Morning! After some dabbling in the genre, FT Alphaville is starting an informal series of simple Q&A interviews.
The basic premise is that we’ll have a hopefully interesting chat with hopefully interesting people doing hopefully interesting things in and around finance, economics and business (sometimes they may be boring people doing interesting things, or interesting people doing boring things as well), and publish a lightly-edited Q&A afterwards. No promises, but we’ll try to do one a week.
We’re kicking things off with a conversation with Guiseppe “Gappy” Paleologo, head of quantitative research at Balyasny Asset Management. We have some ideas for future victims subjects, but feel free to make suggestions in the comments below. The transcript has been edited for clarity and length.
FTAV: Hi Gappy. You’re originally a physicist from Rome. How did you end up in finance?
I came to the US to study for my PhD at Stanford. In the last year of it I had a long internship at Enron, and that immunised me from joining finance for the next 10 years. So I went to IBM Research. After the 2008 crisis, with perfect timing, I decided to enter finance, mostly because of financial reasons. I had kids, the usual stuff. It was not a marriage of passion.
But I went to Axioma, which is a factor risk model provider, and was then hired by Citadel. And surprisingly I had a fantastic time at Citadel, contra expectations. So I came to like what I was doing, and have stuck around since then.
What was IBM Research like?
IBM Research was an incredible place for the first 40 years of its life. I had colleagues who spent, as part of their job, a year in Africa tracking the migration of elephants. I don’t even know why. It’s great to be a research lab attached to a near-monopoly.
There was plenty of research funding, you didn’t even need to ask, and you would just be evaluated once a year by your manager. As a result, there was a lot of really good research that was never monetised.
But in the 1990s, IBM started laying off of thousands because of the transition to personal computers, and that was the beginning of the end. We had to start to recover some of our salary through consulting. It became neither fish nor fowl. It was not fundamental research, it was not applied research, and over the years all the good people left for Google, Yahoo or Wall Street.
I found an interesting paper from your time there, with the title The optimality of the online greedy algorithm in carpool and chairman assignment problems.
That one was with much better mathematicians than me. The carpool problem is actually a beautiful problem that is really a dynamical system that arises in a variety of fields. The travelling salesman problem, the carpool problem, the news vendor problem — they are all beautiful names for classes of applied mathematical formulations. You’re not using the travelling salesman problem for sales, but for chip design.
You’ve worked at a lot of hot places, like Citadel, Millennium, Hudson River Trading and now Balyasny. What are their similarities and differences?
The key to understanding differences in hedge funds is the difference in personality of the founders, because hedge funds are incredibly mimetic organisations. There’s an absolute leader, and the people who report to them tend to mimic the leader. Over time they acquire the same tics and personality traits.
Citadel is in general extremely driven. Ken Griffin must be one of the most competitive people on earth. This obsession, this drive, is a constant at Citadel.
Ken is a big believer in technology, and is a persistent learner. So Citadel fails many times, but eventually gets things very right. And this kind of persistence is a beautiful thing.
How does it compare to Millennium?
In the words of another hedge fund manager, Citadel is like Singapore, and Millennium is like the US. Citadel is small and centralised, and Millennium is large and decentralised.
Someone once told me that Millennium achieved diversification by accident. I don’t know if that’s true, but it has become intrinsically diversified and extremely scalable. It has mastered the art of scale — perhaps at the cost of efficiency. That’s the key to understanding Millennium.
Instead of moulding the portfolio manager in the firm’s way — which is what Citadel does — it adapts the firm to the needs of PMs. And this has in some cases paid off spectacularly well.
I’d also add that Izzy [Englander, Millennium’s founder] is a humble person, and he has incredibly good risk instincts. It’s like some kind of Jedi power.
So what is Hudson River Trading like then? Canada?
HRT is a very different animal. It’s fundamentally a technology company. The technologists are true first-class citizens, which cannot be said about any hedge fund I have been at.
Algo engineers are probably a little more elevated, but a good core developer is highly valued at HRT. Jason Carroll, the one remaining founder at HRT who is still an active partner, was a core developer of legendary skill, by the way. If the most important person at a firm is a core dev it changes the equation.
It’s a very collaborative, friendly place, with a great respect for technical excellence. And they get the tech very right. They’ve tried many things and not everything has worked out. But their technical bets have paid off so spectacularly well. If you ignore Jane Street or Susquehanna — which are really more trading companies than technology companies — then I think HRT is the second-best company [of the big trading firms]. And it has a fair chance of overtaking Citadel Securities.
And Balyasny?
I’ve known BAM since forever. I almost came twice before, but third time’s the charm.
BAM also takes the characteristics of the founder. I once asked Dmitry [Balyasny] to describe the firm in three adjectives, and he said ‘humble, collegial and collaborative’. I think it’s actually true. It’s not into performative kumbaya collaboration. BAM doesn’t use buzzwords. But it is pretty open. And it’s a learning organisation.
Dmitry is at his heart a trader, with great instincts and humility, just like Izzy. As a result, BAM is also quite trading-oriented, and the managers tend to be quite humble, which I like.
I heard that when HRT first approached you that you thought it stood for hormone replacement therapy?
Yes! I had never heard of it, so I Googled it and that was the first thing that came up.
The different styles of quantitative investing and trading can be a bit mystifying to a lot of people. Do you have a rough taxonomy that can help?
The best way is to look at the sources of alpha, and there are really three ways of making money — very broadly speaking — that both quants and fundamental investors use.
The first is risk premia. This is by far the largest of the three. All endowments, all sovereign funds need to allocate to high-capacity fundamental indices or asset classes and seek to get compensated for taking some kind of risk, whether it’s credit risk, liquidity risk, whatever.
The second is informational advantage. It means looking at the same data as everyone else — hopefully, as long as no one is cheating — but by making more informed analysis I can come up with a better forecast of the future. This is at the core of long-short fundamental equities, statistical arbitrage or qualitative credit investing.
The third is arbitrage. At its core, arbitrage is violating the Law Of One Price. It’s a situation when you and I have different constraints or preferences, and someone can exploit this. A good example is exchange-for-physical transactions.
These three classes are not disjoined. There is some overlap. Large firms tend to be successful in more than one area, but most tend to be predominant in one of them.
AQR for example, is primarily a risk premium shop. Two Sigma is primarily a informational shop. Jane Street is an arbitrage shop. HRT is an informational shop, because most of their P&L is in high-frequency trading, and high-frequency trading is an informational strategy.
You have to consider the dominant strategy at each firm and then you have an idea of what the culture is like. Culturally, HRT is not like Jane Street — at all. One is a high-frequency, quantitative, informational strategy shop, and the other is an arbitrage shop.
There’s been a lot of talk about how multi-manager, multi-strategy firms like Balyasny, Citadel and Millennium are growing, while large parts of the rest of the industry stagnate. Do you think this will continue?
In all honesty I don’t actually think the trend is new. It’s just become more conspicuous now and has captured the attention of journalists. But it’s been a trend for something like 20 years. And a good, zero-information prediction is that if something has been going on for 20 years it will probably go on for another 20. I have structural reasons for believing that platforms will only grow in importance.
You’ve written two books on quantitative investing, both of them dedicated to Tofu. I have to ask, who is Tofu?
Tofu is my cat! It’s a British shorthair. Typically I wrote books during my non-competes and my cat keeps me company then.
So your first one is called Advanced Portfolio Management but that one is more of an introduction to quantitative investing, and your second one is called The Elements of Quantitative Investing, but that one is actually more hardcore for quant practitioners?
That’s right. I actually wanted to call the first book a Grimoire of Portfolio Management. But my editor didn’t know the term grimoire, and the rest of the editorial board there thought it had something to do with witchcraft. I thought “Advanced Portfolio Management” was bullshit, but my editor was right. He’s been doing this for 20 years. With the second book, I just liked the ‘Elements of . . . ‘ formulation.
I don’t know. I think grimoire has a bit of extra pizzazz. But why write books about quant investing?
Because I am not a finance person at heart. I don’t care particularly about money or finance, to be completely honest. I am motivated by people, and in my jobs I have had a lot of questions from portfolio managers. So I thought I’d write a short book to explain all the basic concepts to them. The first book is really a love letter to the portfolio managers.
The second book I had always wanted to write. I really believe in the exchange of ideas. I really think we are too obsessed with not sharing ideas. I think you can share a good amount of ideas without damaging anybody, and it’s good for society. I dislike writing. Writing is very painful. But I like to think about problems. And it’s a very basic book. It’s basically a letter to my past self, 10 years ago.
I’ve seen you say ‘boys use copulas, men use volatility and correlations’, in terms of risk management in investing. Can you explain that?
If you talk to a quantitative person who is enamoured with techniques — especially sophisticated techniques — it is a good heuristic to underweight them. The practices of a good quantitative modeller — in any industry, not just in finance — is not to apply the most sophisticated technique. It is to use the simplest technique that works for the problem at hand.
Volatility works. It’s not perfect, but this is not a good reason not to use it. Other metrics seem at face value to be more accurate, but they are often more fragile. Maybe vol is not beautiful, or the final word. But people need simple and common languages to progress.
You’ve mentioned that you’re not really a finance person at heart. If you were starting out again today, would you go into finance?
It’s hard to say. Firstly, it is a sector that is as a whole not growing. Secondly, there are problems that are incredibly interesting in biology, in computer science. Work in AI, work in drug design — even work in logistics. If I was young again I’d work in areas that people are not looking at now.
So if you were going to retire and do something completely different, what would you do?
That’s a good question . . . Unfortunately, this wouldn’t just mean retiring from finance but also retiring from my wife, but I would love to be a monk. Honestly, that’s my ideal job. I’d really love the solitude and contemplation.
Leave a Reply