Leaving Physics: Other Questions and Answers

If you found this page through a search engine, you might want to go back to the beginning of the Leaving Physics Web Site before reading this page.

Some of the questions that I have received from people about my experiences leaving the academic physics world didn't quite fit into the themes or space of the other pages in the Leaving Physics web site. Because I thought they might be of some interest to other readers, I have included edited versions of some of them here, along with my answers.

This list of questions and answers has gotten long, so I've finally organized it by topic:

Preparation for Particular Jobs

Q: What background/skills would be useful for getting a job in the finance industry? Do you have any reading recommendations? What else can I do to prepare for job interviews?

A: The skills that I tended to use in my Wall Street job, and that my group was looking for were: C/C++ programming, numerical analysis programming (e.g. for solving differential equations, solving linear equations, etc), and general mathematical skills like those that a theoretical physicist usually would have. A Wall Street recruiter who wrote to me in December 2005 (Jennifer Levin) said that a good programming background is a definite must for getting a Wall Street job, but not necessarily a background in finance.

If you do want to get some background in finance, you might consider taking a class in financial derivatives, or reading one or more books. The following list was compiled from my own knowledge, and two email correspondents (Thomas Haeusser and Jennifer Levin; the latter sent a list credited to Karhan Akcoglu):

  • John Hull: Futures, Options, and Other Derivative Securities (good overview of the subject)
  • Salih N. Neftci: Introduction to the Mathematics of Financial Derivatives (more math detail than Hull's book, less coverage)
  • Jarrow: Modelling Fixed Income Securities (speicalized book about Black-Scholes equation)
  • Wilmott et al: The Mathematics of Financial Derivatives (specialized book about binomial trees)
  • Timothy Falcon Crack: Heard on the Street: Quantitative Questions from Wall Street Job Interviews
  • Johnson & Reiss: Numerical Analysis (or any other introductory numerical analysis text)
  • Bernt K. Oksendal: Stochastic Differential Equations: An Introduction with Applications

Besides learning about derivatives, you might find it useful to learn about the different functions of the big investment banks: investment banking, stock/bond/commodity trading, asset management, etc. I am not sure how you should go about learning about them, maybe from these companies' web sites or a general intro to finance book or a career description book? Also, your recruiter is often a good resource.

Q: Should I wait until I actually have my PhD to apply for a finance job, or should I do it several months before?

A: I would talk to a recruiter maybe six months ahead and find out what they are currently recommending -- the time it will take depends a lot on the job market. Also, if you call a recruiter, he or she may be able to help you get your resume and skill set in order before you start actually looking.

Q: What about working for a few years in between undergrad and grad school? It seems like getting some money in the bank would not be a bad idea so that I wouldn't have to live in total poverty during grad school, and that having some real work experience under my belt would help my future prospects.

A: My experience of going straight through school without a long break was that I didn't feel like I was living in poverty. Most physics grad students, at the reasonably good schools anyway, have support from either a teaching assistantship, research assistantship, or fellowship, and the experience of other physicists I knew at grad school was that it was plenty to live on (besides being about 4 times what I was living on as an undergrad), and gave us enough income to pay the rent, eat, and go to movies and out to eat if we wanted to. If you're conditioned to having a real job and earning real money and living without roommates before you go to grad school, however, of course you will think it a definite step down. And I wouldn't say that getting a PhD is really torture, either. Working on Wall Street I found to be much much much more stressful. Another consideration is time -- if you take time off between undergrad and grad school, you will add a few more years to the already mounting years before you start your "real" career.

Q: What background/skills would be useful for getting a job in software consulting?

A: Programming in C/C++/Java/PERL/SQL, and experience with HTML; computer science coursework; written and oral communications skills; experience with programming tasks such as database design, graphical user interfaces, designing classes and data structures; experience working on large group projects; experience bringing software from conception to finished, marketable product.

Q: I have heard that having a PhD can sometimes make it hard to find a job -- that many corporations are reluctant to hire a PhD because they have to pay them more. I suppose there is some truth to this, but can you elaborate on it for me, please?

A: Well, I suppose it might be true. I have also heard that some companies may be reluctant to hire physics PhDs, because they wonder if they are capable of doing practical things (due to stereotypes of scientists in white lab coats and such). But I personally have never felt either one to be the case in my job searches.

That said, probably if you are looking to get a job in a company, getting a PhD in physics (assuming you don't already have one) is not the best route. You might be better off first finding companies you might be interested in working in, and reading the job descriptions on their web sites to find out what the best education/training would be.

Q: Is age a factor in getting jobs?

A: Sad to say, it may be. In the U.S., age discrimination is illegal, but that doesn't mean in doesn't happen. In some other countries, it isn't even illegal. In many cases, I believe that people who are around... say, 30 or so, maybe? fit the "image" that interviewers are at least subconsciously looking for, when they consider people for many jobs. So, if you're a lot younger or older than that, you may have to consider how to present yourself as either mature or dynamic for your age, and how to present your somewhat non-traditional experience as something that's a benefit to the company.

For one more perspective on the age factor, a correspondent from India wrote, "If you can add something about the age factor in the change of track issue in your web page, some people would stop dreaming wildly and stick to the job at hand. There are laws, equal opportunity stuff and so on but age discrimination is a reality even in the USA, I suspect."

Skill Improvement

Q: How do I go about improving my oral communication skills?

A: If you are a grad student, try to become a teaching assistant, even part-time for a semester or two. If you are doing research, attend the American Physical Society or other meetings and give a talk. Organize a group of grad students to get together for a weekly "journal club", where each week one of you gives a talk on a current journal article in your field. Take a class in public speaking. Join Toastmasters.

Q: How do I go about improving my written communication skills?

A: Take a writing class. Write papers and get someone who is a good writer to critique them.

My Experiences

Q: What has been your career path, and why?

A: After finishing my PhD, my first job was a two-year post-doc at Bell Labs. It was during my second year there, as I began to look for academic jobs, that I decided that I did not want to be an academic physicist. I also wanted to stay in the greater New York area at the time, and so I began to look for non-academic jobs in the area. I happened to find a notice from a Wall Street recruiter who was used to talking to physicists on an internet job search site (I searched for the keyword "physics"), and called him up. Two weeks later I had a job offer at Goldman, which I took. I'm sorry to say that it wasn't something I really thought about and decided to do -- it just sort of happened.

After two years on Wall Street, I decided that the stress level and the long hours were outweighing the excellent financial compensation I found there, and decided to leave; I also decided that I wanted to leave the New York City area and return to my home state of Washington. I had a friend or two who had worked as a consultant, and I liked the idea of a flexible schedule and being my own boss, so I decided to give it a try. I got my start as a consultant for Goldman, in my old group, so that made it easy to get into the business; once I was somewhat established, I found other clients fairly easily, and also started getting phone calls based on my web site after about four months.

However, after giving consulting a whirl for a year or so (and moving to Seattle in the midst of it), I was bored with the kinds of work I was finding to do as a consultant, and I was also discouraged by having found a couple of potential clients who I felt were unethical in their business practices. So, I gave up consulting and went to work as a Research Scientist for a company called MathSoft in Seattle. I decided that this job offered me a reasonable combination of salary, job satisfaction, and autonomy, and that I could give up the freedom of being a consultant.

After a couple of years there, it was time for A Grand Adventure, so I quit my job at MathSoft, and went to El Salvador to do some volunteer work for a few months. Upon returning, I started working as a full-time telecommuter for a company called Kiodex, that a friend of mine was starting in New York. That lasted for about six months, before I started going crazy from being at home alone all day, and cut back to half time, and then converted from being an employee to being a contractor. At that point, I started finding other clients and am now in business for myself as a Freelance Software Developer.

Q: Can you comment on the work, atmosphere, and satisfaction of the places you have worked since leaving academic physics?

A: First off, see my page on what it's like to work outside of academic physics. But to expand:

  • Bell Labs -- reasonable hours, reasonable pay, interesting work, allows quite a bit of autonomy, lots of collaboration going on between researchers, generally positive experience working there. Others I know doing applied research in physical sciences who are still at Bell Labs or at other companies are quite happy with their jobs. One very nice thing about working at Bell Labs was the quantity and quality of seminars and people visiting the labs.
  • Goldman Sachs -- very long hours, very stressful, very high pay.
  • MathSoft -- OK. Much like working at Bell Labs, except that it's a much smaller company and the research group of about 20 people had a much narrower focus.
  • Kiodex -- full-time telecommuting was too isolating for my tastes.
  • Consulting -- Working at home takes some getting used to, but I have a good balance now between clients I visit and clients for whom I work at home. You do have to spend some time marketing and networking, but in return you have flexible hours and good pay (assuming you can find clients). Not so much of the intellectual stimulation of doing research, and no colleagues around to talk to. Much uncertainty about where your next paycheck might be coming from.

Q: Can you elaborate even more about your experience on Wall Street? For example, what causes the stress - short deadlines, demanding bosses, both?

A: Short deadlines are definitely a factor, but I think that the main thing in my case was that I was working with traders, who are always very stressed out, and it is stressful working with people like that every day. Also, in the derivatives business, the way a company gets ahead of its competition is to come up with new ideas that it can price and its competition hasn't figured out yet, and by figuring out how to price it's competition's ideas quickly. The only way this can work is if the people pricing the derivatives (i.e. the people in my group) are both quick and VERY accurate. It's stressful.

Q: To continue elaborating, how long is a typical work week? Are you expected to work every day? And is it possible to work from home?

A: I worked five days a week, and my minimum day was from 9AM to 7AM, with many days much longer than that. Some people in my group definitely worked much more than that, and I think that most worked somewhat more than that, so I might not be a good example of a typical employee. Since no trading happens on weekends, most people that I know of do not have to work weekends, but I sometimes did log in from home to check on computer runs on the weekends. Most people that I know of on Wall Street are not allowed to work from home during the week, but of course if you are putting in extra time on the evenings and weekends you would be welcome to do that at home, on your commuter train, or wherever suits you.

Q: And what would be the typical mix between symbolic math/thinking and programming in a Wall Street physicist job?

A: It really depends on the firm and the particular job. Some people I know on Wall Street who are physicist types do almost exclusively programming; at the other extreme, some do almost none. Most do some. You are allowed to ask questions at your interview, so find out what they expect before taking the job if this is a concern.

Q: I've started a new non-academic job, and I am experiencing minor "anxiety attacks" from time to time, and feeling that something is missing -- I think I'm missing the academic life, though my current job is interesting and rewarding, I feel that I'm much more appreciated than I was in academia, and I am glad I'm not still in the midst of the terrible working conditions and low job security of academia. What can I do about my anxiety?

A: It seems to me that your anxiety is not about missing the Academic Life as a whole, but only some part or aspect of it. If you can figure out what that part is, I think you will either (a) see a clear way to continue doing that in your life, (b) be able to come to accept that "this is what I gave up" in order to avoid the negative aspects of the academic life, or (c) realize that the thing you are "missing" is only a part of the Ideal Fantasy World Academic Life, not the Real World Academic Life (e.g., if what you are missing is "academic freedom", you might realize that in practice, you're only really free to study what you can get money for, so "academic freedom" may not really exist in the Real World).

What to Expect

Q: What would a PhD with no experience expect to earn in the beginning, including signing bonuses, performance incentives, salary, etc? How much does salary increase with experience? How does it compare to the more traditional (MBA) hires at investment banks?

A: That depends. If the job market is tight, you won't get a job unless you have at least taken a class in derivatives, so asking about salary is irrelevant. And it was in 1994 when I started working on Wall Street, so my information is probably out of date -- talk to your recruiter to find out more. Don't expect a signing bonus. Your salary is likely to stay about the same for years on end, with perhaps an infrequent cost-of-living increase, but your bonus will probably increase quite rapidly, if you perform well. Just how much depends on the company you work for.

If you get into something like derivative price modeling, you will likely find that the traders you are supporting with your modeling will always be earning much more than you. On the other hand, your compensation will likely still be very good, and keep increasing every year; your work will be more interesting and less stressful; and you will be working with smart and interesting people, so you may find, as I did, that the compensation discrepancy was not a major issue.

Q: If a PhD is hired into a modeling job (typical), how difficult would it be to move into more traditional disciplines (if desired), such as investment banking and trading? Would it be easy to gain the knowledge needed (MBA?), and would the employer encourage such a transition?

A: In my experience, that is at least possible at most firms and it would be strongly encouraged in others. If that is what you want to do, you can definitely help your chances by seeking out as much knowledge as possible about what is going on in your department/division/company, by asking questions and keeping your eyes open. My experience is also, however, that once you meet the people who are using your models, you probably won't want to become one of them -- I personally would rather work in a group filled with science/engineering types than MBAs.

Your company might fund you to get an MBA, but in my group/area at Goldman, having an MBA was of no use whatsoever. Generally, MBA programs (this might be changing) focus on management techniques and analysis of company case studies, and as such are mostly useful for investment banking and stock analysis, whereas most physics types are used in modeling having to do with trading (e.g. derivative pricing), which is completely separate.

Q: Why do people on Wall Street change jobs so often? How long would they be unemployed in between jobs? Is the career as fluid for physicist-types working in quantitative positions?

A: There are quite a few factors. Some people are laid off when the market goes down. Some are courted by other firms and get a better offer. Some grow to not like where they are for other reasons and look around. I don't know of anyone that was temporarily unemployed for very long if they had Wall Street experience -- maybe a couple of weeks, unless they wanted time off on purpose. And yes, all of these things seem to apply to quantitative types too -- if the market hits a rough spot, the firm is not likely to need "quants" to support the traders they have to lay off.

Q: I have heard that the interviews on Wall Street are grueling, and that before finalizing a hire, the firm will do a background check. Is this true?

A: Yes. The interviews are quite intense, at least in my experience, and may go on for a couple of days. And most if not all firms will check every fact on your resume or other materials you gave them, do a "routine" background and credit check, and make their offer contingent on their not finding anything fishy. It is an extremely bad idea to misrepresent yourself on your resume in any way.

Q: Speaking of interviews, what do you say when they ask you the inevitable question: Why are you leaving physics?

A: Probably the best thing to do is to be honest -- a good idea in an interview in any case. Making up stories may end up making you sound (and feel) insincere... but of course you don't want to sound desperate either, if that is the honest answer.

Saying that you've always wanted to get into the finance industry is not likely to be believed -- if so, why did you go into physics? Saying that you were always excited about physics for the love of knowledge, but now are not so excited about becoming a professor, writing grants, and the long-term nature of the work might be a better idea.

Another possibility is sort of to evade the question, if your honest answer is not something you want to say; for example, to use the neutral "for personal reasons" answer. This will probably lead people to believe you are trying to find employment in some particular geographic area, because of spouse/family/health/personal reasons. And if the interviewer is a reasonable person, he or she will probably avoid probing further; if not, you may have to say "I would rather not discuss it," and consider whether you really want to work with this unreasonable person. Saying all of this in a cheery tone of voice is probably a good idea -- don't want to give them the idea you are in dire straits.

Q: Isn't another difference between working as an academic physicist and working for a company that physicists tend to be geeks and "know-it-alls", have no imagination, and be overly focused on their research, and that people in corporate settings are less so?

A: I think not.

First of all, I don't agree that you can characterize academic scientists that way. When I was a physics undergrad, I thought for a while that my professors fit that description, until I kept in touch with a few of them after completing their classes, and found out that they were very well read, devoted parents, musicians, artists, etc, and generally very interesting people. The same proved to be true of my fellow undergrads, as well as the graduate students and professors I met during my PhD studies at Cornell, once I got to know them, as well as the people I met at conferences and during a couple of stays at other institutions. I think that some people may be turned off of physics and physicists by the way they're portrayed in the news and entertainment media, combined with the regrettable aura that some professors project to their beginning undergrad students, but I don't think that they're all boring, one-sided geeks.

Secondly, I haven't found that much difference between my co-workers in the academic settings I have been in and the corporate settings I've been in. Of course, most of the groups I have worked in have been made up of other scientists and engineers, so that is not too surprising.

Questions which I cannot answer

Here are some questions I have received which I am unable to answer. Most can be answered by a recruiter familiar with the target industry, or perhaps a university career center. Generally, if you search on Monster or one of the other job sites, a large fraction of the job postings will be from recruiters (you will be able to tell because the posts will not give the name of the company you would work for; instead they will say something like "A large software company in Redmond, WA"). See which recruiters are posting positions that interest you, and either call or send a resume. If you are qualified for that type of position, you should get a call back.

  • Do you have any contacts at companies, or suggestions about which recruiter to use?
  • How is the job market on Wall Street/Industrial Research/Academia?
  • Which companies are hiring physicists right now?
  • Do you have any specific advice for people wanting to get into neural networks? pattern recognition? bayesian inference? (all fields I know very little about -- I've put information about fields I know about on the pages of this site, plus some suggestions that other readers contributed)
  • What jobs can I do with a bachelor's/master's degree in physics?