A Day in the Life of an Exotic Derivatives Trader

Abdel · 6/1/11

Pretty intense. May I ask, what salary this guy must be making if he's working his a** off !?

pandyajwalin · 6/2/11

this guy is thinking of an MFE after such an experience on a daily basis!!
Was wondering- How much time will it take a fresh MFE grad to do the kind of work he is doing!!!

Author · 6/6/11

There's a ton of thinking that goes on all day long. Between pricing products and hedging dynamically, we spend all day discussing our positions, deciding which ones we think will perform well and which won't. It's not trivial to just "get greeks" - these greeks very often represent enormous, multi-million dollar risks, some of which we like and some of which we don't. It's our job to sort out the good from the bad and work out how to make sure our position is made up of all the ones we think are good. We talk about how to improve our positioning in order to reflect our views of the market. In order to come up with views in the first place, we of course glean insight from seeing directly the various flows and where they come from, but we also spend a great deal of time pouring through data, charts, correlations, backtestings, etc. We have a ton of tools that we've made over time that can grab relevant data for us, and we have tools that can come up with a model price for us, but when it comes time to make a decision about a market view and how best to express it, and when it comes time to turn a model price into a dealable rate in a large size, those are very much human decisions. As traders, we're expected to be experts in translating what machines do for us into real life. Our models are all limited (read: wrong) in some way, and so while they're great in getting us the majority of the way to a price, if we just dealt off the model and hedged out the exact greeks of the model every day, we'd lose money. Further, once we deal these products, it's up to the traders to decide how to go about hedging them. There's not just one way to hedge. Sure, there's probably a standard way to hedge each first generation exotic, but the trader knows fine alternatives and can decide which is most appropriate in current market conditions. The trader knows he can get option X done more cheaply than option Y because he knows how similar things are trading in the market, the vast majority of which is not computerized - and even it it were it would be exceedingly difficult to give a computer the judgement it takes to make that decision. If you make the wrong call you can move the market against you and lose money on your hedge. All of the above - and surely I'm leaving more examples out in this brief post - make the job an impossible one for a computer to do profitably.

I apologize that I did not make it clear that we do a lot of thinking on the desk, and I do hope this reply rectifies that mistake. What I was hoping that you would take away from the post was that we work very hard and very quickly. A lot of people out there like to think that traders are these guys that get paid a lot of money to sit at a desk and eat steak while telling sexist jokes all day. Half of the people who believe this despise the industry, and the other half want to be in the industry. I wanted to make it clear that it is not what we do, especially for people who think that the hard work will be over once they get an offer on the trading floor out of school. At times it's true that we don't get to think as much as we like to because we can be exceedingly busy and stressed out, but every single day we sit down at the desk we are expected to share our thoughts about the market and our book's positioning.

Author · 6/6/11

pandyajwalin said:
this guy is thinking of an MFE after such an experience on a daily basis!!
Was wondering- How much time will it take a fresh MFE grad to do the kind of work he is doing!!!

No, I am not thinking of getting an MFE - I'm not sure where you got that from. I already have a master's degree in another quantitative discipline and I'm quite content with that! Surely there are plenty of things I could still learn from an MFE program, but I am learning enough on the job to keep me happy.
For a fresh MFE grad to do this kind of work it would take just a couple years to really get up to speed on a trading desk. (This is assuming of course you get hired as a trader straight away.) You always continue learning but it generally takes 2-3 years to become semi-competent at your job! The same would go for an undergraduate. No matter what educational background you come from you still have no clue what you're doing during your first year.

Author · 6/6/11

It doesn't sound as if we worked for the same desk at all. First, no trader would tolerate the bonus pool being eaten up by someone who didn't work hard and perform, even if it was someone senior on the desk. No one can just sit there and do nothing - there would be mutiny. After all, we're there to make money for ourselves and having overpaid slobs around gets in our way. That's what PnL is for - if you don't do well, everyone knows. Second, the desk I work on is not a prop desk; it seems to me as if you're describing a prop desk. We don't have models that tell us to buy and sell. We have tools that present us with data that can give us an idea of relative value on various different tradable statistics. It helps us with coming up with trade ideas and biasing our spread on a price one way or the other on an illiquid request. We have tools that price exotic options for us. Neither the relative value tools nor the pricers are perfectly tradable. Same as with every other bank.

Our quants do very important work. They develop the models we use to come up with prices, see our risks, and mark our books. They develop environments in which it's easy to use these tools and in which we can view data to help us with idea generation. However, there's always a little tension on desks between quants and traders. Quants can sometimes view traders has arrogant and unintelligent; traders can sometimes view quants as too theoretical, impractical, and detached from the real world. The viewpoints come from the job descriptions and responsibilities. Quants on my desk are responsible for building and maintaining the infrastructure the traders are to use. The traders' task is to use the infrastructure along with expert knowledge of the market to generate revenue for the bank. A quant, as in your example, may come up with a very sensible model which points to a certain trade. The trader might just not take the suggestion. At the end of the day, it's not in the quant's job description to make money in the market - that's the trader's role. If the trade goes badly, the trader can whine about this, that, and the other, but it goes into his PnL and he's responsible for it. The quant ultimately won't be held accountable for risk the trader takes on. Furthermore, the feedback the quants get is generally from the traders who use their models. So of course quants will be hammered by traders if they mess up and never the other way around - traders don't get feedback from quants. Traders get hammered by management when they don't perform. It's just the organizational structure. On the other hand, traders can certainly get ahead of themselves as well. It's all too easy to forget that without quants around, traders would be stuck without the models that they so desperately need to have working in order to do their jobs. Without quants, traders would be lost. Quants certainly know how much they're needed as it's tangible. But then they forget that being a good quant doesn't usually imply being a good trader, and so good traders are needed as well to extract money out of the market. Traders focus more on the latter point and quants on the former. On my desk there have been quants who have transitioned into trading and traders who have transitioned into being quants. Most of the time they find they were better suited in their original seats, though there have been one or two exceptions.

Author · 6/6/11

Just enough to keep me doing what I'm doing.

Author · 6/7/11

Nope. Barely know what you're talking about.

Author · 7/1/11

There are a few ways to get into trading, but none of them are a guarantee. The easiest is probably starting out in a S&T analyst or associate program. In order to get into one of those programs without some sort of connection, you pretty much have to attend a "target school" for the bank in question. Target schools generally include the Ivy League as well as a handful of other top US universities. I don't know as much about getting into trading in Europe or Asia. Anyway, it's 500x more difficult to get an interview from a non-target school. But even if you do go to a target school, it's still incredibly competitive to land an interview. To do this you have to have done very well in school (high GPA), have high test scores, and some solid extra-curriculars. And even if you do land an interview, it's still difficult to make it through the process - which includes multiple rounds - and land the job. For these reasons, never consider a degree from any program from any university to be a sure way of getting into trading at a bulge bracket bank. You could also, as you said, start in another area of the bank which is not as competitive to get into and then hope to move onto the trading desk. From what I've seen, while this is not impossible, it is still very difficult to do. But it may be the best option for some people. The important take away there is don't sign up to do a job you know you're doing to dislike just because you think it'll definitely land you a spot on a trading desk in a couple years. It very well may not. In any case, from what I've seen, most traders began trading just out of school, i.e. they were hired into the above mentioned graduate programs, and a minority have crossed over from different areas such as quant and market risk.

Alice G · 7/3/11

5:30am + almost zero quantitative work? No, thanks.

Author · 7/9/11

Ekongkong, that's not entirely correct. If you get your MFE from a target school, you may be in the running for an analyst or associate program. There are also trading desks that prefer to hire graduate students and recruit directly from certain programs. It will matter a whole lot why it is you're getting an MFE to begin with if your goal is to be a trader. If you're coming straight out of undergrad, it means either you were unable to secure a position in your senior year - either because the university you attended didn't provide you the access you needed to the banks or because you didn't make it through any of the processes - or you studied something unrelated, had different goals, and now want to focus on trading as a career path. In either case, getting the MFE will increase your chances of getting in the door. You will of course have to do a bit of cost/benefit analysis on your own about whether the tuition is worth the extra shot at making it to a desk. Another reason why someone might want an MFE is because they're changing careers. If you want to be a complex derivatives trader, it makes more sense to get an MFE than an MBA (though there are complex derivatives traders with MBAs). In this case it will as well increase your chances of getting onto a trading desk, as desks will prefer you to have some meaningful exposure to the material if you're in your late 20s and want to join up. In any case, it's not like getting an MFE will turn your chances of getting a seat on a desk from 2% to 92% - it's more like from 2% to 12% (please don't analyze these numbers). There just simply aren't as many seats available as there are students who want to fill them. This is why getting into trading is so competitive and why you can't view any degree as a sure shot way to get into trading. Also just want to point the following out - I've come across so many students who feel that simply because they've made it into an MFE program, that means they will definitely get to be a trader just out of school. That idea needs to be wiped away from students' minds. You need to do very well in that program and you need to do legwork on your own to get yourself into interviews. But would an MFE simply increase your chances if you didn't make it in straight out of undergrad? Yes. It's up to the prospective student to judge how much that increase is and weigh that against tuition costs.

Author · 7/25/11

Generally that's true though I suppose it varies between different types of desks. Usually when one refers to a quant at an investment bank, it is a someone who develops pricing models and applications for financial products, which are used by traders to price and risk manage those products. So in that sense, yes, quants support traders. However, there is not just one kind of quant. You can find other quants at banks developing black box trading strategies, supporting risk management (departments that independently monitor trading desks' risk), creating relative value analysis tools and backtesting, and other generally quantitative jobs. You can also find quants in other places, from hedge funds to insurance companies. In particular, quants at hedge funds are probably the most likely to be those who are quant / trader hybrids, designing the logic by which trading decisions should be made as well as executing those trades. The main point, however, is that the term quant refers to a slew of jobs - hence your confusion. As a trader, the kind of quant I come into contact with at work almost exclusively is the kind who supports traders, but that's not the only kind there is.

Author · 8/8/11

It was a masters degree in a quantitative field linked to finance - so not a masters in math or physics or electrical engineering, etc., though you will find those degrees in the field.

abc · 11/25/11

"masters degree in a quantitative field linked to finance"

Masters in economics / quantitative economics/ OR ?

NSelectrical · 12/17/11

could you tell us how you got to be a trader. As in your academic and career path which landed you here. And the schools you attended. Thanks!

Author · 12/18/11

Responding to ABC and NSELECTRICAL:

I haven't given out very specific information here because I do want to stay anonymous - I don't want anyone in the market knowing I'm posting stuff up here. But here's my general background:

I am American, went to a very good but non-target (in NYC, anyway) undergrad and majored in math. I then did a masters degree at a top name target university (think ivy league or equivalent), in as I described earlier as a quantitative field linked to finance. In that degree I studied both applied mathematical material, like stochastic stuffs and programming (simulation and applied stuff like that), as well as finance courses - but in the derivatives / capital markets realm specifically. That's as specific as I'd like to be here. In between academics I did internships at a hedge fund and at a bank in risk management. By the time I was looking for a full time job, hedge funds simply were not hiring (not that that was necessarily what I wanted to do, it just was not an option, and certainly not a the hedge fund at which I had spent time since the division I was working in was shut down) and I knew I did not want to accept the offer to join the bank in risk management full time (because I didn't like risk management, though the people were very nice), so I continued to look for jobs elsewhere. Another bank was hiring at my grad school for a junior trading job. I knew next to nothing about what they actually did, but I interviewed (of course several rounds), got the offer, they made it sound cool and exciting, and so I accepted. And I've been there since and it's actually been pretty cool and exciting, so I guess they've lived up to their promise. It has its ups and downs though. Anyway, hope this response helped!

NSelectrical · 12/21/11

thank you

Author · 1/21/12

Just a quick response to Alex's posting a while back about 5:30am and zero quantitative work. Well, really just about the zero quantitative work since there's no way around the early morning wake up (though I do know people who live close by work and wake up as late as 6:30). It really is not true to say that there is zero quantitative work involved in this job. First of all, you've got to be able to understand pricing and risk management of derivatives, which is itself quantitative. Second, I often find myself scripting up payoff structures in C++ (or something close to it) in order to price non-standard products using a monte carlo engine as well as coding in VBA to create spreadsheet tools that search for trade opportunities or present ways of viewing risk. Third, I work very closely with the quant group in model development, improvement, and implementation, which involves a lot of testing and sifting through quantitative concepts. At the end of the day, we are not quants and so do not do as much rigorous quant work as they do, but we certainly have a good amount of quantitative work to do.

Deepanshu · 6/18/12

Hi Author,

Quite an insightful article on the work life of a trader. thanks!

srm · 6/18/12

This was a really interesting article. Thanks for sharing your perspective. Is there any particular resource/exp that you find most useful, looking back?

A Day in the Life of an Exotic Derivatives Trader

Abdel

Economist

pandyajwalin

Author

Author

Author

Author

Author

Author

Alice G

Author

Author

Author

abc

NSelectrical

Author

NSelectrical

Author

Deepanshu

srm

Similar threads