Lasse Heje Pedersen is a Danish financial economist known for his research on liquidity risk and asset pricing. He is the John A. Paulson Professor of Finance and Alternative Investments at the New York University Stern School of Business. He is also a principal at AQR Capital Management.¹

Michael:

Efficiently Inefficient, the title of your new work . . . you could say it’s new ground, but we’ve all seen where the Nobel Prize was split between Fama and Shiller, and I’m guessing there was a certain inspiration for your title from that. If not, tell me the inspiration.

The mental infection known as political correctness is one of the most dangerous intellectual afflictions ever to attack mankind. It appeals to pseudo-intellectuals everywhere, since it evokes the strong streak of cowardice notable among those wielding academic authority nowadays. Any empty-headed student with a powerful voice can claim someone (never specified) will be hurt by a hitherto harmless term, object or activity and be reasonably assured that the dons and professors in charge will show a white feather and do as the student demands. To a great extent PC is the revenge of the resentful underdog.

Paul Johnson²

Lasse:

Yes, that’s right. So there’s been this long academic debate about whether the markets are efficient, and Eugene Fama, he stands for this view that markets are fully efficient and prices reflect the fundamental value of securities versus prices are inefficient. And Shiller, he sort of represents the opposite point of view, that market prices are irrational, they’re driven by investors’ irrationality and investors push prices far away from their fundamental values.

I have a great respect for both Fama and Shiller. I know them both well, but I wanted to take a different point of view. One that is somewhere between, but also tries to be well-defined, saying that market prices are neither fully efficient nor completely inefficient. But they are inefficient enough that smart asset managers and active investors have a chance to beat the market and be compensated for all the costs they have associated with actually becoming informed about asset prices and trading.

At the same time, market prices are efficient enough that the marginal investor is sort of indifferent about pursuing this active investing, with the benefits being better performance but the costs being transaction costs, asset management fees, and the costs associated with finding a good asset manager. There’s that equilibrium in 
which markets are at the right place, and in between they’re ­efficiently inefficient.

Michael:

George Soros and David Harding are dogmatic about, “The efficient market hypothesis is . . . fill in the blank bad word,” so how did the managers and traders in your book feel about your perspective?

Lasse:

Yes, that’s a great question. Definitely, you’re right that many of these managers, they would certainly not feel that markets are fully efficient. They feel very strongly their job is to beat the market, and they succeed in that job not because they’re just lucky but because markets are not efficient. You’re right, David Harding’s thought is strongly against the efficient market ­hypothesis.

A lot of these guys, at the same time, are sympathetic to this 
view that, yes, there is a [way] to have edge, but you must be very efficient to exploit these market inefficiencies. That’s sort of another interpretation of the title, and I think they’re actually quite sympathetic to this view. They don’t think just anybody off the street can easily beat the market; you have to be really efficient to actually accomplish that.

Michael:

Obviously, you know the managers and their styles, but what about your connection with them?

Lasse:

I certainly knew of all of them, they’re all extremely famous. I knew some of them well. I knew Cliff Asness, he’s my colleague at AQR Capital Management, so he was the one by far I knew the best. I knew Myron Scholes quite well through my meeting him at academic conferences. I also knew John Paulson. He endowed a chair at NYU and I was lucky enough to actually hold the John Paulson chair.

Several of the others I had never met. It was really exciting and fun to meet these guys and get their perspective on how they invest, and try to get an opportunity to ask them exactly how they do it, what motivates them and what are the key drivers of their success.

Michael:

Was there a common theme developing in how they thought regardless of style or technique?

The inability to predict outliers implies the inability to predict the course of history.

Nassim Taleb³

Lasse:

Obviously, they’re all very smart. They are all extremely driven and ambitious. They are all disciplined and clear ­thinkers. In terms of their investment styles and methods, they’re very ­different, but there are some common themes. Even though they use very different words, a lot of them are using trading strategies related to value investing and momentum trading.

Cliff Asness has actually written academic papers and was one of the discoverers of momentum trading in stocks together with a few other academics. He talks a lot about value-momentum. But a guy like George Soros, he’ll talk in very different terms. He will talk about trading on boom–bust cycles, for instance. But when he says he’s riding the boom, in some sense he’s a momentum trader, and when he positions himself for the bust, at that point he effectively becomes a value investor.

When you think about Lee Ainslie, who’s a famous Tiger Cub, he’s done very well in trading stocks globally. Obviously value ­investing and thinking about the quality of those companies is a key driver. And value is also a key driver for a short seller like James Chanos, who thinks about which company is overvalued, which company might be fraudulent or have some little bit of too creative accounting and so on—he wants to short those. And in shorting all value firms he’s basically taking the short side of being a value investor.

Michael:

You bring up a great point about terminology and jargon. I think to the classic phrase that Soros uses—reflexivity. It’s a term with a certain philosophical bent to it. He’s written several books. But a great place where you can take apart some of the jargon and perhaps with your best professorial hat is to describe the different types of momentum. AQR is obviously known for both types. You mentioned Cliff and the type of momentum that he discovered. But time-series momentum, the type of momentum that might be more commonly called trend following, can you compare and contrast the two types of momentum?

Lasse:

What’s called momentum in stock trading—it’s also called cross-sectional momentum—it means looking at the different stocks and seeing which stocks have outperformed other stocks. Then you will tend to go long the stocks that have been outperforming while at the same time shorting stocks that have been underperforming. Then you’re basically hoping that those outperforming stocks over the last, let’s say 6 to 12 months, will continue to outperform over the next month.

That is to be compared with what’s called time-series momentum, which means looking typically at an index, whether it’s a stock index or a currency or commodity, and basically seeing if it’s going up and down. If it’s been trending up over the last year, let’s say. Time-series momentum is looking at markets in isolation, whereas cross-sectional momentum is looking at relative 
outperformance.

The difference is, for instance, let’s say the whole stock market has been going down. In a cross-sectional momentum investment, you might go long stocks that have been dropping less in value. Suppose all the stocks in a given universe have been actually going down. Some of them have been going down less, so those are outperforming. A cross-sectional momentum investor would buy those and short those that drop by more.

Whereas the time-series momentum investor or managed futures type of investor, he would just short all of them. He would say, “They’re all trending down, let me short all of them.”

The more secretive or unjust an organization is, the more leaks induce fear and paranoia in its leadership and planning coterie. This must result in minimization of efficient internal communications mechanisms (an increase in cognitive “secrecy tax”) and consequent system-wide cognitive decline resulting in decreased ability to hold onto power as the environment demands adaptation.

Julian Assange⁴

Michael:

You mentioned the idea these very diverse traders and managers have commonalities. One of the commonalities is gamblers’ ruin, risk of ruin.

Lasse:

Yes, they’re certainly all quite aware of risk management and several of them actually told me about some of their losing trades. 
A lot of them are very happy about their winning trades but they have learned a lot from their losing trades. It’s really important that you learn from your losing trades, but also that you don’t go bankrupt from them.

Michael:

Are there any examples from a particular trader or manager and a losing trade that come to mind?

Lasse:

Yes, certainly. Lee Ainslie talks about one particular losing trade when he had a great investment thesis. His analysis was correct but it turned out the management that were supposed to deliver on these great promises for that particular company, were not capable of doing it. He really learned the importance of management, and has since really paid very close attention to the management of the companies he invests in.

Also Chanos talks about a short position he had that kept moving against him for a variety of reasons. Fortunate for him, he has a position limit. So as the short position moves against him, it gets bigger and this is one of the risks of short selling. When you’re short selling, you lose when the stock price goes up. When the stock price goes up, if you have a fixed number of shares, now you’re actually short more dollars or bigger value and the risk potential increases. Therefore he has a position limit and he would actually reduce his position as the position moved against him and that helped him to limit losses in that particular trade.

Michael:

Most people, if they’re familiar with Julian Robertson and the Tiger Cub story, it’s a fundamental style of thinking and trading. I thought what was interesting was your questioning to Lee Ainslie about his quantitative aspects of his trading methodology, particularly portfolio construction.

Lasse:

Yes, that’s right. He has a number of investment professionals who go out and meet with management and meet with different competitors to try to really understand the industry where each company operates. But at the same time, he told me they actually have a quantitative system to evaluate their positions, the risk exposures, and to help them make the portfolio construction as efficient as possible.

You maniacs! You blew it up! Ah, damn you! God damn you all to hell!

George Taylor 
Planet of the Apes (1968)

Michael:

Looking at all these managers, looking at these traders, their track records, studying their performance, “Is there a best strategy?” There are different strategies for different reasons, but is there a strategy do you think across these managers, these traders, that’s perhaps more repeatable? There’s only one George Soros, there’s only one Paulson. Do you see a particular style across this selection of traders that’s more repeatable?

Lasse:

I think what’s common among these strategies is that there is not this magic effort by this genius, but we can ­actually understand at the core of many of these strategies there is an economic logic behind why they work. But at the same time they require tremendous skill and dedication and hard work to execute them well.

I talk about equity strategies, macro strategies, arbitrage strategies, and for each of these I try to understand what is the basic economics behind why they work, and how you could create a repeatable process that could deliver potential outperformance. It certainly has to be very difficult to do because otherwise these things would be arbitraged away.

Michael:

In your work were you looking for the particular styles or were you looking for the particular manager and trader?

Lasse:

I started with thinking about, “What are the various classic styles” and then for each of these eight styles that I identified I sought to get the most prominent current investor who’s still active. I wrote down these eight names and I thought how am I going to get them to say yes. I started with the people I knew the best, and then when those guys said yes, I had a connection to someone who knew George Soros. Then Soros said yes, and when I started asking the next people it started getting easier when you have Paulson, Asness, Soros, and the like.

Michael:

Something that Cliff Asness brings up, you have as a ­question in your work. You talk about the difference between the real world and the academic world. I like his answer because we can go back to a performance track record, for example, and you can say, “There was a three-year flat period, or there was a ­significant drawdown, but then you came out of it and everything was okay.”

Cliff points out that in the real world it’s not so easy to just look at track records like that.

Lasse:

That’s right, and he talks about when you’re actually going through such a tough period, there’s a time dilation, he calls it, where it seems like time is going much more slowly and it’s really [painful] to have those losses, even if you have greater gains coming out of it. It’s an extremely tough period.

That’s part of the reason these strategies work in that it 
certainly is hard work to do it, but it’s also really stressful and requires a lot of discipline to just continue to do the right thing. When you’re starting to face losses there’s a lot of temptation to panic or do various things that were not part of the original plan, because you think those will . . . be too difficult to handle otherwise.

Michael:

Many traders and managers have to deal with clients and make sure clients are on board with the strategy and understand there can be down periods, etc. Which leads me to bring something up about Warren Buffett.

Many don’t know that Warren Buffett has had multiple 50 percent drawdowns. I don’t know the exact number but if you’re going to produce the type of performance that he’s produced over the years, you have to be willing to take and handle a significant down period. You talk about Buffett and his alpha. You put together significant insights where people can ponder, “Hold on, if Buffett’s a value investor, why is he so much better than everyone else?” You broke it down and pulled apart why he is so much better.

Never underestimate the power of doing nothing.

Winnie the Pooh

Lasse:

Yes, that’s right. We got the data on Buffett’s performance and we also looked at the performance of Berkshire Hathaway stock holdings too. As you know Buffett is running Berkshire Hathaway, and one way to look at his performance is to follow that stock price. But you could also see the publicly traded stocks that he holds. Berkshire Hathaway has to make a filing to the SEC about the publicly traded stocks they own, so we can put together a portfolio and simulate that portfolio at the same time.

The first thing we looked at is just how good is Warren Buffett? How good do you need to be to become one of the richest people in the world? And it turns out his Sharpe ratio is just over 0.7, between 0.7 and 0.8, which surprised a lot of people when we showed them this number, because a lot of people had assumed that he had a Sharpe ratio above 1 or even above 2, as a lot of hedge funds are bragging about numbers in that range. It turns out that he has had tremendously high returns, but he has also had quite ­significant risk, leaving him with a Sharpe ratio which is still incredibly impressive, because it’s about double the Sharpe ratio of the ­market.

For the risk he’s taking, which has been substantial, he’s over the long haul been able to deliver double the return relative to that of the market. It’s really an astonishing performance. Our research doesn’t in any way try to diminish that incredible accomplishment, but when other researchers have looked at it, they have [simply] recognized he’s a value investor. He talks about being a value investor, and you can also regress his return on the returns of being a value investor which is called a value factor.

A reliable way to make people believe in falsehoods is frequent repetition, because familiarity is not easily distinguished from truth. Authoritarian institutions and marketers have always known this fact.

Daniel Kahneman Thinking, Fast and Slow

That factor explains some of his performance, but only a small part, and it leaves [out] a huge alpha that’s a big unexplained part. What we then did is say, “Let’s also regress on the return to what we call a quality investor.” What he buys are not just cheap stocks but he also buys high-quality stocks. He talks about that in his speeches and his writings—that he really likes to buy high-quality companies at a reasonable price.

When we actually regress him on both the standard factor, including value factor, as well as what we call quality minus junk, and also a factor that captures more safe stocks, called betting against beta, then it turns out we can actually explain a large part of his performance, and get us a strategy that looks very much like his performance . . . to explain almost all his alpha.

What we interpret that to be is he is doing what he’s saying he’s doing. Yes, he’s a value investor, he’s a quality investor, he likes to buy stable, profitable, high-quality companies at a reasonable price, and those types of stocks have done well in general. The interesting finding is it’s not just the quality firms that he has been buying that have done well and have been good investments, it’s quality firms in general. Those types of stocks in general have performed well, and he has benefited from being a very early ­investor to recognize that and exploit it. And he’s been able to leverage it through his insurance company in a very clever and very smart way and really perform extremely well as a result of those investment choices and that leverage strategy.

A large fraction of traffic accidents are of the type “driver looked but failed to see.” Here, drivers collide with pedestrians in plain view, with cars directly in front of them, and even run into trains. That’s right—run into trains, not the other way around. In such cases, information from the world is entering the driver’s eyes. But at some point along the way, this information is lost, causing the driver to lose connection with reality. They are looking, but they are not seeing.

Ronald A. Rensink

Michael:

He definitely is using leverage to a degree that perhaps the novice investor might not understand, but obviously he’s doing it in a very smart way—but he’s not a trader devoid of leverage.

Lasse:

Yes, that’s right. He speaks quite negatively about leverage often and certainly there are risks to leverage. If you look at his portfolio and his performance you can see that he uses leverage in two ways. Number one is that you can simply look at his balance sheet and you can see that he has leverage. He has a number of liabilities, he’s been issued bonds, for instance, as one source of leverage, and he has a lot of insurance-related liabilities. And we estimate his leverage is about 1.6 to 1.0. Meaning that for every 
dollar of equity capital he will then invest in $1.60 of companies.

Another way to see that he’s using leverage? You remember I mentioned that we could simulate the performance of his portfolio of publicly traded stocks? We could look at the volatility, the risk of that strategy and then we can look at the risk or the volatility of Berkshire Hathaway’s stock. Sometimes the volatility of Berkshire Hathaway stock is bigger than the volatility of his publicly traded stocks—that suddenly is consistent with the idea that . . . Berkshire applies leverage to the portfolio. If you add leverage to a portfolio, then you will tend to increase the risk and that sort of ranking of the risk is consistent with leverage.

Michael:

It’s just a tool and he’s using the tool wisely.

Lasse:

That’s right.

How should we modify our beliefs in the light of additional information? Do we cling to old assumptions long after they’ve become untenable, or abandon them too readily at the first whisper of doubt? Bayesian reasoning promises to bring our views gradually into line with reality and sync up with the universe. The theorem itself can be stated simply. Beginning with a provisional hypothesis about the world (there are, of course, no other kinds), we assign to it an initial probability called the prior probability or simply the prior. After actively collecting or happening upon some potentially relevant evidence, we use Bayes’s theorem to recalculate the probability of the hypothesis in light of the new evidence. This revised probability is called the posterior probability or simply the posterior.⁵

Michael:

Let me move to a quote from Cliff Asness. “Good quant investment managers can really be thought of as financial economists who have codified their beliefs into a repeatable process. They’re distinguished by their diversification, sticking to their process with discipline and the ability to engineer portfolio characteristics.” For those traders out there today that call themselves more fundamentally driven or value, is there anybody that’s not back testing their world now?

Lasse:

Yes, there’s a lot of discretionary traders that will look at a unique special situation that you might have a very hard time back testing. If you’re an event-driven trader and there’s a new type of event, perhaps you can’t back test that, but perhaps you can still convince yourself that you’re buying something at a discount to its fundamental value. You’re relying on the general principle that ­buying things cheap has worked over the long-term. You might not be able to back test in a more scientific way this type of strategy, or if you’re trading a new type of security and so on. The problem is still the majority of investing is done in a discretionary manner.

Michael:

Do you think guys like Harding, guys like Asness, with this mindset of testing and this quantitative mindset, is it slowly making a dent in the consciousness where perhaps investors in the future are going to say, “We want more than discretionary ability.” Are they winning over the head space? Or is it in the early stages?

Lasse:

I myself am a quant investor. I’m a sworn believer in using scientific methods to evaluate investment strategies. But at the same time I try to keep an open mind. I don’t want to say that one method is better than the other method. You are right the 
more scientific way of investing has made large inroads over the 
last 20 years. But at the same time there is a role for ­discretionary strategies. I personally have great respect for some of the more discretionary traders.

When Chanos, for instance, is finding funny things in the accounting numbers in a way that you can’t really get a computer to do right now, or when other discretionary traders are doing things that cannot easily be codified in a computer because it’s more relying on more soft information or different case-by-case type of analysis. I personally have a great respect for that. So I’m not sure that we’ll gravitate to everybody investing in one way. I think investing fundamentally will have many aspects and it’s driven by people having different methods and different points of view. That’s basically what creates a buyer and a seller. There are all these different ways of doing it.

As the years have unfolded and I have had experience of the seemingly magical phenomenon of trends, my prejudice in favor of this unloved and unheralded investment approach has hardened. To a statistician this is called a Bayesian Philosophy. You start with your prejudices and modify them in the light of experience. I started with a weak belief in trend following because conventional opinion said it couldn’t work. As I saw it work year after year my belief in it has hardened.

David Harding⁶

Michael:

Let’s talk about the role of hedge funds in the economy. The typical understanding is there’s often negative press, especially at chaotic moments. Words like “speculation” will be thrown around in a negative way. But there is a role for hedge funds in the economy and it’s not just necessarily the investors themselves that invest in the particular fund. Some might say, “Why should I be concerned with assorted hedge funds? I’m not invested in these.” Talk about the big picture role of hedge funds.

Lasse:

Hedge funds can play a very positive role in the economy for a number of reasons. First of all, our free market economy basically allocates resources based on prices. If somebody invents a drug, they can cure a particular disease. He may not have enough capital to pursue that cure but he can . . . “this is a very promising idea that others can’t understand” . . . he can then raise a lot of money through the capital markets and then build factories to pursue that idea.

Whereas another inventor who has perhaps a not very useful idea, he might have a difficult time raising money, and therefore the machines and the real capital, the factories, are allocated to the more promising idea. Now that whole notion relies on market prices being relatively efficient. The company with the more promising ideas, their stock price has to be higher. Their market cap has to be higher in order for them to be able to really raise more capital and build more machines.

This is incredibly important for the market economy and hedge funds, and other active investors are tending to buy low and sell high, so they tend to push prices in the right direction and make markets more efficient. Again, that’s just incredibly important. They help to provide liquidity, make it easier to buy and sell for investors, make it easier for people to save for retirement. A lot of large pension funds are invested in these types of vehicles and hopefully hedge funds and other active investors contribute to their performance.

Of course they take large fees and they keep a lot of the benefits for their own sake, too. It’s certainly fair to say that investors should be very concerned about the fees and to what extent they themselves enjoy the benefits of those investment strategies.

Cross-Validation means assessing not only how well a model fits the data it’s given, but how well it generalizes to data it hasn’t seen.

Brian Christian⁷

Michael:

You asked a question to David Harding: “Does your research suggest that it’s better to have the same type of model for every instrument or is it better to have a very specific model for each one?” Talk about his response and your feelings about his response.

When you’re truly in the dark, the best-laid plans will be the simplest. When our expectations are uncertain and the data are noisy, the best bet is to paint with a broad brush, to think in broad strokes.

Brian Christian⁸

Lasse:

I agree with his response that it’s very important to have a very robust model that works across a number of different securities. It’s often tempting to tailor your strategy so it looks like it’s the best for every asset. You might want one ­particular managed futures [industry term for time-series momentum or trend following] strategy for gold, another strategy for wheat, yet another strategy in commodities and so on. If you try to do something like that, you might get extremely good performance in your back test, in your simulation of what would have happened in the past, because you’re able to tailor this. But you might be exploiting a lot of noise, a lot of randomness, and you might end up with a model that works less well going forward in the future, when it really matters, when you’re trading with live ammunition and when it’s really important that your model works.

To avoid that kind of data mining or overfitting of your model, it’s often better to have a simpler strategy. How do you get confidence that your strategy has a chance to work going forward? Well, if the same strategy has worked for gold, not just last year or the last 
10 years, but perhaps the last 50 or 100 years, and it’s the same ­strategy that has worked across all the commodities, not just all the commodities, it’s worked for a bunch of currencies . . . If it’s worked for equities, if it’s worked also in fixed-income markets . . . Then you start to get a conviction that this strategy is not randomness, it’s not just a coincidence of the past that will not repeat itself in the future. Then you start to get some conviction that this type of trend following strategy or managed futures strategy has a chance to work in the future.

Michael:

You are part of the AQR study, “A Century of Evidence on Trend-Following Investing”?

Lasse:

Yes.

Michael:

Were there any elements or insights going through that process, doing that research, doing that study, that surprised you?

Lasse:

It was quite an exciting project because we had been doing a lot of research on what are the best trend following strategies, and showing that some relatively simple trend following strategies actually performed extremely well over the last 20 to 30 years. And they’ve worked on average, not every year or every day, but on average, for every instrument we looked at, which is sort of unusual.

You think that it’s going to work perhaps at the portfolio level, but not necessarily for every instrument in the portfolio. But we found this very consistent performance even at the instrument level. We were thinking, “Wow, this looks really robust, but how can we get even more conviction? How can we get another out-of-sample study?”

Of course, one out-of-sample study is just waiting and looking at the live performance going forward. We didn’t have the patience for that, so we said, “Let’s actually go ahead and collect data. 
Historical data on prices that are earlier than what we looked at before. Rather than looking just over the last 20, 30 years, let’s look at a full century of data.” We were a little bit worried, of course, that what had seemed like such a robust result maybe didn’t work in the 60 to 70 years prior. But to our surprise it worked extremely well, and also historically about as well as in the more recent sample.

Michael:

You talk about the arms race in high-frequency trading. There’s an arms race amongst trend following managers to see how many centuries they can go back! [Laughter] You guys inspired with your first paper. Several others have come right after.

I’m focusing on what can I do for our players so we can have a chance to win the SEC championship tomorrow. That’s really all I’m focusing on. I’ve talked about, ‘Be where your feet are.’ I’m right here, right now. This is what’s important and this is what we have to focus on.

Nick Saban 
Alabama Football Coach

Lasse:

That’s right, yes.

Michael:

One of the things in your conversation with George Soros . . . 
He says that you should not risk a significant part of your capital, but if you have a good run and you’re making a lot of profit, you can risk your profits more than you can risk your [original] capital. That’s a very classic time-series momentum trend following type thinking, a very quant thinking. Did you talk to Soros about the implementation of his strategies?

Lasse:

Yes, he has a different way of implementing, not so quantitative, but he’s also very concerned about the downside risk versus upside risk. What he’s really hoping to find is a strategy or a bet that has a big upside potential and a limited downside potential and then put a lot of money into that bet, like he did on the British pound [see Soros 1992 pound trade].

He talks about [how] he had the courage to really put such an ­enormous position on the British pound, because if the pound broke he would make a lot of money. But given that it was trading in a band there was almost no risk that it would make a large move going against him. And that type of thinking, he told me, inspired John Paulson in the way he sized his subprime bet. I asked John Paulson whether it was true that he got the idea to size the bet he did, the way he did, from Soros, and he confirmed that in that interview.

Much as we bemoan the daily rat race, the fact that it’s a race rather than a fight is a key part of what sets us apart from the monkeys, the chickens—and, for that matter, the rats.

Brian Christian⁹

Michael:

Everyone has to look with deep admiration, because many of the traders that you talk about would not have the latitude to take those kinds of bets. There’s obviously a certain discretionary element. I can use interesting sports metaphors and randy language to describe it, but let’s just say it takes a certain amount of gusto to do what some of those guys [like Soros] do in terms of their big bets, doesn’t it?

Lasse:

It definitely does, yes. And to keep riding at them when they’ve made a lot of money. To not just fold and be satisfied, but keep the bet on until it’s delivered its full ­potential.