The Value Perspective Podcast episode – with Tom Gayner

JTR: Tom Gayner, welcome to The Value Perspective Podcast. It is a pleasure to have you here. How are you?

TG: I am well, thank you so much.

JTR: Tom, where do we find you today?

TG: Fortunately, today I’m in Richmond, Virginia. So ‘home sweet home’.

JTR: Tom, there will not be many people in the investment community who are unaware of your work but, for those who don’t know you or have never heard of Markel, please could you provide us with a brief summary about yourself and the company?

TG: Certainly. Markel is an insurance-based holding company. Public since 1986, the company actually started back in 1930. There were three generations of family leaders who ran Markel and built the insurance business from 1930 up until a decade or so ago. I started at Markel myself in 1990 to help out with the investment side of the operations and the investment side grew and morphed into Markel Ventures, which is where we became a holding company to own several non-insurance industrial and service businesses. We started out in 1986, with a market capitalisation of a little over $30m (£24m) and today it approaches $20bn.

We do business all over the world. We insure creative and interesting things that tend not to have a home in the standard insurance markets. And we build a variety of products and services that do things – for example, we bake bread and rolls that you would find in a fast-food restaurant or on grocery store shelves, we are the largest grower of indoor house plants in the world and we provide IT consulting services, medical services, precast concrete and so on and so on and so on. So it is a long story but I hope that gives you a brief snippet of what Markel certainly does and is willing perhaps to do,

JTR: Tom, as you just mentioned, Markel is known for insuring things that do not have a home in standard insurance markets – what is the most unusual thing you have insured in your 30-year career?

TG: Well, it is a long list as there are hundreds of different product lines and areas we would tend to insure. Still, one of the stories that gets told is, if you remember The Wizard of Oz, Judy Garland’s Dorothy character wears a pair of ruby-red slippers. Apparently, in total, there were six pairs used during filming and one pair was held by a collector at some point in time – and we were the insurer and they were lost. So that was a claim we paid on a product where otherwise, if you were looking to find insurance for collector’s-item ruby-red slippers from The Wizard of Oz – well, I have never tried to Google search that! I don’t know what you would come up with but Markel would be the answer if anybody was willing to do something like that.

JTR: That’s really interesting.

Co-management and the ‘special sauce’ of culture

AW: That’s fantastic. It is a really eclectic and interesting mix of industries and things you insure. I would like to talk about a subject that is very close to our heart, here on the Value Team at Schroders, and that is co-management. Our team has co-heads and all of our portfolios are co-managed. Now, I know you managed Markel’s investment portfolio alone for 17 years – and then you were the co-CEO of Markel Corporation for six years before becoming sole CEO at the beginning of this year. It strikes me, then, there is probably no better person to talk to about the pros and cons of co-management. So what would you say are the ingredients for success with co-management and what are the drawbacks?

TG: Well, even people who are CEOs in a sole role, as I am at this point, are really just running the team. And there is no CEO, there’s no quarterback, there’s no coach, there’s no point guard, there’s no centre of any functioning team that does not have teammates around them. The best teams have heads who have recruited the people around them with some care – and part of that care is not just the skill those people have, but the willingness to listen to them, to involve them in your own thought process and to make sure you are taking advantage of all of the intellectual capital you surround yourself with on the field of play.

So the number-one formula for a co-structure to work – whether it is formally identified as such or whether it is just as a matter of practical reality, as the way the world works – is respect for the people on your team and listening to them and taking their counsel and different points of view and experience into account as you make decisions. That said, one of the things that might be something of an advantage to the sole CEO of that structure – or the head coach or starting quarterback or point guard – is decisions oftentimes need to be made instantaneously. So you should have done your homework and have every bit of background information you possibly can have. But sometimes everybody in the team is going to look at somebody and say, What do we do? And for somebody to say, This is what we’re going to do, and choose the path – even if that’s not a perfect decision – is sometimes what’s called for in a moment. So pluses and minuses to both structures.

AW: It is interesting to hear you talk about that because you can make quicker decisions, certainly, if you are the sole CEO, but you still have that team. It sounds like what you are saying is there is not a huge difference between the mindset of being a co-CEO versus being a CEO – is that fair?

TG: I think that is correct because the important thing is, there is a label that seems rather binary – you are either the sole CEO or you are a co-CEO. Yet the reality is, if you are the sole CEO, it would be foolish not to have a circumstance of co-operation with people you have surrounded yourself with, to take full advantage of the wisdom they bring to the table. So it is not as much two ends of the spectrum as the labels might imply.

JTR: Tom, in your experience how much does business culture and narrative impact decision-making?

TG: Well, I think the ‘secret sauce’ of Markel is the culture that is here. So back in 1986 – when, as a matter of fact, we might technically have had the label of one CEO but, in practical terms, it was like ancient Rome, where there was a ‘triumvirate’ of three ‘consuls’, in the form of Alan Kirschner, Steve Markel and Tony Markel, who effectively were the three CEOs of the business – and as part of the IPO process and going public, they explicitly wrote down our creedal statement, which they called the ‘Markel style’. And in that effort, what they were really trying to do is recognise there would be a time when none of the three of them would be here – but they wanted Markel to still be here and the company to still carry on.

So they wrote down a series of guidelines and principles – the ‘creed’, if you will, of Markel – so that those who came after them, like Richie Whitt and myself, would understand the culture and the values of Markel. And frankly, that informs every single decision and action we make from the top to the bottom of the company. And I don’t think there is anybody of the 20,000-some people who work for the company who don’t have some exposure and awareness and understanding of what the Markel style is all about – and the values and the way in which we do things and hold ourselves accountable. And fortunately, for someone in my position and who has been here 33 years, those roots are deep. For someone who just started this morning. it might not be quite as deep – there is no way it could be – but, over time, those lines should converge.

A ‘doughnut’ truth – and the map is not the territory

AW: Tom, you have previously described financial statements as a ‘doughnut truth’ – so the truth but not the whole truth. Now, this is about understanding what might and might not be visible within financial statements – which I have heard you refer to as the whole story or the narrative. That is really interesting to us because we know narratives can be very compelling – human beings communicate through storytelling and narrative – but they can also be very dangerous. A year or so ago, we had as a guest on the podcast Bethany McLean, who uncovered the Enron scandal – and that was obviously the much darker side of narratives, if you like. Do you agree there can be this friction between the numbers and the narrative – and, if so, how do you account for that when looking at companies?

TG: To start with your premise of a ‘friction’ between the numbers and the narrative – there is an element of truth to that but there doesn’t have to be. So, for instance, I think many in the investment business, especially within the last five or 10 years, have become quite sceptical of the narrative – and for good reason. You just referred to Bethany McLean and Enron – where there was a narrative that convinced people something was true when, in point of fact, it wasn’t. And the numbers perhaps did suggest that – eventually. But the numbers – at least for a while – probably were in concert with the narrative enough that people got fooled.

So the narrative in and of itself is neither good nor bad. We have adopted our own practice to try to have some detachment and non-emotional reaction to what somebody’s narrative is because we have seen – and perhaps all been scarred by – examples of where somebody was pretty good at telling the narrative but the numbers weren’t backing it up. I think it is also often true that there is a pretty good narrative and the numbers do indeed back it up over the fullness of time – in which case, the narrative is not a bad thing at all. And to the extent it allows people to understand and see things and get the picture a little bit more than just looking at the numbers, well, it is fine. So narrative is neither for good nor evil – it is just a tool and can be used either way.

AW: And, in your experience, what exactly is it that tends to be missing from the numbers? What extra information do you obtain from finding out that ‘whole truth’?

TG: Well, let me talk specifically about Markel as an example. I have written about this in our annual report for the last several years and my colleagues groan whenever I start a sentence around here that begins with, ‘You know, I used to be an accountant and ...’! And whatever follows the ‘and’ is where I am pointing out what I believe to be some of the current flaws of GAAP accounting. So, in the realm of GAAP accounting, you are trying to describe economic reality – but remove yourself from accounting presentation and numbers and there is a line that preceded this conversation and probably preceded my lifetime: ‘The map is not the territory.’ So the map is a representation of what the territory is and a good map will give you a good sense of the territory – but, no matter how good the map is, it will never fully describe the reality of the territory.

I think that is what we are talking about here. The numbers that describe the business are essentially maps of the business itself, the human beings running it, the relationships they have with their workforce and clients, the sense of marketing mindshare they have in their customers’ mentality – all those things are rather difficult to quantify immediately but the numbers probably show the direction in which things are travelling over time. So I just think it is very important to recognise there are maps out there, there are territories out there and, if you were doing a Venn diagram, there is some overlap between those two things – but they are not the same things. They are different words and they are different words for a real reason, in that they are different things.

AW: That is a cracking analogy. Thank you.

JTR: Yes – I think we will be stealing that analogy going forward – but we will give you the rights!

TG: I appreciate that! The old joke about that sort of thing is, the first time you quote somebody, you give them full credit; the second time, you just sort of say, ‘It’s said that ...’; and then the third time, you just say it yourself, so I appreciate you at least quoting me for the first time!

Learning the right lessons about probabilities

JTR: We will do our best! Tom, this podcast originated four years ago after we met someone you have mentioned in the past, Annie Duke, after she released her book Thinking in Bets. We thought many of the insights she described from a poker mentality had a lot of applications to the world of investing. That was a revelation and, just off that session, we decided to launch a podcast to address the topic of decision-making under uncertainty – trying to understand how people were making decisions in very uncertain environments and seeing what we could apply ourselves, not only to investing but also to becoming better decision-makers in our day-to-day lives. A central element of the book is, to make better decisions, you need to try and think in terms of probabilities – but, while that is all fine in theory, it is extremely difficult to execute in practice. So how do you approach probabilistic thinking yourself? And do you have any suggestions for anyone trying to do so but maybe not executing it correctly?

TG: Great question – I love this topic. First, Annie Duke’s Thinking in Bets – that is a really good book. She did the world a great favour by writing it because what she did was try to articulate the way she thinks about things. So again, it is an examples of narrative – she was giving you a narrative about the quantitative thought process. It is not the process itself – it is a description and a narrative of it – so there is an example of the narrative being a force for good rather than a force for evil.

To your question about how you apply that in real life – first off, don’t hold yourself to a standard of perfection. As she points out in her book over and over again, when you are playing poker, you need to play in such a way that you can play a lot of hands so the odds and the numbers and the probabilistic decision-making do start to work out for you over time. In any given hand, you can calculate the odds correctly, you can bet correctly, you can do every procedural step correctly – but you can still lose that hand. Now, that reality of poker is the reality of life so the idea that I need to get so good at probabilistic thinking I am going to get that exactly right – that is a false goal. So don’t hold that as the standard.

Still, what you can do – and I would say this was my own path – is you can be curious and always recognisant that, no matter how good I am at this right now, or how much I understand of it right now, it’s probably in my best interest, and it would be a wise thing, to try to get better. And a couple of steps to get better is, one, I was visiting my alma mater of the University of Virginia last night, giving a talk there to the investment club I am involved with up there. And I remember my probability and statistics class and I remember taking my first whack at it and not doing well – at all. I mean, at all – it was a great struggle and I really had a hard time.

I ended up withdrawing from the class, which was a mandatory class – I had to take it – but I did take it with a different teacher. That made all the difference in the world – the light bulbs went on – and the guy who taught it, my second time around, was a sports fan. So he related the idea of probability and statistics to sports and gambling on sports, which made all the sense in the world to me – I connected with that very logical approach and embraced the topic. So, first off, there is a bit of an academic discipline and some math and rigour and process so, if you want to get good at it, take probability and statistics at the college level and make sure you pass and understand it well enough to at least get that checkmark. Then, after that, it is a matter of curiosity.

If you want to sound fancy about it, Bayesian probability is one of the terms – and what Bayesian probability really means in real life, in my opinion, is once you go through a day and you have learned something, when you are looking at the odds of this or that happening tomorrow, you have more data and you have more facts and understanding of what circumstances are. So if you are setting the initial line at the beginning of the day, it should be better than your ability to set it the day before because you learned something yesterday.

So lifelong learning, curiosity, understanding the basic, basic stuff and the disciplines of probability and statistics, reading things like Annie Duke’s books – there are so many books that are so helpful. I am sure it is a long list of things I have read on that topic over the years. And probably – whether it is once a year, once every six months or whatever – there is some new book that has a twist on that sort of thinking and I will probably read that too, just to try to pick up a little extra wisdom and knowledge along the way. I have been doing that my whole life so I think the cumulative effect of that starts to show up over time,

JTR: Annie Duke became very good at probabilistic thinking because she spent many years as a professional poker player so it was embedded in her and she was part of a community of people who were training her to think in terms of probabilities – in your case, do you think probabilistic thinking comes naturally to you because of your time in the insurance business?

TG: I think it is very, very helpful. It is a context and a setting so, for instance, one of the ways you see Annie Duke’s skills on display is at a poker table playing against a lot of other people who are very good at the same thing – and she differentiates herself with her performance. So you can draw a logical conclusion that this woman really knows what she is doing in this realm – and, if you transferred her to a completely foreign realm and made her an organ transplant surgeon or a botanist or something else, she would probably have quite a learning curve to be world-class at one of those disciplines.

But she probably has the horsepower to do it – with the appropriate time and course of study and training and so on. So it manifests itself at the poker table because you can visibly see it – and these things manifest themselves for me within the realm of insurance because I have spent my entire professional life thinking in those sorts of probabilistic terms. But there is a route, even below that, that comes from natural curiosity and a desire to know and understand how the machine works that helps you get better at that sort of thing.

AW: That is really interesting. I would like to dig into learning from outcomes. We have been talking about Annie Duke and poker and you can play a poker hand in a certain way where, 60 times out of 100, you win but there will be 40 times where you do not win – yet you should keep playing that same hand the same way. Now clearly in investment – and in other walks of life, like botany or organ transplants – the feedback loop is much slower than poker so it can be much harder to learn from your outcomes, even though we know, via probabilistic thinking, that outcomes can be a particularly lousy teacher. Over your career, then, how have you tried to ensure you learn the right lessons over time?

TG: Well, by doing it wrong – over and over again – and then the feedback loops do actually start to kick in! One of the terms in Annie Duke’s book is ‘resulting’, where she talks about looking at outcomes and drawing conclusions from them, which sometimes are wrong – and you can draw the wrong conclusions. So, frankly, it is a process that should be both bottom-up and top-down – and you should engage in both dimensions, just to make sure you are as robust as you can be. Recognise that, if you are exclusively bottom-up, you will get some things wrong; and, if you are exclusively top-down, you will get some things wrong. If you engage in a reasonable discipline, both bottom-up and top-down, you will still get some things wrong – but probably fewer than you would have had you only thought about things in one dimension or the other.

And there is just this tendency I observe all the time, where we all have our own biases and we all have our own ways we like to do things. We think it is going to work and we would rather it work that way – so it is confirmation bias. When that works, we talk about how great it is and, when the other method doesn’t work so well, we point out the flaws – See? I told you that doesn’t work. And when the other method does work, and yours doesn’t, you tend to have the same biases at play – like, Oh, they were just lucky that time and my process was good but my outcome was bad. Those may be true statements but that doesn’t excuse you from the responsibility of trying to be intellectually honest and trying to solve the problem both ways.

I think that is largely what Charlie Munger is getting at when he says: ‘Always invert.’ When your mind confronts a problem, you have a natural way in which you start to try to solve that problem – and I think Charlie Munger is saying, Stop. Pause. You can do that – but also do the problem the other way. Do it backwards – invert it from the way you would normally do it and reframe the question. His point is not that that is where the answer comes from and it is not that your original approach is going to be wrong – it is that you have doubled the model’s power by making it more robust and looking at things from both directions.

AW: We do something similar as a team, where we target the cheapest quintile of stocks but we always ask ourselves, Why might this screen be wrong? We call it ‘red-teaming’ – a term we stole from the US Army – where we come at it from completely the other side and try and see why might we be wrong.

TG: Perfect example – and by the way, here is one of the reasons why that is so hard. I have used that exact analogy and talked about the army’s ‘red team’ exercises – and sometimes my colleagues will say, Enough! Why don’t we call it what it really is: the ‘jerk team’! You’re always the guy arguing the devil’s advocate, or what about the other point of view, or what could be wrong about this – that has a human cost and a toll on relationships that is not zero. So you need to be, again, humanly skilful, and have a sense of humour and a sense of modesty and dole out the responsibility of being the captain of the red team from time to time – just so everybody can feel what it is like to be the person arguing the other side of the case. But it is hard to do.

Endowing the traits of endurance and longevity

AW: There do need to be a lot of behavioural safeguards within a team to be able to do that properly. Let’s switch gears and talk about something you have set up recently – an investment fund for students at a couple of colleges in the US that will remain invested in the same companies for a quarter of a century. What was the inspiration for such an incredible idea and, also, what do you feel the students will gain from it? What sort of problems currently faced by students do you think it can help solve?

TG: Well, thank you for asking about that. There was a very nice article in the Wall Street Journal recently where Jason Zweig wrote about my efforts. It is at two schools, the University of Virginia [UVA], which is where I went to school, and a school named Delaware State, which is an HBCU [Historic Black College or University]. I happen to serve on the board of Graham Holdings with the president of Delaware State so I got to know him and developed a personal relationship so I wanted to try to help out there as well.

As you suggest, the mechanics of the fund are fairly straightforward. You have to pay dues – you put $25 of your own money in the pot to be a full-fledged voting member of the fund. I have made a contribution, some of my friends and colleagues have made contributions and the Markel Group is making a contribution – so we are putting some money behind this. So each year, there is a ‘vintage’ year and the students will join the club and put some money in the pot – and, at the end of the year, after a series of speakers and a series of talks, each member of the club gets to pick one stock to buy and that stock will be held for 25 years.

That is the holding period and it cannot be traded – you are locking yourself in for a 25-year time horizon. And, at the end of year 25, the total portfolio created by all of these students making the one decision will be sold and half the money will be rolled into year 26 – to keep it going into perpetuity. And half of the money will be used to fund scholarships at that point in time. And the people who decide who gets the scholarship are the people who made the initial investment decisions – the students in year one. So, at their 25th reunion, they will be distributing those funds and, in a perfect world, if the value of those investments grow at a faster rate than tuition, in the fullness of time – the magic of compound interest being what it is – someday there will be enough money in that pot so that, if you go to UVA or Delaware State, your tuition will be covered.

That is the vision. As I mentioned earlier, we had a meeting at UVA last night where I was speaking and I said, You know, in your academic career here and if you are studying investments and you get a job in the investment industry, you are going to be doing a lot of work on trying to ascertain how good a company is. So you will be looking at return on equity and return on capital and capital allocation decisions and competitive forces and competitive advantages and all those sorts of things. And those are good and worthy and healthy disciplines. And there will be another line of reasoning that talks about the price of the security and doing an analysis of what this price should be – perhaps first what it is in the marketplace and is there a gap there that can be addressed? Either by going long and buying the stock or shorting the stock or taking the arbitrage opportunities – things of that nature – and that is a fine and worthy academic discipline as well. And many of those students will be using those first two lines of reasoning, all day, every day, in their professional career.

But I think what is missing from those two streams is the concept of endurance and longevity. So if this is a good business, why? How long is its competitive advantage likely to persist? Will technological changes sweep it away? Are they vulnerable to new ideas – somebody being creative or doing something different? Or are there things that really create some longevity to the competitive advantage that business would have? And I just think that idea of buying something where you are going to tie yourself to the mast and commit for 25 years will to rule out a lot of shorter-term things that may have a blisteringly hot moment in the sun, but may not endure or last in the fullness of time. And that is not talked about enough as part of the educational process. So this is just one small attempt to try to create another line of thinking – when you’re going through college and trying to determine models and things that might be useful to help you get through life.

JTR: Changing tack again – you have said in the past that dollar-cost averaging has been so important in your strategy. At the risk of being seen to play for the ‘jerk team’, would it not make more sense to just go passive and buy into the cheapest ETF on a monthly basis rather than just averaging down?

TG: It might do – a lot of people do that – and there are far worse ways to go, in many cases. That being said, we started this conversation with the idea of a sole CEO versus co-CEOs and co-managers and then we talked about some other things where binary conditions are established and thought to be true with no room in the middle. There is one fella who used to work around here, who had a great thing he used to say about ‘People who have forgotten there are numbers between zero and 100’ – and this would be one of those things. The idea is not that you have to be completely active – one end of the spectrum – or else go completely passive and just buy the lowest-cost ETF. There are roles for thought and imagination and creativity.

Now, you need to let things have enough time to work themselves out and, when people observe the pace at which I buy and sell, some might accuse me of being passive! Certainly, on any given day, it is pretty unlikely I am buying or selling something. But I do come to work every day and I do read and think and talk to people and try to observe and understand how the world adapts. And I like to give myself the leeway and flexibility and freedom to make changes when it is appropriate to do so. So it is a matter of judgement and balance. I am not a high-frequency trader or anything like that but I do make a few trades every once in a while!

Making a film recommendation you cannot refuse

AW: Tom, you have been a consummate guest but, before we let you go, we ask all our guests two ‘signature’ questions. The first is usually for a book recommendation. We understand you are a real film nut, however, so we will give you special dispensation that it can be a must-see film, if you prefer.

TG: Sure. Let’s go with the movie because, in fact, it is one I insisted my children sit on the couch and watch with me before they went off to college – and that is The Godfather. So if you want one of the best movies ever made, watch The Godfather and understand all of the relationships and power structures and humanity – and lack of humanity – involved in that movie. That is a good starting spot.

AW: I will do that – although my children are a bit too young for it at the moment. And I think yours are as well, Juan. Perhaps we will get the families together in 10 years and watch The Godfather!

JTR: I am going to make a public confession – I have never watched The Godfather.

TG: Well, here is the great news about it – I guess that movie probably came out when I was in my mid to late teens and I remember watching it at that point in time. Then my wife had never seen it and she was probably in her 40s before she made a statement like that to me – and that she had never seen it was unfathomable to me! So we watched it and I hadn’t seen it in 15 years and we both agreed the movie could have been made yesterday – it is timeless. So the fact it was made in the 1970s and you haven’t seen it yet – the good news is you are going to experience it in a fresh way when you watch it later today, which I’m sure you’ll do!

JTR: Absolutely! Tom, our last question, before we let you go, is we like to ask our guests for an example of a bad outcome that was identifiably due to poor process rather than bad luck?

TG: Well, that is an interesting question – and I think it is a good self-examination. Recently, I was doing some of that and I found out something embarrassing about myself – but I’ll share it. So mistakes of omission – things you should have done, but didn’t – tend to be way, way bigger than mistakes of commission. You make a mistake of commission and the feedback loop hits you pretty hard. When you make a mistake of omission, Shhh – you don’t really have to tell anybody about that – it can just be your own little secret. In point of fact, however, the biggest change or the biggest delta in your range of outcomes tends to come about because you didn’t do something you should have done. So I’ll tell my own story.

In 1984 – I was brand new in the investment business – I came across the Carol Loomis article about Warren Buffett in Fortune magazine and I thought it made all the sense in the world. The scales fell from my eyes, it was exactly on-track and seemed a wonderful way of thinking – but then I went and looked at the stock and it was several hundred dollars per share and I thought, Wow, no stock could be worth that. So I didn’t buy it, in 1984, when I first found out about it. It wasn’t until 1990 – six years later – that I bought my first shares in Berkshire Hathaway. And then, sadly enough, it wasn’t until 1998 that I started buying more. So that was an epic mistake of omission about how long it took, for me personally, to make Berkshire a very large position.

Now, fortunately, it is one of those situations where, as the old saying goes, ‘The best time to plant a tree was 20 years ago; the second best time is right now’. So I could make up for my mistake to some degree. But all of the steps I took – the reading, the curiosity – those were all reasonable processes but I just couldn’t bring myself to pull the trigger on something and that was an epic mistake of omission. And I think, if you really want to be your own personal ‘red team’, go back and examine actions you took and didn’t take over the last year, over the last five years, over the last 20 years – and write them down so you have it in front of you. And look at that list and use it as a teaching guide. Is there a pattern here? Is there something I can observe about these mistakes of omission I have made over the last one, five or 10 years that, gosh, in retrospect, I shouldn’t have made? Did I know enough to have acted differently? And with that, then, what are you going to do differently – starting right now?

JTR: That is a great way to close our session. Tom Gayner, thank you very much for coming onto The Value Perspective podcast.