Interview. Bill Nygren on 25 Years of Beating the Market

(upbeat music) Hello and welcome to the Synopsis, a business and investing podcast. My name is Drew Cohen and I am really excited to talk today to Bill Nigran, who is the Chief Investment Officer of U.S. equities at Harris Associates. He manages two of their funds there, both the Oak Mark Fund and the Oak Mark Select Fund, the Oak Mark Fund since 2000, has had an annualized return of roughly 12% a year versus the S&P at around 10%. So beating the S&P by almost two points a year over a very long time frame, so phenomenal record and really excited. Welcome, Bill. - Thanks for having me Drew, great to be with you. - Just to kind of kick us off, I was wondering if you could just start with very broadly your investment philosophy. Everyone kind of talks about value investing, but what does that really mean for you and the way you practice it? - Sure, we absolutely would say that at Oak Mark, we're long-term value investors. I think one of the easy ways to think about it is to say we bring a private equity perspective to public equity investing. And what does that mean? It means we look out farther into the future than most public stock market investors do. We like to think out about seven years into the future. And how do we think it's likely that investors will be valuing this business seven years from now? And then we discount that back to the present and we look for large gaps between our estimate of business value and where the stock price is selling. I think that the difference relative to traditional value investing that's important to highlight is that we don't always end up in the lowest PE or lowest price to book names. And it's not that we intentionally try to avoid those, but we think the true value investing universe is much broader than is defined on gap accounting. And it's largely because so many companies today make growth investments via the income statement. At the start of my career 40 some years ago, if a steel company wanted to grow, they had to build a new plant that went onto the balance sheet. It would be depreciated over the entire life. But that's not how advertising expense, R&D expense, customer acquisition costs are treated. And so a lot of the ideas that we end up owning in Oakmark that you don't see in traditional value accounts are these names where we've adjusted the accounting, we call it Oakmark accounting instead of gap accounting, to where we are trying to treat these intangible growth investments as longer term, as having longer term value to the company. Yeah. And so I think just to broadly summarize there, one of the big differences in the way you quote unquote, value invest is you're thinking of-- you're basically agnostic to where the growth is coming from, whether it's existing cash flows, cash flows coming from actually growing the business. And then you're discounting it back to today. And I believe you said on another podcast you're looking for buying a company at around 60% of what you believe the actual valuation is. Yeah, that's about right. And if you think about what market expectations are, I would say most people would expect that over a four or five year period, you make something like 50% in the S&P 500 or just a broadly diversified basket. So we're looking at the companies that over a four to five year period we think doubling is a likely outcome. And of course, we aren't 100% right, so we never fully achieve that. But I like to think in terms of being able to explain how a company we own in a five year time frame could double without seeming like we've stretched the assumptions too much. So when you were looking out seven years and you're modeling out the cash flow seven years, I mean, we all know that whenever you're valuing a business, it's the valuation as the sum of all discounted cash flows for the entire life of the business. But it kind of reminds me a little bit when I was at CalPo Group, there was a portfolio manager there that says, I really only look out five years. And then I know at that point most investors are only looking out another five years. So I don't, I really only care about 10 years. And so it's kind of interesting how you're kind of coming around a same time frame of seven years. Why is that the right time frame instead of the entire life of the business? Well, if we thought we had equal clarity for year seven and year 30, we would absolutely look out further. I think the biggest difference between fundamental value investors and fundamental growth investors is how far out do we say our crystal ball gets so cloudy that our projections are really not important to the value case. And like you'll read in some growth companies, they'll be trying to look out 15 years, 20 years. And then they discount that back at a really high discount rate. We instead say, after seven years, we're just not comfortable saying that a company isn't going to grow at a normal rate. You could pull that back to five. You could grow it to 10. And we always have internal arguments about in that range what's the right number. But we've settled on seven because that's a longer time period than most other investors are looking at. And it's not so far into the future that it feels like we're doing coin flips. And I hear you that when you get farther out, of course, it's harder to really estimate out what the cash flows are. But aren't you effectively still doing that in your terminal value calculation in that year seven? Yes. But our assumption is generally at year seven that we don't know enough about the business anymore that we don't want to project a multiple that's much different than the market multiple. So if a company's not going to grow its way into a discount multiple by year seven, we look at it and say, you know, that's better for the growth investors than it is for us. I see. So you're really looking at growing earnings to get to a market multiple or hopefully much below that within a shorter time frame than that. Right. And I think one of the reasons that we think about it like this is I've had a fairly long career. And I remember when the safest businesses in the world were landline telephones or newspapers. Those were always the examples of-- well, we know we can look out 20 years because what could possibly happen to disrupt those industries. And we've seen whole industries fail. So I think there's the margin of safety concept that you hear in like Graham and Dodd style value investing. We're kind of applying that by saying there's a margin of safety by not having to go out further than seven years. Estimating that a company will be doing something super normal during that time period. And so when you're then buying at this 40% discount to that valuation that you're getting, and you discounted those cash flows by some rate, isn't that basically the same thing those is discounting all of those cash flows at just a higher rate of return? Probably. Yeah. So then how do you figure out what to discount them at? Is it just a hurdle rate, is it whack, or how do you decide to do that? And we start by using a long term treasury. What kind of premium to that for moderate quality corporate credit is implied in the market. And then we look at how much more we think the S&P should be discounted than a typical corporate bond. And then we'll put a risk category on the company, on a one to five scale, with ones requiring a lower discount rate than the S&P and a five requiring significantly higher. And so after you kind of do put that formula together, why chop off 40% is that just kind of what's worked or makes you comfortable? We don't want false precision. Sometimes it'll make me laugh when I read other value managers say that we're buying this stock at $60, and our value estimate is $75 and $12. It's kind of a reminder to us that there's not precision in this. We're trying to get into the right ballpark. And 40% is a big enough number that if we're off by a standard deviation or even a little bit more than that on our estimate, we've truncated our upside, but we haven't created downside. Yeah, that makes sense. And you're kind of poking at it a little bit in that question, but I want to ask you, is there such a thing as fair value? I-- you know, hard to answer that question. I think it's hard to believe in value investing if you don't think there's some concept of fair value. But the idea that fair value is a specific number that you can compute with precision, I don't think you can get there. Can I look at a company that's selling for $60 today and say, I'm pretty comfortable. A private buyer would pay somewhere between $90 and $109. in 10. Yeah, I think you can estimate value with that kind of precision. Do you think fair value is the same for different investment funds? Because I can imagine maybe an endowment or something with kind of a 7% return mandate. They may be okay assuming a lower rate in their DCF and so their evaluations would be higher. Well, I think if you start saying every investor has a different fair value that's accurately computed, you get to a really odd place. Just because they have a lower required return than some other investors might, I don't think means they should pay more for a specific stock. It might mean that they can be content taking a lower risk that could be achieved through different securities, but I don't think it means they discount alphabet cash flows at 3% while I'm discounting them at 8.5%. Don't you think though that it basically kind of just comes down to everyone has, I'm trying to say basically when you're investing, you're kind of deciding what the proper or like the market hurdle rate is for a stock and then it should be kind of uniform. In that case you're saying it's agnostic of whatever your own capital requirements are your own funding needs. Does that make sense? Yeah. And I think when you think about the process of trying to estimate fair values on a lot of different public equities, precision in the number isn't nearly as important as rank ordering. If you've applied consistent assumptions across a group of a couple hundred businesses, you look at the names that look most attractive, whether they're undervalued by 40% or 20% or 60% isn't nearly as important as that you've identified the ones that are most undervalued. So again at Oakmark, our whole process is on making sure we've rank ordered our ideas in a way that we think reflects how attractive they are. So I just wanted to kind of follow up on that line of thinking a little bit. So I know you invest in a lot of financials. Does that mean that all banks together, if you think about them, they should be willing to lend loans at sort of the same rates regardless of whatever their cost of funding is? Well, again, I don't think just because JP Morgan has very low cost of funding relative to your small local bank means that they have to drop their lending rate to their cost of funds. I think that's why they're more profitable. The right rate to lend at is kind of the same across all the financial institutions. And it's just why the companies that have lower cost of capital like JP Morgan can profitably do so much more lending than companies that have higher cost of capital. And just staying on banks for a moment. So I covered banks when I was at Goldman and it was always interesting to me that you never wanted to get a quote unquote bank multiple because banks always would get lower multiples. So whenever there was any financial institution that had a little bit of tech in it, it wanted to become a fintech and covered by that sort of analyst. My question was, why do banks tend to get lower multiples? Well, first, I think if you look over a longer time period, like 50 years instead of the time period that you were an analyst, there was a time bank multiples were good multiples. Utility stocks would strive to get bank multiples instead of the other way around today, where one of the arguments against owning bank stocks is they're starting to get so heavily regulated they look more like utilities. It's like I'd be really happy if my bank portfolio got utility multiples tomorrow. I think one of the reasons banks get lower multiples is because there is such a hangover from the great financial crisis when we saw the banks having balance sheets that were too levered lending standards that were not tight enough and the damage that that did to the banks. Prior to that, banks kind of traded in what they'd call today a risk off trade. They had betas lower than one. They were viewed as the safe part of the portfolio and pretty much ever since the GFC, they've been viewed as when the market wants to assume risk the bank stocks do well and they're viewed as betas higher than one and riskier names in the portfolio. We think that's kind of deviated from the fundamentals and that's one of the reasons why we own so many banks. Relative to what these companies looked like pre GFC, they tend to have maybe twice as much capital per dollars of assets that they have. They've got tighter bank lending standards. I think the economies of scale argument is favoring large banks more today than it did 30 years ago. Things like mobilization, anti-fraud activities, dealing with government regulations. Those are all more expensive for really big banks than small banks but not by the same factor as the assets are larger. So the bigger the bank gets, the more of an expense advantage they have. I think it's going to take a generation for bank investors to stop focusing so much on the GFC and to think more about what normal recessions are like. Something I've been fond of saying is if you're under about 45 years old, you weren't in the business for the last normal recession and the models that are trained on how companies will behave to a bad economic environment. They've seen the GFC and they've seen COVID. Those are more what I would call generational recessions as opposed to the kind that was common early in my career where you see a quarter down GDP for the next quarter you're arguing about whether this is going to turn into a recession or not. The second GDP print is also negative so by definition now you're in a recession and by the time Wall Street agrees that we're in a recession we're already working our way out of it and you might even see a year with positive GDP growth include a recession. Those types of recessions were modest earnings events for banks not capital events. So I think getting through a couple recessions like that and banks showing how resilient their earnings are is probably a necessary precursor to seeing these companies again get back to being viewed as lower risk businesses and having multiples that are closer to market multiples. And that makes sense and I know you're not a macro guy but I'm wondering if you have any kind of take us to why there hasn't been a normal recession and so long. Well one of the reasons is when we had what we would normally have called a normal recession I think it was back in 22 where the first two quarters of the year were down GDP. Politically we decided to not call it a recession. They said well yeah that's kind of the technical definition but for other reasons we don't think this really fit. You look back on that it didn't touch the banks at all. Credit didn't get markedly worse. We saw very tiny reductions in earnings for that six month period. I think we will see those kinds of recessions again. Once a century pandemic I think it would be a mistake to look at as normal and then the reason we call it the GFC it wasn't quite as bad as the great recession but it was a lot worse than a normal recession so I need to call it great something but the very name suggests that it's not the kind of thing that you expect to see every four or five years. So it's something like banks you know there still is going to be a credit cycle. There's still going to be a little cyclical but you believe generally speaking investors have in their mind a much worse ad for scenario than is really likely to happen. Yes I think investors look at the Fed's adverse scenario stress tests as being kind of what will happen to a bank in a recession and my opinion and Oakmark analyst teams opinion is that it would take a really severe recession not the kind we see a couple times a decade to create those results. So if we think about like a P multiple is basically being a shorthand for a DCF it's kind of easy to put a multiple in a stock where earnings are flat to growing and we could get some sense of the right multiples. It gets tricky though when you have a couple years that are going to be earnings down and there is a little bit of that cyclicality and so I was wondering I don't know if you invest in that many cyclical businesses but banks are arguably a little cyclical so I was wondering how that kind of enters your valuation process. So in any company we're looking at we're trying to make sure we're discounting mid-size. learnings so that you account for both the down periods and the up periods in the estimate that feeds your DCF because the last thing you want to do is value the company on a boom period that's only going to happen maybe a couple times a decade or on a really weak period. So we try to make that adjustment through the income statement and the level of earnings that we discount. So for banks one of the things we look at is how much better than the average do they do in really good years and try to make sure that we're discounting those but then also including how far below average earnings go in the bad years and make sure you account for both and discount normal. Do you ever really know what mid-cycle earnings are until they kind of pass though? Well it's kind of like your question I'm fair value. Do you know exactly of course not but are you getting better? Are you getting a better answer if you try to make those adjustments than if you don't? We think we definitely are. Are there some businesses like I'm thinking of AMD where I tried looking at that and it's it seems like a very wide range of potential estimates you could have for whether or not the chip boom continues in a few years from now or not and I don't know that you're invested in them but are there businesses that the cyclicality is just too much to try to estimate for? I think one of the reasons we've rarely ended up in in the semi-space is that it's we found it always really hard to get comfortable with the level of earnings that that you know we feel highly confident is going to be you know achieved on average over the next five or seven years we we don't like routinely exclude any industries because of that but I think just this the way the process works there there are industries that don't show up on our approved list as much as other industries do. The three things we look for in a company the 40% discount we were talking about management that's aligned with the shareholders and an expectation that value growth plus dividends will at least match the market you know the management criteria of management that think and act like owners we found that really hard to find in the electric utility industry those managers don't tend to think as entrepreneurial as we like and it's one of the reasons we've rarely ended up invested there. Some companies in some cases semi-conductors, biotechnology, the potential earnings generators are so far out in the future we can't get a confidence level on them and it's unlikely that they end up in our portfolio. So we definitely have a bias toward companies that you can project them kind of continuing to do what they've done recently to end up in in our portfolio as opposed to those that have to do something radically different than what they've done in the past. If you were thinking about like the downside scenarios for the cash flows of the business that you would want it to earn in order for you to invest in it how confident are you in that downside scenario is it like 90% if you had to put a number on it? Oh I don't think we ever get to 90% as much as we'd like to be there on any of the companies that we look at. I think you look back over the past decade or so and find what the worst year has been what the factors were that contributed that try to make a reasoned estimate if that occurred today what that how much downside that would create and then also to do the same the same exercise on what could happen on the upside. I think a lot of times you find value investors again unnecessarily limiting their universe by the pride they have in conservatism and they'll spend a lot of time looking at downside cases but not offset that with upside something that again differentiates us a little bit at Oakmark is we prize accuracy in forecasts not conservatism. And when you're talking about your kind of like three aspects within investing one of them you kind of talked about was management alignment and I'm guessing there's an element of making sure incentives are aligned but then probably also gauging the quality of management. Yes both. So we like to spend we spend a lot of time looking at the history of the CEO both at the company were investigating and prior to that also the same for the rest of the management team. We we look at incentives we like them to have denominators you don't want the management that just by making big acquisitions will hit sales growth or earnings growth targets. We want earnings per share growth business value per share return on invested capital not just income and then the other thing we we like to do is we like to have one-on-one face-to-face meetings with management. Not so much to help us improve the accuracy of our DCF but more to get a feel for is this somebody I want to be invested with does it seem like our goals are aligned. Five years down the road they're looking back on their success are they're going to be are the most important metrics to them the same things that we think drive business value and stock price or is there something else that is important to them for their legacy. You know we we want to think that the next dollar they have to invest will get invested into the highest return opportunity whether that's growing the top line or shrinking the share base and you know management that think like that we're very anxious to align with. And I want to look back on the management part but first I because you talked about ROIC and you know we all know Charlem Lunger talks about you on a business with a high enough return on invested capital and they continue to invest at that rate over time your return as an investor will converge to the ROIC. That's over the super long term though if you do the math on that and so is wondering how important is ROIC to your process. It's very important I think it's it's a big factor in judging the quality of a business and therefore the appropriate discount rate. The businesses that have a good moat that have high return on capital are just definitionally better than businesses that have low return capital and you should be willing to pay a higher multiple for it. I think we try to trim the risk of eventually competition eating away at that ROIC by assuming that you can only forecast out seven years and then businesses become normal pretty quickly following that. So yeah I think ROIC is really important but to Charlie's point you know I think investors over the past decade have kind of been trained. If you identify a good business you can just buy it you don't have to worry about what your entry price is because that's how stocks have behaved and I think because it takes so long for the investor return and the companies return to converge I think that's a very risky way to think about investing. Right because you know you could buy Costco which has a great return on invested capital now but at a 50 times multiple you might need to hold it until you're very very old in order for those to converge if it ever does. Right and I think you know at Oakmark we think our holding periods a lot longer than the average investor and it is the average investor might be a year or two we're more like five to seven years but as you say in Costco's case you know it might take 30-40 years for those return lines to converge. Yeah and it's really interesting your investment philosophy because there's like you're talking about a lot of value investors I want to hold a business for the entire life and then they never really reevaluated over that time they don't really want to sell it they're still holding it even if it's very overvalued and there's something kind of a little bit more honest about your your holding period and your time frame that feels a little bit like a happier balance between this idea that you can hold a business for its entire life because we also know that most businesses five years from down the S&P 500 are going to be worst businesses. Yeah and as much as we would like to take the Buffett approach of you just you buy and hold forever if our process is right that we can identify businesses that are selling at large discounts it only makes sense that you'd fund those purchases with the companies you own that are no longer selling at large discounts. To me there's a little bit of a momentum factor that creeps in if you say once I bought the stock it's now anointed as a hold it forever company that just doesn't feel right to me. I think it's important to have sell targets that are based on business value estimates it's important to update those with new information but when a stock exceeds the the price you think it's worth just because you owned it is not a good excuse to continue owning it. Yeah, and I think what investors might be doing in that scenario is they're just looking at that return on invested capital and saying, "That's my return. I'm going to hold it forever and I'm never going to make more or less and hope it just kind of converges to that." Yeah, and there you're basically hoping that over a generation time frame, the competitive environment doesn't change, which to me in a way is a little bit like betting against capitalism. You know, the way capitalism is supposed to work is you have these pockets of very high return on capital that will attract lots of new capital to try and mimic those returns and in the process drive those returns down toward average. I believe that does work. I believe that even works in the new economy that we're in today. And reversion to the mean might be taking longer than it used to, but if you don't believe that concept exists, then you get to some really crazy outcomes on how you should be investing. Yeah, and all that makes sense. I'm going to push back a little bit just for the sake of it. When you are selling, let's say, a great high-quality company, and I'm sure you've gotten this pushback before, the great companies tend to surprise to the upside. And so business might be able to do better than you even expected to do and you're at risk of selling out too soon and missing kind of this long-term compounding period and just look at how much money Warren Buffett made off of C's or Coca-Cola or Moody's maybe a little more mix. But all of these businesses you're holding for a long period of time and you risk basically, once you sell it, you kind of miss some of that upside, which I know you've held Apple for a very long time and that's kind of an example there. Right. First on those very long-term holdings, I think you're right, the returns got very high over an extended period of time. But I think also in all the examples you named, you could argue that by extending the holding period, you ended up owning it during a period where the returns were quite low on some of the consumer product names and things like that. Apple is a name that was in our portfolio. We haven't owned it for some time. But I think it demonstrates the importance of making sure you're adjusting your business value estimate and sell target for all the new information that has come out from the time you purchased it. When we bought Apple, we thought it was selling it a 60 cents on the dollar. We then made something like, I don't remember what it was, 20 times our money in the next decade. We never held it above our sell target during that period. It was that consistently quarter after quarter growth was exceeding our expectations and our value estimate kept climbing. In fact, for most of the decade that we held Apple, it was selling below a market multiple. It wasn't too challenging a thesis. You just had to say, I think Apple's at least an average business. If you believe that, it was cheap. The typical view in the investment community was Apple was over earning and this regression to the mean was going to come. As we've seen in lots of other consumer products, I think the reason it didn't happen is investors underestimated the switching costs to pull someone outside of the Apple ecosystem. There are lots of value investors that at the time they buy a stock, it's based on what the current estimate of business value is and then whether that's achieved in a year or three years or five years, they'll sell it when they hit that number. They're too anxious to take the win. I think we avoid that and kind of get to this middle ground you referred to by aggressively changing our business value estimates based on all the new information that we have. We would have sold Apple for a 50% gain and been out of the stock if we didn't adjust our numbers from what the analyst presented the first time we bought Apple. You could also probably push back by saying most of the gains in Apple, after you guys sold it, I believe around 2021, 2020, has come from multiple expansion, from going from a 20 times multiple to a 30 times multiple stock, not from earnings growth. Exactly. We tend to get queasy with those multiples businesses of that scale because the history has been up until recently. Once you get that big, size starts to become a pretty important negative and starts to move your returns down more toward average. I think one of the reasons we've seen the tremendous growth stock boom and growth managers outperforming value managers is we've seen these very large companies instead of regressing to the mean, they've accelerated their growth rates. We're not in the camp that says the Mag 7 is valued at multiples of what they're worth. It's just the growth rates have been so far beyond our expectations. We looked at something like if seven years ago you had looked at the Mag 7 stocks and made an estimate of what you think earnings would be in 2026, that number would have been half of what the current consensus is. The tremendous performance has been fully backed by fundamental earnings growth and in some cases like Apple, a little bit supplemented by PE expansion. When we get the PE expansion, that troubles us. The names that we continue to hold that we've held for a long time, like an alphabet, they haven't seen the PE expansion. Their stock price has pretty much just followed the growth of the underlying business. Kind of a little bit you're saying people have taken the wrong lessons from history by looking at these stocks. The compounder group of stocks that have continued to grow a lot, even beyond their fair valuation, if you will, because even though there's a lot of scenarios and circumstances, we can point to like that, we're forgetting all of the evidence of companies that have fallen off. If you're looking at the full sample set of those that we're not able to keep that status or much larger group in your losses would have been higher than hanging on to those as a whole. The fingers crossed, they're taking the wrong lessons because if they aren't, then our performance is going to continue to lag there. You think of all the sayings that value investors have come up with over the years. The buy low sell high trees don't grow to the sky. Everything is based on, there's over a long enough time period you get reversion to the mean. Some companies stop doing stupid things and eventually become average and really good companies despite best efforts have returns competed away and they too become closer to average. If we're truly in a winner take most economy where most industries end up dominated by one really successful player and the rest of the companies struggle to exist, that's going to be a really tough environment for people like us who say things over a long enough period of time kind of move toward average. Yeah, and that makes sense and I think a lot of people they look at, Phil Fisher, Motorola, kind of like a growth investor, but a lot of that valuations or a lot of the stock price movement was backed by real growth as we're talking about and maybe Meta is another good example of this kind of more recently where if you're doing a valuation of them in 2022 when revenues just contracted for the first time ever and in the midst of ATT and TikTok competition and all this, you would have maybe been happy assuming they're going back to positive single-digit growth, 10% growth. You were not expecting them to come out with a 30% growth estimate five years later almost. No, and in fact we owned Meta and Facebook before that. During a time period where you could say if you put a relatively modest multiple on Instagram and the blue Facebook and added back cash on the balance sheet that was significantly more than the stock price was selling for and that gave us a lot of comfort that we didn't have to develop much precision about what the value on what's app might be, what the value could be in the metaverse. This was before the hyper scalars was even a term but then that also resulted us in us selling the stock well before today's current prices because our thesis that this should be worth somewhat more than market multiples on Instagram and Facebook. Somewhere in the four or five hundred that thesis didn't work anymore. And kind of getting back to a thread we had a little bit earlier on the quality of management. Does talking to managers hurt or help? Well, we believe it helps us because if it didn't we could save a lot of travel time, a lot of travel money and business travel. I don't think many people say enhances their job satisfaction. But we believe that talking to management and understanding what motivates them, how they think, how they measure success, I think is not I think. It is a very important part of our process, and I think has been beneficial. Our management interviews end up quite a bit different than a typical management interview. Normally analysts are asking about the results of the past quarter, expectations for the next couple of quarters. Our discussions are much more about what are your hopes and dreams that this business might look like five to ten years from now. If you look back at the end of your ten year as CEO, what are the first things you're going to look at to measure whether or not it was as successful as you hoped it could be? And it's interesting the answers you get to that. And you know, the CEOs today are so well prepped for interviews by the IR team that I think a lot of times a typical meeting with them, it's almost like you're hitting the play button, and they're just doing the same interview they've had with numerous other managers. And if you're going to do that, you might as well just watch their last conference presentation. When we get them off script and talking about things that really matter to long-term rates of return, I think you're getting more transparency, more candor, and it's the kind of things that if other investors were sitting in the room with us, they'd be groaning at the questions because our questions aren't going to help them figure out their DCF model. But they're going to help us decide, do we have the alignment that we think is important to the business maximizing returns for the outside shareholders? And how much research do you do before you're ready to invest in a business? A lot. Our analysts will generally be looking at a company for months before it's recommended. They'll be looking at all the publicly available material from that company. They'll be looking at what competitors and customers have had to say about it. They'll be consulting expert networks. We'll generally have met with the management team in person prior to a purchase. Sometimes if we think the valuation is compelling enough, we'll flip that around. And that's the level of work the analyst has done. Then when I accompany an analyst or any of our other portfolio managers accompany them, they will brief us on what they've learned, what they think is important. We'll have a book printed out for us that goes through the resume, all the people that we're meeting, looking for, are there directors in the businesses we know? Do we know people they went to business school with that we can do more background work on, what kind of person is this? There's an extensive amount of work that's done prior to the meeting. And then we try to meet with our holdings in person once a year or at least every two years in their offices, plus also seeing them when they come through Chicago. And we do significant prep for those meetings as well. Have you ever had a business maybe you really wanted to own, but you didn't do full research on, and it's just sold off a lot. And you kind of just felt like you had to take action on that, or would the full research process prevent you from that? Well, if the analyst has done the work to demonstrate why they are confident, the current stock price is truly only 60% or so of what the business is worth. We'll definitely invest at that point. If a stock selling at 60, and I'm highly confident it's worth at least 100, we are happy to defer the work of, you know, whether that more than 100 is 120 or 160 until after we own the stock. If we can't get the range of potential values kind of in a spot that says, no matter what we're buying it at less than 60 cents on the dollar today, that's when we'll wait. So, the same example stock at 60, best guess, 100. If we say, yeah, there's one way of looking at it where it might be 70 and another way where it might be 150, we'll say we need to do more work on it, because we want the lower end of that range to still give us close to the 40% discount. How do you have confidence that you know everything though that you need to know, and there's no sort of hidden risk in that, because even if you could kind of understand why it's sold off, and you're getting a good sense of estimates and valuation, maybe there's some hidden like ATT risk in meta, which I don't think many people would have ever listed that as a risk like, you know, a few months before that happened. Well, I don't believe you can ever get 100% confident that you've identified all the risks before you purchase it, or that we've identified all the potential risks of everything that currently resides in our portfolio. But just like trying to get in the right ballpark for business value estimate, I think you do your best job to minimize the chances that those risks are there. It's analyzing financial statements carefully to make sure there isn't a problem with cash flow matching earnings. Watching to see if days sales and receivables have been expanding recently may be suggesting that customers aren't happy with the product that's been delivered. Again, you don't get to 100%, but anything you do to start to reduce those risks is a return enhancer, and we just try to do as much of it as we can. Does that also kind of go back to something you said where almost all businesses have a little bit of a black box nature, and that's why you need to trust management? Yeah, and that statement goes back to the way a lot of value managers reacted to the financial crisis back in '08, '09, where they said, "I lost a ton of money on banks, and it wasn't my fault. It was that there's a black box that I didn't know what was going on inside of." And I think the reality is, no matter how hard you try, to some degree there's a black box in any business you look at. And you go back to GFC, if I told you this bank you're looking at has most of its assets tied to commercial mortgages and home mortgages, and the loan to value looks pretty good today at 80%, but we're about to see values drop by 20 to 30%. And by the way, this company's levered 20 to 1. I don't think you'd say, "Wow, that was all a black box." What happened back in the GFC is you had the outcome on real estate be significantly more negative than we had previously seen, and the leverage was high and disclosed, but a lot of us hadn't thought through the negatives enough of if real estate prices went down by 20%, how big a problem is that going to be for the banks? Now, I think we know a lot about how General Motors produces light trucks and jeeps, but I'm not going to kid you for a minute that I could go through step by step at the assembly line, what the supply chain exactly looks like. We try to understand some of those risks, but there's still a little bit of a black box. And my point was when I made that statement, I don't think it's fair to single out the financials as a black box and not say there are things you don't know about other businesses that you own. I do think the trust in management takes on exponential importance when you're dealing with companies that are highly levered. I can look at an unlevered business today and be pretty convinced my range of outcomes is tight. Once you get a bank that's got a 9% tangible capital to assets, and it's hard to say, if you don't believe what people have been telling you, you probably shouldn't own it, because I don't think you can just analyze the financial statements to the point that you can say it doesn't matter whether or not they're being honest. And so you said not to just pick on financials, so I'll pick on a technology company. With Meta, I think to just kind of flesh this out, there's sometimes when you're invested in a business and there's maybe some aspect of it, you thought you had a good grip on and then all of a sudden it becomes kind of clear that maybe this is an issue or not, and then you have to research to understand whether or not you want to just trust management, or if there's a conclusion that can happen. And so to just put a little more specifics on that, I'm talking about Meta with all of a sudden app tracking transparency, where Apple rolled that out, and then now basically a lot of ad signal was lost. And the question at the time was whether or not Meta would ever be able to recover that ad signal to get back to where their return on ad spend was for advertisers prior. And I found myself at the time because I was in bed. invested in meta trying to understand ad tech. And you could take two avenues here. You could say I could trust Mark Zuckerberg and team will figure this out. But then you also don't know. Of course, they're in the prerogative that they're gonna tell you they have a good grip on it in a solution. - True. - And so you gotta kind of have an opinion on that. - Yeah, it would be a stretch to say we understood app tracking technology to the point that we could confidently say that they were going to be able to recover from it. We were somewhat dependent on management saying we have a path to do this. It's gonna take us, whatever it was, a year to two years to replicate the data internally. And that one didn't give us a problem. Where we got in trouble with the you have to just trust management was in that time period when you had referenced it earlier where they had a decline for like one of the first times ever decline in cash flow coming out of the business. We had been so focused on the cash flow that's being extracted from Facebook and Instagram is going to go into R&D and we're not going to be able to tell you what the returns on that are. But we trust management that they are taking an economic approach to that investment. At the time when Facebook stock got down to 100 or so and we were looking at should we be doubling or tripling our position in it, not just restoring what the market took away but making it significantly higher. We had concluded the most important thing for us to understand is that this massive new capital spending and R&D spending was economically justified. And at the time Facebook had a new CFO who wasn't really capable of explaining it in terms that to us suggested economics was driving it. It was kind of like, you know, this is what we do. So of course we're going to be spending this. And that kept us from making our Facebook investment one of our largest positions that we just didn't get to a point where we said, you know what, we can trust them that they are economically thinking about how capital is being deployed. You don't hold meta currently though, right? We do not know. So that's right. So we didn't make it as big a position as we should have when it was around 100. We held it into the 400s. And then it got to a point that we could no longer rely on our thesis that businesses like Instagram and Blue Facebook were worth more than the more, were worth at least a market multiple. At to justify prices higher than that, you had to believe a premium multiple on those businesses plus that all of the hyper scaling investments were going to eventually pay off. And we said, might happen, might not happen, but we don't have strong conviction on it any further. Yeah, and it's an interesting case study because I think going back to this idea of great businesses, sometimes they have these little air pockets where the earnings quite haven't caught up to maybe what the current valuation is. And I actually know kind of that moment because I ended up trimming a little bit myself at that time. But I then, you know, a quarter later all of a sudden, they started re-accelerating revenue growth. And then they did it again and again. And now they're saying they're going to grow revenues 30%. And as you were saying, you know, can you really show an ROI on the CAPEX? If you attribute that incremental revenue growth to that CAPEX spending, then all of a sudden, it looks like they actually can show an ROI on all of that spending. And so it kind of did change that thesis a little bit. Yeah, and you know, I can point to a different example where we had kind of a similar pattern with Netflix. And you know, people listening to this might think, you know, he's talking about Meda, Alphabet Netflix. He's not really a value investor. Our argument for owning Netflix was the market was valuing Netflix subs at about $200 a sub. At the same time, it was saying HBO subs were worth something on the order of $1,000. We thought that disconnect didn't make sense. And what would otherwise count as gap earnings was being disguised by very large customer acquisition costs. So that was the thesis for owning it. Netflix too went through a period where they had one or two quarters of negative subscriber growth. Market freaked out. The stock went down from like 200 to 100 something on that scale. And we went through the same process. We went out to California. We sat down with the CFO of Netflix. We talked about what the path to recovery was, how the password sharing was going to be incremental to revenue, how at the ad supported tier was going to be incremental to revenue. And we came back from that meeting, believing management was thinking extremely economically about the future prospects of the business. The stock looked very cheap on the basis of current subscribers. And we fully restored what the market had taken away in terms of position waiting. And that ended up being a very profitable move for our shareholders. And in fact, made more for us than what we didn't make in the meta case where we couldn't get comfortable that management was applying economic thought. And so during these kind of, I guess, air pockets, I'll call them, where there's a little bit of disconnect between the valuation and the current earnings power. Is that a scenario where it sounds like you're now willing to give management more benefited the doubt and really leaning to a blue sky scenario of how good could this actually work? Well, we want to make sure we're still asking about the how good could it be. But I think historically, one of the biggest flaws of value investors has been you look at an event and you say, the market dropped the price of the stock by 25% of my value estimate is only down by 10. So it's really more attractive today than it was yesterday. And we've looked at our own data to say, once the analyst starts lowering the fundamental value estimate, there tends to be an anchoring bias. And there's momentum to that declining value. So the knee jerk tendency to say the stock is overreacted is something we've had to teach ourselves not to follow and to say, this stock is now something that we need to take a fresh look at. And if we didn't own it today, do we want to buy it? I think a lot of times it's easy to freeze and say, I don't know enough to act. I don't know enough to buy it. I don't know enough to sell it. And we try to take the position to say, do I know enough today that I want to own it? And if I can't answer that affirmatively, then it doesn't belong in the portfolio. That makes sense. And I do think that's very common for people to not kind of update their a primary estimates. But is that true to the upside too, though, if something, let's say you expected it to only move up 10% and then it moves up 20%. Does that mean now that that's even a better opportunity or does that mean that, you know, I'm going to be wrong on something else in the future? And I don't like that. I didn't really understand, you know, where it was going with earnings. Again, analyzing our own data, our analysts have that anchoring bias in both directions. And we have increasingly focused on trying to make sure that we root out any potential conservatism in our value estimates before we sell the stock. You know, we don't want to succumb to the momentum idea that it doesn't really matter what the business is worth. You just keep owning it. But we do want to make sure that, you know, if an analyst recommended something saying, yeah, I'm not really sure if the value is 100 or 150, but the stock is at 60. So it's cheap either way that we don't sell it until we've, until we've fleshed out the value more to get comfortable with a tighter range. I did want to talk a little bit more about Netflix, which you were just mentioning because that was this maybe eight or 10 years ago. Now I remember I saw you on, I think it was CMBC, but I was really touched, I guess, by how quickly you kind of moved on from your prior thesis on Netflix where, you know, you kind of had, you were talking a little bit of kind of the regular value thesis, ah, it's 150 billion dollar market cap, like look at the earnings multiple, it's so high. And then you very quickly kind of pointed out, but then I was, you know, had an analyst that was able to point me to the actual math and show, you know, subs if they just raise, you know, our poo a little bit. This is what margins would actually look like. And it's not a really high multiple stock. It's actually really low multiple stock. Yeah, so what happened there? And I don't remember the year on this, but it's quite a while ago. An analyst put Netflix on our idealist to discuss at our weekly stock selection meetings. And I looked at it with an eye roll. And in the first paragraph of his report, he said that he had done an effectively anecdotal survey of friends, people around the office saying, which of your subscription services is the most valuable to you. And Netflix overwhelmingly won that battle over HBO, Sirius XM, Spotify, pretty much any of the other things that people were subscribing to. And at the time, I think the Netflix price was $10 and most of these other monthly subscription services were closer to 20. What the analyst said was, "If Netflix raised its price to match these other services, which customers don't value as highly, it would be selling at 14 times earnings." And then the rest of the report was why it made economic sense for Netflix to use price as a weapon to achieve scale that would be unmatched in video streaming and the eventual advantages it would give them over any of the other video subscribers. So we started thinking, "We've got to come up with a different way to think about business value here because gap accounting isn't doing justice to this." So one way to do it would be to say, "Okay, Netflix could charge $20 and what would the income statement look like?" And that way you'd see it's a lower p than the market. Another way to do it is to say, "Can't really come up with a reason why Netflix sub and an HBO sub should be worth markedly different numbers." And if you looked at earnings instead of how gap did it by saying, "Netflix adds 25 million subs a year," and the subs are probably worth about $800 a piece. That's, what is that, $20 billion of value they're adding. So the stock is selling at about six times the incremental value that is being created. Now, very different way of thinking about earnings than the typical value investor who's saying this is a barely profitable business. In fact, it generates a little cash flow, not even sure their debt is well supported. And our debt team used our equity analysis on Netflix to help get comfortable that the near junk yields on Netflix bonds were an incredible bargain. And we made that a significant holding in the Oakmark bond fund. One of the most important lessons for value investors out of all this is where gap accounting seems to deviate from where real value creation is, you have to be willing to make adjustments to that instead of taking pride that you're limiting your universe to low PE and low price to book stocks. And I think that's a great explanation. And I also think though that you deserve a lot of credit for being able to switch your opinion on that because I know that's an area where people will be still investing in the same low PE stocks their whole life that continue to decline because they're not able to make that transition that you were able to. Well, I think sometimes people get a misperception of what my role is here at Harris Oakmark. And while I've definitely had a strong hand in changing the way we think about valuing businesses and my imprint is on lots of things we've done, this is very much a team approach. It wasn't me saying Netflix looked attractive and asking an analyst to look at it. It wasn't me saying Netflix belongs on our approved list. It wasn't me saying Netflix is going into the portfolio. All those decisions are made on a team basis. And to say the team deserves a lot of credit for that, I wouldn't push back on you at all. Really nice pushing off some of a lot of humility there we'll say. But I wanted to ask a little bit more about kind of the portfolio analyst relationship. And you're talking about your role. So an analyst is the one who's doing all the research. And they're kind of getting you to kind of push back a little bit. But then it's ultimately you that is, I guess with the team, but going to make that decision whether or not that gets inclusion in the portfolio. So if you could kind of say a little bit more about that. So all of our analysts are generalists. We ask them to find stocks that they think are selling at less than 60 cents on the dollar. Were they comfortable that value is going to grow at least as fast as the average business and where management is aligned. They will write three or four page report that explains why they're comfortable that the stock meets our criteria. That gets distributed to all of our investment team on a Monday afternoon. We all spend Monday afternoon and evening trying to poke holes in it. And then in Tuesday morning, the analyst will present that to the entire group. And all of us are trying to shoot holes in the idea. The goal is if the analyst is wrong, we want to maximize the chance that we identify it in that meeting and that we never lose client money on it. At the end of a lengthy discussion that's often very heated, we have three of our senior investment partners vote on whether or not the stock is on the approved list. If they vote no, I can't buy it in any of my portfolio. None of our managers can buy it in any of their portfolios. And I don't sit on that team of three people. Once it's on the list, then it's got another hurdle. So like for example, in the Oakmark fund, the three of us who are portfolio managers will sit down, will talk about every name that's on the approved list that we don't own. And in some way, could it make our portfolio better? And better could mean bringing down the average multiple, bringing up the average growth rate, or not correlating much with the other stocks that we own so that it would reduce the risk of the portfolio. And then if two of the three of us vote that it belongs in the portfolio, it goes in. So for the past three years, we've had that structure a little more than three years. So I'm the minority vote on that committee. If the other two portfolio managers disagree with me, majority rules. So even if I don't like it, it will go into the portfolio. And how do you think about portfolio diversification? What are kind of your limits for an individual stock? So we like to think of the Oakmark fund as a fund that you'd be comfortable telling a friend. They could put most of their assets in that fund and then check back in five to ten years that they don't really have to make a habit of monitoring it closely. So in that context, we think about 50 to 60 names in the portfolio. So the average name is only 2%. And when something gets to double a normal position, that's time to start automatically trimming it. And again, that's in the context of portfolio risk being appropriate that it can be a very large holding for a typical individual. A new name, when we put it in the portfolio, we'll probably go in at about 1%. And over the ensuing months, as our confidence grows in that idea, we will add to it and bring it up to a normal waiting. We probably would stop adding to it at about three. And then it would be market appreciation that would take it up more than that. Do you see the benefit of having that many stocks in your portfolio as overweighing, let's say, if you were to rank them because you talked briefly about that. If you had to rank your 60 stocks and then just pick the top 30, you still think that the diversification benefit of having that other 30, so a total of 60, outweighs that maybe the fact that your best 30 ideas are going to do better. Well, we've seen that in the differential performance between Oakmark and Oakmark Select. Over a very long time period, Oakmark Select has outperformed Oakmark in terms of absolute returns. But it's done so at the cost of a higher standard deviation. And whether or not that standard deviation is appropriate for an investor often depends on what else they own. It's why we encourage somebody who's building a portfolio of portfolios to look at a product like Select where our stock selection is amplified in terms of the effect it has on the returns. And for somebody who's looking more for one-stop shopping, we would steer them toward the Oakmark Fund because those bottom 30 names, we think do enough to reduce the standard deviation that it's worth the modest reduction in portfolio returns. So the things we're thinking about for like the bottom 30 are how much of our top 30 are banks. Are we getting too much single industry exposure? Once we have four or five banks in the portfolio, if the return on a bank looks the same as the return on a food stock, we'd add the food stock just because we'd rather get the portfolio diversification. But we're not averse to having portfolios that look nothing like either the S&P or the Russell value, the two indices most of our shareholders compare us to because we think they are hiring us to look markedly different than the stock market averages do when we think there's a risk return benefit to doing so. So we don't focus on tracking error. is our primary risk, we focus on draw down risk as the risk we're trying to mitigate. That makes perfect sense and you have the perfect example to kind of back up what you were saying there. Just kind of as we wrap up, I want to do talk a little bit about a couple of single stocks. Can you maybe say a little bit about Salesforce? I know right now everyone's worried that AI is going to kill SAS and so maybe just kind of a few high level thoughts on that one. So a few of the things we think about there that maybe aren't in a lot of the research reports you read or certainly the popular press. So one of the arguments is AI is going to write your software and you're going to tear out your enterprise software and put something that Claude generated in there instead. We think we are many years away from compliance departments being comfortable with that. That one of the big benefits that a company gets from using product like Salesforce is the long history of it being it being tested. Being tested for how easily can someone hack into it for potential errors. And there's a big value to that to corporations. There's a big potential cost to tearing out an existing enterprise software plan. So that's part of what gives us comfort. Another part is the argument that AI is going to reduce back office seats to maybe half of what they are today and Salesforce prices on a per seat metric. We think they price on that metric out of convenience today. But they're really pricing based on usage but that tracks closely enough with seats that you don't have to worry about measuring. If there is a meaningful change in seats and not a meaningful change in the amount Salesforce is being used inside a company, we think they'll change the way they price. You know, the pricing has to tie to the economic value it's adding to the enterprise and that's kind of regardless of how many seats it's being used in. And then lastly is valuation. Five years ago, you had to believe Salesforce was a significantly better business than a bank to justify owning it because it was priced at twice the multiple. Where we've gotten to now on PE even, free cash flow yield, Salesforce isn't priced much differently than high quality banks. So we look at that and say there's a much lower hurdle that Salesforce needs to pass to give an average or higher average return. And the nice thing is it doesn't correlate very highly with the industries that we're overweight in. So we get comfortable that we're putting in a higher growth business without increasing our average PE and significantly decreasing the correlation risk of the names in our portfolio. So doesn't fully answer your question, we're not 100% sure that AI is a nothing for them or potentially a positive. But those are the things that lean us in the direction of saying that the current price makes it worth taking on some of these risks today. Yeah, and that makes makes a lot of sense. I did a deep dive into Salesforce and kind of those issues you hit on also kind of similar where I shook out, it was a little more concerned about incumbent software competition and it being easier to kind of get into someone else's lane and potentially with everyone trying to get this like AI platform layer that that might make it easier to get into someone else's cross vertical. But you know, no investment as a risk. That couldn't easily happen. Yeah. And that could be an opportunity for Salesforce. It could. Yeah. Yeah, investing is hard. I did want to ask though, I always had this question on software companies where they have this great model where you're receiving cash upfront. And because of that, it shows as a positive on the cash flow statement from all this deferred revenue. But what do you think about in terms of putting a multiple on that? Because on the one hand, you could say it is real cash they're collecting. But as their growth slows in the future, then that sort of deferred revenue element as a source of cash flow is going to disappear. So if you're putting a multiple on it, you're implying it's there forever. It's not going to be there forever because businesses don't grow forever. So it's just wondering if you have any thoughts on that. I guess we try to look at what's the recurring cash flow and put a multiple on that rather than necessarily giving a deferred revenue a big multiple. We like the SaaS model because it helps do the work that we had to do previously of saying if somebody's buying this new every five years, what are they really thinking the incremental value is? And you're creating a recurring stream of revenue that's easier to value. We also think the longevity of that stream has been improved as these software companies have grown in importance to the enterprises they're in and the cost of tearing them out grows. So I don't think we worry too much about how big the deferred revenue asset is and we're trying to put the value on what we think annual cash flows are. And we think a software company does deserve a higher average multiple. It's capital-light, so ROIC is high, strong recurring revenue nature. It's not like you start each year at zero and it's all fully dependent on new sales. Growth has been significant in the industry. We think that can continue as you keep adding bells and whistles each year that make the product do more for the customer. So our argument would be that whatever you decide that recurring cash flow number is, the terminal multiple on that deserves to be meaningfully higher than where the stocks are today and more in the range of what the S&P 500 is and broadly across the software industry today, the companies are available at the lowest multiples of recurring cash flow that they've ever sold at. When you're talking about that terminal value multiple you're putting on it, are you still comfortable though not fully answering that AI question? I think one of the fears with the software is that if this doesn't happen now, it will happen in 15 years or something from now whatever it is. So our method which isn't perfect of trying to deal with that risk is as we've become, as the potential range of outcomes has grown with AI, and again I think it's important to think of both positive and negative impacts. You've said it's inappropriate to use the lower risk category on this that we had previously applied and so we've added something like 150 basis points to our discount rate because that 15 year out future just isn't as sure as we used to think it was. Just as we kind of come to a close, if we could fit one last little one in on Airbnb if that's all right. Sure. Why do you own Airbnb instead of booking when booking has a lower cash flow multiple and is growing a little bit more? We've got two analysts here that would love to hear that question because that's openly debated here and we think both names are interesting. We've landed on Airbnb largely because of the amount of spending that they are doing in areas that they are not currently profitable where they think they've got a right to win and that that spending should have a good long term return on investment. So we take that spending, add it back to the current earnings that they're showing and when you do that it does look like Airbnb is cheaper than booking. We get comfortable that the take rate on Airbnb in terms of the value that's provided given they're generally dealing with small buyers and small sellers ought to be at least as high as bookings take rate and it's currently lower. So when we make those two adjustments Airbnb comes out looking a little bit cheaper but we are interested in both companies and their points in time before where we've own booking instead of Airbnb and we've kind of vacillated back and forth between them. We think they're both really good businesses. Do you have to take that opinion though that Brian Chesky getting into experiences isn't the same as Mark Zuckerberg spending in the metaverse because you're adding that spending back? I think it's easier for us to believe Airbnb will be successful in experiences. It seems like a natural add on to their existing business and the way Airbnb itself talks about that spending. It implies that they are focused on eventual ROI. Our issue with meta when it was metaverse spending and at that point in time metaverse to us sounded a lot like overgrown teenagers and their parents basement living in a fake world instead of the real world and the returns were so far out in the future that was tough for us to get our arms around. I think with experiences I would bet within three years if those businesses if they haven't been able to show that they can be profitable in those businesses they will redirect that spending. Thank you. I really enjoyed this interview. I thought that your answers were fantastic and I know our listeners will have gotten a lot from it, a lot of great insights. Where should people go if they want to learn more about Oakmark or anything else about you? I think the best spot to go is to our Oakmark.com website. You find a lot of information on all of our funds there. We spend a lot of time on quarterly reports trying to make them both educational and entertaining, which again I think separates us from most of the mutual fund industry. You can find those there and hopefully in the very near term you'll also find a link to this podcast at Oakmark.com. Thanks for having me Drew. I really enjoyed the provocative questions and for you giving me the freedom to give open and to dancers this was a lot of fun. Thank you and I highly recommend that everyone read anything and everything Bill Niger and writes and thank you guys for listening and until next time.