We had the great pleasure of speaking with Professor Bruce N. Greenwald, Founding Director of the Heilbrunn Center for Graham and Dodd Investing at Columbia Business School, about the outlook for value-oriented investing and the newly published second edition of his investment classic, Value Investing: From Graham to Buffett and Beyond.

The second edition comes two decades after the first, with each edition coinciding almost perfectly with a market boom in Internet stocks. Whether the publication of the second edition is a contrarian signal remains to be seen.

Professor Greenwald’s comments in this conversation may surprise some listeners, as his thoughts on the evolution of value investing emphasize how it needs to change in order to remain relevant and successful in the future.

The following transcript has been edited for space and clarity.

John Mihaljevic: It’s a pleasure and honor for me to welcome to this conversation Professor Bruce Greenwald, the Robert Heilbrunn Professor Emeritus of Finance and Asset Management at Columbia Business School, and the Academic Director of the Heilbrunn Center for Graham & Dodd Investing. Professor Greenwald has been recognized for his teaching abilities, and I’ve heard it firsthand from many of his students as well. He has received numerous awards including the Columbia University Presidential Teaching Award. All of us in the value investing community look to you, Professor Greenwald, as somebody who has guided us for many years in our value investing endeavors. You’ve been a steadfast guide and advisor whose wisdom we all appreciate. Welcome.

Professor Bruce Greenwald: Thank you.

Mihaljevic: Also, you have blessed us with another gift, which is the second edition of your best-selling book on value investing; it’s been quite some years since the first edition. How many years exactly?

Greenwald: It’s 20. My timing is perfect: The first edition came out in 1999 when everybody said value investing was dead. Then the second edition has just come out when everybody again says value investing is dead. It’s a 20-year cycle I’ve managed to hit at the peak both times.

Mihaljevic: It might not be optimal for book sales, but it’s optimal for making all of us money going forward. We are probably at an unprecedented point in time in terms of the disparity between growth and value.

Greenwald: That’s right. But I plan to talk today about how this situation is different from 1999. At that time the internet boom had driven everybody crazy, and the valuations were completely devoid of any rationality. For several years they produced enormous returns for non-value investors. This time fundamental changes in the economy will not go away; they are not temporarily crazy. These are not things value investing must adapt to. In that sense, the timing of this book is a little better than the timing of the last one.

The first thing to recognize is that if you look at the statistical evidence – if you look at reasonably well-developed models of value versus growth, you’ll see the value premium is still there but not obvious. Second, if you do the crudest definitions, you could call cheap stocks value stocks then you call the expensive stocks growth stocks – by this crude measure, value has underperformed for as much as 10 years. You have a lot of hitherto successful value investors who have underperformed for 10 to 15 years, so there is a long-term set of issues here about value. When you talk though about value, it’s important to be clear about what it is so you’re not throwing the baby out with the bathwater. Four underlying principles constitute a value approach, and there’s a follow-on corollary of those principles. We probably want to start there and then talk about the corollary.

The first basic value principle, principle zero, is the recognition – you see this in Graham and Dodd, and you see this in all good investors: Investing is a zero-sum game. The average return to all investors must be the average return to all assets, which is essentially what the market return is. That principle applies broadly, and it applies to every asset class because everybody owns everything. The derivatives – the uncovered shorts, the warrants, the options, and others – all of them have a buyer and seller, so they just net out. The first thing you must always ask if you’re a value investor – or any investor – is why you will be the one who outperforms. After all, on average people perform at the market level. The way to think about it is every time you buy a security, you think it’s a good idea because it will do better than others’ security choices or investment choices. At the same time, somebody else is always selling you that security because they think it will do worse than other investment choices. One of you is always wrong! You want to ask yourself, why, as a value investor, you will be, more often than not, on the right side of that trade.

First, the traditional answer has been that value investors look for opportunities to take advantage of deep-seated behavioral irrationalities. We know people overpay for lottery tickets, that they buy them with great abandon, and that they’re crappy investments. Nobody has ever gone broke arranging a lottery. Glamor stocks, the get-rich-quick stocks, will almost always be overvalued to some degree. Second, people don’t like to spend a lot of time thinking about ugly, disappointing, potentially loss-making, and therefore cheap stocks. Focusing on the latter, which is the value approach, will still have some value to it, but statistically that value is diminished. You don’t want to throw out this principle. You want to think intelligently about it.

The second value principle is to know what you’re buying. That’s why it’s called value investing; you must be able to identify the value you’re paying for. Historically, that has been about Graham and Dodd approaches to putting the value on securities. Earnings power, asset value, and triangulating the two are superior because, among other things, they do look carefully at balance sheets when DCF projection-based values don’t. Yet they are and have been better than traditional valuation approaches. You want to have a better approach to deciding what you buy.

The third principle after you know what you’re buying is to be patient. You must be willing to wait to see that there’s a margin of safety, a significant gap between the price you paid and the value you bought. Patience and discipline, in that sense, haven’t gone away.

The problem is that because value investors concentrated on knowing what they were buying, they weren’t willing to buy the future. Almost by default, it became the case that a value investor would not pay for growth. It turns out, if you think about the history of investing – and the history is relatively short in terms of the broad number of investors we think about – investing in growth was, for the most part, not a particular advantage. In the wake of the Depression, in the wake of the Second World War, nobody wanted to take any chances. Not being willing to pay for growth just puts you in the same bucket as everybody else. Then starting in the 1950s, the late 1940s with the Nifty Fifty, you’ve got investors who looked to buy growth companies and looked to buy just the great American corporations like General Motors, General Electric, DuPont, U.S. Steel, RCA, and so on. They paid for the growth. That probably didn’t work because from the 1950s to the late 1980s, growth produced disappointing returns. Returns disappointed because the dominant economic force over that period was globalization. Globalization has you go from a national market where, say, General Motors is the dominant competitor with economies of scale and significant competitive advantages that create barriers to entry, to big global markets nobody dominates and where there are no barriers to entry.

The fundamental fact about growth that seems to be increasingly appreciated is when you don’t have a market protected by barriers to entry. It’s a market where you have, as an incumbent, a competitive advantage, but you’re not going to make money from growth. That’s clear in the case of organic growth and revenue because with revenue growth you typically do better in the short run. But ultimately, other people will see those profits. You’ll get a lot more entry. You’ll get expansion by existing firms, and the profit opportunity will go away. The same thing applies to margin growth where changes in technology reduce costs. If there are no barriers to entry, your profits go up temporarily. Everybody sees it, and they crowd into that market. Again, the benefit of the growth just dissipates. When you think of active growth where you’re investing, if you’re investing in a competitive market, you won’t earn 15% because everybody else will see that opportunity. They’ll take advantage of it, and they’ll drive that 15% return down to a 10% return. Even in active investing, if you don’t have a market protected from competition by barriers to entry, you’ll earn 10% on the investment; you’ll have to pay 10%, if that’s the cost of capital, to the people who provided you with the funds and the net benefit will be zero. Again, growth has no value.

That was the trend; it changed from markets that could be dominated and were protected by, in Buffett terms, moats, to globally competitive markets. Then over time, the value of growth was undermined and the value of growth, therefore, was disappointing. If you didn’t buy growth, you tended to do well.

That trend has reversed. As we move from manufactured goods to services – manufacturing is dying these days the way agriculture died in the early part of the 20th century – you go from big global markets to local service markets where goods are locally produced and consumed. Those local geographic markets are small enough that they can be dominated by local competitors like Walmart. Once Walmart dominates the market and has the benefit of economies of scale and customer captivity to protect the market share that underlies those economies, Walmart will start to earn above the cost of capital. If it expands widely by, say, adding food within the area where it has economies of scale, it will get returns above the cost of capital. If it expands widely to adjacent markets, which is what Walmart did, it will make returns above the cost of capital. If costs go down, entrants simply can’t get the necessary scale in that relatively small market to compete with Walmart, which dominates. You won’t see entrants competing away the benefits of growth in revenue or the benefits of growth in margins as costs go down.

Then, suddenly, as you go from big global markets to small local markets – essentially service-based markets – and this applies to manufacturing firms too. If you look at a company like Deere, the part of the business that makes tractors and sells them in a big, globally competitive product market, has shrunk. What Deere really does is provide local service support. It provides local software that runs those machines off GPS systems and places every seed in the ground individually at optimal times. You have local secondhand markets because the machines last for a long time. Because it can charge a lot for these extremely valuable machines, Deere finances the sale, which is based on local information about lending risks. When Deere dominates the United States or regions within the United States and has 90% of the agricultural equipment in that sector, there’s just no chance anybody will buy a Kubota or an Agco tractor. This principle applies equally to manufactured goods as local services become a bigger part of the package.

You see this in the share of profits in national income. At the end of the 1980s, profits were about 8% to 8.5% of national income. Today, that number is about 13.5% because globalization is essentially being reversed and technology has reinforced this trend. If you think about the old technology companies like IBM, they did everything. They did the chips. They did the software. They did the hardware. They did the peripherals. If you look at the technology companies today, they’re highly specialized. Oracle does only databases. Microsoft does operating systems and adjacent software. Intel does only CPU chips. Oracle does only fonts and graphic material. Google does only search. Once you get disaggregated markets like that and technology has enabled people to put systems together from many different suppliers, those tend to be small niche markets. In a sense, they’re local markets and product geography, and you get dominant competitors in those markets who can keep other people out because they have the scale economies, and they have the customer captivity to deny those customers and deny the necessary scale to entrants.

We live in a world, as those trends continue, where growth turns out because franchises are getting stronger and more valuable than people thought. If you’re not looking at growth – and also, if you’re not looking at intangible assets because that has also changed with these technologies and the trend toward services – you won’t do a good job of buying where the real opportunities are. That is the fundamental change value investors must come to terms with. They must be much better at valuation and intelligently buying growth. They can’t just do what Ben Graham and David Dodd did and say, “Look, we just don’t do growth.”

Mihaljevic: It seems, on the flip side, there’s a lot of value destruction due to the obsolescence of older technologies.

Greenwald: That’s right. Here’s the thing: If you are in a competitive market and you go out of business because you’re earning the cost of capital on your investment and there are no economies of scale – as you go out of business, the operation doesn’t get less profitable, and you typically recover capital to cover the lost profits. That’s because you have, say, $100 million invested, you’re making $10 million a year, that $10 million goes away. But you get your $100 million back because it’s working capital and fixed capital that you just allow to depreciate. Ultimately, you make something close to your $100 million max. In a world of competitive businesses, lack of growth, that is, negative growth – these disruptions don’t kill you. In a world of franchises like newspapers, the destruction of franchises, negative growth is deadly because now your fixed cost is spread over a smaller and smaller base. Your product gets less attractive because the network effects go away or go to some other sector. Because you’re earning 40%, 50%, 60% on the invested capital, the capital you recover comes nowhere near to compensating for the lost business.

You’re right. The second half of this is if you look at dying businesses in a franchise world, you’ll get murdered. That probably hurt a lot of value investors who thought they would get rich buying newspapers and similar businesses.

Mihaljevic: Now we see that in a lot of sectors. An example is energy, which might be down to its lowest in history as a percentage of the S&P. How can we think about looking for value in a sector like that which is gradually getting disrupted; it’s a sector that will shrink in certain ways, and yet it seems investors might have overreacted on the negative side.

Greenwald: Here you must be careful. Those are competitive businesses. You’re talking about commodity businesses. The history of commodity businesses – and this includes energy – is that in the race between improvements in technology and the exhaustion of readily accessible resources (including the atmosphere), improvements in technology have won hands down for 200 years. Even though oil did well for a period starting in roughly 2003, historically it’s done badly. There’s a brief period in the 1970s and early 1980s when it did well, then it did badly for 20 years. If you’re going to buy natural resources intelligently, you must be a strict value investor and not look for growth at all. You want to make sure the asset value is there, make sure the earnings power value is there, and you want to make sure there is a margin of safety.

In commodity businesses, your advantage as a value investor is the second of two forces we talked about. When oil does badly it’s disappointing, it’s no longer glamorous, and we know people overreact. You want to look at that possibility. You want to put a value on what you’re buying that’s better than you can do with one of these speculative DCF calculations based on unreliable projections of future prices. That’s basically doing an asset value, doing a current earnings power value, and triangulating between the two, which is the Graham and Dodd approach, which is better. You must have an appropriate approach to those investments. Third, you must be patient. Once you’ve got a value, you might find that even though it’s a bad sector, it’s still not a bargain. You want to be in that area as a strict value investor.

There’s something else that’s come into play as you start to pay for growth and as you start to think about these changes. When you buy growth, when you look at the possibility that shrinkage will kill you, you’re focusing on the future. Focusing on the future – and that’s also what you’ll do increasingly in this era of disruption and commodity markets – is a hard thing to do. That means – this is the last piece of advice I give people – if you’re going to invest in oil, you better be an oil specialist. You can’t be a generalist. Imagine that I do South Texas Gulf. I trade or invest in South Texas Gulf Coast onshore oil leases. It’s all I do, and I’ve done that for the last 20 years. Imagine you fly in from New York or Germany and buy an oil lease from me. Who do you think will make money in that transaction? It’ll be the expert.

You must specialize by industry, and that is a big change for a lot of value investors. You won’t be able to do all commodities. You might be able to handle two or three – and by geography. The oil companies that do well are not the ones that go all over the world. They tend to be the oil companies with basin specialties. They dominate a small number of basins; they invest in those basins where they have informational and knowledge advantages. You want to have a sense of looking for the disaster. You want to have a good value-based approach to knowing what you’re buying, which is a good valuation technology. You want to be disciplined about it. To do that well, you must be an expert, and that’s a big change for value investors.

Mihaljevic: To continue that idea of specialization, one of the four principles you mentioned is to know what you are buying or the concept of “circle of competence.” Would you say that also applies to investors who are now looking to find value in growth? Because you need to assess growth more, that you need to be…

Greenwald: …Absolutely. Let’s talk a little about how you value growth because that’ll give you a feel for how hard it is and how important it will be to know that business and be specialized. If you think of a growing firm and buying a growing firm, most of the value is way out there in the future. Typically, when you do a DCF on a growth stock, you’ll do five years of cash flow projections, and then you’ll do a terminal value. Somewhere above 80% of all the value will be in the terminal value. The terminal value will be a terminal cash flow times a multiple, and the multiple is one over the difference between the growth rate in that terminal cash flow and the cost of capital. If the growth rate is four and the cost of capital is eight, 4% is the difference. The multiple will be one over four percent or 25%. But suppose you’re off by 1% in either of those numbers. Suppose the growth rate is not 4%, it’s 3% instead. The cost of capital is not 8%, it’s 9% minus 3%, 6%. One over six percent is 16x, not 25x. On the flip side of that, suppose the growth rate is 5% and the cost of capital is 7% and these are 1% errors in projecting a long-term future. Seven minus five is 2%. One over two percent is a 50x multiple. Within a narrow range of forecasted growth rates and what future risks might look like, you can get a 3:1 variation in multiples. You had better be an expert if you play that game.

Also, you cannot put a number on things. If you’re going to buy growth, you must not say, “This is what it’s worth,” because you’ll get those inaccuracy problems I just described. What you must say is, “If I buy at this price, what kind of return will I earn?” You must look at these growth stocks and returns base. What you’ll do is an earnings power. The assets probably won’t matter because it’ll be a franchise business. You won’t invest in any growth if you don’t have confidence in the existence of a franchise or the barriers to entry. You’ll verify that they dominate the market, they have customer captivity or proprietary technology, and that there’s real market share stability. For an entrant to be viable, it must get to, say, a 20% market share. Two-tenths of one percent changes hands of market share every year, so it will take them 100 years to get here, which is a powerful moat. Then you’ll say, okay, as that franchise sits there today, what are the earnings? That’ll be a traditional earnings power calculation. You’ll divide that number by what you’re paying for the company. That’s a cash return, and this is an enterprise value. You say you think you’ll make $100 million and you’re paying $1.2 billion. It’s an 8% earnings return.

You don’t get all of that. You can’t just stop there because the management will give you some of that in cash, and they’ll reinvest some of it. You must have a feel for how much of that you’ll get in cash, which is what the distribution policy looks like now and in the future. Let’s say you think they’ll distribute half of the 8% earnings with growth. You’ll get a 4% cash return. Everything else must come from growth, and the growth will have two sources. One source – and here it’s important to recognize it’s just organic growth because the market is protected by barriers to entry – organic growth in demand and margins from cost reduction will benefit you. But you must know what those numbers will look like. You must know and be able to forecast what these long-run growth rates and revenue will be. Typically, because you’re talking about long-run revenue growth, it won’t be more than 3% above GDP growth. If it’s temporary growing at 20% faster than GDP, that won’t last. You won’t be able to tell when that stops, and that’ll be too tough to call.

You must have a long-term revenue growth rate, and you must have a long-term trend in cost reduction to improve margins. To understand both of those, you better understand industry demand and how it’s evolving; you better understand industry technology and how it’s likely to evolve. Then you’ve got the 4% return they reinvested. What you get for that depends on how good the management is and will be at capital allocation. If they’re terrible at capital allocation – and there are companies out there that typically earn about $0.20 on the dollar for everything they retain; understanding that number requires real industry knowledge – then that 4% will be worth less than 1%. If they have a disciplined strategy – they spend money first on cost reduction and, when they grow, they grow either at the margins of their existing markets where their economies of scale carry over, or like adding food products for Walmart within the markets they dominate – you could have a case where every dollar reinvested earns $2, in which case that 4% would be increasing. But you better understand the evolution of that management, how good they are at capital allocation, and that’s a lot of detail.

Once you have that number – it consists of the cash return, the organic growth return, and the active reinvestment return – you must compare that to the cost of capital because you want a margin of safety. If the cost of capital is seven percent, which is a reasonable cost these days, you want returns of at least 10% or 11%.

Here I come back to the good point you made earlier: These franchises don’t last forever. If this franchise lasts 50 years – and it’s a random event when it does – and it is dying at about 1.5% a year, which is 72 divided by 50. If it’s a 30-year franchise, that franchise has an expected death rate of 2.4%, and your margin of safety must cover that. The ability to make good judgments about how long these franchises will last is industry-specific. Again, you better be a real industry expert.

In every dimension, if you start buying growth and you think you’ll do it as a generalist, then you’ll get in real trouble. If you look at the value investors who’ve done well – these are the ones we’ve included in the book in this edition as opposed to the last edition – they’re all specialized. They’re usually specialized by industry and geography, or they do specialized kinds of securities and kinds of investment.

Mihaljevic: Would you say to be successful these days, we are not just security analysts looking at the numbers, but we also need to be judges of people or management teams more and more? Buffett is famous for saying you should buy a business that’s so good, even a fool could run it. That doesn’t seem to apply anymore in the world where so much of the value depends on the future. You think of an Amazon – if it had stayed a bookstore, it would have had a different outcome than what Jeff Bezos has done. How important has assessing management become?

Greenwald: First, you must understand the industry. Buffett also said when a management with a good reputation meets an industry with a bad reputation, it’s invariably the reputation of the industry that survives. You better understand these trends in detail. Value investors make a common mistake about industries. One of the things I cannot beat out of my students is they say, “Oh, this is a powerful franchise, so I’ll pay a higher multiple for it.” Powerful franchises are all in the earnings, presumably. If you give it a higher multiple, you’re double counting. What you care about is the durability of the franchise, which is the point you raised early in the context of disruption. Also, you care about the rate of growth of that franchise, which depends on management’s ability to make those earnings from the franchise grow. You must start with detailed industry knowledge. Then yes, of course, you must be able to judge management.

But here’s the thing. Knowing about good management without knowing about the industry economics is a tough thing to do. Knowing about good management if you can’t talk to them frequently so that they are a long way away or the managements you’re monitoring are all over the globe is a tough thing to do. I agree with you that you must be good at judging management as well as judging the industry. Jeff Bezos got lucky in some of that because other people did try to grow the way he did, and it didn’t work out. To do that well, you must be locally focused so you can talk to these managements and probably focus on managements who are below the radar otherwise. Then based on your knowledge of the industry, you can judge when they start making stupid decisions.

Mihaljevic: You talked a little about profit margins and how they’ve evolved. Profit margins used to be viewed as one of the most mean-reverting series there was. But that sentiment seems to have changed in the last decade or two. Would you say profit margins can stay “high” forever given that specialization of various businesses? If so, does that mean we have become, in a way, less capitalistic because maybe the businesses have gotten smarter about carving out niches versus competing on a commodity basis?

Greenwald: A good point underlies that question, and it is this: Mean reversion is a characteristic of the timeframe you’re talking about. If you’re talking about business cycle timeframes – periods of five to seven years – then yes, you’ll always get mean reversion. If you got high profits for a short period, they’ll mean-revert over four to six years. If you have low profit, they’ll mean-revert over five or six years. That hasn’t changed, but here’s the thing: There are longer timeframes, and in those long timeframes you typically don’t get mean reversion in profits. If you look from 1946 to 1980, returns on capital and, therefore, profit margins fell consistently. Over that period, they fell by a factor of two. It’s like people forget that over 200 years, commodity prices have continued to fall because they don’t look at those long timeframes. If you’re going to invest in growth because so much of the value is out in the future, you’ll be looking at long-term timeframes. There, you don’t see mean reversion except in low frequencies.

Once you look at the late 1980s and you look at what’s happened since then, first, the nature of the great companies has changed. It’s the UnitedHealthcares now. It’s the Deeres that understand the local component of manufacturing. It’s a service business like Amazon. It’s a service business like Walmart, although Amazon is disrupting it. It’s a non-traditional manufacturer like Google or Microsoft. It’s Apple, where the manufacturing costs are negligible. When you look at these kinds of companies, they dominate niches, small product markets, and they dominate geographies. Google is nowhere in China. Baidu dominates China. The trends we’re looking at that have empowered or created unusual value to growth and have led to this growing share of profitability are long-term trends.

Do I think the trend toward services is going away? I don’t see that. I don’t know what will replace services. But right now, that’s the game, and it’s the game all over the world. The countries that are still manufacturing economies like Japan and like a lot of the Europeans have gotten in real trouble because they haven’t adapted. That’s a trend I don’t see going away, and it’s a trend that empowers companies and empowers smart managements. This is especially true with geography, to stay out of each other’s way. One of the great things happening in agricultural equipment is that these companies are defining their own geography. Deere is dominant in North and South America and now, Australia. Agco sticks to Europe pretty much, and Kubota sticks to wet rice agricultural in Asia. Once that happens – you’re right – competition goes away, so you’ve got dominant competitors in each of these several markets. It’s just not worth it. It’s just not economical to go after each other. That geographic trend won’t go away. Ask yourself this: When you look at social media, what’s the trend? It’s getting increasingly specialized. It’s LinkedIn, it’s Instagram, it’s Tik-Tok that do only specialized things. That trend toward specialization in technology, I don’t see going away either. You must understand these trends and that someday they will change. The only way you’ll understand how they impact your investments is to understand how they impact the industries in which you invest, and that means being an industry expert and how they impact the geographies in which you operate.

Mihaljevic: To ask you a counterpoint, I’ve heard many people say because investors have become so specialized, now there’s value in being a generalist and making connections that the specialists cannot make. What would you say to that?

Greenwald: The empirical evidence is against them. For example, and they’ve known this for a long time, there are these big fund companies that offer industry-specialized funds. If you built a portfolio – it could just be an S&P-type valuated portfolio – of those specialized funds, they typically outperform the generalized funds from the same companies. If you look at the value investors whose performances have not deteriorated, they tend to be the ones who are specialized in this environment. Let’s just go back to the fundamental test. Suppose I’m an expert in agricultural equipment and I invest in agricultural equipment stocks. Suppose you buy your stock from me. You’re buying and I’m selling. Who do you think will be on the right side of that trade? Sometimes I will have missed a broader trend, but this is something value investors tend to do, and it’s tended to benefit them. They tend to look for stability. If an industry is changing rapidly, they put it in the too-tough-to-call column. They look for temporary problems that will go away in a stable economic environment. The things you’re talking about that would benefit a generalist will probably be dramatic changes from outside an industry. They’ll be rare, and they’ll be hard to forecast. The impact of those changes on the industry itself will be hard to forecast. On average, you’re much better off understanding the current stable industry structure, especially when it comes to trying to forecast it over 10, 15, or 20 years. Then you’ll come to that industry with some brilliant insight and do it. If you understand Sweden, you won’t be somebody who does well in a dense economy like China’s. Sweden is a bunch of small markets broken up by mountain chains. China is these dense markets along river valleys on the coast, like New York, and it’s just a different economics. I still say you must start by being a specialist.

If you’re smart, you can do 8 or 10 industries. Warren Buffett has done four, which are consumer non-durables, old media, insurance, and banking. When you look at him in other areas, he hasn’t done nearly as well as in those four areas. Evidence and logic still argue in favor of specialization. But, and here’s the big problem, when you define risk properly as (not upward variation but) downward variation, which is loss, and (not temporary fluctuations or variations but) long-term permanent impairment of capital, it turns out most of the permanent impairment of capital events are unsystematic. It’s like the newspapers going away. They’re not necessarily economy-wide events. That means diversification is important, and it’s hard for specialists to be diversified. There is this other function out there, and it’s not an investment analyst function. It’s a portfolio management function where there must be people who are good at speaking with the industry specialists and eliciting how much confidence they have in their choices, how reliable their projected returns are, and how correlated they are with other returns so you can sensibly build a portfolio. The role of a generalist has not completely gone away, but it has changed a lot.

Mihaljevic: I have a question about being an investor given how broad a swath of investors you see and have followed for a long time. Some value investors have done well through the decades. Some seem to have not adapted to the times, if you will. What is your advice for a value investor about evolving as an investor? There’s this notion of style drift, and that’s a bad thing. It seems a lot of investors have stuck to their guns because they don’t want to be accused of drifting. But how does one evolve in the right way?

Greenwald: The first question concerns what it means to stick to your guns. If you were an asset investor, for example – you are Marty Whitman or people in that game who focused on real assets – real assets have gone away. Assets are overwhelmingly intangibles these days. The old metrics you applied didn’t apply to a lot of businesses. Often, the businesses you had to look at where there were bargains had fundamental problems because you didn’t have a broad choice. If you didn’t adapt, all those old-fashioned asset-based investors got in real trouble.

Second, when you say style drift, you must have a fairly catholic view of your style. When you say don’t drift from a value style, increasingly today that means don’t drift outside your specialty because that’s what will put you on the right side of the train. That’s something Graham and Dodd and Buffett are acutely aware of. Stick to your specialties. The second thing is to embrace a value orientation in looking for stocks within an industry. If you have two or three industry specialties, concentrate on the one that’s the sickest. Those fundamental human tendencies haven’t gone away. Sensible search strategies are where you’ll always start. Once you’ve evolved the sensible specialized value-oriented strategy, don’t go away from that.

Remember that value investors should have superior valuation approaches, and the one thing we know about valuation approaches is one size doesn’t fit all. People get taught in business schools that everything should be a DCF; that’s idiotic. For example, if you’re doing event-based investing, which is an inherently short horizon, and if it’s fixed income so the payoffs are quite clear, you can go ahead and do a DCF because you’re not going far out into the future; the payoffs will be clear, and it will be all about probability weighting of well-defined scenarios. For that, a DCF is a good way to go. If you’re going to invest in competitive businesses like real estate, raw materials, natural resources, agricultural products, and businesses like steel fabrication, you know growth will have no value. If you do a stupid DCF forecast that is incredibly sensitive to the assumptions, it won’t be a good valuation. You’ll want to look for asset values, valuing the intangibles in a sensible way, as well as the tangible assets. You’ll want to look for earnings power values, and you’ll want to triangulate those.

You won’t have a simple multiple you can apply because it’s not that simple. You’ll have to apply this approach, which is the Graham and Dodd approach, which is much superior to a DCF approach, where it’s appropriate. If you’re going to invest in franchises, which will be increasingly a bigger part of the market, you’ll need to do this return-based calculation. When you do the return-based calculation, you must consider things like franchise fees in your margin of safety, and it’s a different way to proceed. In that case, you must accept the fact that making the sale decision will be brutally hard because you won’t have a value you can compare to the market price; these returns are relatively insensitive to market prices. You must have a policy. Buffett’s policy was never to sell. Seth Klarman’s policy originally was never to pay more than 16x earnings. It’s drifted up in this environment, but it’s still – it’s an arbitrary policy, and you must stick with it. You must stick with your valuation discipline and using these approaches as frameworks to do good research and identify exactly what you’re doing. If you drift away from that and start doing multiples or start flashing on things, you’ll get in real trouble. It comes down to sticking with the sound valuation approaches and sound research approaches; these will need to be specialized.

Then, you must be disciplined. You must monitor your behavior. You’ll make a mistake once, and it’s probably forgivable. But if you make the same mistakes repeatedly, you are not keeping track of your own record; you must be careful to monitor your own discipline.

It starts out the way we talk about things, which is having a sensible search strategy where you find the opportunities where you’re likely to be on the right side of the train – sticking with that, having valuation approaches that are both appropriate to each of the opportunities you look at and are applied consistently and rigorously, and having a disciplined approach to a buy decision and having some appropriate disciplined approach to the sell decision. If you don’t drift away from that – there are value dimensions to all of that that we’ve talked about – you will continue to do well. The people I know who’ve done well in this environment – there are people who, over the last 15 years, when value has not done well, have outperformed their markets by 8% to 9% per year. They’ve typically done it by being specialized.

Mihaljevic: Professor Greenwald, this has been wonderful. Thank you for your wisdom and for giving us the gift of the new edition of your book, which is nice for all of us to have either for Thanksgiving or, for those who don’t have it yet, in time for Christmas.

Greenwald: Can I do two valedictory ads?

Mihaljevic: Please.

Greenwald: First, the questions have been just terrific, so I hope your viewers appreciate that as a general principle.

Mihaljevic: Thank you.

Greenwald: You’re specialized in questions, and you do it well. Second, the first book, Value Investing: From Graham to Buffett and Beyond, which sold over 100,000 copies in English and probably a couple hundred thousand or more around the world, is not particularly good. The second edition is a much better book. It’s good enough that, like Competition Demystified, we’ll never do a third edition. If you buy this one, it will be the last one you ever have to buy.

About the author:

Bruce C. Greenwald was Founding Director of the Heilbrunn Center for Graham and Dodd Investing at Columbia Business School from 2001 until his retirement in 2019. In addition to training thousands of students in the mysteries of value investing, he taught oversubscribed courses on the economics of business strategy and globalization. His book Competition Demystified, published in 2005, is still in print. He has also been Chairman of Paradigm Capital Management since its founding in 2007 and the Director of Research at First Eagle Funds from 2007-11, serving as a senior advisor since.

About the second edition:

The first edition of Value Investing: From Graham to Buffett and Beyond was published in 2001. It is still in print, having sold over 100,000 copies. It has been translated into five languages. Business school professors still assign it in their courses. But in the 20 years since the first edition, the economy has changed, the investment world has evolved, and the discipline of value investing has adapted to this new environment. This second edition responds to these developments. It extends and refines an approach to investing that began with Benjamin Graham and David Dodd during the Great Depression and was adapted by Warren Buffett, Charlie Munger, and others to earn returns in an environment in which the opportunity to buy a stock worth a dollar for 50 cents is no longer waiting in plain sight.

The foundation of this book is the course on value investing that Bruce Greenwald taught at Columbia Business School for almost a quarter century. His aim in the course, and our aim in the book, is to help the investor operating in the Graham and Dodd tradition find him or herself on the right side of the trade. The steps include searching for the right securities, valuing them appropriately, honing a research strategy to devote time to the right activities, and wrapping it all within a risk management practice that protects the investor from permanent loss of capital.

The book has been revised throughout, but the biggest change is the addition of more than two chapters on the valuation of growth stocks, which has always been a problem for investors trained in the Graham and Dodd tradition.

Successful value investing practitioners have graced both the course and this book with presentations describing what they really do when they are at work. There are brief descriptions of their practices within, and video presentations available on the website that accompanies this volume.

In addition to a selection of Warren Buffett’s letters, there are presentations by Mario Gabelli, Glenn Greenberg, Paul Hilal, Jan Hummel, Seth Klarman, Michael Price, Thomas Russo, and Andrew Weiss. Although their styles vary, they all are members in good standing of the Graham and Dodd tradition.