This conversation is the first in a special series we are calling Intelligent Investing 100. Every week for the next one hundred weeks we will bring you an interview with a recognized thought leader in the field of investing. Stay tuned for a new release every Thursday at 11:30am ET.
We kick off this series with timeless insights from one of the legends of quality investing, Montreal-based superinvestor François Rochon, president and portfolio manager of Giverny Capital.
Over more than three decades of following the same investment strategy, François has delivered strong outperformance versus the benchmark.
In this wide-ranging conversation, François discusses:
the core tenets of his investment philosophy;
quantitative vs. qualitative factors in the research process;
portfolio construction and risk management;
Giverny’s selling discipline;
AMETEK vs. Stericycle, and the Philip Carret rule;
the Carmax case study;
developing rationality, humility, and patience; and
building a great team at Giverny Capital.
This interview was recorded in May 2021. The following transcript has been edited for space and clarity. (MOI Global members, access all features, including ways to follow up with François.)
John Mihaljevic, MOI Global: François belongs in the echelon of superinvestors in the world of value investing and quality investing. How would you describe your investment philosophy?
François Rochon: Our approach is simple. To me, investing is owning great companies. When we build a portfolio of 20 to 25 names, we have the same attitude as if we were buying for the long run 25 outstanding businesses that we want to own for many, many years. They share some characteristics in terms of that quality. Because we want to own them for many years, they have a strong competitive advantage that should help them continue to create wealth at a higher-than-average rate.
MOI: François, let’s talk a little about your investment process. What I find notable is that you apply both quantitative and qualitative analysis. Could you talk a little more about each of those and how you sequence them in your process?
Rochon: We want to own great businesses, and the best way to identify them is to look at their past numbers, their past performance. We look at what you describe as quantitative. We look at the numbers of the company from the last 5, 10, or 15 years to understand how they did during recessions. We look at average return on capital, margins, debt level, and the quality of profits in terms of the accounting they use. That’s the first step. We want to find strong companies with strong numbers.
But also, we realize the limitations of an approach that focuses too much on the past because past numbers only reward past shareholders. When you buy shares today or you stay a shareholder in a company, it’s the future that will decide whether you get rewarded or not. We want to be certain the ingredients that created the great numbers in the past are still present in the company so the future will be, let’s hope, as good as the past. This is where the qualitative analysis enters. We try to understand why the company performed so well, why it has been able to be so profitable in terms of margins or return on capital. We try to understand the nature of its business model, its competitive position, its competitive advantage. That’s a key part of the analysis to understand if it has a moat and how durable the moat is.
Also, we have learned over the years that one main reason a company does well – does better than the others in its industry – is usually because a management creates great performance. Yes, we want a moat within the business model. But we also want good management because a moat doesn’t usually come from thin air. It has been created by human beings, and it’s also sustained over the years by human beings. We want to be sure we have a management that understands the importance of building a great culture. Another part of the importance of having a great management is how they’ll allocate capital because if you’re a long-term shareholder – let’s say 5 or 10 years – a big part of the rewards will come from the management’s capital allocations. We look at how they bring new products to the market and how they make acquisitions. Were intelligent acquisitions made at sensible prices? Also, they can reward shareholders by either paying dividends or buying back stock at opportunistic moments when the stock trades at a price less than its intrinsic value. The quantitative approach is the first step then we go further to try to see the qualitative ingredients, to see if they are present for the company to continue being outstanding.
MOI: Perhaps it’s most instructive to illustrate this with an example or a case study. I’d love to talk to you a little about CarMax which, as of this conversation, is still one of your top holdings. Could you tell us a little about the genesis of that investment and why you decided CarMax met your criteria?
Rochon: Yes. We invested in CarMax probably 14 years ago – in 2007 the first time. What we liked about CarMax is we believed it had a good culture in which it sold used cars to customers in an ethical way. The prices were fixed, and CarMax would give customers a guarantee. Also, when we bought CarMax, it was a big company, though it only had something like a 3% market share. Not only did this company seem to us to have a good business model, but it was also already one of the main players in the industry. It had changed the industry. But also, this was such a big industry that the company had many years of growth in front of it. Today, 14 years later, CarMax has more than 200 stores.
One thing we were worried about when we first invested – and that’s why we started with the small weight in 2007 – is that we didn’t know how the financial division would do in recessions. I don’t want to say we were lucky, but in some ways we were lucky because we had a good recession in 2008 and 2009 through which the company maintained its profitability. Earnings went down in 2008, but the company was still profitable. It was able to sell its car loans on the secondary market which reassured us that the business model was solid. We’re here 14 years later. If you don’t count 2020, which was a year out of the ordinary, earnings per share have probably grown from 2007 to 2019 at about 16% or 17% per year. The company has been a good wealth creator. It grew sales. It grew profit. It improved margins. It improved return on capital. It even bought back a lot of stock over the years, so it’s been a rewarding company. But the company probably still has less than 4% market share, so it can continue to grow for many years. Last year the company reported its targets for the next five years, and management believes they can sell 2 million cars per year by 2026. That’s a compounded growth of about 10%. They believe they can grow revenues by 12% annually and grow market share to more than 5% by 2025 or 2026. We think they can buy back stock.
One key element for that growth prospect to materialize is probably the company’s success at adapting to new competitors that sell directly on the internet. Over the last three years or so, management laid out an omnichannel approach. You can either buy on the web or buy at the superstores. Or, you can do a combination of both. It was a big project for CarMax to offer the omnichannel approach in all of its markets, but slowly the omnichannel approach emerged. Now the company should be well-positioned to continue its growth. The company should be able to continue growing at a rate similar to its growth rate from 2007 to 2019; that would be around 15% or 16% annually. If you look at the valuation, the stock trades at something like 22x on this year’s earnings, so it’s still a reasonable valuation for a company that we believe can continue to grow at least twice the growth rate of the average company in the S&P 500. This is a company we believe has a strong brand, a strong competitive moat, and a great management that has made good capital allocation decisions over the years. We expect management can continue to make good capital allocation decisions, and the valuation is reasonable.
MOI: In terms of the competitive moat, how important is the appraisal database the company built up over the decades?
Rochon: It’s a very important competitive advantage. Yes, I agree that CarMax can probably price cars better than most of its competitors. Also, it’s an advantage to be able to either sell on the internet or sell in superstores because most people will still probably want to test drive a car before purchasing it. CarMax has a great brand. It has a great reputation. Consumers see CarMax as a reliable place to buy a car. The moat is not just in the operation and the experience the company has accumulated over the years, but also in the brand itself. In the mind of consumers, CarMax is a good place to purchase and sell a car. Sometimes you just want to sell your car to CarMax because you know the price will be fair and they’ll be quick. I have a friend who went to CarMax and sold his car in something like 15 minutes. It’s an easy process because the company has so much experience.
MOI: Can you talk a little more about capital allocation at the company? Is there anything you might prioritize differently than the management?
Rochon: No, the management has done a great job finding a balance of investing in their business – mostly in the omnichannel over the last few years or in opening new stores. They plan to open more than 10 stores a year, which is probably 5% or 6% annual growth. What is left in capital they’ll probably use to continue buying back stock. The balance of capital allocation between investing in the technology for the business, investing in new stores, and buying back stock – I think it’s the right balance.
MOI: How do you think about something like the competitive threat that arose with Carvana gaining prominence in the industry and having a market valuation that would seem to suggest high growth expectations?
Rochon: Yes, it’s been a big threat in some ways to CarMax and other players in the industry. CarMax seemed to take the threat seriously and addressed it with its omnichannel strategy. If you want to buy a car through the internet and only through the internet, you can do it with CarMax as easily as you can with Carvana. CarMax has addressed that competitive threat. Why is Carvana valued more by the market than CarMax? It’s probably because the market loves it when the growth rate is high. But to my knowledge, Carvana has not yet reached a level of profitability, so it’s probably a riskier business model than CarMax’s model. But it’s been a great success. I cannot deny that. I’m happy CarMax took that threat seriously because it changed the industry. Both companies, CarMax and Carvana, can prosper in that industry. It’s a big industry, and both companies are good companies. I don’t think just one of them will survive. Both could do well in the future.
MOI: François, when it comes to portfolio construction, could you talk about any guidelines, any principles you adhere to in terms of the portfolio itself?
Rochon: Yes. We’re quite diversified, though not as much as many pension fund managers or mutual fund managers which sometimes have more than a hundred securities. We have about 20 to 25 securities in the portfolio, and the average weight is probably about 4%. That number is probably the right balance between not being too focused so that if something goes wrong, the company can hurt your portfolio. At the same time, you want it to be concentrated enough so you have chances to beat the index; because the more stocks you have in your portfolio, the harder it will be to beat the index. That number – 20 to 25 names – seems to be the right number to attain that balance between diversifying the risk and simultaneously having the odds to beat the index, the averages. I’ve come to learn also that it’s hard to know the future. You must accept that probably one stock out of three that you purchased, even if you’ve done your work well, won’t work as expected. That’s one reason you want to diversify because you don’t know the future. When you buy a stock, you’re buying the future of a company, and no one knows the future. We try to invest in companies for which the future can be as knowable as possible, and in companies that can control their destinies. They’re not hostages of things they can’t control. Like companies in the natural resources industry, it’s hard for them to do well if the resource doesn’t do well. We try to avoid companies that depend too much on a commodity.
Even though you find companies you believe have good fundamentals with good balance sheets, the business world is tough, competitive, and always changing. You must accept that, and you must manage risk by having a certain amount of diversification. I think 20 to 25 names is a good number. We manage those weights by tending to keep the companies that do well and, as quickly as possible, sell the ones whose fundamentals are disappointing. In other words, we try to sell our losers quickly and keep our winners for many years. The winners will become a larger weight in the portfolio because we have kept them for many years and because they’ve done so well. I’ve sometimes made the mistake of trimming or selling a company that looks a little expensive, that has become a larger weight in the portfolio, and whose valuation seemed a little high. That’s not a winning strategy. The winning strategy is to hold on to your winners and leave them with perhaps a higher valuation than you would like to. But at the same time, we have this rule that if one company does so well that it becomes more than 10% of the portfolio, we’ll trim it. We don’t want to have more than 10% in a single security. The only exception we’ve made in the past is with Berkshire Hathaway because the company itself is so broadly diversified. The intrinsic risk of the business is low. We have lived with higher than 10%. At one point, we had 15% of the portfolio in Berkshire. Today, it’s closer to 11%, but that’s the only exception we’ve made.
MOI: The point you made about being quick to take losses and reluctant to take profits – you write about this in your annual letter and call it the “Philip Carret rule.” Could you talk a little more about that and perhaps tie it in with a couple of examples from your portfolio, in Ametek and Stericycle?
