The Value Perspective Podcast episode – with Jake Taylor 2.0


Hero image

Hi, everyone. We are back from a short holiday and we are back with Jake Taylor, who also guested on The Value Perspective podcast in January 2021. You may know Jake as one of the hosts of the podcast Value: After Hours or as the CEO of Farnam Street Investments. In this episode, he gives Juan an update on Journalytic, the decision-making software he teased in his previous appearance on the show, and discusses how studying the returns from the S&P 500 over the past decade has given him a new framework for thinking about the future. It would not be a TVP conversation without some Bayesian probabilities chat and, finally, Jake touches on investing and climate change from his perspective as a former electrical engineer and his observations from the West Coast of the US. Enjoy!

spotify-podcast-badge-wht-grn-165x40.pngListen on Apple Podcasts

Chapter headings for Jake Taylor on The Value Perspective Podcast

Please click on the link below to jump straight to a chapter


JTR: Jake Taylor, welcome back to The Value Perspective podcast. It is a pleasure to have you back again. How are you?

JT: Juan, it is so good to be back – and thanks for having me on.

JTR: Where do we find you today?

JT: I am back home in California after a little bit of travelling – so trying to dig myself out from under the email pile that builds up while you are on the road and get back to doing some real work again. But it is good to be home.

JTR: Fantastic. We have had you on the pod before – we did our first recording in November 2020 and our episode went live in January 2021. For those who have not listened to that – and I would encourage them to do so! – could you give us a little bit of your background?

JT: Sure. I am the CEO of Farnam Street Investments, which is an RIA [registered investment adviser] in the US. I am also a co-host of a podcast called Value: After Hours – with my two buddies, Toby Carlisle and Bill Brewster – where on Tuesdays we get together and, for an hour, just basically chop it up about markets and different things that are happening in the investment world. And hopefully, by the end of the episode, maybe we are all a little bit smarter – or at least marginally entertained! And I am also the founder of a little side project I have been working on called ‘Journalytic’, which is some software that is hopefully going to help everyone make better investment decisions.

JTR: The thing you left out from that very kind summary is you are also the author of a novel, which you summed up last time around as ‘basically The Karate Kid – but if Mr Miyagi was Warren Buffett’! And you were also the host of another podcast, which I believe you have stopped doing.

JT: I did write a book in 2019 called The Rebel Allocator, where I basically was trying to help young people understand business and investing a little bit better through telling a story. And, yes, I had another podcast, called Five Good Questions, which has had a little bit of a hiatus because I have been so busy with these other projects. So it has been on the shelf and we will see when it comes back. I am not making any promises at the moment but it is a labour of love that I do enjoy doing so, eventually, I think it will come back.

JTR: We are huge fans of Value: After Hours, here on The Value Perspective. Myself, I literally listen to it every single week. I find it very useful, very educational, very entertaining. I already said that on our first pod but it is really good.

JT: Thanks, Juan. I appreciate that.

Track your decisions

JTR: We also touched on Journalytic in our first conversation but it was only in passing in answer to another question and, back then, it was only in its beta-test phase. So what is Journalytic and what has happened with it over the course of the last two years?

JT: Thank you. The whole impetus of the project was really trying to scratch my own itch of improving my own decision-making. So, you know, recording the data I wanted to know about myself and my process that were sort of a pain to do and keep track of – and just lowering the friction to be able to make sure I am keeping track of the things that then allow me to close feedback loops of understanding that let me learn faster about, you know, where are my strengths? Where are my weaknesses?

After I was working on it for a while, it was like, wow – this probably could help other people as well. I wonder if there is maybe even a commercial product that might come of this. And I have a couple of amazing co-founders. One guy is one of my really good friends – one of the best people I know – and he worked in private equity for a number of years, in a family office, and he has a really strong operations background. And then my other co-founder has been in the programming and development world forever, has tons of background in working at all the big tech companies and is just an absolute rockstar.

It is almost like having a genie and, whatever you want to build, he can say, yeah, we can do that. We have some employees now who are also working on it to speed things up. And we are actually in the middle of a seed-round to speed things up even faster – so doing a little bit of early venture capital. It is really exciting and it is super-fun. I think what we are building is really going to help a lot of people, which is what is most important to me. So yes – it is probably the highest upside project I have ever worked on actually.

JTR: That is really interesting. I have to say I had the privilege of being invited to test it a few months ago and I put one of your team in touch with my colleague, Vera German. I think it is full of great tools to improve your own decision-making – for example, it gives you the optionality to create your own journal so you can actually record in real time how you are thinking about different variables for a business and it also gives you the ability to assign different probabilities to expected outcomes. One of the most powerful tools – one I really enjoyed a lot myself – gives you all of these options to run checklists, depending on what you are looking for or what is important to you. So if you are more aligned to the way Charlie Munger tends to think and you want to run a ‘Munger’ checklist, or if you are very much into accounting, it will build a list of the different red flags you need to check. That is very useful.

JT: Yes. The checklist component is, I think, especially good for educational purposes – if you are a bit of a newer investor, we have included tons of stuff in there to learn from. If you got to understand why you asked yourself these questions in a checklist, you would have a pretty thorough understanding of the investment process. Like you said, we have probabilistic predictions you can make really easily. We have ‘contracts’ with yourself that you can make – there is this kind of joke in the investment world that, when it comes time to buy a company at the price you wanted it at, you are not going to want to do it, right? Just because of all the scary macro headlines that will be driving it down to that attractive price.

So one countermeasure to that behavioural bias is to create a contract with yourself – while you are calm, while you are thinking clearly, without all of the headlines that can be scary. And you set a contract that says, OK, if it gets to this price – or this metric or this corporate action, whatever it is you want to do – here is the action I will take. So you are pre-programming yourself for that ‘man overboard’ moment so you are in the driver's seat and controlling your emotions – and, I think, the biggest part of this whole game really is controlling your own emotions. It is really you against yourself – so this is a way to help that.

And then, of course, we have a bunch of the standard things you would expect in a typical kind of journaling, note-taping app and being able to tag things – but what really sets us apart is that, because we know this is specifically for the investment use case, we have the specific actions you can take, like recording your decisions, contracts and a bunch of other things that then provide structured data for us to serve up reports to you about your whole process and close those feedback loops. So it is called ‘Journalytic’ because it is journaling on the front end, and then analytics on the back end – that is kind of the vision.

But what I am most excited about is that Google has this natural-language processing, API, where you can feed it any piece of text and it will then assign a sentiment score – positive or negative – as well as a magnitude. It is based on the word choice and the word order so imagine, Juan, that you are just journaling about a particular investment idea and, as time has gone on, you are putting your thoughts in there and kind of making this your second brain. Well, our programme will then automatically look at how your sentiment has been changing and then overlay that – maybe even to the price – so you can see, gosh, look how much price is driving how I feel about this or how I am interpreting all the data or how I am viewing this company.

And I think what a lot of people are going to find is the price is dramatically driving their sentiment and they need to zoom out from that and stay focused on the fundamentals and the numbers and not let Mr Market sway them so much. But it is hard to do that in real time without seeing, gosh, look how my sentiment is changing, based on the feedback I am getting from the programme. At the end of the day, it is all about trying just to serve you up things about yourself so you know yourself better and can make better decisions.

JTR: Journalytic has been built on your own experience. You started the project and it has evolved over the last two or three years – and that evolution has come on the back of you living throughout the process on the investment decisions you have been making over that time. So how has the tool improved your own individual mental process – not just with investing but maybe life in general too? And then, while this is probably a bit subjective, which built-in feature have you found most useful?

JT: Yeah, it basically lives on my second monitor at all times – and I am just dumping things in there all day long. It is the first tab I open and the last one I close. I have found there are a few things that have been super valuable for me – like, number one has been to do a really good job of tagging different ideas. What it allows me to do then is kind of go down these little rabbit holes into my own thought processes and to see how things connect.

Let’s say I am looking at a particular idea and I can start to see like, hey, I wonder if this sort of rhymes with this other pattern I have seen. Well, that starts to emerge when you start tagging things with different notes – for instance, as a mental model, you could choose anything you want, but let’s use ‘antifragile’. I can look at a company and see like, wow, I think this is maybe starting to look like this other company I have seen that is antifragile. And all these things just start popping out and emerging because you are tagging the world and allowing you to sort it out into a latticework.

The other thing that has been really helpful has actually been feedback on when I say ‘no’ to something – so one of the key elements of recording decisions has been to put in a rejection or a ‘no-buy’, as we call it. I code why I rejected that and then go back and look and the software will automatically calculate what the opportunity cost was of rejecting on that day – because we know what the price is and we know what the price has become since. So we can start to see what the opportunity cost is of all of the things we said ‘no’ to.

And ‘those sins of omission’, as Buffett has said, have been his biggest sins – it is not sins of commission. So, for me, getting a sense of what the sins of omission are – what has been the cost of my filtering process? It just lays out all the stuff that we all know exists out there – like the data are there but we just are not capturing them. Like why did I say ‘no’ to that company? Unless you write it down and code it, you do not know – the bigger picture of all of the times that I coded for that reason and what that has gone on to do? We just do not know what the consequences are. So just laying bare the consequences ...

It is actually kind of comparable to ... I don’t know if you have ever done food-tracking before where you plug in your macros for food and you keep track of, OK, I ate this much protein, this many carbohydrates, this much fat every single day – and you really start to get a sense of, oh my god, I did not realise that was what I was eating every day. Like, that was the total calories and those were the macronutrients. Until you start measuring these things, you just do not have any cognition as to what they are. So, of course, you are not going to learn and make the smarter choice because you are just not giving yourself the data and the feedback.

At the end of the day, the whole project is to provide that kind of clarity about yourself and your process so you can have that awareness and then make the corrective action and change the habit that then leads to the progress you are looking for.

JTR: For anyone interested in hearing more about Journalytic, what should they do?

JT: Well, they can go to and get on our waitlist. We are still in beta phase so it is not completely open yet but we are getting there. What we really want to do is make sure we have something really kick-ass built before we go fully public for everybody but yeah, please reach out and get on there if you are interested in this.

Where do market returns come from?

JTR: That is really interesting. Now, I want to move on to something you discussed in your quarterly note last April and you have built on in your July note and also a Value: After Hours episode – it is this exercise to try to understand and explain what has been behind the 17% annual returns generated by the S&P500 index over the course of the last decade. And it is a really interesting exercise where you decompose the different variables that lead into that specific performance. Can you please walk us through that methodology, which is really powerful, and explain why it can be so useful – not only when thinking about markets, but even individual companies?

JT: In this exercise I am borrowing a lot of numbers from Chris Bloomstran of Semper Augustus, who writes this up in his annual letter to show, OK, just where did the composition of returns come from? So all of this comes from very simple accounting identities – like, we are not making any really big leaps but I find it very helpful to just see, OK, where do returns even come from? Like, what is the answer here?

And what the answer is, is really five factors. Let’s start by imagining those 500 corporations in the S&P500 all rolled up into being one corporation. So what is the total change in sales at the top line? What is the change in margin? What is the change in the share count – you know, the number of claim tickets chasing the ownership of this business? What is the change in the multiple, which is really a factor of sentiment? And then the dividend yield – you know, what was paid out to the owners?

And what we find is that, when we look at all of those components and we sum them up, we end up with the total return to an investor in the S&P500. Just to give you a sense of what those numbers look like over the last 10 years – and let’s call it from 31 December 2011 to 31 December 2021 – that decade, coming into this year, sales grew by 3.1%, which sort of jives with our notion of like GDP growing at 3%. Margin contributed 4% – so we had an expansion in margins over that decade.

Share count reduced to give seven-tenths of a percent attribution to return – so there were buybacks that happened over the last 10 years. The multiple expanded – providing 6.4% That is a huge number – like, that is almost the entirety of the normal amount of return you would get from the ownership of a business, in the six percent-ish range, right? And then dividend yield was 2.4%, which is about where it lives most of the time.

So when we add all those things together, we end up with a 16.6% annualised return over a 10-year period, which is absolutely phenomenal. The question then is, OK, what does the next 10 years look like? We know what it was for the last 10 years – can we do it again? And that was the question I was trying to answer in the April letter. And then in my July letter, it was, OK, if we do not like the answer to that – and there I am spoiling the punchline to the April letter a bit – but if we do not like the answer, what other ways to win might we be looking for, if the answer if it is not going to be from, let’s say, top-line growth or multiple expansion?

JTR: You then went on to do some ... I do not want to use the word ‘predictions’, but you were trying to figure out where that specific index could end up in 10 years’ time, as well as the probability that some of the historical performance may continue into the future. Could you walk us through your mental process in arriving at those numbers?

JT: Well, everyone could come up with their own numbers here – I mean, this is just what I felt comfortable possibly underwriting for the next 10 years. What I came up with was, let’s call it 3%, again, for the sales growth. And, by the way, here is an interesting fact: from 2000 to 2010, which was a ‘lost decade’ for the S&P500, if you recall – like returns went nowhere over that 10 year-period – top-line growth was actually stronger than the 2011-2021 number. Actually, top-line growth was stronger during that period than this boom period. So, for everyone who thinks it is all about revenue, I would say that is probably a bit short-sighted.

Anyway, let’s pretend then that we can expect sales to continue growing at 3% and kind of stays with GDP. Now, margin ... this is a tough one because Jeremy Grantham has said margins are one of the most mean-reverting data-sets in finance, and Warren Buffett has said only a fool would believe profit margins for corporate America can be much more than 6% – yet, currently, we are at 13% and we have been above 6% for a long time.

So either Buffett has lost it or we should maybe expect some reversion to the mean in that number. Now, whether that comes from inflation of cost and energy inputs or reshoring or deglobalisation or who knows why – there are lots of possible ways where we can just not expect corporate America to be as profitable as it has been. So I am fading that by 3% contribution – and even that does not bring us all the way back to 6%. I am just saying, OK, over the next 10 years, I do not think we are going above 13%, that’s for sure. But I am open to arguments against the 3%.

JTR: For me and many other value investors, top-line growth – indeed, growth in general – is more difficult to forecast, just because it is speculative in nature. I mean, it can go anywhere. With margin, though, you have so many data points – you can mean-revert the margin; or you have history on your side so you can look at what the market has done or individual companies; or you could look at the peers of companies for where the sector has been; or you can have a sense of what the base rate is. Yet you are saying margin has been more difficult in the past to gauge than top-line sales growth?

JT: Yes. I think margin has probably resisted reversion to the mean a lot longer and stronger than most people would have expected – certainly more than I would have expected.

JTR: I know you have a theory on why that might have been?

JT: OK, let’s look at the constituent components of the S&P500 – a big chunk of them are made up of the large tech companies, most of which are extremely profitable. So, if they are running 30%-plus margins, in some cases when things are going really well, that is a big weighting on the profit margin for the S&P500 to keep it buoyed up. So it is not surprising but, I think, the question is – do those mean-revert? And, you know, does that come from competition or regulation or who knows what? Or disruption – I mean, this is tech, after all – the whole point of technology was to disrupt incumbents, right?

And 10 years is a long time – a lot can happen in that period. Thus far, they’ve been giants and have not been dethroned but the same could have been said before – the biggest constituents of the S&P 500 turn over, you know, roughly every decade. The nature of markets is it is incredibly hard to stay on top – the world is gunning for you whenever you are on top. We will see – I mean, there is this counter-argument about returns to scale and network effects and I think all those things are very real. It is just, you know – what is the duration of the competitive-advantage period these companies are going to have? That is going to dictate whether margins probably stay as elevated as they have been.

JTR: I’m sorry – I interrupted you earlier. You had top-line growth at 3% and margins coming down – how were you thinking about the other elements of return?

JT: Sure. Share count, which was a contributor in the last decade – I would probably put it flat for the next 10 years. I mean, we added a lot of leverage in the last 10 years to do those buybacks that lowered the share count. I would not be surprised if, at some point, companies actually had to issue equity for survival – we saw that in 2008 – so let’s just put that as a wash at zero. Let’s put yield at the typical 2% that it has always lived around, for dividends – and then there is PE [price/earnings] multiple.

OK, if we look at any valuation metrics, there is no good timing valuation metric – when you think about things like market cap-to-GDP or CAPE or Tobin’s Q, they all stink for timing. They do not tell you anything but they do all sort of rhyme and they all tell you the same thing – that markets are pretty expensive, even with the corrections we have seen so far. So I, personally, am fading that multiple – though even that does not get us all the way back to a long-run kind of average.

But let’s say it is a 4% reduction – I mean, we saw 6.5% for 10 years going up so is it not possible we might see 4% going the other direction for the next 10 years, just to get back to normal? Then you add all that stuff up and you end up with minus 2% annualised, which is almost minus 20% total over a 10-year period. That might well be shocking for a lot of people – to think, 10 years from now, the S&P is 20% lower than it was at the start of 2022.

JTR: But then, over the last seven months, the market is actually down even more than that.

JT: Yes – who knew that, around three months from me writing all this, we would get all of that number? I guess the question now is – are we done? Or is there more pain before we can get to the next bull market? That is a great question but one I do not have a lot of good answers for.

JTR: One of the useful things about this exercise is how it informs your decision-making process – trying to figure out what numbers lie behind some of these assumptions so you are thinking about all these different components and are able to make an informed decision going forward.

JT: Definitely. I just wrote up a case study of a company that was able to do well – even without a large top-line growth and without a bunch of multiple expansion. And both of those – the big top line and the big multiple change – are sort of hallmarks of a bull market. So, if we are not going to have that for the next 10 years, what are some other ways you can win? And, in my case study, this particular company grew at a 6% annualised growth rate from 2005 to 2021.

Profit margins hovered around 17% to 20% so it was a reasonable quality business, if nothing spectacular – but the big thing was the share count went from 80 million down to 23 million shares. So that was a 7.5% CAGR reduction in the share count – just absolute Munger cannibalism. In fact, the company spent $22bn, which was about 85% of its cashflow from operations – basically pointed the capital allocation firehose at buying back shares.

Now, buying back shares is not an unalloyed good – it has to be done at the right price. Otherwise, value is basically walking out the door. So this company was blessed with a relatively low EV-to-EBIT multiple for most of that time period – on average, it was retiring shares at about 11x to 12x EV-to-EBIT. So, pretty cheap and, naturally, if you did not sell back to the company – if you held on for that entire period – you ended up with a 20% CAGR over that time period, which ended up being like a 2,100% return. You absolutely crushed it for almost 20 years.

Of course, it is a lot easier said than done, right? Because the first four years of that time period, the stock went nowhere – it was like 8% cumulative return for that time period. You looked like a bozo. There were multiple 30% drawdowns over that time period and lots of just dead doldrums where nothing was happening – yet, behind the scenes, the company was growing a little bit; the margins stayed steady; it was retiring share count like a beast; and, in the end, lots of accretion was happening under the surface. You just had to be patient and ride it out.

My takeaway from all that is, basically, if you are not going to get huge top-line growth and you are not going to get multiple expansion for the next 10 years, how can you win in a down-market? And the answer is – find a steady, boring, reasonably profitable business that has good capital allocation and is retiring shares when it is cheap. That way, you wake up every morning hoping to see the price is down because you know that means they are buying back at a lower valuation, which means it is even better for you as a remaining shareholder. So that is one pattern that you might be looking for over the next decade as a way to win,  if markets are not going to be giving you this big tailwind.

Shedding light on blind your spots

JTR: I want to build on something you have also written about in the past. At some point around 2015, you came to the conclusion that, based on different variables, value investing itself as a factor was not that attractive – and you were right about that. But, at that point in time, some of the context and the landscape of what you were looking at became a blind spot that stopped you from doing some investments that could have worked very well – and you have mentioned Google as example of that. Now, the exercise you have just described is pretty much about avoiding extrapolation – don’t just take that very good number from the last 10 years and assume it is guaranteed for the next 10. Think about what could be the different drivers behind it –and, just for compliance reasons, that is just one possible scenario that could play out and we live in a world where a lot of things could happen. So my next question is, based on that, what is the best way to avoid blind spots? I guess Journalytic might be one of the key components of that!

JT: The intention of the software is definitely to help reveal those but there are a couple of things to unpack in your question. The first one is the 2015 observation I had, which was that the quality of the value opportunity set at that time period was especially weak. And what that had to do with is ... we often look at how expensive the general market is, right? But what I think gets lost in that is there is a bell curve around that average you have to take into account – and that bell curve can have tighter or wider distribution.

We are talking about valuation here so, when the bell curve is especially tight ... let’s say the market average is 15x but the cheapest 10% – that tail of the bell curve – is at 14x and the most expensive 10% is at 16x. So we have a very, very tight spread around that average. Well, that is one opportunity set. But then, let’s say the market average is 25x – so a much more expensive market – but the cheapest 10% is down at 5x earnings and the most expensive is at 50x. OK, now we have a very widespread.

The important thing about my insight was that it is not just the market average – it is what do the tails look like? So looking at the distribution with a little bit more granularity and looking at the dispersion of valuations – between the most expensive and the cheapest. And what I found was that, in 1999, you had a very expensive market but you also had an incredible dispersion – and what that came from was the narrative at that time that the ‘new economy’ stocks were going to take over the world. So they were very expensive and they were dragging the averages up.

But then you had a bunch of bricks-and-mortar, boring ‘old economy’ stuff that was just being thrown in the garbage – and there was tons of value to be had there. So naturally, there is an entire vintage of value investor gurus who had 1999 start dates – and I think a lot of that had to do with the fact that the tailwind of the opportunity set they had at that time was just so strong. So even in an expensive or what you might think of as a ‘bad’ market – because it was going to correct – you still had a ton of value you could buy at that time. And what we saw was that those guys had positive returns when everyone else had negative returns – and that is how you get to look like a genius, right?

Now, fast-forward to 2008. We had a much tighter dispersion of valuations coming into 2008 and so you did not see that same kind of value outperformance from the guys that were value investors over the next few years because their opportunity set was not that good. And, when I wrote in 2015, it was even tighter – it was an expensive market with a very tight distribution around it. And so what I said was that value opportunity set was not projecting to have very good returns from there.

That ended up coming true but what I was not smart enough to realise was that, well, if the dispersions are tight, that probably means some of the higher-quality, maybe higher-growth, companies are mispriced to the other side and maybe there is actually a lot of cheap optionality embedded in those that I could have bought. And one of them happened to be something like Google where it was mispriced and I wasn’t smart enough to take the leap from, oh, value is not good right now, therefore look somewhere else. I just threw up my hands and said, oh, I guess hold cash because there is not a lot of cheap stuff to buy. I wish I was smarter back then!

JTR: And how does that distribution look like right now?

JT: The last time I did this analysis, it was an expensive market with a reasonably wide distribution between most expensive and cheapest. But I do not think the cheapness of today’s value set is as cheap as the cheapest was in 1999. So if I had to guess and this is very rough – like, we are painting with a crayon here – I would say it is probably outperformance from value, but not amazing absolute returns.

JTR: Interesting. Going back to the original question, then, how do you avoid the blind spots?

JT: Let’s jump back into the Google example. What I missed for the next five or six years from 2015 was that I was trying to be a good kind of Bayesian forecaster – you know, what can I expect for Google's top line? What might I expect for profit margins? What might I expect for share count? So all of that same exercise we talked about earlier – and I was putting a lower growth rate on top line because, when you look at the base rates of big companies, it is really hard to keep growing at such an amazing clip. Like, when you get that big, it is hard to keep putting up 20% annualised growth rates on your top line because you just start to saturate your market – you know, elephants can only get so big and trees can’t grow to the sky.

Well, what I missed was that the base rates for some of these tech companies defy all of the other base rates you would have looked at historically. And that then comes back to our conversation about returns to scale and network effects and some of these other technology-enabled profits that historically have just not been available. So I was using the wrong set of base rates to inform my decision-making – that was a blind spot.

So my mistake came because I was trying to do what I thought was right but I was using the wrong data for my analysis and I therefore ended up being wrong. It was a sin of omission I would pick up later – and I probably would have picked it up sooner had I been a better journaler at that time and been documenting why I was passing. It is possible I would then have discovered the error faster and maybe been able to correct it and not keep making it for five more years – a ‘25% compounded a year’ kind of mistake in opportunity cost!

Acknowledging complex systems

JTR: That is really interesting. You are famous for your ‘Veggie Segment’ in your Value: After Hours podcast and, during our first session, you explained what these Veggies are, why they are so useful and where you source them. This might be an unfair question but, from all the different sciences you have studied, which one is best able to improve your process and help you understand markets and human behaviour? On top of that, which area is your favourite and which do you reference the most?

JT: I am going to punt a little bit and pick one genre, rather than any one segment – and it has to do with complexity and chaos and, you know, complex adaptive systems. The reason that is so key to understanding the world is their non-linearity – so, like, the input can be very small yet you end up with an output that is much outsized – and you can run the same simulation a million times and end up with huge discrepancies in outcomes. As opposed to a linear system where it is same input, same output, same input, same output – so you end up with much tighter bands.

All of which is to say, just stay humble about forecasts you are making and realise the world can be much more random, much more chaotic, much more complex than you might give it credit for – and, certainly, than what you will see on Twitter, say, or CNBC. The world is much less deterministic, I think, than what you would hope or want it to be, which means: bring your overconfidence levels down a lot – do not be overconfident.

There is this idea in complexity theory called the ‘Lyapunov Window’. And that is, for every input you have into a complex system, the tighter you can measure those inputs – like, the specificity with which you can measure them – the little bit longer you have with the predictability and what is going to be the output. So I would say that understanding as much about a business as you can is giving you a better granularity – a better specificity – into the inputs of the model and that then allows you to have a little bit more of a window.

So basically, ‘know what you own’ is what I am trying to say. But you have to recognise that the window is not that big and it is probably impossible to be making predictions and putting terminal values out on businesses 20 years from now – I mean, this is just me but, when I see research that is 15 years from now, saying, you know, the TAM [total addressable market] is going to be this big and they are going to have this percentage of it and this profit structure, I feel like you are fooling yourself a little bit there.

And so having a deep appreciation for complex systems just keeps your ego in check and probably keeps you from making some of those errors that I think – in the last six months, especially – had been revealed as, boy, that was a little bit of wishful thinking and maybe the world is not that predictable. Some of those story or narrative-based stocks have been the ones that have been hammered the most and may have turned into some of the biggest pain points for people. So it just keeps you humble and keeps you into things that, hopefully, you know something about and you have a little bit of predictive power – but not assigning so much that you think you know how the world is going to look in 10 years’ time.

That is, I think, why it took so long for this idea to become mainstream. And what ended up happening was that in the, call it the 1980s, and especially into the 1990s, you had the Santa Fe Institute, which was really pushing this and the reason they were able to make progress was because they set out to be multidisciplinary from the get-go.

So they pulled in physics, which has complexity in it; they pulled in biology, which has complex adaptive systems; they pulled in markets, which are represented by complex adaptive systems; they pulled in computer science, which has that same element – they pulled in all of these different disciplines and recognised that, oh, this explains the phenomena much more than any one discipline did by itself. So they were able to make tremendous progress in their understanding. That is probably also the part that appeals to me the most – that it covers such a wide range. That makes it a lot more fun for me to study and learn about and try to draw analogies from for us as investors.

Putting probabilistic thinking into practice

JTR: That is really good. I heard you talk about The Scout Mindset by Julia Galef and read it right away – and thinking in probabilities has been a constant topic of this show. Could you walk us through your thoughts on the applicability of Bayesian thinking, which you mentioned earlier, when it comes to investing and how challenging it actually is to put into practice?

JT: Yeah. I really liked Julia’s book a lot so I did a segment on it. I think what is really hard about the investment process is we are seeking to find that ‘outside view’ – right? That is, a way of getting out of our own little narrative tunnel and really trying to see the truth of whatever the situation is – whatever the truth is for that business or that marketplace or whatever it is. And what is hard is that we all have blind spots and, by definition, they are hard to fix – because, otherwise, they wouldn’t be a blind spot.

Now, there are multiple ways of seeking the outside view. One is to talk to other people who maybe understand parts of it better than you do. Another is to read widely so you can find different tools – you know, if you only have a hammer, then everything looks like a nail, right? But if you have different tools, you can start to uncover blind spots and try to see things through a different prism. And then you have base rates, which is basically – of all the times this type of situation has occurred, what has been the outcome? And what number of times has that been the outcome?

The trick is to try to refine the base rate you are using to get closer and closer to what the actual truth looks like – and that is what Bayesian probabilities are all about: starting with a top-level base rate and then working your way down with data to truly understand, OK, as I get closer and closer to what the actual phenomenon is, what can I start to expect? And that is really how your brain works as well – like, it starts out with very basic, fundamental sort of Bayesian ideas of, oh, if I drop this ball, then 100% of the time it is going to fall to the ground, and you start building up from there. And eventually, by the time you have a fully-formed brain, hopefully, you are understanding a lot more nuance of the world and those probabilities start being reflected.

And, with the brain being a Bayesian updating machine, this is why surprise is so important – because surprise is the feedback telling you, hey, one of your base rates is wrong. What you thought was going to happen did not happen – like red alert, throw up a flag, we need to update our models. That is why failure is so important – and a big part of the software I am building is to uncover those surprises and kind of rub your nose in your mistakes a bit by going back and looking at what your thinking was at that time so you gain a better understanding of why you were surprised and therefore you update your models better. That is the whole idea of closing the feedback loop to learn faster – you recognise your mistakes faster and then you take measures to fix them.

JTR: This is something we ask everyone – just because probabilities are so difficult to grasp – but, from your own experience, do you have any tips that could help investors become better at adopting probabilities into their mental and investment process?

JT: I think just writing them down more regularly is the key. I mean, a lot of times, probabilities come from intuition – you have all these mental models and there is a lot of stuff happening at the subconscious level that you do not even really have access to with your cognitive, executive function. So just writing them down – and then going back to see how they turned out – will start to close that feedback loop. And intuition does not work unless feedback is provided to start building that intuition up – if you are just making intuitive guesses in a vacuum, I don’t think you ever get better at it. So I think you just have to start writing it down and keeping track a little bit better and then you can start to trust your intuition a little bit more.

An electrical engineer’s take on ESG

JTR: That is really powerful. I am going to be very cheeky now but I cannot pass up the opportunity to ask you about the debate permeating markets over the last two years around the topic of ESG. The reason I bring this up is because you are an electrical engineering engineer by training, if I am right?

JT: Yes – before I became a professional investor, I ran the power grid for the state of California.

JTR: OK, perfect. So, having started your career in the utility space and working with the grid, you understand the maths and the physics behind power generation and what it takes – plus, you live in California, which is front-and-centre with many of the things happening in ESG. So, with that sort of perspective, what are your thoughts around narrative-versus-reality in terms of the transition towards renewables, how feasible that actually is and the second-order consequences of what is happening?

JT: More generally, I feel like ESG is a bit of a bull-market luxury – like, ‘I want my returns but I also want to feel good about how they came about’, right? And I will be very curious to see – if what we talked about earlier comes to pass and maybe the returns for the next 10 years are not particularly strong – does the appetite for ESG then fade a little bit as well? That would be pretty on-point for human nature – but we will see.

More generally about energy, I think one of the difficult realities about our current world is how dependent we are on fossil fuels to create the modern abundance we have. I mean, it is untenable to imagine the current world without fossil fuels as it exists today. The affluence so many people experience ... even food is hugely influenced by fossil fuels – like, we are 100% dependent on feeding seven and a half billion people or whatever it is now using fossil fuels – and that has to do with the energy required, as well as the ammonia that needs to be produced for the fertilisers. Natural gas is a huge component of that too.

I am very hopeful we will be able to transition away from fossil fuels – but there are no silver bullets for that. I think it is going to be a lot of little things – hopefully, nuclear power is one of the answers. Also, increased renewables, better battery storage – actually, reducing food waste would be huge for saving energy. So it is going to be a lot of little things that hopefully add up but it is probably going to take 30 to 50 years, if we stay diligent and work on it. And anyone who wants to do it in the next, you know, five or 10 years, unfortunately, I think what they are implicitly saying is they are OK with starving and keeping poor the lower one to two billion people on this planet.

I find that to be quite objectionable – and I hope that it is just a lack of understanding of the realities of the physics of the world that we are in today and it is not, you know, something more devious than that. I do not think it is. But, I mean, we just have to recognise the reality that the modern world just runs on fossil fuels at the moment, and it is going to take time and ingenuity and human effort and persistence to transition off them. I think we can do it but it is going to create a lot of human misery if we try to go too fast.

JTR: Everyone is concerned about climate change but, when it comes to investing, do people need to differentiate their own concerns on whether or not something is a good or a bad investment? In other words, if you compound both ideas, do you risk getting into a bit of an emotional conundrum, which might or might not be very good for the investment itself? It is not lost on us, for example, here on The Value Perspective, that while Warren Buffett has never been very big on energy, he has made an energy company one of his top holdings over the last 12 months.

JT: More generally, you are saying, don’t let your politics run your portfolio? I think that is probably wise. But, ideally, your politics are run by the data of the world you see around you and, you know, what reality is between economics and human interaction and the policies we think will allow us to coordinate our efforts in the best way we can so the maximum human happiness and utility is created with a minimum amount of externalities for everyone else to deal with. So, if that is your stance, then maybe it is OK to have a little bit of politics in your portfolio. Generally, though, I think it is good advice not to mix those two and not to let that drive the bus for you. Still, it might be a little bit optimistic to think that, you know, physics and economics and the reality of those things are driving everyone’s politics!

Book recommendations

JTR: Jake, we are coming to the end of our session and, as you may remember, we always ask our guests two signature questions. As this is your second time around, however, I will spare you the one on an example of an adverse outcome to a decision that was down to poor process rather than bad  luck. Instead, given you are such an avid reader, could you please give us at least one book recommendation? Last time out, you actually gave us three so that is what you are competing against!

JT: OK, I will do better and give you two this time. The first one is this book called Behave: The Biology of Humans at Our Best and Worst by Dr Robert Sapolsky, which is an examination of human behaviour. He slices the field up into different time periods – so, in real time, he looks at what is happening in your brain, in your blood chemistry, in your physical body and in the environment around you, right as you are making a decision and taking an action.

And then he rewinds it – like, what happened 10 seconds before that? – and he looks at all the same components. And then let’s go back a little further – like, a week ago and then a few years ago and then your childhood. And then, like, while you were in the womb because what was happening to you then will actually later dictate things like how you view the world. And then your genes before that and then even deeper back into, you know, where did your ancestors evolve, what were the inputs that drove their evolution and how is that impacting your behaviour today?

So I am an unapologetic Sapolsky fanboy. He has this amazing class that is online on YouTube. He teaches at Stanford and he has this series that is, like, his entire class for one semester, where he goes through a bunch of neuroscience and neurobiology and this book is a distillation of a lot of his research. He is actually a primatologist by training – like he spends his summers in Africa, studying apes, and then comes back and teaches and writes and does all this stuff. So I love him and this is my favourite of his books. It is a little bit long because he is covering so much ground but the amount of research that has gone into it is just absolutely amazing. So if you are interested in why you do what you do, this book is kind of a must-read.

The second recommendation is a little bit more fun and a little shorter and it is Ed Thorpe’s autobiography, A Man for All Markets: From Las Vegas to Wall Street, How I Beat the Dealer and the Market. And what is fun about that is just to have such a steel-trap mind applied to different problems in the world and how Thorpe solved them – I think that is really helpful for us as we look for what our edge is and, you know, how can we stand out and have a bit of an unfair advantage? Thorpe was always looking for that in every domain he was in so he is also a hero of mine. So I have just picked out two of my favourite people and then picked their books for recommendations.

JTR: Those are fantastic recommendations. Jake Taylor, thank you very much for coming back to The Value Perspective podcast.

JT: Thanks for having me on, Juan. It was so much fun.


Important Information:

The views and opinions displayed are those of Nick Kirrage, Andrew Lyddon, Kevin Murphy, Andrew Williams, Andrew Evans, Simon Adler, Juan Torres Rodriguez, Liam Nunn, Vera German, Tom Biddle and Roberta Barr, members of the Schroder Global Value Equity Team (the Value Perspective Team), and other independent commentators where stated.

They do not necessarily represent views expressed or reflected in other Schroders' communications, strategies or funds. The Team has expressed its own views and opinions on this website and these may change.

This article is intended to be for information purposes only and it is not intended as promotional material in any respect. Reliance should not be placed on the views and information on the website when taking individual investment and/or strategic decisions. Nothing in this article should be construed as advice. The sectors/securities shown above are for illustrative purposes only and are not to be considered a recommendation to buy/sell.

Past performance is not a guide to future performance and may not be repeated. The value of investments and the income from them may go down as well as up and investors may not get back the amounts originally invested.