Meb Faber Show
May 10, 2024

Is the Stock Market Really as Predictable as You Think? | The Cockroach Strategy

Summary

  • Stocks for the Long Run: The hosts challenge the certainty of long-term equity outperformance, using global data to show meaningful odds of negative real returns over 30 years.
  • Real Returns: Emphasis on inflation-adjusted returns and compounding math, highlighting heuristic figures (e.g., 6/5/4/2/1 for US stocks/global stocks/CTAs/bonds/T-bills).
  • Country Case Studies: Japan’s 1989–2019 period shows negative 21% real returns, while Europe (e.g., Italy) had severe drawdowns; US outperformance is framed as exceptional and not guaranteed.
  • Global Diversification: The discussion advocates global equity diversification to mitigate country-specific risks and reduce reliance on US exceptionalism.
  • Risk Management: Planning around the 25th percentile outcome (~2% CAGR real) is recommended, underscoring volatility drag and path dependency on lifetime investment results.
  • Portfolio Construction: Future episodes will explore adding assets like bonds, gold, and trend strategies to build a more robust “Cockroach portfolio.”
  • Market Outlook: A sober view that historical averages (7% real) can mask wide dispersions, urging realistic expectations rather than extrapolating past US returns.
  • No Stock Picks: No specific companies or tickers were pitched; the focus was on asset allocation, global data, and robustness.

Transcript

[Music] [Music] welcome to the Mutiny investing podcast I'm here with my partner Jason Buck we're going to be uh doing the first of a series today talking about the Cockroach portfolio and the thinking that has gone into it and this is something we're excited to start sharing and talking about we've been working on it uh for the better part of a decade now um and you know I started to publish some um notes on our uh website which you can find at muty fun.com insites and are going to be talking through uh some of what we've written there as well as going into a bit more um detail on this podcast So today we're going to start off with everyone's favorite asset class stocks STS STS um and talking about stocks for the long run question mark as opposed to stocks for the long run um declarative so I I'll kick us off and um we'll go back and fourth here so many people probably familiar there's a very well-known book called Socks for the long run um by Jeremy seagull and I think broadly when you look at how many portfolios are constructed today uh stocks for the long run is sort of the the core there and I think there's a uh fairly straightforward logic that uh if you look at most sort of the large ashet classes let's call it stocks bonds gold um historical data sets particularly historical data sets based in the US um Teno stocks as the highest performing long-term asset and so the logic goes uh we want to have a portfolio that's mostly composed of stocks especially if you're relatively young or you have a long time Horizon because of course you want the highest returning thing uh over uh a long period and you know I think that's that's sort of the basis in where people um get to and we want to talk about maybe a little bit more Nuance around some of the thinking there um so a lot of this is based on there was a great paper um written by Anar kova at all in the uh Journal Financial economics uh you can go to our website and we'll link to that um basic Bally looking at uh developed Market stock returns uh from 1841 to 2000 and 19 and what was interesting about this sort of this selection this data set compared to what you typically hear which is typically um I would say you know in my experience the most common thing is usually like US Stocks us like 1920 1929 something like that to present is typically sort of like the data set people quote from um uh for stocks and so I think the sort of couple things that jump out about the the broader data set one is the probability of loss over a 30-year Horizon so Jason Jason knows the answer but what Jason trivia what is the well here I we could do a fun uh like a fun Mr Market or a beauty patch what do you think if I ask the average person the probability of a loss losing money in absolute terms over a 30-year Horizon with stocks what do you think the average answer is what does that what does the median reasonably intelligent investor that's read some investing books think I honestly think if you're going 30-year Horizon most people it's going to be zero people are going to think over 30-year Horizon it's just up and to the right you may have gone through a draw down or a losing period but I think over 30y year Horizon I don't think anybody's gonna think they have losses I think so yeah I think it's zero or one perc I think I think it's incredibly low so in on us return data that's actually pretty close to correct the number was 1.2% uh and that's real return so loss in buying power over 30-year Horizon um over you know across developed markets 1841 to 20 2019 the number is a much higher 12.1% so call it roughly a one in eight chance of a loss in purchasing a decline in real terms of a um Stockon portfolio which I think I did I did a Twitter poll to this fact and uh yes the results were very different that's not that's not that's not what most people would sort of like expect at that level yeah reminds me of like meb's Twitter polls so I was just thinking I should text Meb and have him do a Twitter poll to see like he he would love one like that but I feel like he's probably done one similarly but also like with some of meb's great papers as you know he's using Global Financial data and I think like he has some pretty good her istics in there but maybe I should start with you know in your paper and stuff we're talking about like real returns and what are real returns I mean that's your nominal return minus inflation and so that's what I think people you know really care about your real returns because you need you know what can you eat after inflation and I think rarely do people look at real returns so throughout this we're talking about real returns but I think Meb always has good heris STS too of um the simplest way to think about it is I think 65421 is basically over the last 100 years the US Stock Market has produced 6% real return um the so once again returns after inflation Global stock markets returned 5% real return after inflation kind of shows you that American exceptionalism over the last 100 years and once again that's rear view mirror that we'll get to as well um commodity Trend following produce about 4% real returns um and similarly with uh bonds is 2% real returns and then T bills is 1% real returns it you know there's a little fudge factor I think T bills is actually like 0.78 but you get the general idea it's like 654 uh 2 one or you know so you get an idea of like the different asset classes and what they uh produce in real returns but that's why it's important like we said us is six but you know Global stock markets if we look globally which you're writing about in this essay um it's a very different scenario and you know one of the luckiest things you know if we were talking about our Peter Zan I guess thinking about you know we have like these great oceans on either side of us and all this you know um great growing crop land where we can feed ourselves but like you know especially if we look at Europe going through the world wars they've had some dramatic draw downs and I don't know about you like I knew you know like we all we all hear about viar you know going to zero Russia going to zero those sorts of things but I don't know if Italy jumped out at you man they've been through hell and back it looks like negative 87% draw down 1928 to 1948 negative 84% draw down 1907 to 1927 negative 72% draw down 196 to 1980 that's uh yeah that's brutal Argentina you know we looked at sort of some of the worst um worst scenarios across countries Argentina didn't make that I guess they're probably maybe not in the developed thing but that they also jumped out in real term I can't imagine they must have had some I would expect they'd had some sort of similar periods but yeah some some brutal Italian periods in there and I think you bring up in the essay yeah talking about like Germany Europe and even like Japan and we'll come back to Japan but like you're saying in 1900 you know a priority you know looking forward you know and you know everybody if they were picking the best you know economies in the 1900s that they thought were going to out compete everybody I think you put in there Germany and UK were like number one number two and even in the Western Hemisphere as we know everybody's betting on Argent artina they would have picked Argentina over the US and like you said they're just not they're they got wiped off the map multiple times with their stock market and currency so um it gives you an idea of like at least putting trying to place ourselves you know in that time frame looking forward instead of placing ourselves you know currently and looking backwards with that that benefit of hindsight but one of the other things I love that you did is one of my favorite phrases always is like everybody has all these great economic theories and how Global macro works and then you go well now do Japan and Japan usually throws them all out the window so talk a little bit about Japan a little bit too over over the last few decades yeah I mean Japan it's it is sort of like the poster child case study for like you know stock you know country that I think it's interesting about Japan like Japan's really nice like I haven't spent a ton of time in Japan but like it's not like it's a dump it's like a really the food's great everyone's really it's like a very high function you know like the day-to-day experience is very high functioning there uh from all the time I sp there it's great I love it I actually wanting to go back um but you know they had this 30-year period sort of 1989 to 2019 with negative 21% real returns um uh and you know that's not a I think what's interesting about that is um like oh wow that's so sort of uncommon um and that's actually so that's in the ninth percentile of distribution right so that's that's a one out of 10 outcome slightly worse you know whatever one out of 11 um outcome like that's not in insanely and you know what I mean like rolling a uh rolling two dice or something right this isn't like some crazy one out of a thousand sort of outcome like that's that's in the ninth sort of the ninth percenti distribution so I think that's that's sort of one important point I wanted to stress here is just um you know looking at Global Equity markets broadly historically and again who can say going forward um but the sort of General confidence people have that over even a very long time period like 30 years years the stocks are very likely um to deliver a positive return they are they're certainly likely to deliver a positive return that's what the circle would suggest maybe not as likely as people are generally expecting and I think you know the other thing I wanted to talk about here that I think you know we've talked about a bunch and is is really interesting is like okay you know we could take these really extreme scenarios again Japan's not that streem of an scenario per se but you you could take these like you know one you know first percentile scenarios of of really unlikely things um but just looking at you know I think I was Googling around and that you often see like oh you know you Google what's the average return of stocks or something you know something like that and like typically the number I see getting thrown around in the media is like 7 perish that's like typically what gets quoted and uh now according to the data from this paper that's actually like pretty close um that is that is pretty close to Accurate I don't I think it was like 7.12 or something 7% was like a reasonable um estimate but I think the interesting thing is you know when you talk about um average versus median or you know average versus typical right so you know um I'll bring up my favorite ergodicity topic here right but uh you know the fun ergodicity example is if you're playing a game of Russian Roulette and you have six people play one time or one person plays six times you know the six people play one time five win and one loses if one person who play six times is guaranteed to lose eventually right eventually they pull the trigger and the bullets in the chamber so as a one individual living one life through time you do not get the average uh returns of the markets you get what you get you get whatever happened on that um sort of path you are on so that the median the 50th percentile return um is substantially lower like a little less than 5% 4.91% um which is not I think you said yeah meb's number was 6% us and the US is sort of outperformed right so globally it's going to be a little bit lower um that's not crazy but then I think what's really interesting is the 20th 25th percentile is a ker of 2.02% so compound growth of 2.02% which over a 30-year time Horizon is an 82% return right so you're you know let's say you're 35 years old you're 4 4 years old you're retiring at 70 um you know take whatever's you know if you have a sort of stock F the 25th percentile outcome is an 82% increase uh in real terms and I think that's the one that's the one that I think if you surveyed people would be really surprising it was surprising to me like we worked on this for 10 years and even I was reading I was like Jesus like that yeah like a 75th percent I mean that's crazy 25 in yeah I mean I think I think it is surprising to people and again like the I'll I'll read some of the stats out here the the 75th percentile is an 800% return it's great amazing over 30 years um and that's actually pretty close to the average the 75th per says 7.6% KR I think I said the average or something I think it's something like 7.2% um but again you don't necessarily get the average you get what you get over sort of the the trajectory of what you live on so I think like you know one implication people this we talking about specific countries sort of global Equity diversification right and I think that that is a strong argument for that um sort of approach but I think it also sort of raises this question of like you know if you're setting reasonable expectations right like I think you know planning for a 25th percentile scenario is not like disaster planning right we're not talking about like some highly unlikely um wild scenario like we're talking about you know like flipping a coin and getting heads twice in a row right like that's uh that's the probability of this outcome and there's there's other things in there I think that like you said what if you run surveys people say like a seven 8% return what I actually think is more interesting that is if you actually a lot of the other like kind of surveys people just talk about the actual just arithmetic return just what the annual annual return is like they expect 10 to 12% I I mean usually was you hear 10 a lot I think people maybe just get fixated on like round numbers like 10 but it was either Meb or like Jared dillian or something they ran a survey that hundreds of people and it was up to like 12% was the expected return um but like you said like I think people a lot of times are just thinking about what the annual return is not thinking about the compounded return and like when we get into this ergodicity right the simplest way to think about it's not addition it's multiplication right you have to multiply your returns and that's how that volatility drag that we'll always get into is like it's reducing that compounding so when if people are even saying 10 to 12% um like nominal return like you're saying on the compounding that's more like 78% after inflation you're talking like 5 6% you know depending on what the inflation environment is over those decades that you're that you're using and then the other thing I want to be clear about is like you know we're not anti-patriotic in any way like we hope America like you know shoots the lights out for the next hundred years it's just like looking at a historical representation it's like you know you can't be so certain of that you know a priority and one of the ones I always think is interesting that we we like to bring up is like you know people show a 100-year you know back test of US stocks and it's that that up and to the right and once once again they're not looking at those indiv idual decades or their their their life cycle but it's also um I like to bring up since the Advent of like the Industrial Revolution let's say the mid to late 1800s to now we went from 1 billion people in the workforce to roughly six billion people in the workforce right that's really unprecedented in world history and you know depending on which statistics you look at you know we're supposed to have Peak population of N9 billion so the idea that you're going to you know have a five sixfold increase in the working population is is unlikely and how much that affects GDP and then America's ability to kind of Hoover up Global G GP is kind of unprecedented as well um so that's it's an interesting way of we're looking at does that mean we're we're calling for the demise of America or American Stock Market absolutely not it's just the this is the way we try to think about things these things with the perspective um what I thought was interesting recently is um you know they're putting out you know rip Charlie Munger they're putting out uh the newest version of Charlie munger's Almanac a revised version stripe is uh press is putting that out and on invest like the best they had John Collison doing an interview with uh Charlie I think it was just a few weeks before he passed and what was interesting is you know Buffett's known for recommending you know just indexing but uh Munger brought up exactly what we're talking about he's like you know this last 100 years of American exceptionalism is pretty unprecedented and if you're betting on just a broad index of stocks for the next hundred years he's like I he think he didn't think that was a very good idea um and so just another representation of that um but as we look at these 30-year periods and like okay what's the grand takeaway kind of right like what was like the lessons learned like kind of like the end of essay like what were you thinking about like using these 160 years of developed World example and these 30-year cohorts you know maybe a two% chance of losing money what does that what does that work out to like what's what's the real takeaway so I think a couple things one also there's a section in the the the essay that people can go read it also sort of looks at you know if you started the time sample at different I think they did like four different periods and you get roughly similar results right so it seems fairly robust um and then yeah as you said I think one thing that's like it feels very obvious to people or somewhat intuitive to people that um deal with markets a lot or interact with markets a lot is like it's sort of like what what's priced in or whatever so this is a tangent but like I'm reading a making of the atomic bomb by Richard roadr not which is great I would recommend sort of like a history of uh history of the atomic bomb and history of physics um but like America was like kind I don't know like my I don't know my my it's like a bit of a Backwater in 1900 right like it's not you know I think we like take for granted America the superpower whatever but like uh you know my guess is I don't have anything to substantiate this right but like if you did the equivalent of a Twitter poll in 1900 of like which country is going to be the most successful over the next uh The Next Century like I don't think America does great you know it doesn't do terrible in that poll but like I don't think it's number one um and I think right that that's part of know that's perhaps part of why it did well right like it was underpriced in certain sense that you know it outperformed and so it would make sense that the the returns were better but I think in terms of takeaways I think the biggest thing is just like I think you know sort of looking at this distribution here I think I think the 25th percentile one is like really the most interesting one for people to think about um that's just like a real I mean right that that's just historically and you can make an argument that whatever the world is different and um technology could change everything blah blah but you know you have to sort of looking at this data set say that's historically what it's based on and soort of like if you know if you're making retirement plans whatever it is like and you're doing an all stock portfolio that's sort of the the basis on which you're um you're planning and then I think that gets as we'll get into future episodes the next question is um you know instead of taking an all stock portfolio or taking that approach what are some ways we could diversify that and add some other assets in and what would that look like um to try and you know reduce that 25th percentile outcome effectively to to to create something more robust over time so we can leave that is the teaser uh for the next [Music] episode this podcast is provided for informational purposes only and should not be relied upon as legal business investment or tax advice all opinions expressed by podcast participants are Solly their own opinions and do not necessarily reflect the opinions of mut fund their Affiliates or companies featured due to Industry regulations participants of this podcast are instructed to not make specific trade recommendations nor reference past or potential profits listeners are reminded that managed features commodity trading Forex Trading and other alternative Investments are complex and carry a risk of substantial losses as such they're not suitable for all investors and you should not rely on any of the information as a substitude for the exercise of your own skill and judgment in making a decision on the appropriateness of such Investments visit Mutiny find.com disclaimer for more information