Howard Marks on Avoiding Disaster, Risk Posture, and the AI Bubble

Howard Marks

Thank you, William. With an introduction like that, I'm tempted to say go on.

Howard Marks

I had dinner with a man named David Van Benschoten, who ran the pension fund for General Mills. He was a good friend and client, and he told me that he'd been running the plan for 14 years. In those 14 years, their equity portfolio had never been above the 27th percentile of pension fund equity portfolios or below the 47th — solidly in the second quartile for 14 years in a row. But interestingly, for the 14 years overall, they were in the fourth percentile.

That's incredible maths. You would say, if you bounce back and forth between 27 and 47, on average you're probably about 37. No — fourth. How could that be? The answer is that most investors shoot for the stars and occasionally shoot themselves in the foot and wreck their record. Once you have a big loss, it takes a long time to get back to scratch.

So I wrote the first memo, called "The Route to Performance." I went back and looked at it recently, and it says: simply put, what the pension fund's record tells me is that in equities, if you can avoid the losers and losing years, the winners will take care of themselves. When we started Oaktree in 1995, I wrote that down and it became our motto — and it still is. If you can avoid the losers, the winners will take care of themselves.

Howard Marks

I went back to read the 1940 edition, because the owner of the book asked Seth Klarman to update it, and Seth asked me to do the part on fixed income. I got a little irritated when I saw them calling it a negative art — I thought they were denigrating what I do. But then I realised what they were saying.

If there are 100 bonds out there and they're all 8% bonds, and you know that 90 will pay and 10 will default, it doesn't matter which of the 90 that pay you buy — they all get the same return. The only thing that matters is that you don't buy any of the 10 that default. You improve your performance not by what you buy, but by what you exclude. It is a negative art. In fixed income, where you are promised a return and the only moving part is whether the company keeps its promise, all you have to do is weed out the defaulters and you get the promised return.

That was a very good mindset when I started the high yield bond business in 1978. I didn't have those words for it, but I thought that way. When we moved into other businesses — distressed debt funds in 1988, emerging market equities in 1998 — it wasn't enough merely to avoid losers; we also had to find some winners. But we maintained the motto because the risk-conscious mindset is a great guidepost.

Howard Marks

I had lunch yesterday with Charlie Ellis, who wrote the article in 1975 that gave rise to this whole line of thinking — "The Loser's Game." He said there are two styles of tennis. The professional has to hit a winner to win a point because if he hits a mild return, his opponent will put the point away. Professionals are so good at what they do that it's mostly under their control; they even track something called unforced errors because there are so few.

But the amateur, because they lack that control, has to content themselves with not hitting losers. If I can get the ball over the net and within the bounds of the court ten times in a row, chances are good my opponent will stop at nine. I win the point not by hitting a winner, but by avoiding a loser.

Investing is not like championship tennis. We don't have that much control of the outcome — there's too much randomness, too much uncertainty, too many things that are unknowable. In a game like ours, swinging for the fences can get you carried out. It is better to play within yourself, emphasise consistency, avoid the grand gesture, and strive for steady competency. That is a lot of what we've done.

Howard Marks

One of the quotes I've been using most in the last decade is attributed to Mark Twain: "It ain't what you don't know that gets you into trouble. It's what you know for certain that just ain't true." No sentence that starts with "I don't know but" or "I could be wrong but" ever got anybody into big trouble. You get big trouble when you say, "I'm 100% sure that XYZ" — and then you take bold bets on a premise that turns out to be incorrect.

Long-Term Capital Management operated that way because they thought their method was infallible. It produced tiny returns, but they levered them up into big returns with a lot of borrowed money. When you have a problem on leverage, your losses are magnified and you can be carried out — as they were. Amaranth likewise bet boldly and incorrectly.

Howard Marks

One of my favourite illustrations is: never forget about the six-foot-tall person who drowned crossing the stream that was five feet deep on average. The notion of surviving on average is meaningless and irrelevant. You have to survive every day in order to reach the finish line — which means you have to survive on the worst days.

If you have a portfolio directed at maximising results if everything you hope comes true, you expose yourself to the possibility of being carried out if what actually happens is something different. A lot of investment management decisions come down to this: are you going to try to maximise your gains if things go the way you hope, or minimise your losses if they don't? You can't do both at the same time.

I wrote a memo called "Fewer Losers or More Winners." If you are going to try to win in investing — superior results — how can you get them? You either have more of the things that go up, or fewer of the things that go down, or both. Most people can't do both. The skillful aggressive player might get more winners; the skillful defensive player might have fewer losers. Very few people have the equipment for both. Most have to choose — and that choice should be a conscious one.

Howard Marks

I wrote a memo — I think about eight years ago — called "Calibrating." I used the image of a car's speedometer: zero is no risk, 100 is maximum possible risk. Every person, institution, and money manager should figure out where in that continuum they should normally be. What is the right risk posture for me given this client, my employer, my institution?

For individuals, it is a function of age, wealth, income, the relationship between wealth and needs, number of dependants, level of aspiration, proximity to retirement, and intestinal fortitude. Take all that together and figure out where, from zero to 100, is the right place for you normally. You might say, "I'm young, I don't have many dependants, I'm aggressive, I can stay with it, I have plenty of time to recover from a mistake" — so perhaps an 80 or 85.

There is no lookup table that tells you which combination of assets produces an 85. It is a mindset, a way to direct your thinking. And once you've established a normal posture, the next question is whether you will stay there always or try to vary it as opportunities arise. If you will vary it, where do you want to be today?

Howard Marks

On all these questions there are no right answers — only a range of possibilities, choices, none of them perfect. Most people are not 100% maximisers or 100% preservers. It is a personal choice, and importantly, no right or wrong.

Nick's attitude — never fiddle at all — is a little too idealistic in my opinion. I believe there are good opportunities once in a while, not every day, to become a little more aggressive or a little more defensive. I wouldn't do a lot because it's easy to be wrong, but I wouldn't let all those opportunities go. My son Andrew noted, when I wrote "Mastering the Market Cycle," that I thought our calls had been about right. He said, "Yeah, Dad, that's because you did it five times in 50 years." Certainly not every day.

By the way, there was a study of clients at Fidelity that concluded the best performance belonged to accounts of people who were dead. Fidelity has not been able to find that study, so it may be apocryphal — but you get the point. Overtrading is a mistake. Buy and hold is highly superior. But for people with ability and temperament, there are occasions when you want to behave contrarily — becoming a little more defensive when the market is precarious, a little more aggressive when it is generous.

Howard Marks

You're right that the calls were all in the last 25 years — which means it took me 30 years to get up the nerve and feel I had enough insight to make one. The first was the first day of 2000, about the tech bubble. And I didn't know anything about tech stocks or technology or the internet. I call the process "taking the temperature" because it is about assessing the behaviour of the people around me.

Buffett always says it best: "The less prudence with which others conduct their affairs, the greater the prudence with which we must conduct ours." When everybody is carefree and sees no risks, their exuberance has probably moved prices so high that things are dangerous. Conversely, when people are depressed and all they want is to get out, their pessimism usually renders things so cheap that it is the time to become highly aggressive. Contrarian. Counter-cyclical.

As my son said: five times in 50 years, there were compelling times to do it — moments when the logic was compelling and the probability of being right was high. I never did it without trepidation. I was never certain. But it felt like the right thing, so it was worth trying. Even when you think you're right, you shouldn't assume it's more than 80/20 — maybe it's really 70/30. If you know what you're doing, it's worth trying. But if instead of five times I had tried 500 times? After 56 years, multiplying by 365, I'm approaching 20,000 working days. If I had made a call every four days, my record would be 50/50 at best. You have to wait until it's compelling and then hope you're right.

Howard Marks

I don't think we have become institutionalised. We've just become bigger. My dad used to say that marriage is a wonderful institution for people who like living in institutions — and I don't. My first job was at Citibank, where I spent 16 years, and I learned that I don't like institutional living. They would say, "We can't, because there are institutional constraints." That word "institutional" is a powerful one — when I was a boy, "he lives in an institution" meant an insane asylum.

I try hard to avoid bureaucratic tendencies. I wrote a memo around 2005 called "Dare to Be Great" — mostly a rant against bureaucracy and committees. At 29, I became director of research at Citibank. They put me on five committees that met for a minimum of 16 hours a week, and I almost went mad. Committees go as long as the person who wants them to go longest wants them to go.

Think about it: if you have a brilliant, idiosyncratic insight, can you imagine trying to convince a committee of ten people to support it? If most people in the market are doing A, how can you convince the majority of a committee to do B? You can't. Bureaucracy and great investing are counter-indicated, as the doctors would say.

When Lehman Brothers went bankrupt in mid-September 2008, we had raised the biggest distressed debt fund in history — by a factor of about three — and we had $10 billion sitting on the shelf. Most people thought the financial world was going to melt down. Bruce ran the fund and he bravely invested an average of $450 million a week for the next 15 weeks — $7 billion in one quarter. We could not have convinced a committee. The good news is we had pre-raised the money, so we didn't have to convince the clients that the best time to invest is during a crisis. People can't raise money during a crisis because they're frozen into inaction. It was certainly uncomfortable. We certainly weren't sure we were right. But it seemed like the right thing to do.

Howard Marks

Let's take the counterfactual. When you sit down to start investing, you have to think about what elements will make you a success — and success means doing better than others. You can't say "I'm smart," because everybody else in the business is pretty smart. You can't say "I went to the best schools." You can't say you have a generalizable intelligence that will translate to every asset class. You have to develop a knowledge advantage — usually by developing an approach that is right and that you implement consistently, or by knowing more than the other people.

That might mean having more data, though the SEC's job is to ensure everybody has the same data. Or it might mean doing a better job with the data, or having more insight. Or it might mean going into inefficient markets where information is not evenly distributed. But you need some source of superiority. Investing is an incredibly competitive game. Winning means beating people who are similarly intelligent, numerate, computer-literate, hardworking, and highly motivated. You must have an edge.

To illustrate market efficiency: what if, when I got out of the University of Chicago in 1969, someone had approached me and said, "I'm a bookmaker. I've concluded I can make money on football if I know which team will win the coin toss. I'll give you 15 PhDs and a Cray supercomputer — all you have to do is predict the coin toss." If it's a fair coin, it can't be done. That is an efficient market — a market where everybody knows as much as you do and there is no edge. A less efficient market is one where hard work and skill can pay off.

In August of 1978 I got the phone call that changed my life from the head of the bond department at Citibank. He said there was some guy in California who dealt in something called high yield bonds — could I figure out what that was? They were called junk bonds. Most people wouldn't touch them with a ten-foot pole. I didn't get my first public pension fund account for 18 years — they were reputationally and politically unpalatable. If it's an asset class nobody will buy at any price, maybe it's full of bargains. But if everybody thinks something is great and will gladly pay any price, why should you think you can get a bargain there?

Howard Marks

One of the great things to remember is that there are two kinds of people who lose money in the market: the people who know nothing, and the people who know everything. I hope I never know nothing, but I never think I know everything. Specifically, I believe the macro future is unpredictable.

But there are things that can give us a hint. My second book, "Mastering the Market Cycle," published in 2018, talked about tendencies. We never know what the market is going to do, but we can have a feeling for when its tendency will be to do well or to do poorly, and that tendency is largely determined by where we are in the cycle. When things have been going great for 16 years, PE ratios are high, and bond yield spreads are narrow, we can assess that the tendency of the market may be to do less well — and act accordingly. You don't have to make any predictions.

None of those five calls I talked about earlier was based on a prediction. Each was based on an observation. As I always say: we never know where we're going, but we sure as hell ought to know where we are. Are prices and valuations high? Is risk being ignored? Are people acting in an exuberant, buoyant way? Those questions can tell you what the odds are, even though you don't know what the future holds.

The subtitle of that book — which the publisher buried — was "Getting the Odds on Your Side." You never know what's going to happen, but you can sometimes have a sense of what the odds of a certain outcome are. When the market is high in its cycle, the odds are against you. When it is low, the odds are in your favour. The odds can be in your favour and you can still lose money for the next year or two or three, and vice versa.

Howard Marks

Charlie Munger used to quote the philosopher Demosthenes: "For that which a man wishes, that will he believe." We tend to do that, and it is injurious. Let's say you're a stockbroker paid on commission. You tell yourself there's always something good to buy, so you buy every day for your clients and make a lot of commissions. But some days there is nothing good to buy. You have to accept that, be mature about it, be patient, and hold back.

It all goes with being mature, analytical, patient, insightful, understanding yourself and your biases, understanding the process of how money is made — and then waiting until the odds are on your side to turn up the wick. I believe you should have investments all the time, but sometimes you do it more aggressively and sometimes more defensively. You wait for those golden moments to really turn up the wick and become more aggressive.

Howard Marks

When you look for comparisons, the strongest one — not a perfect comparison, and I am not saying this is true in degree, only in kind — is the TMT, internet.com bubble of 1998 to 2000. The Nifty Fifty was different because it was not centred on one novel technology; it was around established great companies. The 2005–2007 subprime period is not comparable because that was a financial invention, not a technological innovation. But this — like the internet bubble — is a technological innovation that I think is going to change the world.

My recollection, though, is that we had a clearer view of how the internet would change the world. The vision of many in 1999–2000 has mostly come true: e-commerce became a major force. Today, I think we have less clarity. I have never heard anybody tell me specifically how AI is going to change the world — how it is going to be a business, how people are going to make money at it, how it is going to impact life.

Most bubbles are around something new, because the imagination is untethered and can go off in a flight of fancy — you can imagine trees growing to the sky. You are never going to have a bubble in paper stocks or timber stocks; those are too prosaic. People can estimate how many houses will be built and how much wood each needs. It is always something new. In 1969 it was growth stock investing, in 2006 it was subprime mortgages, in 1999 it was the internet, in 1720 it was the South Sea Company, in 1620 it was tulip bulbs in Holland.

Howard Marks

Changing the world and investors making money are not the same thing. Warren Buffett pointed out — I think at his 2000 annual meeting — that there is no doubt the internet would produce a great increase in productivity, but it was not clear it would have a positive impact on profitability. I think the same is true of AI.

I saw a CNN anchor say to a guest that AI has the ability to eliminate half of entry-level jobs. That may be true. If you can eliminate half the entry-level jobs, it could be more productive. But will it be more profitable — and for whom? If different companies are competing to provide the AI service, they may compete on price to the point where it's not profitable for them. Or if those who employ AI compete for market share, all the savings may go to consumers in the form of lower prices. Exactly how a labour-saving tool of AI turns into profits, I don't think anybody can say.

Howard Marks

What I've seen in euphoria after euphoria: first, you shouldn't assume that today's leaders are certain to be the leaders of tomorrow. They may well be, but don't bet your life on it. Second, you shouldn't assume that because the leaders sell at high prices, it's a good idea to invest in the laggards because they're cheaper. People say they have a low probability of success but a big potential payoff, so they buy. That's lottery-ticket mentality. If something has a low probability of success, you should accept that chances are good of unsuccess.

On AI specifically, I am led to believe you can make binary bets on companies that have nothing else going on — sink-or-swim bets — or you can invest in pre-existing great tech companies, which will get moderate benefits from AI if it is successful but will still be in business and profitable if it turns out not to be that big a deal. This brings us back to the very beginning of our conversation. Do you want a novel entrepreneurial pure play with no revenues and no profits today — a moonshot if it works — or do you want to invest in a great tech company that is already making a lot of money, where AI could be incremental but not life-changing? It is a choice. What is your game plan?

Howard Marks

At Oaktree we consider ourselves value investors. What a value investor does is look at a situation — a stock, a company, a bond, a building — and try to figure out its intrinsic value, then see how today's price compares to that value. Intrinsic value invariably comes from the fact that it can make money, be profitable, and produce cash flow.

Say I have a building that produces a million dollars a year in profit. I want to sell it, you want to buy it. I might say $12 million — an 8% return. You might say $8 million — you want 12% because it's risky. We can talk, but we're talking in a range centred around value. If you're doing Bitcoin, gold, diamonds, or paintings, there is no intrinsic value. Think about oil: it was $147 a barrel in mid-2007 and then $35 in January 2008. Nothing about oil changed — it was still black, it still made lights go. Oil doesn't have an intrinsic value; its price is what the market will bear.

How can you invest analytically if you can't base a conclusion on cash flow? You can invest in gold out of superstition. You can invest because you think it will go up, or because you think it will be a store of value, as it always has been. But you can't invest on the basis of something called intrinsic value. The people who bought gold a year ago have made a lot of money. But if you bought gold at the end of 2010, you've had a 7.7% annual return since then. If you bought the S&P at the same time, you've had a 12.7% return. Gold is not a disaster, but you shouldn't be distracted by recent gains. Over time it has been a lackluster investment.

Howard Marks

High yield bonds are part of what I call "lending assets" — debt, fixed income, bonds, notes, loans — instruments where you give somebody your money, they rent it from you, and they promise to pay interest every six months and return the principal at the end. You can calculate the rate of return if they keep that promise. It is called fixed income because the outcome is fixed; it is a contractual relationship promising a fixed return. The only moving piece is the probability that they will keep the promise of interest and principal — and it's the job of the credit analyst to assess that.

For most investors listening, in most areas the amateur should go into funds, ETFs, or some managed product. Charlie Munger said: it's not easy, and anyone who finds it easy is stupid. That applies to what I do, but also to stocks. Investing is a funny business: it is very easy to get an average return, and very hard to get an above-average return. But if you are content with an average return, you can get managed products at a relatively low cost with a high probability of delivering it. High yield bonds currently yield around seven percent. If you're happy with that, there are lots of managed products that will deliver it.

Howard Marks

I've concluded in the last few months that the toughest questions I get are the ones that start with "how." I can tell you what you have to do — keep an even keel, perhaps have a more defensive portfolio — but how to make that decision and how to keep an even keel is harder. Obviously, if you let your emotions run away with you, if you buy when things get exciting (which usually means prices are high) and sell when things get depressing (which usually means prices are low), it is going to be very counterproductive.

Investing is not a fluke or a pachinko game or a roulette wheel. It works over time because economies grow and companies improve their profitability. The most important thing for investors is to get on that gravy train and stay on it — invest early, invest a lot, and don't tamper with it. Having your emotions under control is essential if you are going to do that last thing: don't tamper with it. Getting on the gravy train and staying on it is far more important than picking exactly the right times to get in and out, or picking the exact stocks that will go up the most. The most important thing is to be a long-term investor.

I'm not a futurist; I don't think my vision of the future is bound to be more right than anybody else's. The future is not a single thing that, if you're smart enough, you can figure out. It is a probability distribution — a range of possibilities. Only one thing will happen, but many things can. You should accept that this introduces uncertainty into the equation and resist forming a certainty around one outcome and betting heavily on it, unless you have special expertise, which very few people do.

My favourite fortune cookie said: "The cautious seldom err or write great poetry." Every person has to decide: do I want to try to write great poetry and get rich if my bets are right, or do I want to avoid erring and be sure I'll do okay if my bets are wrong? You can't emphasise both at the same time. Life is uncertain. Investing is uncertain. Are you going to go for winners, or are you going to try to avoid losers?

Howard Marks

We all have to choose what's important to us. Charlie Munger used to say that someone who only cared about working hard and making money was a maniac. You have to decide whether that's for you — and it's not for me. The tritest of all sayings is that nobody on their deathbed ever said, "I wish I had worked more." I believe it is true. That is not how I am living my life. I have pursuits I enjoy greatly and I wouldn't give them up. Once you have enough money — or more than enough — why should you give up part of your enjoyment to have more?

The greatest saying I always use when giving advice to young people is from the writer Christopher Morley: "There is only one success — to be able to live your life in your own way." The hard part is figuring out what your way is. If I had described myself to you 40 years ago, I would not have described the person I am today — maybe I was wrong, or maybe I changed. But our goal should be to get to the end and say, "I am happy with the choices I made." Pursuing work and money and career and prestige because other people are doing it, or because you want to emulate the richest person in the world — you shouldn't do it unless it is really right for you. I don't think it's right for most people. Figure out what's good for you and pursue it.

Howard Marks

Thank you. My idol Warren Buffett always says he skips to work in the morning, and I feel the same way. I'm happy to go to work and I like what I do. I hope to keep doing it for a long time to come.

Howard Marks

Thank you. It's a pleasure speaking with you — let's do it again.