James de Castro Head of Equity Linked Trading Merrill Lynch Hong Kong
Team Latte's Rahul Bhattacharya recently caught up with James de Castro, Head of Equity Linked Trading at Merrill Lynch in Hong Kong and had a chat with him. Here is an excerpt of that chat.

Team Latte :
Of late, over the last few years we have witnessed that there has been a change in the volatility landscape in the equity markets; volatility itself has become quite volatile.
How do you capture such volatility of volatility (or vvol as the traders call it) in any pricing model? What trading strategies and new products can be designed to take advantage of such changing volatility? OK, more precisely put, how exactly are the banks trading volatility in Asia and what products are they buying and selling to capture the vol.

James de Castro:
Pretty much anything with high vol convexity will help capture that. There is considerable debate as to which mathematical representation of the vol surface is most correct or appropriate, but if a trade shows significant vol convexity even on a simple flat vol model it is fairly safe to assume it will have high vol convexity in most reasonable vol surface models too. Generally things with out of the money implied strikes, forward starting resets, out of the money knock in or knockouts, etc.... will have vol convexity. Without going into ¡§exotic land¡¨ even a humble butterfly, put spread, ratio put spread etc. will have convexity as well and have uses in a changing volatility environment.
If you really need to understand how the banks are trading vol in Asia all you need to do is to look at their books. You'll find that the bulk of the institutions here have a reasonably easy way of buying volatility from retail structures (structured notes) because every retail structure in Asia is practically a combination of buying volatility and buying dispersion and then they have a problem at the end of the day that they have too much vol on their book so they need a way to sell it and what many people do is that they try to do warrants. So if all works well you sell the vol through warrants and you buy the vols through retail trades (structured notes) and everything is working fine. The problem is that typically you are a lot better at one than the other or the market at any point in time tries to favour one over the other and then you end up with an imbalance. So the type of vol trading that you do in Asia depends on what kind of product cycle you are in.
One thing that has happened in Asia in the last few years is the Variance swaps. But trade levels and the degree of sophistication is nowhere near what we see in Europe or the U.S.

Team Latte:
Why is that?

James de Castro :
The markets are lot less complete and the transaction costs are high as well. And you have far less developed single name derivatives market. In most markets you just get the index to trade.

Team Latte:
Derivatives trading (and investment strategies) can be broadly, and somewhat loosely, categorized into being either long volatility or short volatility. Many practitioners, including some very respected traders, think that no matter what asset class we trade and what instrument in that asset class we trade, at the end of the day it all boils down to either buying the volatility or selling it. Recently, a trader joked and said that there is only one asset class in all markets and that is volatility.

1. Would you agree with such a categorization?
2. And if it is indeed true then do we have the necessary tools, techniques and processes to measure and model volatility in our pricing and risk management engines?

James de Castro :

1. There is something in the assertion in that ultimately you typically needs things to move for you to make money buying or selling them, so volatility is a very important concept. However, I would not totally agree that it is the only asset class, particularly for less developed markets. In many Asian regulatory regimes getting the right structure for something as simple as going long or short delta can already get you or lose you a lot of edge and bring about or eliminate a lot of risk before the market even moves. In that sense even "liquidity" could be said to be a bit like an asset class too, and regulatory risk is just as risky as the market risk which volatility encapsulates. The market doesn't even need to move for you to lose money on the wrong tax structure.
2. Generally as an industry we keep getting better but we are not fully there. A proper, fully consistent model for asset price behaviour and volatility still remains a work in progress for the industry. We get closer every year, but we will probably never fully arrive, and when we think we have the markets will change their behaviour!

Team Latte:
In many of our discussions with traders and risk managers in Asia we find that there is somewhat of nonchalance towards capturing the risk reversal of an options book. Many trading desks do not use a full blown price-vol matrix to capture the risk reversal (changing vegas) and vega convexity, which in our opinion is an essential tool to manage the risk of an options portfolio. As we know very well from history, at least one very reputable trading desk lost a very significant amount of money in the mid-1990s from trading Japanese equity derivatives (warrants) when the Japanese stocks unexpectedly plummeted in 1996 exacerbating the already large negative vegas and convexity positions of these desks.
Generally speaking, would you recommend traders and risk managers to pay attention to the risk reversals and vega convexity of the book in Asia ? Do you think this is a robust risk management tool? Please comment.

James de Castro :
Generally yes. Things like risk reversals and vol convexity are almost model-independent and thus quite robust risk measures. In Asia you ignore the tails of the distribution at your peril. Likewise the centre of the distribution can be very deadly too when things stop moving like this past summer. If there are only incomplete option markets things quickly get expensive if you are long vol and convexity at the wrong level.

Team Latte:
Is there any kind of derivative product that you have found very challenging to trade, hedge or price in your entire career so far?

James de Castro :
In the early days on type of product that caught everyone in the street, including us, off guard was the reverse floor cliquets, where you receive a minimum of 100 and a maximum of a large number, say, 200 or something minus a series of monthly cliquets,. Traders were surprised because they had not done enough work on volatility convexity practitioners hadn't been accounting for the fact these products were pathologically volatility sensitive. Even if they were not volatility sensitive to start with if you sold 5 vols then they suddenly became very pathologically volatility sensitive. Besides these products were not only volatility sensitive but also involved an enormous amount of forward starting options which were extremely difficult to hedge. In the early days, a lot of multi-asset path dependent options were also very challenging because pricing of those products were a daunting task, unlike today when we can easily run Monte Carlo simulation to price these products.

Team Latte:
You have an undergrad degree in Physics from Oxford , do you think that gives you an edge in today's world? I mean, the fact that you have a quantitative science degree which perhaps allows you a better perspective on products and markets. Is that the case?

James de Castro :
Yes, it does. It's a different way of thinking. Economists are very concerned about their particular model of reality whereas people with physics background would be looking at it from a very different angle, trying to give a mathematical structure, such as trying to solve problems using the principle of symmetry. When I look at a problem I try to solve it using symmetry which could be much faster, of course, I could get lazy. On the other hand an economist would try to solve the same problem using first principles and could be stuck up for hours and not get to crack it. I find a lot of tools and processes from Physics are very helpful. The concept of symmetry is one that I keep using all the time and it is very helpful, it is a very profound concept and the Economists are not trained to think that way.

Team Latte:
Can you give an example of symmetry?

James de Castro :
A very typical example of symmetry is why does the price of a Black-Scholes barrier option looks the way it does. In principle the Black-Scholes is a linear partial differential equation so therefore any linear combination to that equation will be a solution itself. Now if you take the case of a knock-out barrier option, its value is known to be zero at the barrier and you can achieve that zero value by putting an identical option, by reflecting it above this mid point....putting an identical option on the other side and subtracting it. For a Physicist that is a very natural way of looking at things like putting an electron in front of a metal plate, you imagine an electron on the other side of the plate and that sort of a metaphor carries over very well in quantitative finance or option theory. An economist does not naturally think that way because they have not been trained to think that way. Its not their fault, it is part of their mental toolbox when they arrive out of the university.

Team Latte:
Given your predisposition towards quantitative sciences, especially Physics, if you were to choose between two candidates for an option traders job, one with a Physics or Math background and the other with Economics or any other Social Science background, would have a preference for the former?

James de Castro :
Generally, I would tend to find out which of the two is a better lateral thinker and more imaginative and quite often the Physicist or an engineer turns out to be more imaginative. But that is not always the case.....when I left Japan I left the Nikkei book in the capable hands of a fellow with a degree in History. So, it doesn't always work that way. But I have found that generally people with technical background better at this.

Team Latte:
But won't you agree that there is too much importance on math modelling these days amongst the practitioners as well as the "Ph.D.-turned-quant traders". The art of intuitively understanding concepts and product payoffs, distilling the complex math out and retaining simple English like explanation of derivative instruments seems to be missing. I once knew a trader who would routinely ask trainees to explain Black-Scholes in plain English during interviews. But these days, Black-Scholes is Ph.D. stuff. Do you think this is the case? And if so, then being grounded in "math modelling" is counter-productive to being a successful derivatives trader? Would you want to see most of your trainee traders or structurers become ultimately become math modelers?

James de Castro :
I think a good quantitative background is certainly very helpful and over the years I have hired many successful traders with that kind of profile. However, I think when carried too far it can lead to ¡§analysis paralysis¡¨ and focusing to much on optimizing third and fourth order effects when the nice first order arb is staring you right in the face. In Asia the markets are often very incomplete so the extra accuracy is often much smaller than the bid offer spread you can achieve even just on delta. On balance I weigh a solid grounding in reality and markets and the ability to conceptualize in simpler terms just as highly as the more advanced skills. The one exception is advanced exotics where the field has certainly become a lot more technical over the last few years and you need both a quant and a well grounded trader working together to have an effective team, but even there you should remain wary of any trade so complicated that it cannot be distilled into a few simpler tradeable concepts. And finally, regulatory knowledge is always good to have as well!

Team Latte:
Well, talking a bit more about recruiting fresh graduates and trainees, a task, we believe you take very serious, say you are faced with ten fresh graduates who have applied for a job as a trainee trader in the equity derivatives division. Each of them is highly qualified, such as having Ph.D.s or MBAs from top schools and each has great communication skills. However, you would have to choose just one person. What attributes ands skills will you look for in the candidate? How will you go about making the selection?

James de Castro :
I would be interested in finding out about the candidate's personality in what relates to work - work ethic, drive, enthusiasm, and curiosity by reference to for example what the candidate might have done in the past. I would also be very interested in having a good personality fit with the rest of the team, easily investigated by doing interviews with other team members. And I would finally be very interested in the candidate's ability for logical structured thought and for lateral thinking when confronted with unfamiliar circumstances. I always look for interesting logical reasoning puzzles to try out on candidates. It has been my experience that the Asian marketplace will keep throwing up unexpected surprises on a regular basis, so lateral thinking and flexibility are essential to survival. Checking out those three characteristics should help to narrow the field down pretty quickly.

Team Latte:
What excites you Jim? What makes you tick? What keeps you going? You are in a high powered job with cutting edge models and trades, complex markets and products and always surrounded by young and fresh traders and quants. Surely, you need to be on top of things and as we all can see, you are on top of things. How do you do it?

James de Castro :
Finding new things, I guess. I am eternally curious so if I don't learn something new in a day, I find it less exciting than other days. And then solving problems also gives me a sense of great accomplishment. There is a challenge there and you find a way around it. That keeps me going as well. And then the markets, have become so challenging and dynamic of late that I am fascinated by it every working day. There is always something happening one day and something diametrically opposite happening the next day.

Personal Profile
James de Castro has an M.A. in Phyics from Oxford University and has seventeen years of work experience. He is an options trader, a lateral thinker, he loves classical music.

Any comments and queries can be sent through our web-based form.

More Asia's Brightest >>

back to top