Sunday, August 17, 2014

Damian Handzy: Physicist and Co-founder of Investor Analytics, Part I

I met Damian Handzy in 1989 when we were both participants in the National Science Foundation's Research Experiences for Undergraduate program at the Indiana University Cyclotron Facility. Damian and his fellow University of Pennsylvania physics major, Alycia Weinberger, used to say, with characteristic physics arrogance, "as GPA tends toward zero, the physics student goes to [highly-regarded Penn business school] Wharton."

So it's a bit ironic that Damian now works on Wall Street. But he did not get there via the route of the failing physics major. After finishing his bachelor's degree in physics at Penn he won a fellowship to do a PhD in experimental nuclear physics. What's more, he completed it in four years -- three years shorter than average for astronomy and physics PhDs.

Damian is the Chairman and CEO, and also a co-founder, of Investor Analytics, whose clients manage over $380 billion in assets. Here he describes what drew him to physics, how he got through grad school in four years, why he left academia behind and switched to the financial industry, as well as his work on "the Street."  His description of the level of mathematical sophistication in the financial industry is one I found quite surprising.

For the physicist considering a career in finance, Damian points out that the most likely viable path will be through a master's degree in financial engineering, and that a bachelor's degree in physics would be good preparation for the master's degree and subsequent career.

For the physicist about to pick a PhD advisor, you'll be interested to read how Damian went about it.

There will be a part II because, after our first interview, Damian realized he had plenty of advice he was eager to dole out. We've done the part II interview and Angela has transcribed it. Look for it to be posted in the near future.

Q: How did you become interested in Physics?

My father was a mathematics/physics student when I was born -- an event that cut his grad school career short. Growing up, math and physics were just what we talked about. I was always good at math because he helped me get good at math. From an early age, I was taught more advanced math than what I learned in school, and by the time I got to high school, I loved it. When I entered college, in the mid-80s at Penn, I was actually convinced I was going to be premed. In my freshman year, I took physics because I liked it, but I knew I needed to take a bunch of chem and bio. I was not looking forward to the bio part. Chemistry, I thought, was just applied physics. I don’t believe that anymore. It was two nuclear physicists, Fay Ajzenberg-Selove and David Balamuth, that made nuclear physics the absolute most exciting thing I could possibly do, and I was hooked. 

 Q: You went on to do a Ph.D. in Nuclear Physics in Michigan?

Yes, at Michigan State, I worked on heavy ion collisions.

When I got to grad school, one of the first things I did was email my old undergrad advisor, and I asked him, “What are the most important things to think about when picking an advisor for my Ph.D.?” He gave me a list of practical things. I actually went and interviewed, I think, every single nuclear physics professor in the department. I literally sat down, made appointments with everybody, and said I want fifteen, twenty minutes of your time, not more than that. I introduced myself, told them what I’m interested in, and wanted to know about their research and whether they’re looking to get an additional student. One of the things I said was that I wanted to be done in less than four years. I had a bunch of people looking at me like I had two heads. Two of the professors did not kick me out of their offices.

Q: Why did you want to be done so quickly?

I had a fellowship that gave me a little extra money, and it ran out after four years. I decided that the last thing I wanted was to be a fifth-year grad student suddenly on less money than I had for the first four years. I would just feel really bad about that. It was a good motivation for me to finish. At that point, I was absolute: I was going to do my postdoc, probably two, and eventually be a professor someplace. I said, “If I can be done with grad school sooner, why not?”

One of the guys who did not kick me out of his office actually said, If you’re really lucky and you do really hard work and you’re good, you could be done in four years, but even if you’re really good and lucky and work hard, there’s no guarantee.” I accepted that as a not no. He said, “Look, if you’re really seriously considering asking me to be your advisor, you need to know, I’m about to be named Lab Director. That means I’m not going to have a lot of time for you, and you need to know that.”

I thought about it. I said, “Well, basically you just told me that you’re going to show me how to run a 400-person organization by example, and I’m going to get to ask you questions about how to run something really big. I’m going to learn serious management. For the most part, I’m going to be on my own, you’re still going to be able to interact with me, so I’m going to learn from you, but I’m going to have to learn by doing, so I’m going to learn it better. And, after four years, you’re going to have built a lot of political connections all over the physics community as Lab Director, and it’s going to be much easier to get the job that I want. Absolutely, this is exactly who I want to be my advisor.”

Now, I say this a little tongue-in-cheek, but I actually said all three of these things. It’s going to be harder on me, you’re going to help me get a job, and I’m going to learn how to manage people.

Q: That’s a very savvy approach for a beginning graduate student.
I’m going to give credit to my undergrad advisor. Some of the things he said were to make sure that the professor has at least one student who is more senior than you, because that student is the one who’s really going to show you the ropes. For me, that was Mike Lisa, who is now at Ohio State. My advisor also said to make sure that the professor actually graduates students. You don’t want to be there seven years, eight years. You want to move on with your career. Make sure there actually is a thesis project that is viable in the time frame. You’re not going to be starting your experiment four years from now. And, he said that tenure is a touchy topic, because you don’t want to risk your advisor leaving after two years to pursue tenure somewhere else. You almost have to pick someone who is tenured. He did say that the non-tenured professors need students. That’s part of the community, and everyone has to participate in that.

At the time I entered grad school, I was 100% I’m going to be a physics professor. So, the transition came about in my second year. I remember walking to class with one of the other grad students, and he had a copy of the Wall Street Journal under his arm. I remember teasing him, but I was quite serious, why would you ever read such drivel? There is no intellectual content in that; it’s all man-made. There is nothing to be discovered – it’s all money.

Within a year of that statement, I found myself reading finance papers to try to understand how that entire industry works. I had realized that there were a number of things about the route to being a physics professor that didn’t jive with what I wanted out of a career. One of those is geography. I’ve always lived in the Northeast. I knew I wanted to work someplace within a three-hour drive of New York City. That really limits what you can do if you’re a physicist.

What attracted me to physics was solving really hard problems and getting that euphoric high when you understand something that you didn’t understand before. I love that. As an undergrad, you don’t tend to work on problems more than a week or two. What I found was, professors spend a lot of their time managing grants and grant proposals and dealing with very restrictive budgets. And, you don’t get to choose where you’re going to live. I just said, quite frankly, life’s too short. These were things about physics that were pushing me out of professional physics.

There’s one other significant one: I’m selfish, in that I wanted to see the benefits of my labor in my lifetime. I absolutely subscribe to the fact that the world needs more scientists. There’s a piece of me that is in perpetual guilt that I did not continue promoting the body of knowledge of pure science, of pure research. That, to me, is the thing that is going to continue to make humanity better. Medical advances, lifespans, our intellectual curiosity about how the world and universe works, all of those things. That is the thing that drives long-term growth for everybody.

Think about it this way: in the 1890s, J.J. Thomson discovered the electron. Here we are, 120 years later, and the entire world works because we can manipulate that phenomenon. 120 years. The progress that humanity has made in the last 400 years since Galileo picked up the telescope is absolutely astonishing, as compared to the lack of progress in the 30,000 years of civilization before that. Science is the key to unlocking everything.

Q: You mentioned wanting to see the fruits of your labor in your lifetime. Why don’t you talk about what you see these fruits as? What is really satisfying to you, either those you’ve seen or those you hope to see?

Today, I run a medium-sized company that specializes in quantitative risk management for people who run very large portfolios. That means state pension funds – pension funds for state employees, teachers, firemen, policemen, what have you. That means corporate pension funds for people who have worked at large corporations for forty or fifty years and get pension when they retire. The people who manage that portfolio might use our services to understand the risks that they have in order to make sure they have enough money to pay the pensioners. That’s the bottom line. Our clients are also mutual fund managers, hedge fund managers, people who are managing other pools of money for other types of constituents.

I got to Wall Street in ’95. There were a lot of physicists who made the transition right around then. It wasn’t a phenomenon yet on Wall Street, so we were still all kind of unique birds, but Wall Street hired a lot of physicists and mathematicians because derivative investment was going through the roof, and they needed people who understood these things. Today, they have a whole bunch of courses in Masters programs called Financial Engineering, which essentially is what I did, after hours in the early ‘90s, educating myself on finance.

After almost 20 years on Wall Street, I’ve come to the conclusion that economics is in the state that medicine was in the 1700s: belief in humors, belief in leeching. What I mean by that is that economic theory works enough of the time that people aren’t willing to throw it out. But it is blatantly wrong so much of the time that it would never get published in a physics journal. There is very little experimentation to justify the theory. The theories are justified because they are consistent with previous theories, which are grounded in only a little data. Math has been applied to the economy, and statistics have been applied to the economy, as if we understand the causal agents and what’s going on underneath. The fact is, we don’t. No one does. There’s a new branch of science called Complexity Science. It’s an outgrowth of Chaos Theory that studies beehives, ant colonies, cities, traffic patterns. One of the hallmarks is it has plenty of feedback loops, lots of nonlinearities, but you cannot describe it with an equation. There is no equation of state for this stuff. It evolves, it is highly interactive, it’s adaptable, and through time it interacts with itself. The economy is such a system. Applying math as if there was a deterministic system to it is just the wrong approach. That is essentially what modern economics attempts to do: throw a bunch of math at the problem and pretend that we understand how things are going to work. What I mean by that is this: imagine how much more difficult physics would be if protons had personalities

That’s one of the reasons I was attracted to physics. I liked that you could actually know what you know. When you make a trade, someone buys a stock and someone else sells it. For the most part, the person selling the stock is selling because they don’t think it’s going to go up any more. They have a view of the first derivative of the price, and it’s negative. The person buying the stock has a view of the first derivative, and they believe it’s positive. Well, that means you have two parties that fundamentally disagree about the direction, just the sign of the measurement of the first derivative, never mind the magnitude. That’s what the market is all about. That’s happening all the time. You’ve got a lot of different opinions from people analyzing these things, and I don’t mean stupid people. I mean smart people who are trying to figure it out, and the market exists because enough of them disagree. Now, add into that the fact that there are a lot of phenomena around what the pricing might be, around what actually causes volatilities to move, what causes prices to move. There are an enormous number of legitimate scientific studies pointing out that the causal agents are not what the assumptions were. For example, efficiency in the market is a buzzword around here, and has been for about sixty or seventy years. The notion is that volatility exists in the marketplace because of inefficiencies. Meaning, I have information and you don’t have information; that’s inefficiency. We might trade and the price might move a lot because information flow takes time and because the markets are inefficient. That’s a testable claim. You can strongly argue that information flow is much faster today than it was in the nineteen-twenties, thirties, forties, fifties. Everybody has access to everything right away. So, you would therefore expect, in general, that the overall volatility of the markets should be much lower today than in the twenties and thirties. Blatantly false. It’s much higher today than back then. That contradicts one of the core tenets of modern economic theory. And yet, we still use all the results of the modern economic theory to analyze the rest of the market even though we know that the tenets are false. Imagine the theory of gravity where you have to recalculate G every day.

The truths they discover are locally and temporally true, but there are no macro truths. There is a well-known economist out of MIT named Andrew Lo, and we’ve had a couple partnerships with him where we’ve implemented some of his ideas about how to measure things. He has this wonderful quote that says, “In physics, there are three laws that explain 99% of all phenomenon. In economics, we have 99 laws that explain 3%.” That really captures it.

Q: Would you say that you discovered a very rich intellectual territory with difficult problems and a real practical benefit from addressing them?

The answer to that is yes, I have. Now the disappointing part: you don’t get paid on Wall Street for understanding how things work. That is part of the frustration of this career. I don’t have the luxury of spending my days thinking about how stuff works or testing out theories or trying to gain that knowledge. On Wall Street, you are paid for producing something that someone else wants and maximizing the revenue for that. Because that’s the nature of the environment, you have to participate in the game or you don’t have a job. What I’ve attempted to do is bring a unique perspective. It’s a minority viewpoint. Back to the pension funds, so few people understand some of the analytics they use on a regular basis that they use them even when the analytics are completely and totally wrong. 

I’m going to give you a simple example. There’s something on Wall Street called beta. It’s the slope of a regression line between two different sets of data. You make a scatterplot, find the linear regression between them and it’s the slope of that line. When the scatterplot is the returns of your fund, take one of your investments and the returns you made every month, as compared to the returns of the S&P 500, or any other benchmark index that you want to compare yourself to. The notion is that if the slope is greater than one, you have a greater volatility. If the slope is less than one, you have a lower volatility. You can show that is true if and only if the correlation between those two data sets is equal to one. They drop the “if and only if” part. They just drop it. Nobody, except people really steeped in science, even knows that that’s part of the whole process. It gets simplified down to the point where people who manage tens of billions of dollars of pension fund assets do not understand that when the correlation is below one, the slope of the line is less and less useful. Imagine you’ve got a pension fund that benchmarks itself to some index. The correlation to that index is dropping lower and lower, and they want to hedge some of the risks. They’re using that beta number to figure out how much hedging they should buy to hedge the risks in that portfolio, meanwhile beta is a completely meaningless number. They’re spending an enormous amount of money on something when the analysis is devoid of content.

Q: Why doesn’t that get punished?

It does get punished in a market fashion, except that their peers, the other pension funds, are behaving similarly. In this industry, it is not an absolute game. It’s a relative game. What I mean is, during the financial crisis, the US equity market, the stock markets, were down 45%. The average pension fund lost something like 40% of its assets. Everyone was down 40%. Nobody got fired, because you didn’t do worse than that other pension fund down the street. There’s a herd mentality. It simply doesn’t get punished because of the way the system works. The people who get punished are the pensioners, not the people managing the pension funds.

Q: You’re aware of these discrepancies between practice and reality, and that must be part of the reason why your company exists and why you’re successful.

We appeal to that sliver of the marketplace that actually wants to do a better job. Of course, I can’t say categorically that no one wants to do a better job. There are some who legitimately want to understand how to do this better and get better at it. That’s the part we appeal to. Here’s the other edge of the sword, the intellectually honest sword, if you will. It’s a very powerful tool in the sciences to have the peer review, to be intellectually honest, because the process improves. Everyone knows that’s the way to play the game. We have a number of competitors who also put out analytics products, but when it comes to risk, we are the only firm to my knowledge that actually publishes the goodness of fit measures, the confidence intervals, anything like that. That is a challenge. It’s a detriment to us. We’ve actually lost business because the interpretation is not “These guys are really good, they get it, they’re showing me stuff.” No, the interpretation is “Your models don’t work?” Don’t think about having conversations with other physicists, think about having conversations over Thanksgiving dinner at a not-physicist’s house, a relative’s house, a friend’s house, not in the sciences, and having this conversation. That’s kind of how people interpret this stuff because they don’t have the physics training. They have the finance training.

If you ever pick up an economics textbook, you will never see an error bar in it, anywhere. Economists, finance majors, and anyone who goes to business school don’t know the concept of an error bar. There is no concept of uncertainty in any measurement. How do you even begin to have a conversation with somebody about measurements if you can’t have that conversation? My transition from physics to finance was very rocky in the first year because I spoke physics, but I did not speak a word of business. I took a job at a consulting firm called Deloitte and Touche. Here I am, defending my thesis, and two weeks later I am literally in the senior executive office at Metlife, biggest insurance company in New York City, and I’m talking to senior executives of how to use the Internet to change their business. Two weeks. What the hell did I know of how to change a business? But I knew how to use email, I knew how to write HTML, so I was declared the firm’s Internet guru. This is how it happens. I spoke so little business, that when someone in that meeting mentioned the IMF, my immediate thought was, “Why are they talking about intermediate mass fragments?” Not International Monetary Fund, or anything to do with finance.

I also learned early on that if I’m going to talk about the concept, I shouldn’t use the term “error bar,” because immediately people think you’re making mistakes. My first boss almost fired me. When I say my first year was rocky, it was rocky. I now call them confidence intervals. I try to explain the concept to people on the street using personal examples. I’ll talk about the fact that anyone who owns a house doesn’t really know it’s worth. We have an estimate of what the house is worth, but it might be 5% lower than that, it might be 6% above that. We can’t guarantee it to a thousand dollars or ten thousand dollars. It fluctuates. Okay, that’s an error bar. They’re like, “Oh, yeah, yeah, I get that.” But then they don’t apply it to anything else. It only applies to houses. There are a lot of challenges in that regard. There is some satisfaction, in that I actually am educating people about better ways to think about things, and better ways to interpret things. That’s where the satisfaction comes from. But there are very few people who think like physicists on Wall Street. People do not think statistically here. It’s just a very different mindset.

 I’ll tell you a story about a very, very large pension fund. It was an international organization, a multibillion-dollar pension fund. What we were doing was collecting every one of their investments. They had investments scattered across a couple thousand different mutual funds, hedge funds, et cetera, and we gathered, on a daily basis, what securities do they have in that account, in this account, in that account, and then doing the analysis, aggregating all that up, and presenting them macro-level results of what they’re sensitive to and what risks they may have, et cetera. Of course, it didn’t start with complete transparency. Many of these funds said, “No, we’re not giving you anything,” so data collection became a problem. As we were going through it, we were actually having pretty good success with who was willing to share information with us. After about a year, the guy who ran the entire pension fund for this organization called me into his office to say he’s canceling the contract, because he’s decided he doesn’t want that level of information. He only wants to analyze them from a very opaque perspective of what returns they post every month, rather than what investments they make every day. I had a look on my face, kind of like what you do right now. Why would you ever give up detail? And his justification was very simple: if I have that level of detail, then I am ethically obligated to do something about it and avert disasters. If I don’t have that level of detail, I can’t be blamed. He actually said that to me.

Besides the fact that I lost a big account, which, of course, is not a pleasant thing to hear, I’m hearing that the justification for that is, “Damian, you’re trying to reveal truth, and I don’t want to hear the truth.”

 Q: I heard maybe a year ago that you were advising people who manage $380 billion worth of assets. That’s a lot of responsibility.

Most of that money is pension money. Less than half of it is in private investments, hedge funds, money I’ll say is invested on behalf of people who can afford to lose it. The pension funds, on the other hand, are literally the garbage collectors in every town, the paramedics, police officers. These people don’t get paid a lot of money to begin with, and they are dependent on that pension money so they actually can retire and not sell their house and move into a cardboard box. There is a huge ethical responsibility. The smartest people on Wall Street work for hedge funds. Hands down, categorically, there is no question about that. Hedge funds attract the best and the brightest, they pay the best, but you have to be very risk-tolerant to work for a hedge fund, because lots of them go out of business in a given year. But the truth is that the intellectual power is where the money is, where the payments are, where the high salaries are, and the high stakes. That’s hedge fund land, where most of the investors are very high net worth individuals or corporations. I have intellectually challenging conversations with my hedge fund clients on occasion. I am trying to get my pension fund clients over the very basics of how to understand this stuff. 

Q: If I sent a young person to you who has been studying physics and is now interested in finance, what would you tell them and what would you ask them?

I mentioned before that today there are departments of Financial Engineering at universities. In the ‘90s, when I made this transition, physicists and mathematicians were just being introduced to Wall Street. Fast forward twenty years, and some of those people made spectacular careers for themselves. One person in particular, a theoretical physicist from Columbia. He joined Goldman Sachs and started developing some volatility models for them. He made them a boatload of money, so they paid him a boatload of money, and instead of retiring, he was invited by Columbia to start the country’s first school of Financial Engineering. Emanual Derman left physics and went to Wall Street, made phenomenal success for himself, and now has been running this school for the last fifteen years, where essentially he churns out Masters students who understand mathematical modeling in finance. Top-notch Masters students. A lot of other schools compete with them, so they now churn out Masters in Financial Engineering. There is now an entire process by which one enters Wall Street as a “quant,” and it’s no longer through physics or math, but through Financial Engineering.

The bad news is that there are not nearly as many opportunities for physicists right out of physics to come to Wall Street and do what I did twenty years ago. I was lucky enough to catch a wave. I was in the right place at the right time. To be competitive on Wall Street against those people, you actually have to learn finance and modeling the way they have learned it, because they’re learning to speak finance and math and analysis all at the same time. A physicist who comes to Wall Street now looks more like a fish out of water. If you want to make the transition, if you are a Ph.D. or a postdoc or a young professor who’s decided to go into finance, unfortunately the route is now going through one of these programs, or at least being conversant that way. I’m saying that with 90% confidence, so some will be able to come in and get a job. There’s a polish that Wall Street expects, and a way to communicate. You’re just much more successful if you have it, and the cards are stacked against you if you don’t.

Q: After earning an undergraduate degree in physics, you could then go get a Masters in Financial Engineering?

Absolutely. And I would say that thinking like a physicist, not like an economist, is very important. I think an undergraduate degree in Physics prepares you very well.

Q: I know family is very important to you. Could you speak to balancing the competitive industry you’re in with life at home?

When I was in grad school, the typical week was a 70 or 80-hour week. Most Saturdays, 12 or 14-hour days were the norm. And not just for me, because I was trying to get through fast, but I saw a lot of professors doing that. I saw 40, 50, 60-year-olds putting in those kinds of hours, and I said, “Wow. Do I want to be spending my life that way?” If you’re going to be successful in any career, it doesn’t matter what industry, you’re going to be spending about that much time. I’ll say my commute is much longer, because I live in New Jersey and I have to get into Manhattan. I’ve got three kids, so I don’t live in Manhattan, which is prohibitively expensive. So, I’m putting in similar hours, but the nice thing about Wall Street is weekends are sacrosanct. Nobody works Saturdays and Sundays. Let me pause – investment bankers work Saturdays and Sundays. But, that’s also a very high burnout job. Investment bankers put in 80, 100-hour weeks, and that’s typically 25-year-olds right out of M.B.A. programs. You can do that for three, four, five years, and they’re getting paid oodles of money to do that. I really do mean that. The starting salary in that kind of a profession is $300,000 a year. You’re able to stock away a lot of money over four or five years, and then you will do something else afterwards, because you just can’t keep up a 100-hour workweek pace and survive. That’s one subpart of the industry that is known for excruciating hours, but people only do stints for a couple of years there. For most of the street, weekends really are your own. That’s kind of how I look at it. Monday through Friday, I am in total work mode, but tonight I’m going to go home, and probably make some sangria and sit down with my wife, and play some sports with my kids. That’s what I’ll do all weekend. It’s not a bad work-life balance.

Q: Is there anything you wanted to add?

Sometimes I’m asked if I used my physics background in my career, and I kind of look at the person and I say, “Well, since I only have a physics background, I have no other background on which to rely, so of course I do.” I could not imagine living my life not thinking like a physicist. There is a richness to this way of doing things that I am so glad I was trained in. The undergrad experience was a very important part of that, and going through the Ph.D. program changes the way you think. It requires you to learn how to think in a very clear, both theoretical and practical, way. You learn to understand where common mistakes are made, and how deeply you need to push in order to really understand something. Very, very few people have the luxury or the benefit of being able to think like a physicist. I’m actually thrilled that I got that. That’s another reason I promote all that stuff about science on Facebook. That way of thinking really lets you understand stuff deeply. Now, I’ve attempted to apply that. One of the reasons I, personally, think I’m successful is that I’m able to bring a viewpoint to my clients that very few other people are able to bring. Regardless of what field people go into, if they’re able to think like a physicist and keep thinking like a physicist as they go through their careers, they can make a name for themselves and establish a reputation that can bring a lot of value and a very rewarding career.

No comments:

Post a Comment