This conversation is part of our “Wisdom in Books” series and podcast.
We had the pleasure of speaking with fund manager Dan Rasmussen about his new book, The Humble Investor: How to Find a Winning Edge in a Surprising World.
The following transcript has been edited for space and clarity. (MOI Global members, access all features, including ways to follow up with Dan.)
John Mihaljevic: It is a great pleasure to have with me best-selling author Dan Rasmussen, founder and portfolio manager at Verdad. He has a new book out, The Humble Investor: How to Find a Winning Edge in a Surprising World. It will be the subject of our conversation today.
Before starting Verdad, Dan was at Bain Capital Private Equity as well as Bridgewater Associates. He earned an AB from Harvard College summa cum laude and Phi Beta Kappa and an MBA from the Stanford Business School. He also is a contributor to The Wall Street Journal, and his research has been featured in multiple volumes of The Best Investment Writing. In 2017, he was included in the Forbes 30 under 30 list.
Dan, what inspired you to write the book?
Dan Rasmussen: Since I started Verdad, I have been writing a weekly research note that we send out to about 30,000 subscribers. The essence of those research notes is what we are thinking about, working on, and studying internally at Verdad. Whatever the number one problem of the day is that we study and research, we share it with our readers.
After doing that for about 10 years, I thought, “Why don’t I take these and put them together into a book that summarizes everything I have learned in the first 10 years of writing and running this hedge fund?” That’s what the book is. It’s trying to distill my philosophy and share 10 years of original research with a broader audience.
John: You curated and edited some of your writings that resonated the most so far. Is that how a reader should think about the format?
Dan: That is where I started, but ultimately, it is not just a compilation of notes. It is a book that runs from start to finish, with a constant idea but covers all of our most read and best-received research.
John: The book is structured in three parts – laying the groundwork, putting your theories to work, and then investing through the tests of time. Maybe you can talk about what you mean by laying the groundwork.
Dan: In college, I was a historian. I love studying history. The history of financial theory has totally gotten me. I am deeply interested in trying to figure out where the ideas that dominate Wall Street today come from. Why do we think what we think? Why do we learn what we learn in business schools? What are those core ideas? What do we make of them?
I started with trying to understand modern finance theory. I think the essence of modern finance theory goes back to the discounted cash flow model. It’s the first step in the thought process. The discounted cash flow model says that a company’s value is the sum of all of its future cash flows discounted back to the present. Starting in 1930, you can build on that. You start and say, “That’s the essence of valuation.” The next stage was people saying, “If that was true, why wouldn’t everyone just put all of their money in the asset with the highest expected return?”
The idea of risk was then introduced. You don’t put all of your eggs into one basket because there’s risk. That risk can be measured by volatility, so you need to focus not only on returns but also on the Sharpe ratio, which is optimized through diversification. Through there, you are onto the essence of modern portfolio theory. I think the theory is weakest at its very foundation, which is that discounted cash flow model. It is for two reasons, and these are the reasons why I call my book The Humble Investor. It’s a critique of some of these ideas.
The first idea is that you can predict cash flows years into the future, which is the first premise of the discounted cash flow model. If you don’t know future cash flows, then you don’t know the valuation of a firm. The second one is the idea of discount rates – that we can predict discount rates far into the future. If you discount back to the present, then you’ll get the accurate valuation. I have the perhaps radical but eminently commonsense position that you can’t predict cash flows and discount rates years from the future, so the whole theory is a bit of bunkum.
Much of the book is saying, “If we can’t predict the future, if we can’t predict these important financial variables, what do we do? What do we make of that? How do we react? How do we build an investing strategy on more firm foundations?”
The other key component of this is the idea for which Robert Shiller won the Nobel Prize – excess volatility. The idea is that even if we could predict future cash flows and discount rates, the volatility of those time series is only able to explain about 10% to 20% of equity market volatility. The rest is inexplainable by discount rates and earnings. In other words, there is something else beyond earnings and volatility driving the market most of the time.
What I try to argue here is that excessive volatility is the result of the way markets work – markets are driven by humans, and humans have erroneous forecasts about the future. We all go and place a set of trades. We make our predictions about the future. The future happens, and most of us turn out to be wrong in their predictions. Then all of those bets get reset, and that happens on an ever-evolving basis. The volatility of the market is driven by expectation errors. It is driven by investing mistakes, by forecasting mistakes. Yet, when you read finance textbooks or go to an MBA program or whatever it may be, the word “mistake” hardly ever comes up. The idea that our forecasts are going to be wrong most of the time never comes up.
What I’m trying to do with The Humble Investor is say, “How do we react to this? How do we incorporate the idea of the unpredictability of the future, of the certainty of investing mistakes, of the certainty of forecasting mistakes into building a better investment strategy that doesn’t rely on brittle, likely-to-be-wrong forecasting?”
John: One obviously big item is how to build an edge in equity investing. Given what you just described, what were some of your thoughts around finding an edge? How has that evolved over time?
Dan: If you approach equities from the perspective of just building discounted cash flow models, this book is an argument against it – that it’s not the right strategy because it’s built on the wrong theory and it makes the arrogant assumption that you can predict future cash flows and discount rates with a degree of accuracy that I don’t think is justified.
I did a major update of a famous study done in 2004. I updated it with 20 years of new data that looked at the question of the persistence of growth. Do companies that have grown at a certain rate continue growing at that same rate into the future? If you divide companies by their historic growth, does that predict future growth? The answer is not at all. Whether a company has grown fast or slow doesn’t mean anything about how it will grow in the future.
When you talk about stocks, people say, “This company is growing at 6%,” or something like that. I always say, “You don’t know that. You know it has grown at 6%, but the fact that it has grown at 6% doesn’t imply anything about its future growth.” This is probably one of the most interesting or provocative findings from quantitative finance. It has been replicated over and over again. You can’t derive any information about future profits from the historic series of profits – it simply doesn’t predict anything.
That leads me to the next idea. We then say, “What about using a guidance or estimates to predict the future?” One thing we find there is that you first have to segment out short-term and long-term forecasts. Some work by Andrei Shleifer and others has found that long-term forecasts tend to be so wrong that they’re contrarian signals. The higher the long-run forecast, the worse the returns; the worse the long-run forecast, the better the returns. Short-term growth forecasts are more interesting. They get you within the right range 50% of the time; 50% of the time, they are wrong. Growth ends up being way slower or way faster.
When you do a sum product where you say, “What happens when a company predicts its growth rate and then hits that growth number?” It turns out the stock does about average. If you do what you said you were going to do, you earn about an average return. What happens if you say you’re going to grow and then you grow less than that? Obviously, you get punished for it. What happens if you grow much faster? Your stock does much better than the market.
Given that the accuracy rate is about 50%, what ends up happening is firms that predict low growth have about a 50% chance of being right. When they achieve low growth, their stock does about average. The other 50% of the time, the forecast ends up being too conservative; the company does better, and the stock does better than average. Conversely, very high-growth firms are right about 50% of the time, and the stock does about average. However, most of the time they’re too optimistic, and the stock does so badly when they disappoint that the growth forecast doesn’t provide any information because when it’s wrong, the market reaction is big enough to punish your erroneous prediction so much that when you’re right, it was priced in anyways.
What you start to see is that with equities, you’re playing in a world where you’re betting against expectations. I like to say that investing isn’t a game of analysis – it’s a game of meta-analysis. It doesn’t matter what you know or what you’ve figured out; it matters what you know and have figured out relative to what the market has figured out. In this case, we find that buying stocks that have lower expectations, that are priced for more disappointing outcomes, ends up leading to systematically better returns because when the forecasts are wrong, it accrues to your benefit versus when the forecasts are wrong at punishing you. That’s what has happened over long periods in equity markets.
As a caveat, we should say this dynamic has not played out in the US for about 10 years, but internationally and in the history of the US market, this phenomenon or pattern has been very much true.
My idea is that if you want to build a winning edge in equities, the first thing you need to do is look at valuation and focus your attention on the places where the forecasts or what’s priced into the stock valuation is an easy bar to pass rather than focusing on those where expectations are high and consensus optimism is too great.
There’s a second idea, which is, “What does predict future earnings if not prior growth?” It turns out there’s a wonderful metric – gross profit divided by assets, which is a proxy for return on assets or very highly correlated with return on assets or ROIC. You can think of this as among a family of different metrics, generally calculated by dividing an income statement or cash flow item by a balance sheet item, and you’ll get something in this realm.
It turns out that companies with higher gross profit divided by assets, higher return on assets, do tend to exhibit higher long-term future growth, especially higher long-term future growth than what’s priced in. Buying very high-quality businesses as measured by ROIC or gross profit to assets turns out to be another path to achieving the same outcome. Relying on growth projections by discounted cash flow models or on historic growth will lead you astray, whereas relying on these metrics will push you constantly in the right direction in terms of looking for firms that are undervalued, where expectations are too pessimistic, and that are quite high quality, where the fundamental businesses are strong and likely to generate sustainable cash flows years into the future.
John: How does this relate to the idea of base rates?
Dan: The idea is that when you’re thinking about a company, one way to make forecasts – which is the way we experts traditionally do it – is to say, “I’ll analyze this company. I’ll study everything I can possibly figure out about it. Once I know all of that, I’ll develop a forecast.”
Base rates are different. They say, “Rather than doing it that way, you want to look at the historic probabilities.” Take a company, or a tech company, or a big tech company, or whatever set of categorical distinctions you want to make about the company, then say, “What has been the historic distribution of outcomes for companies like this – whether for revenue growth, earnings growth, or stock price return?” Then base your forecast on that categorical set of historical probabilities rather than on your own expert judgment based on detailed analysis of the individual example.
That’s what drives me towards trying to figure out and say, “Rather than looking at DCF models, let’s look at the predictive value of some of these categorizations or things that help us separate stocks into categories.” I argue in the book that two of the most meaningful ones are valuation, that low-valuation stocks do perform quite differently from high-valuation stocks – most of the time better, not in the United States in the last 10 years, but mostly better – and then hot stocks with high profitability or high quality as measured by gross profit assets or any set of those related metrics – which also have done very well over time, including in the US recently – and that by looking therefore for stocks that fit a pattern of stocks which end up doing well rather than trying to use your own individual analysis to understand a company in isolation, you end up doing better.
John: How should we understand the experience of the US over the past decade?
Dan: I love thinking and writing about history. History doesn’t follow an arc. It doesn’t bend in any particular direction. There is no end of history. There are no great forces that will drive it in a specific direction. Marx is wrong. There are no laws of history.
I think history is always and everywhere contingent on human action and human events. We have to look at the last decade with that in mind. We can have all of these base rates and ideas about how the world generally works, but there are also moments in history where great events happen and great people act and shape the future of nations and of the world.
In the past decade of economic history, that has indeed been true. We have gone through a period of technological innovation that seems to happen once a century, maybe even less frequently than that. The confluence of events that have happened – the adoption of the cloud, the rise of mobile computing, SaaS software, social media – all of these related technologies blossomed in the 2010s and since the financial crisis when people had widespread high-speed internet access and high-speed internet access on their phones. We have seen YouTube launch. We have seen the iPhone. Think of Salesforce and the tremendous rise of Nvidia more recently. There has been a huge boom in technological innovation.
The innovators have been rewarded for their innovation. They have earned massive profits that have been growing very rapidly as these new technologies have grown and expanded. Because of that, the traditional relationships between growth and value have not worked since these companies – these innovators – have beaten expectations year after year to a shocking degree.
Ten or 15 years ago, nobody could have predicted how profitable Amazon was going to be in 2024, or Apple, or Facebook, or any of these stocks. They have outperformed what anyone imagined and rather than mean-reverting down, their stocks have kept going up and up, and the valuation multiples have been reset upwards to the point where these are colossal, world-beating companies whose CEOs were invited to the presidential inauguration yesterday.
You therefore say, “What happens next? Is that innovation likely to continue? Will these companies continue to be as dominant forever as they were over the last few years? What is priced in and how does that relate to the future?” As unprofitable as it has been to be a pessimist about technological innovation, the right stance perhaps isn’t pessimism about technological innovation but rather agnosticism, like saying, “I don’t know because, traditionally, when new technological products have been released, the customers need to benefit more than the sellers.” As these adoptions get more widespread, not only does new competition arise, but these new innovations are commoditized and their use to the customers becomes greater and greater as the revenues to the companies become greater and greater.
In the second wave of technological innovation, the rewards don’t go to the innovators – they go to the early adopters and the customers of these companies. That’s likely what we will see. As all these exciting technological innovations spread, widen, and dissipate, we will see rising productivity and rising margins among normal companies that were not innovators but have simply adopted these technologies.
At the same time, the central investment question of 2025 is quite clear. “What is the future of AI?” I think we have two futures – one in which AI is equally or more impactful as an innovation as cloud and mobile and SaaS were in the 2010s, so we see another tech boom, an innovation wave happening in the 2020s just like it happened in the 2010s. The second alternative is that we have some combination of trajectories and things don’t turn out as great as anticipated. I’ll offer three possible paths for AI not being quite as exciting.
The first is that, quite simply, the technology works and is great, but it takes longer than people expect to roll out and be adopted in a widespread way – just like the internet did. Amazon was founded in the 1990s, but it didn’t become the colossus it is for quite a while. Between the 1990s and the time Amazon became a powerhouse, we had a 10-year drought where everything tech was a horrible investment in the stock market. That’s one possibility.
The second possibility relates to the idea of competition neglect, which is that all the big tech companies seem to be building the same thing, without perhaps acknowledging that only one of them will be the real winner. Yes, there will be winners in AI – like there was a winner in search or a winner in social networking – but the also-rans will be burned by their excessive investment in AI as the winners come to dominate.
A third potential future is one in which AI doesn’t work out – we have seen a tremendous moderation in the business cycle in the United States, a strong stabilization of economic growth that has come from a transition from a manufacturing-based economy to a services economy. If AI pans out as everyone says it will, we’ll see the transition of a services economy back into a manufacturing economy where the services and the knowledge are what’s being manufactured, but in every other way, it looks like a traditional manufacturing-type investment where you’re putting a massive amount of costs in – whether that’s buying servers, building out data centers, or investing in the LLM – and those investments are tremendously energy-intensive.
They require a massive investment, for example, in electricity or a power in order to function. Therefore, we see a return to some of the more traditional things we saw on a business cycle where you have booms of overinvestment and then busts as that overinvestment gets burned out, where you have huge capex swings, and where a certain level of cyclicality returns to business in a way that people weren’t anticipating.
Those are only three potential futures – none of which might occur or all of which might in some way – but all of which are to say that history is contingent. We don’t know what will happen, yet the markets are pricing in a great degree of certainty about it. My argument in The Humble Investor is that when other people are certain, we should try to take the other ends of the trade. That’s where the best odds are likely to be.
John: Since we’re on the topic of AI, I am also wondering whether there is a scenario where AI succeeds so much that it ends up obsoleting or crushing the business models of a lot of the players that sell software because AI could theoretically create software at an extremely low cost. How will these companies keep their pricing power if there is all this deflationary pressure from AI-generated competing products?
Dan: That is a brilliant point. You can think about some of the classic arguments for SaaS companies. “The switching costs are too high. It would cost someone so much money to extract our database and turn it into something.”
AI could do that pretty quickly. Switching costs are likely to come down at the very least. To your point, I think AI is most obvious. We use a lot of AI tools internally, primarily for coding. It’s great for coding. It can do a lot of that for you. It can check your code and even write a first draft. It will dramatically accelerate people’s ability to build software.
If software is the number one valued thing where excessive profits are being earned – or at least great profits relative to the investment – what does this mean for it? The argument that AI could contribute to those profits getting competed away seems a pretty sensible one.
John: Let’s talk about geographic diversification and investing globally. Some of the analogous argument to the US experience over the past decade also applies to US equities in general as compared to non-US equities where I think the concentration of market value today in the US is probably quite high historically. What does that tell you about going-forward returns – US versus international?
Dan: It’s wrapped up in exactly the same logic because right now, the US is the high-valuation country, and everywhere else in the world is low valuation, with the exception of India or something. Broadly, Europe, Japan, even China – anywhere outside of the US you want to go – you are buying much lower valuations.
The tension between the two arguments – on the one hand, folks are saying, “The great world-beating companies, the Mag Seven, are all in the United States. That’s why US corporate earnings have been so much higher than anywhere else in the world, and that’s likely to remain so with the US continuing to dominate the world in innovation, continuing in its deregulatory abilities to offer the best place for businesses to be headquartered, and attracting the best talent because of its low-tax regime.”
On the other hand, international markets have gone so cheap that if the US is 65% of the ACWI and international stocks are 35%, by revenue, the US is probably 35% and the ex-US portion is 65%. It’s almost reversed, which gives you a sense of the magnitude of the difference in valuation multiples between the two.
We’re also faced with the same argument we had over value and growth, which is that international diversification hasn’t worked for a long time, and being 100% in the US has. I relate the reasons for that to the historical contingency where the set of companies that drove all the US outperformance are technology companies in the same sector building on the same innovations that are making the same bet on AI – all of them at the same time.
We’re left with the US-international debate as a value-growth debate. I am here making the argument for value and for international, which is really the same argument in the face of being wrong, of losing that bet for 10 or 15 years now, where the only right answer has been the US, and the only answer within the US has been growth. However, just because something has been true even for quite a long time, it doesn’t mean things have changed forever and it will always be true. The base rates would imply that low expectation stocks will do better and high expectation stocks will do worse. The gap in expectations between the US and international is so clear right now, but I think it’s prudent to be more diversified than one might normally be.
As a side note, I was looking recently at the correlations between major indices. US value stocks now have about the same correlation to US growth stocks as international stocks have to US growth stocks, which gives you a sense of how much the market is dominated by one set of stocks, by one trade, by one bet, and everything else is falling to the wayside.
John: Is there a historic analogy here? Have we seen this movie before?
Dan: In various ways, it is said that history rhymes but doesn’t repeat. There are certainly times that rhyme, and I’ll give a few. One is the Nifty 50 bubble where a set of great American companies in the 1960s were thought to be true world-beaters and huge amounts of capital flowed into those stocks. You only had to own a few of them to win. Then you had this period over the next 10 or 15 years of zero-percent equity returns. Many of those companies ended up doing well over the long term. They were great companies, but they got way too overvalued during that period.
A more extreme analogy would be Japan in the 1980s, with people saying the way the Japanese run companies and the Japanese management style is so much better than anywhere else in the world, so why shouldn’t you have all this capital in Japan? Even if 30% of the ACWI is now Japan, it makes sense because Japan’s so great. That could be another analog.
The 1990s could be another one where the dominance of technology and innovation as themes is recurring, but there are counterpoints to each of those. No analogy is perfect. An important point is that even if you say today rhymes with those periods, it’s extremely hard to know. For example, are we in 1995? Are we in 1999? How does our current situation match?
I tend to think that one of the most important things to know about our current environment is simply how low risk everyone thinks it is as measured by high-yield spreads, which is one of my favorite macroeconomic indicators and now pretty much near all-time lows where the market is essentially pricing in that we’ll never have defaults again. That’s probably wrong,
John: That’s the spread between high-yield borrowing costs and AAA?
Dan: And duration-matched Treasuries – yes, exactly.
John: Have we ever been this low or close to this low on the spread?
Dan: The late 1990s and 2007.
John: Do you think there’s anything different – either structurally or in the Fed or whatever the appropriate intervention force would be – that would render defaults a thing of the past?
Dan: We saw a very aggressive reaction from the Fed to COVID. In some ways, it was a brilliant reaction to very explicitly try to prevent the blowing out of the high-yield spread and to prevent defaults by even buying corporate bonds, by buying high-yield bonds. A lot of people seemed to take it as a signal that the Fed was willing to expand its arsenal of tools and be very aggressive to prevent default cycles from getting out of hand, to stop what Ben Bernanke called “the financial accelerator” from getting off and going.
However, lending has been around for a very long time in human history, and we’ve always had default cycles. I don’t think we’ve reached the end of history. There will be default cycles in the future. There would have to be. Capital does get misallocated during certain periods, and bond holders do demand the money back. It seems inevitable that defaults will return. The Fed can’t beat the lending markets forever, but I do think we’re in a period where risk has been building up into the system because the belief that defaults will be so low for so long or will never come back has become so prominent.
John: You have a chapter in the book on bond investing – “Chase Quality, Not Yield.” That would seem to relate to the notion of those spreads being so low that people will just chase that last bit of yield and not focus on quality at all.
Dan: Yes, what people most commonly seem to get wrong about fixed-income markets is that they take off their efficient market hat when they leave the equity market and arrive in the wild world of fixed income because yields seem like such an obvious and compelling story. If one bond offers a 5% yield and the other an 8% yield, they are both contracts that, in theory, guarantee you that coupon payment. It’s logical to say, “The 8% yielding bond should return 8%, and the 5% yielding bond should return 5%.”
In the well-behaved corners of the bond market, that’s true. If you start from AAA government securities and go down to BBB corporate securities, as you walk from the US Treasury bond at 4% to the Microsoft bond at 4.5% to the Exxon or Ford bond at 5%, you generally do earn a little more for taking on a bit of corporate credit risk. However, once you get into a wilder and woolier world where you start to see below-investment grade, you see this concept of a fool’s yield, which is that as yields rise, total realized returns go down so that if you are offered a 6% yielding bond and a 16% yielding bond, history and base rates would show that the 6% yielding bond usually does better than the 16% yielding bond because the magnitude of defaults and downgrades In the latter case is enough to bring down expected returns.
The markets are efficient. If you have to borrow at 16%, you’re probably not a very good borrower. You’re probably pretty risky. By the way, we’ve been lending money to companies for hundreds and hundreds of years, so we’re pretty good now at pricing that risk – even better than we are at pricing equity risk. It’s such an efficient market that you don’t earn an incremental return by being fooled into investing in companies with much higher yields.
Yet, the sexiest thing on Wall Street these days is private credit, which is basically high-interest rate loans banks wouldn’t touch that offer a huge premium to what, say, Treasuries would offer. That story is probably the tip of the spear of excessive promiscuity in corporate lending. It’s probably the place – private equity, private credit – that will be hit the worst if we are to see a default cycle.
John: The Fed has intervened aggressively, which has most likely led to the inflation we’ve seen. I’d love to talk to you about inflation and how investors stay afloat in such an environment.
Dan: We’ve learned a few lessons over the past few years. One of my favorite metrics is this idea of inflation beta. How does an asset react when inflation goes up or down? If you’re trying to build a diversified portfolio, what assets do you need to incorporate in order to survive inflationary environments?
Interestingly enough, a lot of people talk about TIPS (Treasury Inflation-Protected Securities), but one contrarian argument I’d make is that they’re pretty lousy most of the time at protecting against inflation. The reason TIPS are lousy at protecting against inflation is because they are comprised of both a bet on interest rates and a bet on inflation. You earn a specific interest rate and then you earn inflation spread on top of that, so the market is pricing in both. When there’s a lot of inflation, interest rates go up, so the interest rate component of your TIPS goes up, and you lose money. Your inflation component obviously does better, so you make money, but you end up about even. Theoretically, TIPS end up being a rather lousy way for people to protect their money against inflation.
It turns out commodities work much better, especially gold and oil. Both come with their own challenges. Oil in particular is highly volatile and dependent on the business cycle, so gold probably becomes the best option on a Sharpe-adjusted basis for protecting your money against inflation. Think of gold and bonds as your almost equivalents where they’re both low-risk assets, but gold does well in inflationary environments, and bonds do badly in inflationary environments and the opposite in a deflationary environment. Incorporating some element of commodity trading into a broader portfolio and especially thinking about gold is quite important for surviving inflationary times.
John: When it comes to gold, how do you think about its continued relevance going forward in light of Bitcoin and crypto capturing so much mindshare and seemingly being increasingly adopted by institutions? There is now even talk of the US government possibly doing something that would legitimize Bitcoin. Does that affect the attractiveness of gold?
Dan: I don’t think so. They are two very different things. They certainly trade in a highly uncorrelated way.
Gold has two attributes that distinguish it from Bitcoin and crypto. The first is its longevity. There’s a famous paper by Campbell Harvey that shows the amount in gold you pay a Roman centurion is the same amount in gold you pay a US Army captain today. There’s such a long history of gold as a currency. Gold has maintained its value over literally thousands of years. It has a history Bitcoin just cannot rival.
The second is looking at the way these things trade. Gold has about the volatility of bonds – a little more than bonds, but thereabout. Crypto has the volatility of a crazy meme stock. I think of the role of gold as a bond equivalent or a bond alternative, whereas I think of crypto as equities on steroids. They are very different roles these might play in your portfolio.
You could say, “I want to protect some legacy investment that I need to survive for 30, 40, or 100 years. I want to make sure it stays protected from inflation and retains its value, so I’ll put it in gold.” That logic makes a lot of sense, while saying, “I’ll put all of that in Bitcoin for the next 100 years” sounds like putting all of your money in Nvidia for 100 years. It might be an incredible investment for the next one, two, or three years, but how do we know Bitcoin will still be around 100 years from now? We can be fairly confident that gold will.
John: When it comes to equities that would protect against inflation, there is the perception out there you should just invest in the highest-quality businesses that have pricing power and basically run on intangibles. On the other hand, historically, a lot of the assets that have protected well against inflation are, in fact, capital-intensive – like real estate or mining. How do you think about the best sectors for inflation protection?
Dan: Both arguments can be true. You never know which will prevail at any given time, but it’s enough to say that equities won’t necessarily be hurt by inflation and won’t necessarily benefit. It’s very contingent.
There are certain periods where there’s been high inflation and equities have been walloped, and there are other periods where there’s high inflation and equities have done quite well. I wouldn’t spend a huge amount of time thinking about inflation when I’m thinking about my equity portfolio. It doesn’t seem like a huge driver day-to-day, but it’s a massive driver for the bond market.
That’s why I think gold fits as a fixed-income alternative. It belongs in the conversation about what I do with my safe assets, not necessarily the conversation about what I do with my risk assets or what I do with my equity assets, which I think depends far less on inflation than what’s happening in the bond market.
John: You end the book with a chapter on designing a counter-cyclical asset allocation strategy. I’d love to hear about that.
Dan: This has been the great whale of an idea I’ve been pursuing – how to build more robust, counter-cyclical portfolios that work regardless of economic scenario and incorporate a lot of the ideas in the book and ideas I’ve been working on for the last few years.
Where I’ve gotten on the topic of portfolio construction is that if you think about efficient markets – and I love going back to efficient markets as a North Star – they say predicting expected returns is quite hard, maybe even impossible, but it doesn’t say anything about predicting volatility or predicting the correlations between different assets. We have found that volatility and correlations as a result are much more predictable than returns. It’s much easier to predict whether stocks and bonds will be correlated next month than whether stocks or bonds will outperform each other. It’s much easier to predict that bonds will be less volatile than stocks, for example, than it is to predict that stocks will return better than bonds in any given short horizon period.
It also turns out that both volatility and correlations are widely variable. For example, we’ve gone through periods where stocks and bonds are negatively correlated and periods where they’re positively correlated. During those periods, if you took last month’s correlation and forecast it to be next month’s correlation, you would have done relatively well at predicting the correlations between those assets.
The argument I make in this chapter and the type of logic I’ve been working on is that if you’re building a portfolio, yes, you need to have a prediction about expected returns, but a prediction about volatility and a prediction about correlations are also equally important to portfolio optimization and are much more predictable. What I have been trying to do is to code that logic into software and say, “Let’s build Sharpe-optimized portfolios that incorporate multiple asset classes and then factors within the asset classes so we could trade value as an asset, or gold as an asset, or Treasuries as an asset, or momentum as an asset.”
If you can look at what the correlation of each of those is, what the volatility of each of those is, even if your expected return is simply the long-term average of each of those things, you should be able to build more robust portfolios that are reacting in real time to changing macroeconomic conditions, changing correlation structures, changing volatility environments. If you can do all of that consistently and intelligently and get all those basics right, you should be able to build portfolios that work better, have higher Sharpe ratios, and can be optimized to achieve whatever target return or target volatility you’re looking for.
That’s what I’ve been working on. It builds on a lot of the foundations of the work I’ve been talking about in equities and fixed income.
John: What has been the experience so far in terms of success and what you still regard as any big questions to answer or issues to resolve?
Dan: It has become quite clear from trading these types of strategies that the insight about correlations and volatility is right. You can build more diversified, lower-risk portfolios by incorporating a broader set of assets. You can achieve a higher Sharpe ratio through doing this. It’s clear, and there’s a whole variety of assets that a lot of people don’t trade. They think only about stocks or about bonds and stocks, not about commodities, currencies, and the factors.
Once you expand your horizon to consider those things, you start to see that you have a much broader set of tools to diversify your portfolio if you’re willing to use leverage, go short, and do all the other fancy things hedge funds love to do to achieve a smoother path to hopefully equity or equity plus returns.
John: I know you base a lot of your conclusions on quantitative work that you do internally. To what extent would you describe yours as a quant-first firm versus bringing in qualitative judgment as well?
Dan: We’re probably 70% or 80% quantitative, but we try to be logical. We don’t want to unleash machine learning and say that whatever the machine finds is right. We want to build up logical structures and make sure everything we’re doing and everything we’re telling the computer to do makes sense and follows logical economic principles. In that sense, we are perhaps more fundamental than your average quantitative investment firm.
John: Who do you hope to reach with this book?
Dan: I think of the audience as your readers of The Wall Street Journal, The Financial Times, or The Economist – people interested in an in-depth dive into how markets work and the quantitative research on financial markets in an applied way; people interested in behavioral finance and the science of prediction, which hopefully is a large audience.
John: Dan, thank you so much for taking the time to sit down with me for this conversation. I would recommend the book to all of our members. It is a terrific read – The Humble Investor.
This conversation was recorded in January 2025.
Daniel Rasmussen is the founder and portfolio manager of Verdad Advisers, a ~$900M hedge fund. Before starting Verdad, Dan worked at Bain Capital Private Equity and Bridgewater Associates. He is a member of the investment committee of the Trustees of Donations of the Episcopal Church, is the New York Times bestselling author of American Uprising: The Untold Story of America’s Largest Slave Revolt, and in 2017, was named to the Forbes 30 under 30 list. Dan is a contributor to the Wall Street Journal and his investment research has been featured in multiple volumes of The Best Investment Writing. Dan earned an AB from Harvard College summa cum laude and Phi Beta Kappa and an MBA from the Stanford Graduate School of Business.
Share this post