I'm sorry to start here because I know everyone seems to start with a comparison to Berkshire Hathaway, which is quite a good thing. But my question is very specific around the differences in the evolution of Berkshire Hathaway and Markel as a company. Given that Berkshire Hathaway evolved from an investment holding company into insurance, and Markel seems to have done the reverse—insurance first and then an investment holding company—how does this evolution fundamentally change the essence of the company?

Thank you for the comparison which is certainly flattering. If we continue to persist and people continue to make that comparison, that means things are going well. I think there is some degree of form over substance embedded in that question that many people ask and explore. You could describe Berkshire Hathaway as a three-engine craft, much as has been the case for us, where they had an investment operation and an insurance operation, and they also grew an industrial, commercial, retail, and manufacturing set of businesses.

That came from the mindset of Warren Buffett, who with an investment background, investment gene, and investment training, had the notion of a 360-degree range of capital allocation choices. Markel is the same way in that even before I was here, Steve Markel, who was vice chairman—he was the financial guy at Markel and the investment guy—also knew of the Berkshire Hathaway model. From the mindset of being able to allocate capital to its highest and best use, and perhaps beyond the limits of what you were doing right now, that is the genesis of how Berkshire Hathaway and Markel became what they did. The finer points of whether it is an insurance company which became a holding company or a holding company which became an insurance company, I don't spend a lot of time on that. I prefer to spend my time on finding the next highest and best use for the next dollar of incremental capital that we have, and ensuring we invest it there.

My question was also leading towards how the balance sheets are different. Obviously, Berkshire Hathaway today is much larger. But even if you go back and look at a similar time in Berkshire's journey to where Markel is in scale and duration, the capitalization of Markel and Berkshire is quite different. Berkshire is one of one in that sense, where if you look at their gross written premiums and their equity, it is under 20% and has been—I was checking the balance sheet from the 1980s and they were still writing premiums and their surplus was huge—whereas most insurers are 60% plus in terms of that ratio. How does that difference in the capitalization change the ability to allocate capital?

There are some profound implications from that. First, let's start with the math, because at Berkshire and Markel, the compound annual growth rate in the equity capital of the organization has grown at a faster rate than insurance premiums have. If you take that reality and extend it for a long period of time, what happens is the growth rate in the premiums is this, the growth rate in the capital is this. By definition, that makes the ratio you speak of—the amount of premiums relative to the amount of capital—diminish over time. That process has been underway for a long time at Berkshire. It is not so much that was the design of what they were trying to reach; it is the outcome of the natural process. Markel is not as far along in that curve over time.

Because of our relatively small size when we started down that path—in 1986 when the company went public, we sold 500,000 shares at $8 a share, so $4 million was the size of the IPO and the total market cap of the company was $33 million, so it was a pretty small starting point. The size of the insurance business relative to the capital was more traditional or in line with what a lot of other insurers would have been, especially at the time when you could leverage your premiums more than would be the case in current financial markets. We have been on the same path, and the capital has compounded at a 15% rate since the mid-1980s in terms of the IPO. The premiums I think have grown at a slower rate than that. We are playing out that math as well.

From time to time we have done some major acquisitions relative to the size of the company, which in effect re-geared it and set that premium back lower than what it otherwise would have been. But in the fullness of time—10, 20, 30 years—I would expect something of a similar outcome because I would expect that our rate of capital growth, certainly on a per share basis, will be higher than what the rate of premium growth will be. Also, as was the case with Berkshire, that comparison becomes a bit of a hybrid because to what degree does your balance sheet influence and affect some of your non-insurance businesses, which are not regulated on a capital basis in the same way that an insurance business would be?

On that last point, how do you think about that?

I think about it in terms of economics. My training was as a CPA, so I understand the language of GAAP accounting. One of the reasons I got into the investment business is, as I joke, I was an accountant who was more interested in dollars than numbers. There is a big difference between the two.

What accounting should do, and the whole value of good accounting, is serve as a map. It describes the territory of what is really happening in an economic sense. Sometimes, in some matters, accounting is a better map than others. I think my job is to think through what the real economics are, what the real territory is, and not let accounting conventions distort or fail to paint the true picture of what is really happening economically.

There are a number of companies out there, and these would tend to be non-financial companies. I am thinking of something like Home Depot, Lowe's, McDonald's, or Marriott. Those businesses have been very successful for a long period of time, and they have done a lot of share repurchasing over a long period of time. If memory serves, I think the book value per share in those businesses, the equity capital numbers, are negative right now. Economically, that is nonsensical. The idea that if somebody was going to hand you McDonald's and hand you a check to make up for the deficit in the equity - if I were you, I would take that deal. But that doesn't describe reality well.

The point is, when you have a financial business where the balance sheet and the equity capital are important and are well described by GAAP accounting, and you have non-financial businesses where the cash flow statement and the income statement are more relevant and give you a better sense of things, put your focus where your focus should be on any given component or aspect of the business.

What we have done internally for a long time, and what we started to do last year when we recast the financial statements and tried to improve segmentation and clarity for people to understand what we were doing, is think about the intrinsic value per share. We talked about our methodology, where we discussed the balance sheet aspects of the business and how we would think about what that was worth, and we talked about the income statement-focused pieces of the business and how we would think about what they were worth.

We shared our methodology with the world so that people could follow along, and they could choose whether they wanted to use that methodology or something different. It doesn't matter. As long as you have something that is reasonable and you apply it consistently, you get directionally correct information about the direction of travel and the rate of increase. That is what we focus on.

How important is the equity surplus that you have on the balance sheet compared to the insurance premiums you are writing when you think about allocating capital?

It certainly exists and is the sort of thing which rating agencies and regulators look at, and we need to be in compliance with that. We need to be mindful of what sort of ratios they want to keep their ratings at that are appropriate for us to operate the business. But in terms of economically running the business, that is only the first step. We clear those hurdles with some margin of safety, and then we think about economics.

We are not trying to engineer the balance sheet or set much in the way of targets for those ratios. We are thinking about things in terms of economics. Once we clear the hurdles with margin of safety for rating agencies and regulators, I don't want to say we are unfettered, but the way in which we look at things at that point begins to pivot to the pure economics of balancing off what the opportunities are, where you can deploy capital, where it earns a good rate of return.

How would that change the way you look at deploying capital between your rates of return and the risk you can take with that capital as the balance sheet evolves?

One of the things we have stated consistently from day one is that when we write an insurance policy and put up a reserve for future losses, that reserve appears on the balance sheet as a liability - money we expect to pay out in the future. We have consistently always put enough capital and cash investment into very high quality fixed income securities to provide for that liability. If there is a dollar of liability on our balance sheet for a loss reserve, we want a dollar of high quality fixed income to stand behind that particular reserve. That is what really matters.

The biggest number on the right-hand side of the balance sheet is that liability for future reserves. The second biggest number on the right-hand side would be shareholders' equity. It appears on the liability side of the balance sheet due to accounting conventions, but it makes sense. That is an infinite maturity liability. It keeps the balance sheet balanced, and it is the money that belongs to the shareholders. However, there is not a specific amount and a specific time attached to it in the way there would be for long-term debt that might have a fixed coupon and a fixed maturity, or reserves which do not have fixed maturities and fixed coupons but represent money that will be paid out.

If you have an infinite maturity liability on the right-hand side of the balance sheet, then you can have an infinite maturity asset on the left-hand side of the balance sheet. Infinite maturity assets can be publicly traded stocks or wholly owned non-financial businesses - they are the same thing. It is an equity interest in either something you own a partial or controlling interest in. What you own is an interest in a business itself.

If that ratio goes down, where the amount of capital goes up and the amount of reserves and liabilities goes down relative to that ratio, it opens up degrees of freedom to invest more of the assets on the left-hand side of the balance sheet in equity-type securities rather than fixed income securities. There is an interplay. If your ratio starts to suggest that you are overcapitalized - meaning you are not levered as much as one would argue is the efficient frontier of what the leverage should be - we can offset that by increasing the amount of equity ownership interest we have, which would have a higher expected rate of return than what we would expect to get on the fixed income.

Which is what Berkshire had effectively from day 1. In 1988 they had $3.4 billion in book value and $600 million in premiums.

What that meant is that the liabilities - the future policyholder payments they were going to make - were so well capitalized that they did not need to own fixed income securities in the way that, at this point in our development, we still need to own fixed income securities. Berkshire managed to earn excellent returns over decades by being overcapitalized. May the Lord smite us with such a curse of doing that as well.

How would that change the way you invest in public equities in terms of, for example, the level of concentration you can take? If you compare Buffett's portfolio and his balance sheet to your balance sheet and portfolio, how do you think that would change as that ratio shifts?

The way it will change the method by which we invest, the thought process, the discipline - not at all. We will use the same discipline of trying to find good businesses that earn good returns on capital without too much debt, run by management teams with equal measures of talent and integrity, that have capital discipline, good reinvestment opportunities, or are good at acquisitions or capital management, for fair prices. That is the catechism for how we invest.

No VC moonshots, then?

That is not our skill. We don't know how to do that. Others know how to do that, and that is their game, their business. We would not do well if we were trying to do that. The degree of concentration has some cyclicality to it. Buffett is smarter than I am, and certainly Buffett and Munger together for so many years. It would be a mistake, just like it would be a mistake for me to try to do VC moonshot kinds of things. I have to know my limits. As the Clint Eastwood movie line goes, a man has to know his limitations. I am aware of my limitations, and I am not as smart as those two guys were and are. As a consequence of their intelligence, they were able to have a concentrated portfolio by which they earned very good returns, better than the market for a long period of time. If you can do that, that is great. But second best and still pretty good is to have a bit more diversification.

By the way, our diversification—I think people make a bit of an error in that they see 140 or 150 names. The fact of the matter is, in today's portfolio, there are a couple of buckets that represent two-thirds of the portfolio. One of those buckets is Berkshire in and of itself. In fact, I would say Berkshire is the new S&P 500. The S&P 500, when it was first created as an index, was meant to be a broad representation of investment choices and low cost. Today, if you buy an index fund or ETF, do you get low cost? Yes, you do. Do you get a broad representation? I would argue not so much. It has become pretty concentrated. I don't think it is as diversified as people think it is.

With Berkshire, by contrast, is it low cost? Yes, it is. Buffett doesn't pay very much in management fees. Secondly, is it diversified? Yes, it is diversified. You get the insurance business, you get the utility business, you get the energy business, you get the railroad business, you get all these manufacturing businesses. It is the new S&P, so I would say that is a bit of a bucket.

Secondly, we own five of the Magnificent Seven, and we have for a long time because they are spectacular businesses, and that is a meaningful chunk of what we own. Third, we own a lot of the asset managers, whether that is JPMorgan, KKR, Apollo, Brookfield. Those buckets of companies have done a very good job of positioning themselves as both royalty override businesses in the arena that they operate in and building some balance sheet returns as well. Then we own a series of companies that I would call market leaders, whether that is Home Depot and Lowe's or MasterCard and Visa. They are market leaders in those businesses and have very good competitive positions. Those four buckets would account for two-thirds of the portfolio. Economically, I would say it is a little more concentrated than people think it is.

In the world that we are in right now, I think it is fair to say that nobody knows. Sam Goldwyn used to say, nobody knows nothing. In the topsy-turvy world in which we live and the pace of technological and geopolitical changes that are going on out there, this seems to me to be a time when things are changing in much more sudden and substantive ways than what I have experienced in the 40 years I have been in the investment business. As a consequence, I do want to cast the net a bit wider than would otherwise be the case and let the market tell me where I am right and wrong. The signaling of owning a long list of stocks and the signaling that comes from being aware of and paying attention to them and seeing which ones are making new highs—that is the marketplace telling me which things are working.

As Charlie Munger says, the universal algorithm is do more of what works. Leaving the portfolio alone, there is an emergent quality that your winners will become greater percentages of the portfolio and your losers will become smaller percentages if you leave them alone and let the garden grow. If you use the information and the signal that you are getting from observing what things are working and what things aren't, that helps with that process of allocating the next dollar of incremental capital.

How would you reflect on how the way that you have invested in public equities has evolved over 20, 30 years? I was digging up an old 2006 Value Investor Insight report that you may remember. In there it said you had 80 to 90 stocks, 70% were in your top 20, which seems a bit more concentrated. But given how your role has evolved—obviously you are CEO today, you started in a very different position—how would you reflect on how the way you approach public equity investing has changed?

In terms of the way I approach them, the way has not changed. Again, back to the four-part discipline—and I will save you editing time by not repeating it again—but let's stipulate that that approach really hasn't changed. That has been in place for decades now. In that particular 2006 marketplace and the particular size that we were, and the particular opportunity sets that I was observing, the outcome was that the top 20 was 70%. Now the top 20 again—for instance, getting back to that example of Home Depot and Lowe's, would you count that as one position, two positions, 1.3 positions, 1.7 positions? I don't think it is one, and I don't think it is two. I think it is somewhere in between those two numbers.

Speaking of unpopular positions right now, Brown-Forman and Diageo—again, is that one position or two? It is somewhere in between two. Depending on how economically thoughtful you are in counting the number of positions and the inevitable irreducible amount of judgment that you would make by assigning 1.2 or 1.8 to that, I think you would find that economically the portfolio is still relatively concentrated compared to the way most folks do it. But I don't put too fine a pencil on how I would want to quantify that.

So you would look at Visa and MasterCard or Home Depot and Lowe's as effectively one bet. Was that a similar approach you had back in 2006 or before that, where you would have these buckets?

One of the ways I have evolved as an investor is that I have gotten more focused on quality and understanding. Buffett said that being an investor made him a better businessman, and being a businessman made him a better investor. You can read that, but until you actually become one or the other and start doing both, you do not really feel it viscerally.

In 2005, we bought AMF Bakery Systems, which was the first start of Markel Ventures. That was when I became responsible for running an actual business as opposed to just investing in a business. One of the real learnings and developments that have come through 20 years of being in that role as part of an operating team running businesses, is that running a business is hard. The people who do that are dedicated. They work 25 hours a day, they travel all the time. It is immense work.

When you find people who are really good at it and produce good results and do it in the right way, and they do it with that ethos of serving others and selflessness involved, do not lightly discard those relationships. Do not chuck them over the side of the balloon because you think you might have found an incrementally better one. There is immense value in loyalty, in developing and deepening relationships over time. That is true for the businesses we own when they are publicly traded as well.

Home Depot and Lowe's are great examples. The bulk of those positions were bought during the Great Financial Crisis in 2008-2009. Here is a little bit of how my research process works. I had a home with three kids. It was a standard suburban home with a driveway and a basketball net and a pleasant place to live. But a big chunk of the siding went bad, and it was 2008-2009. I did not want to spend any money, but I had to if I wanted the house not to rot. I needed to spend a chunk of money, which was extraordinarily painful, to make my house look like what it looked like before I spent that money.

I was thinking to myself, there was the notion that mortgages were trading for 50 or 60 cents on the dollar. I did not have the mechanism by which I could buy my own mortgage at 50 cents on the dollar. The bank was not willing to take 50 cents on the dollar for my particular mortgage, and I was not skilled at figuring that out. But what I did say is that the housing stock that exists is going to continue to exist. I think it is going to grow over time. It is my lived experience as a homeowner that it requires maintenance and ongoing expenditures all the time, whether you want to do it or not. The two leading companies that were astride and atop that flow were Home Depot and Lowe's.

I started buying them, and they have been immensely successful investments for a long time. That was the starting point, the fundamental insight as to why they would be good. I am not a handyman. I have been inside a Home Depot twice in my whole life to get something to fix. Once when I replaced the shower head, and I was so proud of myself. The other time, we have these cathedral ceilings where the light bulbs are out of reach, so there is this suction cup on the end of a very long stick that you use to change the light bulb. Those were the two times I have been in a Home Depot in my life that I recall.

How has your investment team changed?

Not dramatically. For instance, I worked side by side with Steve Markel, whose office was next to mine. He had operating responsibilities for Markel, but he was the person who was keenly interested in the investment side of the business. As partners, we would talk all day, every day and talk about ideas. I was responsible for managing the portfolio, but certainly kept Steve in the loop. He was and still is a great questioner, always trying to get to the underlying thesis as to why you are doing something or why you are not doing something.

I worked side by side with Steve, and I think I was here 17 years before I hired the first associate to help me. That was Dan Gertner, who had been at Grant's Interest Rate Observer and was one of the folks who early on spotted the troubles brewing in the mortgage market. He was referred to in the Michael Lewis book, The Big Short. I got to know him because I had spoken at Grant's a couple of times, and we connected. Dan joined me.

It was a decade later before Saurabh Madaan joined me. He was here for several years. Saurabh decided he wanted to paint on his own canvas and go out and do his own things. I was sad to see him go, but we are still friendly. He moved to Richmond as part of the process and has remained in Richmond, so I see him with some regularity.

After Saurabh left, a gentleman named Tyler Brown joined. He had been part of the firm here in Richmond in the investment world and had many years of experience. He was known to many members of the Markel team around here for his character, integrity, and skills. He took Saurabh's spot and continues to be with us to this day as a sidekick on both the equity investment portfolio and our investor relations efforts.

How many people are on the public equity side?

On the equity side, it would be myself, Tyler, and Dan. We have three folks who manage the fixed income portfolio, which, as we spoke earlier, is a hybrid. There is part of that which is a treasury function that is matching up against the insurance reserves, and there is part of it which is an investment function and a pool holding tank of liquidity with which we can make active capital allocation decisions. We have some money to do it with because we maintain a bit of a fixed income portfolio from which to either put money in or take some money out as we see opportunities.

How do you think about the insurance book, the structure of the insurance book between long and short tail, and how that relates to the reinvestment rate of the capital into equities and ventures?

If you are talking about insurance float itself, I would limit that to the reserves for future claims. As we talked about earlier, that money we are going to invest in fixed income securities. That money really isn't going to go into equity securities. Now someday, 10, 20, 30 years from now, if we keep going down the path of compounding our equity capital, we will get into a more Berkshire-like position where you have so much equity capital relative to the liabilities that you don't need any fixed income securities. In Berkshire's case, we are on that path. That is a long time before that would probably be the case. I wanted to make sure we were defining the terms of float accurately. The equity securities are funded by the capital of the company, the retained earnings and equity capital on the investment decisions related to whether it is a short tail risk or a long tail risk on a short tail risk.

If you have property coverages and you are exposed to natural catastrophes and a natural catastrophe happens and the next day you need to be writing a check, you need to keep that money pretty short term. That is in firm money market funds, short term fixed income securities, and treasury bills. If you are dealing with liability type claims where the tail might be five, six, seven, eight years, the actuaries of Markel provide us with their estimate of what the duration of the liabilities are. So we do our best to match that with the duration of what the fixed income portfolio is, so they keep in balance between the two.

You always have to have enough of the short term to deal with your short term liabilities. You have a natural ladder process that your longer term securities are becoming shorter term with the passage of every passing day. And of course the liabilities are growing closer every passing day to having to write a check to cover the claim. So we just try to keep it matched.

How do you think about short versus long tail and that mix in the insurance portfolio opportunistically?

We are not a pure property writer. As a percentage of the total amount of insurance liabilities we have, property is a smaller percentage of our book than would be the case for a lot of insurance companies. That has historically been the case. We think of ourselves as a specialist insurance writer, and that is separate and distinct from whether you call it property or casualty.

We are going to do things that are special, that are unique, that other people perhaps are not able or willing to do. Sometimes that is in the property markets, sometimes it is in various things which would fall under non-property as a distinction. We don't have a specific target plan for what we would have in one versus the other. What we look for is the opportunity to write business that makes sense. If you are going to write property business, that by definition has to be at a lower combined ratio target because you are not going to get as much investment income off the float, because you need to keep it shorter and you are not going to earn as much.

If you are writing business that has a longer tail to it, you can put that on the books at a higher expected combined ratio because you are going to benefit from investment income for a longer period of time with business like that.

With the equity capital you have that goes into ventures or public equities, how does that feed into the risk-based models that regulators look at? Because the unrealized gains, I imagine, are not treated the same as ventures cash flow.

When I joined Markel in 1990, the regulators didn't spend a whole lot of time or assign much credit or mind space to what the equity investment portfolio was. They wanted to make sure we had enough high quality fixed income securities to cover those liabilities and do that with a margin of safety that is appropriate for the rating that they would assign us.

Different states and different regulatory bodies would have some different thresholds, but we would always be mindful of those and always operate in such a way that we cleared all those hurdles with a margin of safety. The regulators didn't pay too much attention to the extra capital that was in the equity portfolio. Similarly, when we started down the path of building out the Markel Ventures companies, that was over and above the capital that rating agencies and regulators would be primarily focused on.

It has grown to a point and it has existed for a long enough period of time that it begins to have credibility. We have pointed that out to the regulators and the rating agencies that here is another source of financial strength for Markel. Through time, as you have been doing it longer and longer, it becomes more normal and steady and expected. You can look at long trails of history that would suggest that this is a positive feature of Markel. Different regulators and rating agencies treat that differently, but they are becoming more aware and comfortable that this is a net positive for Markel and giving us some kudos for that.

How does the regulator's view of the capital, the unrealized gains you might get from the public equity portfolio versus the cash flow from ventures, change the way you deploy the capital? All else equal, if you have the same return, how would that impact?

I don't think the answer to that question is mathematical. I think it is a sense of growing trust and confidence that the people at Markel are thoughtful in the way they allocate capital. As a consequence, when you are viewing something, your mindset is important in terms of what you see.

There are four bottles of wine in my life that I have had where I thought, "Wow, that was a good bottle of wine." One of them was on a trip to Paris that I took my wife to as a surprise for her 40th birthday. On the first day we were adjusting to the time difference, and we were in a modest hotel. We went to a grocery store to get some cheese and meats, and we bought a bottle of wine. It was a nice bottle of wine, but not a Grand Cru. We drank that bottle of wine and we both thought, "Wow, that was a fantastic bottle of wine." When we got back home, I looked for that label and tried to recreate that experience, and the wine didn't taste as good, because in that moment of the aura of a surprise trip and an adventure, the wine was a component of the experience, so it tasted better than objectively you would think.

In a similar way, as Markel has built financial credibility over a long period of time, I think the rating agencies and regulators look at what we have done and they look at the track record at the time when they have done it. Whereas in the past, going back 10 or 20 years, they would not have spent a lot of time on it and not thought too much in the way that they would allocate or ascribe credit or kudos to us for having done those sorts of things, now they are further along the path of thinking those kudos are good, not bad.

You recently sold the reinsurance book. Why was that?

We tried but couldn't make it earn the returns that it needed to continue to remain part of the group.

Why was it difficult?

I think there are a variety of reasons for that. Reinsurance, by definition, is one degree of separation removed from making the decision at the primary line. Your ability to control your destiny is at least diminished in some degree by the fact that you are one step removed from the action. That is not a fatal flaw. There are a lot of people who have run good reinsurance businesses for a long period of time, but it takes a certain size and scale and skills to do that specifically. We tried and we continued for several years. Basically, we kept raising the price per unit of risk, which is the directionally correct thing to do. But the way the ball bounced, whether it is US court systems or social inflation or loss cost trends, the risk kept getting more expensive through the last four or five years in such a way that we were never able to catch up and get to the point where we were charging enough per unit of risk.

We thought that it would probably be better to have that book join forces with a larger entity that was building it and doing it at a different scale. When we were looking at the array of where should we allocate our limited capital, should we allocate it in reinsurance where we haven't done well, or should we use those limited resources of capital and time and energy in other parts of our business where we had earned higher rates of return? We did what I think was not fun, but it was the rational thing to do. Getting back to Charlie Munger's universal algorithm, do more of what works and less of what doesn't. That was the rationale behind their choice.

How much of the reinsurance book was the old Alterra acquisition?

The vast majority of it. Where we got into the reinsurance business, by and large, was through the Alterra acquisition.

I went back and looked at Berkshire's reinsurance performance in the early days, and it was terrible. I was looking here from 1975 to 1988, they lost money every year, and a pretty decent amount. In his 1986 letter, he mentioned the word "social inflation." I didn't know social inflation was a permanent fixture of the industry, but it seems like it has been going for years. What is it? I guess this links back to the balance sheet question on how a business like Berkshire, even back in the 80s, can do reinsurance. Maybe a different style to Markel. How do you, in hindsight, review how Markel was approaching reinsurance versus how Berkshire built their reinsurance business?

As Buffett and Munger wrote many times, and it continues to be true, insurance is a tough business. You are putting out a price to cover an expense that is going to happen in the future. You don't know your cost of goods sold with precision at the time. That is by definition the nature of insurance. If insurance is a tough business, reinsurance is a doubly tough business, as witnessed by the results you speak of from Berkshire in that time frame. Now, one of the key things about reinsurance, and here's a particular good bit about it, is the float tends to be higher in reinsurance because you hold onto the money longer. Because of that fact, you are one step removed in the chain.

If you go back to the 80s, interest rates across the board were massively higher than what they have been in recent years. The first mortgage on the first house I bought was 15%. I have scars and memories of that era. But if you get 12%, 14%, 15% on a government bond with no credit risk, the combined ratio at which you can accept a business is higher. In fact, if you look at the insurance industry writ large and talk about the late 80s, when I was becoming familiar with Markel, I was an analyst who covered it from the outside and then joined Markel in 1990, the industry average combined ratio was probably north of 100, because you could write at 102, 103, 104, and invest in the government bonds, and you didn't have great economics, but you were profitable. That was the gig. You produced positive returns on capital. The losses that Berkshire would have experienced in reinsurance back in that era, once you added back the investment income, they might have been profitable. So it is a different era.

He said that most people were writing unprofitable reinsurance because they wanted the 15% yield.

In today's environment, we have bounced up from 0% interest rates, but they are still historically on the low side. The amount of underwriting profitability or losses that you can accept and make economically viable in the reinsurance business is different today than it was 30 or 40 years ago. Reinsurance, just like insurance itself, has to categorically operate at a lower target combined ratio than was previously the case.

To make this specific to Markel, last year we finished at around 94 for the combined ratio. If you look at the history of Markel from the IPO until last year, the average combined ratio over those 39 years is approximately 95. We have been able to compound the capital of the Markel Group for 39 years in the mid-teens rate with an average combined ratio of 95.

However, 95 is not a spectacular combined ratio in today's world because interest rates are lower now than they were through that 39-year period. The targets are different. For the industry and many companies, you see numbers in the very low 90s or in the 80s last year, and those are spectacular results. Part of that is because those companies have done an excellent job and deserve congratulations and kudos for their results. It is also partially because the overall standard—the hurdle rate for what it takes to operate a reasonably successful insurance business when looking at the combined ratio—is lower than it would have been 10, 20, or 30 years ago.

What is a great combined ratio in your mind for Markel?

Low 90s is pretty good. That is an oversimplified answer because you need to look at the complexity of our business. For instance, in a reductio ad absurdum fashion, if you made the statement that property prices are ridiculous and that market is too competitive and everybody is offering property coverages at low rates, and as a consequence we should shrink our property exposures, then the mix of business skews more towards non-property and longer tail. A combined ratio of maybe low to mid-90s could end up producing spectacular economic results in terms of the return on capital for Markel.

On the other side of that coin, if property rates were extremely high and the expected value when you looked at property rates was spectacular, such that we should skew our book more towards property coverages and away from non-property coverages, maybe our target should be 85 or 88. I hesitate to give you a single point estimate because the single point estimate without the context of what kind of business you are writing is misleading, if not wrong.

But you are moving out of reinsurance. That is going to give you a bit of a tailwind.

That is not necessarily by definition of reinsurance versus insurance. It is that we did not cover ourselves with glory in reinsurance. We did not meet our underwriting targets. By virtue of the fact that we are dialing that back, all things being equal, the combined ratio will improve because we have more of what works and less of what did not.

As you move out of reinsurance, how do you expect the restricted assets to change? I think you are at $5 billion now, roughly.

I do not want to quote a number about that, but it would be economically accurate to say that the capital in the reinsurance book right now is associated with the runoff of the reinsurance liabilities that will be around for a number of years. That capital and those reserves are matched up against fixed income securities in the same way that every other insurance liability we have. As those reserves run off, some of that fixed income will be liquidated to pay the claims.

We would, as always is the case, hope to have our reserves stated at a level that is more likely to be redundant than deficient. That is the way we set reserves around here. That money should be able to be invested with high degrees of latitude and flexibility as that portfolio runs off. The layer of capital that you have to have for rating agency and regulatory purposes diminishes as well. It should directionally free up more money.

We use the word "money" instead of "capital" because sometimes the word "capital" means more than one thing. Sometimes it means the regulatory definition of how much you have to have posted to write the business you have—that is an accounting entry. Sometimes people use that for money in the investment. To clear up the muddiness between what that word can mean in different contexts, I am saying we will have money that should be able to be invested more freely as that book runs off.

How long do you think that will take to run out?

It will be a couple of years. The concept of how you should think about that is half-life. Whatever was on the books a year ago, as you get out a year or two, cut that in half, and then another year or two, cut that in half. It will diminish at an increasing rate in terms of absolute dollars as time goes by, but they will be with us for a couple of years.

Switching to look more at opportunity costs, you have recently said clearly you have four ways to reinvest capital, so obviously back into insurance or public equities, into ventures or just repurchase Markel shares, how do you think about that every year?

Opportunity cost is exactly the right phrase to use, and I have two stories about that. First, Charlie Munger had a joke, when somebody asked him, how are you? He would respond, compared to what? So oftentimes there isn't an absolute answer to a question like that.

Second, the phrase "opportunity cost" makes my children shudder because they were raised around a dinner table where that concept was discussed from a very early age. Growing up with three kids in a busy household, whenever somebody wanted to do something, my typical response would be, "If we choose to do that, what are we going to choose to stop doing so we can do that?" They shudder because they know that a choice needs to be made. You cannot just do the next thing without taking some things off the plate.

Here is where I think some of the real magic of Markel exists. We have an insurance business. We have a collection of wonderful non-insurance businesses. We have decades of experience investing in public equities and fixed income securities. We have a more recent history of repurchasing our own shares. When any opportunity comes up in any one of those dimensions, it gets the Charlie Munger question. Everybody is a Type A successful person who runs these businesses. They all have ideas for how to deploy capital. And I say, "Great, but compared to what?" People who have a demonstrated track record of producing good returns with the capital they have are first in line for where the capital goes.

How does that work? For example, hypothetically, let's say the company earns $100 in free cash flow to deploy at the end of the year. Simon might ask for funding to open two more lines of insurance. A ventures business might want more capex to build something.

There is a process, and it is not annual. It would be an exaggeration to say it is every day, but there is a quarterly board meeting process for Markel Group as a whole. There is also a quarterly board meeting process for each and every individual business unit within Markel. Some of those quarters, the reports focus more on the current quarter's business. Some quarters include annual budgets, which cover capital expenditures and acquisition thoughts. Some involve monthly financial reporting. Some involve day-to-day conversation and the normal flow of running the business.

It is not so rigid that we set an annual budget and an annual target for how much we want to fund in acquisitions. The financial resources of the company are available, and if an acquisition looks attractive, you may become aware of it on April 4th or August 7th—something completely independent of December 31st year-ends. It is incumbent on us to talk to one another.

Simon sees an opportunity or hears about something—and this is real, this has happened within recent months—he will pick up the phone and say, "Hey Tom, I saw this" or "We heard about that, and I think it makes sense. We would like to explore it." I might ask a few questions and say, "That sounds great. Learn more, let me know what you find out." That is a normal recurring weekly or monthly process. Maybe not every day, but weekly or monthly, people hear things, see things, get exposed to ideas, and that is what we do. The nature of the business is talking about that.

Let's say Costa Farms or one of the ventures companies wants to build a big capex plan which will cost several hundred million dollars. At some point that comes to you and Andrew. How does that process work?

An example like that would flow through the normal annual budgeting process, where the plans for next year are being set. For a long-term plan such as that, it is not a snap decision. Management does not wake up in the morning, come into the boardroom, and say, "We want to do that." In the normal course of running their business, they are looking at opportunities and places where they should deploy capital. In the normal conversation, the case for that gets built.

There is a formal process where the board of those companies—in the case of Costa Farms, the Markel representatives on that board, which would be Andrew and our presence there—would be the deciding vote for things like that. We will make that decision in the context of the annual budget. Hundreds of millions of dollars, to use your number, would flow through a normal board process.

Would you have a certain requirement for the cost of capital from Markel Group in terms of funding that outside of their cash flow? How do you think about that?

I will tell you a story to illustrate the point. When my middle daughter was in college, she had her first finance class where one of the big projects was to calculate the weighted average cost of capital, the WACC. This was a group project, and they worked hours and hours on it. They came to a decimal point number, turned it in, and dealt with all the frictions of a group project you might imagine with college kids. She got a good grade on it.

After that, around the dinner table, she said to me, "Hey Dad, what do you use for cost of capital when you are looking at things?" I said, "In today's environment, call it about 10%." She almost cried because obviously I gave that as a snap answer without endless amounts of spreadsheets or calculations. But roughly speaking, it is a directionally correct answer, and it salutes the imprecision of the reality of life. It also salutes the precision of the fact that you should not do things unless they meet a certain basic hurdle.

To her credit, she went to her professor and said, "We had the cost of capital project, we did this, and I came up with that number. I asked my dad about it—he is in that world and has done okay over time—and when I asked him, he said 10%." The professor, to his credit, said, "Your dad is right. We just don't know how to teach that."

That was a moment. To your point, we would expect to earn double-digit returns on our capital when we are laying out capital. For 39 years as a public company, we have. In today's interest rate environment, where the 10-year is 4%, if we earn double-digit returns on our capital, I think those are good returns, and those are projects that we will thoughtfully consider. If we do not think we can make 10%, we should put the pencils down and keep looking elsewhere.

I think that lines up with the proxy, if I am not mistaken, in terms of that 10% hurdle rate through the company where that is the hurdle rate in terms of how you look at deploying capital into insurance, ventures, or public equity.

That is correct. To put a fine point on that, my incentive compensation and that of the senior cadre is based on us getting across that 10% hurdle rate. An important distinction from Markel, which I think is a huge component of our competitive advantage, is that this is not a one-year measurement. It is a five-year rolling average measurement. We are long-term people here, and I never want to have the behavior of some short-term ringing of the bell that might come at the expense of longer-term returns.

In a perfect world, we have the founder's mentality, as Simon would call it, where you are thinking about the long-term nature of this business and thinking about things in generational terms, not trying to ring a bell for any one year's results that can often come at the cost of longer-term returns.

Is the average operating income fixed at a point in time or does that change every year?

The target is recast every year, and it is basically the same sort of math. If you take roughly the capital that is being deployed and you think about a 10% hurdle rate for that, where does that number land? It is meant to create a mechanism that, measured another way, would still get you to that hurdle rate concept before the management earns incentives.

Last year it was $1.7 billion operating income. How would that change for this year?

The number that will likely go into this year's proxy would be higher because we made money this past year. We did return some capital, both in the form of redeeming the $600 million of the preferred stock issue—we paid that back—and we bought $400-million worth of our own common shares. We made money, which adds to the capital base on which we need to earn a return. But we also distributed a chunk of that, which reduces the capital base that is out there.

It is meant for the owners of Markel. If you are a shareholder of Markel, you have provided us with capital to operate this business. We need to earn the right to earn your trust to have that capital. What we have said is that unless we earn a double-digit return on that capital, we won't pay ourselves full incentive bonuses against the target. We have to get to double-digit returns before we meet the target bonus targets, and we measure those over five-year rolling averages.

I want to discuss Ventures, which many believe is the most exciting part of Markel. Can you provide a brief bit of context on when and why you moved into controlling interests in businesses? The bakery business was the first one.

Being a student of Buffett and Berkshire, he made the same transition over time where instead of just owning shares in publicly traded companies, at some point it morphed into actually owning controlling interest in companies. It was the same thought process and the same discipline by which he was deciding is this a good business or not? Is this a fair price or not to buy 100 shares of something versus 100% of things. The intellectual discipline is the same. So that DNA was always here at Markel from the very moment that we started and I arrived in 1990. It was 15 years, not until 2005, before a specific circumstance came up where there was the opportunity to buy controlling interest in a business. It happened to be a local Richmond business, run by a CEO who I knew, and it was also a fairly straightforward business where I thought it was within my circle of competence to analyze that business and make judgments about its valuation. The stars aligned in order for us to make that step.

That step was both a financial decision and a learning decision of doing something in the size and scale that even if we were completely foolish and wrong, we could learn something that would not be a fatal error. Fortunately, it has worked out very well. That was the step that began the process of building out the Markel Ventures set of businesses. That first business had a top line of $50 million and EBITDA of $5 million. The Markel Ventures set of businesses last year was more than $5 billion of top line revenues and $600 million worth of EBITDA. It has worked very well and it is the sort of thing that we look to continue to do as time goes by.

How is the criteria different if you are acquiring a business in Ventures versus the minority interest in the public equity?

It is the same criteria. Is this a good business that earns good returns on capital without using too much debt to do it? Is the management team equally endowed with talent and integrity? Because one without the other is worthless. Do they have reinvestment opportunities? Can they grow? And if they can't, we are in control of the capital allocation decisions at Markel rather than at the Ventures company level. Many of these businesses are wonderful businesses, but they have very limited reinvestment opportunity because they are already big in the industry that they are in.

That is fine because what they do with their cash is they redistribute it back to Richmond, and then it is our job to redistribute it. We think about the cash flow characteristics of the business and whether it can be applied within that business unit or whether it should be dividended up to the holding company and then allocated elsewhere through the portfolio. Fourth, and finally, can we acquire it at a reasonable price that provides good returns on the capital we are laying out to buy it? Those four criteria are exactly the same as what we would use to select public equity.

When you say profitable, how do you measure that?

That double-digit return on the capital. If we lay out $100, we want to see that they are making at least $10.

How do you think about organic growth? You mentioned that they don't need to have a huge runway to redeploy capital because you can dividend up to Markel and you can allocate that as you wish.

We are in favor of organic growth. It tends to be very capital efficient because you already have the business and the customer list. Depending on how tangible the product is, you may or may not have capital expenditures to support organic growth. The definition of organic growth can vary. If you talk about Costa, all of their growth is organic—they are growing plants, all carbon-based products, so it is 100% organic. That word, like "capital," can mean different things in different contexts and settings.

The purest sense of organic growth to me would be the ability to do incremental business and earn incremental profits without having to deploy much in the way of incremental capital. It is very capital efficient, so we are in favor of it. Now, the fact of the matter is there are some industries where, if you already have a very large market share, unless that industry grows, it is not feasible to think your market share is going to grow in such a way that it would produce organic growth. But that is acceptable because as long as the businesses are well run and produce good returns, we can take that money back at the holding company and then look around the 360-degree range of opportunities and see where we should deploy that capital productively.

Let's say a business was declining slightly organically, the top line had decent plus-10% EBIT, and you could buy it at a good price, and it ticks all the other boxes—would that fit the criteria for Ventures?

I will add a nuance to that. If you are looking at a business that is in decline to some degree, you are dangerously close to looking at a melting ice cube. In the tough competitive world that I observe out there, that is one of those lessons I have learned over time. I would prefer not to enter into a position where I thought that would be the case. I do not think I am smart enough, swift enough, or skilled enough to manage melting ice cubes, even if the math of how much cash you could recover and how quickly would suggest that it might be an attractive financial return.

This is not our cup of tea. In fact, it is a very important criterion that the business—and again, go back to the original transaction of AMF Bakery Systems—they make ovens and equipment that bake bread. We as human beings have been eating bread for thousands of years. We even eat bread when we tell people we are not eating bread. In the era of the Atkins diet or low carbs, we still eat a lot of bread. That is part of the underwriting criteria. I want to invest in businesses that I think are durable and likely to continue to be needed, doing things that people will want and need in the indefinite future. We do not step into things where we say, "This might have five good years or ten good years, but then it is going to go away." That might happen, but it will not happen on purpose around here.

When I was eyeballing the organic growth of the consumer and industrial segment—I did not look at the financial segment because you have not allocated much capital to that new division—how do you think about the organic growth performance of industrial and consumer Ventures over the last five or six years?

I am very pleased with it. Those results across the board have been very good in aggregate. Obviously, there are some businesses at any given point in a given year, any given moment in a cycle, that are way up or way down. But in aggregate, the portfolio has done very well, and it has done very well from a return on capital point of view. Those businesses have funded all of their appropriate capital expenditures and still dividend money up to the sign-off one.

One of these big $200 million deals that we hypothetically discussed?

I do not remember the exact number, so I cannot say that. But we have had some companies that have done some pretty major work within their footprint, though not a whole lot of them. Any of them that have been done have been funded entirely from within the resources of those existing businesses. They did not even ask any of their other cousins to come up with the money for what they wanted to do. They self-funded all of their capex.

Do you mean individually as a company or as a Ventures portfolio?

Both. But any of the capital expenditures we have done have been done within the context of each individual company.

From the free cash flow of the subsidiary?

Yes.

You mentioned the return on capital. You have here tangible capital, total capital, and adjusted operating income. How do you measure the performance? Do you use a return on invested capital, or do you use return on equity for the Ventures portfolio?

We are not including leverage in that, so it is not our ROE number, it is return on total capital. All of those businesses, some of them have zero debt on their own individual balance sheets. Some of them, which have a little more in the way of physical assets where it is a little more appropriate to use some long-term debt, we have some. But I would suggest that all of them are pretty conservatively financed, and any that have industry competitors or things that look like them, we would tend to operate on the low end of the scale of leverage that we would use.

So that total capital is majority equity then?

That is correct.

Do you use pre- or post-tax for ROI?

I do not think there is a distinction. You get different numbers, but they directionally say the same thing. When I talk about double-digit returns on the equity capital of Markel writ large, that is after tax. Everybody has to get through the after-tax return on capital thresholds.

Do you not adjust the adjusted operating income for tax when you calculate that?

Both of those calculations will give you a number. One is a pre-tax number, one is the after-tax number. They are both conveying the same directional information of whether this is good or not. What I am saying is it has to get to a double-digit return after tax for us to say this is good. The number that is calculated using pre-tax by definition is going to be higher because it does not have the tax burden on it. But we pay a lot of taxes so we are not kidding ourselves that if we get to a double-digit return pre-tax that we are done. We are not. We need to make sure that we have provided for our taxes to get to the double-digit returns.

How do you think about how Ventures can compete against—I imagine they are mainly private equity or strategic buyers to these assets because of the size of them? You have a slide in one of your reports comparing Ventures to a fund model. I think it was 400 basis points in drag that the PE fund has over a five-year period of fully deployed capital. Because of the structures, it is very favorable to Ventures. How do you then think about competing and bidding and multiples you can pay versus private equity?

There is a quantitative and a qualitative answer to that. The numbers that you just quoted speak to the quantitative advantage. We do not have a management fee that we charge those businesses. We do not have a carried interest of compensation for the senior executives of Markel. So there is a cost advantage that is quantifiable that would suggest that the shareholders of the Markel Group will, like for like, end up earning a higher return by owning those businesses through ownership of shares of Markel Group than would be the case if they owned it through a fund structure. That is just math.

The more important thing to me is the qualitative difference. For instance, when I am in a conversation with a potential seller of one of these businesses, oftentimes the conversation inevitably gets to the point where I say, look, there are essentially three kinds of buyers out there. There are the PE firms that are out there and very active. They are and probably will be able to pay you a higher price than what we can pay. They are going to use leverage to do a transaction—that is not right or wrong, it is just the way they would do things. And three, five, seven years down the road, this process of selling the business, which is intense and somewhat disruptive, you are going to go through that all over again because that is the nature of the way things work. So if you are looking for a higher price, they will pay a higher price, but just have eyes wide open of what it is you are signing up for.

The second big category of buyer out there is a strategic buyer, which is code for they are in the same business that you are in. Strategic buyers do not need two CEOs, two CFOs, two heads of sales and two heads of R&D because they already have them. Your team that you have built this business with, their future is probably going to be different than what would be the case if this business carried on as it is. They can pay you more because they will have synergies and those will be cost saves that they factor into the way they would calculate what they can pay. We will not have those. So again, they could probably pay a higher price than what we could pay.

If you want your business to continue to grow and compound over time and have the ethos of long-term decision making and not be running towards a specific exit or change the fates of the people who are part of this organization, there are not a whole lot of buyers like us who really are trying to build long-term generational type businesses. That is fundamentally an attractive proposition for some people, and for other people it is not. It is okay. It is a big world. There is plenty there for both of us. What we have found is that the people who we are for, we are unique. There are not a whole lot of people like us. And generally speaking, the people who have joined us seem to be pretty happy and they do well.

In fact, of the 23 or 24 businesses we bought along the way, I think the original CEO at the time of the deal, including AMF Bakery, is still in place running the business years later. To me that means that they did what they said they were going to do, and we did what we said we were going to do. That is a very important tell, and they tell their friends about us. So in terms of deal flow, they know people in their industry or in their trade associations or in the place where they live, that they provide a reference for us that provides us with conversations and opportunities to talk to people that otherwise would not be the case. I think that is a meaningful competitive advantage over time.

And the last thing I will say about it. So for instance, if someone is thinking about joining Markel Group, one of the things I will say to them is here are the 25 companies we bought. Here are the names of the CEOs. You call them, pick anyone you want and see what they say. So that is a vetting process that I think compounds over time. Relationships compound. And when you are looking at compounding capital, compounding capital is an outcome. The process which builds that outcome is the compounding of relationships and technical, technological skills. Both of those work together in a multiplicative way to create the compounding for the organization as a whole.

And typically, out of the 20%, do you buy 100% or do the owners, sellers, roll some equity into an earn-out or just keep some of the equity?

Once you have seen one deal, you have seen one deal. It is bespoke. Now, from a threshold point of view, it is usually advantageous to buy at least 80% because then you can consolidate on your tax return. We have done deals at that 80% threshold. We have done them at 100%. We have done minority interest as well. But all of them support the notion that at the end, the entity that is designed to last in perpetuity is the Markel Group.

So when we initially set up a deal, we also set up the definitions and the process by which if the management retains some interest, what is the formula, what is the way, what is the mechanism by which they can sell that to Markel Group in the fullness of time. And sometimes those time frames are determined, sometimes they are indeterminate. But if we negotiate a deal, I want negotiation only to take place one time and that is at the inception of the deal. We will not have subsequent negotiations where we are partners. We define that all up front.

How many opportunities do you look at per year?

I do not want to quote a silly number because when you say how many do we look at, the phone rings a lot. How many do we seriously look at? How many do we have the second and third and fourth conversation with? In today's environment where prices in general have been relatively high, call it a dozen in the course of a year, roughly.

You are on the investment committee that uses this. That is when it comes to you, when there is a serious opportunity after Andrew and his team have done some qualification, I imagine?

That is correct. They would oftentimes be the first call, the first look through their networks, through their conversations, through the things that they do, through their daily work, through their conversations with the CEOs of the business. It is more and more the case that they would be the first call. Sometimes somebody approaches me about something, we would talk about it as a group and we would decide let's learn a little bit more or that's not for us.

What would be great in your mind in terms of Ventures performance for the next five years?

If we continue to earn double-digit returns on capital and we continue to find opportunities to deploy that capital at those kinds of rates, that is what success looks like to me from a financial perspective. I don't have a numerical target or a specific goal. It is go to work every day, be thoughtful and do what should be done and don't do what shouldn't be done. The second part is more subtle but it is the more important part. Avoid the errors of commission that Buffett and Munger talk about - the institutional imperative. People think that they have to do things. Our team that oversees these businesses has plenty to do. They have 20-some businesses to look after. They have a full plate. So if there are a paucity of other opportunities to add to the list, that is okay. We have work to do on what we have already assembled.

How many errors do you think you've made in Ventures portfolio?

Some. Our batting average is pretty good if you ask me. Call it 20, the 20-some deals we've done. There are a couple that have had some challenges to them. That's just reality.

What were the lessons from that?

In those four criteria - the business that earns good returns on capital with relatively low debt, the management teams with measures of integrity and talent - the business ended up experiencing more challenges than what we foresaw. In today's world, the competitive landscape changes more rapidly and substantively than previously.

Is that mainly industrial?

I think that is true across the landscape of business writ large. No matter what business you are in, the winds of change are strong. So sometimes conditions change from what they had historically been and what we would have thought they would be going forward. Similarly, I have made some mistakes about the people and we have had some different points of view with some of the management teams.

It is one thing to date somebody, it is another thing to marry somebody. I am not perfect. I make mistakes and we will probably continue to make mistakes. We will try to make those mistakes in such a way that we don't make any fatal mistakes. We always learn something in order to make a better decision the next time we have a decision to make.