks. There’s, again, Fama and French paper from 1993, common risk factors in the returns on stocks and bonds, it’s the Journal of Financial Economics paper and they identified their two sources of higher expected returns in fixed income. Those were term and credit and as far as I know in fixed income factor models, these are still the workhorse factors that explain differences in returns. If you go on the portfolio visualizer website that lots of our listeners like to play around in, they have those factors available to use in regression.
Cameron Passmore: Term being the amount of time until maturity?
Ben Felix: Yeah. Longer term to maturity, bonds with a longer term to maturity have higher expected returns than shorter and riskier bonds, so lower credit bonds have higher expected returns then safer bonds. That’s the term and credit premiums. Now, given that information, given that there are differences in expected returns in bonds, one approach that we could take to build fixed income portfolios is to maintain constant term and credit exposure over time, which is exactly what we do with stocks.
What we do, not what everybody does. Some people try and time factors in stocks. We don’t think that makes a whole lot of sense. You could try and do the same thing. Static exposure, okay, let’s overweight term and credit relative to, say, the market in order to increase expected returns. But now, the difference, and this is mind-blowingly fascinating or at least it was to me when I figured it out, there’s a huge difference in the ability to forecast expected returns in bonds versus stocks. I’ll explain why that exists.
If we take the basic evaluation equation, price equals expected future cash flows divided by the discount rate, same thing we can use to talk about stocks, but also bonds. If we talk about stocks, if price changes, it’s really hard to say definitively whether the price change was caused by changes in the market’s estimate of future cash flows or if it was changes in risk aversion. Can’t say.
Cameron Passmore: Generally speaking, overall.
Ben Felix: Yeah, if stock prices drops, it’s stock index. The S & P 500 stocks fall in price. We can’t say whether that price drop was due to increasing risk aversion, so an increase in discount rate, or deteriorating expectations for future cash flows or some combination. I mean, realistically, it’s some combination, but you can’t really disentangle those in stocks. I mean, that just makes sense, or at least it does to me. With bonds though, it starts to get a lot more interesting, because as the name of the asset class might suggest, the cash flow component, expected future cash flows is pretty much fixed. You kind of know what you’re getting on the cash flow side. So it’s a little bit more reasonable to expect that differences in price changes or yield changes on bonds are due to differences in risk aversion, differences in the risk premium for owning bonds.
That’s not the academic explanation. We’re going to talk about that, but that’s sort of an intuitive, logical explanation for why we may be able to forecast differences in bond returns better than stock returns, because we’re more certain that differences in yields today are based on differences in risk aversion, meaning differences in expected returns.
Cameron Passmore: Safe to say it’s also a change in the cost of capital for the companies, effectively? Price of bonds go down –
Ben Felix: Yeah. Well if we talk about differences in expected returns, yeah. That’s exactly what it is.
Cameron Passmore: Expected returns is the cost of capital to the company.
Ben Felix: Roughly, yeah. Or at least there’s a relationship there. Now, it is important to note that with stocks, over longer periods of time, there is a relationship between price and expected returns. If we use the Shiller price earnings or just price earnings, even generally speaking, it doesn’t have to be Schiller. There is a relationship between current prices and 10 year realized returns. So it’s not like there’s no relationship in stocks, but you’re not going to use a 10 year market timing signal. It doesn’t tell you a whole lot to know that year expected returns are lower. It doesn’t mean you should get out of the market today. You can’t use Shiller PE as a timing mechanism, but with bonds, I guess it does become a bit of an empirical question, which is one of the things we’re going to talk about.
With bonds, the short term forecasts … so with Shiller PE, 10 years you’re going to explain about 40% of the differences, 40% of the variation in future returns using Shiller PE with stocks over a 10 year forecast period. Over one year periods, it’s zero. There’s no forecasting power with Shiller PE. With bonds, it starts to become more like 40% for one year forecasts. So it starts to become a much more meaningful signal. If we want to start talking about, I mean, it sounds like market timing, I guess it kind of is, if we want to start talking about timing the exposure to risk factors in fixed income. So there’s a theoretical question. Does make sense in an empirical question? And we’re going to dive into both for both term and credit, because they’re a little bit different, both on the theory side and the empirical side.
But just to step back, what we’re talking about here is saying okay, there are factors in fixed income, but it doesn’t actually make sense to maintain consistent exposure to them because it’s possible to forecast risk premiums in a short enough window that varying the maturity of a bond portfolio and the credit exposure of a bond portfolio can increase expected returns. With stocks, the signals are too noisy. You can’t do it. With bonds, maybe possible.
Okay, so we’ll start with term. The yield curve, that’s a term lots of people I’m sure have heard at least. The yield curve is a chart that describes the term structure of bond yields. So a normal yield curve is upward sloping, which just means that shorter maturity bonds, so on the shorter end of the maturity spectrum, have lower yields than longer maturity bonds. Makes sense. And then that means simple story is that longer maturity bonds are riskier. So in the normal upward sloping yield curve is exactly that. Now a bond’s realized return has three primary components. There’s the yield, the current yield, there’s the expected capital appreciation or depreciation over the holding period based on the slope of the yield curve, and that’s a bit of a tricky one to think about, but I’m just going to leave it the way I said it.
The third piece is the return due to future interest rate changes. And we can’t predict future interest rates. That’s been examined thoroughly on the empirical side, and you just can’t. So there’s no point in trying, but the first two components, the first two pieces that I mentioned, the current yield and the expected capital appreciation or depreciation, those are things that we can observe today by looking at yield curves, and they combine to form something called the forward rate. So when we start getting into the academic work, by Fama, largely, or at least starting with Fama, that’s what he’s looking at. He’s looking at the forward rate, and he’s asking if current forward rates contain reliable information about future differences in bond returns. So here we’re back at that forecasting idea using forward rates. And I’m going to talk more about what those are. I’m sure people are scratching their heads. I don’t know how many people talk about forward rates over the dinner table.
Cameron Passmore:It’s clear so far.
Ben Felix: Just as a side note, this is … I’ve been wanting to do this topic for awhile, and there are a handful of papers on it that I knew existed that I had tried to read a bunch of times, and it probably took me five times of sitting down and reading these papers to actually at least feel like I could talk about it or understand it enough to talk about it. Forward rates compare two hypothetical investments. You can purchase a bond with a longer maturity, which would have a higher yield, or purchase a bond with a shorter maturity and reinvest the proceeds of the shorter maturity bond into a new bond with a maturity that matches the remaining maturity of the longer maturity alternative. I said maturity a lot of times there. Basically you can hack together a short-term bond now and a bond in the future to equal the current yield on a longer term bond. The forward rate sets the required return on the second bond in the hacked together scenario that you would need to earn to equal the current rate on the longer term bond. That make sense?
Cameron Passmore: Yes.
Ben Felix: So if you take two bonds with the same yield of maturity, one is yielding 2% for a one-year maturity and the other yields 3% for a two year maturity, so two different bonds, a shorter term and a longer term. The longer term has a higher yield, the shorter term has a lower yield. The one-year forward rate in this case is the rate on a second one-year bond in the future. So this bond doesn’t exist yet. It’s the bond that will exist in the future to set the combined rate of the two shorter maturity bonds equal to the longer maturity bond. So in this case, the forward rate is 4%, because if you’re getting 2% on a one-year bond and reinvesting the proceeds at maturity into a new one-year bond at 4%, the resulting yield is 3%, which matches the current two year bond, which is yielding 3%.
Now this also means, interestingly, that the steeper the yield curve is, the higher the forward rate is going to be, the higher that second rate is going to have to be equal the longer maturity bond if there’s a steeper slope on the yield curve.
Ben Felix: Precisely. That all made sense. I’m not going to mess it up. It makes sense.
Ben Felix: Good. We’re talking about forecasting here, as I’ve mentioned a couple of times, so there’s some pretty big theoretical questions and some pretty big empirical questions that we have to start asking about how much sense this actually makes. So Fama had a paper in 1976 in the Journal of Financial Economics titled Forward Rates as Predictors of Future Spot Rates, where he showed mathematically that the forward rate in excess of the current short rate can be written as the expected change in the future spot rate and the expected term premium on bonds of of matched maturity. There were two components there that Fama showed mathematically, the forward rate in excess of the current short rate can be written as the expected change in the future spot rate.
So expected change in the future spot rate is one piece, and the expected term premium on bonds of matched maturities. That’s the second piece. So there are two components there that matter, and the distinction is important, because if the expected change in the future spot rate is the only thing that determines forward rates, then the only way to get an upward sloping yield curve is through an expectation of higher spot rates in the future. So if that’s true, if the only reason that the yield curve is upward sloping is because there are higher expected spots rates in the future, then there is no term premium. The yield on a long-term bond just equals the average of the expected future short term interest rates over the life of the bond in that theory. And these are actually two competing theories.
Like I just said, if that is the case, if forward rates are explained by expected future spot rates alone, then there is no term premium. And if that’s true, then it would make sense just to continuously roll short-term bonds forward, because that would have the exact same expected return as buying a longer term bond.
Cameron Passmore: Without the long-term bond volatility and price, potential volatility in price.
Ben Felix: That’s probably true. So in that case, you, there would be no point in pursuing any kind of term premium. Now the alternative view … and this is where it starts to become the interesting empirical question. The alternative view is that forward rates do contain information about an expected term premium. In other words, differences in forward rates between bonds of different maturities forecast differences in expected returns, and those expected return differences can change over time as the forward rate changes. So here’s where we get to the empirical question, which is do forward rates forecast expected term premiums? And the evidence does support a reliable relationship between forward rates and expected term premiums. In other words, the shape of the yield curve does have information about differences in expected bond returns. It’s pretty heavy stuff.
Cameron Passmore: It’s pretty tactical.
Ben Felix: Yeah. Hopefully it’s okay. I mean it’s not the first time we’ve been technical, so I’m not too concerned.
Cameron Passmore: No, no, no. I’m not saying to be concerned. It’s interesting.
Ben Felix: A steeper yield curve forecasts a higher expected term premium, which is important when we start talking about forecasting. Should you take on term risk? Well, the answer to the question depends on how steep the yield curve is at a point in time. A flatter yield curve forecasts a lower term premium. Now you’re taking term risks either way. The question is does that risk have an expected premium associated with it at that time, which varies through time. I found a 1984 paper, again in the Journal of Financial Economics, titled Term Premiums in Bond Returns. And in this paper, he found reliable evidence that forward rates contain information about expected term premiums. And he found … the way he supported that is that when yield curves have been steep, there has been a reliable difference in subsequent average returns between longer and shorter maturities, and when the yield curve is flat, those return differences have been much, much smaller.
I found a 1987 paper where he found similar results for longer maturities and Campbell and Schiller found similar results again in 1991. Now, dimensional has a white paper … this paper is really cool because it takes these academic ideas. And what they do in that paper, and this is one of the reasons I wanted to talk about this in the podcast. I don’t like talking about things that are only relevant to clients in the podcast, for example, because I know a lot of our listeners are not clients. So where does this start to become relevant to DIY investors? And it’s from this dimensional paper. So they had a 2016 paper where they built a variable maturity strategy using indexes. So they’re not doing black box stuff, they’re not doing any secret sauce stuff. It’s just like, okay, if we take indexes, short term indexes and intermediate term indexes, can we build a variable maturity strategy using those indexes depending on the shape of the yield curve and earn an excess return?
So what they did is if the yield curve was upward sloping, and they defined upward sloping as more than a 10 basis point spread between intermediate and short-term bonds. If the yield curve was upward sloping, they allocated up to 100% of the portfolio to intermediate term bonds. And if the yield curve was flat, they allocated up to 50% to short term bonds. And they’re just rolling forward this variable maturity strategy from, I think it was 1974 until 2015 is when they ran their simulation. So they found in that sample a 12 basis point excess return with a lower standard deviation compared to a static benchmark, which was just 50:50, or it was just intermediate credit and government bonds. It was 50 credit, 50 government, both intermediate term. There was a premium, which is what you’d expect for taking term risk only at the times when there was an expected premium for doing so, or when that expected premium was larger.
And I do just want to reiterate that this is all based on the current shape of the yield curve. There’s no forecasting going on here other than maybe forecasting expected premiums, but there’s no forecasting the future shape of the yield curve. It’s saying based on the current shape of the yield curve, is there an expected term premium? And if the answer is yes, you take on the term risk. And in a Monte Carlo’s simulation, they showed that you can get a premium using just indexes, which obviously follows … you can maybe do the same thing with ETFs. So that’s term. The next one I want to talk about is credit. Again, if you look at credit premiums through time, even without doing any timing, there are meaningful credit premiums. I’ve seen some papers that argue that the credit premium is fully explained by equity risk factors, and therefore there’s no sense in taking credit risk on the fixed income side. I’ve not seen research that comes to that conclusion while using a variable credit strategy.
So an interesting thing to look into, but I haven’t found it yet. And then if the credit premium changes over time, it’s maybe not surprising that running analysis on full period returns for a credit index that you wouldn’t see a whole lot of benefit. If you’re taking credit risk when it’s compensated and when it’s not compensated, I can see why it wouldn’t look so good in historical analysis. But when you split that out, you’d expect that to change. Now I looked at … just because I had the data, and I don’t have the numbers in front of me. We’ll put them up on the YouTube video. Barclays has intermediate indexes from treasuries all the way down to BA, which is the lowest rating for investment grade bonds.
And they also have a high yield index and the relationship is monotonic. And the return increases from treasuries to high yield are significant. So there’s definitely a difference in expected returns, even static, even without varying credit exposure. Now where it gets interesting is that the yield spread, the difference in yield between credit bonds and government bonds, has been far from constant. It’s been pretty volatile. The spread, the yield spread has been volatile. It’s like the relative cheapness of credit bonds relative to government bonds is maybe a way to think about it. Now again, we get into two possible explanations for this time bearing yield difference, this time bearing yield spread. It could be changes in the default probabilities and corporate bonds. And in that case, kind of like the cashflow mechanism for stocks, if cashflow expectations decrease and expected returns don’t change, then you can get prices dropping without a corresponding increase in expected returns. Same thing for corporate bonds.
Cameron Passmore: Oh.
Ben Felix: Yeah, right?
Cameron Passmore: Yeah, so it’s more than just a value story.
Ben Felix: Well, it could be. And this is one of the empirical questions. So maybe the time bearing yield spread is because default probabilities increase, but that would result in falling prices for corporate bonds without a corresponding increase in expected returns. So you probably wouldn’t want to take credit risk in that case. The other possibility is the differences in default probabilities between government and corporate remain constant, but the default premium changes, the risk premium changes. Now in that case, in that case if default premiums remain constant, but the risk premium increases, you would want to take credit risk when the spread increases.
Cameron Passmore: How do you know the difference?
Ben Felix: Great question. I wish I was smart enough to be the one to come up with how you test that, but there are a couple of papers that we’re going to talk about in a second that explain how you know the difference. Before I get to those, the expected return on a bond can be summarized as current yield times one, minus the probability of default, plus the probability of default, plus the probability of default, times the expected recovery rate, minus one. Recovery rate being how much you get back if the issuer defaults. So differences in expected returns depend on yields, default probabilities, and recovery rates. If yields change, while default probabilities and recovery rates remain the same, then there is a clear relationship between credit spreads and expected returns. So yields change is like prices changing. If yields change while default probabilities and recovery rates remain the same, then there is a clear relationship between credit spreads and expected returns. Expected returns go up, prices go down. And if the prices are going down because of the risk premium, and not because of an increase in default probabilities or a decrease in recovery rates, there is a risk premium, or and expected premium that we may want to try to capture.
The empirical work, there’s a 2010 paper titled Corporate Bond Default Risk: A 150 Year Perspective. And in that paper, and this is the answer to your question, Cameron, they found that credit spreads have little explanatory power for default rates. So credit spreads increase, it does not have a corresponding increase default rates. Now that also suggests, kind of like we were just talking about, that the variation in credit spreads is mostly driven by the variation in expected credit premiums, not by default rates.
Now there’s a 2014 paper titled What Drives the Cross Section of Credit Spreads, a variance decomposition approach. And in that paper they found in 40 years of US investment grade corporate bond data, that over 90% of the variation in credit spreads can be attributed to variation in expected credit premiums. I did not dig into their methodology. Presumably it was somewhat similar to the other one.
Now in the same dimensional paper that I mentioned, where they test these strategies with indexes, they also did a variable credit strategy. So they took a cap weighted credit, and government bond index, portfolios as the benchmark. And then they built a variable credit strategy using the same underlying indexes, but they increased the weight in credit out to a maximum when credit spreads were wide, and then decrease down to a minimum when credit spreads were narrow. So they’re never eliminating credit risk, and they’re never going all credit, but they’re varying the credit exposure based on how wide credit spreads are at a point in time. And with that strategy, they found a 17 basis point annualized return increase over that same 1974 through 2015 period.
So, two different premiums were in dimensional simulations they found 12 basis point annualized extra return, and a 17 basis point, so 12 for term and 17 for credit. And in the same paper they said, “Let’s take this to the next step. Let’s vary term and credit at the same time, depending on the credit spreads and the shape of the yield curve, and see what that does. So in this case, they built a variable credit and variable maturity strategy that could overweight it’s allocation to corporate bonds or government bonds by a maximum of 40%, based on credit yield spreads, varied its maturity based on the shape of the yield curve like we’ve been talking about. And they benchmarked that against 50% intermediate government and 50% intermediate credit bond indexes. And in this case, it actually seems like it’s an additive for the other two premiums. I don’t know if that’s always going to be true, but in this case it was 30 basis points of extra annualized return with a lower standard deviation.
And then the other really interesting part is it only a slightly lower allocation over the period to government bonds. So keeping in mind their varying credit exposure, so you’re taking more credit risks, sometimes less credit risk other times, they ended up with a higher realized return, a lower standard deviation, and only a slightly lower average allocation to government bonds. So for the full period, it’s not … You weren’t taking that much more credit risk. You were just taking it at the times when it was compensated. And that resulted in the premium with less volatility, which is pretty cool.
Now, the most important part. And we had a conversation Cameron last year with somebody about this, where you can’t look at bonds and say, “Well bonds, they have no expected return.” And the reason that you can’t do that is because these premiums that we’ve been talking about, term and credit, they are independent of yields. It’s not about the absolute level of yield. Yields are low, sure. But if there’s a credit spread and there’s a credit premium, or if the credit spread’s wide then there’s a credit premium, and if the yield curve is upward sloping, then there’s a term premium. So to say that bonds as a whole have no expected return, that that would maybe be true if yields are zero and there’s no credit spread and the yield curve is flat, then bonds had no expected returns. Otherwise there are multiple sources of expected returns in fixed income.
So to conclude that if the fed funds rate is zero, we can’t say the bond yields are zero, because there is a credit spread and there is a yield curve. If the credit spread is wide and the yield curve is upward sloping, then there are more than one source of expected returns.
Cameron Passmore: And not all bonds are government bonds.
Ben Felix: Yeah. A lot of them are, just because of the way markets tend to be. But yeah, there’s a whole opportunity set outside of that. And even in that though, there’s a term premium. You don’t even need credit to find additional expected returns.
Now you think about that in the context of expected returns for bonds, like financial planning purposes right now, PWL is using 0.57% in real terms for bond expected returns. And if we just use the 30 basis points that Dimensional found in their variable credit and variable maturity strategy, that’s a lot. An extra 30 basis points on 57 of a real expected bond return. So that’s the positive side. Why could this be a good thing to do?
There’s also a negative side, not negative in terms of the strategy we’re talking about, but negative in terms of the alternative, which is just cap weighted indexing. You can make similar arguments for stocks, but I think with the forecast ability of risk premiums in bonds, it becomes a little bit more interesting. So if you think about all the information that we just talked through, and then think about a cap weighted bond index fund, when credit spreads narrow, so credit bonds have relatively higher prices, just from what we just talked there, we know that the credit premium is low if the spread is narrow. But what’s a market cap weighted bond index doing if credit spread’s narrow?
Cameron Passmore: Own more of them. Or it’ll be a bigger weight of the portfolio.
Ben Felix: It’s increasing the weight in credit because their prices have increased. So a variable credit strategy is doing the opposite. If credit spreads narrow, it’s decreasing the weight in credit, whereas a cap weighted bond index fund is increasing the weight in credit. And likewise for term, if the yield curve flattens, again, what happens?
Cameron Passmore: So a cap weighted bond index is effectively lowering it’s expected returns based on this framework.
Ben Felix: Yeah, yeah. So yield curve flattens, all of a sudden longer maturities prices are going up, and a cap weighted bond index fund is therefore owning a higher allocation to these longer maturity bonds that have a lower term premium. So a cap weighted bond index fund, it’s kind of working against you, at least in terms of the premiums. So again, that’s a reason that this strategy could be interesting for ETF investors.
Now, from Dimensional’s perspective, and this is getting more into the stuff that’s relevant to clients and not really anybody else. Dimensional does this around the world. So not just looking at credit spreads and yield curves in Canada, they’re looking everywhere. So if a country has particularly attractive yield curve, they can go look there. One of the reasons that that works is something called covered interest rate parody. Not going to get too much into the details, but basically if Germany has an upward sloping yield curve, but they’re short. The short end of their yield curve is negative, people go, “Oh, I don’t want negative bonds.”
If you hedge foreign fixed income back to your home currency, your short rate, the starting point of the yield curve, ends up being the same as your home yield curve. So Canada has a short rate at 0.5% or something, but Germany is negative whatever, 0.25, but you hedge the German yield curve back to Canadian dollars, you start at the same short rate as Canada, but you get the shape of the German yield curve. So if it’s upward sloping, maybe you want to be taking term risk in Germany. You can only do that if you can hedge back to your home currency.
Cameron Passmore: So you’re basically bringing foreign yield curve shapes home.
Ben Felix: Shapes and spreads, all that stuff. Yeah. I’m grateful. Dimensional is doing that in their portfolios. But I think for an ETF investor to do that on their own, I think that’d be pretty tricky.
Now, a concept that I want to explore more, and it’s kind of along the same lines as the model ETF portfolios that we’ve done in the past. I think it’d be really interesting to see if it would be possible to build a variable credit, variable maturity strategy for a DOI investor using fixed income ETFs. So as opposed to just owning the aggregate bond ETF and the reasons you might not want to do that are what we just talked about in terms of expected returns, it could be interesting to look at the idea of taking four ETFs, credit, short-term credit, intermediate term credit, short-term government, intermediate term government, and building an ETF strategy that tries to capitalize on some of these premiums. So I have not done that, but it’s something that I plan on looking at. The products exist. I went that far. The products to do this, at least in Canada exist.
And there’s a whole other big question here, which is why would somebody own bonds? And I know at least in the rational minded community, there’s a lot of people who would just say that they wouldn’t. Keep in mind that with the premiums we just talked about, you’re not giving up on the volatility reduction benefits of bonds. So if someone wants to own bonds, then adding these excess returns that we’re talking about could make a lot of sense.
Cameron Passmore: I agree. All right. So let’s call it our planning topic this week. You wanted to talk about expected return assumptions.
Ben Felix: We’ve done this one before. A while ago, actually. Probably quite a while ago, we talked about it on the podcast. So I thought it was worth bringing up again. Since Brayden started working with us, this is something he spent quite a bit of time on, or we spent quite a bit of time on together, just figuring out in the ideal financial planning model, if we were to build one, because we are, how would we model expected returns?
There are aspects of Monte-Carlo, which we currently use. I think it’s a fantastic tool, and it’s much better than just using a fixed historical return, for example. But there are aspects of Monte Carlo that have always made me a little bit uneasy. And so we’ve been trying to address some of those. You can fix that by just making more conservative assumptions in the Monte Carlo analysis, which is something that we’ve also done. I think there are also modeling approaches to overcome some of those shortcomings, which is some of what I want to talk about.
Now, the reason this matters is because one of the most important assumptions that you make in financial decisions is expected returns. A lot of decisions people make would change materially with different expected return assumptions, which makes them pretty important, I would say. Now, something that I see online every now and then, which is terrifying, is people just assuming a 10% expected return because that’s what the US market has returned historically. Which it has, but does that make that a good expected return? I don’t know about that. And there’s big implications. Somebody might under save, they might retire too early. They might spend too much in retirement if they’re assuming these high returns. Now, on the other side, assuming too low of a return, that can also be detrimental. You might over save, retire too late, not enjoy freedom for as long, or not spend enough. We’ve spent so much time recently talking about wellbeing and all that stuff, and all of this is central to that.
Now the other piece of this whole thing is that it’s not just the average return that matters. Even if we say 10%, you’re not going to get 10% every year. So that sequence of returns also matters a lot. Bad returns early on can be really hard to recover from. So pretty obvious, I think, that the choice of not only the expected return, the expected average return is important, but the model for expected returns over time is also really important.
Historical analysis, so using historical data, like the 4% rule stuff that Bill Bangin’s done, which is, as we’ve talked about on the podcast, including with him, hugely important research in the field of financial planning, but it’s got some major drawbacks in the limited data availability. There are only so many historical periods to look at. And then the other big one, particularly in the US is that price earnings levels historically have been a lot lower, the mean average Shiller price earnings ratio is like 16 or something, in the US going back to 1871 when Shiller starts computing the data. Or maybe even higher than that. I haven’t checked in awhile. And prices just keep going up.
Two big questions to think about in determining financial planning assumptions. One is do expected returns change over time, or are they constant? And then the other one is if they are time-varying, if they change over time, is there anything that predicts differences in future expected returns? A lot of people are thinking, “Well yeah, we know that Shiller price earnings or some measure of price earnings has some predictive power, and yeah it does. If expected returns are constant, then using the historical average is pretty sensible, if they’re constant. If they’re time-varying and have some predictability, then it might make sense to look for information and prices regarding expected returns, at least from a financial planning perspective.
Now, the empirical research on this does suggest that there is predictability in returns. And honestly, the first time I read … It was in actually a Lubos Pastor paper, where he said as a statement in part of the hypothesis he was testing, that returns are predictable. And I was like, “What? Returns are predictable? It seems to go against everything that I’ve ever learned.” And I think he was referencing a FAMA paper. And then I started to uncover this whole body of literature about return predictability and stock markets. Which sounds weird to say, but it’s kind of not what it sounds like. Cliff talked about this. He had a paper a while ago titled, “An old friend, the stock markets Shiller PE.” He said this title do analysis is not much help to market timers, but it can help set reasonable expectations. If your long-term plan calls for a stock market return of 10% nominal, you’re basically rooting for the absolute best case in history to play out again, and rooting for something drastically above the average. This could happen, but unless you’re casual with forecasting some giant positive market surprise, we believe you should plan for lower average returns. You could, for example, save more or spend less or change your portfolio structure.
Now there’s a distinction here between, like I said, return predictability. There’s short-term predictability and longer-term predictability. In the shorter term, monthly returns kind of thing, there’s no predictability. And this was a big part of FAMA, and FAMA’s 1970 argument for markets being efficient. The returns are not predictable. Now, longer term predictability. This is where there is a lot more evidence. It’s just not evidence in favor of timing the market. It’s average returns over longer periods of time can have some predictability based on current conditions. Now, does that mean markets are inefficient? It doesn’t. It just means expected returns aren’t constant. Kind of like we were talking about with fixed income a second ago.
It’s actually really similar, interestingly, to what we were talking about with bonds, which makes sense. So the valuation equation suggests the prices change based on cashflow expectations and risk aversion. Like we mentioned earlier, it’s really hard to split out what causes price changes, but it’s tricky to say that the equity risk premium is constant through time. It’s kind of like saying that price changes are solely based on cashflow expectations. So, John Cochran had a paper in 2006 that looked at that exact question, and he was addressing another body of literature that was saying that there is no predictability in stock returns. So Cochran said, if both returns and dividend growth are unforecastable, then present value logic implies that the price dividend ratio is a constant, which empirically it surely is not.
Alternatively, since the dividend yield is stationary, one of dividend growth or price growth must be forecastable to bring the dividend yield back following a shock. We cannot just ask our returns forecastable, we must ask which of dividend growth or returns is forecastable. Different part of the paper, he says that the point estimates say that every time of high prices, low dividend yields in the past 80 years has been resolved by low subsequent returns, and not by higher dividend growth.
Cameron Passmore: Interesting.
Ben Felix: So, he’s kind of saying a lot of the price variation is differences in risk aversion, as opposed to cashflow forecastability. Fama had a paper in 1990, and he found in that paper that time-varying expected returns and expected returns shocked capture 30% of the variants in annual NYSE-valuated returns. CFA Research Foundation Book, Financial Market History, Anthiel Manon , he looked at the literature on whether equity risk premiums are constant or time-varying. He had a little anecdote to illustrate his general finding, and he looked at the 2000 tech boom as his anecdote. So he said, “Periods of sustained declines in required returns boost contemporaneous returns, the so-called discount rate effect, and are especially dangerous for investors who believe in constant and expected returns. Our rear-view mirror perspective have made the equity premium seem highest at the end of a long bull market, just when market valuation ratios were flashing red.” After the bust in the early ’00s, it was evident that forward-looking valuation measures had given an empirically and logically better signal than had historical average returns.”
So from a financial planning perspective, it starts to get pretty important. If you’re using historical returns as an expected return, as realized returns throughout the path of your investing life… As historical returns get higher, future expected returns may be getting lower. But if you’re using historic returns as your expectation for the future, as they go up you’re also assuming higher returns in the future. Can be a little bit deadly.
That’s what Fama looked at this in his paper that he does every year, Equity Risk Premiums, Determinants, Estimation and Implications, so in the 2020 edition, he looked at different forecasting metrics to see how they perform, which is a pretty intuitive but interesting exercise. So he used implied equity risk premiums, which is the earnings yield idea. He doesn’t use Shiller CAPE. He thinks Shiller CAPE is a useless measure, which is a whole other interesting side note, but similar. He basically says, if you just use earnings yield without using the Shiller adjustment, you get the same result. That’s why he doesn’t think it’s useful. But he finds that risk premiums from earnings yield are positively correlated with realized five and 10 year returns, but historical equity risk premium are negatively correlated. So if you’re planning savings amounts and cashflow spending from a portfolio, you can see why this is really important. We’re going from a positive correlation on one hand, to a negative correlation on the other. The sequence of returns and withdrawals is extremely important.
So what we’ve been working on, and the model that we’ve created to hopefully do a better job capturing the stuff that we’ve been talking about, we… Currently at PWL, and I think that what we’re doing right now is a really good compromise, but currently what we’re doing is using a constant expected return assumption in Monte Carlo analysis, and it’s based on an average, a simple average of current expected returns from earnings yield and historical returns. It captures the idea that expected returns are lower when prices are higher, but it also captures the idea that over the long-term, you might not expect returns to be low forever just because they’re low today, I guess, although, it doesn’t capture the time variance of that relationship.
Now, the other problem with Monte Carlo is that there’s no serial correlation. So serial correlation is that predictability idea. If prices are high, expected future returns are low. Monte Carlo doesn’t see that. Same thing with bootstrap. If we take the historical data and just pull randomly sample months out of the historical data, you kill any serial correlation, which makes your distribution of outcomes a lot wider than what we actually see in the historical data where the distributions are a lot tighter. So what we’ve done is we’ve created this expected returns model, where it starts with earnings yield. So it starts with the basically inverse of the Shiller P/E as the starting point for the equity risk premium, but then we’ve allowed that to drift over time as prices change. So it captures the higher probability of lower expected returns today when prices are high, but it also allows for expected returns to change over time in the model to reflect the fact that if prices fall in the simulation, or if earnings rise, then expected equity returns can increase.
So when you think about the path dependence aspect, what you get in this model is lower expected returns today, but still the chance at increasing expected returns in the future if things “normalize,” if prices go back to some more historically normal level. This approach relative to using a constant expected return without any predictability and all that stuff, the impact is largest for longer retirement periods. So, for modeling a 60-year period, this updated model gives them a more favorable outlook over long periods of time, largely because of the predictability. We’ve so far found, and this is all preliminary research, but we’ve found up to a 30% higher success rate. Even though we’re starting with a lower expected return, the success rate’s a lot higher because the predictability over long periods of time, the mean reversion, basically, over long periods of time, has a really big impact.
Cameron Passmore: That’s high impact.
Ben Felix: It’s high impact. But the crazy thing is that if you do really long periods using Monte Carlo, it’s kind of scary for me to say, “Well, look, it’s this much better if we model it this way,” but at the same time, this is a lot closer to the historical reality. And this is including… We’ve tested this with the Dimson, Marsh, Staunton global data set from 1900 to 2019, including two countries that disappeared off the face of the… Well, of the stock market, at least, not the earth. We’ve used that as a guidepost. So we don’t want to use the type of mean reversion that the US has had. That’s maybe too extreme. But if we look at global data, is it reasonable that we get something similar over the next 100 years?
I mean, if we get mean aversion, if returns just crater and never come back, I think we’re in that situation again where it’s like, “We’ve got bigger problems,” that old saying. So modeling it this way, I don’t know, I think it seems pretty reasonable. But it starts to change a lot of stuff. Like views on the 4% rule for very long periods of time, it’s not going to increase it to 4%. When I’ve done this with Monte Carlo, I think I’ve found like a 2% safe withdrawal rate. Using this new model, it’s going to be a bit higher than that. So I think that’s where this starts to become pretty important. So anyway, as we continue to work on this, hopefully we’ll be able to start sharing more of the research. But we’ve spent a lot of time thinking about it, so I thought I’d thought it.
Cameron Passmore: Pretty cool. Good to move on to the Talking Cents segment? First question, if you couldn’t have your first choice job, what other types of jobs would you consider? I know I’d love to be a surgeon. A very good friend of mine’s a surgeon. I think it’s an incredible profession where you get to really dig into something mastery, high-impact, purposeful, saving lies. That’s what I would do. How about you?
Ben Felix: I’ve always thought about medicine. When I finished engineering, my dream was to go and study medicine, but I chose the path of basketball instead, which took me to Carleton because Carleton in Canada has like the Duke or UNC of basketball in Canada. They didn’t have a medicine program. So because of that, I did an MBA. But to answer the question, going into medicine would be pretty cool. I also think about, especially with all the fun I’ve been having with building the bots, doing some kind of engineering work could be pretty interesting. Yeah, so maybe one of those two.
Cameron Passmore: Here’s an interesting question. Would you rather have a free place to live the rest of your life or never have to pay for food again?
Ben Felix: I eat a lot of food. I don’t know.
Cameron Passmore: That’s what I was just thinking.
Ben Felix: Yeah. Yeah, I eat a lot. I spend a lot of money.
Cameron Passmore: You get free food for your kids for the rest of their lives.
Ben Felix: I spend a lot on food. My parents used to joke. My dad worked at a boarding school, and so I went there. And they used to always joke that they still had to pay… It was a discounted tuition rate, I think, but still not inexpensive. And they always joke that it was cheaper to send me to the boarding school because I got to eat at their cafeteria, than it would have been to keep me at home.
Cameron Passmore: Yeah. I don’t even know how to answer this question.
Ben Felix: Well, you got to pick one. I love food.
Cameron Passmore: I mean, free food or… never really thought of it. But you get to the point, right? Free food. So, on the cards, I’ve been talking to Rebecca, who was on a couple weeks ago, and they are looking into getting us co-branded cards, we hope. Let’s see if we can get a good rate to pass along to our listeners inside the merchandise store. So, more to follow on that.
Ben Felix: That’d be an awesome product to add.
Cameron Passmore: I agree. Bad advice of the week comes from listener, Sean, in Cambridge, Ontario, who got his hoodie. And this goes back to Bitcoin again. So, he came across an article from CASHAY, C-A-S-H-A-Y, which is a personal finance website that, “Helps people in all stages of life manage individual and family milestones.” And the article was entitled, Here’s How to Incorporate Bitcoin into Your Retirement Savings. So, the article is from December 7th, 2020, and of course the price of Bitcoin I think has basically doubled since then. But the article suggests that now that the value of Bitcoin has increased over 4,000% since its inception, some retirement savers may wonder if it’s smart. So, retirement savers may wonder if it’s smart to invest in cryptocurrency for their golden years.
And it’s interesting. They interviewed a couple people as to what are their arguments to why holding crypto now make sense for retirement portfolios. And these are the arguments that the people they interviewed had. So number one, “Consistent annual returns over the last five years, 121% annually.” Although, there was a period in there where Bitcoin did fall 70% from 2018 into 2019. So that was the number one reason, five years of good returns. At number two, “It’s here to stay.” So you can take that for what it’s worth. And, “The best time to invest in cryptocurrency is now rather than later, especially for those approaching retirement.” So to quote the article here, “Obviously, there is the idea that it could go down also, but if you stay safe with a crypto that is just gradually increasing, you are more than likely going to make a large profit if you have it for a number of years,” said this person they interviewed. “It will be a smart move for those who are looking to retire in the next three to five years, as you could almost double your income just by moving it to crypto.”
Ben Felix: Oh, man. When you started with this, I wasn’t convinced it was necessarily bad advice because I’m not sold that Bitcoin is the stupidest thing in the world. I’m not sold that it’s going to go to zero. It may end up being like a gold type-
Cameron Passmore: We are not saying that. I agree with you.
Ben Felix: Humans are like that. But this is beyond that. This isn’t just saying this might be a thing that sticks around and maybe keeps pace with inflation. This is saying that you’re going to double your income.
Cameron Passmore: But how has that staying safe and gradually increasing if you’re going to double in the next three to five years?
Ben Felix: I would never recommend it. I just can’t wrap my head around investing in cryptocurrencies, or gold, as listeners know. But if someone said, “I’m going to allocate 1% or 5% of my portfolio to Bitcoin,” I don’t think I’d call them too crazy. I wouldn’t do it. Our clients aren’t going to do it.
Cameron Passmore: So, Sean emailed me. As he said in his email, “So how much of your retirement income do you need to invest in crypto to double your income?” It’s a pretty big chunk, right?
Ben Felix: Well, it depends what the-
Cameron Passmore: Yeah, I guess if it gets 4,000% return, you’re right. But don’t worry, if you’re three to five years away from retirement, now is definitely the time to take on more risk. So, there you go. If you have bad advice for us to share, send us a note and we will send you a hoodie.
Ben Felix: That’s pretty irresponsible. As far as bad advice goes, that’s pretty bad. It’s one thing to say the mistake of 1% is in your portfolio, it’s a whole other thing to say, “Pile in, because you’re going to double your income three years before you retire.” Yeah, that’s something. If you’re enjoying the podcast, leave us a review on iTunes. I think it helps. I think it helps with the visibility of the podcast on the podcast platforms. So, more listeners makes the community bigger and probably better. So yeah, if you’re joining the podcast, leave a review on iTunes or wherever you’re listening. Anyways, as always, thanks for listening.
Books From Today’s Episode:
We Need to Talk: A Memoir About Wealth — https://www.amazon.ca/Hit-Refresh-Rediscover-Microsofts-Everyone/dp/0062959727/
Links From Today’s Episode:
Rational Reminder on iTunes — https://itunes.apple.com/ca/podcast/the-rational-reminder-podcast/id1426530582.
Jennifer Risher — http://jenniferrisher.com/
‘Bitcoin consumes ‘more electricity than Argentina’ — https://www.bbc.com/news/technology-56012952
‘Common risk factors in the returns on stocks and bonds’ — https://www.sciencedirect.com/science/article/abs/pii/0304405X93900235
‘Forward rates as predictors of future spot rates’ — https://www.sciencedirect.com/science/article/abs/pii/0304405X76900271
‘Term premiums in bond returns’ — https://www.sciencedirect.com/science/article/abs/pii/0304405X8490014X
‘Corporate Bond Default Risk: A 150-Year Perspective’ — https://www.nber.org/papers/w15848
‘What Drives the Cross‐Section of Credit Spreads?: A Variance Decomposition Approach’ — https://onlinelibrary.wiley.com/doi/abs/10.1111/jofi.12524
Episode with David Blanchett — https://rationalreminder.ca/podcast/137
Episode with William Bengen — https://rationalreminder.ca/podcast/135
Episode with Lubos Pastor — https://rationalreminder.ca/podcast/124
‘An Old Friend: The Stock Market’s Shiller P/E’ — https://www.aqr.com/Insights/Research/White-Papers/An-Old-Friend-The-Stock-Markets-Shiller-PE
‘Stock Returns, Expected Returns, and Real Activity’ — https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1540-6261.1990.tb02428.x
Aswath Damodaran — http://pages.stern.nyu.edu/~adamodar/
‘Equity Risk Premiums: Determinants, Estimation and Implications – The 2020 Edition’ — https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3550293
‘Here’s how to incorporate Bitcoin into your retirement investments’ — https://www.cashay.com/bitcoin-retirement-investments-192200853.html